Sunteți pe pagina 1din 324

Med Clin N Am 88 (2004) xv–xvi

Preface

Type 2 Diabetes Mellitus

Alan J. Garber, MD, PhD


Guest Editor

Diabetes has become a plague upon modern America. It is the most ex-
pensive chronic illness in the country and has become a scourge of multiple
inpatient medical and surgical services. Indeed, diabetes is the fourth most
common comorbid condition in hospitals and accounts for prolonged
lengths of stay and excess costs for virtually one quarter of all hospital
admissions in the United States. More importantly, however, is the bur-
geoning number of diabetic patients. Each year, approximately 1 million
new cases of type 2 diabetes are diagnosed, and even type 1 diabetes appears
to be increasing in incidence in the United States. Because of our ability to
reduce or prevent the costly chronic complications of diabetes, as shown by
the results of recent clinical trials, newer, more aggressive standards of care
for both glucose control—as well as blood pressure, lipids, and other risk
factors for these chronic complications—have been promulgated recently
by a variety of organizations, chiefly the American Association of Clinical
Endocrinologists, The National Cholesterol Education Program, and the
American Diabetes Association. As new classes of compounds and new
agents within each class develop, thereby expanding our therapeutic options
for patients with diabetes, the potential for drug interactions and drug dif-
ficulties seems to mount geometrically.
In this issue of the Medical Clinics of North America, we have endeavored
to bring to practicing clinicians the most modern strategies by which to

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.05.003
xvi A.J. Garber / Med Clin N Am 88 (2004) xv–xvi

manage patients with diabetes by meeting these goals and guidelines. We


hope to improve the risk of complications and to better the lives of those
patients who have diabetes, both type 1 and type 2.

Alan J. Garber, MD, PhD


Departments of Medicine, Biochemistry and Molecular Biology
and Molecular and Cellular Biology
Baylor College of Medicine
Chief of Endocrinology
Diabetes, and Metabolism
The Methodist Hospital
6550 Fannin Street, Suite 1045
Houston, TX 77030, USA
E-mail address: agarber@bcm.tmc.edu
Med Clin N Am 88 (2004) 787–835

Pathogenesis of type 2 diabetes mellitus


Ralph A. DeFronzo, MD
Diabetes Division, University of Texas Health Science Center, 7703 Floyd Curl Drive,
San Antonio, TX 78229, USA

Normal glucose homeostasis


A discussion of the pathogenesis of type 2 diabetes mellitus must start
with a review of mechanisms involved in the maintenance of normal glucose
homeostasis in the basal or postabsorptive state (10–12 h overnight fast)
and following ingestion of a typical mixed meal [1–9]. In the postabsorptive
state the majority of total body glucose disposal takes place in insulin-
independent tissues. Thus, approximately 50% of all glucose use occurs in
the brain, which is insulin-independent and becomes saturated at a plasma
glucose concentration of approximately 40 mg/dL [10]. Another 25% of
glucose disposal occurs in the splanchnic area (liver plus gastrointestinal
tissues), which is also insulin-independent. The remaining 25% of glucose
use in the postabsorptive state takes place in insulin-dependent tissues,
primarily muscle, and to a lesser extent adipose tissue. Basal glucose use,
approximately 2.0 mg/kg/min, is precisely matched by the rate of endog-
enous glucose production (Fig. 1). Approximately 85% of endogenous
glucose production is derived from the liver, and the remaining 15% is
produced by the kidney. Glycogenolysis and gluconeogenesis contribute
equally to the basal rate of hepatic glucose production.
Following glucose ingestion, the increase in plasma glucose concentration
stimulates insulin release, and the combination of hyperinsulinemia and
hyperglycemia (1) stimulates glucose uptake by splanchnic (liver and gut)
and peripheral (primarily muscle) tissues and (2) suppresses endogenous
(primarily hepatic) glucose production (Box 1) [1–9].
The majority (80%–85%) of glucose uptake by peripheral tissues occur
in muscle, with a small amount (4%–5%) metabolized by adipocytes.
Although fat tissue is responsible for only a small amount of total body
glucose disposal, it plays a very important role in the maintenance of total
body glucose homeostasis by regulating the release of free fatty acids (FFA)
from stored triglycerides (see discussion below) and through the production
of adipocytokines that influence insulin sensitivity in muscle and liver

0025-7125/04/$ - see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.013
788 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Fig. 1. Postabsorptive state. Glucose production and glucose use in the normal human in the
postabsorptive state. (From DeFronzo RA. Pathogenesis of type 2 diabetes mellitus: metabolic
and molecular implications for identifying diabetes genes. Diabetes 1997;5:117–9; with
permission.)

[11–14]. Insulin is a potent antilipolytic hormone, and even small increments


in the plasma insulin concentration markedly inhibit lipolysis, leading to
a decline in the plasma level of free fatty acid [12]. The decline in plasma
FFA concentration augments muscle glucose uptake and contributes to the
inhibition of hepatic glucose production. Thus, changes in the plasma FFA
concentration in response to increased plasma levels of insulin and glucose
play an important role in the maintenance of normal glucose homeostasis
[11–14].
Glucagon also plays a central role in the regulation of glucose
homeostasis [9,15]. Under postabsorptive conditions, approximately half
of total hepatic glucose output is dependent on the maintenance of normal
basal glucagon levels, and inhibition of basal glucagon secretion with
somatostatin causes a reduction in hepatic glucose production and plasma

Box 1. Factors responsible for the maintenance of normal


glucose tolerance in healthy subjects
A. Insulin secretion
B. Tissue glucose uptake
1. Peripheral (primarily muscle)
2. Splanchnic (liver plus gut)
C. Suppression of HGP
1. Decreased FFA
2. Decreased glucagons
D. Route of glucose administration
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 789

glucose concentration. After a glucose-containing meal, glucagon secretion


is inhibited by hyperinsulinemia, and the resultant hypoglucagonemia
contributes to the suppression of hepatic glucose production and mainte-
nance of normal postprandial glucose tolerance.
The route of glucose entry into the body also plays an important role in the
distribution of administered glucose and overall glucose homeostasis
[9,16,17]. Intravenous insulin exerts only a small stimulatory effect on
splanchnic (liver plus gut) glucose uptake, whereas intravenous glucose
augments splanchnic glucose uptake in direct proportion to the increase in
plasma glucose concentration. In contrast, oral glucose administration
markedly enhances splanchnic glucose uptake. The portal signal that
stimulates hepatic glucose uptake after glucose ingestion is directly pro-
portional to the negative hepatic artery–portal vein glucose concentration
gradient [9]. As this gradient widens, the splanchnic nerves are stimulated, and
this activates a neural reflex in which vagal activity is enhanced, and
sympathetic nerves innervating the liver are inhibited. These neural changes
augment liver glucose uptake and stimulate hepatic glycogen synthase, while
simultaneously inhibiting glycogen phosphorylase. In contrast to intravenous
glucose/insulin administration, in which muscle accounts for the majority
(80%–85%) of glucose disposal, the splanchnic tissues are responsible for
the removal of approximately 30%–40% of an ingested glucose load. Glucose
administration through the gastrointestinal tract also has a potentiating effect
on insulin secretion. Thus, the plasma insulin response following oral glucose
is approximately twice as great as that following intravenous glucose, despite
equivalent increases in the plasma glucose concentration. This increase in
effect is related to the release of glucagon-like peptide (GLP)-1 and glucose-
dependent insulinotropic polypeptide (GIP) (previously called gastric in-
hibitory polypeptide) from the intestinal tissues cells [18,19].

Glucose homeostasis in type 2 diabetes mellitus


Type 2 diabetic subjects manifest multiple disturbances in glucose
homeostasis, including: (1) impaired insulin secretion; (2) insulin resistance
in muscle, liver, and adipocytes; and (3) abnormalities in splanchnic glucose
uptake [1,2,20,21].

Insulin secretion in type 2 diabetes mellitus


Impaired insulin secretion is found uniformly in type 2 diabetic patients in
all ethnic populations [1,2,20–29]. Early in the natural history of type 2
diabetes, insulin resistance is well established but glucose tolerance remains
normal because of a compensatory increase in insulin secretion. This dynamic
interaction between insulin secretion and insulin resistance has been well
documented. Within the normal glucose tolerant population, approximately
790 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

20%–25% of individuals are severely resistant to the stimulatory effect of


insulin on glucose uptake (Fig. 2) (measured with the euglycemic insulin
clamp), and subjects in the lowest quartile of insulin sensitivity are as insulin
resistant as type 2 diabetics (see Fig. 2). Insulin secretion (measured with the
hyperglycemic clamp technique) in these insulin-resistant, nondiabetic
individuals, however, is increased in proportion to the severity of the insulin
resistance, and glucose tolerance remains normal. Thus, the beta cells are able
to ‘‘read’’ the severity of insulin resistance and adjust their secretion of insulin
to maintain normal glucose tolerance.
In type 2 diabetics, the fasting plasma insulin concentration is normal or
increased and basal insulin secretion (measured from C-peptide kinetics) is
elevated. The relationship between the fasting plasma glucose (FPG) and

Fig. 2. (A) Whole-body rate of glucose disposal during euglycemic insulin clamps in 32 women
divided according to quartiles of insulin sensitivity. * P  0.001 for each quartile versus the
adjacent quartile. (B) Time course of plasma insulin response during the hyperglycemic clamp in
the same 32 women divided into quartiles of insulin sensitivity. Insulin secretion rose
progressively from the highest to the lowest quartile of insulin sensitivity (P  0.01). , Quartile
1; 6, quartile 2; , quartile 3; C, quartile 4. (From Diamond MP, Thornton K, Connolly-
Diamond M, Sherwin RS, DeFronzo RA. Reciprocal variations in insulin-stimulated glucose
uptake and pancreatic insulin secretion in women with normal glucose tolerance. J Soc Gynecol
Invest 1995;2:708–15.)
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 791

insulin concentrations resembles an inverted U shape or horseshoe [1,2].


Because this curve resembles Starling’s curve of the heart, it has been
referred to as Starling’s curve of the pancreas. As the fasting glucose rises
from 80 to 140 mg/dL, the fasting plasma insulin concentration increases
progressively, reaching a peak value 2.0–2.5-fold greater than in normal
weight, nondiabetic, age-matched controls. The progressive rise in fasting
plasma insulin level can be viewed as an adaptive response of the pancreas
to offset the progressive deterioration in glucose homeostasis. When the
FPG exceeds 140 mg/dL, the beta cell is unable to maintain its elevated rate
of insulin secretion, and the fasting insulin concentration declines pre-
cipitously. This decrease in fasting insulin level has important physiologic
implications, because it is at this point that hepatic glucose production (the
primary determinant of the FPG concentration) begins to rise.
The relationship between the mean plasma insulin response during an
OGTT and the FPG concentration also resembles an inverted U-shaped curve
(Fig. 3) [1,2]. The curve, however, is shifted to the left compared with the basal
insulin secretory rate, and the glucose-stimulated insulin response begins to
decline at a fasting glucose concentration of approximately 120 mg/dL. A
typical type 2 diabetic subject with a FPG level of 150–160 mg/dL secretes an
amount of insulin similar to that in a healthy nondiabetic individual; however,
a ‘‘normal’’ insulin response in the presence of hyperglycemia and underlying
insulin resistance is markedly abnormal. At FPG levels in excess of 150–160
mg/dL, the plasma insulin response, when viewed in absolute terms, becomes

Fig. 3. Starling’s curve of the pancreas for insulin secretion. In normal-weight patients with
IGT and mild diabetes, the plasma insulin response to OGTT increases progressively until the
fasting glucose reaches 120 mg/dL. Thereafter, further increases in the fasting glucose
concentration are associated with a progressive decline in insulin secretion. (From DeFronzo
RA. Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A collusion responsible for
NIDDM. Diabetes 1988;37(6):667–87; with permission.)
792 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

insulinopenic. Finally, when the fasting glucose exceeds 200–220 mg/dL, the
plasma insulin response to a glucose challenge is markedly blunted. Nonethe-
less, the fasting hyperinsulinemia persists despite FPG concentrations as high
as 250–300 mg/dL, and 24-hour integrated plasma insulin and C-peptide
profiles in lean type 2 diabetic patients remain normal. These normal day-long
values result from the combination of elevated fasting and decreased
postprandial insulin and C-peptide secretory rates [30,31].
It should be emphasized that, even though the plasma insulin response is
increased in absolute terms early in the development of type 2 diabetes (FPG
 140 mg), this does not mean that beta-cell function is normal. The beta
cell responds to an increment in plasma insulin (DI) by an increment in
plasma glucose (DG) and this response is modulated by the severity of
insulin resistance, that is, the more severe the insulin resistance, the greater
the insulin response. When this index of beta-cell function is plotted against
the 2-hour plasma glucose concentration during the OGTT, the loss of
60%–70% of beta-cell function can be appreciated in individuals with
impaired glucose tolerance. In fact, normal glucose tolerant individuals in
the upper tertile of glucose tolerance (2-h plasma glucose, 120–140 mg/dL)
already have lost 50% of their beta-cell function [32].
The natural history of type 2 diabetes, starting with normal glucose
tolerance, insulin resistance, and compensatory hyperinsulinemia, with pro-
gression to impaired glucose tolerance (IGT) and overt diabetes mellitus, has
been observed in a variety of populations including whites, Native Americans,
Mexican Americans, and Pacific Islanders, and in the rhesus monkey, an
animal model that closely resembles type 2 diabetes in humans [1, 2,20–28,33–
35]. These population studies have demonstrated a strong association between
obesity and type 2 diabetes, leading to the new-world syndrome of
‘‘diabesity.’’ In high-risk populations, the progression from normal to IGT
is associated with marked increases in both fasting and glucose-stimulated
plasma insulin levels and a decrease in tissue sensitivity to insulin (Fig. 4). The
progression from IGT to type 2 diabetes with mild fasting hyperglycemia
(120–140 mg/dL, 6.7–7.8 mmol/L) is heralded by an inability of the beta cell to
maintain its previously high rate of insulin secretion in response to a glucose
challenge without further or only minimal deterioration in tissue sensitivity to
insulin. Increased basal insulin secretion and fasting hyperinsulinemia,
however, are maintained until the FPG exceeds 140 mg/dL. A similar pattern
of insulin secretion has been observed during the development of diabetes in
the rhesus monkey [33]. The aging monkey becomes obese, and a high
percentage of monkeys develop typical type 2 diabetes. The earliest detectable
abnormality (preceding the onset of diabetes mellitus) is a decrease in tissue
sensitivity to insulin, with a compensatory increase in fasting and glucose-
stimulated plasma insulin concentrations. With time, the high rate of insulin
secretion cannot be maintained, the beta cell starts on downward slope of
Starling’s curve (see Fig. 3), and fasting hyperglycemia and glucose in-
tolerance ensue. In summary, these studies are consistent in demonstrating
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 793

Fig. 4. Summary of the plasma insulin (top, ) and plasma glucose (bottom, C) responses
during a 100-g OGTT and tissue sensitivity to insulin (top, C) in control (CON), obese
nondiabetic (OB), obese glucose intolerant (OB-GLU INTOL), obese hyperinsulinemic diabetic
(OB-DIAB Hi INS), and obese hypoinsulinemic diabetic subjects (OB-DIAB Lo INS). See text
for a detailed discussion. (From DeFronzo RA. Lilly lecture 1987. The triumvirate: beta-cell,
muscle, liver. A collusion responsible for NIDDM. Diabetes 1988;37(6):667–87; with
permission.)

that hyperinsulinemia precedes the development of type 2 diabetes, and


hyperinsulinemia is a strong predictor of the development of IGT and type 2
diabetes. It should be emphasized, however, that overt diabetes (fasting
glucose  126 mg/dL) does not develop in the absence of a significant defect in
beta-cell function. The nature of this beta-cell defect is considered in
subsequent sections.

Type 2 diabetes with hypoinsulinemia


A large body of clinical and experimental evidence documents that
hyperinsulinemia and insulin resistance precede the onset of type 2 diabetes.
Nonetheless, a number of studies have shown that absolute insulin deficiency,
with or without impaired tissue insulin sensitivity, can lead to the development
794 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

of type 2 diabetes. This scenario is best exemplified by patients with maturity


onset diabetes of youth (MODY) [36–38]. This familial subtype of type 2
diabetes is characterized by early age of onset, autosomal dominant in-
heritance with high penetrance, mild to moderate fasting hyperglycemia, and
impaired insulin secretion.
MODY originally was described by Fajans and coworkers [39], and
subsequently it was demonstrated that MODY1 resulted from a nonsense
mutation in exon 7 of the hepatic nuclear factor (HNF)4a gene. It later was
demonstrated that the occurrence of MODY in French families resulted
from mutations in the glucokinase gene on chromosome 7p (MODY2) [40].
Six specific mutations in different genes have been implicated in the MODY
profile, including glucokinase and five transcription factors [36–40]:
MODY1, HNF4a; MODY2, glucokinase; MODY3, HNF1a; MODY4,
insulin promoter factor 1; MODY5, HNF1b; and MODY6, neurogenic
differentiation 1/beta-cell E-box transactivator 2. The hallmark defect in
MODY individuals is impaired insulin secretion in response to glucose and
other secretagogues; however, peripheral tissue resistance to insulin and
abnormalities in hepatic glucose metabolism also have been shown to play
some role in the development of impaired glucose homeostasis [41,42].
Although glucokinase mutations are characteristic of MODY2, genetic
studies in typical older-onset type 2 diabetic individuals have shown that
glucokinase mutations account for less than 1% of the common form of
type 2 diabetes [43].
Cerasi [21] and Luft [44] and Hales [45] and coworkers proposed that
insulin deficiency represents the primary defect responsible for glucose
intolerance in typical type 2 diabetics who do not have glucokinase or other
MODY mutations. According to these investigators, impaired early insulin
secretion leads to an excessive rise in plasma glucose concentration, and the
resultant hyperglycemia is responsible for late hyperinsulinemia. Hales and
colleagues [45] have demonstrated that many lean whites with mild fasting
hyperglycemia (140 mg/dL, 7.8 mmol/L) are characterized by insulin
deficiency at all time points during an OGTT. An impaired early insulin
response also has been a characteristic finding in Japanese Americans who
progress to type 2 diabetes [46]. Unfortunately, none of these studies provide
information about insulin sensitivity. In whites, several investigating groups
[47,48] have demonstrated normal insulin sensitivity in a minority of type 2
diabetic individuals, and it has been suggested that up to 50% of African-
American type 2 diabetic patients who reside in New York City are
characterized by severely impaired insulin secretion and normal insulin
sensitivity [49]. A similar defect in insulin secretion has been described in
black African type 2 diabetics living in Cameroon [50]. In summary, it is
clear that impaired insulin secretion, in the absence of insulin resistance, can
lead to the development of full-blown type 2 diabetes, but it remains to be
clarified how frequently a pure beta-cell defect results in typical type 2
diabetes in the general population.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 795

First-phase insulin secretion


In response to intravenous glucose, insulin is secreted in a biphasic
pattern, with an early burst of insulin release within the first 10 minutes
followed by a progressively increasing phase of insulin secretion that persists
as long as the hyperglycemic stimulus is present [51]. This biphasic insulin
response is not observed after taking oral glucose because of the more
gradual rise in plasma glucose concentration. Loss of first-phase insulin
secretion is a characteristic and early abnormality in patients destined to
develop type 2 diabetes. In most type 2 diabetic subjects, a reduction in the
early phase of insulin secretion during the OGTT (0–30 min) and during the
intravenous glucose tolerance test (0–10 min) becomes evident when FPG
concentration exceeds 110–120 mg/dL (6.1–6.7 mmol/L) [34,35,51–53].
During the OGTT, the defect in early insulin secretion is most obvious if
the incremental plasma insulin response at 30 min is expressed relative to the
incremental plasma glucose response at 30 min (DI30 ‚DG30). Although the
first-phase insulin secretory response to intravenous glucose characteristi-
cally is diminished or lost in type 2 diabetes, this defect is not consistently
observed until the FPG concentration rises to approximately115–120 mg/dL
(6.4–6.7 mmol/L). The defect in first-phase insulin response can be partially
restored with tight metabolic control [54,55], indicating that at least part of
the defect is acquired, most likely secondary to glucotoxicity or liptoxicity
[11,56–59] (see subsequent discussion). Loss of the first phase of insulin
secretion has important pathogenic consequences, because this early burst of
insulin primes insulin target tissues, especially the liver, that are responsible
for the maintenance of normal glucose homeostasis [60,61].

Causes of impaired insulin secretion in type 2 diabetes mellitus


Progression from normal glucose tolerance to IGT to type 2 diabetes with
mild fasting hyperglycemia (120–140 mg/dL, 6.7–7.8 mmol/L) is charac-
terized by hyperinsulinemia (see Figs. 3 and 4) [1,2,32]. When the fasting
glucose concentration exceeds approximately 120 mg/dL (6.7 mmol/L) and
approximately140 mg/dL (7.8 mmol/L), respectively, the fasting and
glucose-stimulated plasma insulin levels decline progressively. A number
of pathogenic genetic and acquired factors have been implicated in the
progressive impairment in insulin secretion. Pancreatic beta cells are in
a constant state of dynamic change, with continued regeneration of islets
from ductal endothelial cells of the exocrine pancreas and simultaneous
apoptosis [62]. Multiple abnormalities have been shown to disturb the
delicate balance between islet neogenesis and apoptosis (see subsequent
discussion).
Studies in first degree relatives of type 2 diabetic patients and in twins
have provided strong evidence for the genetic basis of beta-cell dysfunction
[63–66]. Impaired insulin secretion also has been shown to be an inherited
trait in Finnish families with type 2 diabetes mellitus with evidence for
a susceptibility locus on chromosome 12 [67].
796 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Both ‘‘glucotoxicity’’ [2,56] and ‘‘lipotoxicity’’ [11,57–59] are among the


acquired defects that can lead to impaired insulin secretion (Fig. 5). The
glucotoxicity hypothesis is supported by the observation that improved
glycemic control, however it is achieved (diet, insulin therapy, sulfonylureas,
metformin), leads to enhanced insulin secretion [54,55]. More direct support
of the glucotoxicity hypothesis comes from animal studies in which diabetic
rats were treated with phlorizin, a potent renal tubular glucose transporter
inhibitor that reduces the plasma glucose concentration without altering
other circulating substrate levels [68]. When administered to partially
pancreatectomized, chronically hyperglycemic diabetic rats, phlorizin re-
stores normoglycemia and results in a dramatic improvement in both first-
and second-phase insulin secretion. Conversely, when nondiabetic rats with
a reduced beta-cell mass are exposed in vivo to a chronic physiologic
increment in plasma glucose concentration of as little as 15 mg/dL, insulin
secretion by the pancreas perfused in vitro is inhibited by 75% [69,70]. These
provocative results suggest that a minimal elevation in mean plasma glucose
concentration, in the presence of a reduced beta-cell mass, can lead to
a major impairment in insulin secretion by the remaining pancreatic tissue.
Prolonged beta cell exposure to high glucose concentrations in vitro also has
been shown to impair insulin gene transcription, leading to decreased insulin
synthesis and secretion [71].
Lipotoxicity [11,57–59,72] also has been implicated as an acquired cause
of impaired beta-cell function, as individuals progress from IGT to overt
type 2 diabetes mellitus. Short-term exposure of beta cells to physiologic
increases in free fatty acids stimulates insulin secretion. Within the beta cell,
long-chain fatty acids are converted to their fatty acyl-CoA derivatives,
which lead to increased formation of phosphatidic acid and diacylglycerol.
These lipid intermediates activate specific protein kinase C isoforms, which
enhances the exocytosis of insulin. Long-chain fatty acyl-CoA also stimulate

Fig. 5. Pathogenetic factors implicated in the progressive impairment in insulin secretion in


type 2 diabetes mellitus. TNFa, tumor necrosis factor-a.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 797

exocytosis, cause closure of the Kþ-ATPase channel, stimulate Ca2þ-


ATPase and increase intracellular calcium, thus augmenting insulin secre-
tion. In contrast to these acute effects, chronic beta cell exposure to elevated
fatty acyl-CoA inhibits insulin secretion through operation of the Randle
cycle. Increased fatty acyl-CoA levels within the beta cells also stimulate
ceramide synthesis, which augments inducible nitric-oxide synthase. The
resultant increase in nitric oxide increases the expression of inflammatory
cytokines, including interleukin-1 and tumor necrosis factor a, which impair
beta-cell function and promote beta cell apoptosis.
Most recently, deficiency of or resistance to ‘‘incretins’’ have been
implicated in the pathogenesis of beta-cell dysfunction in type 2 diabetic
patients [18,19,73–78]. When glucose is administered through the gastroin-
testinal route, a much greater stimulation of insulin secretion is observed
compared with a similar level of hyperglycemia created with intravenous
glucose [73]. This observation prompted a search for the responsible
‘‘incretins’’ or gut-derived hormones that enhance glucose-stimulated in-
sulin secretion following the oral route of glucose administration. Two
gastrointestinal hormones, GIP and GLP-1, have been shown to account for
more than 90% of the incretin effect observed following glucose or mixed-
meal ingestion [74–78]. Both GIP and GLP-1 are released from endocrine
cells of the duodenum and jejunum in response to intraluminal carbohy-
drate but not in response to circulating glucose. The stimulation of insulin
secretion by both GIP and GLP-1 is dependent on the ambient glucose
concentration, which is greater when plasma glucose concentration is high
and absent when the plasma glucose concentration returns to basal levels.
Antibodies that neutralize GIP and GLP-1 impair glucose tolerance in
a variety of animal species, including primates. Although the amount of
GLP-1 released is considerably less than that of GIP, GLP-1 is such a potent
potentiator of insulin secretion that it is thought to be the major incretin. In
type 2 diabetic humans, the GIP response to glucose ingestion is normal,
indicating the presence of beta-cell resistance to the incretin-effect of GIP. In
contrast, the GLP-1 response to oral glucose is reduced. Acute intravenous
administration of GLP-1 in type 2 diabetic patients enhances the post-
prandial insulin secretory response, and chronic continuous GLP-1 admin-
istration restores near-normal glycemia in type 2 diabetic patients [74,79].
GLP-1 also has been shown to augment islet regeneration in rodents [80].
Amylin (islet amyloid polypeptide [IAPP]) has been implicated in
progressive beta-cell failure in type 2 diabetes mellitus [81–83]. IAPP, which
is packaged with insulin in secretory granules and co-secreted into the
sinusoidal space, is the precursor for the amyloid deposits that are
frequently observed in type 2 diabetic and occur spontaneously in diabetic
monkeys. At very high doses, amylin has been shown to inhibit insulin
secretion by the perfused rat pancreas in vitro. Elevated plasma islet
amyloid polypeptide levels have been demonstrated in type 2 diabetic
subjects, obese glucose-intolerant subjects, glucose-intolerant first-degree
798 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

relatives of type 2 diabetic patients, and in animal models of diabetes [81–


83]. Following its secretion, amylin accumulates extracellularly in close
proximity to the beta cell, and it has been suggested that amylin deposits
cause beta-cell dysfunction. Although it is attractive, this theory has been
challenged by Bloom and coworkers [84], who failed to find any inhibitory
effect of amylin on insulin secretion when the peptide was infused in
pharmacologic doses in rats, rabbits, and humans. Studies in transgenic
mice [85], which express the gene encoding either human or rat IAPP under
control of an insulin promoter, also mitigate against an important role of
IAPP in the development of beta-cell dysfunction. Thus, although pancre-
atic and plasma amyloid polypeptide levels were significantly elevated in
these transgenic mice, hyperglycemia and hyperinsulinemia did not develop.
A recent provocative review [86] even suggests that IAPP in the islets of
Langerhans may serve a protective role under conditions of increased
insulin secretion. In summary, definitive evidence that amylin contributes to
beta-cell dysfunction in human type 2 diabetes remains elusive, although
recent evidence [81] suggests that the combination of elevated plasma FFA
levels and amylin hypersecretion may interact synergistically to impair
insulin secretion and cause beta-cell injury.
The number of beta cells within the pancreas is an important determinant
of the amount of insulin that is secreted. Most [87–90] but not all [91,92]
studies have demonstrated a modest reduction (20%–40%) in beta-cell mass
in patients with long-standing type 2 diabetes. Obesity, another insulin-
resistant state, is characterized by a significant increase in beta-cell mass
[88], and the majority of type 2 diabetics are overweight. Thus, even a modest
reduction (20%–40%) in beta-cell mass is most impressive. On routine
histologic examination, the islets of Langerhans appear normal with the
exception of beta-cell degranulation [87–90]. Insulitis is not observed. The
factors responsible for the decrease in beta-cell mass in type 2 diabetics
remain to be identified. Several studies suggest that new islet formation from
exocrine ducts is reduced in type 2 diabetic individuals [93]. In animal model
of diabetes, there is evidence that islet neogenesis is reduced and beta-cell
apoptosis is accelerated [94]. Although recent studies with well-matched
controls (age, gender, and obesity) suggest that beta-cell mass is reduced,
even during the early stages of the development of type 2 diabetes, it seems
likely that factors in addition to beta-cell loss must be responsible for the
impairment in insulin secretion.
Low birth weight is associated with the development of IGT and type 2
diabetes in a number of populations [95]. Developmental studies in animals
and humans have demonstrated that poor nutrition and impaired fetal
growth (small babies at birth) are associated with impaired insulin secretion
or reduced beta-cell mass. Fetal malnutrition also can lead to the de-
velopment of insulin resistance later in life [96]. One could hypothesize that
an environmental influence, for example, impaired fetal nutrition leading to
an acquired defect in insulin secretion or reduced beta-cell mass, when
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 799

superimposed on insulin resistance, could eventuate in type 2 diabetes later


in life. Thus, during the normal aging process, with the onset of obesity or
with a worsening of the genetic component of the insulin resistance, the beta
cell would be called on to augment its secretion of insulin to offset the defect
in insulin action. If beta-cell mass (or function) is reduced (or impaired) by
an environmental insult during fetal life, this would lead to the development
of IGT and eventually to overt type 2 diabetes. Although such a defect
would limit the maximum amount of insulin that could be secreted, it would
not explain the progressive decline in insulin secretion in response to
physiological stimuli as individuals progress from IGT to type 2 diabetes
(see Fig. 4).

Insulin resistance and type 2 diabetes


In cross-sectional studies and long-term, prospective longitudinal studies,
hyperinsulinemia has been shown to precede the onset of type 2 diabetes in
all ethnic populations with a high incidence of type 2 diabetes [1,2,20–
27,34,35,97–102]. Studies using the euglycemic insulin clamp, minimal
model, and insulin suppression techniques have provided direct quantitative
evidence that the progression from normal to impaired glucose tolerance is
associated with the development of severe insulin resistance, whereas plasma
insulin concentrations, both in the fasting state and in response to a glucose
load (see Figs. 3 and 4) are increased when viewed in absolute terms (see
above discussion of insulin secretion). It should be emphasized, however,
that, even though the absolute insulin secretory rate is increased, beta-cell
sensitivity to glucose is markedly reduced in individuals with IGT.
Himsworth and Kerr [103], using a combined oral glucose and in-
travenous insulin tolerance test, were the first to demonstrate that tissue
sensitivity to insulin is diminished in type 2 diabetic patients. In 1975,
Reaven and colleagues [104], using the insulin suppression test, provided
further evidence that the ability of insulin to promote tissue glucose uptake
in type 2 diabetes was severely reduced. A defect in insulin action in type 2
diabetes also has been demonstrated with the arterial infusion of insulin into
the brachial artery (forearm muscle) and femoral artery (leg muscle) as well
as with radioisotope turnover studies, the frequently sampled intravenous
glucose tolerance test, and the minimal model technique [1,2,5,105–107].
DeFronzo et al [1,2,5,12,105,108,109], using the more physiologic eugly-
cemic insulin clamp technique, have provided the most conclusive documen-
tation that insulin resistance is a characteristic feature of lean, as well as obese,
type 2 diabetic individuals. Because diabetic patients with severe fasting
hyperglycemia (180–200 mg/dL, 10.0–11.1 mmol/L) are insulinopenic (see
Fig. 3), and because insulin deficiency is associated with the emergence of
a number of intracellular defects in insulin action, these initial studies focused
on diabetics with mild to modest elevations in the FPG concentration (mean,
150  8 mg/dL, 8.3  0.4 mmol/L). Insulin-mediated whole-body glucose
800 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

disposal in these lean diabetics was reduced by approximately 40%–50%,


providing conclusive proof of the presence of moderate to severe insulin
resistance. Three additional points are noteworthy: (1) lean type 2 diabetics
with more severe fasting hyperglycemia (198  10 mg/dL) have a severity of
insulin resistance that only is slightly greater (10%–20%) than diabetics with
mild fasting hyperglycemia; (2) the defect in insulin action is observed at all
plasma insulin concentrations, spanning the physiologic and pharmacologic
range (Fig. 6); and in (3) diabetic patients with overt fasting hyperglycemia
even maximally stimulating plasma insulin concentrations (under euglycemic
conditions) cannot elicit a normal glucose metabolic response. With a few
exceptions, the majority of investigators have demonstrated that lean type 2
diabetic subjects are resistant to the action of insulin [24–27,29,34,35,97,
101,107,110–112]. The ability of glucose (hyperglycemia) to stimulate its own
uptake, that is, the mass action effect of hyperglycemia, also is impaired in type
2 diabetics [113].

Site of insulin resistance in type 2 diabetes


Maintenance of normal whole-body glucose homeostasis requires a nor-
mal insulin secretory response and normal tissue sensitivity to the in-
dependent effects of hyperinsulinemia and hyperglycemia to augment
glucose uptake [1–7]. The combined effects of insulin and hyperglycemia
to promote glucose disposal are dependent on three tightly coupled
mechanisms (see Box 1): (1) suppression of endogenous (primarily hepatic)
glucose production; (2) stimulation of glucose uptake by the splanchnic

Fig. 6. Dose-response curve relating the plasma insulin concentration to the rate of insulin-
mediated whole-body glucose uptake in control (C) and type 2 diabetic () subjects. * P  0.01
versus control subjects. (From Groop LC, Bonadonna RC, DelPrato S, Ratheiser K, Zyck K,
Ferrannini E, DeFronzo RA. Glucose and free fatty acid metabolism in non-insulin-dependent
diabetes mellitus. Evidence for multiple sites of insulin resistance. J Clin Invest 1989;84(1):205–
13; with permission.)
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 801

(hepatic plus gastrointestinal) tissues; and (3) stimulation of glucose uptake


by peripheral tissues, primarily muscle. Muscle glucose uptake is regulated
by flux through two major metabolic pathways: glycolysis (of which 90%
represents glucose oxidation) and glycogen synthesis.

Hepatic glucose production


In the postabsorptive state, the liver of healthy subjects produces glucose at
the rate of 2.0 mg/kg/min [1,2,12,114]. This glucose efflux is essential to meet
the needs of the brain and other neural tissues, which use glucose at a constant
rate of approximately 1.0 to 1.2 mg/kg/min [10,115]. Brain glucose uptake is
insulin-independent and accounts for approximately 50% to 60% of glucose
disposal under fasting conditions. Therefore, brain glucose uptake occurs at
the same rate during absorptive and postabsorptive periods, and it is not
altered in type 2 diabetes. During glucose ingestion, insulin is secreted into the
portal vein where it is taken up by the liver and suppresses hepatic glucose
output. If the liver does not perceive this insulin signal and continues to
produce glucose, there will be two inputs of glucose into the body, one from the
liver and a second from the gastrointestinal tract, and marked hyperglycemia
will ensue.
In type 2 diabetics with mild fasting hyperglycemia (140 mg/dL), the
postabsorptive level of hyperinsulinemia is sufficient to offset the hepatic
insulin resistance and maintain a normal basal rate of hepatic glucose output
[114]. In diabetic subjects with mild to moderate fasting hyperglycemia (140–
200 mg/dL, 7.8–11.1 mmol/L), however, basal hepatic glucose production is
increased by approximately 0.5 mg/kg/min (Fig. 7) [1,2,12,114]. Thus, during
overnight sleeping hours (20:00 h to 08:00 h), the liver of an 80-kg diabetic
individual with modest fasting hyperglycemia adds approximately 30 g of
additional glucose to the systemic circulation. The increase in basal hepatic
glucose production (HGP) is closely correlated with the severity of fasting
hyperglycemia (see Fig. 7) [1,2,12,114], and this has been demonstrated in
numerous studies [116–118]. In conclusion, in type 2 diabetics with overt
fasting hyperglycemia (140 mg/dL, 7.8 mmol/L), an excessive rate of hepatic
glucose output is the major abnormality responsible for the elevated FPG
concentration.
Under postabsorptive conditions, the fasting plasma insulin concentration
in type 2 diabetics is 2- to 4-fold greater than in nondiabetic subjects [1,2].
Because hyperinsulinemia is a potent inhibitor of HGP, it is obvious that
hepatic resistance to the action of insulin must be present to explain the
excessive output of glucose by the liver. Because hyperglycemia per se exerts
a powerful suppressive action on HGP, the liver also must be resistant to
the inhibitory effect of hyperglycemia on hepatic glucose output, and this has
been well documented.
The dose response relationship between hepatic glucose production and
the plasma insulin concentration has been examined with the euglycemic
802 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Fig. 7. Summary of hepatic glucose production (HGP) in 77 normal-weight type 2 diabetic


subjects () with fasting plasma glucose concentrations ranging from 105 to greater than 300 mg/
dL. Seventy-two control subjects matched for age and weight are shown by filled circles. In the 33
diabetic subjects with fasting plasma glucose levels less than 140 mg/dL (shaded area), the mean
rate of hepatic glucose production was identical to that of control subjects. In diabetic subjects
with fasting plasma glucose concentrations greater than 140 mg/dL, there was a progressive rise in
hepatic glucose production that correlated closely (r, 0.847; P  0.001) with the fasting plasma
glucose concentration. (From DeFronzo RA, Ferrannini E, Simonson DC. Fasting hyperglycemia
in non-insulin-dependent diabetes mellitus: contributions of excessive hepatic glucose production
and impaired tissue glucose uptake. Metabolism 1989;38(4):387–95; with permission.)

insulin clamp technique and radioisotopic glucose (Fig. 8) [12]. The


following points deserve emphasis: (1) the dose-response curve relating
inhibition of HGP to the plasma insulin level is very steep, with a half-
maximal insulin concentration (ED50) of approximately 30 to 40 lU/mL; (2)

Fig. 8. Dose-response curve relating the plasma insulin concentration to the suppression of
hepatic glucose production in control (C) and type 2 diabetic () subjects with moderately
severe fasting hyperglycemia. * P  0.05; ** P  0.01 versus control subjects. (From Groop LC,
Bonadonna RC, DelPrato S, Ratheiser K, Zyck K, Ferrannini E, DeFronzo RA. Glucose and
free fatty acid metabolism in non-insulin-dependent diabetes mellitus. Evidence for multiple
sites of insulin resistance. J Clin Invest. 1989;84(1):205–13; with permission.)
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 803

in type 2 diabetics, the dose-response curve is shifted to the right, indicating


resistance to the inhibitory effect of insulin on hepatic glucose production.
Elevation of the plasma insulin concentration to the high physiologic range
(100 lU/mL), however, can overcome the hepatic insulin resistance and
cause a near normal suppression of HGP; and (3) the severity of the hepatic
insulin resistance is related to the level of glycemic control. In type 2
diabetics with mild fasting hyperglycemia, an increment in plasma insulin
concentration of 100 lU/mL causes a complete suppression of HPG;
however, in diabetic subjects with more severe fasting hyperglycemia, the
ability of the same plasma insulin concentration to suppress HGP is
impaired. These observations indicate that there is an acquired component
of hepatic insulin resistance, which becomes progressively worse as the
diabetic state decompensates over time.
Hepatic glucose production can be derived from either glycogenolysis or
gluconeogenesis [9]. Using the hepatic vein catheter technique, the uptake by
the liver of gluconeogenic precursors, especially lactate, has been shown to
increased in type 2 diabetic subjects [119], and this has been confirmed with
radioisotope turnover studies using radiolabeled lactate, alanine, glutamine,
and glycerol [120,121]. More recent studies using 13C-labeled magnetic
resonance imaging [122] and D2O [123] have confirmed that approximately
90% of the increase in HGP above baseline can be accounted for by
accelerated gluconeogenesis. A variety of mechanisms has been shown to
contribute to the increase in hepatic gluconeogenesis, including hypergluca-
gonemia, enhanced sensitivity to glucagon, increased circulating levels of
gluconeogenic precursors (lactate, alanine, glycerol), increased FFA oxida-
tion, and decreased sensitivity to insulin.
Because of the inaccessibility of the liver in humans, it has been difficult to
examine the role of key enzymes involved in the regulation of gluconeogenesis
(pyruvate carboxylase, phosphoenol-pyruvate carboxykinase), glycogenoly-
sis (glycogen phosphorylase), and net hepatic glucose output (glucokinase,
glucose-6-phosphatase). Considerable evidence from animal models of type 2
diabetes and some evidence in humans, however, has implicated increased
activity of phosphoenolpyruvate carboxykinase and glucose-6-phosphatase
in the accelerated rate of hepatic glucose production [123,124].
The kidney possesses all of the gluconeogenesis enzymes required to
produce glucose, and estimates of the renal contribution to total endogenous
glucose production have ranged from 5% to 20% [125,126]. These varying
estimates of the contribution of renal gluconeogenesis to total glucose pro-
duction are largely related to the methodology used to measure renal glucose
production [127]. One unconfirmed study suggests that the rate of renal
gluconeogenesis is increased in type 2 diabetics with fasting hyperglycemia
[128], but studies using the hepatic vein catheter technique have shown that all
of the increase in total body endogenous glucose production (measured with
[3-3H]glucose) in type 2 diabetics can be accounted for by increased hepatic
glucose output (measured by the hepatic vein catheter technique) [5].
804 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Peripheral (muscle) glucose uptake


Muscle is the major site of insulin-stimulated glucose disposal in humans
[1–3,5,129,130]. Under euglycemic conditions, studies using the euglycemic
insulin clamp in combination with femoral artery or vein catheterization
have shown that approximately 80% of total body glucose uptake occurs in
skeletal muscle. In response to a physiologic increase in plasma insulin
concentration (80–100 lU/mL), leg muscle glucose uptake increases
progressively in healthy subjects and reaches a plateau value of approxi-
mately 10 mg/kg leg weight/min (Fig. 9) [5]. In contrast, in lean type 2
diabetic subjects, the onset of insulin action is delayed by approximately 40
min, and the amount of glucose taken up by the leg is markedly blunted,
even though the insulin infusion is continued for an additional 60 min to
allow insulin to more fully express its biologic effects. During the last hour
of the insulin clamp study, the rate of glucose uptake was reduced by 50% in
the type 2 diabetic group. These results provide conclusive evidence that
muscle represents the primary site of insulin resistance during euglycemic
insulin clamp studies performed in type 2 diabetic subjects. Using the
forearm and leg catheterization techniques, a number of investigators have
demonstrated a decrease in insulin-stimulated muscle glucose uptake.
Studies using positron emission tomography have provided additional
support for the presence of severe muscle insulin resistance in type 2
diabetic subjects.

Fig. 9. Time course of change in leg glucose uptake in type 2 diabetic () and control (C)
subjects. In the postabsorptive state, glucose uptake in the diabetic group was significantly
greater than that in control subjects. The ability of insulin (euglycemic insulin clamp) to
stimulate leg glucose uptake, however, was reduced by 50% in the diabetic subjects. * P \ 0.05;
** P \ 0.01. (From DeFronzo RA, Gunnarsson R, Bjorkman O, Olsson M, Wahren J. Effects
of insulin on peripheral and splanchnic glucose metabolism in noninsulin-dependent (type II)
diabetes mellitus. J Clin Invest 1985;76(1):149–55; with permission.)
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 805

Splanchnic (hepatic) glucose uptake


Because of the difficulty in catheterizing the portal vein in humans, glucose
disposal by the liver has not been examined directly. Use of the hepatic vein
catheterization technique in combination with the euglycemic insulin clamp,
however, has allowed investigators to examine the contribution of the
splanchnic (liver plus gastrointestinal) tissues to overall glucose homeostasis
in lean type 2 diabetic subjects [3,5,129]. In the postabsorptive state, there is
a net release of glucose from the splanchnic area (ie, negative balance) in both
control and diabetic subjects, reflecting glucose production by the liver.
When insulin is infused while maintaining euglycemia, splanchnic glucose
output is promptly suppressed (reflecting inhibition of HGP) and, within 20
min, the net glucose balance across the splanchnic region decreases to zero
(i.e., no net uptake or release). After 2 hours of sustained hyperinsulinemia,
the splanchnic area manifests a small net uptake of glucose, approximately
0.5 mg/kg/min (i.e., positive balance), which is virtually identical to the rate
of splanchnic glucose uptake during in the basal state. These results indicate
that the splanchnic tissues, like the brain, are largely insensitive to insulin
with respect to the stimulation of glucose uptake. There were no differences
between diabetic and control subjects in the amount of glucose taken up by
the splanchnic tissues at any time during the insulin clamp study.
In summary, under conditions of euglycemic hyperinsulinemia, very little
infused glucose is taken up by the splanchnic (and therefore hepatic) tissues.
Because the difference in insulin-mediated total body glucose uptake
between type 2 diabetic and control subjects during the euglycemic insulin
clamp study was 2.5 mg/kg/min, it is obvious that a defect in splanchnic
(hepatic) glucose removal cannot account for the impairment in total body
glucose uptake following intravenous glucose/insulin administration; how-
ever, following glucose ingestion, the gastrointestinal route of glucose entry
and the resultant hyperglycemia conspire to enhance splanchnic (hepatic)
glucose uptake [4,9,16,17,131] and, under these conditions, diminished
hepatic glucose uptake has been shown to contribute to impaired glucose
tolerance in type 2 diabetes (see discussion below).

Summary: whole-body glucose use


A summary of insulin-mediated whole-body glucose metabolism during
the euglycemic insulin clamp is depicted in (Fig. 10). The height of each bar
represents the total amount of glucose taken up by all tissues in the body
during the insulin clamp in control and type 2 diabetic subjects. Net
splanchnic glucose uptake (quantitated by hepatic vein catheterization) is
similar in both groups and averaged 0.5 mg/kg/min. Adipose tissue glucose
uptake accounts for no more than 5% of total glucose disposal. Brain
glucose uptake, estimated to be 1.0–1.2 mg/kg/min in the postabsorptive
state, is unaffected by hyperinsulinemia. Muscle glucose uptake (extrapo-
lated from leg catheterization data) in control subjects accounts for
806 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Fig. 10. Summary of glucose metabolism during euglycemic insulin (100 lU/mL) clamp studies
performed in normal-weight type 2 diabetic and control subjects (see text for a more detailed
discussion). NIDD, non-insulin-dependent diabetes. (From DeFronzo RA. Pathogenesis of type
2 diabetes mellitus: metabolic and molecular implications for identifying diabetes genes.
Diabetes 1997;5:117–269; with permission.)

approximately 75%–80% of glucose uptake by the body. In type 2 diabetic


subjects, the largest part of the impairment in insulin-mediated glucose
uptake is explained by the defect in muscle glucose disposal. Numerous
studies have demonstrated that adipocytes from type 2 diabetics are
resistant to insulin, but because the total amount of glucose taken up by
fat cells during the insulin clamp is small [130], even if adipose tissue of type
2 diabetic subjects took up no glucose, it could, at best, explain only a small
fraction of the defect in whole-body glucose metabolism.

Glucose disposal during OGTT


During daily life, the normal route of glucose entry into the body is
through the gastrointestinal tract. To assess tissue glucose disposal
following glucose ingestion, Ferrannini, DeFronzo, and colleagues [131–
133] administered oral glucose combined with hepatic vein catheterization to
healthy control subjects to examine splanchnic glucose metabolism. The oral
glucose load and endogenous glucose pool were labeled with [1-14C]glucose
and [3-3H]glucose, respectively, to quantitate total body glucose disposal
(from tritiated glucose turnover) and endogenous HGP (difference between
the total rate of glucose appearance, measured with tritiated glucose, and
the rate of oral glucose appearance, measured with [1-14C]glucose).
During the 3.5 hours after glucose (68 g) ingestion: (1) 28% (19 g) of the
oral load was taken up by splanchnic tissues; (2) 72% (48 g) was removed by
nonsplanchnic tissues; (3) of the 48 g taken up by peripheral tissues, the
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 807

brain (an insulin-independent tissue) disposed of 22% (15 g or 1 mg/kg/min)


of the total glucose load; and (4) basal HGP declined by 53% [131]. Similar
percentages for splanchnic glucose uptake (24%–29%) and suppression of
HGP (50%–60%) in normal subjects have been reported by other inves-
tigators [8,134–136]. The contribution of skeletal muscle to the disposal of
an oral glucose load has been reported to vary from a low of 26% [135] to
a high of 56% [136], with a mean of 45% [8,131,134–136]. These results
demonstrate a number of important differences between oral and in-
travenous glucose administration. After glucose ingestion, HGP is less
completely suppressed, most likely because of activation of local sympa-
thetic nerves that innervate the liver, peripheral tissue (primarily muscle)
glucose uptake is quantitatively less important [3], and splanchnic glucose
uptake quantitatively is much more important.
When an oral glucose is given to type 2 diabetic individuals, marked
glucose intolerance is observed, and this results from decreased tissue
(muscle) glucose uptake and impaired suppression of HGP. The uptake of
glucose by the splanchnic tissues is similar in diabetic and control groups.
Impaired suppression of HGP accounts for approximately one third of the
defect in total-body glucose homeostasis, whereas reduced peripheral
(muscle) glucose uptake accounts for the remaining two thirds. It should
be noted that, even though the total amount of glucose taken up by the
splanchnic region in type 2 diabetics is ‘‘normal,’’ splanchnic glucose
metabolism is quite abnormal. Because hyperglycemia per se enhances
splanchnic (hepatic) glucose uptake in proportion to the increase in plasma
glucose concentration, and because the rise in plasma glucose concentration
in diabetics is excessive, the splanchnic glucose clearance (splanchnic glucose
uptake ‚ plasma glucose concentration) following glucose ingestion is
markedly reduced in type 2 diabetic subjects. Using a combined insulin
clamp/OGTT technique, an impairment in glucose uptake by the splanchnic
tissues in type 2 diabetics has been demonstrated directly [137].
In summary, following glucose ingestion both impaired suppression of
HGP and decreased muscle glucose uptake are responsible for the glucose
intolerance of type 2 diabetes. The efficiency of the splanchnic (hepatic)
tissues to take up glucose (as reflected by the splanchnic glucose clearance)
also is impaired in type 2 diabetic individuals.

Summary: insulin resistance in type 2 diabetes


Insulin resistance in muscle and liver is a characteristic feature of type 2
diabetes mellitus. In the basal state, the hepatic insulin resistance is
manifested by overproduction of glucose despite fasting hyperinsulinemia,
and the increased rate of hepatic glucose output is the primary determinant of
the elevated FPG concentration in type 2 diabetic individuals. Although
muscle glucose uptake in the postabsorptive state is increased when viewed in
absolute terms, the efficiency with which glucose is taken up (ie, the glucose
808 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

clearance) by muscle is diminished. During insulin-stimulated conditions,


both decreased muscle glucose uptake and impaired suppression of HGP
contribute to the glucose intolerance.

Dynamic interaction between insulin sensitivity and insulin secretion in type 2


diabetes
Insulin resistance is present in approximately 25% of the adult population
[138–140]. The majority of these individuals, however, have normal glucose
tolerance because the pancreatic beta cells are able to read the severity of
insulin resistance and appropriately augment their insulin secretory rate. This
dynamic interaction between insulin sensitivity and insulin secretion is
demonstrated by results obtained in healthy, lean, young normal-glucose-
tolerant women who received a euglycemic insulin clamp (1 mU/kg/min) and
were stratified into quartiles based on the rate of insulin-mediated glucose
disposal (see Fig. 2A) [141]. Women in the lowest quartile were as insulin
resistant as type 2 diabetic individuals. Insulin secretion was measured on
a separate day with a þ125-mg/dL hyperglycemic clamp (see Fig. 2B). Women
who were the most insulin resistant (quartile 1) had the highest fasting plasma
insulin concentrations and highest early and late-phase plasma insulin res-
ponses (see Fig. 2B). Conversely, women who were the most insulin sensitive
(quartile 4) had the lowest plasma insulin response. A very strong positive
correlation (r, 0.79, P 0.001) was observed between the severity of insulin
resistance and the insulin secretory response. Similar results relating the
plasma insulin response and the severity of insulin resistance have been
reported in normal-glucose-tolerant subjects with the minimal model tech-
nique and the insulin suppression/OGTT.
The dynamic interaction between insulin secretion and insulin sen-
sitivity in type 2 diabetic individuals has been the subject of intensive
investigation. DeFronzo [2] studied lean (ideal body weight  120%) and
obese (ideal body weight  125%) subjects with varying degrees of glucose
tolerance: group I, obese subjects (n = 24) with normal glucose tolerance;
group II, obese subjects (n = 23) with impaired glucose tolerance; group III,
obese subjects (n = 35) with overt diabetes, who were subdivided into those
with a hyperinsulinemic response and those with a hypoinsulinemic response
during an OGTT; group IV, normal weight type 2 diabetics (n = 26); and
group V, normal weight subjects (n = 25) with normal glucose tolerance (see
Fig. 4). All subjects received a euglycemic insulin (100 lU/mL) clamp to
quantitate whole-body insulin sensitivity and an OGTT to provide a measure
of glucose tolerance and insulin secretion. The insulin clamp was performed
with indirect calorimetry to quantitate rates of glucose oxidation and
nonoxidative glucose disposal, which primarily reflects glycogen synthesis.
In normal weight type 2 diabetic subjects, insulin-mediated whole-body
glucose uptake was reduced by 40%–50%, and the impairment in insulin
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 809

action resulted from defects in both glucose oxidation and glycogen


synthesis. It is particularly noteworthy that obese, normal glucose tolerant
individuals were as insulin resistant as the lean normal-weight diabetic
subjects (see Fig. 4) and that the insulin resistance resulted from defects in
both glucose oxidation and glycogen synthesis. Thus, from the metabolic
standpoint, obesity and type 2 diabetes closely resemble each other. Similar
results concerning reduced whole-body insulin sensitivity in obese and type
2 diabetic individuals have been reported by other investigators [142,143].
Despite nearly identical degrees of insulin resistance, the normal-weight
diabetic subjects (see Fig. 4) had fasting hyperglycemia and marked glucose
intolerance, whereas the obese nondiabetic individuals had normal oral
glucose tolerance. This apparent paradox is explained by the plasma insulin
response during the OGTT (see Fig. 4). Compared with control subjects, the
obese group secreted more than twice as much insulin, and this hyper-
insulinemic response was sufficient to offset the insulin resistance. In
contrast, the pancreas of normal-weight diabetic subjects, when faced with
the same challenge, was unable to augment its secretion of insulin
sufficiently to compensate for the insulin resistance. This imbalance between
insulin secretion and the severity of insulin resistance in liver and muscle
resulted in a frankly diabetic state, with fasting hyperglycemia and marked
glucose intolerance.
The coexistence of obesity and diabetes in the same individual resulted in
a severity of insulin resistance that was only slightly greater than that in
either the normal-weight diabetic or nondiabetic obese groups (see Fig. 4).
Although hyperinsulinemic and hypoinuslinemic obese diabetic subjects
were equally insulin resistant, the glucose intolerance was much worse in the
hypoinsulinemic group, and this was related entirely to the presence of
severe insulin deficiency (see Fig. 4).
The interaction between insulin secretion and insulin resistance in lean,
obese, and diabetic groups can be summarized as follows. In the obese
nondiabetic subjects, tissue sensitivity to insulin (Fig. 4, top) is markedly
reduced, but glucose tolerance (bottom) remains normal because the
pancreas is able to augment its secretion of insulin (top) to offset the defect
in insulin action. The development of IGT, the mean plasma glucose
concentration during the OGTT, increases only minimally because the
pancreas is able to further augment its secretion of insulin to counteract the
deterioration in insulin sensitivity. Progression from IGT to overt diabetes is
signaled by a decrease in insulin secretion without any worsening of insulin
resistance (see Fig. 4). The obese diabetic has tipped over the top of
Starling’s curve of the pancreas and is now on the descending portion (see
Figs. 3 and 4). Although, compared with nondiabetic control subjects, the
plasma insulin response is increased, the hyperinsulinemia is insufficient to
offset the severity of insulin resistance. In the normal-weight diabetic group,
there is a further decline in glucose tolerance, which results from a greater
impairment in insulin secretion without any additional deterioration in
810 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

insulin sensitivity. Last, the obese diabetic group with a low insulin response
manifests the greatest glucose intolerance owing to the presence of marked
insulin deficiency without further worsening of insulin sensitivity (see Fig. 4).
The natural history of type 2 diabetes described above (see Fig. 4) is
consistent with that described by other investigators in humans and
monkeys [1,2,20,22–27,29,33–35,97–99,102,144–146]. In lean subjects span-
ning a wide range of glucose tolerance, Reaven et al [97] demonstrated that
the progression from normal glucose tolerance to IGT was signaled by the
development of severe insulin resistance, which was largely counterbalanced
by increased insulin secretion. Progression from IGT to type 2 diabetes was
associated with a marked decline in insulin secretion with no (or only slight)
further deterioration in tissue sensitivity to insulin (Fig. 11). A similar
sequence of events has been documented prospectively in Pima Indians,
Pacific Islanders, and rhesus monkeys.
In summary, insulin resistance is an early and characteristic feature of the
natural history of type 2 diabetes in high-risk populations. Overt diabetes
develops only when the beta cells are unable to appropriately augment their
secretion of insulin to compensate for the defect in insulin action. It should
be recognized, however, that there are well-described type 2 diabetic
populations in whom insulin sensitivity is normal at the onset of diabetes,
whereas insulin secretion is severely impaired. This insulinopenic variety of
type 2 diabetes appears to be more common in African Americans, elderly
subjects, and lean whites. In this latter group, it is important to exclude type
1 diabetes, because approximately 10% of white individuals with older onset
diabetes are islet cell antibody, or glutamic acid decarboxylase, positive.

Fig. 11. Insulin-mediated glucose clearance (measured with the insulin suppression test) and the
plasma insulin response (measured with an OGTT) in controls (top), in subjects with IGT
(bottom), and in type 2 diabetic individuals (top) with varying severity of glucose intolerance (see
text for a more detailed discussion). (Data from Reaven GM, Hollenbeck CB, Chen YDI.
Relationship between glucose tolerance, insulin secretion, and insulin action in non-obese
individuals with varying degrees of glucose tolerance. Diabetologia 1989;32:52–5.)
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 811

Role of the adipocyte in the pathogenesis of type 2 diabetes mellitus: the


harmonious quartet
The majority ( 80%–90%) of type 2 diabetics in the United States are
overweight or obese [147]. Both lean and especially obese type 2 diabetics
are characterized by day-long elevation in plasma free fatty-acid concen-
tration, which fails to suppress normally following ingestion of a mixed meal
or oral glucose load [30]. FFA are stored as triglycerides in adipocytes and
serve as an essential energy source during fasting conditions. Insulin is
a potent antilipolytic hormone and restrains the release of FFA from the
adipocyte by inhibiting the enzyme hormone-sensitive lipase [11,12]. The fat
cells of type 2 diabetics (and nondiabetic obese individuals) are markedly
resistant to the inhibitory effect of insulin on lipolysis. In the postabsorptive
state, the rate of lipolysis (as reflected by impaired suppression of
radioactive palmitate turnover) is increased despite plasma insulin levels
that are 2- to 4-fold elevated. Moreover, the ability of exogenous insulin to
inhibit the elevated basal rate of lipolysis and to reduce the plasma FFA
concentration is markedly impaired. Many studies have shown that
chronically elevated plasma FFA concentrations cause insulin resistance
in muscle and liver and impair insulin secretion (Fig. 12) [1,2,11,14,58,148–
152]. Thus, increased plasma FFA levels can cause or aggravate the three
major pathogenic disturbances that are responsible for impaired glucose
homeostasis in type 2 diabetic individuals, and the time has arrived for the
‘‘triumvirate’’ (muscle, liver, beta cell) to be joined by the ‘‘fourth
musketeer’’ [152] to form the ‘‘harmonious quartet’’ (Fig. 13) [11]. In
addition to the FFA that circulate in plasma in increased amounts, type 2
diabetic and obese nondiabetic individuals have increased stores of
triglycerides in muscle and liver, and the increased fat content correlates
closely with the presence of insulin resistance in these tissues [153–156].
Triglycerides in liver and muscle are in a state of constant turnover, and the
intracellular metabolites of triglycerides and FFA (ie, fatty acyl-CoA,
ceramides, and diacylglycerol) have been shown to impair insulin action

Fig. 12. Etiology of type 2 diabetes mellitus (T2DM). The deleterious effect of chronically
elevated plasma FFA concentrations on basal or insulin-suppressed rate of hepatic glucose
production, insulin-stimulated glucose uptake in muscle, and glucose-stimulated insulin
secretion.
812 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

Fig. 13. Harmonious quartet. Insulin resistance in adipocytes, muscle, and liver in combination
with impaired insulin secretion by the pancreatic beta cells, represent the four major organ
system abnormalities that play a central role in the pathogenesis of type 2 diabetes mellitus.

in both liver and muscle [11,157–159]. Evidence also has accumulated to


implicate lipotoxicity as an important cause of beta-cell dysfunction (see
earlier discussion) [11,58,150,151]. The sequence of events whereby elevated
plasma FFA and increased lipid deposition in tissues cause insulin resistance
and promote beta-cell failure has been referred to as ‘‘lipotoxicity,’’ and
several recent in depth reviews of this subject have been published [11,58].

Cellular mechanisms of insulin resistance


The cellular events through which insulin initiates its stimulatory effect
on glucose metabolism start with binding of the hormone to specific
receptors that are present on the cell surface of all insulin target tissues
[2,160–162]. After insulin has bound to and activated its receptor, ‘‘second
messengers’’ are generated, and these second messengers activate a cascade
of phosphorylation-dephosphorylation reactions that eventually result in
the stimulation of intracellular glucose metabolism. The first step in glucose
use involves activation of the glucose transport system, leading to glucose
influx into insulin target tissues, primarily muscle. The free glucose, which
has entered the cell, subsequently is metabolized by a series of enzymatic
steps that are under the control of insulin. Of these, the most important are
glucose phosphorylation (catalyzed by hexokinase), glycogen synthase
(which controls glycogen synthesis), and phosphofructokinase (PFK) and
pyruvate dehydrogenase (PDH) (which regulate glycolysis and glucose
oxidation, respectively).

Insulin receptor/insulin receptor tyrosine kinase


The insulin receptor is a glycoprotein that consists of two a-subunits and
two b-subunits linked by disulfide bonds (Fig. 14) [2,160–162]. The two a-
subunits of the insulin receptor are entirely extracellular and contain the
insulin-binding domain. The b-subunits have an extracellular domain,
a transmembrane domain, and an intracellular domain that expresses
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 813

Fig. 14. Insulin transduction system. Insulin receptor and the cascade of intracellular signaling
molecules that have been implicated in insulin action (see text for a more detailed discussion).

insulin-stimulated kinase activity directed toward its own tyrosine residues.


Phosphorylation of the b-subunit, with subsequent activation of insulin
receptor tyrosine kinase, represents the first step in the action of insulin on
glucose metabolism. Mutagenesis of any of the three major phosphorylation
sites (at residues 1158, 1163, and 1162) impairs insulin receptor kinase
activity, leading to a decrease in the metabolic and growth-promoting effects
of insulin [163,164].

Insulin receptor signal transduction


Following its activation, insulin receptor tyrosine kinase phosphorylates
specific intracellular proteins, of which at least nine have been identified
[160,165]. In muscle insulin-receptor substrate (IRS)-1 serves as the major
docking protein that interacts with the insulin receptor tyrosine kinase and
undergoes tyrosine phosphorylation in regions containing specific amino
acid sequence motifs that, when phosphorylated, serve as recognition sites
for proteins containing src-homology 2 (SH2) domains. Mutation of these
specific tyrosines severely impairs the ability of insulin to stimulate muscle
glycogen synthesis, glucose oxidation, and other acute metabolic- and
growth-promoting effects of insulin [164]. In liver, IRS-2 serves as the
primary docking protein that undergoes tyrosine phosphorylation and
mediates the effect of insulin on hepatic glucose production, gluconeogen-
esis, and glycogen formation [166].
In muscle, the phosphorylated tyrosine residues of IRS-1 mediate an
association with the 85-kDa regulatory subunit of phosphatidylinositol
3-kinase (PI3K), leading to activation of the enzyme (see Fig. 14) [160–
162,165,167]. PI3K is composed of an 85-kDa regulatory subunit and a 110-
kDa catalytic subunit. The latter catalyzes the 39 phosphorylation of PI
4-phosphate and PI 4,5-diphosphate, resulting in the stimulation of glucose
814 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

transport. Activation of PI3K by phosphorylated IRS-1 also leads to


activation of glycogen synthase through a process that involves activation of
protein kinase B/Akt and subsequent inhibition of kinases, such as glycogen
synthase kinase-3, and activation of protein phosphatase 1 (PP1) (also called
glycogen synthase phosphatase, see later discussion). Inhibitors of PI3K
impair glucose transport and block the activation of glycogen synthase and
hexokinase (HK)-II expression [160–162,165,167–169]. The action of insulin
to increase protein synthesis and inhibit protein degradation also is
mediated by PI3K.
Other proteins with SH2 domains, including the adapter protein Grb2
and Shc, also interact with IRS-1 and become phosphorylated following
exposure to insulin [160–162,165]. Grb2 and Shc link IRS-1/IRS-2 to the
mitogen-activated protein kinase (MAPK)-signaling pathway (see Fig. 14),
which plays an important role in the generation of transcription factors and
promotes cell growth, proliferation, and differentiation [160,165]. Inhibition
of the MAPK kinase pathway prevents the stimulation of cell growth by
insulin but has no effect on the metabolic actions of the hormone [170].
Under anabolic conditions, insulin augments glycogen synthesis by
simultaneously activating glycogen synthase and inhibiting glycogen phos-
phorylase [171,172]. The effect of insulin is mediated through the PI3K
pathway, which inactivates kinases such as glycogen synthase kinase-3 and
activates phosphatases, particularly PP1. PP1 is believed to be the primary
regulator of glycogen metabolism. In skeletal muscle, PP1 associates with
a specific glycogen-binding regulatory subunit, causing dephosphorylation
(activation) of glycogen synthase. PP1 also phosphorylates (inactivates)
glycogen phosphorylase. The precise steps that link insulin receptor tyrosine
kinase/PI3K activation to stimulation of PP1 have yet to be defined. Studies
[160,173] have demonstrated convincingly that inhibitors of PI3K inhibit
glycogen synthase activity and abolish glycogen synthesis.

Insulin receptor signal transduction defects in type 2 diabetes


Insulin receptor number and affinity
Both receptor and postreceptor defects have been shown to contribute to
insulin resistance in individuals with type 2 diabetes mellitus. Some studies
have demonstrated a modest 20%–30% reduction in insulin binding to
monocytes and adipocytes from type 2 diabetic patients, but this has not
been a consistent finding [1,174–177]. The decrease in insulin binding is
caused by a reduction in the number of insulin receptors without change
in insulin receptor affinity. Some caution, however, should be used in
interpreting these studies because muscle and liver not adipocytes are the
major tissues responsible for the regulation of glucose homeostasis in vivo,
and insulin binding to solubilized receptors obtained from skeletal muscle
and liver has been shown to be normal in obese and lean diabetic individuals
[175,176,178]. Moreover, a decrease in insulin receptor number cannot be
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 815

demonstrated in over half of type 2 diabetic subjects, and it has been difficult
to demonstrate a correlation between reduced insulin binding and the
severity of insulin resistance [179–181]. A variety of defects in insulin
receptor internalization and processing have been described in syndromes of
severe insulin resistance and diabetes. The insulin receptor gene, however,
has been sequenced in a large number of type 2 diabetic patients from
diverse ethnic populations and, with very rare exceptions, physiologically
significant mutations in the insulin receptor gene have not been observed
[182,183]. This excludes a structural gene abnormality in the insulin receptor
as a cause of common type 2 diabetes mellitus.

Insulin receptor tyrosine kinase activity


Insulin receptor tyrosine kinase activity has been examined in skeletal
muscle, adipocytes, and hepatocytes from normal-weight and obese diabetic
subjects. Most [1,175,176,179,184,185] but not all [178] investigators have
found a reduction in tyrosine kinase activity (Fig. 15) that cannot be explained
by alterations in insulin receptor number or insulin receptor binding affinity.
Restoration of normoglycemia by weight loss, however, has been shown to
correct the defect in insulin receptor tyrosine kinase activity [186], suggesting
that the defect in tyrosine kinase is acquired secondary to some combination of
hyperglycemia, distributed intracellular glucose metabolism, hyperinsuline-
mia, and insulin resistance, all of which improved after weight loss. Exposure
of cultured fibroblasts to high glucose concentration also has been shown to

Fig. 15. Insulin signaling cascade in T2DM. Effect of insulin on insulin receptor (top) and IRS-
1 tyrosine phosphorylation (bottom) and the association of IRS-1 with the p85 regulatory
subunit of PI3K and PI3K activity in muscle from T2DM and control (CON) subjects. Data are
expressed as percentages of the mean insulin-stimulated values in the control groups. Open bars,
basal state; filled bars, insulin-stimulated state; * P  0.05, T2DM versus CON.
816 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

inhibit insulin receptor tyrosine kinase activity [187]. Because insulin receptor
tyrosine kinase activity assays are performed in vitro, the results of these
assays could provide misleading information with regard to insulin receptor
function in vivo. To circumvent this problem, investigators have used the
euglycemic hyperinsulinemic clamp with muscle biopsies and anti-phospho-
tyrosine immunoblot analysis to provide a ‘‘snap shot’’ of the insulin-
stimulated tyrosine phosphorylation state of the receptor in vivo [185]. In
insulin-resistant obese nondiabetic and type 2 diabetic subjects, a substantial
decrease in insulin receptor tyrosine phosphorylation has been demonstrated;
however, when insulin-stimulated insulin receptor tyrosine phosphorylation
was examined in normal-glucose-tolerant insulin-resistant individuals (off-
spring of two diabetic parents) at high risk for developing type 2 diabetes,
a normal increase in tyrosine phosphorylation of the insulin receptor was
observed [188]. These findings are consistent with the concept that impaired
insulin receptor tyrosine kinase activity in type 2 diabetic patients is acquired
secondary to hyperglycemia or some other metabolic disturbance.

Insulin-signaling (IRS-1 and PI3K) defects


In insulin-resistant obese nondiabetic subjects, the ability of insulin to
activate insulin receptor and IRS-1 tyrosine phosphorylation in muscle is
modestly reduced, whereas in type 2 diabetics insulin-stimulated insulin
receptor and IRS-1 tyrosine phosphorylation are severely impaired (see
Fig. 15) [185]. Association of the p85 subunit of PI3K with IRS-1 and
activation of PI3K also are greatly attenuated in obese nondiabetic and type
2 diabetic subjects compared with lean healthy controls (see Fig. 15)
[185,189,190]. The decrease in insulin-stimulated association of the p85
regulatory subunit of PI3K with IRS-1 is closely correlated with the
reduction in insulin-stimulated muscle glycogen synthase activity and in
vivo insulin-stimulated glucose disposal [185]. Impaired regulation of PI3K
gene expression by insulin also has been demonstrated in skeletal muscle
and adipose tissue of type 2 diabetic subjects [191]. In animal models of
diabetes, an 80%–90% decrease in insulin-stimulated IRS-1 phosphoryla-
tion and PI3K activity has been reported [192].
In the insulin-resistant, normal glucose tolerant offspring of two type 2
diabetic parents, IRS-1 tyrosine phosphorylation and the association of p85
protein/PI3K activity with IRS-1 are markedly decreased despite normal
tyrosine phosphorylation of the insulin receptor; these insulin signaling
defects are correlated closely with the severity of insulin resistance measured
with the euglycemic insulin clamp technique [188]. In summary, impaired
association of PI3K with IRS-1 and its subsequent activation are charac-
teristic abnormalities in type 2 diabetics, and these defects are correlated
closely with in vivo muscle insulin resistance. A common mutation in the
IRS-1 gene (Gly-972-Arg) has been associated with type 2 diabetes, insulin
resistance, and obesity, but the physiologic significance of this mutation
remains to be established [193].
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 817

Insulin resistance of the PI3K signaling pathway contrasts with an intact


stimulation of the MAPK pathway by insulin in insulin-resistant type 2
diabetic and obese nondiabetics individuals [185,189]. Physiologic
hyperinsulinemia increases mitogen-activated protein kinase/extracellular
signal-regulated kinase 1 activity (MEK-1) and extracellular signal-regulated
kinase1/2 phosphorylation activity (ERK) similarly in lean healthy subjects,
insulin-resistant obese nondiabetic, and type 2 diabetic patients. Intact
stimulation of the MAPK pathway by insulin in the presence of insulin
resistance in the PI3K pathway may play an important role in the develop-
ment of atherosclerosis [185]. If the metabolic (PI3K) pathway is impaired,
plasma glucose levels rise, resulting in increased insulin secretion and hyper-
insulinemia. Because insulin receptor function is normal or only modestly
impaired, especially early in the natural history of type 2 diabetes, this leads to
excessive stimulation of the MAPK (mitogenic) pathway in vascular tissues,
with resultant proliferation of vascular smooth muscle cells, increased
collagen formation, and increased production of growth factors and
inflammatory cytokines [194,195].

Glucose transport
Activation of the insulin signal transduction system in insulin target
tissues stimulates glucose transport through a mechanism that involves
translocation of a large intracellular pool of glucose transporters (associated
with low-density microsomes) to the plasma membrane and their subsequent
activation after insertion into the cell membrane [196,197]. There are five
major, different facilitative glucose transporters (GLUT) with distinctive
tissue distributions (Table 1) [198,199]. GLUT4, the insulin regulatable
transporter, is found in insulin-sensitive tissues (muscle and adipocytes), has
a Km of approximately 5 mmol/L, which is close to that of the plasma
glucose concentration, and is associated with HK-II [198,199]. In adipocytes
and muscle, GLUT4 concentration in the plasma membrane increases
markedly after exposure to insulin, and this increase is associated with
Table 1
Classification of glucose transport and HK activity according to their tissue distribution and
functional regulation
Organ Glucose transporter Hexokinase computer Classification
Brain GLUT1 HK-I Glucose dependent
Erythrocyte GLUT1 HK-I Glucose dependent
Adipocyte GLUT4 HK-II Insulin dependent
Muscle GLUT4 HK-II Insulin dependent
Liver GLUT2 HK-IVL Glucose sensor
Glucokinase beta cell GLUT2 HK-IVB (glucokinase) Glucose sensor
Gut GLUT3-symporter — Sodium dependent
Kidney GLUT3-symporter — Sodium dependent
Data from DeFronzo RA. Pathogenesis of type 2 diabetes mellitus: metablic and molecular
implications for identifying diabetes genes. Diabetes 1997;5:177–269.
818 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

a reciprocal decline in the intracellular GLUT4 pool. GLUT1 is the


predominant glucose transporter in the insulin-independent tissues (brain
and erythrocytes) but also is found in muscle and adipocytes. GLUT1 is
located primarily in the plasma membrane where its concentration is
unchanged following exposure to insulin. It has a low Km (1 mmol/L)
and is well suited for its function, which is to mediate basal glucose uptake.
It is found in association with HK-I [200]. GLUT2 is the predominant
transporter in liver and pancreatic beta cells where it is found in association
with a specific hexokinase, HK-IV or glucokinase [201]. GLUT2 has a very
high Km (15–20 mmol/L), which allows the glucose concentration in cells
expressing this transporter to increase in direct proportion to the increase in
plasma glucose concentration. This unique characteristic allows these cells
to function as glucose sensors.
In adipocytes and muscle of type 2 diabetic patients, glucose transport
activity is severely impaired [179,196,197,202–204]. In adipocytes from
human and rodent models of type 2 diabetes, GLUT4 mRNA and protein
content are markedly reduced, and the ability of insulin to elicit a normal
translocation response and to activate the GLUT4 transporter after insertion
into the cell membrane is decreased. In contrast to the adipocytes, muscle
tissue from lean and obese type 2 diabetic subjects exhibits normal or
increased levels of GLUT4 mRNA expression and normal levels of GLUT4
protein, thus demonstrating that transcriptional and translational regulation
of GLUT4 is not impaired [205,206]. In contrast to the normal expression of
GLUT4 protein and mRNA in muscle of type 2 diabetic subjects, every study
that has examined adipose tissue has reported reduced basal and insulin-
stimulated GLUT4 mRNA levels and decreased GLUT4 transporter number
in all subcellular fractions. These observations demonstrate that GLUT4
expression in humans is subject to tissue-specific regulation. Using a novel
triple-tracer technique, the in vivo dose-response curve for the action of
insulin on glucose transport in forearm skeletal muscle has been examined in
type 2 diabetic subjects, and insulin-stimulated inward muscle glucose
transport has been shown to be severely impaired [207,208]. Impaired in
vivo muscle glucose transport in type 2 diabetics also has been demonstrated
using magnetic resonance imaging [209] and positron emission tomography
[210]. Because the number of GLUT4 transporters in the muscle of diabetic
subjects is normal, impaired GLUT4 translocation and decreased intrinsic
activity of the glucose transporter must be responsible for the defect in muscle
glucose transport. Large populations of type 2 diabetics have been screened
for mutations in the GLUT4 gene [211]. Such mutations are very uncommon
and, when detected, have been of questionable physiologic significance.

Glucose phosphorylation
Glucose phosphorylation and glucose transport are tightly coupled
phenomena [212]. Hexokinase isoenzymes (HK-I–IV) catalyze the first
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 819

committed step of glucose metabolism, the intracellular conversion of free


glucose to glucose-6-phosphate (Glu-6-P) (see Table 1) [198–200,213]. HK-I,
HK-II, and HK-III are single-chain peptides that have a very high affinity
for glucose and demonstrate product inhibition by Glu-6-P. HK-IV, also
called glucokinase, has a lower affinity for glucose and is not inhibited by
Glu-6-P. Glucokinase (HK-IVB) represents the glucose sensor in the beta
cell, whereas HK-IVL plays a central role in the regulation of hepatic
glucose metabolism.
In human skeletal muscle, HK-II transcription is regulated by insulin,
whereas HK-I mRNA and protein levels are not affected by insulin [214–
216]. In response to physiologic euglycemic hyperinsulinemia of 2 to 4
hours’ duration, HK-II cytosolic activity, protein content, and mRNA levels
increase by 50% to 200% in healthy nondiabetic subjects, and this is
associated with the translocation of hexokinase II from the cytosol to the
mitochondria. In forearm muscle, insulin-stimulated glucose transport
(measured with the triple-tracer technique) is markedly impaired in lean
type 2 diabetics [207,208], but the rate of intracellular glucose phosphory-
lation is impaired to an even greater extent, resulting in an increase in the
free glucose concentration within the intracellular space that is accessible to
glucose. These observations indicate that in type 2 diabetic individuals,
although both glucose transport and glucose phosphorylation are severely
resistant to the action of insulin, impaired glucose phosphorylation (HK-II)
appears to be the rate-limiting step for insulin action. Studies using 31P
nuclear magnetic resonance in combination with [1-14C]glucose also have
demonstrated that both insulin-stimulated muscle glucose transport and
glucose phosphorylation are impaired in type 2 diabetic subjects, but the
defect in transport exceeds the defect in phosphorylation [209]. Because of
methodologic differences, the results of the triple-tracer technique [207,208]
and magnetic resonance imaging [209] studies cannot be reconciled at
present. Nonetheless, these studies are consistent in demonstrating that
abnormalities in both muscle glucose phosphorylation and glucose transport
are well established early in the natural history of type 2 diabetes and cannot
be explained by glucose toxicity.
In healthy nondiabetic subjects, a physiologic increase in the plasma
insulin concentration for as little as 2 to 4 hours increases muscle HK-II
activity, gene transcription, and translation [214]. In lean type 2 diabetics, the
ability of insulin to augment HK-II activity and mRNA levels are markedly
reduced compared with controls [215]. Decreased basal muscle HK-II activity
and mRNA levels and impaired insulin-stimulated HK-II activity in type 2
diabetic subjects have been reported by other investigators [216,217]. A
decrease in insulin-stimulated muscle HK-II activity also has been described
in subjects with IGT [218]. Several groups have looked for point mutations in
the HK-II gene in individuals with type 2 diabetes, and, although several
nucleotide substitutions have been found, none are close to the glucose and
ATP binding sites and none have been associated with insulin resistance
820 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[218–220]. Thus, an abnormality in the HK-II gene is unlikely to explain the


inherited insulin resistance in common variety type 2 diabetes mellitus.

Glycogen synthesis
Following phosphorylation by hexokinase II, glucose either can be
converted to glycogen or enter the glycolytic pathway. Of the glucose that
enters the glycolytic pathway, approximately 90% is oxidized, and the
remaining 10% is released as lactate. At low physiologic plasma insulin
concentrations, glycogen synthesis and glucose oxidation contribute equally
to glucose disposal; however, with increasing plasma insulin concentrations,
glycogen synthesis predominates [1,2,221]. Impaired insulin-stimulated
glycogen synthesis is a characteristic finding in all insulin-resistant states,
including obesity, IGT, diabetes, and diabesity in all ethnic groups, and
accounts for the majority of the defect in insulin-mediated whole-body
glucose disposal [1,2,12,98,210,222–224]. Impaired glycogen synthesis also
has been documented in the normal-glucose tolerant offspring of two
diabetic parents, in the first-degree relatives of type 2 diabetic individuals,
and in the normoglycemic twin of a monozygotic twin pair in which the
other twin has type 2 diabetes [65,98,225].
Glycogen synthase is the key insulin-regulated enzyme that controls the
rate of muscle glycogen synthesis [171,173,216, 226–228]. Insulin activates
glycogen synthase by stimulating a cascade of phosphorylation-dephos-
phorylation reactions (see above discussion of insulin receptor signal
transduction), which ultimately lead to the activation of PP1 (also called
glycogen synthase phosphatase). The regulatory subunit of PP1 has two
serine phosphorylation sites, called site 1 and site 2. Phosphorylation of site
2 by cAMP-dependent protein kinase inactivates PP1, whereas phosphor-
ylation of site 1 by insulin activates PP1, leading to the stimulation of
glycogen synthase. Phosphorylation of site 1 of PP1 by insulin in muscle is
catalyzed by insulin-stimulated protein kinase (ISPK)-1. Because of their
central role in muscle glycogen formation, the three enzymes, glycogen
synthase, PP1, and ISPK-1, have been extensively studied in individuals
with type 2 diabetes.
Glycogen synthase exists in an active (dephosphorylated) and an inactive
(phosphorylated) form [171–173]. Under basal conditions, total glycogen
synthase activity in type 2 diabetic subjects is reduced, and the ability of
insulin to activate glycogen synthase is severely impaired [185,229–231]. The
ability of insulin to stimulate glycogen synthase also is diminished in the
normal glucose-tolerant, insulin-resistant relatives of type 2 diabetic
individuals [232]. In insulin-resistant nondiabetic and diabetic Pima Indians,
activation of muscle PP1 (glycogen synthase phosphatase) by insulin is
severely reduced [233]. Because PP1 dephosphorylates glycogen synthase,
leading to its activation, a defect in PP1 appears to play an important role in
the muscle insulin resistance of type 2 diabetes mellitus.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 821

The effect of insulin on glycogen synthase gene transcription and trans-


lation in vivo has been studied extensively. Most studies have demonstrated
that insulin does not increase glycogen synthase mRNA or protein expression
in human muscle [214,234,235]. Glycogen synthase mRNA and protein
levels, however, are decreased in muscle of type 2 diabetic patients, partly
explaining the decreased glycogen synthase activity [235,236]. The major
abnormality in glycogen synthase regulation in type 2 diabetes is its lack of
dephosphorylation and activation by insulin, as a result of insulin receptor
signaling abnormalities (see previous discussion).
The glycogen synthase gene has been the subject of intensive investigation,
and DNA sequencing has revealed either no mutations or rare nucleotide
substitutions that cannot explain the defect in insulin-stimulated glycogen
synthase activity [237–239]. The genes encoding the catalytic subunits of PP1
and ISPK-1 have been examined in Pima Indians and Danes with type 2
diabetes [240,241]. Several silent nucleotide substitutions were found in the
PP1 and ISPK-1 genes in the Danish population, but the mRNA levels of
both genes were normal in skeletal muscle. No structural gene abnormalities
in the catalytic subunit of PP1 were detected in Pima Indians. Thus, neither
mutations in the PP1 and ISPK-1 genes nor abnormalities in their translation
can explain the impaired enzymatic activities of glycogen synthase and PP1
that have been observed in vivo. Similarly, there is no evidence that an
alteration in glycogen phosphorylase plays any role in the abnormality in
glycogen formation in type 2 diabetes [242].
In summary, glycogen synthase activity is severely impaired in type 2
diabetic individuals, and the molecular cause of the defect most likely is
related to impaired insulin signal transduction.

Glycolysis/Glucose oxidation
Glucose oxidation accounts for approximately 90% of total glycolytic
flux, whereas anaerobic glycolysis accounts for the other 10%. The two
enzymes PFK and PDH play pivotal roles in the regulation of glycolysis and
glucose oxidation, respectively. In type 2 diabetic individuals, the glycolytic/
glucose oxidative pathway has been shown to be impaired [243]. Although
one study [244] has suggested that PFK activity is modestly reduced in
muscle biopsies from type 2 diabetic subjects, most evidence indicates that
the activity of PFK is normal [230,235]. Insulin has no effect on muscle PFK
activity, mRNA levels, or protein content in either nondiabetic or diabetic
individuals [235]. PDH is a key insulin-regulated enzyme with activity in
muscle that is acutely stimulated by insulin [245]. In type 2 diabetic patients,
insulin-stimulated PDH activity has been shown to be decreased in human
adipocytes and in skeletal muscle [245,246].
Obesity and type 2 diabetes mellitus are associated with accelerated FFA
turnover and oxidation [1,2,12,247], which would be expected, according to
the Randle cycle [248], to inhibit PDH activity and consequently glucose
822 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

oxidation. Therefore, it is likely that the observed defects in glucose oxidation


and PDH activity are acquired secondary to increased FFA oxidation and
feedback inhibition of PDH by elevated intracellular levels of acetyl-CoA and
reduced availability of NAD. Consistent with this scenario, the rates of basal
and insulin-stimulated glucose oxidation are not reduced in the normal
glucose-tolerant offspring of two diabetic parents and in the first-degree
relatives of type 2 diabetic subjects, whereas it is decreased in overtly diabetic
subjects.
Summary
In summary, postbinding defects in insulin action primarily are re-
sponsible for the insulin resistance in type 2 diabetes. Diminished insulin
binding, when present, is modest and secondary to down-regulation of the
insulin receptor by chronic hyperinsulinemia. In type 2 diabetic patients
with overt fasting hyperglycemia, a number of postbinding defects have
been demonstrated, including reduced insulin receptor tyrosine kinase
activity, insulin signal transduction abnormalities, decreased glucose trans-
port, diminished glucose phosphorylation, and impaired glycogen synthase
activity. The glycolytic/glucose oxidative pathway is largely intact and,
when defects are observed, they appear to be acquired secondary to
enhanced FFA/lipid oxidation. From the quantitative standpoint, impaired
glycogen synthesis represents the major pathway responsible for the insulin
resistance in type 2 diabetes and is present long before the onset of overt
diabetes, that is, in normal glucose-tolerant, insulin-resistant prediabetic
subjects and in individuals with IGT. Recent studies link the impairment
in glycogen synthase activation to a defect in the ability of insulin to
phosphorylate IRS-1, causing a reduced association of the p85 subunit of PI
3-kinase with IRS-1 and decreased activation of the enzyme PI3K.

References
[1] DeFronzo RA. Pathogenesis of type 2 diabetes mellitus: metabolic and molecular
implications for identifying diabetes genes. Diabetes 1997;5:177–269.
[2] DeFronzo RA. Lecture Lilly. The triumvirate: beta cell, muscle, liver. A collusion
responsible for NIDDM. Diabetes 1988;37:667–87.
[3] DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J, Felber JP. The effect of insulin
on the disposal of intravenous glucose: results from indirect calorimetry. Diabetes 1981;
30:1000–7.
[4] DeFronzo RA, Ferrannini E. Regulation of hepatic glucose metabolism in humans.
Diabetes Metab Rev 1987;3:415–59.
[5] DeFronzo RA, Gunnarsson R, Bjorkman O, Olsson M, Wahren J. Effects of insulin on
peripheral and splanchnic glucose metabolism in non-insulin dependent diabetes mellitus.
J Clin Invest 1985;76:149–55.
[6] Mari A, Wahren J, DeFronzo RA, Ferrannini E. Glucose absorption and production
following oral glucose: comparison of compartmental and arteriovenous-difference
methods. Metabolism 1994;43:1419–25.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 823

[7] Mandarino L, Bonadonna R, McGuinness O, Wasserman D. Regulation of muscle


glucose uptake in vivo. In: Jefferson LS, Cherrington AD, editors. Handbook of
physiology. The endocrine system, vol. II The endocrine pancreas and regulation of
metabolism. Oxford: Oxford University Press; 2001. p. 803–48.
[8] Mitrakou A, Kelley D, Veneman T, Jensen T, Pangburn T, Reilly J, et al. Contribution of
abnormal muscle and liver glucose metabolism to postprandial hyperglycemia in NIDDM.
Diabetes 1990;39:1381–90.
[9] Cherrington AD. Control of glucose uptake and release by the liver in vivo. Diabetes
1999;48:1198–214.
[10] Grill V. A comparison of brain glucose metabolism in diabetes as measured by positron
emission tomography or by arteriovenous techniques. Ann Med 1990;22:171–5.
[11] Bays H, Mandarino L, DeFronzo RA. Role of the adipocyte, free fatty acids, and ectopic
fat in pathogenesis of type 2 diabetes mellitus: peroxisomal proliferator-activated receptor
agonsits provide a rational therapeutic approach. J Clin Endocrinol Metab 2004;89:
463–78.
[12] Groop LC, Bonadonna RC, Del Prato S, Ratheiser K, Zych K, Ferrannini E, DeFronzo
RA. Glucose and free fatty acid metabolism in non-insulin dependent diabetes mellitus.
Evidence for multiple sites of insulin resistance. J Clin Invest 1989;84:205–15.
[13] Bergman RN. Non-esterified fatty acids and the liver: why is insulin secreted into the
portal vein? Diabetologia 2000;43:946–52.
[14] Boden G. Role of fatty acids in the pathogenesis of insulin resistance and NIDDM.
Diabetes 1997;46:3–10.
[15] Baron AD, Schaeffer L, Shragg P, Kolterman OG. Role of hyperglucagonemia in
maintenance of increased rates of hepatic glucose output in type II diabetics. Diabetes
1987;36:274–83.
[16] DeFronzo RA, Ferrannini E, Hendler R, Wahren J, Felig P. Influence of hyper-
insulinemia, hyperglycemia, and the route of glucose administration on splanchnic
glucose exchange. Proc Natl Acad Sci USA 1978;75:5173–7.
[17] Ferrannini E, Wahren J, Felig P, DeFronzo RA. Role of fractional glucose extraction in
the regulation of splanchnic glucose metabolism in normal and diabetic man. Metabolism
1980;29:28–35.
[18] Drucker DJ. Glucagon-like peptides. Diabetes 1998;47:159–69.
[19] Holst JJ, Gromada J, Nauck MA. The pathogenesis of NIDDM involves a defective
expression of the GIP receptor. Diabetologia 1997;40:984–6.
[20] Polonsky KS, Sturis J, Bell GI. Non-insulin-dependent diabetes mellitus: a genetically
programmed failure of the beta cell to compensate for insulin resistance. N Engl J Med
1996;334:777–83.
[21] Cerasi E. Insulin deficiency and insulin resistance in the pathogenesis of NIDDM: is
a divorce possible? Diabetologia 1995;38:992–7.
[22] Sicree RA, Zimmet P, King HO, Coventry JO. Plasma insulin response among Nauruans.
Prediction of deterioration in glucose tolerance over 6 years. Diabetes 1987;36:179–86.
[23] Saad MF, Knowler WC, Pettitt DJ, Nelson RG, Mott DM, Bennett PH. Sequential
changes in serum insulin concentration during development of non-insulin-dependent
diabetes. Lancet 1989;i:1356–9.
[24] Saad MF, Knowler WC, Pettitt DJ, Nelson RG, Mott DM, Bennett PH. The natural
history of impaired glucose tolerance in the Pima Indians. N Engl J Med 1988;319:
1500–5.
[25] Haffner SM, Miettinen H, Gaskill SP, Stern MP. Decreased insulin secretion and
increased insulin resistance are independently related to the 7-year risk of NIDDM in
Mexican-Americans. Diabetes 1995;44:1386–91.
[26] Weyer C, Hanson RL, Tataranni PA, Bogardus C, Pratley RE. A high fasting plasma
insulin concentration predicts type 2 diabetes independent of insulin resistance. Evidence
for a pathogenic role of relative hyperinsulinemia. Diabetes 2000;49:2094–101.
824 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[27] Weyer C, Tataranni PA, Bogardus C, Pratley RE. Insulin resistance and insulin secretory
dysfunction are independent predictors of worsening of glucose tolerance during each
stage of type 2 diabetes development. Diabetes Care 2000;24:89–94.
[28] Pimenta W, Korytkowski M, Mitrakou A, Jenssen T, Yki-Jarvinen H, Evron W, et al.
Pancreatic beta-cell dysfunction as the primary genetic lesion in NIDDM. Evidence from
studies in normal glucose-tolerant individuals with a first degree NIDDM relative. JAMA
1995;273:1855–61.
[29] Eriksson J, Franssila-Kallunki A, Ekstrand A, Saloranta C, Widen E, Schalin C, et al. Early
metabolic defects in persons at increased risk for non-insulin-dependent diabetes mellitus.
N Engl J Med 1989;321:337–43.
[30] Reaven GM, Hollenbeck C, Jeng C-Y, Wu MS, Chen Y-DI. Measurement of plasma
glucose, free fatty acid, lactate, and insulin for 24 hours in patients with NIDDM. Diabetes
1988;37:1020–4.
[31] Garvey WT, Olefsky JM, Rubenstein AH, Kolterman OG. Day-long integrated urinary
C-peptide excretion. Diabetes 1988;37:590–9.
[32] Gastaldelli A, Ferrannini E, Miyazaki Y, Matsuda M, DeFronzo RA. Beta-cell
dysfunction and glucose intolerance: results from the San Antonio metabolism (SAM)
study. Diabetologia 2004;47:31–9.
[33] Hansen BC, Bodkin NH. Heterogeneity of insulin responses: phases leading to type 2
(noninsulin-dependent) diabetes mellitus in the rhesus monkey. Diabetologia 1986;29:
713–9.
[34] Kahn SE. Clinical Review 135. The importance of b-cell failure in the development and
progression of type 2 diabetes. J Clin Endocrinol Metab 2001;86:4047–58.
[35] Bergman RN, Finegood DT, Kahn SE. The evolution of b-cell dysfunction and insulin
resistance in type 2 diabetes. Eur J Clin Invest 2002;32:35–45.
[36] Polonsky KS. Lilly Lecture 1994. The beta cell in diabetes: from molecular genetics to
clinical research. Diabetes 1995;44:705–17.
[37] Bell GI, Polonsky KS. Diabetes mellitus and genetically programmed defects in b-cell
function. Nature 2001;414:788–91.
[38] McCarthy MI, Froguel P. Genetic approaches to the molecular understanding of type 2
diabetes. Am J Physiol 2002;283:E217–25.
[39] Bell GI, Zian K, Newman M, Wu S, Wright L, Fajans S, Spielman RS, Cox NJ. Gene for
non-insulin-dependent diabetes mellitus (maturity-onset diabetes of the young subtype) is
linked to DNA polymorphism on human chromosome 20q. Proc Natl Acad Sci USA
1991;88:1484–8.
[40] Froguel PH, Zouali H, Vionnet N, Velho G, Vaxillaire M, Sun F, et al. Familial
hyperglycaemia due to mutations in glucokinase: definition of a subtype of diabetes
mellitus. N Engl J Med 1993;328:697–702.
[41] Beck-Nielsen H, Nielsen OH, Pedersen O, Bak J, Faber O, Schmitz O. Insulin action and
insulin secretion in identical twins with MODY: evidence for defects in both insulin action
and insulin secretion. Diabetes 1988;37:730–5.
[42] Mohan V, Sharp PS, Aber VR, Mather HM, Kohner EM. Insulin resistance in maturity-
onset diabetes of the young. Diabetes Metab 1988;13:193–7.
[43] Elbein SC, Hoffman M, Qin H, Chiu K, Tanizawa Y, Permutt MA. Molecular screening
of the glucokinase gene in familial type 2 (non-insulin-dependent) diabetes mellitus.
Diabetologia 1994;37:182–7.
[44] Efendic S, Grill V, Luft R, Wajngot A. Low insulin response: a marker of pre-diabetes.
Adv Exp Med Biol 1988;246:167–74.
[45] Davies MJ, Metcalfe J, Gray IP, Day JL, Hales CN. Insulin deficiency rather than
hyperinsulinaemia in newly diagnosed type 2 diabetes mellitus. Diabet Med 1993;10:
305–12.
[46] Chen K-W, Boyko EJ, Bergstrom RW, Leonetti DL, Newell-Morris L, Wahl PW, et al.
Earlier appearance of impaired insulin secretion than of visceral adiposity in the
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 825

pathogenesis of NIDDM. 5-year follow-up of initially nondiabetic Japanese-American


men. Diabetes Care 1995;18:747–53.
[47] Arner P, Pollare T, Lithell H. Different etiologies of Type 2 (non-insulin-dependent)
diabetes mellitus in obese and non-obese subjects. Diabetologia 1991;34:483–7.
[48] Ferrannini E, Natali A, Bell P, Cavallo-Perin P, Lalic N, Mingrone G. Insulin resistance
and hypersecretion in obesity. European Group for the Study of Insulin Resistance
(EGIR). J Clin Invest 1997;100:1166–73.
[49] Banjeri MA, Lebovitz HE. Insulin action in black Americans with NIDDM. Diabetes
Care 1992;15:1295–302.
[50] Mbanya J-CN, Pani LN, Mbanya DNS, Sobngwi E, Ngogang J. Reduced insulin
secretion in offspring of African type 2 diabetic patients. Diabetes Care 2000;23:1761–5.
[51] DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying
insulin secretion and resistance. Am J Physiol 1979;6:E214–23.
[52] Hales CN. The pathogenesis of NIDDM. Diabetologia 1994;37:S162–8.
[53] Brunzell JD, Robertson RP, Lerner RL, Hazzard WR, Ensinck JW, Bierman EL, et al.
Relationships between fasting plasma glucose levels and insulin secretion during
intravenous glucose tolerance tests. J Clin Endocrinol 1976;46:222–9.
[54] Vague P, Moulin J-P. The defective glucose sensitivity of the B cell in insulin dependent
diabetes. Improvement after twenty hours of normoglycaemia. Metabolism 1982;31:
139–42.
[55] Kosaka K, Kuzuya T, Akanuma Y, Hagura R. Increase in insulin response after
treatment of overt maturity onset diabetes mellitus is independent of the mode of
treatment. Diabetologia 1980;18:23–8.
[56] Rossetti L, Giaccari A, DeFronzo RA. Glucose toxicity. Diabetes Care 1990;13:610–30.
[57] Unger RH. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM. Genetic and
clinical implications. Diabetes 1995;44:863–70.
[58] McGarry JD. Banting lecture 2001: dysregulation of fatty acid metabolism in the etiology
of type 2 diabetes. Diabetes 2002;51:7–18.
[59] Shimabukuro M, Zhou Y-T, Levi M, Unger RH. Fatty acid induced b cell apoptosis:
a link between obesity and diabetes. Proc Natl Acad Sci USA 1998;95:2498–502.
[60] Bruce DG, Chisholm DJ, Storlien LH, Kraegen EW. Physiological importance of
deficiency in early prandial insulin secretion in non-insulin dependent diabetes. Diabetes
1988;37:736–44.
[61] Luzi L. Effect of the loss of first phase insulin secretion on glucose production and
disposal in man. Am J Physiol 1989;257:E241–6.
[62] Bonner-Weir S. b-cell turnover. Its assessment and implications. Diabetes 2001;50:S20–4.
[63] Gautier J-F, Wilson C, Weyer C, Mott D, Knowler WC, Cavaghan M, et al. Low acute
insulin secretory responses in adult offspring of people with early onset type 2 diabetes.
Diabetes 2001;50:1828–33.
[64] Vauhkonen N, Niskanen L, Vanninen E, Kainulainen S, Uusitupa M, Laakso M. Defects
in insulin secretion and insulin action in non-insulin-dependent diabetes mellitus are
inherited. Metabolic studies on offspring of diabetic probands. J Clin Invest 1997;100:
86–96.
[65] Vaag A, Henriksen JE, Madsbad S, Holm N, Beck-Nielsen H. Insulin secretion, insulin
action, and hepatic glucose production in identical twins discordant for non-insulin-
dependent diabetes mellitus. J Clin Invest 1995;95:690–8.
[66] Barnett AH, Spilipoulos AJ, Pyke DA, Stubbs WA, Burrin J, Alberti KGMM. Metabolic
studies in unaffected co-twins of non-insulin-dependent diabetics. BMJ 1981;282:1656–8.
[67] Watanabe RM, Valle T, Hauser ER, Ghosh S, Eriksson J, Kohtamaki K, Ehnholm C,
Tuomilehto J, Collins FS, Bergman RN, Boehnke M. Familiality of quantitative metabolic
traits in Finnish families with non-insulin-dependent diabetes mellitus. Finland-United
States Investigation of NIDDM Genetics (FUSION) Study investigators. Hum Hered
1999;49(3):159–68.
826 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[68] Rossetti L, Shulman Gi, Zawalich W, DeFronzo RA. Effect of chronic hyperglycemia on
in vivo insulin secretion in partially pancreatectomized rats. J Clin Invest 1987;80:
1037–44.
[69] Leahy JL, Bonner-Weir S, Weir GC. Minimal chronic hyperglycemia is a critical
determinant of impaired insulin secretion after an incomplete pancreatectomy. J Clin
Invest 1988;81:1407–14.
[70] Leahy JL, Cooper HE, Weir GC. Impaired insulin secretion associated with near
normoglycemia. Diabetes 1987;36:459–64.
[71] Robertson RP, Olson IK, Zhang H-J. Differentiating glucose toxicity from glucose
desensitization: a new message from the insulin gene. Diabetes 1994;43:1085–9.
[72] Prentki M, Corkey BE. Are the beta cell signaling molecules malonyl-CoA and cytosolic
long-chain acyl-CoA implicated in multiple tissue defects of obesity and NIDDM?
Diabetes 1996;45:273–83.
[73] Creutzfeldt W. The incretin concept today. Diabetologia 1979;16:75–85.
[74] Drucker DJ. The glucagon-like peptides. Endocrinology [minireview] 2001;142:521–7.
[75] Nauck MA, Bartels E, Orskov C, Ebert R, Creutzfeldt W. Additive insulinotropic effects
of exogenous synthetic human gastric inhibitory polypeptide and glucagon-like peptide-1-
(7–36) amide infused at near-physiological insulinotropic hormone and glucose
concentrations. J Clin Endocrinol Metab 1993;76:912–7.
[76] Vilsboll T, Krarup T, Deacon CF, Madsbad S, Holst JJ. Reduced postprandial
concentrations of intact biologically active glucagon-like peptide 1 in type 2 diabetic
patients. Diabetes 2001;50:609–13.
[77] Ahren B, Larsson H, Holst JJ. Effects of glucagon-like peptide-1 on islet function and
insulin sensitivity in noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab
1997;82:473–8.
[78] D’Alessio DA, Vogel R, Prigeon R, Laschansky E, Koerker D, Eng J, Ensinck JW.
Elimination of the action of glucagon-like peptide 1 causes an impairment of glucose
tolerance after nutrient ingestion by healthy baboons. J Clin Invest 1996;97:133–8.
[79] Nanauck MA, Kleine N, Orskov C, Holst JJ, Willms B, Creutzfeldt W. Normalization of
fasting hyperglycaemia by exogenous glucagon-like peptide 1 (7–36 amide) in type 2 (non-
insulin-dependent) diabetic patients. Diabetologia 1993;36:741–4.
[80] Xu G, Stoffers DA, Habener JF, Bonner-Weir S. Exendin-4 stimulates both beta-cell
replication and neogenesis, resulting in increased beta-cell mass and improved glucose
tolerance in diabetic rats. Diabetes 1999;48:2270–6.
[81] Kahn SE, Andrikopoulos S, Verchere CB. Islet amyloid. A long-recognized but
underappreciated pathological feature of type 2 diabetes. Diabetes 1999;48:241–53.
[82] Johnson KH, O’Brien TD, Betysholtz C, Westermark P. Islet amyloid, islet-amyloid
polypeptide, and diabetes mellitus. N Engl J Med 1989;321:513–8.
[83] Howard CF. Longitudinal studies on the development of diabetes in individual Macaca
nigra. Diabetologia 1986;29:301–6.
[84] Bretherton-Watt D, Gilbey SG, Ghatei MA, Beacham J, Bloom SR. Failure to establish
islet amyloid polypeptide (amylin) as a circulating beta cell inhibiting hormone in man.
Diabetologia 1990;33:115–7.
[85] Hoppener JWM, Verbeek JS, de Koning EJP, Oosterwijk C, van Hulst KL, Visser-
Vernooy HJ, et al. Chronic overproduction of islet amyloid polypeptide-amylin in
transgenic mice: lysosomal localization of human islet amyloid polypeptide and lack of
marked hyperglycaemia or hyperinsulinaemia. Diabetologia 1993;36:1258–65.
[86] Gebre-Medhin S, Olofsson C, Mulder H. Islet amyloid polypeptide in the islets of
Langerhans: friend or foe? Diabetologia 2000;43:687–95.
[87] Gepts W, Lecompte PM. The pancreatic islets in diabetes. Am J Med 1981;70:105–14.
[88] Kloppel G, Lohr M, Habich K, Oberholzer M, Heitz PU. Islet pathology and the
pathogenesis of type I and type 2 diabetes mellitus revisited. Surv Synth Path Res 1985;4:
110–25.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 827

[89] Clark A, Wells CA, Buley ID, Cruickshank JK, Vanhegan RI, Matthews DR, et al. Islet
amyloid, increased alpha-cells, reduced beta-cells and exocrine fibrosis: quantitative
changes in the pancreas in type 2 diabetes. Diabetes Res 1988;9:151–9.
[90] Butler AE, Janson J, Bonner-Weir S, Ritzel RA, Butler PC. Beta-cell deficit and increased
beta-cell apoptosis in humans with type 2 diabetes. Diabetes 2003;52:102–10.
[91] Stefan Y, Orci L, Malaisse-Lagae F, Perrelet A, Patel Y, Unger R. Quantitation of
endocrine cell content in the pancreas of non-diabetic and diabetic humans. Diabetes
1982;31:694–700.
[92] Rahier J, Sempoux C, Moulin P, Guiot Y. No decrease of the B cell mass in type 2
diabetic patients. Diabetologia 2000;43(Suppl 1):A65.
[93] Janson J, Butler AE, Bonner-Weir S, Ritzel RA, Sultana C, Butler PC. Failure of
compensatory increase in new islet formation in humans with type-2 diabetes mellitus.
Diabetes 2002;51(Suppl 2):A377.
[94] Finegood DT, McArthur D, Kojwang M, Thomas J, Topp BG, Leonard T, et al. b-cell
mass dynamics in Zucker diabetic fatty rats: rosiglitazone prevents the rise in net cell
death. Diabetes 2000;50:1021–9.
[95] Eriksson UJ. Lifelong consequences of metabolic adaptations in utero? Diabetologia
1996;39:1123–5.
[96] Phillips DIW. Insulin resistance as a programmed response to fetal under nutrition.
Diabetologia 1996;39:1119–22.
[97] Reaven GM, Hollenbeck CB, Chen YDI. Relationship between glucose tolerance, insulin
secretion, and insulin action in non-obese individuals with varying degrees of glucose
tolerance. Diabetologia 1989;32:52–5.
[98] Gulli G, Ferrannini E, Stern M, Haffner S, DeFronzo RA. The metabolic profile of
NIDDM is fully established in glucose-tolerant offspring of two Mexican-American
NIDDM parents. Diabetes 1992;41:1575–86.
[99] Martin BC, Warram JH, Krolewski AS, Bergman RN, Soeldner JS, Kahn RC. Role of
glucose and insulin resistance in development of type 2 diabetes mellitus: results of a
25-year follow-up study. Lancet 1992;340:925–9.
[100] Hara H, Egusa G Yamakido M, Kawate R. The high prevalence of diabetes mellitus and
hyperinsulinemia among the Japanese-Americans living in Hawaii and Los Angeles.
Diabetes Res Clin Pract 1994;24(Suppl 1):S37–42.
[101] Lillioja S, Nyomba BL, Saad MF, Ferraro R, Castillo C, Bennett PH, et al.
Exaggerated early insulin release and insulin resistance in a diabetes-prone population:
a metabolic comparison of Pima Indians and Caucasians. J Clin Endocrinol Metab
1991;73:866–76.
[102] Lillioja S, Mott DM, Howard BV, Bennett PH, Yki-Jarvinen H, Freymond D, Nyomba
BL, Zurlo F, Swinburn B, Bogardus C. Impaired glucose tolerance as a disorder of insulin
action. Longitudinal and cross-sectional studies in Pima Indians. N Engl J Med 1988;318:
1217–25.
[103] Himsworth HP, Kerr RB. Insulin-sensitive and insulin-insensitive types of diabetes
mellitus. Clin Sci (Lond) 1939;4:120–52.
[104] Ginsberg H, Kimmerling G, Olefsky JM, Reaven GM. Demonstration of insulin
resistance in untreated adult-onset diabetic subjects with fasting hyperglycemia. J Clin
Invest 1975;55:454–61.
[105] Butterfield WJH, Whichelow MJ. Peripheral glucose metabolism in control subjects and
diabetic patients during glucose, glucose-insulin, and insulin sensitivity tests. Diabetologia
1965;1:43–53.
[106] Katz H, Homan M, Jensen M, Caumo A, Cobelli C, Rizza R. Assessment of insulin
action in NIDDM in the presence of dynamic changes in insulin and glucose
concentration. Diabetes 1994;43:289–96.
[107] Bergman RN, Lecture Lilly. Toward physiological understanding of glucose tolerance:
minimal-model approach. Diabetes 1989;38:1512–26.
828 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[108] DeFronzo RA, Deibert D, Hendler R, Felig P. Insulin sensitivity and insulin binding to
monocytes in maturity-onset diabetes. J Clin Invest 1982;63:939–46.
[109] DeFronzo RA, Sherwin RS, Hendler R, Felig P. Insulin binding to monocytes and insulin
action in human obesity, starvation, and refeeding. J Clin Invest 1978;62:204–13.
[110] Firth R, Bell P, Rizza R. Insulin action in non-insulin-dependent diabetes mellitus: the
relationship between hepatic and extrahepatic insulin resistance and obesity. Metabolism
1987;36:1091–5.
[111] Campbell PJ, Mandarino LJ, Gerich JE. Quantification of the relative impairment in
actions of insulin on hepatic glucose production and peripheral glucose uptake in non-
insulin dependent diabetes mellitus. Metabolism 1988;37:15–21.
[112] Bogardus C, Lillioja S, Howard BV, Reaven G, Mott D. Relationships between insulin
secretion, insulin action, and fasting plasma glucose concentration in non-diabetic and
noninsulin-dependent subjects. J Clin Invest 1984;74:1238–46.
[113] Del Prato S, Simonson DC, Sheehan P, Cardi F, DeFronzo RA. Studies on the mass effect of
glucose in diabetes. Evidence for glucose resistance. Diabetologia 1997;40:687–97.
[114] DeFronzo RA, Ferrannini E, Simonson DC. J Fasting hyperglycemia in non-insulin-
dependent diabetes mellitus: contributions of excessive hepatic glucose production and
impaired tissue glucose uptake. Metabolism 1989;38:387–95.
[115] Huang SC, Phelps ME, Hoffman EJ, Sideris K, Selin CJ, Kuhl DE. Non-invasive
determination of local cerebral metabolic rate of glucose in man. Am J Physiol 1980;238:
E69–82.
[116] Best JD, Judzewitsch RG, Pfeiffer MA, Beard JC, Halter JB, Porte D. The effect of
chronic sulfonylurea therapy on hepatic glucose production in non-insulin-dependent
diabetes mellitus. Diabetes 1982;31:333–8.
[117] Fery F. Role of hepatic glucose production and glucose uptake in the pathogenesis of
fasting hyperglycemia in Type 2 diabetes: normalization of glucose kinetics by short-term
fasting. J Clin Endocrinol Metab 1994;78:536–42.
[118] Jeng C-Y, Sheu WHH, Fuh MM-T, Chen I, Reaven GM. Relationship between hepatic
glucose production and fasting plasma glucose concentration in patients with NIDDM.
Diabetes 1994;43:1440–4.
[119] Waldhausl W, Bratusch-Marrain P, Gasic S, Korn A, Nowotny P. Insulin production
rate, hepatic insulin retention, and splanchnic carbohydrate metabolism after oral glucose
ingestion in hyperinsulinemic type II (non-insulin dependent) diabetes mellitus.
Diabetologia 1982;23:6–15.
[120] Nurjhan N, Consoli A, Gerich J. Increased lipolysis and its consequences on
gluconeogenesis in noninsulin-dependent diabetes mellitus. J Clin Invest 1992;89:169–75.
[121] Stumvoll M, Perriello G, Nurjhan N, Bucci A, Welle S, Jansson P-A, et al. Glutamine and
alanine metabolism in NIDDM. Diabetes 1996;45:863–8.
[122] Magnusson I, Rothman D, Katz L, Shulman R, Shulman G. Increased rate of
gluconeogenesis in type II diabetes: a 13C nuclear magnetic resonance study. J Clin Invest
1992;90:1323–7.
[123] Gastaldelli A, Baldi S, Pettiti M, Toschi E, Camastra S, Natali A, et al. Influence of
obesity and type 2 diabetes on gluconeogenesis and glucose output in humans:
a quantitative study. Diabetes 2000;49:1367–73.
[124] Clore JN, Stillman J, Sugerman H. Glucose-6-phosphatase flux in vitro is increased in
type 2 diabetes. Diabetes 2000;49:969–74.
[125] Gerich JE, Meyer C, Woerle HJ, Stumvoll M. Renal gluconeogenesis. Its importance in
human glucose homeostasis. Diabetes Care 2001;24:382–91.
[126] Ekberg K, Landau BR, Wajngot A, Chandramouli V, Efendic S, Brunengraber H, et al.
Contributions by kidney and liver to glucose production in the postabsorptive state and
after 60 h of fasting. Diabetes 1999;48:292–8.
[127] Moller N, Rizza RA, Ford GC, Nair KS. Assessment of postabsorptive renal glucose
metabolism in humans with multiple glucose tracers. Diabetes 2001;50:747–51.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 829

[128] Meyer C, Stumvoll M, Nadkarni V, Dostou J, Mitrakou A, Gerich J. Abnormal renal


and hepatic glucose metabolism in type 2 diabetes mellitus. J Clin Invest 1998;102:619–24.
[129] DeFronzo RA, Ferrannini E, Hendler R, Felig P, Wahren J. Regulation of splanchnic
and peripheral glucose uptake by insulin and hyperglycemia. Diabetes 1983;32:35–45.
[130] Frayn KN, Coppack SW, Humphreys SM, Whyte PL. Metabolic characteristics of
human adipose tissue in vivo. Clin Sci 1989;76:509–16.
[131] Ferrannini E, Bjorkman O, Reichard GA Jr, Pilo A, Olsson M, Wahren J, et al. The
disposal of an oral glucose load in healthy subjects. Diabetes 1985;34:580–8.
[132] Katz LD, Glickman MG, Rapoport S, Ferrannini E, DeFronzo RA. Splanchnic and
peripheral disposal of oral glucose in man. Diabetes 1983;32:675–9.
[133] Ferrannini E, Simonson DC, Katz LD, Reichard G Jr, Bevilacqua S, Barrett EJ, et al.
The disposal of an oral glucose load in patients with non-insulin dependent diabetes.
Metabolism 1988;37:79–85.
[134] Firth RG, Bell PM, Marsh HM, Hansen I, Rizza RA. Postprandial hyperglycemia in
patients with non-insulin-dependent diabetes mellitus. J Clin Invest 1986;77:1525–32.
[135] Kelley D, Mitrakou A, Marsh H, Schwenk F, Benn J, Sonnenberg G, et al. Skeletal muscle
glycolysis oxidation, and storage of an oral glucose load. J Clin Invest 1988;81:1563–71.
[136] Jackson RA, Roshania RD, Hawa MI, Sim BM, DiSilvio L. Impact of glucose ingestion on
hepatic and peripheral glucose metabolism in man: an analysis based on simultaneous use
of the forearm and double isotope techniques. J Clin Endocrinol Metab 1986;63:541–9.
[137] Ludvik B, Nolan JJ, Roberts A, Baloga J, Joyce M, Bell Jo M, et al. Evidence for
decreased splanchnic glucose uptake after oral glucose administration in non-insulin-
dependent diabetes mellitus. J Clin Invest 1997;100:2354–61.
[138] Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults:
findings from the third National Health and Nutrition Examination Survey. JAMA 2002;
287:356–9.
[139] DeFronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome responsible for
NIDDM, obesity, hypertension, dyslipidemia, and ASCVD. Diabetes Care 1991;14:
173–94.
[140] Reaven GM, Brand RJ, Ida Chen Y-D, Mathur AK, Goldfine I. Insulin resistance and
insulin secretion are determinants of oral glucose tolerance in normal individuals.
Diabetes 1993;42:1324–32.
[141] Diamond MP, Thornton K, Connolly-Diamond M, Sherwin RS, DeFronzo RA.
Reciprocal variations in insulin-stimulated glucose uptake and pancreatic insulin
secretion in women with normal glucose tolerance. J Soc Gynecol Invest 1995;2:708–15.
[142] Reaven GM, Chen Y-DI, Donner CC, Fraze E, Hollenbeck CB. How insulin resistant are
patients with non-insulin-dependent diabetes mellitus? J Clin Endocrinol Metab 1985;61:
32–6.
[143] Bogardus C, Lillioja S, Howard BV, Reaven G, Mott D. Relationships between insulin
secretion, insulin action, and fasting plasma glucose concentration in non-diabetic and
noninsulin-dependent subjects. J Clin Invest 1984;74:1238–46.
[144] Dowse GK, Zimmet PZ, Collins VR. Insulin levels and the natural history of glucose
intolerance in Nauruans. Diabetes 1996;45:1367–72.
[145] Haffner SM, Stern MP, Hazuda HP, Pugh JA, Patterson JK. Hyperinsulinemia in
a population at high risk for non-insulin-dependent diabetes mellitus. N Engl J Med 1986;
315:220–4.
[146] Haffner SM, Miettinen H, Stern MP. Insulin secretion and resistance in nondiabetic
Mexican Americans and non-Hispanic whites with a parental history of diabetes. J Clin
Endocrinol Metab 1996;81:1846–51.
[147] Mokdad AH, Ford ES, Bowman BA, Nelson DE, Engelgau MM, Vinicor F, et al.
Diabetes trends in the United States, 1990–1998. Diabetes Care 2000;23:1278–83.
[148] Thiebaud D, DeFronzo RA, Jacot E, Golay A, Acheson K, Maeder E, et al. Effect of long-
chain triglyceride infusion on glucose metabolism in man. Metabolism 1982;31:1128–36.
830 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[149] De Kelley, Mandarino LJ. Fuel selection in human skeletal muscle in insulin resistance.
A reexamination. Diabetes 2000;49:677–83.
[150] Kashyap S, Belfort R, Gastadelli A, Pratipanawatr T, Berria R, Pratipanawatr W, et al.
Chronic elevation in plasma free fatty acids impairs insulin secretion in non-diabetic
offspring with a strong family history of T2DM. Diabetes 2003;52:2464–74.
[151] Carpentier A, Mittelman SD, Bergman RN, Giacca A, Lewis GF. Prolonged elevation of
plasma free fatty acids impairs pancreatic beta-cell function in obese nondiabetic humans
but not in individuals with type 2 diabetes. Diabetes 2000;49:399–408.
[152] Reaven GM. The fourth Musketeer: from Alexendre Dumas to Calude Bernard.
Diabetologia 1995;38:3–13.
[153] Goodpaster BH, Thaete FL, Kelley BE. Thigh adipose tissue distribution is associated
with insulin resistance in obesity and in type 2 diabetes mellitus. Am J Clin Nutr 2000;71:
885–92.
[154] Greco AV, Mingrone G, Giancaterini A, Manco M, Morroni M, Cinti S, et al. Insulin
resistance in morbid obesity. Reversal with intramyocellular fat depletion. Diabetes 2002;
51:144–51.
[155] Ryysy L, Hakkinen AM, Goto T, Vehkavaara S, Westerbacka J, Halavaara J, et al.
Hepatic fat content and insulin action on free fatty acids and glucose metabolism rather
than insulin absorption are associated with insulin requirements during insulin therapy in
type 2 diabetic patients. Diabetes 2000;49:749–58.
[156] Bajaj M, Surraamornkul S, Piper P, Hardies LJ, Glass L, Cersosimo E, et al. Decreased
plasma adiponectin concentrations are closely related to hepatic fat content and hepatic
insulin resistance in pioglitazone-treated type 2 diabetic patients. J Clin Endocrinol
Metab 2004;89:200–6.
[157] Itani SI, Ruderman NB, Schmieder F, Boden G. Lipid-induced insulin resistance in
human muscle is associated with changes in diacylglycerol, protein kinase C, and IjB-a.
Diabetes 2002;51:2005–11.
[158] Ellis BA, Poynten A, Lowy AJ, Furler SM, Chisholm DJ, Kraegen EW, et al. Long-chain
acyl-CoA esters as indicators of lipid metabolism and insulin sensitivity in rat and human
muscle. Am J Physiol 2000;279:E554–60.
[159] Schmitz-Peiffer C, Craig DL, Biden TJ. Ceramide generation is sufficient to account for
the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal muscle cells
pretreated with palmitate. J Biol Chem 1999;274:24202–10.
[160] Saltiel AR, Kahn CR. Insulin signaling and the regulation of glucose and lipid
metabolism. Nature 2001;414:799–806.
[161] Pessin JE, Saltiel AR. Signaling pathways in insulin action: molecular targets of insulin
resistance. J Clin Invest 2000;106:165–9.
[162] Whitehead JP, Clark SF, Urso B, James DE. Signaling through the insulin receptor. Cur
Opin Cell Biol 2000;12:222–8.
[163] Ellis LE, Clauser E, Morgan ME, Roth RA, Rutter WJ. Replacement of insulin receptor
tyrosine residues 1162 and 1163 compromises insulin-stimulated kinase activity and
uptake of 2-dexoyglucose. Cell 1986;45:721–32.
[164] Chou DK, Dull TJ, Russell DS, Gherzi R, Lebwohl D, Ullrich A, et al. Human insulin
receptors mutated at the ATP-binding site lack protein tyrosine kinase activity and fail to
mediate postreceptor effects of insulin. J Biol Chem 1987;262:1842–7.
[165] Virkamaki A, Ueki K, Kahn CR. Protein-protein interaction in insulin signaling and the
molecular mechanisms of insulin resistance. J Clin Invest 1999;103:931–43.
[166] Kerouz NJ, Horsch D, Pons S, Kahn CR. Differential regulation of insulin receptor
substrates-1 and -2 (IRS-1 and IRS-2) and phosphatidylinositol 3-kinase isoforms in liver
and muscle of the obese diabetic (ob/ob) mouse. J Clin Invest 1997;100:3164–72.
[167] Sun XJ, Miralpeix M, Myers MG Jr, Glasheen EM, Backer JM, Kahn CR, et al. The
expression and function of IRS-1 in insulin signal transmission. J Biol Chem 1992;267:
22662–72.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 831

[168] Cross D, Alessi D, Vandenheed J, McDowell H, Hundal H, Cohen P. The inhibition of


glycogen synthase kinase-3 by insulin or insulin-like growth factor 1 in the rat skeletal
muscle cell line L6 is blocked by wortmannin but not rapamycin. Biochem J 1994;303:
21–6.
[169] Osawa H, Sutherland C, Robey R, Printz R, Granner D. Analysis of the signaling
pathway involved in the regulation of hexokinase II gene transcription by insulin. J Biol
Chem 1996;271:16690–4.
[170] Lazar DF, Wiese RJ, Brady MJ, Mastick CC, Waters SB, Yamauchi K, Pessin JE,
Cuatrecasas P, Saltiel AR. Mitogen-activated protein kinase kinase inhibition does not
block the stimulation of glucose utilization by insulin. J Biol Chem 1995;270:20801–7.
[171] Dent P, Lavoinne A, Nakielny S, Caudwell FB, Watt P, Cohen F. The molecular
mechanisms by which insulin stimulates glycogen synthesis in mammalian skeletal muscle.
Nature 1990;348:302–7.
[172] Newgard CB, Brady MJ, O’Doherty RB, Saltiel AR. Organizing glucose disposal.
Emerging roles of the glycogen targeting subunits of protein phosphatase-1. Diabetes
2000;49:1967–77.
[173] Sheperd PR, Nave BT, Siddle K. Insulin stimulation of glycogen synthesis and glycogen
synthase activity is blocked by wortmannin and rapamycin in 3T3L1 adipocytes: evidence
for the involvement of phosphoinositide 3 kinase and p70 ribosomal protein S6 kinase.
Biochem J 1995;305:25–8.
[174] Freidenberg GR, Henry RR, Klein HH, Reichart DR, Olefsky JM. Decreased kinase
activity of insulin receptors from adipocytes of non-insulin-dependent diabetic studies.
J Clin Invest 1987;79:240–50.
[175] Caro JF, Sinha MK, Raju SM, Ittoop O, Pories WJ, Flickinger EG, et al. Insulin receptor
kinase in human skeletal muscle from obese subjects with and without non-insulin
dependent diabetes. J Clin Invest 1987;79:1330–7.
[176] Caro JF, Ittoop O, Pories WJ, Meelheim D, Flickinger EG, Thomas F, et al. Studies on
the mechanism of insulin resistance in the liver from humans with non-insulin-dependent
diabetes. Insulin action and binding in isolated hepatocytes, insulin receptor structure,
and kinase activity. J Clin Invest 1986;78:249–58.
[177] Trichitta V, Brunetti A, Chiavetta A, Benzi L, Papa V, Vigneri R. Defects in insulin-
receptor internalization and processing in monocytes of obese subjects and obese
NIDDM patients. Diabetes 1989;38:1579–84.
[178] Klein HH, Vestergaard H, Kotzke G, Pedersen O. Elevation of serum insulin
concentration during euglycemic hyperinsulinemic clamp studies leads to similar
activation of insulin receptor kinase in skeletal muscle of subjects with and without
NIDDM. Diabetes 1995;344:1310–7.
[179] Kashiwagi A, Verso MA, Andrews J, Vasquez B, Reaven G, Foley JE. In vitro insulin
resistance of human adipocytes isolated from subjects with non-insulin-dependent
diabetes mellitus. J Clin Invest 1983;72:1246–54.
[180] Lonnroth P, Digirolamo M, Krotkiewski M, Smith U. Insulin binding and responsiveness
in fat cells from patients with reduced glucose tolerance and type II diabetes. Diabetes
1983;32:748–54.
[181] Olefsky JM, Reaven GM. Insulin binding in diabetes. Relationships with plasma insulin
levels and insulin sensitivity. Diabetes 1977;26:680–8.
[182] Moller DE, Yakota A, Flier JS. Normal insulin receptor cDNA sequence in Pima Indians
with non-insulin-dependent diabetes mellitus. Diabetes 1989;38:1496–500.
[183] Kusari J, Verma US, Buse JB, Henry RR, Olefsky JM. Analysis of the gene sequences of the
insulin receptor and the insulin-sensitive glucose transporter (GLUT4) in patients with
common-type non-insulin-dependent diabetes mellitus. J Clin Invest 1991;88:1323–30.
[184] Nolan JJ, Friedenberg G, Henry R, Reichart D, Olefsky JM. Role of human skeletal
muscle insulin receptor kinase in the in vivo insulin resistance of noninsulin-dependent
diabetes and obesity. J Clin Endocrinol Metab 1994;78:471–7.
832 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[185] Cusi K, Maezono K, Osman A, Pendergrass M, Patti ME, Pratipanawatr T, et al. Insulin
resistance differentially affects the PI 3-kinase and MAP kinase-mediated signaling in
human muscle. J Clin Invest 2000;105:311–20.
[186] Freidenberg GR, Reichart D, Olefsky JM, Henry RR. Reversibility of defective adipocyte
insulin receptor kinase activity in non-insulin dependent diabetes mellitus. Effect of
weight loss. J Clin Invest 1988;82:1398–406.
[187] Kellerer M, Kroder G, Tippmer S, Berti L, Kiehn L, Mosthaf L, et al. Troglitazone
prevents glucose-induced insulin resistance of insulin receptor in rat-1 fibroblasts.
Diabetes 1994;43:447–53.
[188] Pratipanawatr W, Pratipanawatr T, Cusi K, Berria R, Adams JM, Jenkinson CP, et al.
Skeletal muscle insulin resistance in normoglycemic subjects with a strong family history
of type 2 diabetes is associated with decreased insulin-stimulated insulin receptor
substrate-1 tyrosine phosphorylation. Diabetes 2001;50:2572–8.
[189] Krook A, Bjornholm M, Galuska D, Jiang XJ, Fahlman R, Myers MG, et al.
Characterization of signal transduction and glucose transport in skeletal muscle from type
2 diabetic patients. Diabetes 2000;49:284–92.
[190] Kim Y-B, Nikoulina S, Ciaraldi TP, Henry RR, Kahn BB. Normal insulin-dependent
activation of Akt/protein kinase B, with diminished activation of phosphoinositide
3-kinase, in muscle in type 2 diabetes. J Clin Invest 1999;104:733–41.
[191] Andreelli F, Laville M, Ducluzeau P-H, Vega N, Vallier P, Khalfallah Y, et al. Defective
regulation of phosphatidylinositol-3-kinase gene expression in skeletal muscle and
adipose tissue of non-insulin-dependent diabetes mellitus patients. Diabetologia 1999;42:
358–64.
[192] Folli F, Saad JA, Backer JM, Kahn CR. Regulation of phosphatidylinositol 3-kinase
activity in liver and muscle of animal models of insulin-resistant and insulin-deficient
diabetes mellitus. J Clin Invest 1993;92:1787–94.
[193] Hitman GA, Hawrammi K, McCarthy MI, Viswanathan M, Snehalatha C, Ramachan-
dran A, et al. Insulin receptor substrate-1 gene mutations in NIDDM: implication for the
study of polygenic disease. Diabetologia 1995;38:481–6.
[194] Hsueh WA, Law RE. Insulin signaling in the arterial wall. Am J Cardiol 1999;84:21–4J.
[195] Jiang ZY, Lin YW, Clemont A, Feener EP, Hein KD, Igarashi M, Yamauchi T, White
MF, King GL. Characterization of selective resistance to insulin signaling in the
vasculature of obese Zucker (fa/fa) rats. J Clin Invest 1999;104:447–57.
[196] Shepherd PR, Kahn BB. Glucose transporters and insulin action. Implications for insulin
resistance and diabetes mellitus. N Engl J Med 1999;341:248–57.
[197] Garvey WT. Insulin action and insulin resistance: diseases involving defects in insulin
receptors, signal transduction, and the glucose transport effector system. Am J Med 1998;
105:331–45.
[198] Gl Bell, Kayano T, Buse JB, Burant CF, Takeda J, Lin D, Fikomoto H, Seino S.
Molecular biology of mammalian glucose transporters. Diabetes Care 1990;13:198–200.
[199] Joost H-G, Bell GI, Best JD, Birnbaum MJ, Charron MJ, Chen YT, et al. Nomenclature
of the GLUT/SLC2A family of sugar/polyol transport facilitators. Am J Physiol
Endocrinol Metab 2002;282:E974–6.
[200] Rogers PA, Fisher RA, Harris H. An electrophoretic study of the distribution and
properties of human hexokinases. Biochem Genet 1975;13:857–66.
[201] Matchinsky FM. Banting Lecture 1995. A lesson in metabolic regulation inspired by the
glucokinase glucose sensor paradigm. Diabetes 1996;45:223–41.
[202] Garvey WT, Huecksteadt TP, Mattaei S, Olefsky JM. Role of glucose transporters in the
cellular insulin resistance of type II non-insulin dependent diabetes mellitus. J Clin Invest
1988;81:1528–36.
[203] Zierath JR, He L, Guma A, Odegoard Wahlstrom E, Klip A, Wallenberg-Henriksson H.
Insulin action on glucose transport and plasma membrane GLUT4 content in skeletal
muscle from patients with NIDDM. Diabetologia 1996;39:1180–9.
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 833

[204] Krook A, Bjornholm M, Galuska D, Jiang XJ, Fahlman R, Myers MG, et al.
Characterization of signal transduction and glucose transport in skeletal muscle from type
2 diabetic patients. Diabetes 2000;49:284–92.
[205] Pedersen O, Bak J, Andersen P, Lund S, Moller DE, Flier JS, et al. Evidence against
altered expression of GLUT1 or GLUT4 in skeletal muscle of patients with obesity or
NIDDM. Diabetes 1990;39:865–70.
[206] Eriksson J, Koranyi L, Bourey R, Schalin-Jantti C, Widen E, Mueckler M, et al. Insulin
resistance in type 2 (non-insulin-dependent) diabetic patients and their relatives is not
associated with a defect in the expression of the insulin-responsive glucose transporter
(GLUT-4) gene in human skeletal muscle. Diabetologia 1992;35:143–7.
[207] Bonadonna RC, Del Prato S, Saccomani MP, Bonora E, Gulli G, Ferrannini E, et al.
Transmembrane glucose transport in skeletal muscle of patients with non-insulin-
dependent diabetes. J Clin Invest 1993;92:486–94.
[208] Bonadonna RC, Del Prato S, Bonora E, Saccomani MP, Gulli G, Natali A, et al. Roles of
glucose transport and glucose phosphorylation in muscle insulin resistance of NIDDM.
Diabetes 1996;45:915–25.
[209] Cline GW, Petersen KF, Krssak M, Shen J, Hundal RS, Trajanoski Z, et al. Impaired
glucose transport as a cause of decreased insulin stimulated muscle glycogen synthesis in
type 2 diabetes. N Engl J Med 1999;341:240–6.
[210] Williams KV, Price JC, Kelley DE. Interactions of impaired glucose transport and
phosphorylation in skeletal muscle insulin resistance. A dose-response assessment using
positron emission tomography. Diabetes 2001;50:2069–79.
[211] Choi WH, O’Rahilly S, Rees A, Morgan R, Flier JS, Moller DE. Molecular scanning of
the insulin-responsive glucose transporter (GLUT 4) gene in patients with non-insulin
dependent diabetes mellitus. Diabetes 1991;40:1712–8.
[212] Perriott LM, Kono T, Whitesell RR, Knobel SM, Piston DW, Granner DK, et al.
Glucose uptake and metabolism by cultured human skeletal muscle cells: rate-limiting
steps. Am J Physiol Endocrinol Metab 2001;281:E72–80.
[213] Printz RL, Ardehali H, Koch S, Granner DK. Human hexokinase II mRNA and gene
structure. Diabetes 1995;44:290–4.
[214] Mandarino LJ, Printz RL, Cusi KA, Kinchington P, O’Doherty RM, Osawa H, et al.
Regulation of hexokinase II and glycogen synthase mRNA, protein, and activity in
human muscle. Am J Physiol 1995;269:E701–8.
[215] Vogt C, Ardehali H, Iozzo P, Yki-Jarvinen H, Koval J, Maezono K, et al. Regulation of
hexokinase II expression in human skeletal muscle in vivo. Metabolism 2000;49:814–8.
[216] Pendergrass M, Koval J, Vogt C, Yki-Jarvinen H, Iozzo P, Pipek R, et al. Insulin-induced
hexokinase II expression is reduced in obesity and NIDDM. Diabetes 1998;47:387–94.
[217] Ducluzeau P-H, Perretti N, Laville M, Andreelli F, Vega N, Riou J-P, et al. Regulation
by insulin of gene expression in human skeletal muscle and adipose tissue. Evidence for
specific defects in type 2 diabetes. Diabetes 2001;50:1134–42.
[218] Lehto M, Huang X, Davis EM, Le Beau MM, Laurila E, Eriksson KF, et al. Human
hexokinase II gene: exon-intron organization, mutation screening in NIDDM, and its
relationship to muscle hexokinase activity. Diabetologia 1995;38:1466–74.
[219] Laakso M, Malkki M, Kekalainen P, Kuusito J, Deeb SS. Polymorphisms of the human
hexokinase II gene: lack of association with NIDDM and insulin resistance. Diabetologia
1995;38:617–22.
[220] Echwald SM, Bjorbaek C, Hansen T, Clausen JO, Vestergaard H, Zierath JR, et al.
Identification of four amino acid substitutions in hexokinase II and studies of
relationships to NIDDM, glucose effectiveness, and insulin sensitivity. Diabetes 1995;
44:347–53.
[221] Thiebaud D, Jacot E, DeFronzo RA, Maeder E, Jequier E, Felber JP. The effect of
graded doses of insulin on total glucose uptake, glucose oxidation, and glucose storage in
man. Diabetes 1982;31:957–63.
834 R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835

[222] Golay A, DeFronzo RA, Ferrannini E, Simonson DC, Thorin D, Acheson K, et al.
Oxidative and non-oxidative glucose metabolism in non-obese Type 2 (non-insulin
dependent) diabetic patients. Diabetologia 1988;31:585–91.
[223] Lillioja A, Mott DM, Zawadzki JK, Young AA, Abbott WG, Bogardus C. Glucose
storage is a major determinant of in vivo ‘insulin resistance’ in subjects with normal
glucose tolerance. J Clin Endocrinol Metab 1986;62:922–7.
[224] Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman RG. Quantitation of
muscle glycogen synthesis in normal subjects and subjects with non-insulin-dependent
diabetes by 13C nuclear magnetic resonance spectroscopy. N Engl J Med 1990;322:223–8.
[225] Rothman DL, Magnusson I, Cline G, Gerard D, Kahn CR, Shulman RG, et al. Decreased
muscle glucose transport/phosphorylation is an early defect in the pathogenesis of non-
insulin-dependent diabetes mellitus. Proc Natl Acad Sci USA 1995;92:983–7.
[226] Yki-Jarvinen H, Mott D, Young AA, Stone K, Bogardus C. Regulation of glycogen
synthase and phosphorylase activity by glucose and insulin in human skeletal muscle.
J Clin Invest 1987;80:95–100.
[227] Frame S, Cohen P. GSK3 takes centre stage more than 20 years after its discovery.
Biochem J 2001;359(Pt 1):1–16.
[228] Cohen P. The Croonian Lecture 1999. Identification of a protein kinase cascade of major
importance in insulin signal transduction. Proc R Soc Lond B Biol Sci 1999;354:485–95.
[229] Damsbo P, Vaag A, Hother-Nielsen O, Beck-Nielsen H. Reduced glycogen synthase
activity in skeletal muscle from obese patients with and without type 2 (non-insulin-
dependent) diabetes mellitus. Diabetologia 1991;34:239–45.
[230] Mandarino LJ, Wright KS, Verity LS, Nichols J, Bell JM, Kolterman OG, et al. Effects of
insulin infusion on human skeletal muscle pyruvate dehydrogenase, phosphofructokinase,
and glycogen synthase. Evidence for their role in oxidative glucose metabolism. J Clin
Invest 1987;80:655–63.
[231] Thorburn AW, Gumbiner B, Bulacan F, Wallace P, Henry RR. Intracellular glucose
oxidation and glycogen synthase activity are reduced in non-insulin-dependent (type II)
diabetes independent of impaired glucose uptake. J Clin Invest 1990;85:522–9.
[232] Vaag A, Henriksen JE. Beck-Nielsen Decreased insulin activation of glycogen synthase in
skeletal muscles in young non-obese Caucasian first-degree relatives of patients with non-
insulin-dependent diabetes mellitus. J Clin Invest 1992;89:782–8.
[233] Nyomba BL, Freymond D, Raz I, Stone K, Mott DM, Bogardus C. Skeletal muscle
glycogen synthase activity in subjects with non-insulin-dependent diabetes mellitus after
glyburide therapy. Metabolism 1990;39:1204–10.
[234] Pratipanawatr T, Cusi K, Ngo P, Pratipanawatr W, Mandarino LJ, DeFronzo RA.
Normalization of plasma glucose concentration by insulin therapy improves insulin-
stimulated glycogen synthesis in type 2 diabetes. Diabetes 2002;51:462–8.
[235] Vestergaard H, Lund S, Larsen FS, Bjerrum OJ, Pedersen O. Glycogen synthase and
phosphofructokinase protein and mRNA levels in skeletal muscle from insulin-resistant
patients with non-insulin-dependent diabetes mellitus. J Clin Invest 1993;91:2342–50.
[236] Vestergaard H, Bjocbaek C, Andersen PH, Bak JF, Pedersen O. Impaired expression of
glycogen synthase mRNA in skeletal muscle of NIDDM patients. Diabetes 1991;40:1740–5.
[237] Majer M, Mott DM, Mochizuki H, Rowles JC, Pederson O, Knowler WC, et al. Association
of the glycogen synthase locus on 19q13 with NIDDM in Pima Indians. Diabetologia 1996;
39:314–21.
[238] Orho M, Nikua-Ijas P, Schalin-Jantti C, Permutt MA, Groop LC. Isolatation and
characterization of the human muscle glycogen synthase gene. Diabetes 1995;44:1099–105.
[239] Bjorbaek C, Echward SM, Hubricht P, Vestergaard H, Hansen T, Zierath J, et al. Genetic
variants in promoters and coding regions of the muscle glycogen synthase and the insulin-
responsive GLUT4 genes in NIDDM. Diabetes 1994;43:976–83.
[240] Bjorbaek C, Fik TA, Echward SM, Yang P-Y, Vestergaard H, Wang JP, et al. Cloning of
human insulin-stimulated protein kinase (ISPK-1) gene and analysis of coding regions
R.A. DeFronzo / Med Clin N Am 88 (2004) 787–835 835

and mRNA levels of the ISPK-1 and the protein phosphatase-1 genes in muscle from
NIDDM patients. Diabetes 1995;44:90–7.
[241] Procharzka M, Michizuki H, Baier LJ, Cohen PTW, Bogardus C. Molecular and linkage
analysis of type-1 protein phosphatase catalytic beta-subunit gene: lack of evidence for its
major role in insulin resistance in Pima Indians. Diabetologia 1995;38:461–6.
[242] Schalin-Jantti C, Harkoenen M, Groop LC. Impaired activation of glycogen synthase in
people at increased risk for developing NIDDM. Diabetes 1992;41:598–604.
[243] Del Prato S, Bonadonna RC, Bonora E, Gulli G, Solini A, Shank M, et al.
Characterization of cellular defects of insulin action in type 2 (non-insulin-dependent)
diabetes mellitus. J Clin Invest 1993;91:484–94.
[244] Falholt K, Jensen I, Lindkaer Jensen S, Mortensen HB, Volund A, Heding LG, Norskov
Petersen P, Falholt W. Carbohydrate and lipid metabolism of skeletal muscle in type 2
diabetic patients. Diabet Med 1988;5:27–31.
[245] Mandarino LJ, Madar Z, Kolterman OG, Bell JM, Olefsky JM. Adipocyte glycogen
synthase and pyruvate dehydrogenase in obese and type II diabetic patients. Am J Physiol
1986;251:E489–96.
[246] Kelley D, Mokan M, Mandarino L. Intracellular defects in glucose metabolism in obese
patients with noninsulin-dependent diabetes mellitus. Diabetes 1992;41:698–706.
[247] Groop L, Bonadonna R, Simonson DC, Petrides A, Hasan S, DeFronzo RA. Effect of
insulin on oxidative and non-oxidative pathways of glucose and free fatty-acid
metabolism in human obesity. Am J Physiol 1992;263:E79–84.
[248] Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty acid cycle: its role in
insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet 1963;i:785–9.
Med Clin N Am 88 (2004) 837–846

The metabolic syndrome


Alan J. Garber, MD, PhDa,b,*
a
Departments of Medicine, Biochemistry and Molecular Biology, and
Molecular and Cellular Biology, Baylor College of Medicine,
6550 Fannin Street, Suite 1045, Houston, TX 77030, USA
b
Department of Endocrinology, Diabetes, and Metabolism,
The Methodist Hospital Houston, TX, USA

The metabolic syndrome is a collection of frequently associated cardio-


vascular risk factors that tend to aggregate in selected patient populations
and that, together, increase coronary and cardiovascular mortality, and total
mortality. The term ‘‘metabolic syndrome’’ has been used variously by
disparate organizations to connote different entities. Some, such as the World
Health Organization, have used this term to indicate a state of insulin
resistance with pervasive dysmetabolism secondary to the state of insulin
resistance. Conceptualization of the metabolic syndrome as a unique, high-
risk cardiovascular state, as defined by the National Cholesterol Education
Program (NCEP), is gaining acceptance as the basis for diagnosing the
metabolic syndrome [1]. The NCEP definition is based on clustering multiple
metabolic abnormalities associated with insulin resistance but does not
require a measure of insulin sensitivity. Table 1 cites the current criteria for
the diagnosis of the metabolic syndrome, according to NCEP guidelines. A
diagnosis is made by the presence of three or more abnormalities. Although
each abnormality may be associated with underlying insulin resistance, no
requirement exists for an experimental demonstration of insulin resistance or
a measurement of impaired insulin-mediated glucose disposal. Several of
these clinical features, such as low high-density lipoprotein (HDL) cholesterol
levels and elevations of triglyceride-rich lipoproteins in fasting serum speci-
mens, are well known to be interrelated. Furthermore, these conditions also
tend to be more prevalent in obese individuals as well as in those with essential
hypertension, making this syndrome increasingly prevalent in patients with
one or more traditional risk factors for cardiovascular disease.

* Baylor College of Medicine, 6550 Fannin Street, Suite 1045, Houston, TX 77030.
E-mail address: agarber@bcm.tmc.edu

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.001
838 A.J. Garber / Med Clin N Am 88 (2004) 837–846

Table 1
Definition of metabolic syndrome
Risk factor Defining level
Abdominal obesity
Men Waist >40 inches
Women Waist >35 inches
Triglycerides 150 mg/dL
HDL cholesterol
Men \40 mg/dL
Women \50 mg/dL
Blood pressure 130/85 mm Hg
Fasting glucose 110 mg/dL
NCEP criteria (any 3–5 symptoms).

The metabolic syndrome, as defined by the NCEP, is not to be confused


with the insulin resistance syndrome or the presence of underlying insulin
resistance, although such resistance is likely in patients with the metabolic
syndrome. Insulin resistance is an evolving entity, which may or may not be
associated with aspects of dysmetabolism secondary to prolonged insulin
resistance that appear later. A useful example of this evolution is the slow,
progressive decline of pancreatic b-cell function under conditions of ongoing
or worsening insulin resistance. This decline leads ultimately to hypergly-
cemia and a clinical diagnosis of type 2 diabetes. Many years are required
for the evolution of this process before diabetes can be diagnosed [2]. There
is, therefore, an antecedent period of compensated hyperinsulinemic
euglycemia in which patients clearly are insulin resistant but in whom
dysmetabolism of carbohydrates may be difficult to detect. A similar latency
may also exist with respect to the evolution of dyslipidemia secondary to
impaired clearance of triglyceride-rich lipoproteins or accelerated removal
or delayed synthesis of HDL cholesterol. For example, teenaged women
with polycystic ovary syndrome clearly have, as part of the underlying
pathogenesis of this disorder, an important element of insulin resistance.
However, at the time of diagnosis, these young women rarely have obvious
hypertriglyceridemia, low HDL cholesterol, or overtly abnormal carbohy-
drate metabolism. With prolonged persistence of unrelieved insulin re-
sistance, however, they do have a much higher likelihood of developing the
stigmata of the metabolic syndrome. Thus, it may therefore be possible that
the metabolic consequence of early b-cell failure in the presence of
prolonged insulin resistance leads to the abnormal metabolism required to
diagnose the metabolic syndrome.
A weaker association exists between insulin resistance and essential
hypertension. Various authors have concluded that approximately 50% to
75% of patients with essential hypertension have underlying insulin
resistance. Whether such resistance begins before the onset of hypertension
is somewhat unclear, although obesity is a clear risk factor for the
A.J. Garber / Med Clin N Am 88 (2004) 837–846 839

development of hypertension, and obesity is also a risk factor for the


development of insulin resistance. It is likely that the insulin resistance
precedes the hypertension in such cases, although not in all cases of essential
hypertension, even when hypertension is fully developed.
Although it is currently under reexamination, impaired fasting glucose (a
fasting glucose 110 mg/dL) is one of the five diagnostic criteria used for
determining the presence of the metabolic syndrome. In contrast, some
authors have suggested that patients with otherwise normal fasting glucose
levels may demonstrate impaired glucose tolerance at the point of a 2-hour
postglucose challenge, during an oral glucose tolerance test. Although
a recent consensus development conference of the American College of
Endocrinology recommended the inclusion of a 2-hour postglucose value of
greater than 140 mg/dL as a potential alternative to the 110-mg/dL fasting
glucose level, this addition has not been included in current definitions of the
metabolic syndrome because glucose tolerance testing is believed to be too
difficult for practitioners and potentially impractical in routine clinical
settings. However, physicians may wish to offer such testing where possible
to clarify patient status with the metabolic syndrome and to ascertain
potential impairments of carbohydrate metabolism or even overt diabetes.
Suggestions have been made to modify diagnostic criteria for the
metabolic syndrome by including elevated levels of highly sensitive C-
reactive protein as a potential criterion for the diagnosis of the metabolic
syndrome [3]. Although individuals with three or more criteria for metabolic
syndrome were clearly shown to have increased levels of C-reactive protein,
such elevations of C-reactive protein may not be independent of the
underlying insulin resistance, which is surely present in patients having
three or more diagnostic features of the metabolic syndrome. Thus, the use
of elevated C-reactive protein may overlap existing diagnostic components
and, therefore, may not add additional insight about the nature of any
underlying dysmetabolism. Instead, the use of highly sensitive C-reactive
proteins as an additional estimate of cardiovascular risk, independent of the
underlying diagnosis of the metabolic syndrome, should be retained as
potentially more useful than by its inclusion within the metabolic syndrome.
A diagnosis of the metabolic syndrome is, however, tantamount to a
diagnosis of excess cardiovascular risk, whether or not C-reactive proteins
are elevated. In the Nurses Health Survey, event-free survival in individuals
with the metabolic syndrome was clearly lower than in individuals without
the metabolic syndrome [2]; however, it should also be noted that
individuals with elevated C-reactive protein and a diagnosis of metabolic
syndrome had even lower rates of event-free survival than did individuals
with the metabolic syndrome but with low levels of C-reactive protein.
Thus, a diagnosis of the metabolic syndrome based on the criteria of the
NCEP is highly effective in predicting excess cardiovascular risk. For
example, in an analysis of the Framingham Study offspring, diagnosis of the
metabolic syndrome approximately doubled cardiovascular risk for males
840 A.J. Garber / Med Clin N Am 88 (2004) 837–846

and tripled such risk for females with the metabolic syndrome compared
with those without the metabolic syndrome. A greater adverse cardiovas-
cular impact of the metabolic syndrome was seen by Lakka et al [4] in the
Kuoppio Study (see [4]), in which a diagnosis of the metabolic syndrome
increased coronary artery mortality and cardiovascular mortality by three-
fold and doubled total mortality in this patient population. An even greater
cardiovascular risk was noted in the Mexico City Study.
The metabolic syndrome may also be an outstanding risk predictor of
future diabetes. In the Framingham Offspring Study, a recent analysis
suggested that a diagnosis of the metabolic syndrome predicted a fivefold
8-year future risk of diabetes in males and a sixfold increase in the future
risk of diabetes in females. Although the accuracy of the metabolic syndrome
as a predictor of excess risk of cardiovascular disease is not surprising, its
robust ability to predict an excess risk of future diabetes is, in fact,
somewhat unexpected. This degree of accuracy may provide insight into the
precise nature of the ‘‘metabolic syndrome.’’ Because it is likely that the
metabolic syndrome describes at least one aspect of underlying insulin
resistance (clearly the period in which multiple metabolic abnormalities have
evolved, likely secondary to b-cell failure resulting from an underlying
insulin resistance state), then the metabolic syndrome presents a slice in time
in the evolution of the insulin resistant state into an area of gross pathology,
such as type 2 diabetes or clinical atherosclerosis.
Therefore, the metabolic syndrome likely characterizes a state in time of
latent underlying disease evolution that may be viewed as preclinical
diabetes or preclinical atherosclerosis. Based in large measure on epidemi-
ologic analyses, such a diagnosis clearly represents a high probability
transition state in the evolution from normality to gross pathology resulting
from insulin resistance. It is important to recognize a transition state
because transition states are generally more reversible than end-stage
pathology. Alternatively, one may view the high predictability of the
metabolic syndrome of future type 2 diabetes as the likelihood that the
metabolic syndrome is already type 2 diabetes. This view derives from
the observation that glucose is a continuous variable that proceeds from
completely normal to completely abnormal, with the diagnosis of diabetes
being a somewhat arbitrary event in the middle of this evolution. Prior
expert committees have set arbitrary diagnostic criteria for type 2 diabetes at
levels of 126 mg/dL, based in part on the recognized increase in the
development of the typical or classic microvascular complications of
diabetes, such as retinopathy. Nonetheless, although it is an expert opinion,
the diagnosis point for diabetes remains arbitrary. This is not to say that
diabetes may not occur at an earlier point or at lower glucose levels before
the development of very substantial rates of hyperglycemia. As seen in the
special analysis of the NHANES III database, there was a fivefold increase
in the incidence of retinopathy in the fasting glucose interval of 110 to 119
mg/dL compared with the interval of 100 to 109 mg/dL. It is therefore
A.J. Garber / Med Clin N Am 88 (2004) 837–846 841

possible that much lower levels of fasting hyperglycemia may be associated


with milder degrees of type 2 diabetes than the level currently used for the
diagnosis of type 2 diabetes (126 mg/dL). Similarly, the United Kingdom
Prospective Diabetes Study (UKPDS) used a fasting glucose level of 108
mg/dL to diagnose diabetes and encountered a 21% incidence of retinop-
athy in their newly diagnosed patients [5]. The metabolic syndrome may
overlap, to a variably greater extent, the diagnosis of type 2 diabetes and
may ultimately be replaced by a more aggressive diagnosis of this disorder.
The prevalence of the metabolic syndrome is truly staggering. Although
there may be an epidemic of type 2 diabetes in the United States, the
proportion of Americans with the metabolic syndrome is far larger. Using the
criteria of the NCEP and applied to the results of the NHANES III database
updated to the 2000 census, Ford et al [6] estimated that approximately 47
million Americans met the current criteria of the metabolic syndrome. As high
as 40% of the population 60 years and older have the metabolic syndrome.
The syndrome is higher in minority patients compared with white patients.
The number of Americans with the metabolic syndrome clearly exceeds the
number of Americans with type 2 diabetes and those with impaired glucose
tolerance. Thus, the number at risk for the future development of type 2
diabetes is much greater than previously suspected. Indeed, the striking
prevalence of the metabolic syndrome in the US population, together with its
extraordinary predictability of future diabetes, may be a useful rationale for
aggressive interventions with regard to treatment of the metabolic syndrome,
not only as a high-risk cardiovascular state, but also, as a serious premonitory
state of type 2 diabetes. Either or both of these views clearly justifies
intervention, both with regard to hygienic measures, such as lifestyle
modification including diet and exercise, together with the possibility of
pharmacologic intervention to interdict the evolution to type 2 diabetes and
to mitigate the excess risk of cardiovascular events.

Obesity
The foundation of the metabolic syndrome is excess deposition of
adipose tissue in visceral or abdominal compartments, which gives rise to
an underlying insulin resistant state. Although the NCEP definition of the
metabolic syndrome does not directly measure insulin sensitivity, much of
the diagnostic criteria outlined previously are associated with the dysme-
tabolism of insulin resistance. Insulin resistance may be directly addressed
by lifestyle modification. Such modifications include regular, vigorous
aerobic exercise as frequently as possible. The insulin sensitizing benefits
of exercise require, for example, five or more sessions per week of
approximately 3 to 5 miles per day, consisting of a brisk walking pace of
3 to 4 miles per hour. Light-weight training is also useful in supporting
muscle bulk and assisting with weight reduction. Of course, before
842 A.J. Garber / Med Clin N Am 88 (2004) 837–846

beginning regular daily exercise, patients with high risk of coronary disease
should undergo exercise cardiac testing to exclude subclinical or atypical
coronary ischemia.
Exercise alone without aggressive calorie restriction is usually insufficient
to produce weight loss. Indeed, exercise may increase appetite, which may
more than compensate for the increased caloric expenditure and oxygen
consumption associated with the exercise. For that reason, aggressive
dietary modification with limitation of concentrated sweets, saturated fats,
and reduction of portion sizes seems essential if weight reduction is to be
achieved. Weight reduction to achieve a body mass index (BMI) less than 30
is essential if the metabolic syndrome is to be modified. Nonobese but
overweight individuals have a substantially reduced risk of the metabolic
syndrome. Of course, reduction of overweight individuals to normal weight
(27 BMI) would be ideal, but this level may be difficult to achieve in the
markedly obese population in whom even small degrees of weight loss
would be beneficial for many of the parameters of the metabolic syndrome,
especially the hypertriglyceridemia and hyperglycemia.
Lifestyle modification with diet and exercise greatly magnifies the effect of
pharmacologic agents designed to intervene in dyslipidemia, hypertension,
or abnormal carbohydrate tolerance. Because the metabolic syndrome is
also a high-risk predictor of future diabetes, intervention to prevent the
development of overt type 2 diabetes seems appropriate. The role of diet and
exercise as part of a thorough lifestyle modification program is unsurpassed
for prevention of type 2 diabetes in patients with impaired glucose tolerance.
Although they are not strictly the same as the patient population with the
metabolic syndrome, patients in the Diabetes Prevention Program (DPP)
showed a 58% reduction in the risk of development of type 2 diabetes by
a 7% loss of body weight and regular daily exercise. Pharmacologic
therapies were also useful for the prevention of type 2 diabetes because
these therapies may have reduced the underlying insulin resistance. These
therapies, however, will be discussed later as part of future considerations in
the management of the metabolic syndrome. Pharmacologically assisted
weight loss may also be considered. Orlistat (Xenical) therapy reduces fat
absorption, produces a 3% to 8% weight loss, and reduces the development
of type 2 diabetes. It should be considered for those failing life style
modifications.

Treatment of dyslipidemia
The metabolic syndrome does not necessarily include hypercholesterol-
emia, as characterized by a markedly increased mass of low-density
lipoprotein (LDL) cholesterol. Instead, a shift in the size distribution of
LDL particles toward a smaller, denser and inherently more atherogenic
LDL particle, as the result of the underlying insulin resistance state, seems
A.J. Garber / Med Clin N Am 88 (2004) 837–846 843

present in many of these patients. Diagnosis that focuses on hypertriglycer-


idemia and low HDL cholesterol levels aids in recognizing that these factors
traditionally herald the presence of a severe, underlying redistribution of
LDL particle size toward a smaller, more atherogenic particle size. Because
patients with the metabolic syndrome represent a high-risk cardiovascular
state, precise quantitation of this state by assessment of the 10-year
cardiovascular risk using the Framingham risk engine (found on the
NHLBI website) is most useful in determining appropriate action regarding
reduction of LDL mass. Risks greater than 20% per 10 years equate to an
extremely high risk state, equivalent to the risk of known coronary disease,
and therefore merit reduction of LDL cholesterol below 100 mg/dL, using
interventions with statins. Use of a statin should not be predicated on
a titration of slight elevations to just below 100 mg/dL. Instead, if the
decision is made to intervene with a b-hydroxy-b-methylglutaryl-coenzyme
A (HMG-CoA) reductase inhibitor, then a full dose, designed to produce
a 30% to 40% reduction of LDL mass, such as 40 mg of pravastatin or
simvastatin, should be used. This recommendation reflects, in part, results
from randomized controlled trials demonstrating that such doses custom-
arily produce a significant reduction in coronary events of approximately
30% or more and also the recognition that in high-risk patients with
minimally elevated LDL levels there may likely be an altered particle
distribution of LDL cholesterol, which may merit a somewhat more
aggressive therapeutic intervention. For example, in the Heart Protection
Study, patients with LDL cholesterol less than 100 mg/dL at the time of
randomization benefited equally from 40 mg of simvastatin, with pro-
portionate reductions in coronary event reduction, as did those patients with
substantially higher levels of LDL cholesterol at the time of randomization
to simvastatin.
Once LDL goals have been attained, then residual elements of the
dyslipidemia should be addressed. The second treatment goal for the
management of dyslipidemia should be correction of low levels of HDL
cholesterol. This goal may be attained by lowering triglycerides with a fibric
acid derivative if these triglycerides are substantially elevated (250 mg/dL).
Previously, the use of gemfibrozil was encouraged, owing to the publication
of two large-scale interventional trials demonstrating coronary event
reduction with gemfibrozil. Recently, however, it has been recognized that
gemfibrozil inhibits the glucuronidation of many statins thereby raising
statin blood levels, resulting in elevated serum levels of such statins. This
may considerably complicate safety considerations in the use of gemfibrozil
in patients already treated with HMG-CoA reductase inhibitors. Thus, in
such patients, fenofibrate (Tricor) is preferred. The smallest possible dose
that produces acceptable modification of triglyceride levels should be used.
In those patients with low HDL cholesterol (40 mg/dL in males; 50
mg/dL in females) in whom triglycerides are not markedly elevated, direct
intervention with nicotinic acid may be useful. Generally, for every 1 g of
844 A.J. Garber / Med Clin N Am 88 (2004) 837–846

nicotinic acid administered, a 10% increase in HDL cholesterol can be


expected. Nicotinic acid, used as part of a treatment program with statins,
has produced remarkable event reductions in trials such as HATS.
Presently, combination therapy tablets using lovastatin and slow-release
nicotinic acid are available and prove effective for correcting dyslipidemia in
patients with the metabolic syndrome. Treatment with nicotinic acid should
be initiated and titrated slowly to avoid precipitating overt symptomatology
secondary to the vasoreactive effects of nicotinic acid or to suddenly
decompensate glycemic control owing to the increased insulin resistance
resulting from nicotinic acid therapy. If carbohydrate tolerance deteriorates
during nicotinic acid therapy, antidiabetic therapy should be instituted early
and aggressively to restrain hyperglycemia from complicating the underlying
metabolic syndrome or from producing a recrudescence of the hyper-
triglyceridemia or the low HDL cholesterol. In patients with overt
hyperglycemia, thiazolidinediones are preferred for correction of hypergly-
cemia in patients with low HDL cholesterol. Nicotinic acid should be
withheld until at least 6 months of antidiabetic therapy has been completed
with thiazolidinediones, such as rosiglitazone or pioglitazone. Biguanides,
such as metformin, may also be useful in improving insulin sensitivity and
preventing weight gain in patients on nicotinic acid and thiazolidinediones.
Combination therapy products of both rosiglitazone and metformin may be
especially useful as initial treatment for new onset hyperglycemia in such
patients, with secondary adverse effects of hyperglycemia as the consequence
of nicotinic acid therapy.

Treatment of hypertension
Numerous agents have been established as excellent first-line therapies
for hypertension in patients with diabetes and in nondiabetic patients with
dyslipidemia. Ramipril, used in the HOPE Study, showed excellent
antiatherogenic effects in diabetic patients with one additional coronary
risk factor and in patients with coronary disease. Captopril and atenolol
were equally effective for macrovascular risk reduction in patients with
diabetes in the UKPDS. Calcium channel blockers such as felodipine
(Plendil) also proved effective in diabetic patients in the HOT Trial. Finally,
thiazide diuretics were found to be useful in the ALLHAT Study. Thus,
a variety of agents can be chosen for patients with diabetes. Because most
patients with type 2 diabetes evolve from a preexisting metabolic syndrome,
the use of antihypertensive agents might be expected in patients with the
metabolic syndrome, as they have been proven to be efficacious in patients
with diabetes. Because endothelial dysfunction is an important part of the
underlying insulin resistance state, which appears to be widely present in the
metabolic syndrome, angiotensin-converting enzyme inhibitors and aldo-
sterone receptor antagonists are useful in improving hypertension and
A.J. Garber / Med Clin N Am 88 (2004) 837–846 845

relieving endothelial dysfunction in this population. Other agents may be


equally efficacious in improving hypertension but may lack the ability to
improve insulin resistance or endothelial dysfunction. These agents, in-
cluding thiazide diuretics, b-blockers, cardioselective b-blockers, and
potassium channel blockers, should be reserved for second and third line
(or beyond) therapies. Part of the clinical issue in patients with diabetes is
that the treatment target for antihypertensive therapy is a diastolic pressure
less than 80 mm Hg. Attainment of such a target, which probably should
extend to patients with prediabetic states, such as the metabolic syndrome, is
the need for multiple antihypertensive agents, including an average of three,
needed for diabetic patients in the United States, and at least two or more in
the UKPDS. Thus, much of the discussion revolving around which one of
a number of the foregoing antihypertensive agents should be considered the
best antihypertensive agent in the metabolic syndrome or other prediabetic
states is rendered moot and obviated by the fact that most of these agents
will be required in combination to achieve the kinds of blood pressure
targets now envisioned for such patients. One should therefore consider
beginning with an angiotensin-converting enzyme or angiotensin II receptor
blocker. A rapid dose escalation to include b-blockers, thiazide diuretics,
calcium channel blockers, and other agents should be introduced to gain
rapid control of the hypertensive state and improve endothelial dysfunction
in these patients.

Treatment of impaired carbohydrate tolerance


The central diagnostic point of abnormal carbohydrate tolerance used in
the NCEP 2001 Guidelines is a fasting glucose level greater than 110 mg/dL.
This is known as impaired fasting glucose and may or may not be
accompanied by impaired glucose tolerance (2-hour postchallenge glucose
140–199 mg/dL). Although impaired glucose tolerance is clearly associated
with the presence of underlying insulin resistance, impaired fasting glucose
may or may not be associated with insulin resistance. It is, instead, more
likely to be associated with declining b-cell function. There are no
prospective randomized trials suggesting that pharmacologic therapy of
impaired fasting glucose yields treatment benefits either with regard to
prevention of adverse coronary outcomes or the prevention of progression
to type 2 diabetes. Nonetheless, it seems clear that pharmacologic
modification may present a successful approach to this particular health
problem because pharmacologic intervention with regard to impaired
glucose tolerance has been demonstrated to be successful with regard to
the prevention of type 2 diabetes. Studies such as the DPP and the TRIPOD
studies have demonstrated efficacy of both metformin, in the former, and
a thiazolidinediones in the latter, with regard to diabetes prevention. The
thiazolidinediones therapy arm of the DPP study showed a striking,
prolonged benefit of thiazolidinediones therapy. It would therefore suggest
846 A.J. Garber / Med Clin N Am 88 (2004) 837–846

that in patients failing beneficial lifestyle modifications or having insufficient


benefit from lifestyle interventions, chemoprevention of diabetes may be
warranted. On the other hand, it is completely unknown whether chemo-
prevention of diabetes with metformin or thiazolidinediones, or both, would
have a beneficial effect with regard to the adverse cardiovascular outcomes
associated with the metabolic syndrome. Future studies to clarify this
important point are clearly necessary.

Summary
The metabolic syndrome is a collection of cardiovascular risk factors that
denote a high-risk multifactorial adverse cardiovascular state, which is
largely the result of obesity and the resulting insulin resistance. It is
diagnosed by the presence of multiple risk factors, such as hypertriglycer-
idemia, low HDL cholesterol, hypertension, essential hypertension, abnor-
mal fasting glucose levels, and abdominal or visceral obesity. This state also
denotes a high-risk state for the evolution to type 2 diabetes. Treatment for
the metabolic syndrome should be focused primarily on lifestyle modifica-
tion, with reduction of the underlying obesity and insulin resistance. The
treatment of individual coronary risk factors is clearly warranted in such
patients because the metabolic syndrome represents a high-risk cardiovas-
cular state equal to 20% or greater 10-year risk of coronary disease. In such
patients, it may be possible that impaired glucose tolerance also exists. This
diagnosis requires a glucose tolerance test to be performed. Chemopreven-
tion of deterioration of impaired glucose tolerance to a future diabetic state
has been successful with metformin or thiazolidinediones, as well as with
aggressive lifestyle modification. Indeed, combination of both chemo-
prevention and lifestyle modification may prevent future cases of diabetes
if instituted early in the course of diagnosis of impaired glucose tolerance.
Whether such treatments benefit the adverse cardiovascular outcomes
remain to be decided.

References
[1] Executive summary of the Third Report of the National Cholesterol Program (NCEP)
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in
Adults. JAMA 2001;285:2486–97.
[2] Lebovitz H. Diabetes Reviews 1999.
[3] Ridker PM, et al. Circulation 2003;107:391–7.
[4] Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, et al.
The metabolic syndrome and total and cardiovascular disease mortality in middle-aged
men. JAMA 2002;288:2709–16.
[5] United Kingdom Prospective Diabetes Study No. 34. Lancet 1998.
[6] Ford ES, Giles WH, Dietz WH. Prevalence of the metabolic syndrome among US adults:
findings from the third National Health and Nutrition Examination Survey. JAMA 2002;
287:356–9.
Med Clin N Am 88 (2004) 847–863

Oral antidiabetic agents: 2004


Harold E. Lebovitz, MD*
Department of Medicine, State University of New York Health Science Center at Brooklyn,
450 Clarkson Avenue, Brooklyn, NY 11203, USA

The 1980s and 1990s were marked by unprecedented advances in


knowledge about the pathogenesis of type 2 diabetes and its chronic
complications. Landmark intervention studies in type 2 diabetic patients
proved that intensive control of glycemia, blood pressure, and plasma lipids
significantly decreased microvascular and macrovascular complications
[1–4]. New drugs were introduced that decreased blood glucose, blood
pressure, and low-density lipoprotein cholesterol levels. The Steno II
research study performed from 1993 through 2001 showed that the use of
contemporary treatments to implement aggressive blood glucose, blood
pressure, plasma lipid, and antithrombotic target goals in type 2 diabetic
patients with microalbuminuria resulted in a greater than 50% reduction in
clinically relevant macrovascular and microvascular complications [5].
Given these impressive advances it is somewhat surprising that several
recent publications comparing the characteristics and treatment profiles of
diagnosed type 2 diabetic populations in the National Health and Nutrition
Examination Surveys of 1988 to 1994 and 1999 to 2000 show a worsening of
the characteristics and no significant improvement in either glycemic control
or total metabolic control [6,7]. The difference between the two surveys
shows that type 2 diabetes has increased in males more than females, is
occurring at a younger age, and with a body mass index that is higher (Table
1). Treatment has changed in that diet-only therapy decreased from 27.4%
to 20.2% of the type 2 diabetic population. Insulin-only therapy decreased
from 24.2% to 16.4%, oral antidiabetic agents increased from 45.4% to
52.5%, and combination oral antidiabetic agent plus insulin increased from
3.1% to 11%. Adequate glycemic control as estimated by hemoglobin (Hb)
A1c less than 7% decreased from 44.5% of the population in 1988 to 1994 to
35.8% in 1999 to 2000.

* 416 Henderson Avenue, Staten Island, NY 10310.


E-mail address: hlebovitz@attglobal.net

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.05.002
848 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

Table 1
Comparison of characteristics of adult diabetic patients 20 years and older in NHANES III
(1988–1994) and NHANES 1999–2000
NHANES III NHANES 1999–2000
Prevalence of diabetes 5.4 6.1
Age standardized (%)
Men (%) 43.2 50
BMI (kg/m2) 29.9 32.3
Age at diagnosis of diabetes 50.7 46.7
Race
Non-Hispanic white (%) 74.6 59.8
Other races (%) 25.4 40.2
Abbreviation: BMI, body mass index.
Data from Saydah SH, Fradkin J, Cowie CC. Poor control of risk factors for vascular
disease among adults with previously diagnosed diabetes. JAMA 2004;291:335–42; and Koro
CE, Bourgeois N, Bowlin SJ, Fedder DO. Glycemic control from 1988 to 2000 among US
adults diagnosed with type 2 diabetes. Diabetes Care 2004;27:17–20.

More detailed analyses of the National Health and Nutrition Examina-


tion Surveys 1999 to 2000 data set show that only 37% of the diabetic
population had a HbA1c less than 7% with 63% having a HbA1c greater
than 7% and 37.2% one greater than 8% [7]. In that same survey 40.4% had
a blood pressure greater than 140/90 mm Hg; 51.8% had a total cholesterol
greater than or equal to 200 mg/dL. Only 7.3% of the diabetic population
had HbA1c less than 7%, blood pressure less than 130/80 mm Hg, and total
cholesterol less than 200 mg/dL.
Understanding the causes responsible for this failure to improve the
metabolic regulation of type 2 diabetic patients is critical to designing more
effective treatment strategies. Table 2 delineates the causes into patient-
related, health care provider–related, disease-related, and therapeutic agent–
related. Patient-related causes include the enormous increase in obesity that
has occurred in the last 10 years [8] and the large problem with patients’
failure to comply with treatment programs designed to achieve metabolic
regulation. Health care provider causes include failure to recognize the
importance of early aggressive treatment and a lack of understanding of the

Table 2
Causes responsible for failure to improve metabolic control in type 2 diabetic patients
Responsible cause Example
Patient Increasing obesity of the population
Difficulty in compliance with complicated regimens
Health care provider Lack of understanding importance of early aggressive control
Failure to understand significance of multiple defects
Disease Progressive beta cell failure
Therapeutic agents Inability to restore normal physiology
Side effects
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 849

interaction of the pathogenic mechanisms responsible for type 2 diabetes,


which dictate that combination drug therapy is almost essential to achieve
glycemic and blood pressure control. A recent report from Kaiser-
Permanente Northwest noted that a nonrandomized retrospective 1994 to
2002 database analysis identified that it took an average of 14 months before
type 2 diabetic patients who were on metformin monotherapy and had
HbA1c greater than 8% were switched or had additional therapies added to
their regimen [9]. For patients taking sulfonylurea monotherapy the average
time was 20 months. The most important disease-related cause is the
progressive nature of the loss of beta cell function, which seems to be
characteristic of type 2 diabetes [10]. Therapeutic agent–related causes
include the inability of many agents to modify metabolic processes to restore
normal physiologic regulation [11] and the occurrence of significant side
effects that limit the use of many agents [12].

The importance of early aggressive treatment


The major pathogenetic mechanisms that are responsible for the
metabolic abnormalities of type 2 diabetes are insulin resistance, deficient
insulin secretion, and glucose and lipid toxicity. Time course studies have
shown that insulin resistance occurs early and although subtle changes in
beta cell function are noted quite early, clinically significant insulin secretory
inadequacy follows several years later [13]. Glucose and lipid toxicity further
contribute to insulin resistance and beta cell insufficiency after derangement
of intermediary metabolism has occurred.
Several clinical observations dictate that both the clinical course of type
2 diabetes and the development of microvascular complications are most
favorably altered by very early and aggressive treatment of both the insulin
resistance and the hyperglycemia. Several intervention studies in individuals
with impaired glucose tolerance have shown that decreasing the compen-
satory hyperinsulinemia presumably by reducing insulin resistance or
lowering postprandial plasma glucose rises can delay or prevent the
development of new-onset type 2 diabetes [14–17]. In the one study
(TRIPOD study), the data show that reducing insulin resistance markedly
decreased the rate of loss of beta cell function [16]. These data suggest that
early interventions that decrease the stress of increased insulin secretion on
the beta cell can delay or prevent the progressive decline of beta cell function
in type 2 diabetic patients. Preservation of endogenous insulin secretory
function is associated with better glycemic control. The initial treatment of
type 2 diabetic patients with insulin resistance should be to decrease the
insulin resistance.
Hyperglycemia is the primary cause of microvascular complications in
all diabetic patients. Both the Diabetes Control and Complication Trial
(DCCT) in type 1 diabetic patients and the United Kingdom Diabetes
850 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

Prevention Study (UKPDS) in type 2 diabetic patients showed that


intensive glycemic control compared with ordinary glycemic control
significantly reduced clinically relevant retinopathy, nephropathy, and
neuropathy [1,18]. For every 1% decrease in HbA1c there was approxi-
mately a 30% decrease in microvascular complications. It is apparent in
examining the time course of the development of the microvascular
complications that the intensive treatment had to be in place for 3 to 4
years in the DCCT or 9 years in the UKPDS before any benefit became
evident [1,18]. These data suggested that the damage caused by early
hyperglycemia takes several years before its impact lessens. The concept
that the vascular bed has a memory of its previous exposure to blood
glucose levels and that this memory plays a major role in the blood vessels’
biology for many subsequent years has been confirmed by the Epidemi-
ology of Diabetic Complications Study [19].
The Epidemiology of Diabetic Complications Study is a long-term
follow-up of the patients who participated in the DCCT study. At the
conclusion of the DCCT both the intensively treated and the ordinarily
treated type 1 diabetic cohorts returned to their primary care centers for
their long-term care. Most of the patients in both cohorts agreed to return to
the research centers once a year to have examinations for HbA1c and retinal
and kidney function. Within 2 or 3 years of care in their primary care centers
the two cohorts had mean HbA1c of approximately 8%. That is, the
intensively treated group could not continue to achieve mean HbA1c values
of 7.1% to 7.2% outside of the intensive research setting. The ordinary
control cohort was allowed to be more aggressive in their management and
they improved from their previous mean HbA1cof 9%. In years 4 to 7 both
cohorts had mean HbA1c of approximately 8.1%. Despite the equal HbA1c,
the rate of progression of retinopathy continued to be twofold greater and
nephropathy sixfold greater in follow-up years 5 and 6 in the former
ordinary control cohort than the former intensively treated cohort. These
data strongly support the concept that early aggressive glycemic control is
essential to maximally reduce microvascular complications.

Strategies for treating hyperglycemia in type 2 diabetic patients


Management of the metabolic abnormalities of patients with type 2
diabetes involves correction of hyperglycemia, blood pressure, lipid
abnormalities, and the components of the metabolic syndrome. This article
deals selectively with the management of hyperglycemia by oral agents,
although it is obvious that such management interfaces with combination
and insulin treatment and must take into account their associated effects on
the other metabolic abnormalities. This being the case, the following
principles must be considered when implementing any treatment program
for hyperglycemia.
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 851

1. Type 2 diabetes needs to be diagnosed as early as possible, preferably


in the presymptomatic phase by screening if possible.
2. Insulin resistance or the presence of the metabolic syndrome should
be an indication to start treatment. Lifestyle modification should
be the first line of intervention. Pharmacologic treatment can be
considered if impaired glucose tolerance is present and should be
implemented if type 2 diabetes is diagnosed and lifestyle modification
has been inadequate to normalize glycemic control.
3. Combination therapy with oral agents that have different modes of
action should be initiated in type 2 diabetic patients if and when the
HbA1c exceeds 6.5%.
4. Most oral antidiabetic agents provide 75% or more of their antihy-
perglycemic activity at 50% of the maximally recommended dose.
Because side effects are frequently dose related, the use of
submaximal doses of two agents with different modes of actions
achieves better glycemic control with fewer side effects than maximal
doses of a single agent.
5. In choosing oral agents in treating hyperglycemia, consideration
should be given in each patient to their potential nonglycemic
metabolic effects.
6. If a combination of two oral agents is insufficient to achieve a target
glycemic goal of HbA1c less than or equal to 7%, addition of a third
oral agent is unlikely to achieve it if the HbA1c is greater than or
equal to 8% to 8.5%. If the HbA1c exceeds 8% to 8.5% the addition
of basal insulin before the evening meal or at bedtime or neutral
protamine Hagedorn insulin at bedtime is preferable.
7. The percent decrease in HbA1c with every oral therapy is directly cor-
related with the baseline HbA1c when the treatment is started. That is,
the greatest drop occurs with the highest baseline values and the lowest
decrease with the lowest baseline values.

Specific oral therapies: agents to treat insulin resistance


Two classes of oral antidiabetic agents reduce insulin resistance:
biguanides (metformin) and thiazolidinediones (pioglitazone and rosiglita-
zone) [19]. These agents decrease insulin resistance by different mecha-
nisms and have preferential effects on different tissues [20–22]. The
thiazolidinediones act primarily on adipose tissue to decrease the meta-
bolic consequences of obesity. It seems that increases in visceral adipose
tissue result in excessive release of free fatty acids and cytokines (eg, tumor
necrosis factor-a) and reduce release of the adipose tissue hormone
adiponectin. The result is altered hepatic metabolism and interference in
the transmission of the insulin signal within insulin-sensitive cells. These
effects are responsible for the insulin resistance associated with obesity.
The thiazolidinediones act on adipose tissue to decrease plasma free fatty
852 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

acids, decrease tumor necrosis factor-a secretion, and improve adiponectin


secretion, all of which improve insulin action in muscle, adipose tissue, and
liver. Metformin increases insulin action directly in insulin-sensitive cells
by a still undetermined mechanism. Metformin exerts a major effect in
increasing insulin sensitivity in liver and only a minor effect in skeletal
muscle [23]. Thiazolidinediones have a major effect in increasing insulin
sensitivity in muscle and a somewhat lesser effect in liver [23]. Overall
thiazolidinediones are about 70% more effective as a total body insulin
sensitizer than metformin [24].
Improvement in insulin sensitivity in insulin-resistant type 2 diabetic
patients allows endogenous insulin to be more effective and treatment of
such patients with insulin sensitizers improves glycemic control. The
magnitude of improvement depends on the amount of beta cell function
that is present at the time therapy is started. The mean decrease in HbA1c
that occurs with either metformin or thiazolidinedione monotherapy in
patients with a mean baseline HbA1c of approximately 9% is 1.5% [25].
About 30% of such patients achieve a HbA1c less than or equal to 7% on
maximal doses of sensitizing drugs. In the remainder of the patients, insulin
secretory function is too low and either insulin secretagogues or insulin
replacement needs to be added to the insulin sensitizer to achieve target
glycemic goals. The addition of insulin secretagogues to insulin sensitizers
can decrease HbA1c an additional 1% to 1.4% [26]. The addition of an
insulin sensitizer to an insulin treatment regimen can decrease HbA1c by
a mean of 0.8% to 1.2% [27] but the amount is influenced by how much the
insulin dose is reduced.
There are several significant advantages to using insulin sensitizers as part
of the therapeutic treatment program. As monotherapy, thiazolidinediones
or metformin do not cause serious hypoglycemia because endogenous
insulin secretion is still glucose dependent. Even therapy combining
a thiazolidinedione and metformin is rarely associated with hypoglycemia
[28]. Treatment of insulin-resistant type 2 diabetic patients with insulin
sensitizers improves many of the components of the metabolic syndrome
[20,21]. The effects of metformin and thiazolidinediones on the components
of the metabolic syndrome differ significantly as shown in Table 3. The
effects of the two thiazolidinediones, pioglitazone and rosiglitazone, seem to
be quite similar. There have been some reports suggesting that pioglitazone
has a greater effect in reducing plasma triglycerides than does rosiglitazone;
however, these have not been well-designed studies and there is need for
a double-blind, randomized head-to-head comparator study to validate
whether there is any significant difference. Metformin is the only drug used
to treat hyperglycemia that has been shown to reduce macrovascular
complications of type 2 diabetes [2]. Metformin treatment of overweight
type 2 diabetic patients in the UKPDS reduced cardiovascular-related
deaths (42%) and myocardial infarctions (39%). Sulfonylurea treatment
and insulin treatment had no significant benefits on those cardiovascular
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 853

Table 3
Comparison of the effects of thiazolidinediones and metformin on insulin resistance and the
components of the Metabolic Syndrome
Activity Metformin Thiazolidinediones
Glycemic control
FPG ## ##
HbA1c ## ##
FPI # ##
Body weight # "
Visceral fat # 0
Insulin sensitivity
Peripheral  ""
Liver "" "
Dyslipidemia
LDL cholesterol  "
LDL particle size 0 "
HDL cholesterol  ""
Triglyceride  #
Lipoprotein (a) 0 "
FFA  ##
Endothelial function
Vasodilation " ""
Blood pressure 0 #
Adhesion molecules #
Muscle proliferation #
Procoagulant state
PAI-1 # #
Fibrinogen
Inflammation
C-reactive protein # ##
Mesangial function
Microalbuminuria 0 #
Abbreviations: FFA, free fatty acid; FPG, fasting plasma glucose; FPI, fasting plasma
insulin; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PAI, plasminogen
activator inhibitor. 0, no effect; #, decrease; ##, marked decrease; ", increase; "", marked
increase.
Data from Lebovitz HE, Banerji MA. Treatment of insulin resistance in diabetes mellitus.
Eur J Pharmacol 2004;490:135–46.

events even though they lowered HbA1c the same 0.6% as metformin.
Clinical trials examining the effects of thiazolidinedione treatment on the
development of clinical cardiovascular events in type 2 diabetic patients are
in progress.
There are additional major differences between the effects of thiazoli-
dinediones and metformin. Metformin treatment is usually associated with
some weight loss. Most clinical studies show a mean weight loss with
metformin treatment of approximately 1 to 2 kg [29]. Thiazolidinedione
treatment is usually associated with a mean weight gain of 1.5 kg at
854 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

intermediate doses and 3.5 kg at high doses [30]. The side effects of
metformin include abdominal discomfort and diarrhea, which occur in
10% to 20% of patients on maximal doses, and a very rare occurrence of
lactic acidosis [29]. The gastrointestinal symptoms can be minimized by
initiating metformin therapy with 500 mg with the evening meal,
increasing the dose to 500 mg twice a day, and slowly titrating the dose
to 1000 mg twice a day with meals. The longer-acting preparations, such
as Glucophage XR, cause less gastrointestinal side effects. The few cases
of metformin-induced lactic acidosis that have been reported occurred in
individuals with decreased renal function, symptomatic heart failure on
therapy, or an underlying metabolic acidosis [31]. The major side effects
of thiazolidinedione treatment are an increase in adipose tissue and fluid
retention. The increase in adipose tissue is selective for the subcutaneous
region [32]. The visceral adipose tissue either is unchanged or decreases
slightly. The fluid retention manifests itself as mild to moderate peripheral
edema, which occurs in 4% of patients on monotherapy [33]. When
thiazolidinediones are combined with insulin, edema is seen in approxi-
mately 15% of the patients [33]. A rare type 2 diabetic patient on
thiazolidinediones develops heart failure. Such patients are usually older,
are on maximal doses of the thiazolidinedione, are also taking insulin,
and have a prior history of cardiovascular disease or renal failure [33].
The mechanism seems to be the increase in plasma volume in type 2
diabetic patients who had asymptomatic compensated heart failure.
Treatment of the fluid retention with loop diuretics is of limited benefit.
Angiotensin-converting enzyme inhibitors and aldosterone antagonists
have been used with limited success to treat the fluid retention [33].
According to a recent consensus panel convened by the American
Diabetes and American Heart Associations patients who are at risk for
developing congestive heart failure and could benefit from thiazolidine-
dione treatment should be started on very low doses, titrated up slowly,
and have careful monitoring of body weight and edema [33]. If excess
weight gain or edema is noted despite attempts to treat it, the
thiazolidinedione should be discontinued.
Metformin is administered twice a day and the maximal benefits on
glycemic control are seen at 2000 mg/d. Pioglitazone is administered once
daily in doses of 15, 30, or 45 mg/d. Rosiglitazone is administered either
once or twice a day in doses ranging from 2, 4, or 8 mg/d.

Specific oral therapies: agents to increase insulin secretion


In insulin-resistant type 2 diabetes, hyperglycemia occurs when insulin
secretion is inadequate to overcome the degree of insulin resistance that
is present. The natural history of the development of type 2 diabetes is
a progressive decline in beta cell function. This was clearly documented in
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 855

the UKPDS, which also showed that this decline in beta cell function was
not altered by treatment with diet, metformin, sulfonylureas, or insulin [10].
The decline in beta cell function is caused by a decreased ability of
glucose to cause closure of an ATP-dependent potassium channel in the
plasma membrane of beta cell [34,35]. The normal mechanism for glucose
stimulation of insulin secretion involves the following steps [34,35]. Glucose
from the plasma compartment is rapidly transported into the beta cell,
phosphorylated, and metabolized to generate ATP. Special enzymes in the
beta cell (Glut-2 glucose transporter and glucokinase) allow this process to
occur quantitatively so that the intracellular ATP:ADP ratio accurately
reflects the plasma glucose concentration. The plasma membrane of the beta
cell contains a potassium channel whose function is regulated by the
ATP:ADP ratio and a voltage-dependent calcium channel. When the plasma
glucose is low the potassium channel is open and extruding potassium from
the beta cell. The plasma membrane is appropriately polarized and the
calcium channel is closed. As the plasma glucose rises, the ATP:ADP ratio
increases, which causes the ATP-dependent potassium channels to close.
Closure of the ATP-dependent potassium channel causes depolarization of
the adjacent plasma membrane, which results in opening of the voltage-
dependent calcium channels in that portion of the depolarized membrane.
The open calcium channels allow calcium to be transported from the
extracellular compartment into the cytoplasm of the beta cell. Increases in
cytosolic calcium ion concentrations linearly increase the release of insulin
from the beta cell granule into the plasma compartment. The impaired
release of insulin, which is characteristic of the early stages of type 2
diabetes, is a delayed and impaired ability of plasma glucose fluctuations
appropriately to regulate closure of the ATP-dependent potassium channel.
As the duration of type 2 diabetes increases, the functional defect in beta
cells is compounded by an increase in apoptosis and a decrease in beta cell
mass. The mechanism responsible for the decrease in mass is unclear but the
consequence is that type 2 diabetic patients become more insulin deficient
with time.
In the first 4 or 5 years after the onset of clinical type 2 diabetes the
functional defect in insulin secretion can be ameliorated by pharmacologic
agents, which act directly on the ATP-dependent potassium channel and
augment its closure by glucose. These agents are the insulin secretagogues,
of which there are three distinct classes in clinical use: (1) sulfonylureas, (2)
meglitinides, and (3) phenylalanine derivatives. All of these agents interact
with the regulatory subunit of the ATP-dependent potassium channel and
directly stimulate its closure [34,36]. There are subtle differences in the
specific binding site and characteristics of binding among the three classes of
insulin secretagogues, which modulate differences in their pattern of insulin
release [35,37].
The commonly prescribed sulfonylureas include glyburide, glipizide, and
glimepiride. The glyburide is formulated as regular and micronized. The
856 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

glipizide is formulated as short acting and slow release. Table 4 lists these
formulations and their characteristics. The meglitinide that is available is
repaglinide and the phenylalanine derivative that is marketed is nateglinide.
Their characteristics are also listed in Table 4.
The clinically relevant insulin secretory defects in type 2 diabetes are (1)
a 30- to 60-minute delay in meal-mediated insulin secretion, (2) a deficient
quantity of insulin secretion, and (3) a progressive loss in beta cell function
with time [34,36,38]. Administration of sulfonylureas does not significantly
alter the delay in meal-mediated insulin secretion.
Sulfonylurea treatment augments fasting insulin secretion and the second
or late phase (after 60 minutes) of meal-mediated insulin secretion [34,36].
The consequence of these pharmacologic actions is a lowering of the fasting
plasma glucose but very little decrease in the postprandial plasma glucose
excursion. Sulfonylureas do not stimulate insulin biosynthesis. The two new
insulin secretagogues, repaglinide [39,40] and nateglinide [41,42], were
specifically designed to increase early meal-mediated insulin secretion. They
are able to do so because they are rapidly absorbed and have rapid binding
kinetics to the regulatory subunit of the ATP-dependent potassium channel
[35,37,39–42]. They can be taken at the time of the meal and are able to
facilitate early meal-mediated insulin secretion. This provides for a lower
early and a shorter duration of postprandial glucose rise. These insulin
secretagogues in contrast to sulfonylureas decrease the postprandial plasma
glucose excursions [39–43].

Table 4
Properties of commonly prescribed insulin secretagogues
Duration
Generic name Daily dose of action Comments
Glyburide 2.5–20 mg >24 h Absorption is incomplete and variable
Hypoglycemia is the most common and severe
Blocks ischemic preconditioning
Glyburide 1–8 mg >24 h More consistent absorption
micronized
Glipizide 2.5–20 mg >12 h Relatively short acting
Hypoglycemia less severe and about half as
frequent as glyburide
Glipizide- 5–20 mg 24 h Hypoglycemia and weight gain reported to be
GITS quite low
Glimepiride 1–8 mg 24 h Hypoglycemia frequency and severity \ half that
(Amaryl) of glyburide
No interference with ischemic preconditioning
Repaglinide 1–4 mg with 5–6 h Low incidence of hypoglycemia
(Prandin) each meal Weight gain less than with sulfonylureas
Nateglinide 60–120 mg 3–4 h Very low incidence of hypoglycemia
(Starlix) with each Little data on weight gain
meal
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 857

Sulfonylurea drugs have a long duration of action because of their


pharmacokinetic properties and their binding kinetics to the regulatory
subunit of the ATP-dependent potassium channel [34,36]. Repaglinide has
a prolonged low-affinity binding to the regulatory subunit of the ATP-
dependent potassium channel, and although it has a rapid onset of insulin
release it also continues stimulating some insulin secretion for several hours,
which explains its additional effect in lowering fasting plasma glucose
[37,44]. In contrast, nateglinide has a rapid onset and short duration of
insulin secretory action because it rapidly dissociates from the regulatory
subunit of the ATP-dependent potassium channel [37,41]. Its beneficial
effects are primarily in lowering postprandial plasma glucose excursions
[41].
Understanding the pharmacology of the various insulin secretagogues
explains the differences in their clinical effects (Table 5) [34,36]. Sulfonylur-
eas have a prolonged duration of action. Their primary beneficial effects are
to decrease fasting and between-meal hyperglycemia. They have minimal
effects on postprandial glucose excursions. Because of their prolonged
actions they can lead to severe fasting and late postprandial hypoglycemia
[34,45,46]. This is particularly so when meals are delayed or missed [47].
Severe sulfonylurea-induced hypoglycemia is a rare complication that is
more likely to occur in older individuals and in those with underlying
cardiovascular or kidney disease [36,45,46]. When it occurs, patients need
treatment for several days in an emergency ward or intensive care unit.
Glyburide treatment has been reported to have a significantly greater
frequency of severe hypoglycemia than other sulfonylureas [34,45,46].
Nateglinide has only minor effects on fasting and between-meal hyper-
glycemia but is very effective in reducing postprandial glucose excursions
[41,48]. Because of its rapid onset and short duration of action it must be
taken with each meal. A major advantage is that missed meals and delayed
meals are unlikely to cause significant hypoglycemia. Indeed, reported rates
of hypoglycemia are quite low. Because of its minimal effects on fasting
hyperglycemia nateglinide is usually given in combination with other
antihyperglycemic agents [42].
Repaglinide has properties that combine the benefits of both sulfonylur-
eas and nateglinide. It primarily facilitates early meal-mediated insulin
secretion but it also has some modest prolonged insulin secretory function
[43,44,47]. As a result, it lowers fasting plasma glucose and also decreases
postprandial glucose excursions [40,44]. It is taken with each meal. If meals
are delayed or missed there is a minimal likelihood of having significant
hypoglycemia [47]. Clinical trials comparing treatment with repaglinide
versus a sulfonylurea show similar degrees of glycemic control with
repaglinide causing somewhat less hypoglycemia and weight gain [44].
The beneficial effects of sulfonylureas on glycemic control occur during
the first several years following diagnosis. The effects decrease with duration
of diabetes. In the UKPDS trial the percentage of type 2 diabetic patients
858 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

Table 5
Comparison of classes of insulin secretagogues
Sulfonylureas Repaglinide Nateglinide
Dosing Once or twice daily With each meal With each meal
FPG # 50 to 60 mg/dL # 50 to 60 mg/dL # 20 mg/dL
PPG excursion Slight effect Moderate effect Major effect
PP insulin secretion " late phase " early and late phases 10 min to 4 h
Hypoglycemia Fasting and late PP Less than sulfonylureas Uncommon
Weight gain 1 to 3 kg Less than sulfonylureas ?
HbA1c # 1.5% # 1.5% # 0.8%
Cost Inexpensive Expensive Expensive
Abbreviations: FPG, fasting plasma glucose; PP, postprandial; PPG, postprandial plasma
glucose excursion.

that achieved HbA1c less than 7% on sulfonylurea monotherapy decreased


from 50% after 3 years to 34% after 6 years to 24% after 9 years [49]. Many
type 2 diabetic patients need insulin replacement therapy rather than insulin
secretagogue therapy after 5 to 10 years of clinical disease [50,51].

Specific oral therapies: agents to delay carbohydrate digestion and


absorption
Postprandial plasma glucose levels are determined by the nutrients being
ingested, the gastrointestinal hormones being secreted in response to the
nutrients, pancreatic insulin and glucagon secretion, and the responsiveness
of the liver and peripheral tissues to the secreted insulin and glucagons.
Changing the digestion of complex carbohydrates decreases the rate of
monosaccharide absorption and lowers postprandial plasma glucose and,
secondarily, plasma insulin levels [52].
Complex carbohydrates are digested to oligosaccharides in the small
intestine by pancreatic lipase. The oligosaccharides are then cleaved to
monosaccharides by a group of enzymes (a-glucosidases) located in the
brush border of the enterocytes and the monosaccharides are absorbed [52].
Normally, most of the digestion and absorption of carbohydrates occurs in
the duodenum and upper jejunum.
Competitive inhibitors of the a-glucosidase enzymes have been made and
when given to diabetic patients decrease the postprandial plasma glucose
rise [52–54]. These drugs decrease the rate of digestion of oligosaccharides in
the duodenum and upper jejunum. The oligosaccharides proceed through
the small intestine where they are slowly cleaved and absorbed. This results
in a decrease in the mean peak postprandial glucose of approximately 50
mg/dL [52–54]. The mean fasting plasma glucose during chronic therapy
with a-glucosidase inhibitors is reduced approximately 20 mg/dL and the
mean HbA1c by 0.5% [52–54].
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 859

The major a-glucosidase inhibitor used is acarbose, which is a nonabsorb-


able agent. The recommended dose is 50 mg taken with the start of each
meal. The distal jejunum and ileum normally have very low concentrations
of a-glucosidase enzymes, which means that during the initiation of
treatment much of the undigested oligosaccharide passes into the colon
where it is metabolized by the bacteria to short-chain fatty acids, hydrogen,
methane, and carbon dioxide. That causes the side effects of abdominal
pain, diarrhea, and flatulence. The side effects can be minimized by initiating
therapy with 25 mg with the evening meal. After a week or so the dose can
be increased to 25 mg with breakfast and the evening meal. If minimal or no
symptoms occur the dose titration can be slowly increased to 50 mg with
each meal. Occasional patients may require 100 mg with each meal. Dose
increases are determined by the 1-hour postprandial plasma glucose levels,
which ideally should be less than or equal to 180 mg/dL.
Acarbose and miglitol are the a-glucosidase inhibitors on the market. The
usual dose is 50 mg with each meal. The a-glucosidase inhibitors can be
combined with all other classes of antihyperglycemic therapy.

Combination therapies
Combination therapy should be the treatment of choice in most patients
with type 2 diabetes. Acceptance of the importance of combination therapy
has spawned the production of several fixed combinations, such as glyburide
and metformin (Glucovance), metformin and glipizide (Metaglip), metfor-
min and rosiglitazone (Avandemet), and glimepiride and rosiglitazone
(Avandaril). These fixed combinations provide for better compliance and
lower cost because they require one copay.
The use of metformin and a thiazolidinedione provides maximum effects
on insulin resistance with diminished side effects. This combination is quite
useful in patients in the early stages of hyperglycemia when insulin resistance
is the predominant defect and insulin secretion is still well maintained. In the
more advanced stages of hyperglycemia when insulin secretory deficiency is
more marked, a combination of an insulin secretagogue and one or more
insulin sensitizers is more appropriate. a-Glucosidase inhibitors or nategli-
nide can be added specifically to improve recalcitrant postprandial
hyperglycemia. The choice between adding a third oral agent or adding
basal insulin therapy should be dictated by the cost and the magnitude of the
HbA1c at the time the therapy change is being considered.

Summary
The appropriate management of patients with type 2 diabetes presents
many challenges to health care providers. The disease is increasing at
860 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

alarming rates because of the population’s changing lifestyle. The first


several years of type 2 diabetes are likely to be unrecognized and untreated.
The pathophysiology of type 2 diabetes dictates that treatment of insulin
resistance should be an early and central focus for every therapeutic
program. Effective treatment of insulin resistance can slow the rate of beta
cell deterioration and improve the cardiovascular risk factors, which are
part of the metabolic syndrome. When beta cell function deteriorates it is
necessary to improve insulin availability either by adding an insulin
secretagogue or providing insulin replacement. Treatment must focus on
control of both fasting and postprandial plasma glucose levels if glycemic
targets are to be met and chronic complications are to be avoided.
Appropriate control of blood pressure, lipids, and the predilection to
thrombotic disease are equally important targets. The pharmacologic tools
currently available are capable of allowing most patients with type 2
diabetes to achieve good metabolic control. Implementation of early
combination therapy is essential if glycemic targets are to be met.

References
[1] UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulphony-
lureas or insulin compared with conventional treatment and risk of complications in
patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.
[2] UK Prospective Diabetes Study Group. Effect of intensive blood glucose control with
metformin on complications in overweight patients with type 2 diabetes (UKPDS 34).
Lancet 1998;352:854–65.
[3] Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S, et al. Effects of
intensive blood-pressure lowering and low-dose aspirin in patients with hypertension:
principal results of the Hypertension Optimal Treatment (HOT) randomized trial. Lancet
1998;351:1755–62.
[4] Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of
cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomized
placebo-controlled trial. Lancet 2002;360:7–22.
[5] Gaede P, Vedel P, Larsen N, Jensen GVH, Parving H-H, Pederson O. Multifactorial
intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med
2003;348:383–93.
[6] Saydah SH, Fradkin J, Cowie CC. Poor control of risk factors for vascular disease among
adults with previously diagnosed diabetes. JAMA 2004;291:335–42.
[7] Koro CE, Bourgeois N, Bowlin SJ, Fedder DO. Glycemic control from 1988 to 2000
among US adults diagnosed with type 2 diabetes. Diabetes Care 2004;27:17–20.
[8] Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among
US adults, 1999–2000. JAMA 2002;288:1723–7.
[9] Brown N. Diabetes 2003;(Suppl 1):A61–2.
[10] UK Prospective Diabetes Study Group. Overview of 6 years therapy of type 2 diabetes:
a progressive disease (UKPDS 16). Diabetes 1995;44:1249–58.
[11] Lebovitz HE. Insulin secretogogues: sulfonylueas, meglitinides and phenylalanine deriva-
tives. In: LeRoith D, Taylor SI, Olefsky JR, editors. Diabetes mellitus: a fundamental and
clinical text. 3rd edition. Philadelphia: Lippincott Williams & Wilkins; 2004. p. 1107–22.
[12] Lebovitz HE. Oral therapies for diabetic hyperglycemia. Endocrinol Metab Clin North
Am 2001;30:909–33.
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 861

[13] Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin secretory
dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin
Invest 1999;104:787–94.
[14] Tuomilehto J, Lindstrom J, Eriksson JG, Hamalainen H, Ilanne-Parikka P, Keinanen-
Kiukaanniemi S, et al. Finnish Diabetes Preventation Group. Preventation of type 2
diabetes by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J
Med 2001;344:1343–50.
[15] Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al.
Diabetes Prevention Program Group. Reduction in the incidence of type 2 diabetes with
lifestyle intervention or metformin. N Engl J Med 2002;346:393–403.
[16] Buchanan TA, Xiang AH, Peters RK, Kjos SL, Marroquin A, Goico J, et al. Preservation
of pancreatic b-cell function and prevention of type 2 diabetes by pharmacological
treatment of insulin resistance in high-risk hispanic women. Diabetes 2002;51:2796–803.
[17] Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laasko M. STOP-NIDDM
Trial Research Group. Acarbose for the prevention of type 2 diabetes the STOP-NIDDM
randomized trial. Lancet 2002;359:2072–7.
[18] Diabetes Control and Complications Trial Research Group. The effect of intensive
treatment of diabetes on the development and progression of long-term complications in
insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.
[19] Writing Team For The Diabetes Control And Complications Trial/Epidemiology Of
Diabetes Interventions And Complications Research Group. Sustained effect of intensive
treatment of type 1 diabetes mellitus on development and progression of diabetic
nephropathy: the Epidemiology of Diabetes Interventions and Complications (EDIC)
study. JAMA 2003;290:2159–67.
[20] Lebovitz HE, Banerji MA. Treatment of insulin resistance in diabetes mellitus. Eur J
Pharmacol 2004;440:135–46.
[21] Lebovitz HE. Rationale for and role of thiazolidinediones in type 2 diabetes mellitus. Am J
Cardiol 2002;90:34G–41G.
[22] Lebovitz HE. Treating hyperglycemia in type 2 diabetes: new goals and strategies. Cleve
Clin J Med 2002;69:809–20.
[23] Inzucchi SE, Maggs DG, Spollett GR, Page SL, Rife FS, Walton V, et al. Efficacy and
metabolic effects of metformin and troglitazone in type 2 diabetes mellitus. N Engl J Med
1998;338:867–72.
[24] Yu JG, Kruszynska YT, Mulford MI, Olefsky JM. A comparison of troglitazone and
metformin on insulin requirements in euglycemic intensively insulin-treated type 2 diabetic
patients. Diabetes 1999;48:2414–21.
[25] Lebovitz HE, Dole JF, Patwardhan R, Rappaport EB, Freed MI. Rosiglitazone
monotherapy is effective in patients with type 2 diabetes. J Clin Endocrinol Metab 2001;
86:280–8.
[26] Kipnes MS, Krosnick A, Rendell MS, Egan JW, Mathisen AL, Schneider RL. Pioglitazone
hydrochloride in combination with sulfonylurea therapy improves glycemic control in
type 2 diabetes mellitus: a randomized, placebo-controlled study. Am J Med 2001;111:
10–7.
[27] Raskin P, Rendell M, Riddle MC, Dole JF, Freed MI, Rosenstock J, for the
Rosiglitazone Clinical Trials Study Group. A randomized trial of rosiglitazone therapy in
patients with inadequately controlled insulin-treated type 2 diabetes. Diabetes Care 2001;
24:1226–32.
[28] Fonseca V, Rosenstock J, Patwardhan R, Salzman A. Effect of metformin and
rosiglitazone combination therapy in patients with type 2 diabetes mellitus: a randomized
controlled trial. JAMA 2000;283:1695–702.
[29] Bailey CJ, Turner RC. Drug therapy: metformin. N Engl J Med 1996;334:574–9.
[30] Lebovitz HE, Banerji MA. Insulin resistance and its treatment by thiazolidinediones.
Recent Prog Horm Res 2001;56:265–94.
862 H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863

[31] DeFronzo RA, Goodman AM. Efficacy of metformin in patients with non-insulin-
dependent diabetes mellitus. N Engl J Med 1995;333:541–9.
[32] Kelley IE, Walsh K, Han TS, et al. Effect of a thiazolidinedione compound on body fat
and fat distribution of patients with type 2 diabetes. Diabetes Care 1999;22:288–93.
[33] Nesto RW, Bell D, Bonow RO, Fonseca V, Grundy SM, Horton ES, et al. Thiazolidinedione
use, fluid retention, and congestive heart failure: a consensus statement from the American
Heart Association and American Diabetes Association. October 7, 2003. Circulation 2003;
108:2941–8.
[34] Lebovitz HE, Melander A. Sulfonylureas: basic aspects and clinical use. In: DeFronzo R,
editor. International textbook of diabetes. 3rd edition. Chichester (UK): John Wiley and
Sons Ltd; 2004.
[35] Fuhlendorff J, Rorsman P, Kofod H, Brand CL, Rolin B, MacKay P, et al. Stimulation of
insulin release by repaglinide and glibenclamide involves both common and distinct
processes. Diabetes 1998;47:345–51.
[36] Lebovitz HE. Insulin secretogogues: sulfonylueas, meglitinides and phenylalanine deriva-
tives. In: LeRoith D, Taylor SI, Olefsky JR, editors. Diabetes mellitus: a fundamental and
clinical text. 3rd edition. Philadelphia: Lippincott Williams & Wilkins; 2004. p. 1107–22.
[37] Hu S, Wang S, Fanelli B, Bell PA, Dunning BE, Geisse S, et al. Pancreatic b-cell KATP
channel activity and membrane-binding studies with nateglinide: a comparison with
sulfonylureas and repaglinide. J Pharmacol Exp Ther 2000;293:444–52.
[38] Pimenta W, Korytkowski M, Mitrakou A, et al. Pancreatic beta-cell dysfunction as the
primary genetic lesion in NIDDM: evidence from studies in normal glucose-tolerant
individuals with a first-degree NIDDM relative. JAMA 1995;273:1855–61.
[39] Guay DRP. Repaglinide, a novel, short-acting hypoglycemic agent for type 2 diabetes
mellitus. Pharmacotherapy 1998;18:1195–204.
[40] Jovanovic L, Dailey G III, Huang W-C, Strange P, Goldstein BJ. Repaglinide in type
2 diabetes: a 24-week fixed-dose efficacy and safety study. J Clin Pharmacol 2000;40:
49–57.
[41] Hanefeld M, Dickinson S, Bouter KP, Guitard C. Rapid and short-acting mealtime insulin
secretion with nateglinide controls both prandial and mean glycemia. Diabetes Care 2000;
23:202–7.
[42] Horton ES, Foley J, Clinkingbeard C, Foley J, Mallow S, Shen S. Nateglinide alone and in
combination with metformin improves glycemic control by reducing mealtime glucose
levels in type 2 diabetes. Diabetes Care 2000;23:1660–5.
[43] Owens DR, Ismail I, Luzio SD, et al. Increased prandial insulin secretion after
administration of a single preprandial oral dose of repaglinide in patients with type 2
diabetes. Diabetes Care 2000;23:518–23.
[44] Marbury T, Huang W-C, Strange P, Lebovitz HE. Repaglinide versus glyburide: a one
year comparison trial. Diabetes Res Clin Pract 1999;43:155–66.
[45] Shorr RI, Ray WA, Daugherty WA, et al. Individual sulfonylureas and serious
hypoglycemia in older people. J Am Geriatr Soc 1996;44:751–5.
[46] Van Staa T, Abenhaim L, Monette J. Rates of hypoglycemia in users of sulfonylureas.
J Clin Epidemiol 1997;50:735–41.
[47] Damsbo P, Marbury TC, Clauson P, Windfeld K. A double-blind randomized comparison
of meal-related glycemic control by repaglinide and glyburide in well-controlled type 2
diabetic patients. Diabetes Care 1999;22:789–94.
[48] Walter YH, Spratt DI, Garreffa S, McLeod JF. Mealtime glucose regulation by nateglinide
in type-2 diabetes mellitus. Eur J Clin Pharmacol 2000;56:129–33.
[49] Turner RC, Cull CA, Frighi V, Holman RR. Glycemic control with diet, sulfonylurea,
metformin, or insulin in patients with type 2 diabetes mellitus: progressive requirements for
multiple therapies (UKPDS 49). JAMA 1999;281:2005–12.
[50] Yki-Jarvinen H, Dressler A, Zieman M. Study Group HOE 901/3002. Less nocturnal
hypoglycemia and better post dinner glucose control with bedtime insulin glargine
H.E. Lebovitz / Med Clin N Am 88 (2004) 847–863 863

compared with bedtime NPH insulin during combination therapy in type 2 diabetes.
Diabetes Care 2001;23:1130–6.
[51] Yki-Jarvinen H, Ryysy L, Nikkila K, Tulokas T, Vanamo R, Heikkila M. Comparison of
bedtime insulin regimens in patients with type 2 diabetes mellitus: a randomized, controlled
trial. Ann Intern Med 1999;130:389–96.
[52] Lebovitz HE. Alpha-glucosidase inhibitors as agents in the treatment of diabetes. Diabetes
Reviews 1998;6:132–45.
[53] Chiasson JL, Josse R, Hunt J, Palmason C, Rodger NW, Ross SA, et al. The efficacy of
acarbose in the treatment of patients with non-insulin dependent diabetes mellitus. Ann
Intern Med 1994;121:928–35.
[54] Holman RR, Cull CA, Turner RC. A randomized double-blind trial of acarbose in type 2
diabetes shows improved glycemic control over 3 years. Diabetes Care 1999;22:960–4.
Med Clin N Am 88 (2004) 865–895

Insulin therapy in type 2 diabetes


Trent Davis, MD, Steven V. Edelman, MD*
Section of Diabetes/Metabolism, Veterans Affairs San Diego HealthCare System,
3350 La Jolla Village Drive 111G, San Diego, CA 92161, USA

Diabetes mellitus affects approximately 18 million people in the United


States, which is approximately 6% of the overall population, and over
800,000 new cases are diagnosed annually [1]. Diabetes may actually be
more endemic than these figures indicate because there are no symptoms in
the early stages of the disease, and potentially one undiagnosed individual
exists for every one that is identified [2]. Of the total diabetic population,
85% to 90% of individuals have type 2 diabetes whereas 10% to 15% have
type 1 diabetes [3].
Type 2 diabetes leads to a tremendous amount of death and disability
and uses a large portion of the health care dollar [4]. Although diabetes is
associated with multiple disorders with distinct pathologic mechanisms,
insulin resistance is the common denominator and is associated with several
comorbidities, including obesity, hypertension, and vascular disease. The
natural history of the disease is often complicated by various microvascular
and macrovascular sequelae that can lead to blindness, end-stage renal
disease, lower-extremity amputation, and atherosclerosis resulting in heart
attack or stroke [3,5]. Although most of the human suffering is caused by
end-stage microvascular disease, 80% of diabetics die of macrovascular
cardiovascular disease. There is now clear evidence from the United
Kingdom Prospective Diabetes Study (UKPDS) and the Kumamoto study
that improved glycemic control through intensive diabetes management
delays the onset and significantly retards the progression of microvascular
complications in patients with type 2 diabetes mellitus [6,7]. The results from
the UKPDS are reassuring in that, although intensive treatment with insulin
was associated with increased weight gain and hypoglycemia, there is no
evidence of any harmful effect of insulin on cardiovascular outcomes, which
has been a controversial issue. An epidemiologic analysis of the UKPDS

* Corresponding author.
E-mail address: svedelman@vapop.ucsd.edu (S.V. Edelman).

0025-7125/04/$ - see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.005
866 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

data shows a continuous association between the risk of cardiovascular


complications and glycemia, such that for every percentage point of decrease
in HbA1c (eg, from 9% to 8%), there is a 25% reduction in diabetes-related
deaths, a 7% reduction in all-cause mortality, and an 18% reduction in
combined fatal and nonfatal myocardial infarction [6].
To achieve glycemic goals in patients with type 2 diabetes, we now have
multiple pharmacologic agents with different mechanisms of action, in-
cluding sulfonylureas, meglitinides, metformin, a-glucosidase inhibitors,
thiazolidinediones, and insulin. It must be emphasized, however, that unlike
patients with type 1 diabetes, who have no significant insulin secretion and
hence require insulin therapy from the onset of their disease, a prominent
feature in the early stages of the disease for patients with type 2 diabetes
is insulin resistance with hyperinsulinemia. Therefore, improving insulin
sensitivity be means of caloric restriction, exercise, and weight management
early in the disease process will benefit type 2 diabetics. When these measures
fail, glycemic goals can often be achieved with oral agents used alone or
in combination with each other. When patients are diagnosed late in the
natural history, however, there is progressive loss of pancreatic beta-cell
function and endogenous insulin secretion, making diurnal glycemic control
difficult. At this late stage, most patients require exogenous insulin therapy
to achieve optimal glucose control. The American Diabetes Association
(ADA) now recommends that the glycemic objective for patients with type 2
diabetes to normalize glycemia and glycosylated hemoglobin concentrations
should be similar to that for type 1 diabetes.

Pathogenesis and natural history of type 2 diabetes


Of the Americans diagnosed with type 2 diabetes, 80% to 90% are obese,
and the remainder are lean [8]. The genesis of hyperglycemia in type 2
diabetes involves a triad of abnormalities: excessive hepatic glucose pro-
duction, impaired pancreatic insulin secretion, and peripheral resistance
to insulin action, occurring principally in liver and muscle tissue [9]. The
severity of these abnormalities and their contribution to the degree of
hyperglycemia can vary considerably, causing heterogeneity in the metabolic
expression of the diabetic state. Such differences are best exemplified by the
lean and obese varieties of type 2 diabetes, which have the same underlying
pathophysiologic basis but differ in the extent to which each abnormality
contributes to the development of the hyperglycemic state. Of these ab-
normalities, peripheral insulin resistance to insulin action and impaired
pancreatic beta-cell secretion are early and primary abnormalities, whereas
increased hepatic glucose production is a late and secondary manifestation.
Early in their disease, patients with type 2 diabetes compensate for increased
insulin resistance at the tissue level by increasing pancreatic beta-cell insulin
secretion [10]. When this compensation is no longer adequate to overcome
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 867

the insulin resistance, blood glucose levels begin to rise. Over the course
of the disease, endogenous insulin levels slowly begin to decrease and,
ultimately, many patients with type 2 diabetes are unable to achieve optimal
glycemic control with oral agents [11].
In subjects with type 2 diabetes who are lean, impaired insulin secretion is
the predominant defect, and insulin resistance tends to be less severe than in
the obese variety [12]. On the other hand, insulin resistance and hyper-
insulinemia are the classical abnormalities of obese persons with type
2 diabetes [12]. In type 2 diabetes, insulin secretion is often excessive
compared with the nondiabetic situation but is still insufficient to overcome
the insulin resistance that is present. It is important to understand and
appreciate these fundamental differences when considering insulin therapy
in type 2 diabetes. Based on this knowledge, lean type 2 diabetic subjects
usually fail oral agents faster and will require considerably less insulin to
control their hyperglycemia than their obese counterparts. In contrast, large
doses of exogenous insulin are the rule in the obese form of this disorder
when euglycemia is desired [13].
The need for large amounts of exogenous insulin in obese type 2 diabetes
also raises the question of the most appropriate methods of insulin delivery.
Under normal circumstances, insulin is secreted from the pancreas into the
portal vein, going directly to the liver in which a large first-pass extraction
of portal insulin occurs [14]. When insulin is injected subcutaneously,
absorption occurs directly into the peripheral circulation, without the initial
effects of hepatic extraction. Therefore, the tissues are exposed to greater
levels of insulin than if insulin was provided by the portal route. Because the
primary target of exogenous insulin is the liver, type 2 diabetes may be
uniquely suited to delivery of insulin through the portal vein. Such a
situation occurs when insulin is delivered intraperitoneally, and the majority
of insulin is absorbed into the portal circulation [15]. Intraperitoneal insulin
delivery systems will not be discussed in this section, however, this method
holds considerable promise in type 2 diabetes because of the more
physiologic delivery of insulin and because of selective and effective
inhibition of hepatic glucose output, with less peripheral insulinemia than
occurs with subcutaneous insulin injections [16].

Intensive insulin therapy


Successful insulin management requires an educated and motivated
patient, as well as the participation of a multidisciplinary health care team.
Intensive insulin therapy requires a substantial input of physician and
support staff time, which has a significant economic impact on the health
care system [17]. Although long-term data on costs are not yet available,
projections suggest that substantial savings from the high costs of end-
stage disease could be achieved by following ADA guidelines [18].
868 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

Furthermore, as discussed later, we now have the opportunity to use


combinations of insulin and a variety of oral antidiabetic agents with
differing mechanisms of action. The use of these potent combinations
permit us to safely and effectively lower blood glucose levels and to
achieve ADA target glycemic levels with relatively low risk for hypogly-
cemia or weight gain.
In addition to the natural history of type 2 diabetes, there is heterogeneity
in the pathophysiology of type 2 diabetes mellitus that may influence when
patients require insulin. Some patients who have been diagnosed with type 2
diabetes may actually have a condition more closely related to insulin-
dependent or type 1 diabetes, with severe insulinopenia. Many of these pa-
tients have been shown to have islet cell antibody positivity or antibodies to
glutamic acid decarboxylase, with a decreased C-peptide response to
glucagon stimulation and a propensity for primary oral medication failure
[19]. These individuals are now labeled with the condition latent autoimmune
diabetes in adults [20]. There are also wide geographic and racial differences
that may influence the need for insulin therapy. For example, Asian patients
with type 2 diabetes tend to be thinner, are diagnosed with diabetes at an
earlier age, fail oral hypoglycemic agents much sooner, and are more
sensitive to insulin therapy than the classic centrally obese patient in the
United States and some parts of Europe [21].
The goals of therapy should be tailored to individual patients. Candidates
for intensive management should be motivated, compliant, and educable,
and be without other medical conditions and physical limitations that
preclude accurate and reliable home glucose monitoring (HGM) and insulin
administration; caution is advised in patients who are aged or have hypo-
glycemic unawareness. Other limitations to achieving normoglycemia
may include high titers of insulin antibodies, especially in patients with a
history of intermittent use of insulin of animal origin. The site of insulin
injection also may change the pharmacokinetics, and absorption can be
highly variable, especially if lipohypertrophy is present. The periumbilical
area has been shown to be one of the most desirable areas to inject insulin
because of the rapid and consistent absorption kinetics observed at this
location; however, rotating the injection site is usually advised [22]. It is also
advisable to inject in the same body location for a certain meal time (ie,
triceps fat pad for breakfast, abdomen for lunch, and upper thighs for
dinner) [23].
In summary, before starting insulin therapy, the patient should be well
educated in the techniques of HGM, proper insulin administration, and self-
adjustment of the insulin dose, if appropriate, as well as knowledgeable
about dietary and exercise strategies, including carbohydrate counting. The
patient and family members also need to be informed about hypoglycemia
prevention, recognition, and treatment. Initial and ongoing education by a
diabetes management team, including a certified diabetes educator, is crucial
for long-term success and safety.
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 869

Insulin treatment strategies


Combination therapy
Combination therapy usually refers to the use of daytime oral antidia-
betic agents together with a single injection of intermediate or long-acting
insulin at bedtime. Several studies [24–36] have looked at the safety and
efficacy of combination therapy. For many of the reasons mentioned earlier,
the analysis of studies to evaluate the efficacy and safety of combination
therapy is difficult. Several review articles using meta-analysis conclude that
combination therapy results in only modest improvements in glucose con-
trol and contribute to increased medical costs of diabetes management
compared with insulin therapy alone. These earlier studies, however, were
conducted when sulfonylureas were the only available type of oral agent.
Because of heterogeneity in type 2 diabetes together with variability in the
design and clinical situations of previous studies, however, the use of meta-
analysis may be inappropriate for making generalized statements regarding
this form of therapy [27,37]. Based on several recent reports, the use of
combination therapy has been quite successful in selected patients, especially
with the newer oral agents used alone or together with insulin [26–
29,35,36,38–41].
For a number of practical reasons, combination therapy may be
beneficial. The patient does not need to learn how to mix different types
of insulin, and patient compliance and acceptance are better with a single
injection than with multiple injections of insulin. Combination therapy also
requires a lower total dose of exogenous insulin than regimens of two or
three injections per day. Combination therapy also contributes to less
weight gain and peripheral hyperinsulinemia. Last, combination therapy is
ideally suited to suppress excessive hepatic glucose production overnight.
The rationale for combination therapy with insulin and sulfonylureas is
based on the assumption that, if evening insulin lowers the fasting glucose
concentration to normal, then daytime oral agents will be more effective
in controlling postprandial hyperglycemia and maintaining euglycemia
throughout the day. Metabolic profiles of patients who have type 2 diabetes
have demonstrated that fasting blood glucose contributes significantly to
daytime hyperglycemia [42]. In addition, the fasting blood glucose concen-
tration is highly correlated with the degree of hepatic glucose production
during the early morning hours [13]. Hepatic glucose output is directly
decreased by insulin [43] and indirectly inhibited by the ability of insulin to
reduce adipose tissue lipolysis, with lower concentrations of free fatty acids
and gluconeogenesis [41]. Also, the peak of bedtime intermediate-acting
insulin coincides with the onset of the dawn phenomenon (early morning
resistance to insulin caused by diurnal variations in growth hormone
and possibly in levels of norepinephrine), which usually occurs between
3 and 7 AM. Bedtime insulin also increases the morning serum insulin
870 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

concentration and may assist in reducing the post-breakfast glucose rise in


addition to the fasting value.

Combination of insulin and sulfonylurea agents


In one of the first large studies demonstrating the efficacy of insulin/
sulfonylurea combination therapy, Yki-Jarvinen et al [38] compared com-
bination therapy with regimens of two and four insulin injections per day
in patients with type 2 diabetes. These patients were on submaximal doses
of glyburide (12.5 mg/d), glipizide (20 mg/d), and metformin (1.4 g/d),
with fasting blood glucose concentrations at approximately 225 mg/dL and
mean fasting serum C-peptide values of 0.66 nmol/L. After 3 months, all
treatment groups had similar reductions in mean diurnal glucose con-
centrations and glycosylated hemoglobin levels (1.6%–1.9%) compared
with the control group, who were taking oral agents alone. The group
treated with a combination of oral agents and bedtime neutral protamine
Hagedorn (NPH) insulin, however, had the least weight gain (1.2  0.5 kg)
of any group and a 50% to 65% lower increment in mean diurnal serum-free
insulin concentrations. There was no evidence of severe hypoglycemia with
combination therapy, and patient acceptance was excellent.
Several other recent publications [26,28,29,31,34,40] also support the
additional efficacy and safety of combination therapy in patients who are
inadequately controlled by oral hypoglycemic agents alone. A recent study
conducted by Riddle and Schneider [36] demonstrates the efficacy and safety
of a combination consisting of 70% NPH insulin and 30% regular insulin
(70/30) insulin at dinnertime and sulfonylurea therapy. In this study, 145
type 2 diabetics with uncontrolled hyperglycemia (fasting plasma glucose
level [FPG] 180–300 mg/dL), on maximum sulfonylurea therapy (glimepir-
ide, 8 mg orally, twice daily) were randomized to placebo plus insulin or
glimepiride plus insulin for 6 months. The dose of 70/30 insulin at
dinnertime was titrated to keep fasting fingerstick capillary blood glucose
to less than 120 mg/dL. At 24 weeks, HbA1c levels decreased significantly
and similarly in both groups (9.9%–7.6%). The combination therapy
group, however, needed nearly 35% less insulin than the insulin-only group
(49 versus 78 units) and achieved glycemic control faster, with fewer
dropouts (3% versus 15%, P \ 0.01). Surprisingly, weight gain was similar
(4.0 kg) in both groups.

Insulin and metformin


Weight gain is a constant occurrence in most clinical trials in which
insulin or sulfonylureas, or both agents, are used to treat type 2 diabetes.
Although it was attenuated with combination therapy with sulfonylureas in
the study by Yki-Jarvinen [38], weight gain remains a problem because it can
exacerbate insulin resistance and hyperinsulinemia. The use of metformin in
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 871

this situation may prove advantageous because its use is associated with
reduced weight gain.
The safety and efficacy of metformin in combination with insulin has
been demonstrated in a recent multicenter study by Yki-Jarvinen et al [35].
In this placebo-controlled study, 96 type 2 diabetics who were poorly
controlled with oral sulfonylurea therapy (mean glycosylated hemoglobin
value 9.9%  0.2%; mean fasting plasma glucose level 214  5 mg/dL) were
randomized to 1 year of treatment with bedtime intermediate-acting insulin
plus either glyburide (10.5 mg), metformin (2 g), glyburide and metformin,
or a second injection of intermediate-acting insulin in the morning. Patients
were taught to adjust the bedtime insulin dose on the basis of fasting glucose
measurements. At 1 year, body weight remained unchanged in patients
receiving bedtime insulin plus metformin (mean change 0.9  1.2 kg) but
increased by 3.9  0.7 kg, 3.6  1.2 kg, and 4.6  1.0 kg, respectively, in
patients receiving bedtime insulin plus glyburide, bedtime insulin plus both
oral drugs, and bedtime and morning insulin. In addition, the greatest
decrease in the glycosylated hemoglobin value was observed in the bedtime
insulin and metformin group (from 9.7  0.4% to 7.2  0.2%, a difference
of 2.5  0.4 percentage points) at 1 year (P  0.001 compared with
baseline and P  0.05 compared with other groups). This group also had
significantly fewer symptomatic and biochemical cases of hypoglycemia
(P  0.05) than the other groups. The authors conclude that combination
therapy with bedtime insulin plus metformin not only prevents weight gain
but also seems superior to other bedtime insulin regimens, with respect to
improvements in glycemic control and frequency of hypoglycemia.
In a more recent study [44] of approximately 390 type 2 diabetics, the
combination of insulin and metformin led to a significant improvement in
glycemic control that was greater than with insulin alone. The mean daily
glucose level decreased from 141  34 to 137  31 mg/dL in the insulin-only
group (mean decrease 0.16; 95% confidence interval [CI]; 10–4 mg/dL)
and from 141  40 to 140  31 mg/dL in the metformin group (P = 0.006
versus placebo; mean decrease 1.04; 95% CI; 27 to 9 mg/dL). The
mean daily glucose level decreased by 13 mg/dL more in the metformin
group compared with the placebo group Fig 1.

Insulin and thiazolidinediones


The glitazones are potent insulin sensitizers and are, therefore, well suited
for use in insulin-resistant patients with type 2 diabetes. In several early
studies, troglitazone was documented to not only improve glycemic control
but also to reduce exogenous insulin requirements in obese patients with
type 2 diabetes [45,46]; however, troglitazone was withdrawn from the
US market as a result of an increased risk of severe idiosyncratic liver
damage. Presently there are two glitazones available, rosiglitazone and
pioglitazone, for clinical use in the US, and several more are in development.
872 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

Fig. 1. (A) Blood glucose levels measured at home. (B) Change in blood glucose levels
measured at home. Data are means with SD error bars. For each time point indicated, the first
and the second bars show values at baseline and the third and the fourth show values at
16 weeks. Blood glucose levels in the metformin group compared with the placebo group are all
significantly lower at 16 weeks (P  0.05). The change in glucose values is also significantly
greater in the metformin than in the placebo group at all times during the day (P  0.05). (From
Wulffele MG, Kooy K, Lehert P. Bets D Ogterop JC, Van Der Burg BB, Donker AJM,
Stehouwer CDA. Combination of insulin and metformin in the treatment of type 2 diabetes.
Diabetes Care 2002;25(12):2133–40; with permission.)

In one 16-week study, Rubin et al [47] demonstrated that the daily addition
of 15 and 30 mg of pioglitazone to the regimen of patients receiving a median
dose of 61 units of insulin resulted in mean FPG reductions of 36 and 49 mg/
dL and HbA1c reductions of 0.7% and 1.0%, respectively, compared with
placebo. The insulin-sparing properties of rosiglitazone were shown in a 6-
month study conducted by Raskin et al [48]. They demonstrated that the
addition of 2 and 4 mg orally twice daily of rosiglitazone improved HbA1c
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 873

levels by 0.6% and 1.2%, respectively, compared with placebo, in 312


patients with type 2 diabetes who were uncontrolled on approximately 70
units of insulin daily (baseline HbA1c 9%). Moreover, insulin requirements
were also reduced by approximately 5 and 10 units, respectively, in the two
groups treated with rosiglitazone, in keeping with the insulin sensitizing
effects of the glitazones. In summary, both rosiglitazone and pioglitazone
improve glucose control in poorly controlled, insulin-treated patients with
type 2 diabetes mellitus. There have been no reports of insulin added to
subjects treated with glitazones alone.

Insulin and a-glucosidase inhibitors


The addition of acarbose to insulin therapy may be an option in patients
who have pronounced postprandial hyperglycemia. The first long-term
controlled study to demonstrate a beneficial effect of acarbose in patients
on insulin therapy was reported by Chiasson et al [49]. Of the total number
of patients in this study, 91 were receiving insulin and had glycosylated
hemoglobin values greater than 7%. Postprandial plasma glucose levels at
90 minutes were significantly reduced to 282 mg/dL with the addition of
acarbose, compared with 331 mg/dL seen with insulin alone. Glycosylated
hemoglobin values decreased by 0.4% in the acarbose group, but, as ex-
pected, no significant decreases in fasting plasma glucose levels were seen.
Acarbose may be initiated in patients on insulin treatment by starting with
a low dose of 25 mg with breakfast and titrating up by 25 mg weekly to 50
to 100 mg three times daily with meals (100 mg three times daily for
patients 60 kg body weight), depending on gastrointestinal tolerance and
efficacy.

Insulin glargine and oral agents


The long-acting analog insulin glargine was studied in comparison with
NPH insulin in 756 patients with type 2 diabetes in an open-label, 24-week,
multicenter study [50]. In this study, patients who were inadequately
controlled on oral agents including sulfonylurea, metformin, and glitazones
were randomized to receive either bedtime insulin glargine or NPH insulin,
and the doses were adjusted to obtain a target fasting glucose level of less
than 100 mg/dL (5.6 mmol/L). At the conclusion of the trial, the median
daily dose of insulin was approximately 0.45 IU/kg of body weight in both
groups. The two forms of insulin produced a similar improvement in HbA1c
(6.96 versus 6.97%) and similar reductions in fasting glucose levels (117
versus 120 mg/dL); however, the incidence of mild nocturnal hypoglycemia
was significantly lower among patients treated with insulin glargine than in
the group treated with NPH insulin (P \ 0.001) [50]. There was a reduction
of approximately 45% of nocturnal hypoglycemia with glargine compared
874 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

with NPH [50]. Treatment with NPH or glargine in addition to oral therapy
in type 2 diabetic patients resulted in a decrease of fasting glucose in both
groups, reaching a plateau by 12 weeks. HbA1c declined at a predictably
slower rate, stabilizing after 18 weeks (Fig. 2) [50].

Fig. 2. (A) FPG and (B) HbA1c during the study. Values in both figures are means; error bars
indicate SE. (From Riddle MC, Rosenstock J, Gerich J. The Treat-to-Target Trial: Randomized
addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes
Care 2003;26(11):3080–6; with permission.)
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 875

Selection of patients most likely to succeed on combination treatment


The most common type of patient in whom combination therapy will
succeed is the one who is failing oral treatment without significant glucose
toxicity and has some evidence of responsiveness to oral agents. Patients
have a higher likelihood of success using combination therapy if they are
obese, have had overt diabetes for less than 10 to 15 years, are diagnosed
with type 2 diabetes after the age of 35, do not have fasting blood glucose
values consistently over 250 to 300 mg/dL, and have evidence of endogenous
insulin secretory ability. Although standard measurement conditions and C-
peptide concentrations have not been established for this clinical situation,
a fasting C-peptide concentration (0.2 nmol/L) or glucagon-stimulated
level (0.40 nmol/L) indicates some degree of endogenous insulin secretory
ability [51,52]. Patients with type 2 diabetes diagnosed before the age of 35
more often have atypical forms of diabetes. Patients who have had diabetes
for more than 10 to 15 years tend to have a greater chance of beta-cell
exhaustion and, thus, tend to be less responsive to oral hypoglycemic agents
and combination therapy. Thin patients are more likely to be hypoinsuli-
nemic and often respond inadequately to oral agents, which lead to
combination therapy failure. In addition, markedly elevated fasting glucose
concentration is often associated with a concomitant decrease in endoge-
nous insulin secretory ability, which renders oral agents ineffective. The
actual number of patients who might respond favorably to combination
therapy is unknown but is estimated to be between 20% and 40%.

Initiating combination therapy


Calculation of the initial bedtime dose of intermediate-acting insulin can
be based on clinical judgment or various formulas based on the fasting
blood glucose concentration or body weight. For example, the average
fasting blood glucose (mg/dL) can be divided by 18 or body weight (kg) can
be divided by 10 to calculate the initial dose of NPH or insulin glargine to be
started at bedtime [43]. Also, 5 to 10 units of insulin can be safely started for
thin patients, and 10 to 15 units can be started for obese patients at bedtime,
as an initial estimated dose. In either case, the dose is increased in
increments of 2 to 5 units every 3 to 4 days until the morning fasting blood
glucose concentration is consistently in the range of 70 to 120 mg/dL [53].
The best time to give the evening injection of intermediate-acting insulin
is between 10 PM and midnight. Insulin glargine has been shown to be
effective when taken either in the morning or evening. Many reliable pa-
tients can make their own adjustments using HGM.
Based on the results of HGM, combination therapy can be altered to
reduce hyperglycemia at identified times during the day. For example, a
common situation seen with daytime oral agents and bedtime intermediate-
acting insulin therapy is an improvement in the fasting, pre-lunch, and pre-
dinner blood sugar values, although the post-dinner blood glucose
876 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

concentration remains excessively high (200 mg/dL). In this clinical


situation, an injection of premixed insulin (70/30 or 75/25 mix) can be
given before dinner instead of a bedtime dose of intermediate- or long-acting
insulin. This regimen will often improve the post-dinner blood glucose
values because the premixed insulin contains rapid-acting analogs yet allows
overnight glucose control secondary to the intermediate-acting component.
With this regimen, however, one must be more cautious about early
morning hypoglycemia because the intermediate-acting insulin given before
dinner will exert its peak effect earlier. In the experience of these authors,
this has not been a major clinical problem in obese patients with type 2
diabetes compared with those with type 1 diabetes mellitus. Normally the
dose of bedtime intermediate- or long-acting insulin can be converted to the
dose of premixed insulin, dose per dose, and adjustments can be made
through HGM.

Dose adjustment
Once the fasting blood glucose concentrations are consistently in a desir-
able range, the pre-lunch, pre-dinner, and bedtime blood sugar values must
be monitored to determine if the oral hypoglycemic agents are maintaining
daylong glycemia. It is recommended that after the addition of evening
insulin patients continue to take the maximal dose of the oral sulfonylurea
agent. If the daytime blood glucose concentrations become excessively low,
the dose of oral medication must be reduced. The morning dose of sul-
fonylurea should be reduced or discontinued first. This situation is common
because glucose toxicity may be reduced because of improved glucose
control, leading to enhanced sensitivity to both oral agents and insulin. If
the pre-lunch and pre-dinner blood glucose concentrations remain exces-
sively high on combination therapy, it is likely that the oral agents are not
contributing significantly to glycemic control throughout the day. In this
situation, a more conventional or intensive regimen of two injections per
day is indicated.

Multiple-injection regimens
One of the most common insulin regimens used in type 2 diabetes mellitus
is the split-mixed regimen consisting of a pre-breakfast and pre-dinner dose
of intermediate- and fast-acting insulin. This split-mixed regimen of two
injections per day is often inadequate for patients with type 1 or lean
patients with type 2 diabetes and can result in persistent early morning
hypoglycemia and fasting hyperglycemia. Such problems do not appear to
occur as frequently in obese type 2 diabetes. This is likely caused by
pathophysiologic differences, particularly in endogenous insulin secretory
ability, insulin resistance, and counter-regulatory mechanisms in type 1 and
type 2 diabetes.
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 877

In a landmark trial with type 2 diabetes by Henry et al [54], daylong gly-


cemia and glycosylated hemoglobin were essentially normalized by 6 months
of intensive treatment with a split-mixed insulin regimen. In this study,
14 typical obese patients with type 2 diabetes mellitus (age 59  2 years;
duration of diabetes 7  2 years; body mass index 31  2 kg/m2; fasting
blood glucose concentration 283  13 mg/dL) failing therapy with oral
antidiabetic agents were intensively managed with pre-breakfast and pre-
dinner NPH and regular insulin over a 6-month period. The insulin dose
was adjusted based on HGM results of four injections per day. Glycemic
control was rapidly achieved within 1 month and was maintained for the
duration of the study.
The average total insulin dose needed to maintain glycemic control
approached 100 units per day, with approximately 50% of the total dose
required before breakfast and 50% before dinner. The ratio of NPH to
regular insulin was approximately 75%:25%. There was a very low incidence
of mild hypoglycemic reactions, which decreased as the study progressed,
and no reactions were severe or required assistance. In addition, patient
compliance and sense of well being were excellent. Near-normalization of
the glycosylated hemoglobin, however, led to some adverse effects in these
patients. The mean serum insulin concentration obtained during 24-hour
metabolic profile studies increased from 308  80 pmol/L at baseline to
510  102 pmol/L (P  0.05) at completion of the 6-month study. The
exacerbation of hyperinsulinemia by exogenous insulin therapy was strongly
correlated with weight gain throughout the study. Despite biweekly visits
with the study dietitian and instructions to reduce the daily caloric intake,
a mean weight gain of approximately 9 kg or 18.8 pounds occurred.
Interestingly, the total daily insulin dose was 86 þ 13 units at 1 month and
100 þ 24 units at 6 months, despite minimal additional improvement in
glycemic control during that period. Most of the improvement in glycemic
control was caused by the suppression of basal hepatic glucose production
(from 628 þ 44 to 350 þ 17 lmol/m2/min, P  0.001), with a more modest
but significant improvement in peripheral glucose uptake (from 1418 þ 156
to 1657 þ 128 lmol/m2/min, P  0.05), as determined by the glucose clamp
technique.
This study emphasizes a number of important aspects of intensive glucose
control with insulin in obese subjects with type 2 diabetes. First, the average
daily dose of insulin needed to control such patients approximates 1 unit per
kilogram of body weight. Second, the total daily insulin requirement can be
split equally between the pre-breakfast and pre-dinner injections. Third, the
split-mixed regimen in patients with type 2 diabetes is usually devoid of the
common problems seen with this regimen in type 1 diabetes, particularly
early morning hypoglycemia and fasting (pre-prandial) hyperglycemia.
Fourth, both severe and mild hypoglycemic events are much less fre-
quent in patients with type 2 compared with patients with type 1 diabetes
undergoing intensive insulin therapy. And finally, weight gain with
878 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

peripheral hyperinsulinemia occurs, which may contribute to metabolic and


vascular complications.
A similar but larger 3-month clinical trial [38] compared a split-mixed
combination with a multiple-injection regimen consisting of pre-meal
regular and bedtime NPH insulin injections. Both the split-mixed and
multiple-injection regimen treatment groups achieved equivalent and near-
normal glycosylated hemoglobin values. These therapies, however, were
associated with weight gain of 0.8  0.05 and 2.9  0.05 kg, a 39% and 36%
increase in mean diurnal serum-free insulin levels, and a total daily insulin
dose of 43 and 45 units, respectively. The authors demonstrated that the
change in body weight was negatively correlated with the change in
glycosylated hemoglobin values and positively correlated with the mean
diurnal serum-free insulin values. The differences between these two studies
with regard to total insulin requirements, mean insulin concentrations, and
weight gain are primarily the result of differences in patient characteristics.
Patients in the latter study were leaner (body mass index 29 versus 31 kg/
m2), had lower baseline fasting blood glucose values (225 versus 283 mg/dL)
and reduced baseline mean diurnal serum-free insulin values (138 versus 308
pmol), and were previously treated with submaximal doses of sulfonylureas,
compared with the patients in the former study. In addition, the latter study
was conducted over a shorter period of time (3 months versus 6 months).
Another long-term (5-year) clinical trial using a split-mixed regimen of two
injections per day in 102 nonobese type 2 diabetic patients demonstrated
that excellent glycemic control could be achieved with intensive split-dose
insulin without significant hypoglycemia but at the expense of progressive
weight gain [55]. All these studies clearly demonstrate the efficacy of various
insulin regimens and the adverse consequences of such therapy.

Premixed insulin approach: rapid-acting insulin analogs


Rapid-acting insulin analogs are also available as manufactured, pre-
mixed insulin formulations. One such insulin preparation is Humalog Mix
75/25, which is a fixed-ratio mixture of 25% rapid-acting insulin lispro and
75% novel protamine-based intermediate-acting insulin called neutral
protamine lispro (NPL). NPL was developed to solve the problem of
instability with prolonged storage that occurs with NPH combined with
insulin. Studies of the pharmacokinetic and pharmacodynamic profiles of
NPL show they are comparable to those of NPH insulin [56].
Humalog Mix 75/25 was studied in comparison to premixed human
insulin 70/30 in 89 patients with type 2 diabetes in a 6-month randomized,
open-label, two-period crossover study [57]. All patients had been previously
treated with mixed insulin therapies, including short- or rapid-acting and an
intermediate- or long-acting insulin, twice daily for at least 30 days before
enrollment. During a 2 to 4 week lead-in period, patients were treated with
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 879

human insulin 70/30. The patients were randomized to receive one of two
treatment sequences: therapy twice per day with Humalog Mix 75/25 in-
jected before morning and evening meals for 3 months, after which they
were crossed over to receive human insulin 70/30 using the same dosing
frequency for an additional 3 months, or the alternate treatment sequence.
Patients performed self-monitoring blood glucose (SMBG) at scheduled
intervals during the study period (preprandial, 2-h postprandial, and
occasional 3 AM readings) and recorded this information along with any
hypoglycemic episodes in a study diary. Mean insulin doses were similar or
identical between treatments. Blood glucose values after the morning meal
were significantly lower during treatment with Humalog Mix 75/25
(Humalog Mix 75/25 8.95  2.17 versus human insulin 70/30 10.00 
2.28 mmol/L, P = 0.017). Treatment with Humalog Mix 75/25 produced
similar significant blood glucose results 2 hours after the evening meal as
well (Humalog Mix 75/25 9.28  2.15 versus human insulin 70/30 10.27 
2.76 mmol/L, P = 0.014). Blood glucose results at other time points, HbA1c
levels, daytime hypoglycemia, and nocturnal hypoglycemia were not sig-
nificantly different between treatments. Compared with human insulin 70/
30, twice-daily injections of Humalog Mix 75/25 in patients with type 2
diabetes resulted in improved postprandial glycemic control after the
morning and evening meals, similar overall glycemic control, and the added
convenience of administration immediately before meals.
Insulin aspart, another rapid-acting insulin analog, is available in a
premixed formulation with a protamine-retarded insulin aspart called
Novolog Mix 70/30 (70% insulin aspart protamine suspension and 30%
insulin aspart). A comparison study [58] of the pharmacokinetic and
pharmacodynamic parameters of the Novolog Mix 70/30 and human
insulin 70/30 in healthy patients showed that the faster onset and greater
peak action of insulin aspart was preserved in the aspart mixture.
Another study [59] compared premixed aspart mixture 70/30 with
premixed human insulin 70/30 administered twice daily in a randomized
12-week open-label trial in 294 patients with type 1 and type 2 diabetes.
Patients were instructed to inject the human insulin 70/30 30 minutes before
morning and evening meals and the premixed aspart mixture 10 minutes
before morning and evening meals. SMBG levels and hypoglycemia
incidence were recorded in diaries. Patients required a small increase in
the total daily aspart mixture dose compared with human insulin 70/30
(mean difference at 12 weeks [95% CI 0.01; 0.05]), P  0.01; 0.03 U/kg.
There was no significant difference in HbA1c between groups, yet the mean
blood glucose values after treatment with the aspart mixture showed
statistically significant treatment differences after breakfast, before lunch,
after dinner, and at bedtime. Blood glucose values were approximately
1.0 mmol/L lower compared with the human insulin 70/30 group at each
time point (P  0.05). The incidence of hypoglycemia was not found to be
different between the two groups, and weight gain was not significant during
880 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

the study period with either type of insulin. Treatment with twice-daily
premixed aspart mixture 70/30 resulted in similar overall glycemic control;
yet postprandial control improved without additional hypoglycemia and
with injections immediately before meals compared with premixed human
insulin 70/30 given 30 minutes before the meal.
In a more recent study that focused on changes in lipid levels, Schwartz
et al [60] compared insulin 70/30 mix taken twice per day plus metformin
versus triple oral therapy (secretagogues, metformin, and thiazolidine-
diones) and clearly demonstrated that insulin plus metformin are superior
in lowering total cholesterol and triglycerides levels. The baseline values for
total cholesterol, low-density lipoprotein, high-density lipoprotein, and
triglycerides indicated no differences between the triple OHA and insulin/
metformin groups. By the end of the study (week 24) significant decreases in
total cholesterol and triglycerides were evident in the insulin plus metformin
group (P = 0.038 and 0.033, respectively, compared with the triple oral
therapy group). Subjects in the triple oral therapy group showed a small
increase in cholesterol and less of a decrease in triglyceride levels [60]. For
glucose control, both groups had similar FPG values at the beginning of the
study. After 24 weeks of treatment, the changes from baseline mean FPG
values were 55 and 65 mg/dL for the triple oral therapy and insulin plus
metformin, respectively [60]. Baseline HbA1c values were 9.62  1.25% for
subjects in the triple oral therapy and 9.65  1.62% in the insulin group.
HbA1c values at weeks 2 and 6 demonstrated the efficacy of both treatments;
however, insulin plus metformin treatment achieved improvements in
HbA1c values at weeks 2 and 6 (9.03  1.35% and 8.11  1.20%,
respectively) that were significantly greater than the response to triple oral
therapy (P = 0.001 and 0.001, respectively). At weeks 12 and 24, no
statistically significant difference in HbA1c between the two groups were
observed (final values at week 24 were 7.59  1.4% for triple oral therapy
and 7.59  1.25% for insulin plus metformin [P = 0.772]) (Fig. 3) [60].
Along with SMBG, the use of rapid-acting premixed insulin analogs is
convenient and can be beneficial in reducing postprandial hyperglycemia
and in helping patients achieve glycemic control without the increased
incidence of hypoglycemia. In addition, protocols are currently underway to
assess the efficacy of using rapid-acting premixed insulin analogs three times
per day before breakfast lunch and dinner, based on HGM data. This
regimen is more related to a basal bolus strategy discussed below.

Basal–bolus strategy
The basal–bolus insulin strategy, which can be used in patients with
either type 1 or type 2 diabetes, incorporates the concept of providing
continuous basal insulin levels throughout the day and night with brief
increases in insulin levels at the time of meal ingestion by bolus doses.
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 881

Fig. 3. (A) Mean FPG values at screening and weeks 12 and 24 by treatment group. No
statistically significant changes were observed between the triple oral therapy (•) and insulin
plus metformin (). (B) Mean  SEM changes for the total cholesterol, HDL, LDL, and
triglycerides at week 24. * Statistically significant (P  0.05) reduction in total cholesterol and
triglyceride levels in the insulin plus metformin group compared with the triple oral therapy
group. HDL, high-density lipoprotein; LDL, low-density lipoprotein; OHA, XXX. (From
Schwartz S, Sievers R, Strange P Lyness W, Hollander P. Insulin 70/30 mix plus metformin
versus triple oral therapy in the treatment of type 2 diabetes after failure of two oral drugs:
efficacy, safety, and cost analysis. Diabetes Care 2003;26(8):2598–603; with permission.)

The use of pre-meal regular insulin with bedtime NPH as the basal insulin
has been a common strategy for intensive insulin therapy in the United
States, but because regular insulin should be administered 20 to 40 minutes
before meals, a risk of hypoglycemia exists if the meal is delayed. If regular
882 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

insulin is given just before a meal, high postprandial glucose levels and
delayed hypoglycemia may result. A strategy that provides for some
flexibility in the mealtime administration of insulin with the use of rapid-
acting insulin analogs, lispro or aspart, administered immediately before
meals, and long-acting insulin, such as glargine, ultralente, lente, or NPH as
the basal insulin. These regimens that use multiple doses of intermediate-
acting insulin such as NPH can be associated with unpredictable nocturnal
hypoglycemia and day-to-day instability of blood glucose patterns in part
because of intrapatient variability of the effect of subcutaneous injected
insulin and the patient’s peak action profile [61]. NPH, which exhibits peak
action 5 to 7 hours after administration, has also been used in combination
with rapid-acting insulin analogs, commonly given at least twice daily,
although the disadvantages of NPH used in this manner are similar to those
associated with Ultralente [62]. Because of its time to peak action, NPH
should be given every 6 hours or 4 times per day to be effective as a basal
insulin, in many patients [63].
Improved mealtime glucose control with the rapid-acting analogs has
exposed the gaps in basal insulin coverage provided by therapy with the
traditional intermediate- and long-acting insulin preparations. Taking
a basal insulin analog with a relatively constant and flat pharmacokinetic
profile such as insulin glargine once per day will result in a more
physiologic pattern of basal insulin replacement. Insulin glargine in
combination with a rapid-acting insulin analog has demonstrated effective
glycemic control and a lower incidence of nocturnal hypoglycemia [64] than
other insulin preparations currently used for basal insulin supplementation
[64–68].
Patients on multiple-injection basal–bolus regiments should use carbo-
hydrate counting to estimate their pre-meal bolus dose of a rapid-acting
analog. In addition, a correction factor should be determined by HGM
before and after rapid-acting insulin boluses. For example, a typical
insulin-resistant subject with type 2 diabetes may need 1 unit of lispro or
aspart for every 8 g of carbohydrate compared with a 1:15 ratio for a lean
insulin-sensitive person with type 1 diabetes. A typical correction factor
would be 1 unit of lispro or aspart to bring down the blood glucose value
to 25 mg/dL compared with a person with type 1 diabetes whose cor-
rection factor is 1 unit of lispro or aspart to bring down the blood glucose
value to 50 mg/dL. The carbohydrate to insulin ratio and correction factor
may be different depending on the time of the day and degree of
hyperglycemia.
The availability of mealtime and basal insulin analogs, combination
therapy with oral agents, and the use of insulin regimens comprising
basal and mealtime (bolus) insulin components that better simulate
normal insulin secretion represent important advances in insulin therapy.
All of these approaches can have a significant impact on treatment
outcomes.
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 883

External insulin pump therapy


External insulin pump therapy or continuous subcutaneous insulin
infusion (CSII) has been traditionally used mainly in people with type 1
diabetes. However, insulin pump therapy is extremely valuable in patients
with type 2 diabetes who require insulin but who have not achieved glycemic
control with subcutaneous injections or who are seeking for a more flexible
lifestyle. All of the benefits that are enjoyed by patients with type 1 diabetes
are shared with people with type 2 diabetes. Many experts believe that
because of the more physiologic delivery of insulin, glucose control is
achieved with less insulin than was needed with the subcutaneous insulin
regimen. This may be caused by a reduction in glucose toxicity and im-
provement of insulin resistance and beta-cell secretory function as a result of
improved glycemic control with pump therapy. Weight gain is less of an
issue because the patient is using less insulin than was used before insulin
pump therapy. In addition, with the reduction of hypoglycemic events there
is less overeating to compensate for excessive insulin. Last, it is possible that
pump therapy may result in less strain placed on the pancreatic beta-cells of
patients with type 2 diabetes, and this may help with overall glycemic
control because a functioning beta-cell can also autoregulate against hyper-
and hypoglycemia, as seen in non-diabetic individuals.
Many older patients with the diagnosis of ‘‘insulin-requiring type 2
diabetes’’ have acute, true, late-onset type 1 diabetes. The literature
documents large groups of patients with insulin-requiring type 2 diabetes
who were tested for anti-glutamic acid decarboxylase antibodies with a
positivity rate of approximately 5% to 8%. These individuals are thinner at
the time of diagnosis, generally do not respond well to oral agents, and
require insulin, although they do not present in severe diabetic ketoacidosis.
These patients generally should be put on an intensive insulin injection
regimen, and insulin pump therapy should be considered.
Insulin pump therapy allows for increased flexibility in meal timing and
amounts, increased flexibility in the time and intensity of exercise, improved
glucose control while traveling across time zones or with variable working
schedules, and quality of life in terms of self-reliance and control.
Because pumps use only regular and fast-acting insulin, there is no
peaking of injected intermediate- and long-acting insulins, which do not
provide as constant a basal rate caused by variable absorption and
pharmacokinetics. Insulin glargine is an exception in that it serves as
excellent basal insulin. Variable insulin absorption and pharmacokinetics
are probably responsible for up to 50% to 60% of the day-to-day
fluctuation in blood glucose values in individuals using multiple-injection
regimens with various insulin types. Insulin pump therapy allows for more
regular insulin absorption and pharmacokinetic profile, resulting in im-
proved reproducibility in insulin availability and reduced fluctuations in
glycemic control [62].
884 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

Presently, there is a paucity of clinical trials using insulin pumps in type 2


diabetes, but pump therapy is a viable option in insulin-requiring patients
with type 2 diabetes who are unable to achieve adequate glycemic control
with multiple-injection regimens. Although some studies demonstrate
metabolic benefits of pump therapy in type 2 diabetes, all are limited by
a relatively short period of evaluation and a small number of heterogeneous
subjects. Interpretation of these studies is further confounded by the
random assignment of subjects to dissimilar conventional insulin regimens,
making comparison between studies difficult.
Garvey et al [69] studied the effect of intensive insulin therapy on insulin
secretion and insulin action before and after 3 weeks of CSII therapy in 14
patients with type 2 diabetes (age 50  3 years, duration of diabetes 7.8 
2.1 years, and 119% ideal body weight). In 3 weeks of therapy, the mean
fasting plasma blood glucose and glycosylated hemoglobin values fell 46%
and 38%, respectively. The mean daily insulin dose stabilized at approxi-
mately 110 units/d, and there was a 74% increase in the insulin-stimulated
glucose disposal rate and a 45% reduction in hepatic glucose output to
mean levels similar to those of normal subjects. In addition, there were
significant improvements in both endogenous insulin and C-peptide
secretion. This study demonstrated that pump therapy was feasible and ef-
fective at improving metabolic control and reversing glucose toxicity in
these poorly controlled subjects with type 2 diabetes.
Jennings et al [70] randomized 20 type 2 diabetic subjects (median age 61
years, duration of diabetes 6 years, and percentage of ideal body weight
120%) to either CSII or twice-daily injections of regular and NPH insulin for
4 months. Glycemic control improved in both groups, although there was
a 30% reduction in the glycosylated hemoglobin in the CSII-treated group
and only a 17% reduction in the twice-daily injection-treated group. There
were no significant differences between the two groups in median daily
insulin requirement (0.58 versus 0.65 units/kg), weight gained (4.5 versus
4.2 kg), prevalence of mild hypoglycemic reactions, or patient acceptance. In
addition, in the CSII group 58% of the total daily insulin requirement was
given as a basal infusion, with the remainder as pre-meal bolus injections
using insulin algorithms. This ratio of basal to bolus insulin requirements are
similar to the rates commonly used in type 1 diabetes, but there are
characteristics of pump therapy that are very different in type 2 diabetes.
In a more recent study, Pouwels et al [71] prospectively studied 8 patients
with poorly controlled (HbA1c 12.0  1.7%) type 2 diabetes with high
insulin requirements (1.92  0.66 U/kg/d). The subjects where aggressively
treated with intravenous (IV) insulin for approximately 1 month followed by
12 months of CSII therapy. Insulin sensitivity and secretion were measured
before and after the IV insulin treatment phase.
Euglycemia was achieved after 12 days of IV therapy, and the insulin
requirements eventually reduced from 1.7  0.09 to 1.1  0.06 U/kg/d (P 
0.005) during the IV treatment phase of the protocol. Whole-body glucose
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 885

uptake increased from 12.7  5.7 to 22.4  8.8 lmol/kg/min (P  0.0005), and
in 1 month of intensive therapy, the HbA1c dropped to 8.9%  1.2% with no
significant change in the patients’ body weight. After 6 and 12 months of CSII
therapy, the mean HbA1c values were 7.1  0.6% and 8.3  1.4%, respectively
(P  0.001 versus pretreatment values for all time points) [71].
In another recent study [72], 132 CSII naıı̈ve type 2 diabetics were
randomized to the pump or multiple daily injections (MDI). This study
showed that pump therapy provided efficacy and safety equivalent to MDI
therapy. Lower 8-point blood glucose values were shown by the CSII group at
most time points (values were only significant 90 min after breakfast; 167 
47.5 mg/dL versus 192  65.0 mg/dL for CSII and MDI, respectively;
P = 0.019) (Fig. 4).
In summary, insulin pump therapy has not been fully evaluated in
patients with type 2 diabetes. From published studies, however, it is
apparent that CSII therapy can safely improve glycemic control while
limiting hypoglycemia. CSII may be particularly useful in treating patients
with type 2 diabetes who do not respond satisfactorily to more conventional
insulin treatment strategies.

Alternative insulin delivery systems


Inhaled insulin is currently under development by several pharmaceutical
companies for use in people with type 1 and type 2 diabetes. The insulin is

Fig. 4. Baseline and end-of-study 8-point blood glucose profiles (mean  SEM) for the intent-
to-treat population. Dashed lines represent baseline profiles; solid lines represent end-of-study
profiles; •, means for CSII; n, means for MDI therapy. Number of patients at each time point:
CSII, 56–63; MDI, 54–59. *P, 0.02. BB, before breakfast; B90, 90 minutes after breakfast;
BL, before lunch; L90, 90 minutes after lunch; BD, before dinner; D90, 90 minutes after dinner;
BE, at bedtime. (From Raskin P, Bode BW, Marks JB, Hirsh IB, Weinstein RL, McGill
JB, Peterson GE, Mudaliar SR, Reinhardt RR. Continuous subcutaneous insulin infusion and
multiple daily injection are equally effective in type 2 diabetes: a randomized, parallel-group,
24-week study. Diabetes Care 2003;26(9):2598–603; with permission.)
886 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

contained in a pellet and is vaporized in an inhaler, which aerosolizes the


liquid insulin. Inhaled insulin can also be delivered to the pulmonary
microvasculature as a dry powder system and inhaled through a mouthpiece.
It provides the obvious incentive for diabetic patients to use insulin without
the need for injections.
Cefalu et al [73] conducted a randomized, open-label, 3-month study with
26 patients (16 men, 10 women; average age, 51.1 years) with type 2 diabetes
(average duration of diabetes, 11.2 years). Patients received inhaled insulin
before each meal plus a bedtime injection of ultralente insulin, performed
HGM, and adjusted their insulin dose weekly. The target level for
preprandial plasma glucose was 100 to 160 mg/dL. At the end of 3 months,
inhaled insulin treatment significantly improved glycemic control compared
with baseline, and mean HbA1c levels decreased by 0.07%. Hypoglycemic
events were mild, and patients showed no significant weight gain or change
in pulmonary function compared with baseline. Thus in this study, pul-
monary delivery of insulin in type 2 diabetic patients who require insulin
improved glycemic control was well tolerated and demonstrated no short-
term adverse pulmonary effects [74].
A new Aerodose insulin inhaler proved to be comparable to sub-
cutaneous injections through overlapping dose-response curves, with con-
sistent relative bioavailability and relative biopotency. The inhaler delivered
a pharmacologically predictable insulin dose to type 2 diabetics, similar
to that with subcutaneous insulin injections [75]. The Aerodose inhaler used
regular insulin (Humulin R), U-500, whereas Humulin R, U-100 was used
for subcutaneous injections. Serum insulin levels before exogenous insulin
administrations were similar between inhaled and subcutaneously inject
insulin (t, baseline, P = 0.12). At the end of dosing (t, 0 min), serum insulin
levels were significantly higher for inhalation treatments than for sub-
cutaneously injected treatments, indicating rapid, systemic insulin absorp-
tion following inhalation. The area under the curve (AUC)0-8h and
maximum serum insulin concentration demonstrated a clear dose-response
relationship for the three doses of inhaled insulin and the three doses of
subcutaneously injected insulin (Fig. 5) [75].
Insulin can also be taken orally by capsules, enterocoated with a soybean
trypsin inhibitor that prevents insulin degradation. This approach has
clinical potential, but large clinical trials have not been carried out.
Chemically modified human insulin, called hexyl insulin, using proprietary
conjugation technology to improve its stability and oral absorption has
shown promise. Preliminary results reported that in healthy human
volunteers, hexyl insulin caused dose-dependent glucose lowering, was safe,
and was well tolerated. In a small clinical study, oral insulin illustrates the
similarities and differences among hexyl insulin monoconjugate (HIM)2 oral
insulin, subcutaneous insulin, and placebo. All three curves are indistin-
guishable from each other during the first hour postdose. The placebo curve
then separates from the other two curves, displaying a significantly higher
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 887

Fig. 5. Glucose infusion rate (GIR) registered following administration of inhaled insulin (¤,
80 units; n,160 units; •, 240 units) and subcutaneous injection (}, 8 units; , 16 units; , 24
units) in patients with type 2 diabetes. GIRs have been averaged over 30-minute periods. Data
points are means  SE (n = 16) at each time point for low, medium, and high doses for inhaled
and injected insulin. (From Kim D, Mudaliar S, Chinnapongse S, Chu N, Boies SM, Davis T,
Perera AD, Fishman RS, Shapiro DA, Henry R. Dose-response relationships of inhaled insulin
delivered via the Aerodose insulin inhaler and subcutaneously injected insulin in patients with
type 2 diabetes. Diabetes Care 2003;26(10):2842–7; with permission.)

peak excursion. The HIM2 and subcutaneous insulin curves remain nearly
indistinguishable for at least another hour. During the fourth hour
postdose, the HIM2 curve clearly separates from the subcutaneous insulin
curve, becoming nearly identical to the placebo curve, as the glucose
excursion values in all three groups decline toward baseline. Peripheral
plasma insulin revealed an initial peak in peripheral plasma insulin
concentrations following administration of HIM2; however, this initial
insulin peak was caused by one patient who had a rapid peak of insulin. The
median insulin Cmax values, following administration of HIM2 and sub-
cutaneous regular insulin, were nearly identical (Fig. 6) [76].

Amylin analog: a novel injectable peptide that compliments the action


of insulin
Destruction and dysfunction of pancreatic beta-cells, resulting in
absolute and relative insulin deficiency, represent key abnormalities in the
pathogenesis of type 1 and type 2 diabetes, respectively [77]. Following the
888 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 889

discovery of amylin in 1987, a second beta-cell 37-amino acid hormone that


is co-secreted with insulin in response to nutrient stimuli, it was realized that
diabetes represents a state of bi-hormonal beta-cell deficiency and that a lack
of amylin action may contribute to abnormal glucose homeostasis.
Experimental studies show that amylin acts as a neuroendocrine hormone
that complements the effects of insulin in postprandial glucose regulation
through several centrally mediated effects. These include a suppression of
postprandial glucagon secretion and a vagus mediated regulation of gastric
emptying, thereby helping to control the inflow of endogenous and
exogenous glucose, respectively. In animal studies, amylin also reduces
food intake and body weight, consistent with an early satiety effect [78].
Insulin is the major hormonal regulator of glucose disposal. Preclinical
and clinical studies indicate that amylin complements the effects of insulin
by regulating the rate of glucose inflow to the bloodstream, suppressing
glucagons secretion and inducing satiety.
Pramlintide is a soluble, nonaggregating, injectable, synthetic analog of
human amylin currently under development for the treatment of type 1 and
insulin-using type 2 diabetes. Long-term clinical studies have consistently
demonstrated that prandial subcutaneous. injections of pramlintide, in
addition to the current insulin regimen, reduce HbA1c and body weight in
type 1 and type 2 diabetic patients, without an increase in insulin use or in
the incidence of severe hypoglycemia [79].
Treatment with 120 lg twice per day of pramlintide in subjects with type
2 diabetes led to a sustained reduction from baseline in HbA1c (0.68 and
0.62% at weeks 26 and 52, respectively) that was significantly greater than
that in the placebo group (P \ 0.05) (Fig. 7) [80]. The greater reduction of
HbA1c observed with pramlintide was not accompanied by an increase in
body weight. Instead, patients in both pramlintide treatment groups
experienced a sustained reduction in body weight that was significantly
different from placebo at week 26 (both P  0.05) [80]. In the group of
subjects who took 120 lg twice daily, the reduction in body weight was
sustained to week 52 (P  0.05 versus placebo).
The most commonly observed side effects were gastrointestinal-related,
mainly mild nausea, which typically occurred on initiation of treatment and

=
Fig. 6. 0.5 mg/kg and 1.0 mg/kg HIM2 dose groups; pooled data. (A) Mean plasma glucose
excursion versus time profiles and (B) mean plasma insulin concentration versus time profiles.
At time 0, patients received 0.5 or 1.0 mg/kg oral HIM2, 8 units subcutaneous regular insulin or
oral placebo. At 30 min, patients began ingesting the standardized meal (Boost Plus). Patients
ingested the entire meal over a 10-minute period. Postprandial plasma glucose excursions and
insulin concentrations were determined from blood samples collected a the time points
indicated. Data are expressed as means  SE (n = 12 patients). , oral HIM2 (0.5) and 1.0 mg/kg
dose groups combined); , 8 units of subcutaneous regular insulin; , oral placebo. (From Kipnes
M, Dandona P, Tripathy D, Still JG, Kosutic G. Control of postprandial plasma glucose by an
oral insulin product (hexyl-insulin monoconjugate 2 [HIM2]) in patients with type 2 diabetes.
Diabetes Care 2003;26(2):421–6; with permission.)
890 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

Fig. 7. Change from baseline in mean HbA1c (A) and weight (B) (intent-to-treat population).
* P  0.05 for treatment arm versus placebo. , placebo; n, 90 lg twice daily; •, 120 lg. (From
Hollander PA, Levy P, Fineman MS, Maggs DG, Shen LZ, Strobel SA, et al. Pramlintide as
an adjunct to insulin therapy improves long-term glycemic and weight control in patients with
type 2 diabetes: a 1-year randomized controlled trial. Diabetes Care 2003;26(3):784–90; with
permission.)

resolved within days or weeks. Amylin replacement with pramlintide as an


adjunct to insulin therapy is a novel physiological approach toward
improved long-term glycemic and weight control in patients with type 1
and type 2 diabetes.

Summary
Type 2 diabetes is a common disorder often accompanied by numerous
metabolic abnormalities leading to elevated rates of cardiovascular morbid-
ity and mortality. Improved glycemia will delay or prevent the development
of microvascular disease and reduce many or all of the acute and subacute
complications that worsen the quality of daily life. Exogenous insulin is
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 891

usually the last line of treatment used to normalize glycosylated hemoglobin


in patients with type 2 diabetes who have failed other therapeutic modalities.
Not all patients are candidates for aggressive insulin management; therefore,
the goals of therapy should be tailored to the individual. Candidates for
intensive management should be motivated, compliant, educable, and
without other medical conditions and physical limitations that would
preclude accurate and reliable HGM and insulin administration.
In selected patients, combination therapy with insulin and oral antidia-
betic medications can be an effective method for normalizing glycemia
without the need for rigorous insulin regimens. The most common clinical
situation in which combination therapy can be successful occurs in patients
who are failing daytime oral agents therapy and still show some evidence of
responsiveness to the medications. Bedtime intermediate- and long acting-
insulin are administered and progressively increased until the fasting blood
glucose concentration is normalized. Additional benefits of combination
therapy include ease of administration, excellent patient compliance and
safety, and lower exogenous insulin requirements with less peripheral
hyperinsulinemia and weight gain. If combination therapy is not successful,
a split-mixed regimen of an intermediate- and a fast-acting insulin equally
divided between the pre-breakfast and pre-dinner periods can be effective
especially in obese patients.
For patients who do not achieve glucose control on combination or split-
mixed regimens, an intensive basal bolus multiple-injection regimen is
indicated. Continuous subcutaneous insulin infusion pumps can be partic-
ularly useful in treating patients with type 2 diabetes mellitus who do not
respond satisfactorily to more conventional treatment strategies. The use of
fast-acting insulin analogs should be used in the majority of insulin-requiring
diabetics because of its more physiologic pharmacokinesis. Inhaled insulin
and the amylin analog pramlintide also hold promise to intensively control
glycemia in patients with insulin-requiring type 2 diabetes.
The glycemic objectives for patients with type 2 diabetes should be similar
to those for patients with type 1 diabetes, namely, to normalize glycemia and
glycosylated hemoglobin without causing undue weight gain or hypoglyce-
mia or adversely affecting the quality of daily life. This is best achieved in
a multidisciplinary setting using complementary therapeutic modalities that
include a combination of diet, exercise, and pharmacologic therapy.
Emphasis should be placed on diet and exercise initially, and throughout
the course of management as well, since even modest success with these
therapies will enhance the glycemic response to both oral antidiabetic agents
and insulin.

References
[1] American Podiatric Medical Association. E-foot FAQS [frequently asked questions]. 2002.
Available at: http://www.apma.org/faqsdiab.html. Accessed March 12, 2003.
892 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

[2] Bennett PH, Rewers MJ, Knowler WC. Epidemiology of diabetes mellitus. In: Porte DJ,
Sherwin RS, editors. Ellenberg and Rifkin’s diabetes mellitus. Stamford: Appleton &
Lange; 1997. p. 373–97.
[3] Sherwin RS. Diabetes mellitus. In: Goldman L, Bennett JC, editors. Cecil textbook of
medicine. Philadelphia: WB Saunders; 2000. p. 1263–85.
[4] American Academy of Family Physicians. Number of visits to family physicians by the 45
most frequent primary diagnoses. 2002. National Ambulatory Medical Care Survey 2000.
Department of Health and Human Services, Centers for Disease Control and Preven-
tion. National Center for Health Statistics. Available at: http://www.aafp.org/practice/
icd9code.html. Accessed March 12, 2003.
[5] Hiatt WR. Atherosclerotic peripheral arterial disease. In: Goldman L, Bennett JC, editors.
Cecil textbook of medicine. Philadelphia: WB Saunders; 2000. p. 357–62.
[6] Prospective Diabetes Study UK (UKPDS) Group. Intensive blood-glucose control with
sulphonylureas or insulin compared with conventional treatment and risk of complications
in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.
[7] Shichiri M, Kishikawa H, Ohkubo Y, Wake N. Long-term results of the Kumamoto Study
on optimal diabetes control in type 2 diabetic patients. Diabetes Care 2000;23(Suppl 2):
B21–9.
[8] Galloway JA. Treatment of NIDDM with insulin agonists or substitutes. Diabetes Care
1990;13:1209.
[9] DeFronzo RA. Insulin resistance, hyperinsulinemia, and coronary artery disease: A com-
plex metabolic web. J Cardiovasc Pharm 1992;20(Suppl 11):S1–16.
[10] Weyer C, Bogardus C, Mott DM, Pratlet RE. The natural history of insulin secretory
dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin
Invest 1999;104:787–94.
[11] American Diabetes Association. The pharmacologic treatment of hyperglycemia in
NIDDM. Diabetes Care 1995;18:1510–8.
[12] Caro JF. Insulin resistance in obese and nonobese man (Clinical Review 26). J Clin
Endocrinol Metab 1991;73:691.
[13] Genuth S. Insulin use in NIDDM. Diabetes Care 1990;13:1240.
[14] Duckworth WC, Saudek CD, Henry RR, Veterans Affairs Study Group. Perspectives in
diabetes: why intraperitoneal delivery of insulin with implantable pumps in NIDDM?
Diabetes 1992;41:657.
[15] Saudek CD, Duckworth WC, Veterans Affairs Study Group. The Department of Veterans
Affairs implanted insulin pump study. Diabetes Care 1992;15:567.
[16] Department of Veterans Affairs Implantable Insulin Pump Study Group. Implantable
insulin pump vs multiple-dose insulin for non-insulin-dependent diabetes mellitus. JAMA
1996;276:1322–7.
[17] American Diabetes Association. Clinical practice recommendations american diabetes
association 1999. Diabetes Care 1999;22(Suppl 1):2.
[18] Colwell JA. Controlling type 2 diabetes: are the benefits worth the costs? JAMA 1997;
278(20):1700.
[19] Tuomi T, Groop LC, Zimmet PZ, Rowley MJ, Knowles W, Mackay IR. Antibodies to
glutamic acid decarboxy - lase reveal latent autoimmune diabetes mellitus in adults with
a non-insulin-dependent onset of disease. Diabetes 1993;42:359.
[20] American Diabetes Association. Diabetes Dictionary. 2004. Available at: http://www.
diabetes.org/diabetesdictionary.jsp. Accessed on October 4, 2003.
[21] Yu A, Wu PS, Edelman SV. The natural history of non-insulin dependent diabetes mellitus
in a Filipino migrant population. Presented at the 3rd International Diabetes Federation,
Western Pacific Regional Congress: Hong Kong; September 25–28, 1996.
[22] Vora JP, Burch A, Peters JR, Owens DR. Relationship between absorption of radiolabeled
soluble insulin, subcutaneous blood flow, and anthropometry. Diabetes Care 1992;15:
1484–93.
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 893

[23] Edelman SV. Taking control of your diabetes. Caddo (OK): Professional Communica-
tions, Inc., 2000.
[24] Bailey TS, Mezitis NHE. Combination therapy with insulin and sulfonylureas for type 2
diabetes. Diabetes Care 1990;13:687.
[25] Lebovitz HE, Pasmantier R. Combination insulin-sulfonylurea therapy. Diabetes Care
1990;12:667.
[26] Groop LC, Widen E, Ekstrand A, Saloranta C, Franssila-Kallunki A, Schalin-Jantti C,
et al. Morning or bedtime NPH insulin combined with sulfonylurea in treatment of
NIDDM. Diabetes Care 1992;15:831.
[27] Pugh JA, Wagner ML, Sawyer J, Ramirez G, Tuley M, Friedberg SJ. Is combination
sulfonylurea and insulin therapy useful in NIDDM patients? A metaanalysis. Diabetes
Care 1992;15:953.
[28] Riddle MC. Evening insulin strategy. Diabetes Care 1990;13:676.
[29] Taskinen M-R, Sane T, Helve E, Karonen SL, Nikkila EA, Yki-Jarvinen H. Bedtime
insulin for suppression of overnight free-fatty acid, blood glucose, and glucose production
in NIDDM. Diabetes 1989;38:580.
[30] Karlander SG, Gutniak MKM, Efendic S. Effects of combination therapy with glyburide
and insulin on serum lipid levels in NIDDM patients with secondary sulfonylurea failure.
Diabetes Care 1991;14:963.
[31] Mezitis NHE, Heshka S, Saitas V, Bailey TS, Costa R, Pi-Sunyer FX. Combination therapy
for NIDDM with biosynthetic human insulin and glyburide. Diabetes Care 1992;15:265.
[32] Vigneri R, Trischitta V, Italia S, Mazzarino S, Rabuazzo MA, Squatrito S. Treatment of
NIDDM patients with secondary failure to glyburide: Comparison of the addition of either
metformin or bedtime NPH insulin to glyburide. Diabetes Metab 1991;17:232.
[33] Raskin P. Combination therapy in NIDDM. N Engl J Med 1992;327:1453.
[34] Marks JB. Combination glyburide and insulin treatment in secondary sulfonylurea failure.
Clinical Diabetes 1994;12:21.
[35] Yki-Jarvinen H, Ryysy L, Nikkila K, Tulokas T, Vanamo R, Heikkila M. Comparison
of bedtime insulin regimens in patients with type 2 diabetes mellitus. A randomized,
controlled trial. Ann Int Med 1999;130(5):389–96.
[36] Riddle MC, Schneider J. Beginning insulin treatment of obese patients with evening 70/30
insulin plus glimepiride versus insulin alone. Glimepiride Combination Group. Diabetes
Care 1998;21(7):1052–7.
[37] Johnson JL, Wolf SL, Kabadi UM. Efficacy of insulin and sulfonylurea combination
therapy in type II diabetes. A meta-analysis of the randomized placebo-controlled trials.
Arch Intern Med 1996;156:259–64.
[38] Yki-Jarvinen H, Kauppila M, Kujansuu E, Lahti J, Marjanen T, Niskanen L, et al.
Comparison of insulin regimens in patients with non-insulin-dependent diabetes mellitus.
N Engl J Med 1992;327:1426.
[39] Riddle MC, Hart J, Bingham P, Garrison C, McDaniel P. Combined therapy for obese type 2
diabetes: suppertime mixed insulin with daytime sulfonylurea. Am J Med Sci 1992;303:151–6.
[40] Shank ML, Del Prato S, DeFronzo RA. Bedtime insulin/daytime glipizide: effective
therapy for sulfonylurea failures in NIDDM. Diabetes 1995;44:165–72.
[41] Koivisto VA. Insulin therapy in type 2 diabetes. Diabetes Care 1993;16(Suppl 3):S29.
[42] Riddle MC. Evening insulin strategy. Diabetes Care 1990;13:676.
[43] Turner RC, Holman RR. Insulin use in NIDDM. Rationale based on pathophysiology
of disease. Diabetes Care 1990;13:1011.
[44] Wulffele MG, Kooy K, Lehert P, Bets D, Ogterop JC, Van Der Burg BB, et al.
Combination of insulin and metformin in the treatment of type 2 diabetes. Diabetes Care
2002;25(12):2133–40.
[45] Buse JB, Gumbiner B, Mathias NP, Nelson DM, Faja BW, Whitcomb RW, and the
Troglitazone Insulin Study Group. Troglitazone use in insulin-treated type 2 diabetic
patients. Diabetes Care 1998;21(9):1455–61.
894 T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895

[46] Schwartz S, Raskin P, Fonseca V, and the Troglitazone Exogenous Insulin Study Group.
Effects of troglitazone in insulin-treated patients with type 2 diabetes. Ann Intern Med
1998;338:861–6.
[47] Rubin C, Egan J, Schneider R, et al. Combination therapy with pioglitazone and
insulin in patients with type 2 diabetes [abstract 0474]. Diabetes 1999;48(Suppl 1):A110.
[48] Raskin P, Dole JF, Rappaport EB. Rosiglitazone improves glucose control in poorly
controlled, insulin treated patients with type 2 diabetes mellitus [abstract 0404]. Diabetes
1999;48(Suppl 1):A95.
[49] Chiasson J-L, Josse RG, Hunt JA, Palmason C, Rodger NW, Ross SA, et al. The efficacy
of acarbose in the treatment of patients with non–insulin-dependent diabetes mellitus:
a multicenter controlled clinical trial. Ann Intern Med 1994;121:928–35.
[50] Riddle MC, Rosenstock J, Gerich J. The Treat-to-Target Trial: Randomized addition of
glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care
2003;26(11):3080–6.
[51] Polonsky KS, Rubenstein AH. C-peptide as a measure of the secretion and hepatic
extraction of insulin. Pitfalls and limitations. Diabetes 1984;33:486.
[52] Greco AC, Caputo S, Bertoli A, Ghirlanda G. The beta cell function in NIDDM patients
with secondary failure: A three year follow-up of combined oral hypoglycemic and insulin
therapy. Horm Metab Res 1992;24:280.
[53] Edelman SV, Henry RR. Diagnosis and management of type 2 diabetes. Caddo (OK):
Professional Communications, Inc.; 2002. p. 132.
[54] Henry RR, Gumbiner B, Ditzler T, Wallace P, Lyon R, Glauber HS. Intensive
conventional insulin therapy for type II diabetes. Metabolic effects during a 6-mo out-
patient trial. Diabetes Care 1993;16(1):21–31.
[55] Kudlacek S, Schernthaner G. The effect of insulin treatment on HbAlc, body weight and
lipids in type 2 diabetic patients with secondary failure to sulfonylureas. A five year follow-
up study. Horm Metab Res 1992;24:478.
[56] Roach P, Woodworth JR. Clincal pharmacokinetics and pharmacodynamics of insulin
lispro mixtures. Clini Pharmacokinet 2002;41(13):1043–57.
[57] Roach P, Yue L, Arora V for the Humalog Mix25 Study Group. Improved postprandial
glycemic control during treatment with Humalog Mix25, a novel protamine-based insulin
lispro formulation. Diabetes Care 1999;22(8):1258–61.
[58] Jacobsen LV, Sogaard B, Riis A. Pharmacokinetics and pharmacodynamics of a premixed
formulation of soluble and protamine-retarded insulin aspart. Eur J Clin Pharmacol 2000;
56(5):399–403.
[59] Boehm BO, Home PD, Brehend C, Kamp NM, Lindholm A. Premixed insulin aspart 30
vs. premixed human insulin 30/70 twice daily: a random mixed trial in type 1 and type 2
diabetic patients. Diabet Med 2002;19:393–9.
[60] Schwartz S, Sievers R, Strange P, Lyness W, Hollander P. Insulin 70/30 mix plus
metformin versus triple oral therapy in the treatment of type 2 diabetes after failure of two
oral drugs: efficacy, safety, and cost analysis. Diabetes Care 2003;26(8):2238–43.
[61] Leahy JT. Intensive insulin therapy in type 1 diabetes mellitus. In: Leahy JT, Cefalu WT,
editors. Insulin therapy. New York: Marcel Dekker, Inc.; 2002. p. 87–112.
[62] Lepore M, Pampanelli S, Fanelli C, et al. Pharmacokinetics and pharmacodynamics of
subcutaneous injection of long-acting human insulin analog glargine, NPH insulin, and
ultralente human insulin and continuous subcutaneous infusion of insulin lispro. Diabetes
2000;49:2142–8.
[63] Rossetti P, Pampanelli S, Fanelli C, Porcellati F, Costa E, Torlone E, et al. Intensive
replacement of basal insulin in patients with type 1 diabetes given rapid-acting insulin
analog at mealtime. Diabetes Care 2003;26(5):1490–7.
[64] Yki-Järvinen H, Dressler A, Ziemen M for the HOE 901/3002 Study Group. Less
nocturnal hypoglycemia and better post-dinner glucose control with bedtime insulin
T. Davis, S.V. Edelman / Med Clin N Am 88 (2004) 865–895 895

glargine compared with bedtime NPH insulin during insulin combination therapy in type 2
diabetes. Diabetes Care 2000;23:1130–6.
[65] Raskin P, Klaff L, Bergenstal R, Halle JP, Donley D, Mecca T. A 16-week comparison of
the novel insulin analog insulin glargine (HOE 901) and NPH human insulin used with
insulin lispro in patients with type 1 diabetes. Diabetes Care 2000;23:1666–71.
[66] Heinemann L, Linkeschova R, Rave K, Hompesch B, Sedlak M, Heise T, et al. Time-
action profile of the long-acting insulin analog insulin glargine (HOE 901) in comparison
with those of NPH insulin and placebo. Diabetes Care 2000;23:644–9.
[67] Rosenstock J, Park G, Zimmerman J; US Insulin Glargine (HOE 901) Type 1 Diabetes
Investigator Group. Basal insulin glargine (HOE 901) versus NPH insulin in patients with
type 1 diabetes on multiple daily insulin regimens. Diabetes Care 2000;23:1137–42.
[68] Pieber TR, Eugene-Jolchine I, Derobert E. Efficacy and safety of HOE 901 versus NPH
insulin in patients with type 1 diabetes. Diabetes Care 2000;23(2):151–62.
[69] Garvey WT, Olefsky JM, Griffin J, Hamman RF, Kolterman OG. The effect of insulin
treatment on insulin secretion and insulin action in type 2 diabetes mellitus. Diabetes 1985;
34:222.
[70] Jennings AM, Lewis KS, Murdoch S, Talbot JF, Bradley C, Ward JD. Randomized trial
comparing continuous subcutaneous insulin infusion and conventional insulin therapy in
type II diabetic patients poorly controlled with sulfonylureas. Diabetes Care 1991;14:738.
[71] Pouwels JJ, Tack CJ, Hermust AR, Lutterman JA. Treatment with intravenous insulin
followed by continuous subcutaneous insulin infusion improves glycaemic control in
severely resistant type 2 diabetic patients. Diabetes Med 2003;20:76–9.
[72] Raskin P, Bode BW, Marks JB, Hirsch IB, Weinstein RL, McGill JB, Peterson GE,
Mudaliar SR, Reinhardt RR. Continuous subcutaneous insulin infusion and multiple daily
injection therapy are equally effective in type 2 diabetes: a randomized, parallel-group, 24-
week study. Diabetes Care 2003;26(9):2598–603.
[73] Cefalu WT, Skyler J, Kourides IA, Landschulz WH, Balagtas CC, Cheng S, et al. Inhaled
human insulin treatment in patients with type 2 diabetes mellitus. Ann Intern Med 2001;
134:203–7.
[74] Nathan D. Inhaled insulin for type 2 diabetes [editorial]. Ann Intern Med 2001;134(3):
242–4.
[75] Kim D, Mudaliar S, Chinnapongse S, Chu N, Boies SM, Davis T, et al. Dose-response
relationships of inhaled insulin delivered via the Aerodose insulin inhaler and sub-
cutaneously injected insulin in patients with type 2 diabetes. Diabetes Care 2003;26(10):
2842–7.
[76] Kipnes M, Dandona P, Tripathy D, Still JG, Kosutic G. Control of postprandial plasma
glucose by an oral insulin product (HIM2) in patients with type 2 diabetes. Diabetes Care
2003;26(2):421–6.
[77] Edelman SV, Kruger DF. A new look at glucose control: the case for therapy with insulin
and an amylin analog. Practical Diabetology 1998;16(4):9–14.
[78] Buse JB, Weyer C, Maggs DG. Amylin replacement with pramlintide in type 1 and type
2 diabetes: a physiological approach to overcome barriers with insulin therapy. Clin
Diabetes 2002;20:137–44.
[79] Fineman M, Gottlieb A, Skare S, Kolterman O. Pramlintide as an adjunct to insulin
therapy improved glycemic and weight control in people with type 2 diabetes during
treatment for 52 weeks. Diabetes 2000;49(Suppl 1):A106.
[80] Hollander PA, Levy P, Fineman MS, Maggs DG, Shen LZ, Strobel SA, et al. Pramlintide
as an adjunct to insulin therapy improves long-term glycemic and weight control in pa-
tients with type 2 diabetes: a 1-year randomized controlled trial. Diabetes Care 2003;26(3):
784–90.
Med Clin N Am 88 (2004) 897–909

Dyslipidemia in type 2 diabetes


Ronald M. Krauss, MDa,b,c,*,
Patty W. Siri, PhD, MSa
a
Children’s Hospital Oakland Research Institute, 5700 Martin Luther King Jr Way,
Oakland, CA 94609, USA
b
Department of Genome Sciences, Lawrence Berkeley
National Laboratory, One Cyclotron Road, Berkeley, CA 94720, USA
c
Department of Nutritional Sciences, University of
California at Berkeley, M/S 3104, Berkeley, CA 94720, USA

The cluster of lipid abnormalities associated with type 2 diabetes is


defined by increases in triglyceride (TG) and small, dense low-density
lipoprotein (LDL) concentrations and decreases in high-density lipoprotein
(HDL) cholesterol. Plasma LDL cholesterol levels are generally normal
because the increase in the number of small, dense LDL particles is
accompanied by a reduction in large LDL particles. Each of the features
of diabetic dyslipidemia has been associated with increased risk of
cardiovascular disease, the leading cause of death in type 2 diabetics.
Although attempts at decreasing cardiovascular disease morbidity and
mortality have usually been focused on lowering LDL cholesterol, it is
important to consider other features of the abnormal lipid profile typical of
type 2 diabetes that may contribute to the disease process, particularly TG-
rich remnants and small, dense LDL. These factors are not detected
by standard lipid testing, but part of the benefits observed in coronary
artery disease-prevention trials may be because of improvements of these
parameters.
This article reviews the processes believed to occur in the development of
diabetic dyslipidemia and discusses some of the underlying mechanisms for
the syndrome. The relationship between facets of the typical lipid profile
exhibited by patients with type 2 diabetes and risk of cardiovascular disease
are described, and options for the management and treatment of diabetic
dyslipidemia are discussed.

* Corresponding author. Children’s Hospital, Oakland Research Institute, 5700 Martin


Luther King Jr Way, Oakland, CA 94609.
E-mail address: rmkrauss@lbl.gov (R.M. Krauss).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.004
898 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

Pathophysiology of diabetic dyslipidemia


Insulin resistance likely underlies the lipid changes associated with type 2
diabetes, and increasing resistance to insulin, even in persons considered to
have ‘‘normal’’ insulin sensitivity, has been associated with higher concen-
trations of cholesterol and TG and lower concentrations of HDL cholesterol
[1]. Importantly, insulin resistance and type 2 diabetes often occur along
with other metabolic abnormalities such as obesity, hypertension, and
hypercoagulability. This grouping of abnormalities has been referred to as
the metabolic syndrome or ‘‘syndrome X’’ and has been associated with an
increased risk for atherosclerosis [2,3]. Other abnormalities may be asso-
ciated with poorly controlled diabetes, for example, glycosylation of lipo-
proteins and other proteins involved in lipoprotein metabolism. A discussion
of these abnormalities, however, is beyond the scope of this article.

Association of insulin resistance and hepatic very low-density lipoprotein


secretion
Resistance to insulin may contribute to the atherogenic dyslipidemia of
diabetes by increasing the hepatic secretion of very low-density lipoprotein
(VLDL). Metabolic tracer studies have documented overproduction of
VLDL TG in insulin-resistant patients with hypertriglyceridemia [4,5].
Additionally, several recent studies demonstrate increased secretion of
apolipoprotein (apo) B in type 2 diabetes [6,7]. The increased secretion of
apoB-containing lipoprotein particles may be the result of increased free
fatty acid (FFA) flux to the liver [8]. Also, insulin-resistant persons have
been shown to lack sensitivity to the suppressive effects of insulin on apoB
secretion [9,10]. This resistance to insulin may be at the level of the
regulation of apoB degradation or inhibition of microsomal triglyceride
transfer protein activity, a protein identified as a key component of the
VLDL assembly process [11].
Because of increased endogenous secretion of apoB-containing lipopro-
tein particles, the increased plasma levels of TG can drive a metabolic
process that results in reduced HDL cholesterol levels and LDL particles
that are smaller and more dense. In a substrate-driven reaction, cholesterol
ester transfer protein (CETP) exchanges VLDL TG for HDL cholesterol.
TG-rich HDL particles are hydrolyzed by hepatic lipase (HL) and, as
a result, are rapidly catabolized and cleared from plasma [12]. HDL
particles are heterogeneous and are classified by particle sizes that range
from small, dense HDL3 to larger HDL2 particles [13]. Typically, the
reduced plasma levels of HDL in patients with type 2 diabetes manifest as
reductions in the HDL2 subspecies with relative or absolute increases in
HDL3.
Increased concentrations of VLDL in plasma also result in the increased
production of small, dense LDL particles. As many as seven distinct LDL
subspecies, which differ in metabolic behavior and pathologic roles, have
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 899

been identified (Fig. 1) [14]. Plasma VLDL levels correlate positively with
increased density and decreased size of LDL [15,16], and increased
concentrations of small, dense LDL in turn have been shown to be
associated with reduced plasma HDL levels [17]. Importantly, the residence
time of small, dense LDL in plasma may be prolonged, given their relatively
reduced affinity for the LDL receptor [14].

Effects of insulin resistance on lipid and lipoprotein clearance


Impaired clearance of lipid and lipoprotein particles represents another
important mechanism by which insulin resistance can lead to abnormal lipid
profiles. Insulin resistance has been associated with impaired lipoprotein
lipase (LPL) and increased HL activity [18–20]. LPL is synthesized in muscle
and adipose tissue and interacts with TG-rich lipoproteins in capillary
endothelial cell beds where it hydrolyzes TG into FFA. The resultant

LPL
IDL 2 LDL I
Liver

TG
TG
CETP
LPL LPL HL Smaller
1 2 VLDL 2 IDL 1 LDL II LDL

TG TG
CETP
LPL LPL/HL HL
VLDL 1 Remnants LDL III Smaller
LDL

TG
CETP
larger LPL
Remnants LPL/HL HL
VLDL 1 LDL IV Smaller
LDL

Fig. 1. Hypothetical metabolic scheme incorporating proposed pathways for the production of
major LDL subclasses I, II, III, and IV. The properties of triglyceride-rich lipoproteins secreted
by the liver are determined by the operation of multiple pathways (1 and 2) and by hepatic TG
availability. In pathway 1, triglyceride-rich VLDL-1 and triglyceride-poor IDL-2 particles are
coordinately produced. VLDL-1 production results from a discrete quantity of lipid on
a precursor particle. Lipolysis of VLDL-1 yields remnants, which in turn yield LDL-III by
hepatic lipase. Further remodeling of these particles may occur by CETP-mediated triglyceride
enrichment and hepatic lipase-mediated lipolysis. Pathway 2, which results in the production of
VLDL-2, is distinct from pathway 1 and gives rise to IDL-1 and LDL-II by lipolysis. Further
processing by CETP-mediated transfer of triglycerides into LDL-II and lipolysis by hepatic
lipase yields smaller and denser LDL products. CE, cholesteryl esters; LPL, lipoprotein lipase.
(From Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL
heterogeneity. J Lipid Res 2002;43:1363–79; with permission.)
900 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

lipoprotein particles are reduced in core volume and surface and are either
cleared through remnant removal pathways or moved along the delipidation
pathway where they are converted into less buoyant LDL particles. HL is
responsible for the hydrolysis of phospholipids in LDL and HDL particles;
its increased activity in the setting of insulin resistance has been associated
with smaller and denser LDL particles and a decrease in HDL2 particles
because the latter are more rapidly cleared from plasma.

Lipid effects on insulin sensitivity


Insulin resistance and type 2 diabetes mellitus are often characterized by
increased plasma FFA concentrations because of increased adipose tissue
efflux or impaired insulin-mediated skeletal muscle uptake [21,22]. The fact
that FFA levels are elevated in individuals with impaired glucose tolerance
suggests that insulin resistance associated with increased FFA levels occurs
before the onset of hyperglycemia [23].
Elevations of plasma FFA concentrations may interfere with glucose
metabolism by impairing glucose uptake and by use in muscle [24]. In
addition, infusion of lipid emulsions and heparin to increase ambient FFA
concentrations has been shown to increase gluconeogenesis in patients with
insulin resistance and type 2 diabetes [25]; however, glycogenolysis was
simultaneously dampened in these patients so that net glucose production
remained unchanged. Increased FFA concentrations in plasma may also
impair hepatic insulin extraction [26], thereby contributing to peripheral
hyperinsulinemia in insulin resistance.
At the level of the pancreatic b-cell, FFA acutely increases glucose-
stimulated insulin secretion, whereas chronic exposure has been associated
with relatively impaired insulin secretion [27]. Finally, in the presence of
insulin resistance, FFA in the form of TG is deposited in muscle and the
liver, heart, and pancreas where it may impair organ function [28]. Notably,
agents that lower elevated FFA, such as the thiazolidinediones (TZDs), have
been shown to improve insulin sensitivity in muscle, liver, and adipose tissue
[29,30].

Dyslipidemia and cardiovascular disease risk


The association of plasma TG and risk of coronary artery disease is based
primarily on the results of epidemiologic studies. A meta-analysis of 17
population-based prospective studies showed that an increase in plasma TG
of 1 mmol/L, or 89 mg/dl, was associated with increased coronary disease
risk in both men and women (14% and 37%, respectively), after adjustment
for HDL cholesterol and other risk factors [31]. Direct atherogenic effects of
TG-rich particles, particularly IDL and remnant particles, may account for
the contribution of plasma TG levels to coronary heart disease risk [32,33].
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 901

The association between reduced HDL cholesterol levels and increased


risk of heart disease is well established [34]. HDL functions in cellular
cholesterol efflux and has direct antioxidative as well as anti-inflammatory
properties. Reduced HDL cholesterol levels are often accompanied by
elevations in plasma TG levels [35], a process mediated by CETP [36]. In
combination, high TG and low HDL cholesterol levels have been strongly
associated with an increased risk of coronary heart disease in a number of
prospective clinical studies [37–39]. The results of the Quebec Cardiovascu-
lar Study led to the speculation that the cardioprotective effects of high
HDL cholesterol levels is more attributable to HDL2 than to HDL3 particles
[35]. The nature of this association, however, is probably most dependent on
the characteristics of the HDL particle distribution specific to the popula-
tion being studied. The Veterans Affairs HDL Intervention Trial (VAHIT)
[40] conducted in men with HDL cholesterol levels below 40 mg/dl show
decreased coronary events associated with increased HDL3 particles after
treatment with gemfibrozil; thus, both HDL2 and HDL3 probably have
cardioprotective effects. Notably, as discussed earlier, people with type 2
diabetes tend to have small HDL particles, which is likely a function of the
increased transfer of HDL cholesterol for VLDL TG mediated by CETP.
Small HDL particle size has also been found to be characteristic of the
dyslipidemic profile typical of individuals with visceral obesity and insulin
resistance [13].
Small, dense LDL particles may confer increased atherogenicity by virtue
of their intrinsic physicochemical and metabolic properties, including
reduced LDL receptor affinity [41,42], greater propensity for transport into
the subendothelial space [43], increased binding to arterial wall proteogly-
cans, and susceptibility to oxidative modifications [44,45]. Although these
are in vitro findings, they support the concept that small, dense LDL
contributes to arterial damage in patients with the characteristic dyslipide-
mia associated with diabetes.

Management of diabetic dyslipidemia


Current recommendations for the management of dyslipidemia in
patients with type 2 diabetes include lifestyle interventions targeted at
improving behaviors known to have adverse effects on insulin sensitivity and
lipid parameters [46]. Diet, exercise, weight loss, and smoking cessation
particularly can significantly improve the dyslipidemic profile typically
associated with type 2 diabetes and insulin resistance [47,48]. Of interest is
the accumulating evidence that supports the concept that a diet low in
carbohydrate can promote weight loss and improve lipid and lipoprotein
profiles [49–51]. In those persons in whom lifestyle interventions have only
limited effectiveness, pharmacologic therapies may be necessary to reach
treatment goals. Several classes of medications are described.
902 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

Lipid-lowering agents
Statins (HMG-CoA reductase inhibitors)
Statins inhibit HMG-CoA reductase, the rate-limiting enzyme in choles-
terol biosynthesis, thereby increasing LDL receptor activity and stimulating
the clearance of lipoprotein particles from plasma. Statins also inhibit the
secretion of lipoprotein particles from the liver. In addition to lowering
LDL, statins have been shown to lower plasma TG levels and to raise HDL
cholesterol, albeit to varying degrees. All subclasses of LDL and IDL are
reduced to an equivalent extent, although there have been reports of greater
lowering of small LDL in conjunction with TG reduction [52]. Most studies,
however, have not demonstrated a reversal of the small, dense LDL
phenotype with statin treatment.

Fibrates (peroxisome proliferator-activated receptor-a agonists)


Peroxisome proliferator-activated receptor (PPAR)-a agonists reduce
levels of TG-rich lipoproteins primarily by up-regulating the transcription of
genes involved in TG clearance and FFA oxidation. Among the changes
that are induced by treatment with PPAR-a agonists are increased tran-
scription of LPL and apoCII, an activator of LPL, and decreased
transcription of apoCIII, an inhibitor of LPL-mediated lipolysis. Fibrates
have also been shown to increase HDL cholesterol levels, probably by
increasing the production of HDL apoproteins and reducing the transfer of
HDL cholesterol for VLDL TG [53]. Effects of fibrates on LDL cholesterol
have been inconsistent; in general, any observed reductions were small. Of
interest is a strong body of evidence that demonstrates that fibrates can
effectively reverse the atherogenic dyslipidemia phenotype in patients with
type 2 diabetes [54–57].
A combination of statin and fenofibrate has been shown to be
particularly effective in treating patients with type 2 diabetes and combined
hyperlipidemia (ie, increased LDL cholesterol, low HDL cholesterol, and
elevated TG concentrations) [58]. Similar improvements in lipid profiles
have also been demonstrated in patients with combined hyperlipidemia and
the metabolic syndrome; in this group, concentrations of small, dense LDL
were also significantly reduced. Treatment with a statin-fenofibrate combi-
nation therapy has raised some concerns with regard to its toxicity,
particularly with gemfibrozil [59]. Caution should therefore be exercised
especially in patients with impaired renal function.

Niacin
Niacin, or nicotinic acid, has been shown to significantly reduce TG
concentrations, increase HDL cholesterol levels, and increase LDL particle
size and buoyancy. Niacin reduces FFA release from adipose tissue and
suppresses the hepatic production of VLDL. These changes effectively
decrease plasma TG levels and reduce the number of small, dense LDL
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 903

particles. The effect of niacin on HDL cholesterol levels has been shown to
be mediated by reduced activity of the receptor responsible for the intra-
hepatic degradation of HDL [60]. Reducing the uptake of HDL elongates
the half-life of these particles in plasma.
Concern regarding the adverse effects of niacin on glycemic control has
discouraged its use in type 2 diabetics. However, a recent clinical trial [61]
indicates that the worsening of glycemic control with niacin treatment may
be limited to high doses. Extended-release niacin administered at 1000 and
1500 mg/d was effective in significantly increasing HDL cholesterol levels
and decreasing TG levels, relative to placebo treatment. Only the group of
patients receiving 1500 mg/d of niacin showed adverse changes in glycemic
control; concentrations of glycosylated hemoglobin increased from 7.2% at
baseline to 7.5% after 16 weeks of niacin therapy (P = 0.048, relative to
controls). Although four patients discontinued therapy because of inadequate
glycemic control, most patients were able to tolerate niacin therapy.

Antidiabetic agents
Thiazolidinediones
TZDs (PPAR-c agonist) have been shown to lower glucose concen-
trations in type 2 diabetes by improving peripheral insulin sensitivity and
thereby promoting glucose uptake. TZDs have also been shown to reduce
hepatic glucose production [62] and decrease plasma FFA levels [63]. As an
agonist of PPAR-c, TZD stimulates genes involved in adipocyte differen-
tiation. Although not indicated for the treatment of dyslipidemia, these
agents have been shown to improve lipid abnormalities associated with type
2 diabetes. Use of rosiglitazone and pioglitazone has been shown to increase
HDL cholesterol levels in patients with diabetes. Although TZDs tend to
increase total and LDL cholesterol, they induce favorable changes in LDL
particle size and susceptibility to oxidation [64]. Plasma TG levels are
generally decreased with pioglitazone therapy [65–67] and have also been
shown to be reduced with rosiglitazone in hypertriglyceridemic patients [68].
In a randomized controlled clinical trial, treatment with rosiglitazone for 8
weeks led to significant increases in HDL cholesterol (6%) and HDL2 levels
(13%) and a shift of 71% of patients initially presenting with small, dense LDL
to a phenotype of more buoyant and, thus, less atherogenic LDL particles.
Combination therapy with atorvastatin led to an even greater increase in HDL
(5%) as well as significant decreases in LDL cholesterol (ÿ30%) and TG
(ÿ27%) [69]. Pioglitazone has also been shown to be effective in ameliorating
the syndrome of dyslipidemia in several studies conducted in patients with
type 2 diabetes [70,71]. A recent study [72] in nondiabetic patients with arterial
hypertension shows a significant reduction in small, dense LDL after 16 weeks
of pioglitazone treatment. At baseline, 66% of the study group had elevations
in small, dense LDL; and treatment with pioglitazone resulted in a 22%
reduction in the levels of small, dense LDL particles.
904 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

Metformin
Metformin is a biguanide that has been proven to be an effective
hypoglycemic agent [73]. The drug lowers fasting glucose and insulin levels
by improving insulin sensitivity, inhibiting hepatic gluconeogenesis, and
stimulating peripheral glucose uptake [74]. Importantly, this agent has been
shown to decrease cardiovascular disease risk in type 2 diabetics, in-
dependently of its effects on glycemic control [75]. Studies have shown that
metformin improves lipoprotein profiles in patients with type 2 diabetes and
insulin resistance [76]. It decreases plasma FFA as well as TG[77], decreases
total and LDL cholesterol [78,79], and maintains [78] or increases [77,79]
HDL cholesterol levels. Improvements in the components of type 2 diabetes-
associated dyslipidemia likely contribute to the cardioprotective effect of
metformin.

Summary
Abnormal lipid metabolism often presents in patients with type 2
diabetes. Resistance to insulin likely underlies the changes that occur in
lipid parameters. Increased FFA availability can drive the secretion of TG-
rich lipoproteins from the liver, and this increased TG substrate can lead to
reductions in HDL cholesterol as well as the conversion of LDL particles
into particles that are smaller and denser. The combination of elevated TG,
reduced HDL cholesterol, and smaller and denser LDL particles typifies the
dyslipidemia associated with type 2 diabetes and insulin resistance. Each of
these lipid changes has been independently associated with increased
cardiovascular disease risk. A variety of treatment options exist, and
changes in lifestyle and behaviors known to adversely alter lipid metabolism
should be key components of any program designed to manage atherogenic
dyslipidemia. Effective pharmacologic therapies include lipid-lowering
agents, such as statins, fibrates, and niacin, as well as antidiabetic agents,
such as the thiazolidinediones and metformin. Treatment with these agents
alone or in combination has been shown to correct the lipid abnormalities
associated with type 2 diabetes and reduce the risk of cardiovascular disease.

References
[1] Garvey WT, Kwon S, Zheng D, Shaughnessy S, Wallace P, Hutto A. Effects of insulin
resistance and type 2 diabetes on lipoprotein subclass particle size and concentration
determined by nuclear magnetic resonance. Diabetes 2003;52(2):453–62.
[2] DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syndrome responsible for
NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease.
Diabetes Care 1991;14(3):173–94.
[3] Reaven GM. Pathophysiology of insulin resistance in human disease. Physiol Rev 1995;
75(3):473–86.
[4] Nikkila EA, KM. Plasma triglyceride transport kinetics in diabetes mellitus. Metabolism
1973;22:1–22.
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 905

[5] Taskinen MR, Beltz WF, Harper I, Fields RM, Schonfeld G, Grundy SM, et al. Effects of
NIDDM on very-low-density lipoprotein triglyceride and apolipoprotein B metabolism.
Studies before and after sulfonylurea therapy. Diabetes 1986;35(11):1268–77.
[6] Duvillard L, Pont F, Florentin E, Galland-Jos C, Gambert P, Verges B. Metabolic
abnormalities of apolipoprotein B-containing lipoproteins in non-insulin-dependent
diabetes: a stable isotope kinetic study. Eur J Clin Invest 2000;30(8):685–94.
[7] Cummings MH, Watts GF, Umpleby AM, Hennessy TR, Naoumova R, Slavin BM, et al.
Increased hepatic secretion of very-low-density lipoprotein apolipoprotein B-100 in
NIDDM. Diabetologia 1995;38(8):959–67.
[8] Laws A, Hoen HM, Selby JV, Saad MF, Haffner SM, Howard BV. Differences in insulin
suppression of free fatty acid levels by gender and glucose tolerance status. Relation to
plasma triglyceride and apolipoprotein B concentrations. Insulin Resistance Atheroscle-
rosis Study (IRAS) Investigators. Arterioscler Thromb Vasc Biol 1997;17(1):64–71.
[9] Lewis GF, Uffelman KD, Szeto LW, Steiner G. Effects of acute hyperinsulinemia on
VLDL triglyceride and VLDL apoB production in normal weight and obese individuals.
Diabetes 1993;42(6):833–42.
[10] Malmstrom R, Packard CJ, Caslake M, Bedford D, Stewart P, Yki-Jarvinen H, et al.
Defective regulation of triglyceride metabolism by insulin in the liver in NIDDM.
Diabetologia 1997;40(4):454–62.
[11] Fisher EA, Ginsberg HN. Complexity in the secretory pathway: the assembly and secretion
of apolipoprotein B-containing lipoproteins. J Biol Chem 2002;277(20):17377–80.
[12] Hopkins GJ, Barter PJ. Role of triglyceride-rich lipoproteins and hepatic lipase in
determining the particle size and composition of high density lipoproteins. J Lipid Res
1986;27(12):1265–77.
[13] Pascot A, Lemieux I, Prud’homme D, Tremblay A, Nadeau A, Couillard C, et al. Reduced
HDL particle size as an additional feature of the atherogenic dyslipidemia of abdominal
obesity. J Lipid Res 2001;42(12):2007–14.
[14] Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL heterogeneity.
J Lipid Res 2002;43(9):1363–79.
[15] McNamara JR, Campos H, Ordovas JM, Peterson J, Wilson PW, Schaefer EJ. Effect
of gender, age, and lipid status on low density lipoprotein subfraction distribution.
Results from the Framingham Offspring Study. Arterioscler 1987;7(5):483–90.
[16] McNamara JR, Jenner JL, Li Z, Wilson PW, Schaefer EJ. Change in LDL particle size is
associated with change in plasma triglyceride concentration. Arterioscler Thromb 1992;
12(11):1284–90.
[17] Krauss RM, Williams PT, Lindgren FT, Wood PD. Coordinate changes in levels of human
serum low and high density lipoprotein subclasses in healthy men. Arterioscler 1988;8(2):
155–62.
[18] Tan CE, Foster L, Caslake MJ, Bedford D, Watson TD, McConnell M, et al. Relations
between plasma lipids and postheparin plasma lipases and VLDL and LDL subfraction
patterns in normolipemic men and women. Arterioscler Thromb Vasc Biol 1995;15(11):
1839–48.
[19] Watson TD, Caslake MJ, Freeman DJ, Griffin BA, Hinnie J, Packard CJ, et al.
Determinants of LDL subfraction distribution and concentrations in young normolipi-
demic subjects. Arterioscler Thromb 1994;14(6):902–10.
[20] Zambon A, Austin MA, Brown BG, Hokanson JE, Brunzell JD. Effect of hepatic lipase on
LDL in normal men and those with coronary artery disease. Arterioscler Thromb Vasc
Biol 1993;13(2):147–53.
[21] Boden G. Role of fatty acids in the pathogenesis of insulin resistance and NIDDM.
Diabetes 1997;46(1):3–10.
[22] Kelley DE, Simoneau JA. Impaired free fatty acid utilization by skeletal muscle in non-
insulin-dependent diabetes mellitus. J Clin Invest 1994;94(6):2349–56.
906 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

[23] Bluher M, Kratzsch J, Paschke R. Plasma levels of tumor necrosis factor-alpha,


angiotensin II, growth hormone, and IGF-I are not elevated in insulin-resistant obese
individuals with impaired glucose tolerance. Diabetes Care 2001;24(2):328–34.
[24] Dresner A, Laurent D, Marcucci M, Griffin ME, Dufour S, Cline GW, et al. Effects of free
fatty acids on glucose transport and IRS-1-associated phosphatidylinositol 3-kinase
activity. J Clin Invest 1999;103(2):253–9.
[25] Lewis GF, Carpentier A, Adeli K, Giacca A. Disordered fat storage and mobilization in
the pathogenesis of insulin resistance and type 2 diabetes. Endocr Rev 2002;23(2):201–29.
[26] Wiesenthal SR, Sandhu H, McCall RH, Tchipashvili V, Yoshii H, Polonsky K, et al. Free
fatty acids impair hepatic insulin extraction in vivo. Diabetes 1999;48(4):766–74.
[27] Carpentier A, Mittelman SD, Lamarche B, Bergman RN, Giacca A, Lewis GF. Acute
enhancement of insulin secretion by FFA in humans is lost with prolonged FFA elevation.
Am J Physiol 1999;276(Pt 1):E1055–66.
[28] Unger RH, Orci L. Diseases of liporegulation: new perspective on obesity and related
disorders. FASEB J 2001;15(2):312–21.
[29] Mayerson AB, Hundal RS, Dufour S, Lebon V, Befroy D, Cline GW, et al. The effects of
rosiglitazone on insulin sensitivity, lipolysis, and hepatic and skeletal muscle triglyceride
content in patients with type 2 diabetes. Diabetes 2002;51(3):797–802.
[30] Miyazaki Y, Mahankali A, Matsuda M, Mahankali S, Hardies J, Cusi K, et al. Effect of
pioglitazone on abdominal fat distribution and insulin sensitivity in type 2 diabetic
patients. J Clin Endocrinol Metab 2002;87(6):2784–91.
[31] Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular
disease independent of high-density lipoprotein cholesterol level: a meta-analysis of
population-based prospective studies. J Cardiovasc Risk 1996;3(2):213–9.
[32] Krauss R. Triglycerides and atherogenic lipoproteins: rationale for lipid management.
J Med 1998;105(Suppl 1A):S58–62.
[33] Krauss RM. Atherogenicity of triglyceride-rich lipoproteins. Am J Cardiol 1998;81(Suppl
4A):B13–7.
[34] Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD, Castelli WP, Knoke JD, et al. High-
density lipoprotein cholesterol and cardiovascular disease. Four prospective American
studies. Circulation 1989;79(1):8–15.
[35] Lamarche B, Despres JP, Moorjani S, Cantin B, Dagenais GR, Lupien PJ. Triglycerides
HDL-cholesterol as risk factors for ischemic heart disease Results from the Quebec
cardiovascular study. Atherosclerosis 1996;119(2):235–45.
[36] Tall A. Plasma lipid transfer proteins. Annu Rev Biochem 1995;64:235–57.
[37] Assmann G, Schulte H. Relation of high-density lipoprotein cholesterol and triglycerides
to incidence of atherosclerotic coronary artery disease (the PROCAM experience).
Prospective Cardiovascular Munster study. Am J Cardiol 1992;70(7):733–7.
[38] Jeppesen J, Hein HO, Suadicani P, Gyntelberg F. Relation of high TG-low HDL cho-
lesterol and LDL cholesterol to the incidence of ischemic heart disease. An 8-year follow-
up in the Copenhagen Male Study. Arterioscler Thromb Vasc Biol 1997;17(6):1114–20.
[39] Manninen V, Tenkanen L, Koskinen P, Huttunen JK, Manttari M, Heinonen OP, et al.
Joint effects of serum triglyceride and LDL cholesterol and HDL cholesterol con-
centrations on coronary heart disease risk in the Helsinki Heart Study. Implications for
treatment. Circulation 1992;85(1):37–45.
[40] Rubins HB, Robins SJ, Collins D, Fye CL, Anderson JW, Elam MB, et al. Gemfibrozil for
the secondary prevention of coronary heart disease in men with low levels of high-density
lipoprotein cholesterol. Veterans Affairs High-Density Lipoprotein Cholesterol Interven-
tion Trial Study Group. N Engl J Med 1999;341(6):410–8.
[41] Campos H, Arnold KS, Balestra ME, Innerarity TL, Krauss RM. Differences in receptor
binding of LDL subfractions. Arterioscler Thromb Vasc Biol 1996;16(6):794–801.
[42] Galeano NF, Milne R, Marcel YL, Walsh MT, Levy E, Ngu’yen TD, et al. Apoprotein B
structure and receptor recognition of triglyceride-rich low density lipoprotein (LDL) is
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 907

modified in small LDL but not in triglyceride-rich LDL of normal size. J Biol Chem 1994;
269(1):511–9.
[43] Bjornheden T, Babyi A, Bondjers G, Wiklund O. Accumulation of lipoprotein fractions
and subfractions in the arterial wall, determined in an in vitro perfusion system.
Atherosclerosis 1996;123(1–2):43–56.
[44] Chait A, Brazg RL, Tribble DL, Krauss RM. Susceptibility of small, dense, low-density
lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein
phenotype, pattern B. Am J Med 1993;94(4):350–6.
[45] de Graaf J, Hak-Lemmers HL, Hectors MP, Demacker PN, Hendriks JC, Stalenhoef AF.
Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein
subfraction in healthy subjects. Arterioscler Thromb 1991;11(2):298–306.
[46] American Diabetes Association. Management of dyslipidemia in adults with diabetes.
Diabetes Care 2003;26(Suppl 1):S83–6.
[47] Kraus WE, Houmard JA, Duscha BD, Knetzger KJ, Wharton MB, McCartney JS, et al.
Effects of the amount and intensity of exercise on plasma lipoproteins. N Engl J Med 2002;
347(19):1483–92.
[48] Williams PT, Krauss RM, Vranizan KM, Wood PD. Changes in lipoprotein subfractions
during diet-induced and exercise-induced weight loss in moderately overweight men.
Circulation 1990;81(4):1293–304.
[49] Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, Mohammed BS, et al. A
randomized trial of a low-carbohydrate diet for obesity. N Engl J Med 2003;348(21):
2082–90.
[50] Samaha FF, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, et al. A low-
carbohydrate as compared with a low-fat diet in severe obesity. N Engl J Med 2003;
348(21):2074–81.
[51] Westman EC, Yancy WS, Edman JS, Tomlin KF, Perkins CE. Effect of 6-month
adherence to a very low carbohydrate diet program. Am J Med 2002;113(1):30–6.
[52] McKenney JM, McCormick LS, Schaefer EJ, Black DM, Watkins ML. Effect of niacin
and atorvastatin on lipoprotein subclasses in patients with atherogenic dyslipidemia. Am J
Cardiol 2001;88(3):270–4.
[53] Guerin M, Bruckert E, Dolphin PJ, Turpin G, Chapman MJ. Fenofibrate reduces
plasma cholesteryl ester transfer from HDL to VLDL and normalizes the atherogenic,
dense LDL profile in combined hyperlipidemia. Arterioscler Thromb Vasc Biol 1996;
16(6):763–72.
[54] Frost RJ, Otto C, Geiss HC, Schwandt P, Parhofer KG. Effects of atorvastatin versus
fenofibrate on lipoprotein profiles, low-density lipoprotein subfraction distribution, and
hemorheologic parameters in type 2 diabetes mellitus with mixed hyperlipoproteinemia.
Am J Cardiol 2001;87(1):44–8.
[55] Guerin M, Le Goff W, Frisdal E, Schneider S, Milosavljevic D, Bruckert E, et al. Action of
ciprofibrate in type IIb hyperlipoproteinemia: modulation of the atherogenic lipoprotein
phenotype and stimulation of high-density lipoprotein-mediated cellular cholesterol efflux.
J Clin Endocrinol Metab 2003;88(8):3738–46.
[56] Feher MD, Caslake M, Foxton J, Cox A, Packard CJ. Atherogenic lipoprotein phenotype
in type 2 diabetes: reversal with micronised fenofibrate. Diabetes Metab Res Rev 1999;
15(6):395–9.
[57] Vakkilainen J, Steiner G, Ansquer JC, Perttunen-Nio H, Taskinen MR. Fenofibrate lowers
plasma triglycerides and increases LDL particle diameter in subjects with type 2 diabetes.
Diabetes Care 2002;25(3):627–8.
[58] Athyros VG, Papageorgiou AA, Athyrou VV, Demitriadis DS, Kontopoulos AG.
Atorvastatin and micronized fenofibrate alone and in combination in type 2 diabetes with
combined hyperlipidemia. Diabetes Care 2002;25(7):1198–202.
[59] Davidson MH. Combination therapy for dyslipidemia: safety and regulatory consid-
erations. Am J Cardiol 2002;90(Suppl 10B):K50–60.
908 R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909

[60] Kamanna VS, Kashyap ML. Mechanism of action of niacin on lipoprotein metabolism.
Curr Atheroscler Rep 2000;2(1):36–46.
[61] Grundy SM, Vega GL, McGovern ME, Tulloch BR, Kendall DM, Fitz-Patrick D, et al.
Efficacy, safety, and tolerability of once-daily niacin for the treatment of dyslipidemia
associated with type 2 diabetes: results of the assessment of diabetes control and evaluation
of the efficacy of niaspan trial. Arch Intern Med 2002;162(14):1568–76.
[62] Inzucchi SE, Maggs DG, Spollett GR, Page SL, Rife FS, Walton V, et al. Efficacy and
metabolic effects of metformin and troglitazone in type II diabetes mellitus. N Engl J Med
1998;338(13):867–72.
[63] Olefsky JM. Treatment of insulin resistance with peroxisome proliferator-activated
receptor gamma agonists. J Clin Invest 2000;106(4):467–72.
[64] Fonseca VA. Management of diabetes mellitus and insulin resistance in patients with
cardiovascular disease. Am J Cardiol 2003;92(Suppl 4A):J50–60.
[65] Kipnes MS, Krosnick A, Rendell MS, Egan JW, Mathisen AL, Schneider RL. Pioglitazone
hydrochloride in combination with sulfonylurea therapy improves glycemic control in
patients with type 2 diabetes mellitus: a randomized, placebo-controlled study. Am J Med
2001;111(1):10–7.
[66] Aronoff S, Rosenblatt S, Braithwaite S, Egan JW, Mathisen AL, Schneider RL.
Pioglitazone hydrochloride monotherapy improves glycemic control in the treatment of
patients with type 2 diabetes: a 6-month randomized placebo-controlled dose-response
study. The Pioglitazone 001 Study Group. Diabetes Care 2000;23(11):1605–11.
[67] Einhorn D, Rendell M, Rosenzweig J, Egan JW, Mathisen AL, Schneider RL. Pioglitazone
hydrochloride in combination with metformin in the treatment of type 2 diabetes mellitus:
a randomized, placebo-controlled study. The Pioglitazone 027 Study Group. Clin Ther
2000;22(12):1395–409.
[68] Gomez-Perez FJ, Fanghanel-Salmon G, Antonio Barbosa J, Montes-Villarreal J, Berry
RA, Warsi G, et al. Efficacy and safety of rosiglitazone plus metformin in Mexicans with
type 2 diabetes. Diabetes Metab Res Rev 2002;18(2):127–34.
[69] Freed MI, Ratner R, Marcovina SM, Kreider MM, Biswas N, Cohen BR, et al. Effects of
rosiglitazone alone and in combination with atorvastatin on the metabolic abnormalities in
type 2 diabetes mellitus. Am J Cardiol 2002;90(9):947–52.
[70] Pavo I, Jermendy G, Varkonyi TT, Kerenyi Z, Gyimesi A, Shoustov S, et al. Effect of
pioglitazone compared with metformin on glycemic control and indicators of insulin
sensitivity in recently diagnosed patients with type 2 diabetes. J Clin Endocrinol Metab
2003;88(4):1637–45.
[71] Rosenblatt S, Miskin B, Glazer NB, Prince MJ, Robertson KE. The impact of pioglitazone
on glycemic control and atherogenic dyslipidemia in patients with type 2 diabetes mellitus.
Coron Artery Dis 2001;12(5):413–23.
[72] Winkler K, Konrad T, Fullert S, Friedrich I, Destani R, Baumstark MW, et al.
Pioglitazone reduces atherogenic dense LDL particles in nondiabetic patients with arterial
hypertension: a double-blind, placebo-controlled study. Diabetes Care 2003;26(9):2588–94.
[73] Bailey CJ, Turner RC. Metformin. N Engl J Med 1996;334(9):574–9.
[74] Cusi K, Consoli A, DeFronzo RA. Metabolic effects of metformin on glucose and lactate
metabolism in noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1996;
81(11):4059–67.
[75] Prospective Diabetes Study UK. (UKPDS) Group. Effect of intensive blood-glucose
control with metformin on complications in overweight patients with type 2 diabetes
(UKPDS 34). Lancet 1998;352(9131):854–65.
[76] Kirpichnikov D, McFarlane SI, Sowers JR. Metformin: an update. Ann Intern Med 2002;
137(1):25–33.
[77] Reaven GM, Johnston P, Hollenbeck CB, Skowronski R, Zhang JC, Goldfine ID, et al.
Combined metformin-sulfonylurea treatment of patients with noninsulin-dependent
diabetes in fair to poor glycemic control. J Clin Endocrinol Metab 1992;74(5):1020–6.
R.M. Krauss, P.W. Siri / Med Clin N Am 88 (2004) 897–909 909

[78] Robinson AC, Burke J, Robinson S, Johnston DG, Elkeles RS. The effects of metformin
on glycemic control and serum lipids in insulin-treated NIDDM patients with suboptimal
metabolic control. Diabetes Care 1998;21(5):701–5.
[79] DeFronzo RA, Goodman AM. Efficacy of metformin in patients with non-insulin-
dependent diabetes mellitus. The Multicenter Metformin Study Group. N Engl J Med
1995;333(9):541–9.
Med Clin N Am 88 (2004) 911–931

Endothelial dysfunction and


hypertension in diabetes mellitus
Paresh Dandona, MD, DPhil, FRCP, FACP, FACC*,
Ajay Chaudhuri, MBBS, MRCP, Ahmad Aljada, PhD
Division of Endocrinology, Diabetes, and Metabolism, State University of New York
at Buffalo and Kaleida Health, 3 Gates Circle, Buffalo, NY 14209, USA

The maintenance of the vascular tone and luminal diameter of a blood


vessel is dependent on the net balance between vasoconstrictor and
vasodilator forces. This is relevant both to the basal state and to states
following a challenge or stress. The major endothelial vasodilatory factors
acting on the vascular smooth muscle are nitric oxide (NO), prostacyclin
(PGI2), and hyperpolarizing factor, whereas the major vasoconstrictor
forces are endothelin-1, angiotensin II, norepinephrine, serotonin (5-HT),
and thromboxane (TX)A2 [1–7]. Although the net balance between
vasoconstriction and vasodilation determines the tone of the blood vessel,
the vasodilatory–vasoconstrictive response following a challenge may also
be determined by the intrinsic mechanical and biological properties of the
vascular smooth muscle. In both diabetes mellitus and obesity, vascular
reactivity, defined as the response to a challenge, is known to be abnormal,
whereas the basal vascular tone and blood flow is usually normal; however,
in some cases the blood flow may even be supranormal [8–12]. Following
ischemia, CO2 challenge, thermal challenge, or exercise, the diabetic or the
obese subject is unable to increase blood flow or to vasodilate in a manner
similar to that in normal subjects [8,13–20]. Subnormal vasodilation has
been shown to occur following CO2 challenge in cerebral blood flow [16],
after exercise in muscle blood flow [17], thermal challenge in skin blood flow
and transcutaneous pO2 [18], and following exercise in the myocardium
[19,20].
The mechanisms involved in abnormal reactivity may involve both
the endothelium and the vascular smooth muscle. Consistent with the
endothelial damage known to occur in diabetics, the level of secretion and

* Corresponding author.
E-mail address: pdandona@kaleidahealth.org (P. Dandona).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.006
912 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

the bioavailability of NO and PGI2, the two major vasodilators, is


diminished [8,14,21–24]. This appears to be particularly relevant to their
secretion/release following a challenge, for example, from acetylcholine,
CO2, ischemia, thermal stress, or exercise. There is a simultaneous increase
in the synthesis, secretion, and action of the vasoconstrictors, endothelin-1,
TXA2, and sensitivity to angiotensin II [25–27]. Plasma concentrations of 5-
HT are increased whereas intra-platelet 5-HT content is depleted [6,28,29];
thus, there is an increase in vasoconstrictors whereas there is a decrease in
vasodilators. In diabetes, TXA2 may be derived both from hyperaggregable
platelets and from the arterial wall [30–33].
It is of interest that the basal vascular tone and flow in diabetics is often
not subnormal. In fact, the basal blood flow, as measured in the diabetic
foot, may be increased; however, the ability of the flow and the arterial
diameter to be increased following challenge is markedly reduced [8,34].

Action of nitric oxide, acetylcholine, and insulin


NO is a vasodilator, and its vasodilatory action is mediated by a sequence
of steps. NO activates guanylate cyclase, which converts guanosine tri-
phosphate into c-GMP in vascular smooth muscle cells (VSMC), which in
turn causes the VSMC to relax. This sequence results in vasodilation [35,36].
Acetylcholine (ACh), acting through a muscarinic receptor, stimulates NO
release. Insulin causes vasodilation through the sequential phosphorylation
of its receptor, the activation of phosphatidylinositol (PI) 3-kinase, Akt
kinase, NO synthase (NOS) activation, and NO generation [37–39]. Insulin
also causes an increase in NOS expression [40]. Thus, both ACh and insulin
cause endothelium-dependent NO-mediated vasodilation. Post-ischemic
vasodilation is also endothelium dependent and is mediated by NO.
Endothelial function as mediated by NO can be tested by any of the three
modalities ACh, insulin, and ischemia. PGI2, a potent vasodilator and an
inhibitor of platelet aggregation, exerts these effects through the activation
of adenyl cyclase and the formation of c-AMP, which causes the relaxation
of VSMC and the inhibition of platelet aggregation [41,42].

Effect of hyperglycemia, increase in free fatty acids, diabetes, and obesity


High glucose concentrations cause a reduction in NOS expression in
endothelial cells [43,44]. In humans, acute hyperglycemia causes a reduction
in postischemic and ACh-induced vasodilation [45]. Intake of glucose (75 g)
also results in an increase in the generation of reactive oxygen species
(ROS), an increase in intranuclear nuclear factor kappa B (NF-jB), and
a decrease in total cellular inhibitor of kappa B (IjB) in mononuclear cells
(MNC), reflecting inflammation at the cellular level [46,47]. Increased ROS
generation may decrease NO bioavailability and thus alter the vascular
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 913

responses. An acute increase in free fatty acids (FFAs) also causes an


inflammatory response, as reflected in an increase in ROS generation by
MNC and an increased intranuclear NF-jB binding and NF-jB expression
by MNC [48]. There is also an increase in macrophage migration inhibitory
factor and C-reactive protein, along with a decrease in postischemic
vasodilation by the brachial artery. An increase in FFAs also causes a fall
in insulin-induced increase in femoral artery blood flow [49], in addition to
causing an induction of a state of insulin resistance (reduction of insulin
induced glucose uptake) [50]. In the plasma of insulin-resistant states of
obesity and type 2 diabetes, FFA concentration is increased. This may also
contribute to abnormal vascular reactivity because FFAs induce a state of
insulin resistance and inflammation in addition to reducing NO release by
the endothelium. FFAs also reduce vascular PGI2 secretion as well as
rendering it unstable [24,51,52]. Similarly, hyperglycemia of diabetes may
contribute to altered vascular responses in that condition.
Diabetes is associated with reduced postischemic, ACh-, and insulin-
induced vasodilation in both arterial and venous systems [14,53–55]. Obesity
is also associated with diminished vasodilation following ischemia and
insulin [56]. Both obesity and diabetes cause marked increases in oxidative
stress and have underlying inflammation. ROS, superoxide (O2) in
particular, combine with NO and thus reduce its bioavailability [35,36].
The insulin-resistant obese person probably also has a reduced action of
insulin on NO release and e-NOS suppression. A part of this effect may be
caused in part by the inhibitory effect of the proinflammatory cytokine,
tumor necrosis factor (TNF)a, on e-NOS expression, and in part because of
the suppression by TNFa of insulin receptor expression and insulin receptor
phosphorylation [57–59]. Plasma TNFa is known to be increased in obesity
and diabetes type 2 [58,60,61]. It is possible that other proinflammatory
cytokines such as interleukin-6, known to be increased in obesity and type 2
diabetes, may exert a similar inhibitory effect on e-NOS and NO release
[62,63].
Hyperactive, hyperaggregable platelets may also contribute to a vasocon-
strictor and proinflammatory state because the platelets contain norepineph-
rine, 5-HT, and histamine, all of which are potentially proinflammatory. In
addition, platelets have arachidonic acid, which is liberated from phospho-
lipids by phospholipase A2 on platelet activation. Arachidonic acid is
converted to proinflammatory prostaglandins, including TXA2, and leuko-
trienes [64–66]. Platelets are also abundant in CD40, which with its ligand
CD40L in leucocytes and the endothelium activates inflammatory cascades
[67–71]. This too may affect NO release and vascular reactivity. The platelets
are known to be hyperaggregable in diabetes and obesity and may therefore
contribute to abnormalities of vascular reactivity in these conditions.
Thus, the state of diabetes with associated hyperglycemia, increase in
FFA concentrations, relative or absolute lack of insulin, and insulin
resistance is a proconstrictor state that would support hypertension. It is
914 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

still not totally clear, however, what the specific mediator of hypertension in
diabetes is.

Treatment of hypertension in diabetes mellitus


Diabetes mellitus and hypertension coexist in approximately 30% of type
1 and 20% to 60% of type 2 diabetics [72,73]. In type 1 diabetes, hypertension
usually occurs in association with nephropathy whereas in type 2 diabetes, it
may be present at diagnosis or even before hyperglycemia, as a part of the
metabolic syndrome. Hypertension is twice as common in the type 2 diabetic,
compared with the nondiabetic [74]; it is present in 85% of subjects with
nephropathy [75], and the coexistence of these two conditions is associated
with an increase in the risk of retinopathy, renal failure, and cardiovascular
disease (CVD) [74]. An increase of 5 mm Hg in systolic or diastolic blood
pressure increases the risk of CVD by 20% to 30%; a diastolic blood pressure
above 70 mm Hg increases the risk of retinopathy; and end-stage renal
disease is 5 to 6 times more common in the hypertensive diabetic [75–77].
In contrast to type 1 diabetes, in which microvascular complications are
the major cause of mortality and morbidity, ischemic heart disease and
stroke account for approximately 70% of the morbidity and mortality
associated with type 2 diabetes [74]. Treatment of hypertension decreases the
risk of CV and microvascular complications largely in the diabetic,
compared with the nondiabetic, and, although the benefit of lowering blood
pressure is clear, the effect of individual agents beyond their antihyperten-
sive effect is debatable.

Benefits of blood pressure lowering


Numerous trials have compared the effect of blood pressure (BP) control
on CV outcomes. In these trials, diabetics have been a part of subgroup
analysis or were the primary group studied, and control of blood pressure
with an active agent was compared with a placebo (Table 1).
In the Systolic Hypertension in The Elderly Program (SHEP) study [78],
a difference of 9.8 mm Hg in systolic and 2.2 mm Hg in diastolic blood
pressure in a diabetic subgroup (583 subjects) treated with chlorthalidone
and atenolol or reserpine, compared with placebo, resulted in a relative
reduction of 34% in CV events and a trend toward lower total mortality.
The Systolic Hypertension in Europe (Syst-Eur) study [79], which compared
patients treated with nitrendipine or placebo showed a benefit on CV
outcomes with a decrease in systolic BP of 8.6mm Hg and diastolic BP of
3.9 mm Hg in a diabetic subgroup of 492 subjects. Cardiovascular mortality
was reduced by 70%, CV events by 62%, and strokes by 69%. The
incidence of CV events in the placebo-treated diabetics was twice that of
the nondiabetics. These trials demonstrated that diuretics and calcium
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 915

Table 1
Benefits of blood pressure control; placebo-controlled trials

Eligibility BP difference
(mm Hg)
Step therapy (BP at entry Duration
Trial vs placeboa mm Hg) (yrs) Sy Di CVDb Microb
SHEP 1. Chlorthalidone
2. Atenolol HTN (170/77) 5 9.8 2.2 # 34% N/A
3. Reserpine
Syst-Eur 1. Nitrendipine HTN (175/85) 2 8.6 3.9 # 69% N/A
2. Enalapril
3. Hctz
HOPE Ramipril CVD/CVDRF 4.5 2.4 1 # 25% #16%c
(142/80)
RENAAL Losartan DM, nephro 3.4 2 0 # 10% (NS) # 21%d
(152/82)
IPDM Irbesartan HTN, DM, 2 3 0 N/A # 70%e
microalbumin
(153/91)
Abbreviations: BP difference, difference in BP between the treated and placebo groups; Di,
diastolic; DM, diabetes mellitus; HOPE, Heart Outcomes and Prevention Evaluation; HTN,
hypertension; IPDM, Irbesartan in Patients with type 2 Diabetes and Microalbuminuria;
nephro, nephropathy; NS, nonsignificant; N/A, not available; RENAAL, Reduction of
Endpoints in NIDDM with Angiotensin II Antagonist Losartan; RF, risk factor; SHEP,
Systolic Hypertension in the Elderly Program; Syst-Eur, Systolic Hypertension in Europe; Sy,
systolic.
a
Agents added sequentially if BP goal was not achieved.
b
Reduction in CVD (cardiovascular disease) and Micro (microvascular complications) in
the treated group compared with placebo.
c
Reduction in overt nephropathy albumin/creatinine >36 mg/mmol, laser therapy, dialysis.
d
Nephropathy, albumin/creatinine >300 mg/g, creatinine 1.3–3 mg/dl, 21% decrease in
doubling of serum creatinine and end-stage renal disease.
e
70% reduction in the risk of overt nephropathy for irbesartan 300 mg.

channel blocker (CCB) agents were safe and effective in reducing blood
pressure and in reducing CV complications.
In a subgroup analysis of the Heart Outcomes and Prevention Evaluation
(HOPE) study [80], subjects with diabetes and an additional risk factor for
atherosclerosis or a history of CV disease were randomized to ramipril,
10 mg or placebo. Subjects had a baseline BP of 142/80 mm Hg, the ramipril
treated group had a BP fall of 2.4/1 mm Hg at the end of the study
compared with the placebo-treated group. Ramipril-treated patients had
a 25% reduction in CV outcomes, 24% reduction in total mortality, and
a 16% reduction in microvascular endpoints. This effect of ramipril was
significant even after adjusting for the blood pressure differences. Another
arm of this study demonstrated a significant ramipril-induced decrease in
progression of carotid intimal-media thickness [81].
In the Reduction of Endpoints in NIDDM with the Angiotensin II
Antagonist Losartan (RENAAL) study [82], 1500 subjects with type 2
916 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

diabetes and nephropathy received losartan or a placebo. The blood pressure


differences were minimal, but there was a 21% reduction in the risk of renal
outcomes (doubling of serum creatinine or end-stage renal disease) in subjects
treated with losartan. There was also a reduction of 32% in hospitalization
for heart failure. In the Irbesartan in Patients with Type 2 Diabetes and
Microalbuminuria (IPDM) study [83], 590 type 2 diabetic and hypertensive
subjects were randomized to irbesartan, 300 or 150 mg or placebo, with the
primary aim to assess the effect on progression to macroalbuminuria. Systolic
blood pressure was 3 and 2mm Hg lower in the irbesartan 300 mg and 150 mg
groups, respectively, compared with placebo. This resulted in a 70% re-
duction in the progression to nephropathy in the group treated with
irbesartan 300 mg. Benefits persisted even after adjustment for blood pressure
differences. These trials were not powered to detect improvements in CV
outcomes.
On the basis of these trials, we can conclude that CV outcomes are
improved in the diabetic with the lowering of blood pressure from baseline.
They also show that angiotensin-converting enzyme (ACE)-inhibitors and
angiotensin II receptor antagonist (ARB) agents reduce CV and microvas-
cular complications with minimal changes in blood pressure.

Intensity of blood pressure control


Three trials have assessed the effects of targeting different blood pressure
levels on macro- and microvascular complications in diabetics (Table 2). In
the Hypertension Optimal treatment (HOT) trial [84], 19,000 patients were
randomized to three groups aimed at a diastolic blood pressure of 80, 85,
or 90 mm Hg. There was a substantial decrease in diastolic blood pressure in
all three groups (24.3, 22.3, and 20.3 mm Hg, respectively), with a mean
diastolic blood pressure of 81.1, 83.2, and 85.2 mm Hg that was achieved
by a five-step titration with felodipine followed by an ACE inhibitor or
b-blocker. There was no difference in CV outcomes among the three groups
in the nondiabetic population; however, in the 1500 diabetics in this trial,
there was a 51% reduction in CV events in the group who achieved
a diastolic blood pressure of 81 mm Hg compared with 85 mm Hg (a 4-mm
Hg difference). Significant reductions were seen even between the groups
(with a 2-mm Hg difference). More than three antihypertensive agents were
required to reach a diastolic blood pressure of less than 80 mm Hg.
In the United Kingdom Prospective Diabetes Study (UKPDS) [85], 1148
subjects with a mean blood pressure of 160/94 mm Hg were randomized into
a target blood pressure group of 180/105 or 150/85 mm Hg. The first group
achieved a blood pressure of 154/87 mm Hg whereas the second group
achieved 144/82 mm Hg. This respective difference of 10 mm Hg systolic and
5 mm Hg diastolic was associated with a 34% reduction in macrovascular
complications, including a significant reduction in any diabetes related
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 917

Table 2
Trials targeting different levels of BP control
Achieved BP
Target BP
(mm Hg)
(mm Hg)
Trial Step therapya control Active Control Active CVDb Microb
HOT 1. Felodipine 5 mg
2. b-Blocker or Diastolic BP Diastolic BP 144/85 140/81 # 51% N/A
ACE inhibitor
3. " Felodipine
to 10 mg
4. " step2 drug \90 \80
5. Diuretic
UKPDS 1. Captopril
or atenolol
2. Furosemide
3. Nifedipine SR \180/105 \150/85 154/87 144/82 # 34%c # 37%d
4. Methyldopa
5. Prazosin
ABCD 1. Nisolodipine Diastolic BP Diastolic BP 138/86 132/78 NSe NS
or enalapril
2. Metoprolol
3. Hctz \75 80–90
Abbreviations: ABCD, Apropriate Blood Pressure Control in Diabetes; HOT, Hypertension
Optimal treatment; UKPDS, United Kingdom Prospective Diabetes Study.
a
Agents added sequentially if BP goal was not achieved.
b
Reduction in CVD (cardiovascular disease) and Micro (microvascular complications) in
the active group compared with the control group.
c
Included a # diabetes related death by 32%, stroke by 44%, heart failure by 56%.
d
Included a 47% reduction in deterioration of visual acuity.
e
Reduction in mortality by 49%.

endpoint, deaths related to diabetes, stroke, and heart failure. There was
also a 37% decrease in microvascular complications, primarily because of
a decrease in retinal photocoagulation, and an improvement in visual acuity.
It was estimated that for every 10 mm Hg of decrease in systolic blood
pressure, there is a 12% reduction in a diabetes related endpoint, a 15%
reduction in death because of diabetes, an 11% reduction in myocardial
infarction, and a 13% decrease in microvascular complications [86]. There
was no threshold of risk, and the best outcomes were seen in subjects with
a systolic BP of less than 120 mm Hg. The benefits of better blood pressure
control were greater than that of improvement in glycemic control
(glycosylated hemoglobin 7.9% versus 7.0%), with a 2- to 5-fold greater
absolute risk reduction and much lower numbers needed to treat [87]. The
benefits of blood pressure lowering, however, were seen in both the
conventional and intensively treated groups, and the benefits of glycemic
control were seen in the aggressively and conservatively targeted blood
pressure groups [85]. Thus, treatment of a single risk factor does not
eliminate the risk of complications, and all risk factors need to be
918 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

aggressively treated in diabetes to reduce the risk of micro- and macro-


vascular complications. As in the HOT trial, more than two antihyperten-
sive agents were required to reach the lower BP, and 29% of the subjects
required three agents.
In the Appropriate Blood Pressure Control in Diabetes (ABCD) [88], 470
patients with diabetes were randomized to an intensive (diastolic blood
pressure 75 mm Hg) or a moderate control (diastolic blood pressure of 80–
89 mm Hg) group. The primary aim was to assess the effect of blood
pressure control on nephropathy. The intensive group achieved a blood
pressure of 132/78 mm Hg whereas the moderate control group
achieved a blood pressure of 138/86 mm Hg. There was a 50% reduction
in total mortality between the groups that could not be explained on the
basis of CV outcomes. No difference was seen in microvascular endpoints.
Mean glycosylated hemoglobin was greater than 11% during the 5 years of
this trial, and there was only a 45% adherence to the assigned study
medication. These factors may have contributed to the lack of benefit on
micro- and macrovascular outcomes with more intensive blood pressure
control in this trial.
Analysis of other clinical trials over the past decade, including the
Modification of Diet in Renal Disease (MDRD) trial [89], has shown that
the lower the blood pressure, the greater is the preservation of renal function
in subjects with kidney disease, irrespective of diabetes. In subjects with
proteinuria greater than 1 g/d and renal insufficiency, a blood pressure lower
than 125/75 mm Hg is associated with a slower rate of decline in kidney
function. There is evidence to suggest an increase in CV and renal disease
risk with systolic BP greater than 127 mm Hg and diastolic BP greater than
83 mm Hg [90,91]. In the RENAAL study, a baseline systolic BP of 140 to
159 mm Hg was associated with an increased risk of end-stage renal disease
or death by 38% compared with subjects with a systolic BP of 130 mm Hg
[92]. Moreover, in the trials of intensive blood pressure control in diabetics,
there is no indication of an adverse effect on CV or renal events with a lower
BP [85]. This is in contrast to a recent meta-analysis of trials in non-diabetic
kidney disease, in which a systolic BP less than 110 mm Hg in subjects with
a daily protein excretion of greater than 1 g/d was associated with an in-
crease in the risk of disease progression [93].
On the basis of these observations, the American Diabetes Association,
National Kidney Foundation, and the Joint National Committee VII
recommend a blood pressure target of 130/80 mm Hg in diabetics [94–96].

Comparison of specific antihypertensive therapies on cardiovascular outcomes


Numerous trials have compared newer antihypertensive agents of ACE I,
CCB, and ARB classes with each other or against diuretics and b-blockers
(Table 3). In a substudy of the ABCD trial, nisoldipine was compared with
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 919

Table 3
Trials comparing specific anti-hypertensive therapies on cardiovascular outcomes
Trial Reference Active CVDa Microa
ACE I vs calcium channel blockers
FACET Amlodipine Fosinopril # 51% N/A
ABCD Nisolidipine Enalapril # 57% N/A
STOP-2 Isradipine/felodipine Enalapril/lisinopril NSb N/A
ACE I vs b-blocker/diuretic
STOP-2 b-Blocker  diuretic Enalapril/lisinopril NS N/A
CAPP b-Blocker  diuretic Captopril # 41%c N/A
UKPDS Atenolol Captopril NS NS
ALLHAT Chlortahlidone Lisinopril NSd NS
Calcium channel blocker vs b-blocker/diuretic
NORDIL b-Blocker  diuretic Diltiazem NSe N/A
INSIGHT Coamilozide Nifedipine GITS NS N/A
ALLHAT Chlorthalidone Amlodipine NSf N/A
STOP-2 b-Blocker  diuretic Enalapril/lisinopril NS N/A
ARB vs b-blocker
LIFE Atenolol Losartan # 24% N/A
ARB vs calcium channel blocker
IDNT Amlodipine Irbesartan NS # 23%
Abbreviations: ABCD, Appropriate Blood pressure Control in Diabetes; ALLHAT, Anti-
hypertensive and Lipid-Lowering treatment to prevent Heart Attack Trial; CAPP, Captopril
Prevention Project; FACET, Fosinopril versus Amlodipine Cardiovascular Events Trial;
IDNT, Irbesartan Diabetic Nephropathy trial; INSIGHT, International Nifedipine GITS
Study: Intervention as a Goal in Hypertension Treatment; LIFE, Losartan Intervention for
Endpoint reduction; NORDIL, Nordic Diltiazem; NS, non significant; N/A, not available;
STOP-2, Swedish Trial in Old Patients with Hypertension-2; UKPDS, United Kingdom
Prospective Diabetes Study.
a
Reduction in CVD (cardiovascular disease) and Micro (microvascular complications) in
the active group compared to the reference group.
b
Reduction in myocardial infarction by 49% in ACE I group.
c
Reduction in myocardial infarction by 66%.
d
22% increase in heart failure in lisinopril group.
e
20% reduction in stroke with diltiazem.
f
42% increase in heart failure with amlodipine.

enalapril [88]. Blood pressure reduction was similar in both arms, but by the
end of the trial, half the subjects were not taking the initially assigned
treatment. In an intention-to-treat analysis, the subjects in the nisoldipine
group had a 5 times higher risk of myocardial infarction (a secondary
endpoint) that was attributed to a beneficial effect of ACE inhibition rather
than an adverse effect of nisoldipine.
In the Fosinopril versus Amlodipine Cardiovascular Events Trial
(FACET) [97], 380 diabetic subjects were randomly assigned to Fosinopril
or Amlodipine. Although the Amlodipine group had a lower systolic blood
pressure than the Fosinopril group, combined CV events were significantly
lower in the fosinopril group primarily because of a difference in the number
920 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

of subjects with hospitalized angina (0 versus 4). The trial was designed to
study treatment-related differences in serum lipids and diabetes control, with
CV endpoints as secondary outcomes. This was also an open-label study,
and if the blood pressure goal was not attained on the assigned study drug,
the other study drug was added. The best outcomes were seen in the subjects
receiving both drugs (108 subjects).
In the Swedish Trial in Old Patients with Hypertension (STOP)-2 [98],
CCB, ACE I, and diuretics plus b-blockers were compared with each other.
Blood pressure lowering was similar between the groups with no difference
in combined CV events or total mortality. However, myocardial infarction
was significantly lower and strokes non-significantly higher in patients
treated with ACE I compared with CCB.
The largest hypertension trial to date, the Antihypertensive and Lipid
Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) [99],
compared the effectiveness of a thiazide diuretic (chlorthalidone), an ACE
I (lisinopril), a CCB (Amlodipine), and an a1-blocker (Doxazosin) as an
initial treatment for hypertension in patients at high risk for a CV event. The
doxazosin arm was suspended early because of an increase in the incidence
of heart failure. The primary outcome was combined nonfatal and fatal
myocardial infarction. Secondary outcomes were all-cause mortality, stroke,
combined coronary heart disease (primary outcome, coronary revascular-
ization or angina with hospitalization), and combined CVD (combined
coronary heart disease, stroke, angina without hospitalization, heart failure,
and peripheral arterial disease). Systolic blood pressure was the lowest in the
diuretic group whereas diastolic blood pressure was the lowest in the
Amlodipine group. In a pre-specified subgroup analysis of 12,063 type 2
diabetics, the incidence of heart failure was the lowest in the diuretic group
compared with lisinopril (RR 1.22 [confidence interval {CI}, 1.05–1.42]) and
Amlodipine (RR 1.42 [CI, 1.23–1.64]). Combined CVD was higher in the
ACE I group compared with the diuretic group (RR 1.08 [CI, 1.00–1.17])
primarily because of a higher risk of heart failure and stroke in the black
population; systolic blood pressure was 4 mm Hg higher than the diuretic
group in this population. There was no difference between the groups with
regard to other outcomes in the diabetic population. Forty percent of the
patients required two medications and one third did not reach the BP goal
of 140/90 mm Hg. Microvascular endpoints are also being evaluated, but
they have not yet been reported. There have been numerous explanations of
the results of the ALLHAT. One explanation is the withdrawal of diuretics
in patients receiving this treatment in the ACE inhibitor and Amlodipine
arm at the beginning of the study, which may have unmasked heart failure
in these patients and therefore resulted in the higher incidence of this
complication in these groups. The other explanation is the inability to use
diuretics in the ACE inhibitor arm, a combination that is commonly used in
clinical practice to enhance the BP-lowering effects of ACE I in blacks. On
the basis of these arguments, it has been suggested that ALLHAT should be
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 921

viewed as a trial that showed diuretics, CCB, and ACE I to be equally


efficacious as initial hypertensive agents. In contrast to ALLHAT, a recent
Australian trial, the Second Australian National Blood Pressure study
(ANBP)-2 [100] has shown a superiority of ACE inhibition over diuretics in
an elderly white population (RR 0.89 [CI, 0.79–1.00]), with benefits seen
primarily in the male subgroup. Overall, there was an 11% reduction in the
first CV event or death from any cause, with a 32% reduction in the
incidence of myocardial infarction in the ACE I group; however, it should
be noted that the population studied in ANBP-2 was different from that of
ALLHAT. Also, ANBP-2 was an open-label study, and the number of
events in this study was much lower.
The UKPDS study compared ACE I with b-blockers [101]. In this study,
there was no difference in microvascular or macrovascular outcomes in
subjects randomized to either captopril or atenolol. Blood pressure lowering
was similar between the two groups (143/81 mm Hg in atenolol versus 144/
83 mm Hg in captopril). Smaller studies in type 2 diabetics have also not
shown a consistent superiority of ACE inhibition over b-blockers with re-
spect to renal outcomes [102].
Studies comparing calcium channel blockers to b-blockers and diuretics
include the Nordic Diltiazem Trial (NORDIL) and The International
Nifedipine GITS study: Intervention as a Goal in Hypertension Treatment
(INSIGHT) [84,103]. In both the NORDIL (diltiazem versus b-blocker with
and without diuretic) and INSIGHT (long-acting nifedipine versus coami-
lozide) trials, there were no differences in CV outcomes between the groups
studied. NORDIL randomized 727 patients, whereas INSIGHT included
1302 patients with type 2 diabetes.
Three trials have assessed the effect of ARBs in diabetic patients with
nephropathy, assessing CV endpoints as secondary outcomes. The Irbesar-
tan Diabetic Nephropathy trial showed the superiority of irbesartan over
amlodipine or placebo in terms of renal outcomes but did not show any
difference in CV outcomes between the agents [104]. There was a 23%
reduction in heart failure requiring hospitalizations. In the Losartan
Intervention for Endpoint Reduction (LIFE) study, there was 24% re-
duction in CV endpoints, 37% reduction in CV-related deaths, and 39%
reduction in all-cause mortality in 1195 hypertensive diabetic subjects with
left ventricular hypertrophy treated with losartan, compared with those
treated with atenolol [105]. Albuminuria was less frequent in the losartan
group. Although systolic blood pressure was 2 mm Hg lower in the Losartan
group, benefits were independent of this effect.

Effect of antihypertensive therapy on microvascular complications


A large placebo-controlled clinical trial in type 1 diabetics with overt
nephropathy (urinary protein 500 mg/24 h) has shown a beneficial effect of
922 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

captopril on slowing the progression of renal disease [106]. There was a 50%
reduction in the endpoints of death, dialysis, transplantation, and doubling
of serum creatinine with captopril, compared with placebo. Although mean
arterial blood pressure was lower by 2 mm Hg in the captopril group, this
effect was shown to be independent of the BP lowering. Studies in type 1
diabetics with microalbuminuria in both hypertensive and normotensive
populations have also shown a reduction in progression to macroalbumi-
nuria and in regression to normoalbuminuria [107]. A recent trial [108] has
also shown an improvement in microalbuminuria unrelated to ACE
inhibition in type 1 diabetics. In this trial, one of the factors associated
with regression of microalbuminuria was a systolic BP less than 115 mm Hg.
ACE I has also been shown to reduce the risk of retinopathy and
neuropathy in type 1 diabetes [109,110].
In type 2 diabetes, ACE I, ARB, b-blockers and nondihydropyridine
CCBs have been shown to reduce the progression to nephropathy and
decline in renal function in micro- and macroalbuminuric normotensive and
hypertensive subjects [82,83,102,106,107,111]. Data regarding a renoprotec-
tive effect in overt nephropathy, however, are most convincing for ARBs
because they have been tested in large placebo-controlled clinical trials in
which the benefit has been shown to be independent of their BP-lowering
effect. Small trials [112–114] have also shown a beneficial effect of combining
ACE I and ARB, and thiazolidinediones and spironolactone on proteinuria.
Although there is little evidence regarding the prophylactic use of ACE I or
ARBs to prevent microalbuminuria, there was a nonsignificant decrease in
the development of albuminuria in the type 2 diabetics in the HOPE and
LIFE studies [80,98].

Choice of antihypertensive regimen


Clinical trial data have shown that to reach the recommended blood
pressure goals in type 2 diabetics, usually two and frequently three agents
are required. Treatment choices depend on coexistent conditions, side
effects, and contraindications.
On the basis of the evidence, in the absence of any complications of
diabetes, a thiazide diuretic along with an ACE I or ARB is recommended
for initial therapy in a type 2 diabetic with hypertension. Thiazide diuretics
are as efficacious as any other antihypertensive, are cost effective, and they
work synergistically with ACE I, ARBs, b-blockers, and CCBs. Potassium-
sparing diuretics such as spironolactone can also be used because they have
been shown to improve prognosis in heart failure and after myocardial
infarction, a benefit attributed to the antagonism of aldosterone [115].
There is strong evidence of a CV protective effect of ACE I and ARB,
independent of their BP-lowering effect. These agents are also effective in
reducing the progression of nephropathy in micro- and macroalbuminuric
subjects; they improve the prognosis following myocardial infarction and in
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 923

heart failure, the risks of which are high in a diabetic; and they limit the side
effects of diuretics (hypokalemia) and CCBs (edema). Moreover, therapy
with ACE I or ARB is well tolerated. Presently the only indication for
choosing an ARB over an ACE I is in patients with overt nephropathy and
those with left ventricular hypertrophy in view of the observations in
RENAAL, IDNT, and LIFE studies. Therapies with the two agents are
interchangeable if there are side effects, most commonly a cough with ACE
I. In blacks, there is also a higher incidence of angioedema with these agents.
With the initiation of ACE I therapy, a noncontinuous increase in creatinine
of up to 30% from baseline can be expected, and this should not prompt
discontinuation of this therapy [116]. Although it is associated with an
increase in the incidence of hyperkalemia, ACE I therapy in subjects with
creatinine levels greater than 3 mg/dL can delay the time to dialysis and
confer CV protection in these high-risk patients [116,117].
Combination therapy with a diuretic and ACE I or ARB should be
considered if blood pressure lowering of greater than 15 mm Hg systolic or
10 mm Hg diastolic is required; however, it should be initiated carefully in
subjects at a risk for orthostatic hypotension. In subjects with coronary
artery disease (CAD) and heart failure, b-blockers should be instituted
along with ACE I or ARB because they are known to improve prognosis in
these conditions [118]. If BP is inadequately controlled with a combination
of ACE I or ARB and diuretic, the present authors still favor the use of b-
blockers as the third-line agent, even in the absence of CAD or heart failure.
This recommendation is based on the increased prevalence of asymptomatic
heart disease and autonomic neuropathy in diabetics, the increased risk of
CAD and heart failure in this population, and the evidence of equal efficacy
to ACE inhibition in the UKPDS. b-blockers, however, have been shown to
be less efficacious in the elderly, less tolerated in clinical studies, and
associated with mild weight gain in the UKPDS trial.
In subjects with proteinuria, there is evidence that suggests a beneficial
effect of nondihydropidine CCB, and in the National Kidney Foundation
guidelines, these agents are the preferred third-line agents in the treatment of
hypertension. There is also evidence regarding a synergistic effect of CCBs
when combined with ACE I. In both the HOT and HOPE trials,
approximately 45% of the subjects were taking this combination therapy,
and the best outcomes in FACET were seen in the group taking this
combination. CCBs are also more effective than b-blockers and ACE I in
lowering BP in the elderly. CCBs have a neutral effect on metabolic
parameters, and there are no contraindications to dihydropyridine CCBs;
however, dihydropyridine CCBs should be used in combination with ACE I
in the presence of nephropathy, and nondihydropyridine CCB should be
combined cautiously with b-blockers in the elderly because b-blockers might
lead to bradyarrhythmias. If further BP lowering is required, any other
agent can be used; however, clonidine should not be used with b-blockers
for numerous reasons including an increased likelihood of bradycardia.
924 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

Apart from blood pressure control, it is also important to aggressively


improve glycemic control and target hyperlipidemia. Aspirin therapy should
be prescribed in any diabetic with CVD or a diabetic over 30 years of age
with an additional CVD risk factor. Such a comprehensive approach has
resulted in a 53% reduction in macrovascular complications, 61% reduc-
tion in nephropathy, 58% reduction in retinopathy and 63% reduction in
autonomic neuropathy in a small study [119].
In conclusion, lowering blood pressure to a target of 130/80 mm Hg
reduces CV and microvascular complications in diabetes. Therapy with more
than one agent is required to reach these goals in this population. In the
absence of any complications or contraindications, treatment should be
initiated with a diuretic combined with an ACE I or an ARB. The choice of
a third-line agent depends on coexistent factors. b-blockers are preferred in
the young and for their cardioprotective effects, whereas CCBs are preferred
in the elderly and for their renoprotective effect.

Antioxidant and anti-inflammatory effects of antihypertensive drugs


Recent observations on the actions of b-blockers and ARBs are worth
noting. Carvedilol and nadolol have been shown to suppress ROS gene-
ration by leucocytes. This may contribute to their antihypertensive effect
through a reduction in O2 formation and an increase in NO bioavailability.
The angiotensin II receptor blocker, valsartan, has been shown to
suppress not only ROS generation by leucocytes but also to exert a potent
anti-inflammatory effect at the cellular and molecular level [120]. It is
possible that ACE inhibitors will exert similar effects. Because oxidative
stress and inflammation are cardinal to atherogenesis, it is possible that the
novel effects of b-blockers and ARBs contribute to their beneficial action on
CV outcomes.

The potential role of antidiabetic drugs in the treatment of hypertension


In this discussion of antihypertensive drugs and their effect on CV and
renal outcomes, it is important to mention that, in the UKPDS study [121],
metformin has been shown to significantly improve CV outcomes, and more
recently, it has been shown to exert an anti-inflammatory action. Thiazo-
lidinediones have been shown to have an anti-inflammatory effect [122,123],
and thus in the long term they may be antiatherogenic. They also reverse
abnormalities in vascular reactivity in both the obese and type 2 diabetics
while improving microalbuminuria. Insulin has been shown to be a vasodi-
lator with antiplatelet activity as well as exerting a potent anti-inflammatory
effect [124–126]. Its usage improves vascular reactivity in the long term [127].
Thus, the use of these drugs may facilitate the action of other antihyper-
tensive drugs. This, however, has yet to be proved.
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 925

References
[1] Furchgott RF, Zawadzki JV. The obligatory role of endothelial cells in the relaxation of
arterial smooth muscle by acetylcholine. Nature 1980;288:373–6.
[2] Mombouli JV, Vanhoutte PM. Endothelium-derived hyperpolarizing factor(s): updating
the unknown. Trends Pharmacol Sci 1997;18:252–6.
[3] Levine E. The endothelin peptides. In: Sowers J, editor. Endocrinolgy of the vasculature.
Totawa, NJ: Humana Press; 1996. p. 49–66.
[4] Luscher TF, Barton M. Biology of the endothelium. Clin Cardiol 1997;20:II-3–II-10.
[5] Cockcroft JR, O’Kane KP, Webb DJ. Tissue angiotensin generation and regulation of
vascular tone. Pharmacol Ther 1995;65:193–213.
[6] Barradas MA, Gill DS, Fonseca VA, Mikhailidis DP, Dandona P. Intraplatelet serotonin
in patients with diabetes mellitus and peripheral vascular disease. Eur J Clin Invest 1988;
18:399–404.
[7] Benyo Z, Gorlach C, Wahl M. Interaction between nitric oxide and thromboxane A2 in
the regulation of the resting cerebrovascular tone. Adv Exp Med Biol 1999;471:373–9.
[8] Johnstone MT, Creager SJ, Scales KM, Cusco JA, Lee BK, Creager MA. Impaired
endothelium-dependent vasodilation in patients with insulin- dependent diabetes mellitus.
Circulation 1993;88:2510–6.
[9] De Vriese AS, Verbeuren TJ, Van de Voorde J, Lameire NH, Vanhoutte PM. Endothelial
dysfunction in diabetes. Br J Pharmacol 2000;130:963–74.
[10] Rocchini AP, Key J, Bondie D, Chico R, Moorehead C, Katch V, et al. The effect of
weight loss on the sensitivity of blood pressure to sodium in obese adolescents. N Engl J
Med 1989;321:580–5.
[11] Hall JE, Brands MW, Dixon WN, Smith MJ Jr. Obesity-induced hypertension. Renal
function and systemic hemodynamics. Hypertension 1993;22:292–9.
[12] Carroll JF, Huang M, Hester RL, Cockrell K, Mizelle HL. Hemodynamic alterations in
hypertensive obese rabbits. Hypertension 1995;26:465–70.
[13] McVeigh GE, Brennan GM, Johnston GD, McDermott BJ, McGrath LT, Henry WR,
et al. Impaired endothelium-dependent and independent vasodilation in patients with
type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 1992;35:771–6.
[14] Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA. Impaired nitric oxide-
mediated vasodilation in patients with non- insulin-dependent diabetes mellitus. J Am
Coll Cardiol 1996;27:567–74.
[15] Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P, Park JY, et al. Microvascular
and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes. Diabetes
1999;48:1856–62.
[16] Dandona P, James IM, Newbury PA, Woollard ML, Beckett AG. Cerebral blood
flow in diabetes mellitus: evidence of abnormal cerebrovascular reactivity. BMJ 1978;2:
325–6.
[17] Menon RK, Grace AA, Burgoyne W, Fonseca VA, James IM, Dandona P. Muscle blood
flow in diabetes mellitus. Evidence of abnormality after exercise. Diabetes Care 1992;15:
693–5.
[18] Gaylarde PM, Fonseca VA, Llewellyn G, Sarkany I, Thomas PK, Dandona P.
Transcutaneous oxygen tension in legs and feet of diabetic patients. Diabetes 1988;37:
714–6.
[19] Tune JD, Richmond KN, Gorman MW, Feigl EO. Control of coronary blood flow
during exercise. Exp Biol Med (Maywood) 2002;227:238–50.
[20] Tsujimoto G. Impaired coronary microvascular function in diabetics. Ann Nucl Med
2000;14:165–72.
[21] Calver A, Collier J, Vallance P. Inhibition and stimulation of nitric oxide synthesis in the
human forearm arterial bed of patients with insulin-dependent diabetes. J Clin Invest
1992;90:2548–54.
926 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

[22] Hendra T, Betteridge DJ. Platelet function, platelet prostanoids and vascular prostacyclin
in diabetes mellitus. Prostaglandins Leukot Essent Fatty Acids 1989;35:197–212.
[23] Shepherd GL, Lewis PJ, Blair IA, de Mey C, MacDermot J. Epoprostenol (prostacyclin,
PGI2) binding and activation of adenylate cyclase in platelets of diabetic and control
subjects. Br J Clin Pharmacol 1983;15:77–81.
[24] Jeremy JY, Mikhailidis DP, Dandona P. Simulating the diabetic environment modifies in
vitro prostacyclin synthesis. Diabetes 1983;32:217–21.
[25] Sarman B, Toth M, Somogyi A. Role of endothelin-1 in diabetes mellitus. Diabetes
Metab Rev 1998;14:171–5.
[26] Udvardy M, Torok I, Rak K. Plasma thromboxane and prostacyclin metabolite ratio in
atherosclerosis and diabetes mellitus. Thromb Res 1987;47:479–84.
[27] Ferriss JB, O’Hare JA, Kelleher CC, Sullivan PA, Cole MM, Ross HF, et al. Diabetic
control and the renin-angiotensin system, catecholamines, and blood pressure.
Hypertension 1985;7:II58–63.
[28] Malyszko J, Urano T, Knofler R, Taminato A, Yoshimi T, Takada Y, et al. Daily
variations of platelet aggregation in relation to blood and plasma serotonin in diabetes.
Thromb Res 1994;75:569–76.
[29] Winocour PH, Klimiuk P, Grennan A, Baker RD, Weinkove C. Platelet and plasma
vasoactive amines in type 1 (insulin-dependent) diabetes mellitus with and without
vascular disease. Ann Clin Biochem 1990;27(Pt 3):238–43.
[30] Sterin-Borda L, Borda ES, Gimeno MF, Lazzari MA, del Castillo E, Gimeno AL.
Contractile activity and prostacyclin generation in isolated coronary arteries from
diabetic dogs. Diabetologia 1982;22:56–9.
[31] Davi G, Catalano I, Averna M, Notarbartolo A, Strano A, Ciabattoni G, et al.
Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N Engl J
Med 1990;322:1769–74.
[32] Jeremy JY, Mikhailidis DP, Thompson CS, Barradas MA, Dandona P. Platelet
thromboxane A2 synthesizing capacity is enhanced by fasting but diminished by diabetes
mellitus in the rat. Diabetes Res 1988;8:177–81.
[33] Takahashi R, Shiraki M, Morita I, Ito H, Murota S, Orimo H. Platelet thromboxane
synthesizing activity in non-insulin-dependent diabetes: correlation with diabetic
retinopathy and diabetic treatment. Prostaglandins Leukot Med 1985;17:149–58.
[34] Smits P, Kapma JA, Jacobs MC, Lutterman J, Thien T. Endothelium-dependent vascular
relaxation in patients with type I diabetes. Diabetes 1993;42:148–53.
[35] Moncada S. Nitric oxide. J Hypertens Suppl 1994;12:S35–9.
[36] Michel T, Feron O. Nitric oxide synthases: which, where, how, and why? J Clin Invest
1997;100:2146–52.
[37] Zeng G, Quon MJ. Insulin-stimulated production of nitric oxide is inhibited by
wortmannin. Direct measurement in vascular endothelial cells. J Clin Invest 1996;98:
894–8.
[38] Zeng G, Nystrom FH, Ravichandran LV, Cong LN, Kirby M, Mostowski H, et al. Roles
for insulin receptor, PI3-kinase, and Akt in insulin-signaling pathways related to
production of nitric oxide in human vascular endothelial cells. Circulation 2000;101:
1539–45.
[39] Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM. Activation of
nitric oxide synthase in endothelial cells by Akt- dependent phosphorylation. Nature
1999;399:601–5.
[40] Aljada A, Dandona P. Effect of insulin on human aortic endothelial nitric oxide synthase.
Metabolism 2000;49:147–50.
[41] Nicosia S, Oliva D, Bernini F, Fumagalli R. Prostacyclin-sensitive adenylate cyclase and
prostacyclin binding sites in platelets and smooth muscle cells. Adv Cyclic Nucleotide
Protein Phosphorylation Res 1984;17:593–9.
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 927

[42] Rovati GE, Giovanazzi S, Negretti A, Nicosia S. Prostacyclin effects on adenylate cyclase
in platelets and vascular smooth muscle: interaction with an inhibitory receptor or partial
agonism? Adv Prostaglandin Thromboxane Leukot Res 1995;23:263–5.
[43] Chakravarthy U, Hayes RG, Stitt AW, McAuley E, Archer DB. Constitutive nitric oxide
synthase expression in retinal vascular endothelial cells is suppressed by high glucose and
advanced glycation end products. Diabetes 1998;47:945–52.
[44] Ding Y, Vaziri ND, Coulson R, Kamanna VS, Roh DD. Effects of simulated
hyperglycemia, insulin, and glucagon on endothelial nitric oxide synthase expression.
Am J Physiol Endocrinol Metab 2000;279:E11–7.
[45] Giugliano D, Marfella R, Coppola L, Verrazzo G, Acampora R, Giunta R, et al.
Vascular effects of acute hyperglycemia in humans are reversed by L-arginine. Evidence
for reduced availability of nitric oxide during hyperglycemia. Circulation 1997;95:
1783–90.
[46] Mohanty P, Hamouda W, Garg R, Aljada A, Ghanim H, Dandona P. Glucose challenge
stimulates reactive oxygen species (ROS) generation by leucocytes. J Clin Endocrinol
Metab 2000;85:2970–3.
[47] Aljada A, Mohanty P, Ghanim H, Abdo T, Tripathy D, Chaudhuri A, et al. Increase of
nuclear factor-kb (NF-kb) and decrease in inhibitor kb (Ikb) in mononuclear cells
following a mixed meal: evidence for a pro-inflammatory effect. In Amer J Clon Nutr
2002, in press.
[48] Tripathy D, Aljada A, Ghanim H, Tufail S, Chaudhuri A, Dandona P. Acute elevation of
plasma free fatty acids increases reactive oxygen species (ROS) generation by
polymorphonuclear cells, induces nuclear factor-kappa B (NF-kB) and impairs brachial
artery reactivity in healthy subjects. Diabetes 2002;51(Suppl 2):A318.
[49] Lundman P, Tornvall P, Nilsson L, Pernow J. A triglyceride-rich fat emulsion and free
fatty acids but not very low density lipoproteins impair endothelium-dependent vaso-
relaxation. Atherosclerosis 2001;159:35–41.
[50] Boden G, Chen X, Ruiz J, White JV, Rossetti L. Mechanisms of fatty acid-induced
inhibition of glucose uptake. J Clin Invest 1994;93:2438–46.
[51] Forster W, Beitz J, Hoffmann P. Stimulation and inhibition of PGI2 synthetase activity
by phospholipids (PL), cholesterol esters (CE), unesterified fatty acids (UFA) and
lipoproteins (LDL and HDL). Artery 1980;8:494–500.
[52] Mikhailidis DP, Mikhailidis AM, Barradas MA, Dandona P. Effect of nonesterified fatty
acids on the stability of prostacyclin activity. Metabolism 1983;32:717–21.
[53] Clarkson P, Celermajer DS, Donald AE, Sampson M, Sorensen KE, Adams M, et al.
Impaired vascular reactivity in insulin-dependent diabetes mellitus is related to
disease duration and low density lipoprotein cholesterol levels. J Am Coll Cardiol 1996;
28:573–9.
[54] Khan F, Elhadd TA, Greene SA, Belch JJ. Impaired skin microvascular function in
children, adolescents, and young adults with type 1 diabetes. Diabetes Care 2000;23:
215–20.
[55] Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin
resistance is associated with endothelial dysfunction. Implications for the syndrome of
insulin resistance. J Clin Invest 1996;97:2601–10.
[56] Feldman RD, Bierbrier GS. Insulin-mediated vasodilation: impairment with increased
blood pressure and body mass. Lancet 1993;342:707–9.
[57] Aljada A, Ghanim H, Assian E, Dandona P. Tumor necrosis factor-alpha inhibits insulin-
induced increase in endothelial nitric oxide synthase and reduces insulin receptor content
and phosphorylation in human aortic endothelial cells. Metabolism 2002;51:487–91.
[58] Hotamisligil GS, Budavari A, Murray D, Spiegelman BM. Reduced tyrosine kinase
activity of the insulin receptor in obesity-diabetes. Central role of tumor necrosis factor-
alpha. J Clin Invest 1994;94:1543–9.
928 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

[59] Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM. IRS-1-
mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alpha- and
obesity-induced insulin resistance. Science 1996;271:665–8.
[60] Dandona P, Weinstock R, Thusu K, Abdel-Rahman E, Aljada A, Wadden T. Tumor
necrosis factor-alpha in sera of obese patients: fall with weight loss. J Clin Endocrinol
Metab 1998;83:2907–10.
[61] Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis
factor-alpha: direct role in obesity-linked insulin resistance. Science 1993;259:87–91.
[62] Pickup JC, Mattock MB, Chusney GD, Burt D. NIDDM as a disease of the innate
immune system: association of acute- phase reactants and interleukin-6 with metabolic
syndrome X. Diabetologia 1997;40:1286–92.
[63] Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, interleukin
6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327–34.
[64] Colwell JA, Halushka PV, Sarji K, Levine J, Sagel J, Nair RM. Altered platelet function
in diabetes mellitus. Diabetes 1976;25:826–31.
[65] Colwell JA, Nair RM, Halushka PV, Rogers C, Whetsell A, Sagel J. Platelet adhesion and
aggregation in diabetes mellitus. Metabolism 1979;28:394–400.
[66] Colwell JA, Halushka PV. Platelets, prostaglandins, and coagulation in diabetes mellitus.
Mt Sinai J Med 1982;49:215–22.
[67] Schonbeck U, Libby P. CD40 signaling and plaque instability. Circ Res 2001;89:
1092–103.
[68] Hakkinen T, Karkola K, Yla-Herttuala S. Macrophages, smooth muscle cells,
endothelial cells, and T-cells express CD40 and CD40L in fatty streaks and more
advanced human atherosclerotic lesions. Colocalization with epitopes of oxidized low-
density lipoprotein, scavenger receptor, and CD16 (Fc gammaRIII). Virchows Arch
2000;437:396–405.
[69] Melter M, Reinders ME, Sho M, Pal S, Geehan C, Denton MD, et al. Ligation of CD40
induces the expression of vascular endothelial growth factor by endothelial cells and
monocytes and promotes angiogenesis in vivo. Blood 2000;96:3801–8.
[70] Yarwood H, Mason JC, Mahiouz D, Sugars K, Haskard DO. Resting and activated T
cells induce expression of E-selectin and VCAM-1 by vascular endothelial cells through
a contact-dependent but CD40 ligand-independent mechanism. J Leukoc Biol 2000;68:
233–42.
[71] Thienel U, Loike J, Yellin MJ. CD154 (CD40L) induces human endothelial cell
chemokine production and migration of leukocyte subsets. Cell Immunol 1999;198:87–95.
[72] Arauz-Pacheco C, Parrott MA, Raskin P. The treatment of hypertension in adult patients
with diabetes. Diabetes Care 2002;25:134–47.
[73] Nishimura R, LaPorte RE, Dorman JS, Tajima N, Becker D, Orchard TJ. Mortality
trends in type 1 diabetes. The Allegheny County (Pennsylvania) Registry 1965–1999.
Diabetes Care 2001;24:823–7.
[74] Wingard D, Barnett Connor E. Diabetes in america. Group NDD, editors. Bethesda,
MD: National Institute of Diabetes and Digestive Kidney Disease; 1995. p. 429–448.
[75] Nelson R, Knowler W, Pettit D, Bennett P. Kidney Disease in Diabetes, Diabetes in
america, National Institutes of Health Publication No. 95–1468. ed 2. Bethesda, MD:
National Institutes of Health; 1995. p. 349–400.
[76] MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, et al. Blood pressure,
stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure:
prospective observational studies corrected for the regression dilution bias. Lancet 1990;
335:765–74.
[77] Janka HU, Warram JH, Rand LI, Krolewski AS. Risk factors for progression of
background retinopathy in long-standing IDDM. Diabetes 1989;38:460–4.
[78] Curb JD, Pressel SL, Cutler JA, Savage PJ, Applegate WB, Black H, et al. Effect of
diuretic-based antihypertensive treatment on cardiovascular disease risk in older diabetic
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 929

patients with isolated systolic hypertension. Systolic Hypertension in the Elderly Program
Cooperative Research Group. JAMA 1996;276:1886–92.
[79] Tuomilehto J, Rastenyte D, Birkenhager WH, Thijs L, Antikainen R, Bulpitt CJ, et al.
Effects of calcium-channel blockade in older patients with diabetes and systolic
hypertension. Systolic Hypertension in Europe Trial Investigators. N Engl J Med 1999;
340:677–84.
[80] Effects of ramipril on cardiovascular and microvascular outcomes in people with diabetes
mellitus: results of the HOPE study and MICRO-HOPE substudy. Heart Outcomes
Prevention Evaluation Study Investigators. Lancet 2000;355:253–9.
[81] Lonn E, Yusuf S, Dzavik V, Doris C, Yi Q, Smith S, et al. Effects of ramipril and vitamin
E on atherosclerosis: the study to evaluate carotid ultrasound changes in patients treated
with ramipril and vitamin E (SECURE). Circulation 2001;103:919–25.
[82] Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects
of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and
nephropathy. N Engl J Med 2001;345:861–9.
[83] Parving HH, Lehnert H, Brochner-Mortensen J, Gomis R, Andersen S, Arner P. The
effect of irbesartan on the development of diabetic nephropathy in patients with type 2
diabetes. N Engl J Med 2001;345:870–8.
[84] Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, Julius S, et al. Effects of
intensive blood-pressure lowering and low-dose aspirin in patients with hypertension:
principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT
Study Group. Lancet 1998;351:1755–62.
[85] Tight blood pressure control and risk of macrovascular and microvascular complications in
type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group. BMJ 1998;317:703–13.
[86] Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, Cull CA, et al. Association
of systolic blood pressure with macrovascular and microvascular complications of type 2
diabetes (UKPDS 36): prospective observational study. BMJ 2000;321:412–9.
[87] Snow V, Weiss KB, Mottur-Pilson C. The evidence base for tight blood pressure control
in the management of type 2 diabetes mellitus. Ann Intern Med 2003;138:587–92.
[88] Estacio RO, Jeffers BW, Hiatt WR, Biggerstaff SL, Gifford N, Schrier RW. The effect of
nisoldipine as compared with enalapril on cardiovascular outcomes in patients with non-
insulin-dependent diabetes and hypertension. N Engl J Med 1998;338:645–52.
[89] Peterson JC, Adler S, Burkart JM, Greene T, Hebert LA, Hunsicker LG, et al. Blood
pressure control, proteinuria, and the progression of renal disease. The Modification of
Diet in Renal Disease Study. Ann Intern Med 1995;123:754–62.
[90] Klag MJ, Whelton PK, Randall BL, Neaton JD, Brancati FL, Ford CE, et al. Blood
pressure and end-stage renal disease in men. N Engl J Med 1996;334:13–8.
[91] Klag MJ, Whelton PK, Randall BL, Neaton JD, Brancati FL, Stamler J. End-stage renal
disease in African-American and white men. 16-year MRFIT findings. JAMA 1997;277:
1293–8.
[92] Bakris GL, Weir MR, Shanifar S, Zhang Z, Douglas J, van Dijk DJ, et al. Effects of
blood pressure level on progression of diabetic nephropathy: results from the RENAAL
study. Arch Intern Med 2003;163:1555–65.
[93] Jafar T, Stark P, Scmid C, Landa M, Maschio G, de Jong P, et al. Progression of Chronic
Kidney Disease: The Role of Blood Pressure Control, proteinuria, and Angiotensin-
Converting enzyme Inhibition. Ann Intern Med 2003;139:244–52.
[94] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The
Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation,
and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289:2560–72.
[95] Arauz-Pacheco C, Parrott MA, Raskin P. Treatment of hypertension in adults with
diabetes. Diabetes Care 2003;26(Suppl 1):S80–2.
[96] Bakris GL, Williams M, Dworkin L, Elliott WJ, Epstein M, Toto R, et al. Preserving
renal function in adults with hypertension and diabetes: a consensus approach. National
930 P. Dandona et al / Med Clin N Am 88 (2004) 911–931

Kidney Foundation Hypertension and Diabetes Executive Committees Working Group.


Am J Kidney Dis 2000;36:646–61.
[97] Tatti P, Pahor M, Byington RP, Di Mauro P, Guarisco R, Strollo G, et al. Outcome
results of the Fosinopril Versus Amlodipine Cardiovascular Events Randomized Trial
(FACET) in patients with hypertension and NIDDM. Diabetes Care 1998;21:597–603.
[98] Lindholm LH, Hansson L, Ekbom T, Dahlof B, Lanke J, Linjer E, et al. Comparison of
antihypertensive treatments in preventing cardiovascular events in elderly diabetic
patients: results from the Swedish Trial in Old Patients with Hypertension-2. STOP
Hypertension-2 Study Group. J Hypertens 2000;18:1671–5.
[99] Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting
enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-
Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002;288:2981–97.
[100] Wing LM, Reid CM, Ryan P, Beilin LJ, Brown MA, Jennings GL, et al. A comparison of
outcomes with angiotensin-converting–enzyme inhibitors and diuretics for hypertension
in the elderly. N Engl J Med 2003;348:583–92.
[101] Efficacy of atenolol and captopril in reducing risk of macrovascular and microvascular
complications in type 2 diabetes: UKPDS 39. UK Prospective Diabetes Study Group.
BMJ 1998;317:713–20.
[102] Nielsen FS, Rossing P, Gall MA, Skott P, Smidt UM, Parving HH. Long-term effect of
lisinopril and atenolol on kidney function in hypertensive NIDDM subjects with diabetic
nephropathy. Diabetes 1997;46:1182–8.
[103] Brown MJ, Palmer CR, Castaigne A, de Leeuw PW, Mancia G, Rosenthal T, et al.
Morbidity and mortality in patients randomised to double-blind treatment with a long-
acting calcium-channel blocker or diuretic in the International Nifedipine GITS study:
Intervention as a Goal in Hypertension Treatment (INSIGHT). Lancet 2000;356:
366–72.
[104] Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, et al. Renoprotective
effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due
to type 2 diabetes. N Engl J Med 2001;345:851–60.
[105] Lindholm LH, Ibsen H, Dahlof B, Devereux RB, Beevers G, de Faire U, et al.
Cardiovascular morbidity and mortality in patients with diabetes in the Losartan
Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial
against atenolol. Lancet 2002;359:1004–10.
[106] Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of angiotensin-converting-
enzyme inhibition on diabetic nephropathy. The Collaborative Study Group. N Engl J
Med 1993;329:1456–62.
[107] Should all patients with type 1 diabetes mellitus and microalbuminuria receive
angiotensin-converting enzyme inhibitors? A meta-analysis of individual patient data.
Ann Intern Med 2001;134:370–9.
[108] Perkins BA, Ficociello LH, Silva KH, Finkelstein DM, Warram JH, Krolewski AS.
Regression of microalbuminuria in type 1 diabetes. N Engl J Med 2003;348:2285–93.
[109] Chaturvedi N, Sjolie AK, Stephenson JM, Abrahamian H, Keipes M, Castellarin A, et al.
Effect of lisinopril on progression of retinopathy in normotensive people with type 1
diabetes. The EUCLID Study Group. EURODIAB Controlled Trial of Lisinopril in
Insulin-Dependent Diabetes Mellitus. Lancet 1998;351:28–31.
[110] Malik RA, Williamson S, Abbott C, Carrington AL, Iqbal J, Schady W, et al. Effect of
angiotensin-converting-enzyme (ACE) inhibitor trandolapril on human diabetic neurop-
athy: randomised double-blind controlled trial. Lancet 1998;352:1978–81.
[111] Bakris GL, Copley JB, Vicknair N, Sadler R, Leurgans S. Calcium channel blockers
versus other antihypertensive therapies on progression of NIDDM associated nephrop-
athy. Kidney Int 1996;50:1641–50.
[112] Sato A, Hayashi K, Naruse M, Saruta T. Effectiveness of aldosterone blockade in
patients with diabetic nephropathy. Hypertension 2003;41:64–8.
P. Dandona et al / Med Clin N Am 88 (2004) 911–931 931

[113] Rossing K, Jacobsen P, Pietraszek L, Parving HH. Renoprotective effects of adding


angiotensin II receptor blocker to maximal recommended doses of ACE inhibitor in diabetic
nephropathy: a randomized double-blind crossover trial. Diabetes Care 2003;26:2268–74.
[114] Parulkar AA, Pendergrass ML, Granda-Ayala R, Lee TR, Fonseca VA. Nonhypogly-
cemic effects of thiazolidinediones. Ann Intern Med 2001;134:61–71.
[115] Klein L, O’Connor CM, Gattis WA, Zampino M, de Luca L, Vitarelli A, et al.
Pharmacologic therapy for patients with chronic heart failure and reduced systolic
function: review of trials and practical considerations. Am J Cardiol 2003;91:18F–40F.
[116] Bakris GL, Weir MR. Angiotensin-converting enzyme inhibitor-associated elevations in
serum creatinine: is this a cause for concern? Arch Intern Med 2000;160:685–93.
[117] Mann JF, Gerstein HC, Pogue J, Bosch J, Yusuf S. Renal insufficiency as a predictor of
cardiovascular outcomes and the impact of ramipril: the HOPE randomized trial. Ann
Intern Med 2001;134:629–36.
[118] Goldstein S. Beta-blockers in hypertensive and coronary heart disease. Arch Intern Med
1996;156:1267–76.
[119] Gaede P, Vedel P, Larsen N, Jensen G, Parving H-H, Pedersen O. Multifactorial
Intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med
2003;348:383–93.
[120] Dandona P, Kumar V, Aljada A, Ghanim H, Syed T, Hofmayer D, et al. Angiotensin II
Receptor Blocker Valsartan Suppresses Reactive Oxygen Species Generation in
Leukocytes, Nuclear Factor-kappaB, in Mononuclear Cells of Normal Subjects: Evidence
of an Antiinflammatory Action. J Clin Endocrinol Metab 2003;88:4496–501.
[121] Effect of intensive blood-glucose control with metformin on complications in overweight
patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS)
Group. Lancet 1998;352:854–65.
[122] Ghanim H, Garg R, Aljada A, Mohanty P, Kumbkarni Y, Assian E, et al. Suppression of
Nuclear Factor-kappaB and Stimulation of Inhibitor kappaB by Troglitazone: Evidence
for an Anti-inflammatory Effect and a Potential Antiatherosclerotic Effect in the Obese.
J Clin Endocrinol Metab 2001;86:1306–12.
[123] Mohanty P, Aljada A, Ghanim H, Tripathy D, Syed T, Hofmeyer D, et al. Rosiglitazone
improves vascular reactivity, inhibits reactive oxygen species (ROS) generation, reduces
p47phox subunit expression in mononuclear cells (MNC) and reduces C reactive protein
(CRP) and monocyte chemotactic protein-1 (MCP-1): evidence of a potent anti-
inflammatory effect. Diabetes 2001;50(Suppl 2):A68.
[124] Trovati M, Anfossi G, Cavalot F, Massucco P, Mularoni E, Emanuelli G. Insulin directly
reduces platelet sensitivity to aggregating agents. Studies in vitro and in vivo. Diabetes
1988;37:780–6.
[125] Hiramatsu K, Nozaki H, Arimori S. Reduction of platelet aggregation induced by
euglycaemic insulin clamp. Diabetologia 1987;30:310–3.
[126] Dandona P, Aljada A, Mohanty P, Ghanim H, Hamouda W, Assian E, et al. Insulin
Inhibits Intranuclear Nuclear Factor kappaB and Stimulates IkappaB in Mononuclear
Cells in Obese Subjects: Evidence for an Anti- inflammatory Effect? J Clin Endocrinol
Metab 2001;86:3257–65.
[127] Vehkavaara S, Makimattila S, Schlenzka A, Vakkilainen J, Westerbacka J, Yki-Jarvinen
H. Insulin therapy improves endothelial function in type 2 diabetes. Arterioscler Thromb
Vasc Biol 2000;20:545–50.
Med Clin N Am 88 (2004) 933–945

Gonadal and erectile dysfunction


in diabetics
Rabih A. Hijazi, MDa,b,
Marion Betancourt-Albrecht, MDa,
Glenn R. Cunningham, MDa,b,*
a
Department of Medicine, Veterans Affairs Medical Center, Houston, TX 77030, USA
b
Departments of Medicine, Molecular, and Cellular Biology, Baylor College of Medicine,
Houston, TX 77030, USA

The high prevalence of erectile dysfunction (ED) in patients with diabetes


is caused mainly by vascular and neurologic conditions; nevertheless,
hypogonadism may also contribute to ED, as well as to changes in mood,
libido, body composition, and bone density.

Hypogonadism and diabetes


Aging and obesity are both associated with diabetes and with hypo-
gonadism, so in an aging and obese population it can be difficult to
determine if there is a direct or an indirect relationship between diabetes and
hypogonadism. Cross-sectional and epidemiologic studies indicate a cause-
effect relationship between obesity and type 2 diabetes. Although the
prevalence of type 2 diabetes increases with age, fat mass also increases
[1–3]. Ohlson et al [4] studied 792 Scandinavian patients prospectively for
13.5 years. They found that waist circumference and degree of obesity are
independent risk factors for the development of diabetes. Serum total
testosterone levels are greatly affected by changes in sex hormone bind-
ing globulin (SHBG); according to study findings, approximately 40% of
circulating testosterone in men is bound to SHBG. Moderate obesity can
reduce SHBG levels and total testosterone without causing a deficiency in
bioavailable or free testosterone levels. Thus, measurement of only total
testosterone levels can result in an unwarranted diagnosis of hypogonadism.

* Corresponding author. Research Service 151 VA Medical Center 2002 Holcombe


Boulevard, Houston, TX 77030.
E-mail address: glennc@bcm.tmc.edu (G.R. Cunningham).

0025-7125/04/$ - see front matter. Published by Elsevier Inc.


doi:10.1016/j.mcna.2004.04.007
934 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

In contrast, aging is associated with an increase in SHBG levels.


Measurement of only total testosterone levels in a normal weight, aging
male can give a false impression of eugonadism, yet bioavailable or free
testosterone levels may be low.
Age, obesity, and the assay used to measure testosterone will affect the
diagnosis of hypogonadism. Clearly, aging and obesity are important in the
development of diabetes and of hypogonadism. Our analysis will focus on
the interaction of these conditions and attempt to explain possible
mechanisms for observations reported in the literature.

Hypogonadism and aging


Multiple cross-sectional and longitudinal studies have shown that serum
testosterone levels in men decrease with age [5–10]. Tenover et al [11] studied
older men and noted that pituitary secretion of leuteinizing hormone (LH)
was intact, but testicular secretion of testosterone was impaired in some
older men; however, the majority of hypogonadal men over the age of 60
have low or inappropriately normal LH levels [12]. Luboshitzky et al [13]
found less pulsatile testosterone and more LH were secreted in middle-aged
men at night compared with young men. The association between
testosterone rhythm and the rapid-eye movement level of sleep was
disrupted. They suggest that the decline in nocturnal testosterone is a
possible result of combined testicular and pituitary hypogonadism.
Several cross-sectional studies have shown an increase of SHBG with
aging as well as a decrease in testosterone and free testosterone levels that is
independent of body mass index (BMI) [5–7]; and longitudinal studies have
confirmed these findings [8–10]. Feldman et al [10] found that total
testosterone levels decreased at a rate of 0.8% per year whereas testosterone
bound to albumin and free testosterone (bioavailable testosterone) fell at
a rate of 2% per year, and SHBG levels increased at 1.6% per year.

Hypogonadism and type 2 diabetes


Low total testosterone levels predict the development of type 2 diabetes.
In 1996, Haffner et al [14] reported that low levels of SHBG and total
testosterone might be a risk factor for developing type 2 diabetes in men. Oh
et al [15] conducted a prospective study that included 294 men aged 55 to 89
years. Patients were evaluated with a 75-g glucose tolerance test, and fasting
and 2-hour glucose and insulin levels were assessed. Low levels of total
testosterone in men predicted the development of type 2 diabetes within 8
years. Subjects with low total testosterone had a 2.7-fold increase risk for
diabetes after correcting for age, BMI, and systolic blood pressure (Fig. 1).
The Massachusetts Male Aging Study evaluated 1709 men from 1987 to
1989 and re-evaluated them 7–10 years later [16]. Each 1 standard-deviation
fall in free testosterone (4 ng/mL) increased the risk by 1.58-fold (P  0.02),
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 935

OR (95% Cl)

Total 2.7 (1.1 - 6.6)*


testosterone
1.6 (0.6 - 4.8)

Bioavailable 0.8 (0.3 - 2.2)

testosterone 2.9 (1.1 - 8.4)*

Total 1.8 (0.8 - 4.3)

estradiol 1.3 (0.4 - 4.6)

Bioavailable 0.7 (0.3 - 1.8)


estradiol
1.9 (0.7 - 5.3)

0.0 1.0 5.0 10.0

Fig. 1. Association of endogenous sex hormones with incident diabetes by multiple logistic
regression, adjusted for baseline age, BMI, and systolic blood pressure among older men (solid
box) and women (open box) without diabetes. Sex hormone differences were analyzed as the
lowest quartile versus the top three quartiles in men and the highest quartile versus the bottom
three quartiles in women (*P  0.05). (From Oh JY, Barrett-Connor E, Wedick NM, Wingard
DL; Rancho Bernardo Study. Endogenous sex hormones and the development of type 2
diabetes in older men and women: the Rancho Bernardo Study. Diabetes Care 2002;25:55–60;
with permission.)

and each 1 standard-deviation fall in SHBG (16 nmol/L) increased the risk
by 1.89-fold (P  0.02). The latter study, however, did not have baseline
glucose and insulin levels but relied on self-reporting of diabetes.
Low testosterone levels are common in type 2 diabetics, but this could be
because of aging or caused by the high prevalence of obesity in this
population. The Rancho Bernardo study evaluated 985 men, aged 40 to 79
years, between 1972 and1974 [17]. The 110 diabetic men had significantly
lower levels of testosterone and SHBG compared with 875 nondiabetic men.
A significant difference remained after adjustment for age (Table 1).
Twenty-one percent of diabetic men compared with 13% of nondiabetic
men had testosterone levels below 350 ng/mL. The same investigator
identified 44 men with untreated type 2 diabetes mellitus over a 15-month
period. Thirty-two of these men were identified with a 75-g oral glucose
tolerance test [18]. The diabetics were compared with 44 age- and BMI-
matched men. Total testosterone levels were lower in the diabetics (P  0.05)
when compared with each of the control groups. Bioavailable testosterone
levels were lower than controls that were not matched for BMI but were
936 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

Table 1
Crude, age-adjusted, and age- and BMI-adjusted mean testosterone and SHBG levels by
diabetic status
Nondiabetic Diabetic
(n = 875) (n = 110) P value
Testosterone (ng/mL) 5.41 4.97
Age-adjusted 5.45 4.85 0.01
Age- and BMI-adjusted 5.35 5.10 0.08
Percentage with 21% 13% 0.05
testosterone 3.5 ng/mL
SHBG (10ÿ8 M) 329 285
Age-adjusted 327 282 0.01
Age- and BMI-adjusted 326 291 0.05
Data from Barrett-Connor E, Khaw KT, Yen SS. Endogenous sex hormone levels in older
adult men with non-insulin-dependent diabetes mellitus. Ann Intern Med 1992;117:807–11.

similar to levels in BMI-matched controls. Tan and Pu [19] evaluated


a geriatric population (mean age 73) and found that 64% of diabetic men
compared with 38% of nondiabetics had total testosterone levels below
300 ng/mL. In another study, the Rancho Bernardo group evaluated 775 men
over the age of 55 using a 75-g oral glucose tolerance test [20]. Subjects were
categorized as normal, impaired fasting glucose, impaired glucose tolerance,
and type 2 diabetes based on World Health Organization criteria. Fourteen
percent had type 2 diabetes. These patients were older and had somewhat
greater waist–hip ratios, even though their BMIs were similar to men with
a normal glucose tolerance test. Total and bioavailable testosterone were
inversely associated with BMI and waist–hip ratios. Only bioavailable
testosterone was inversely related to plasma glucose levels.

SHBG
SHBG levels are affected by several conditions. There is a strong
association between increased total body fat, subcutaneous and visceral
adiposity, and decreased plasma levels of SHBG [21,22]. SHBG levels were
similar in diabetic and nondiabetic men with similar fat content, but SHBG
levels vary inversely with hyperinsulinism in nondiabetic subjects [23].
SHBG levels seem to be an indicator of general adiposity rather than an
index of altered insulin-glucose homeostasis in morbidly obese subjects.
Hyperinsulinism also decreases SHBG synthesis as found by testing cultured
hepatic cells [24]. These observations have been interpreted to show that
obesity causes insulin resistance and hyperinsulinism, and hyperinsulinism
decreases SHBG levels. Hypothyroidism and the nephrotic syndrome also
reduce SHBG levels. In addition to aging, factors that are known to increase
SHBG include estrogen [25], hyperthyroidism [26], some anticonvulsants
[27], a high phytoestrogen diet [28], and hepatic cirrhosis [29].
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 937

Testosterone and obesity


There is a highly negative correlation between plasma-free testosterone
and BMI and a positive correlation between estradiol and BMI in obese and
men of normal weight [30]. Gapstur et al [31] studied 483 black and 695
white males and observed an age-associated decrease in circulating
testosterone and an increase in SHBG beginning in the third decade of life.
The increase in adiposity, especially central obesity, was associated with
a decrease in total testosterone and SHBG levels. Phillips et al [23] measured
and determined the interrelationships of sex hormones, insulin, and
adiposity in 80 apparently healthy men. Testosterone levels and the serum
estradiol-to-testosterone ratio correlated with insulin level. After controlling
for visceral adiposity, only the estradiol-to-testosterone ratio remained
significant. On multiple regression analysis, insulin was was found to be
significantly associated with estradiol and testosterone levels, and these
correlations were independent of visceral adiposity and age.
Weight loss increases testosterone levels. In a study of 11 obese men, weight
loss increased total testosterone, free testosterone, and SHBG levels without
affecting estradiol, estrone, or LH [32]. Dexamethasone suppressed estradiol,
estrone, and SHBG, but it had no effect on LH, total, and free testosterone in
the obese men [33]. This indicates that estrone and estradiol levels are
dependent on adrenal steroids that provide the substrates for aromatase
conversion to estrone and estradiol. In another study [34], obese men were
treated for 6 months with a hypocaloric diet and dexfenfluramine. Weight and
fasting insulin levels were reduced whereas total and free testosterone and LH
concentrations were increased. Thus, significant weight loss decreases insulin
and increases SHBG total and free testosterone concentrations.

Other hormones
Leptin is a peripheral hormone that has a role in energy balance and
energy stores. In animal models it has been shown to reflect total adipose
mass [35]. Leptin levels are directly related to total body fat but are highly
variable among individuals [36]. Inter-individual and age differences in
serum leptin levels may be related more to testosterone levels than to BMI
[37]. Leptin levels increase with age regardless of changes in body fat,
possibly because of the fall in testosterone levels. The negative correlation
between leptin and testosterone in men is consistent with the hypothesis that
testosterone may inhibit adipocyte ob gene transcription.
Adiponectin is an adipose-specific secretory protein, with antidiabetic
and antiatherogenic properties. Androgens appear to decrease plasma
adiponectin [38]. The hypoadiponectinemia induced by androgens may
increase insulin resistance and atherosclerosis in men. In mice, castration
but not ovariectomy increases adiponectin levels, and treatment with
testosterone reduced adiponectin levels in the castrated mice.
938 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

Erectile dysfunction in diabetics


Erectile dysfunction is the inability to satisfactorily achieve or maintain
an erection for sexual activity. It is a common complaint in adult men, and
its prevalence increases with age. Men with diabetes are particularly afflicted
with erectile dysfunction. Although numerous studies have been conducted
in affected men, and diabetic animal models have been developed to explore
pathologic mechanisms of ED, our understanding remains fragmented. We
also have witnessed radical change in the clinical management of ED with
the introduction of sildenafil (Viagra) in 1998. A phosphodiesterase (PDE)
type 5 inhibitor frequently is used as a first-line empirical therapy, with little
regard to cause or diagnosis.

Epidemiology
It is estimated that 40% to 60% of men with diabetes have erectile
dysfunction. Typically, ED affects middle-aged or older diabetic males. In
one survey [39] of 541 diabetic men aged 20 to 59 years, the prevalence of
ED was 6% for diabetic men age 20 to 24 years and 52% for those 55 to 59
years old. The prevalence of ED was evaluated in 365 men, age 21 years or
older, with a minimum duration of diabetes of 10 years, and whose diabetes
was diagnosed before age 30 [40]. Most of these patients had type 1 diabetes.
The prevalence was 1.1% in the 21 to 30-year-old group and 47.1% in those
43 years old or older (Fig. 2). In addition to increasing age, ED was
associated with smoking, obesity, poor glycemic control, low HDL,
neuropathy, and retinopathy.

100
Prevalence of Erectile Dysfunction (%)

75

50

25

0
21-30 >30-40 >40-50 >50-60 >60
Age (yr)

Fig. 2. Relationship of age to the prevalence of erectile dysfunction in the Wisconsin


Epidemiologic Study of Diabetic Retinopathy (1990–1992). (From Klein R, Klein BEK, Lee
KE, Moss SE, Cruikshanke KJ. Prevalence of self-reported erectile dysfunction in people with
long-term IDDM. Diabetes Care 1996;19:135–41; with permission.)
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 939

Causes of ED
Normal erections are mediated by an increase in parasympathetic tone
and a reduction in sympathetic tone [41,42]. These neural pathways are
modulated by visual, auditory, olfactory, tactile, and psychologic stimuli.
Nonadrenergic, noncholinergic nerve endings release nitric oxide, which
stimulates cyclic guanosine monophosphate (cGMP) in the smooth muscle
of arterioles supplying the corpora cavernosa and of smooth muscle cells
lining the sinusoids of the corpora cavernosa. cGMP stimulates processes
that cause intracellular calcium levels to fall, resulting in smooth muscle
relaxation. The combination of increased blood flow to and the engorge-
ment of the corpora cavernosa causes compression of the veins that drain
the corpora cavernosa against the inelastic tunica albuginea. This essentially
prevents venous drainage of the corpora and causes penile rigidity.
ED in diabetics often is multifactorial. It can result from nerve damage
(eg, neuropathy), impaired blood flow (eg, vasculopathy), or psychological
factors. Some factors that can affect cavernosal smooth muscle relaxation
have been assessed in animal models. For instance, streptozotocin-induced
diabetic rats have increased penile cGMP and cyclic adenosine mono-
phosphate (cAMP) levels [43]. This suggests that defective relaxation or
increased contraction of cavernosal smooth muscle is the cause of impaired
erections in this model. Although an increase in connective tissue in the
corpora cavernosa has been observed in diabetic men with ED, many
diabetic men respond to PDE5 inhibitor, which acts by increasing cGMP.
Other diabetic men respond to intracavernosal injections of prostaglandin
(PG)E1, which increases cAMP in the corpora cavernosa and causes penile
erection. These observations indicate that the smooth muscle is able to relax
and fibrosis is not the limiting condition, at least in the earlier stages of ED.

Evaluation
Many patients with diabetes will not volunteer information about their
sexual function. Thus, it is very important for the clinician to inquire about
ED in all diabetic patients. If detected, one should not assume that it is only
caused by neuropathy or vasculopathy.
The provider should conduct a comprehensive history and physical
examination, trying to identify other possible causes of ED, such as med-
ications, substance abuse, alcohol, hypogonadism, thyroid dysfunction, pel-
vic trauma, Peyronie’s disease, or psychogenic causes. Laboratory workup
should include a complete blood count and screening chemistries to rule out
systemic disease, lipid profile, glycosylated hemoglobin, total testosterone,
LH, and a thyroid-stimulating hormone level. If the total testosterone is less
than 300 ng/dL, a second specimen should be obtained between 7 and 10 AM,
and total testosterone LH and prolactin should be determined. Measure-
ment of bioavailable or free testosterone by equilibrium dialysis should be
considered for borderline values or when there is reason to believe that
940 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

SHBG levels may be altered. Referral to an endocrinologist is appropriate in


a setting of low or normal LH with a level of morning total testosterone of
less than 200 ng/dL, hyperprolactinemia, or abnormal thyroid function
tests. Depression is particularly common in diabetics, and it can be an
important factor in the development of ED. It is important to differentiate
psychogenic from organic ED because therapeutic interventions clearly
differ. In most cases, however, it is less important to differentiate between
specific organic causes (eg, neurogenic versus vasculogenic) because treat-
ments usually are similar. In summary, one should seek to identify
conditions that may respond to specific therapy such as medication-induced
ED, hypogonadism, pelvic trauma, Peyronie’s disease, or psychogenic ED.
In the absence of these conditions, it is reasonable to consider empirical
treatment. More elaborate testing, such as a duplex Doppler study, to assess
the possibility of arterial insufficiency or a venous leak or sleep studies to
assess possible psychogenic ED are expensive and usually not warranted in
the initial evaluation. They should be considered for patients who fail to
respond to noninvasive or minimally invasive therapies.

Treatment
There are numerous modalities of therapy for ED, none of which is
specific for diabetes. To our knowledge, there are no studies that
demonstrate tight glycemic control improves ED in diabetics. In fact,
a Veterans Affairs cooperative study [44] involving 153 men who received
standard versus intensive treatment of their diabetes actually found that
erectile dysfunction increased from 53% to 73% in the standard treatment
group and from 51% to 73% in the intensive treatment group, over 2 years.
There are studies, however, showing that tight control can improve
endothelial function, and in the early stages, endothelial dysfunction may
be the cause of ED. In view of the other potential benefits, good glycemic
control is imperative.
Oral PDE5 inhibitors are the most prescribed drugs used to treat ED. These
agents inhibit PDE5, which metabolizes cGMP to inactive 59-GMP. cGMP
causes a reduction in intracellular calcium levels and relaxation of the smooth
muscle in the arterioles supplying the corpora cavernosa and the trabecular
smooth muscle lining the sinusoids of the corpora cavernosa. Sildenafil, the
first PDE5 inhibitor to receive approval for clinical use, inhibits degradation
of cGMP, thereby facilitating and prolonging penile erections. Unlike PGE1,
it does not cause penile erection; thus, sexual stimulation is necessary. In
a double blind, flexible dose, placebo-controlled trial involving 268 diabetics
men with ED, 56% of the men taking sildenafil reported improved erections
compared with 10% in the placebo arm [45]. Head-to-head comparisons in
this population are limited at this time; however, randomized, placebo-
controlled studies have been conducted in diabetics in different patient
populations with three PDE5 inhibitors (Table 2) [46,47].
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 941

Table 2
PDE5 inhibitors efficacy in diabetes based on the responses to ‘‘Has treatment improved your
erections after 12 weeks?’’
Answering ‘‘yes’’ Answering ‘‘yes’’
Drug [Ref.] Dose placebo (%) therapy (%)
Sildenafil [43] 100 mg 10 (n = 121) 56 (n = 131)
Vardenafil [44] 10 mg 13 (n = 133) 57 (n = 137)
Vardenafil [44] 20 mg 13 (n = 133) 72 (n = 131)
Tadalafil [45] 10 mg 25 (n = 71) 56 (n = 73)
Tadalafil [45] 20 mg 25 (n = 71) 64 (n = 72)
These studies were performed on different patient groups and are not head-to-head
comparisons.

Side effects of PDE5 include headaches, flushing, dizziness, dyspepsia,


rhinitis, and visual disturbances (Table 3). Men taking nitrates and a PDE5
inhibitor are at high risk of syncope and hypotension; therefore, the use of
nitrates is a contraindication to the use of a PDE5 inhibitor.
Another effective agent is PGE1 (alprostadil), usually used as a second-
line agent, stimulates adenyl cyclase and increases cAMP levels in
cavernosal smooth muscle. This in turn reduces intracellular calcium and
causes muscle relaxation. The US Food and Drug Administration approved
the intracavernosal administration of PGE1 for ED in 1996 [48] and
intraurethral administration in 1997 [49]. In a placebo-controlled study of
961 men (of whom 20% were diabetic), 65% treated with intraurethral
PGE1 successfully had intercourse versus 19% in the placebo group [49].
Lower success rates, however, have been noted during nonprotocol use of
intraurethral PGE1.
The intracavernosal administration route of PGE1 is more effective. In a
6-month study involving 683 men, intracavernosal PGE1 was prescribed
for men with psychogenic, neurogenic, vasculogenic, and combined causes
of ED. Participants reported sexual activity after 94% of injections. Both
partners rated sexual activity as satisfactory after 87% (malesÕ responses) and
86% (femalesÕ responses) of the injections [48]. Many if not most patients are
reluctant to self-administer alprostadil by injection into the corpora
cavernosa, but those who do find this an effective treatment for ED. Both

Table 3
Sildenafil: most common side effects
Placebo Sildenafil
Event (%) (n = 725) (%) (n = 734)
Flushing 1 10
Dyspepsia 2 7
Nasal congestion 2 4
Abnormal vision 0 3
Headache 4 16
Available at http://www.viagramd.wm/pi/proPackInsert.asp.
942 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

routes of PGE1 administration cause penile pain in approximately 10% of


patients. A small percentage of men develop fibrotic penile plaques, and
a very small percentage of men develop priapism. Combining a lower dose of
PGE1 with phentolamine and papaverine can further reduce the frequency of
priapism; however, these agents must be formulated by a pharmacist.
Many diabetic men with ED use vacuum constrictor devices. Several
devices have been developed that use vacuum pressure to promote arterial
inflow and occlusive rings to inhibit venous outflow from the penile corpora
cavernosa. A certain amount of dexterity is required to use these devices
effectively. Another problem is that men cannot ejaculate because the
occlusive ring that prevents venous drainage also compresses the penile
urethra and prevents ejaculation.
Vacuum devices have enjoyed some popularity in the US and abroad,
where reports of efficacy in creating erections have been reported to be as
high as 67%. Satisfaction with vacuum-assisted erections has varied between
25% and 49%. In a study [50] of 54 diabetic men with ED, 44 chose vacuum
therapy. After 2 months, 75% (33 men) were able to have satisfactory
intercourse using the vacuum constrictor device. Three other men achieved
satisfactory erections, but their partners found the device unacceptable.
There are subtle differences between the devices that make some easier to
use than others [51].
If the noninvasive or minimally invasive treatments are not effective,
a surgical prosthesis should be considered. Two types of prostheses are the
semi-rigid prosthetic rods and the hydraulic inflatable prostheses. Surgical
complications, which include erosion and infection, have been reported in
3% to 8% of men with the semi-rigid rods [52]. The inflatable prostheses are
more physiologically natural and better accepted, but these prostheses have
a small but significant incidence of mechanical failure [53].
Neither arterial nor venous vascular surgery has proven to be successful
as treatments for ED, except in younger men with a history of pelvic
trauma. Thus, vascular surgery is rarely performed for ED in diabetic men.
Although most men with diabetes have one or possibly more organic
causes of ED, psychological factors are also often present. Psychosexual
counseling appears to be a helpful adjunct to any therapy.

Summary
We now have a variety of treatments for the diabetic with ED. If
hypogonadism is present and there are no contraindications, testosterone
replacement therapy should be considered. In addition to the effects of
testosterone replacement therapy on libido and erections, testosterone
improves body composition, bone density, and mood. The hematocrit,
digital rectal examination, and prostate-specific antigen level, however, must
be monitored carefully in men 60 and older and in men 50 and older if they
are African American or have a family history of prostate cancer because
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 943

testosterone replacement can cause excessive erythrocytosis and may cause


a prostate cancer to become clinically detectable. Other specific therapies for
ED should be offered when appropriate. Most diabetic men will achieve
erections with a PDE5 inhibitor, intracavernosal administration of PGE1, or
a vacuum constrictor device. When none of these therapies is satisfactory,
patients should be considered for a surgical prosthesis.

References
[1] Data and trends: diabetes surveillence system: prevalence of diagnosed diabetes by age,
United States, 1980–2000. Available at: www.cdc.gov/diabetes/statistics/prev/national/
fig3.htm. Accessed May 18, 2004.
[2] Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing
epidemics of obesity and diabetes in the United States. JAMA 2001;286:1195–200.
[3] Gallagher D, Ruts E, Visser M, Heshka S, Baumgartner RN, Wang J, et al. Weight
stability masks sarcopenia in elderly men and women. Am J Physiol 2000;279:E366–75.
[4] Ohlson L, Larsson B, Svardsudd K, Welin L, Eriksson H, Wilhelmsen L, et al. The
influence of body fat distribution on the incidence of diabetes mellitus. 13.5 years of follow-
up of the participants in the study of men born in 1913. Diabetes 1985;34:1055–8.
[5] Gray A, Feldman HA, McKinlay JB, Longcope C. Age, disease, and changing sex
hormone levels in middle-aged men: results of the Massachusetts Male Aging Study. J Clin
Endocrinol Metab 1991;73:1016–25.
[6] Deslypere JP, Vermeulen A. Leydig cell function in normal men: effect of age, life-style,
residence, diet, and activity. J Clin Endocrinol Metab 1984;59:955–62.
[7] Morley JE, Charlton E, Patrick P, Kaiser FE, Cadeau P, McCready D, et al. Validation of
a screening questionnaire for androgen deficiency in aging males. Metabolism 2000;49:
1239–42.
[8] Morley JE, Kaiser FE, Perry HM III, Patrick P, Morley PM, Stauber PM, et al.
Longitudinal changes in testosterone, luteinizing hormone, and follicle-stimulating
hormone in healthy older men. Metabolism 1997;46:410–3.
[9] Harman SM, Metter EJ, Tobin JD, Pearson J, Blackman MR. Longitudinal effects of
aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal
Study of Aging. J Clin Endocrinol Metab 2001;86:724–31.
[10] Feldman HA, Longcope C, Derby CA, Johannes CB, Araujo AB, Coviello AD, et al. Age
trends in the level of serum testosterone and other hormones in middle-aged men:
longitudinal results from the Massachusetts Male Aging Study. J Clin Endocrinol Metab
2002;87:589–98.
[11] Tenover JS, Matsumoto AM, Plymate SR, Bremner WJ. The effects of aging in normal
men on bioavailable testosterone and luteinizing hormone secretion: response to
clomiphene citrate. J Clin Endocrinol Metab 1987;65:1118–26.
[12] Korenman SG, Morley JE, Mooradian AD, Davis SS, Kaiser FE, Silver AJ, et al.
Secondary hypogonadism in older men: its relation to impotence. J Clin Endocrinol Metab
1990;71:963–9.
[13] Luboshitzky R, Shen-Orr Z, Herer P. Middle-aged men secrete less testosterone at night
than young healthy men. J Clin Endocrinol Metab 2003;88:3160–6.
[14] Haffner SM, Shaten J, Stern MP, Smith GD, Kuller L, for the MRFIT Research Group.
Low levels of sex hormone-binding globulin and testosterone predict the development of
non-insulin-dependent diabetes mellitus in men. Multiple Risk Factor Intervention Trial.
Am J Epidemiol 1996;143:889–97.
[15] Oh JY, Barrett-Connor E, Wedick NM, Wingard DL. Endogenous sex hormones and the
development of type 2 diabetes in older men and women: the Rancho Bernardo Study.
Diabetes Care 2002;25:55–60.
944 R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945

[16] Stellato RK, Feldman HA, Hamdy O, Horton ES, McKinlay JB. Testosterone, sex
hormone-binding globulin, and the development of type 2 diabetes in middle-aged men:
prospective results from the Massachusetts Male Aging Study. Diabetes Care 2000;23:
490–4.
[17] Barrett-Connor E, Khaw KT, Yen SS. Endogenous sex hormone levels in older adult men
with diabetes mellitus. Am J Epidemiol 1990;132:895–901.
[18] Barrett-Connor E. Lower endogenous androgen levels and dyslipidemia in men with non-
insulin-dependent diabetes mellitus. Ann Intern Med 1992;117:807–11.
[19] Tan RS, Pu SJ. Impact of obesity on hypogonadism in the andropause. Int J Androl 2002;
25:195–201.
[20] Goodman-Gruen D. Barret-Connor E: Sex differences in the association of endogenous sex
hormone levels and glucose tolerance status in older men and women. Diabetes Care 2000;
23:912–8.
[21] Couillard C, Gagnon J, Bergeron J, Leon AS, Rao DC, Skinner JS, et al. Contribution of
body fatness and adipose tissue distribution to the age variation in plasma steroid hormone
concentrations in men: the HERITAGE Family Study. J Clin Endocrinol Metab 2000;85:
1026–31.
[22] Abate N, Haffner M, Garg A, Peshock R, Grundy S. Sex steroid-hormones, upper body
obesity and insulin resistance. J Clin Endocrinol Metab 2002;87:4522–7.
[23] Phillips GB, Jing T, Heymsfield SB. Relationships in men of sex hormones, insulin,
adiposity, and risk factors for myocardial infarction. Metabolism 2003;52:784–90.
[24] Plymate SR, Jones RE, Matej LA, Friedl KE. Regulation of sex hormone binding globulin
(SHBG) production in Hep G2 cells by insulin. Steroids 1988;52:339–40.
[25] Anderson DC. Sex-hormone-binding globulin. Clin Endocrinol 1974;3:69–96.
[26] Lovejoy JC, Smith SR, Bray GA, Veldhuis JD, Rood JC, Tulley R. Effects of
experimentally induced mild hyperthyroidism on growth hormone and insulin secretion
and sex steroid levels in healthy young men. Metabolism 1997;46:1424–8.
[27] Motta E. Epilepsy and hormones. Neurol Neurochir Pol 2000;33(Suppl 1):S31–6.
[28] Adlercreutz H, Hockerstedt K, Bannwart C, Bloigu S, Hamalainen E, Fotsis T, et al. Effect of
dietary components, including lignans and phytoestrogens, on enterohepatic circulation and
liver metabolism of estrogens and on sex hormone binding globulin (SHBG). J Steroid
Biochem Mol Biol 1987;27:1135–44.
[29] Selby C. Sex hormone binding globulin: origin, function and clinical significance. Ann Clin
Biochem 1990;27:532–41.
[30] Vermeulen A, Kaufman JM, Deslypere JP, Thomas G. Attenuated luteinizing hormone
(LH) pulse amplitude but normal LH pulse frequency, and its relation to plasma
androgens in hypogonadism of obese men. J Clin Endocrinol Metab 1993;76:1140–6.
[31] Gapstur SM, Gann PH, Kopp P, Colangelo L, Longcope C, Liu K. Serum androgen
concentrations in young men: a longitudinal analysis of associations with age, obesity, and
race. The CARDIA male hormone study. Cancer Epidemiol Biomarkers Prev 2002;11:
1041–7.
[32] Strain GW, Zumoff B, Miller LK, Rosner W, Levit C, Kalin M, et al. Effect of massive
weight loss on hypothalamic-pituitary-gonadal function in obese men. J Clin Endocrinol
Metab 1988;66:1019–23.
[33] Zumoff B, Strain GW, Miller LK, Rosner W, Levit CD, Miller EH, et al. Partial reversal of
the hypogonadotropic hypogonadism of obese men by administration of cortico-
suppressive doses of dexamethasone. Int J Obes 1988;12:525–31.
[34] Lima N, Cavaliere H, Knobel M, Halpern A, Medeiros-Neto G. Decreased androgen
levels in massively obese men may be associated with impaired function of the gonadostat.
Int J Obes 2000;24:1433–7.
[35] Frederich RC, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS. Leptin levels
reflect body lipid content in mice: evidence for diet-induced resistance to leptin action. Nat
Med 1995;1:1311–4.
R.A. Hijazi et al / Med Clin N Am 88 (2004) 933–945 945

[36] Baumgartner R, Walters D, Morley J, Patrick P, Montoya GD, Garry PJ. Age related
changes in sex hormones affect the sex difference in serum leptin independently of changes
in body fat. Metabolism 1999;48:378–84.
[37] Vettor R, De Pergola G, Pagano C, Englaro P, Laudadio E, Giorgino F, et al. Gender
differences in serum leptin in obese people:relationships with testosterone, body fat
distribution and insulin sensitivity. Eur J Clin Invest 1997;27:1016–24.
[38] Nishizawa H, Shimomura I, Kishida K, Maeda N, Kuriyama H, Nagaretani H, et al.
Androgens decrease plasma adiponectin, an insulin–sensitizing adipocyte-derived protein.
Diabetes 2002;51:2734–41.
[39] McCulloch DK, Campbell IW, Wy FC, Prescott RJ, Clarke BF. The prevalence of diabetic
impotence. Diabetologia 1980;18:279.
[40] Klein R, Klein BEK, Lee KE, Moss SE, Cruickshanks KJ. Prevalence of self-reported
erectile dysfunction in people with long-term IDDM. Diabetes Care 1996;19:135–41.
[41] Andersson KE, Wagner G. Physiology of penile erection. Physiol Rev 1995;75:191.
[42] Lue TF. Erectile dysfunction. N Engl J Med 2000;342:1802–13.
[43] Miller MAW, Morgan RJ, Thompson CS, Mikhailidis DP, Jeremy JY. Hydrolysis of cyclic
guanosine monophosphate and cyclic adenosine monophosphate by the penis and aorta of
the diabetic rat. British Journal of Urology 1996;78:252–6.
[44] Azad N, Emanuele NV, Abraira, Henderson WG, Colwell J, Levin SR, et al. The effects of
intensive glycemic control on neuropathy in the VA cooperative study on type II diabetes
mellitus (VA CSDM). J Diabetes Complications 1999;13:307–13.
[45] Rendell MS, Rajfer J, Wicker PA, Smith MD, for the Sildenafil Diabetes Study Group.
Sildenafil for treatment of erectile dysfunction in men with diabetes. A randomized
controlled trial. JAMA 1999;281:421–6.
[46] Goldstein I, Young JM, Fischer J, Bangerter K, Segerson T, Taylor T, for the Vardenafil
Diabetes Study Group. Vardenafil, a new phosphodiesterase type 5 inhibitor, in the
treatment of erectile dysfunction in men with diabetes. Diabetes Care 2003;26:777–83.
[47] Saenz de Tejada I, Anglin G, Knight JR, Emmick JT. Effects of tadalafil on erectile
dysfunction in men with diabetes. Diabetes Care 2002;25:2159–64.
[48] Linet OI, Ogring FG, for the Alprostadil Study Group. Efficacy and safety of
intracavernosal alprostadil in men with erectile dysfunction. NEJM 1996;334:873–7.
[49] Padma-Nathan H, Hellstrom WJG, Kaiser FE, Labasky RF, Lue TF, Nolton WE, et al.
Treatment of men with erectile dysfunction with transurethral alprostadil. N Engl J Med
1997;336:1.
[50] Cookson MS, Nadig PW. Long-term results with vacuum constriction device. J Urol 1993;
149:290–4.
[51] Price DE, Cooksey G, Jehu D, Bentley S, Hearnshow JR, Osborn DE. The management of
impotence in diabetic men by vacuum tumescence therapy. Diabet Med 1991;8:964–7.
[52] Kearse WS, Sago AL, Peretsman SJ, Bolton JO, Holcomb RG, Reddy PK, et al. Report of
a multicenter clinical evaluation of the Dura-II penile prosthesis. J Urol 1996;334:873.
[53] Garber BB. Inflatable penile prosthesis: results of 150 cases. Br J Urol 1996;78:933.
Med Clin N Am 88 (2004) 947–999

Diabetic neuropathies
A.I. Vinik, MD, PhD*, Anahit Mehrabyan, MD
Departments of Internal Medicine and Pathology/Neurobiology, The Strelitz
Diabetes Institutes, Eastern Virginia Medical School, 855 West Brambleton Avenue,
Norfolk, VA 23510, USA

Diabetic neuropathies (DNs) are a heterogeneous group of disorders


and present a wide range of abnormalities. They are among the most com-
mon long-term complications of diabetes, and are a significant source of
morbidity and mortality [1]. Estimates of the prevalence of neuropathy vary
substantially, depending on specific diagnostic criteria [2,3]. In the United
States, prevalence estimates have ranged from 5% to 100% [1,2,4–6]. In the
classic study by Pirart [7] of a cohort of 4400 patients, prevalence was found
to reach approximately 45% after 25 years. Using this estimate, about 7
million individuals in the United States alone are likely to be afflicted with
DN. This estimate, however, was done before the understanding that pain in
[1,4]. It is the most common form of neuropathy in the developed countries
of the world, accounts for more hospitalizations than all the other diabetic
complications combined, and is responsible for 50% to 75% of non-
traumatic amputations [4,5]. DN is a set of clinical syndromes that affect
distinct regions of the nervous system, singly or combined. It may be silent
and go undetected, while exercising its ravages, or it may present with
clinical symptoms and signs that although nonspecific and insidious with
slow progression, also mimic those seen in many other diseases. It is
diagnosed by exclusion.
The true prevalence is not known and reports vary from 10% to 90% in
diabetic patients, depending on the criteria and methods used to define
neuropathy. Twenty-five percent of patients attending a diabetes clinic
volunteered symptoms; 50% were found to have neuropathy after a simple
clinical test, such as the ankle jerk or vibration perception test; almost 90%
tested positive to sophisticated tests of autonomic function or peripheral
sensation [8]. It is grossly underdiagnosed by endocrinologists and non-
endocrinologists. Neurologic complications occur equally in type 1 and type

* Corresponding author.
E-mail address: vinikai@evms.edu (A.I. Vinik).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.009
948 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

2 diabetes mellitus and additionally in various forms of acquired diabetes


[2]. The major morbidity associated with somatic neuropathy is foot ulcera-
tion, the precursor of gangrene and limb loss. Neuropathy increases the risk
of amputation 1.7-fold; 12-fold if there is deformity (itself a consequence of
neuropathy); and 36-fold if there is a history of previous ulceration [9].
There are 85,000 amputations in the United States each year, one every
10 minutes, and neuropathy is considered to be the major contributor in
87% of cases. It is also the most life-spoiling of the diabetic complications
and has tremendous ramifications for the quality of life of the person
with diabetes. Once autonomic neuropathy sets in, life can become quite dis-
mal and the mortality rate approximates 25% to 50% within 5 to 10 years
[10,11].

Classification
Diabetic neuropathy is not a single entity but a number of different
syndromes, ranging from subclinical to clinical manifestations depending on
the classes of nerve fibers involved. According to the San Antonio
Convention [12], the main groups of neurologic disturbance in diabetes
mellitus include (1) subclinical neuropathy, determined by abnormalities in
electrodiagnostic and quantitative sensory testing, (2) diffuse clinical neu-
ropathy with distal symmetric sensorimotor and autonomic syndromes, and
(3) focal syndromes.
Subclinical neuropathy is diagnosed on the basis of (1) abnormal elec-
trodiagnostic tests with decreased nerve conduction velocity (NCV) or de-
creased amplitudes; (2) abnormal quantitative sensory tests (QST) for
vibration, tactile, thermal warming, and cooling thresholds; and (3) quan-
titative autonomic function tests revealing diminished heart rate variation
with deep breathing, Valsalva’s maneuver, and postural testing. The different
clinical presentations of diabetic neuropathy are schematically illustrated in
Fig. 1.

Natural history
The natural history of neuropathies separates them into two very
distinctive entities: those that progress gradually with increasing duration
of diabetes, and those that remit usually completely. Sensory and autonomic
neuropathies generally progress, whereas mononeuropathies, radiculopa-
thies, and acute painful neuropathies, although symptoms are severe, are
short-lived and tend to recover [13]. Progression of DN is related to
glycemic control in both type 1 and type 2 diabetes [14,15]. It seems that the
most rapid deterioration of nerve function occurs soon after the onset of
type 1 diabetes and within 2 to 3 years there is a slowing of the progress with
a shallower slope to the curve of dysfunction. In contrast, in type 2 diabetes,
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 949

Large fiber Small fiber Proximal Acute mono Pressure


Neuropathy Neuropathy motor Neuropathies Palsies
Neuropathy
Sensory loss: 0→ +++ Sensory loss: 0 → + Sensory loss: 0 → + Sensory loss: 0 → + Sensory loss in Nerve
(Touch, vibration) (thermal , allodynia) Pain: + → +++ Pain: + → +++ distribution: + → +++
Pain: + → +++ Pain: + → +++ Pain: + → ++

Tendon reflex: Tendon reflex: N


Tendon reflex: N → Tendon reflex: N →

Proximal Motor deficit: Motor deficit: Tendon reflex: N


+ → +++ + → +++ Motor deficit: + → +++


Motor deficit: 0
Motor deficit: 0 → +++

Fig. 1. Schematic representation of different clinical presentations of diabetic neuropathy.

slowing of NCVs may be one of the earliest neuropathic abnormalities and


often is present even at diagnosis [16]. After diagnosis, slowing of NCV
generally progresses at a steady rate of approximately 1 m/s/y, and the level
of impairment is positively correlated with duration of diabetes. Although
most studies have documented that symptomatic patients are more likely to
have slower NCVs than patients without symptoms, these do not relate to
the severity of symptoms. In a long-term follow-up study of type 2 diabetes
patients [17], electrophysiologic abnormalities in the lower limb increased
from 8% at baseline to 42% after 10 years, with a decrease in sensory and
motor amplitudes, indicating axonal destruction was more pronounced than
the slowing of the NCVs. An increase of about 2 points in an 80-point
clinical scale can be expected per year. These scales contain information of
motor, sensory, and autonomic signs and symptoms. Using objective
measures of sensory function, such as the vibration perception threshold
test, the rate of decline in function has been reported as 1 to 2 vibration units
per year. There now seems to be a decline, however, in this rate of evolution.
For example, in the recent nerve growth factor study, the vibration
perception threshold at the beginning of the study in the placebo group
was identical to that at the end of 1 year [18,19]. It seems that host factors
pertaining to general health and nerve nutrition are changing. This is
particularly important in doing studies on treatment of DN, which have
always relied on differences between drug treatment and placebo and have
apparently been successful because of the decline in placebo-treated patients
[20]. Based on the earlier estimates of change, clinically meaningful loss of
vibration perception and conduction velocity was estimated to take at least
3 years, dictating a future need to carry out studies over a longer period
950 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

of time. It is also important to recognize that DN is a disorder wherein the


prevailing abnormality is loss of axons that electrophysiologically translates
to a reduction in amplitudes and not conduction velocities, and changes in
NCV may not be an appropriate means of monitoring progress or
deterioration of nerve function. It has always been advocated that diabetes
affects the longest fibers first, hence the increased predisposition in taller
individuals [21]. Now it seems that small-fiber involvement may herald the
onset of neuropathy and even diabetes. Small-fiber function is not detectable
using standard electrophysiology and requires measurement of sensory,
neurovascular, and autonomic thresholds and cutaneous nerve fiber density
[22,23].
There are few data on the longitudinal trends in small-fiber dysfunction.
Much remains to be learned of the natural history of diabetic autonomic
neuropathy. Recently, Karamitsos et al [24] reported that the progression of
diabetic autonomic neuropathy is significant during the 2 years subsequent
to its discovery. The mortality for diabetic autonomic neuropathy has been
estimated to be 44% within 2.5 years of diagnosing symptomatic autonomic
neuropathy [10]. A meta-analysis [25] reveals that the mortality rate after
5.8 years of diabetes with symptomatic autonomic neuropathy was 29%.

Pathogenesis
Fig. 2 summarizes the current view of the pathogenesis of DN. This figure
depicts multiple etiologies including metabolic, vascular, autoimmune,
oxidative stress, and neurohormonal growth-factor deficiency.
Detailed discussion of the different theories is beyond the scope of this
article; the reader is referred to several excellent recent reviews [26–28]. DN
is a heterogeneous disease with widely varying pathology, however,
suggesting differences in pathogenic mechanisms for the different clinical

GENETIC Glucotoxicity
APoE4 and
AR Z2 alleles
ACE polymorphism Lipotoxicity Epigenetic
Toll rec polymorphism AGEs PARPs etc
Endothelial Functional Progressive
Initiating
Injury Changes Pathological
Event Changes
INFLAMMATION
Oxidative/Nitrative
Stress months to years
PKC
Selectins
VCAMS
IL6, TNFα, NFκB
ROS, nitrotyrosines

Fig. 2. Pathogenesis of diabetic neuropathy: the initiation and progression of diabetes.


A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 951

syndromes. Recognition of the clinical homologue of these pathologic


processes is the first step in achieving the appropriate form of intervention.

Metabolic memory hypothesis


The prolonged impact of the early metabolic environment on the devel-
opment and progression of diabetes-related complications is modulated by
a process of ‘‘metabolic memory.’’ The development of vascular complica-
tions of diabetes begins with an underlying genetic predisposition, which
when acted on by initiating events, such as overfeeding or smoking, results
in inflammatory changes that may precede hyperglycemia. Inflammation
and hyperglycemia unleash a cascade of events that effect cellular proteins,
gene expression, and cell-surface receptor expression in the endothelium,
ultimately resulting in progressive pathologic changes and subsequent
vascular complications (see Fig. 2).

Clinical presentation
An international consensus meeting on the outpatient diagnosis and
management of DN agreed on a simple definition of DN as ‘‘the presence of
symptoms or signs of peripheral nerve dysfunction in people with diabetes
after the exclusion of other causes’’ [12]. It was also agreed that neuropathy
cannot be diagnosed without a careful clinical examination: absence of
symptoms cannot be equated with absence of neuropathy because asymp-
tomatic neuropathy is common. The importance of excluding nondiabetic
causes was emphasized in the Rochester Diabetic Neuropathy Study in
which up to 10% of peripheral neuropathy in diabetic patients was deemed to
be of nondiabetic causation [29]. A more detailed definition of neuropathy
had previously been agreed at the San Antonio Consensus Conference:
‘‘diabetic neuropathy is a descriptive term meaning a demonstrable disorder,
either clinically evident or sub-clinical, that occurs in the setting of diabetes
mellitus without other causes for peripheral neuropathy. The neuropathic
disorder includes manifestations in the somatic or autonomic parts of the
peripheral nervous system’’ [12]. It is generally agreed that DN should not be
diagnosed on one symptom, sign, or test alone: a minimum of two
abnormalities (from symptoms, signs, nerve conduction abnormalities,
quantitative sensory tests, or quantitative autonomic tests) is recommended
by Dyck [30]. It is, however, woefully underdiagnosed by endocrinologists
and nonendocrinologists. In the GOAL A1c study [31] identification of the
absence of neuropathy in 7000 patients was fairly adequate but was only
accurate in the presence of mild neuropathy one third of the time and reached
75% only if neuropathy was severe. Clearly, there is a need for education of
the means whereby neuropathy may be diagnosed.
The spectrum of clinical neuropathic syndromes described in patients with
diabetes mellitus includes dysfunction of almost every segment of the somatic
peripheral and autonomic nervous system [32]: it has been said that ‘‘knowing
952 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

neuropathy means to know the whole of medicine.’’ Each syndrome can be


distinguished by its pathophysiologic, therapeutic, and prognostic features.

Focal neuropathies
Mononeuritis and entrapment syndromes
Mononeuropathies occur primarily in the older population; their onset is
generally acute, associated with pain; and their course is self-limiting, re-
solving within 6 to 8 weeks. These are caused by vascular obstruction after
which adjacent neuronal fascicles take over the function of those infarcted by
the clot [28]. Mononeuropathies must be distinguished from entrapment syn-
dromes that start slowly, progress, and persist without intervention (Table 1).
Common entrapment sites in diabetic patients involve median, ulnar,
radial, femoral, lateral cutaneous nerves of the thigh, peroneal, and medial
and lateral plantar nerve. Carpal tunnel syndrome occurs three times as
frequently in a people with diabetes compared with a normal healthy
population [33,34], and its increased prevalence in diabetes may be related to
diabetic cheiroarthropathy [35], repeated undetected trauma, metabolic
changes, or accumulation of fluid or edema within the confined space of the
carpal tunnel [32]. It is found in up to one third of patients with diabetes
[36]. If recognized, the diagnosis can be confirmed by electrophysiologic
study and therapy is simple with surgical release. The mainstays of
nonsurgical treatment are resting the wrist aided by the placement of a wrist
splint in a neutral position for day and night use, and the addition of anti-
inflammatory drug medications. Surgical treatment consists of sectioning
the volar carpal ligament [37]. The decision to proceed with surgery should
be based on several considerations, including severity of symptoms,
appearance of motor weakness, and failure of nonsurgical treatment.

Diffuse neuropathies
Proximal motor neuropathies (diabetic amyotrophy, femoral neuropathy)
For many years proximal neuropathy has been considered as a compo-
nent of DN. Its pathogenesis was ill understood [38], and its treatment was

Table 1
Mononeuritis versus entrapment
Mononeuritis Entrapment
Onset sudden Onset gradual
Usually single nerve but may be multiple Single nerves exposed to trauma
Common nerves: CN III, VI, VII, ulnar, Common nerves: median, ulnar, peroneal,
median, peroneal medial, and lateral plantar
Not progressive and resolves spontaneously Progressive
Treatment symptomatic Treatment, rest, splints, diuretics,
steroid injections, and surgery for paralysis
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 953

neglected with the anticipation that the patient would eventually recover,
albeit over a period of some 1 to 2 years, suffering considerable pain, weak-
ness, and disability. The condition has a number of synonyms: proximal
neuropathy, femoral neuropathy, diabetic amyotrophy, and diabetic neu-
ropathic cachexia. Proximal motor neuropathy can be clinically identified
based on recognition of these common features: (1) primarily affects the
elderly; (2) gradual or abrupt onset; (3) begins with pain in the thighs and
hips or buttocks; (4) followed by significant weakness of the proximal
muscles of the lower limbs with inability to rise from the sitting position
(positive Gower’s maneuver); (5) begins unilaterally and spreads bilaterally;
(6) coexists with distal symmetric polyneuropathy (DSPN); and (7)
spontaneous muscle fasciculation, or provoked by percussion. The condi-
tion is now recognized as being secondary to a variety of causes unrelated to
diabetes, but which have a greater frequency in patients with diabetes than
the general population. It includes patients with chronic inflammatory
demyelinating polyneuropathy, monoclonal gammopathy, circulating GM1
antibodies, and antibodies to neuronal cells and inflammatory vasculitis
[39,40]. It was formerly thought to resolve spontaneously in 1.5 to 2 years,
but now, if found to be immune-mediated, can resolve within days on
immunotherapy. The condition is readily recognizable clinically with pre-
vailing weakness of the iliopsoas, obturator, and adductor muscles, together
with relative preservation of the gluteus maximus and minimus and
hamstrings [41]. Those people affected have great difficulty rising out of
chairs unaided and often use their arms to assist themselves. Heel or toe
standing is surprisingly good. In the classic form of diabetic amyotrophy,
axonal loss is the predominant process and the condition coexists with
DSPN [42]. Electrophysiologic evaluation reveals lumbosacral plexopathy
[41]. In contrast, if demyelination predominates and the motor deficit affects
proximal and distal muscle groups, the diagnosis of chronic inflammatory
demyelinating polyneuropathy, monoclonal gammopathy of unknown
significance, and vasculitis should be considered [43,44]. It seems probable
that these conditions occur more commonly in people with diabetes [45–47].
Vinik [48] (Fig. 3) pointed out that almost half the patients with proximal
neuropathies have a vasculitis and all but 9% have chronic inflammatory
demyelinating polyneuropathy or monoclonal gammopathy of unknown
significance or a ganglioside antibody syndrome [49]. Sharma et al [46]
examined over 1000 patients with neurologic disorders and found that
chronic inflammatory demyelinating polyneuropathy was 11 times more
frequent among their diabetic than nondiabetic population.
Biopsy of the obturator nerve reveals deposition of immunoglobulin,
demyelination, and inflammatory cell infiltrate of the vasa nervorum [50].
Cerebrospinal fluid protein content is high and there is an increase in the
lymphocyte count. Treatment options include intravenous immunoglobulin
for chronic inflammatory demyelinating polyneuropathy, plasma exchange
for monoclonal gammopathy of unknown significance, steroids and
954 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Vasculitis
9% CIDP
17% MGUS
Diabetes

22% 52%

Vinik AI Clin in Geriatric Med 1999,; 15: 293-320.

Fig. 3. Disabling peripheral neuropathies in older adults. (From Vinik A. Diagnosis and
management of diabetic neuropathy. Clin Geriatr Med 1999;15:293–320; with permission.)

azathioprine for vasculitis, and withdrawal from drugs or other agents that
may have caused a vasculitis. It is important to divide proximal syndromes
into these two subcategories, because the chronic inflammatory demyelin-
ating polyneuropathy variant responds dramatically to intervention [43,51],
whereas amyotrophy runs its own course over months to years. Until more
evidence is available, they should be considered separate syndromes.
These conditions need to be distinguished from spinal stenosis syn-
dromes. There is encroachment on nerve roots as they emerge from the
spinal cord, osteophytes may cause compression, with aging there is
hypertrophy of the ligamentum flavum and disk dehydration, and there
may even be some form of arachnoiditis. When the compression involves the
vascular system claudication typical occurs walking downhill, is relieved by
bending forward, and occurs at the watershed level between T12 and L1/2.
Nerve root compression is more typical at L5/S1 and in difficult cases it may
be necessary to obtain an MRI of the lumbosacral spine. Diagnosis is
critical because therapy may be simple physical therapy or surgical decom-
pression if symptoms are severe or there is motor paralysis.

Distal symmetric polyneuropathy


Distal symmetric polyneuropathy is the most common and widely
recognized form of DN. The onset is usually insidious but occasionally is
acute, following stress or initiation of therapy for diabetes. DSPN may be
either sensory or motor, and involve small fibers, large fibers, or both [52].
Fig. 4 is a simplified version of the peripheral nervous system. Also shown in
Fig. 5 is the usual clinical presentation of the large- and small-fiber
neuropathies.
Small nerve fiber dysfunction usually occurs early and often is present
without objective signs or electrophysiologic evidence of nerve damage [53].
It is manifested early with symptoms of pain and hyperalgesia in the lower
limbs, followed by a loss of thermal sensitivity and reduced light touch and
pinprick sensation [32].
There is now evidence that DSPN may be accompanied by loss of
cutaneous nerve fibers that stain positive for the neuronal antigen PGP 9.5
(Fig. 6) [54] and impaired neurovascular blood flow [55]. There are,
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 955

Motor Sensory Autonomic

Myelinated Myelinated Thinly Un- Thinly Un-


myelinated myelinated myelinated myelinated
Aα Aα/β Aδ C C
C

Large Small

Muscle Touch, Cold Warm Heart Rate, Blood


control Vibration, perception, perception Pressure,
Position Pain Pain Sweating,
perception GIT,GUT, function

Fig. 4. A simplified view of the peripheral nervous system.

however, a variety of ways in which small-fiber neuropathies can present.


Clinical manifestations of small-fiber neuropathies include the following:
 Symptoms are prominent. Pain is of the C-fiber type. It is burning and
superficial and associated with allodynia (ie, interpretation of all stimuli
as painful)
 Late in the condition there is hypoalgesia
 Defective warm thermal sensation
 Defective autonomic function with decreased sweating, dry skin, im-
paired vasomotion and blood flow, and a cold foot
 There is remarkable intactness of reflexes, motor strength
 Electrophysiologically silent
 Loss of cutaneous nerve fibers using PGP 9.5 staining
 Diagnosed clinically by reduced sensitivity to 1 g Semmes-Weinstein
monofilament and pricking sensation using the Waardenburg wheel or
similar instrument
 Abnormalities in thresholds for warm thermal perception, neurovascu-
lar function, pain, quantitative sudorimetry, and quantitative autono-
mic function tests
 Risk is foot ulceration and subsequent gangrene (there are 65,000
amputations in the United States each year, 1 every 2 minutes; 50% are
preventable)

Neuropathy

Large Fiber Small Fiber

Motor Pain
Vibration Autonomic
Position sense Thermal
Touch/pressure Produces symptoms
Interferes with QOL and ADL and leads to
morbidity and mortality

Fig. 5. Clinical presentation of the large- and small-fiber neuropathies.


956 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Fig. 6. Loss of cutaneous nerve fibers that stain positive for the neuronal antigen PGP 9.5 in
sensory neuropathy. (A) Normal density of epidermal nerve fibers (arrows) in back. (B) Slightly
reduced density and abnormal nerve fiber swellings (arrows) in proximal thigh. (C) Complete
clearance of nerve fibers in calf.

Pain in diabetic neuropathies


Overall, approximately 10% of patients with diabetes experience persis-
tent pain from neuropathy [56]. Pain syndromes that last less than 6 months
to a year are classified as acute, and include the insulin neuritis syndrome,
which occurs often at the beginning of therapy for diabetes and is self-
limiting. Pain syndromes lasting longer than 6 months to a year are classified
as chronic [57]. The pain can be ongoing; spontaneous; or hyperalgesic (ie,
increased response to a painful stimulus). It can be severe and sometimes
intractable. Pain may be stimulus-independent or stimulus-evoked (Table 2).
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 957

Table 2
Sensory signs and symptoms of stimulus-independent and stimulus-evoked pain
Stimulus-independent pain Stimulus-evoked pain
Patient expresses symptoms Physician elicits signs after mechanical, thermal,
on presentation or chemical stimulation
Involves continuous burning sensation; Includes hyperalgesia or allodynia
also may include:
Intermittent shooting, lancinating
sensations
Electric-shock–like pain
Dysesthesias

Neuropathic pain and quality of life


When pain is not adequately controlled, mood and sleep disturbances are
common [58]. A prospective study of 105 patients with painful DN showed
that pain interfered substantially with sleep and enjoyment of life. Patients
also reported that pain interfered with their normal work, mood, and
activities of daily living, including walking [59]. Although the pain of DN
may resolve spontaneously, pain that persists for more than 3 months is
unlikely to do so, and it can last for years with ongoing and significant
disruptions of a patient’s quality of life.
I don’t like peripheral neuritis—it interferes with work.
R.D. Lawrence, 1923
Individuals with DN were shown to have significantly lower quality of
life scores using a validated Norfolk quality of life tool compared with those
without DN [60]. Small-fiber symptoms had greater effects on quality of life,
whereas features of large-fiber dysfunction, such as weakness and co-
ordination, were more likely to affect activities of daily living [61].
Treatment of pain has been shown to improve patients’ quality of life
[62]. It has become abundantly clear, however, that quality of life in DN is
not simply a function of the nerve damage to the somatic or autonomic
nervous systems, and quality of life needs to be measured independently of
the subjective and objective symptoms and signs of neuropathy and the
quantitative sensory tests and electrophysiology if any form of therapy is to
be properly evaluated [61].

Pain presentation as a confounding issue


The presentation and character of pain in DN can be highly diverse,
although it typically worsens at night. For example, in two randomized
clinical trials, patients described their pain as burning; pins and needles;
shooting; aching; jabbing; sharp; cramping; tingling; cold; or allodynia (pain
response to a stimulus not normally associated with pain) in nature. Of
particular note is the observation that certain types of pain (eg, burning,
958 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

pins and needles, tingling, and sharp knife-like or stabbing pain) derive from
C-fiber dysfunction and may have a different origin and respond differently
to different medications (eg, gabapentin [62] and the PKC-inhibitor LY
333531) [63]. These differences have led to the development of improved
tools for the quantitation of symptoms of DNs [64].
Even more hazardous to the interpretation of pain relief is the fact that
drugs used for treatment (eg, nerve growth factor) induce local sensitivity at
the site of injection, which is construed as pain. Other drugs (eg, topiramate)
cause paresthesias interpreted as pain, leading to the apparent failure of two
trials and the success of a third that specifically defined the nature of pain
and the site to exclude the nonneuropathic pain syndromes listed later
(Table 3) [48].
Pain tolerance is uniquely individual. Quantifying pain is subjective and
can be quite variable, depending on the terminology used to define pain, the
instrument used to measure pain, and the patient’s ability to describe his or
her pain. Moreover, few patients can discern whether their pain is
neuropathic or nonneuropathic in origin, and careful clinical evaluation is
critical for devising suitable management strategies.

Issues of different causes of pain in diabetes


In any painful syndrome in diabetes, appropriate attention to the
underlying condition is a critical aspect of overall management and to the
evaluation of new therapies. Physicians must be able to differentiate painful
DN from other conditions with which it may be confused and that may
coexist in patients with diabetes. The most common of these are claudica-
tion, Morton’s neuroma, Charcot’s neuroarthropathy, fasciitis, osteoarthri-
tis, and radiculopathy (see Table 3) [48,58].

Current perspectives on pathophysiology of pain: management implications


Management of painful DN and other pain syndromes is changing as
research elucidates underlying pathophysiologic mechanisms. The complex-
ities of pain syndromes and advances in basic pain research have
contributed to an evolving concept of pain and strategies for its manage-
ment. Krause and Backonja [58] defined neuropathic pain as ‘‘a group of
disorders characterized by pain due to dysfunction or disease of the nervous
system at a peripheral level, a central level, or both.’’

Acute painful neuropathy


Some patients develop a predominantly small-fiber neuropathy, which is
manifested by pain and paresthesias early in the course of diabetes (Fig. 7).
It may be associated with the onset of insulin therapy and has been termed
‘‘insulin neuritis’’ [65]. By definition it has been there for less than 6 months.
Symptoms often are exacerbated at night and are manifested in the feet
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 959

Table 3
Common pain syndromes similar to painful diabetic neuropathy
Condition Key characteristics and differentiating features
Claudication Doppler ultrasonography confirms clinical diagnosis of arterial
occlusion
Patients with diabetes may present with normal extremities and
absent foot pulses
Peripheral arterial occlusion with underlying atherosclerosis
Usually intermittent, worsened by walking; remits with rest;
other signs or symptoms suggest arterial insufficiency
Morton’s neuroma Benign neuroma formation on third plantar interdigital nerve
Generally unilateral
More frequent in women
Pain elicited when pressure is applied with the thumb between
the first and fourth metatarsal heads
Osteoarthritis Can be secondary to diabetes mellitus, but pain is usually
gradual in onset and in one or two joints
Differential diagnosis based on x-ray
Morning stiffness, diminished joint motion, and flexion
contractures are characteristic
Pain worsens with exercise and improves with rest
Radiculopathy can result
Radiculopathy Can be caused by diabetes, but can also result from arthritis
or metastatic disease
Neurologic examinations and imaging can localize lesion site
Pain can occur in thorax, extremities, shoulder, or arm,
depending on site of lesion
Charcot’s May result from osteopenia caused by increased blood flow
neuroarthropathy following repeated minor trauma in individuals
with diabetic neuropathy
Warm to hot foot with increased blood flows
Decreased warm sensory perception, vibration detection
Plantar fasciitis Pain in the plantar region of the foot
Tenderness along the plantar fascia when ankle is dorsiflexed
Shooting or burning in the heel with each step
Worsening pain with prolonged activity
Often associated with calcaneal spur on radiography
Tarsal tunnel syndrome Caused by entrapment of the posterior tibial nerve
Pain and numbness radiate from beneath the medial
malleolus to the sole
Clinical examination includes percussion, palpation for
possible soft tissue matter, nerve conduction studies, MRI

more than the hands. Spontaneous episodes of pain can be severely


disabling. The pain varies in intensity and character. In some patients, the
pain has been variably described as burning, lancinating, stabbing, or sharp.
Paresthesias or episodes of distorted sensation, such as pins and needles,
tingling, coldness, numbness, or burning, often accompany the pain [52].
The lower legs may be exquisitely tender to touch, with any disturbance of
the hair follicles resulting in excruciating pain. Because pain can be
aggravated by repeated contact of the lower limbs with foreign objects,
960 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

• Pain is C-fiber type, burning, superficial,


allodynia
• Early hyperesthesia and hyperalgesia, late
hypo-
• Impaired warm thermal and pain
thresholds
• Decreased sweating
• Normal strength, reflexes and EMG!!!
Vinik, A.I., Erbas, T., Stansberry, K., Pittenger, G, Small Fiber Neuropathy and Neurovascular Disturbances in
Diabetes Mellitus, Exp. Clin. Endocrinol Diabetes, Vol. 109, Suppl 2, S451-S473, 2001

Fig. 7. Clinical presentation of the large and small neuropathies. (From Vinik AI, Erbas T,
Stansberry K, Pittenger G. Small fiber neuropathy and neurovascular disturbances in diabetes
mellitus. Exp Clin Endocrinol Diabetes 2001;109(Suppl 2):S451–73, Ó J.A. Barth Verlag in
Georg Thieme Verlag KG; with permission.)

even basic daily activities, such as sitting at a desk may be disrupted. Pain
often occurs at the onset of the disease and is often worsened by initiation of
therapy with insulin or sulfonylureas [65].
It may be associated with profound weight loss and severe depression
that has been termed ‘‘diabetic neuropathic cachexia’’ [66]. This syndrome
occurs predominantly in male patients and may occur at any time in the
course of both type 1 and type 2 diabetes. It is self-limiting and invariably
responds to simple symptomatic treatment. Such conditions as Fabry’s
disease, amyloid, HIV infection, heavy metal poisoning (eg, arsenic), and
excess alcohol consumption should be excluded. It does overlap with the
idiopathic variety of acute painful small-fiber neuropathy that is also
a diagnosis by exclusion [67].

Chronic painful neuropathy


There is another variety of painful polyneuropathy with onset occurring
later often years in the course of the diabetes, in which the pain persists for
longer than 6 months and becomes debilitating (see Fig. 6). This condition
may result in tolerance to narcotics and analgesics, and finally to addiction.
It is extremely resistant to all forms of intervention, and most frustrating to
both patient and physician. In this simplified scheme (Fig. 8), the spinal cord
is illustrated as an oval.

Normal situation, no pain. Central terminals of unmyelinated primary


C-afferents project into the dorsal horn and make contact with secondary
pain-signaling neurons. Low-threshold mechanoreceptive primary A beta-
afferents project without synaptic transmission into the dorsal columns (not
shown) and also contact secondary afferent dorsal horn neurons.

Peripheral sensitization, central sensitization processes in peripheral nocicep-


tors (peripheral sensitization, star in the periphery), leading to spontaneous
burning pain, static mechanical hyperalgesia, and heat hyperalgesia. This
spontaneous activity in nociceptors induces secondary changes in the central
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 961

Normal C-fiber C-fiber loss Central


sensitization Nociceptor disinhibition
sensitization Cold hyperalgesia

Sympat

+
_

NA

C Aβ C Aβ /Aδ C Aβ C Aβ
Fig. 8. Schematic representation of the generation of neuropathic pain. In this simplified
scheme, the spinal cord is illustrated as an oval.

sensory processing, leading to spinal cord hyperexcitability (central sensi-


tization, star in spinal cord) that causes input from mechanoreceptive A
beta-fibers (light touching) and A delta-fibers (punctuate stimuli) to be
perceived as pain (dynamic and punctuate mechanical allodynia). Moreover,
afferent terminals in the periphery or afferent stomata in the dorsal root
ganglion acquire sensitivity to norepinephrine by expressing A-receptors at
their membrane. Activity in postganglionic sympathetic neurons is now
capable of activating afferent neurons by the release of norepinephrine.

Synaptic reorganization after C-nociceptor degeneration. Nociceptor func-


tion may be selectively impaired and the fibers degenerate after nerve lesion.
Accordingly, the synaptic contacts between central nociceptor terminals and
secondary nociceptive neurons are reduced. Central terminals from intact
mechanoreceptive A beta-fibers start to sprout to form novel synaptic
contacts with the ‘‘free’’ central nociceptive neurons. This anatomic
reorganization in the dorsal horn causes input from mechanoreceptive A
beta-fibers (light touching) to be perceived as pain (dynamic mechanical
allodynia). In such patients, temperature sensation is profoundly impaired
in areas of severe allodynia.

Central disinhibition and cold hyperalgesia. Normally, cold stimuli are


conveyed by A delta-fibers and cold pain by C fibers. A selective damage
of cold-sensitive A delta-fibers leads to a loss of central inhibition mediated
by interneurons (disinhibition), resulting in cold hyperalgesia.
Pathophysiologic changes in the nervous system can produce symptoms
defined as either negative, such as loss of sensory quality, or positive, such as
spontaneous pain. Patients with neuropathic pain usually present with both.
Absence of pain sometimes may not be caused by improvement in
neuropathy, but to a consequence of neuronal loss. Physicians must exclude
962 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

progression of neuropathy when patients report loss of pain. Neuropathic


pain can manifest as stimulus-independent pain or as stimulus-evoked or
stimulus-dependent pain, whose underlying mechanisms are likely to differ.
Similarly, the mechanisms responsible for hyperalgesia and allodynia
differ. Hyperalgesia is defined as increased pain response to a normally pain-
ful stimulus. Allodynia is said to occur when pain is provoked by a stimulus
not normally painful. This is related to the different nerve pathways
implicated in these various categories. For example, aberrations of the C
and A delta-fibers may result in the burning or prickling sensations of
stimulus-independent pain or of hyperalgesia. Under pathologic conditions,
touch-sensitive A beta-fibers may cause stimulus-independent dysesthesias or
paresthesias or stimulus-evoked allodynia.

Neuropharmacology of pain
The neuropharmacology of pain is also becoming better understood. For
example, recent data suggest that c-aminobutyric acid, voltage-gated sodium
channels, and glutamate receptors may be involved in the pathophysiology
of neuropathic pain. Many of the newer agents (called antineuropathic
agents) have significant effects on these neurophysiologic mechanisms.
The growing knowledge about the neural and pharmacologic basis of
neuropathic pain is likely to have important treatment implications, including
development and refinement of a symptom-mechanism-based approach to
neuropathic pain, and implementation of novel treatment strategies using the
newer antiepileptic agents, which may address the underlying neurophysio-
logic aberrations in neuropathic pain, allowing the clinician to increase the
likelihood of effective management.
The mechanism for pain in small-fiber neuropathy is not well understood.
Hyperglycemia may be a factor in lowering the pain threshold. The
condition may appear soon after initiation of therapy [65]. A striking
amelioration of symptoms with the intravenous administration of insulin
can be achieved [68]. There is a sequence in DN, beginning when nerve
function (A beta and C fiber) is intact and there is no pain. With damage to
C fibers there is sympathetic sensitization and peripheral autonomic
symptoms are interpreted a painful. With death of C fibers there is
nociceptor sensitization and A beta-fibers conduct all varieties of peripheral
stimuli, such as touch, and these are interpreted as painful (eg, allodynia).
With time there is reorganization at the cord level and the patient
experiences clod hyperalgesia and ultimately even with the death of all
fibers pain is registered in the cerebral cortex, whereupon the syndrome
becomes chronic without the need for peripheral stimulation (see Fig. 8 for
explanation of the stages of pain). Disappearance of pain may not
necessarily reflect nerve recovery but rather nerve death. When patients
volunteer the loss of pain, progression of the neuropathy must be excluded
by careful examination.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 963

Large-fiber neuropathies
Large-fiber neuropathies may involve sensory or motor nerves. These
tend to be the neuropathies of signs rather than symptoms. Large fibers
subserve motor function, vibration perception, position sense, and cold
thermal perception. Unlike the small nerve fibers these are the myelinated,
rapidly conducting fibers that begin in the toes and have their first synapse
in the medulla oblongata. They tend to be affected first because of their
length and the tendency in diabetes for nerves to ‘‘die back.’’ Because they
are myelinated, they are the fibers represented in the EMG, and subclinical
abnormalities in nerve function are readily detected. The symptoms may be
minimal: sensation of walking on cotton, floors feeling ‘‘strange,’’ inability
to turn the pages of a book, or inability to discriminate among coins.
Clinical presentation of large-fiber neuropathies includes the following:
 Impaired vibration perception (often the first objective evidence) and
position sense
 Depressed tendon reflexes
 A delta-type deep-seated gnawing, dull, like a toothache in the bones of
the feet, or even crushing or cramp-like pain
 Sensory ataxia (waddling like a duck)
 Wasting of small muscles of feet with hammertoes (intrinsic minus feet
and hands) with weakness of hands and feet
 Shortening of the Achilles tendon with pes equinus
 Increased blood flow (hot foot)
Most patients with DSPN, however, have a mixed variety of neuropathy
with both large and small nerve fiber damages. In the case of DSPN, a ‘‘glove
and stocking’’ distribution of sensory loss is almost universal [32]. Early in
the course of the neuropathic process, multifocal sensory loss also might be
found. In some patients, severe distal muscle weakness can accompany the
sensory loss resulting in an inability to stand on the toes or heels. Some
grading systems use this as a definition of severity.

Diagnosis and differential diagnosis of neuropathy


The diagnosis of DN rests heavily on a careful history, for which a number
of questionnaires have been developed by Young et al [6], Dyck [30], Vinik
and Mitchell [69], and others [70,71]. The initial neurologic evaluation should
be directed toward the detection of the specific part of the nervous system
affected by diabetes (see Fig. 1). Bedside neurologic examination is quick and
easy but provides nominal or ordinal measures and contains substantial
interindividual and intraindividual variation. For example, it is useless to
measure vibration perception with a tuning fork other than one that has
a frequency of 128 Hz. Similarly, using a 10-g monofilament is good for
predicting foot ulceration, as is the Achilles reflex, but both are insensitive to
the early detection of neuropathy and a 1-g monofilament increases the
964 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

sensitivity from 60% to 90% [72]. Sensory function must be evaluated on both
sides of the feet and hands if one wants to be sure not to miss entrapment
syndromes. A Tinel’s sign is not only useful for carpal tunnel problems, but
can be applied to the ulnar notch, the head of the fibula, and below the medial
tibial epicondyle for ulnar, peroneal, and medial plantar entrapments,
respectively. The 1988 San Antonio conference on DN and the 1992
conference of the American Academy of Neurology [12] recommended that
at least one parameter from each of the following five categories are measured
to classify DN: (1) symptom profiles, (2) neurologic examination, (3) QST, (4)
nerve conduction study, and (5) autonomic function testing. A number of
simple symptom screening questionnaires are available to record symptom
quality and severity. A simplified neuropathy symptom score that was used in
the European prevalence studies could also be useful in clinical practice [6,73].
The Michigan Neuropathy Screening Instrument is a brief 15-item question-
naire that can be administered to patients as a screening tool for neuropathy
[74]. Other similar symptom scoring systems have also been described [6].
Simple visual analog or verbal descriptive scales may be used to follow
patients’ responses to treatment of their neuropathic symptoms [71,75,76]. It
must always be remembered, however, that identification of neuropathic
symptoms is not useful as a diagnostic or screening tool in the assessment of
DN, as shown by Franse et al [77].
The QST and quantitative autonomic function tests are objective indices
of neurologic functional status. Combined, these tests cover vibratory,
proprioceptive, tactile, pain, thermal, and autonomic function. An in-
ternational group of experts in DN held a consensus meeting to develop
guidelines for the management of diabetic peripheral neuropathy by the
practicing clinician [12]. This clinical staging is in general agreement with
that proposed by Dyck [78] for use in both clinical practice and
epidemiologic studies or controlled clinical trials. The clinical ‘‘no neurop-
athy’’ is equivalent to Dyck’s N0 or N1a; ‘‘clinical neuropathy’’ is equivalent
to N1b, N2a, or N2b; and ‘‘late complications’’ is equivalent to Dyck N3.
There have been a number of other relevant reports, including two on
measures for use in clinical trials to assess symptoms [75] and QST [79]. The
strengths of QST are well documented [80], but the limitations of QST are
also clear. No matter what the instrument or procedure used, QST is only
a semiobjective measure, which is affected by the subject’s attention,
motivation, and cooperation, and by anthropometric variables, such as
age, gender, body mass, and history of smoking and alcohol consumption.
Expectancy and subject bias are additional factors that can exert a powerful
influence on QST findings. Further, QST is sensitive to changes in structure
or function along the entire neuraxis from nerve to cortex; it is not a specific
measure of peripheral nerve function [80].
The American Academy of Neurology reported on the use of QST for
clinical and research purposes [79] suggesting that these tests could be used
as an ancillary but were not sufficiently robust for routine clinical use.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 965

A number of simple symptom screening questionnaires are available to


record symptom quality and severity. A simplified neuropathy symptom
score that was used in the European prevalence studies could also be useful
in clinical practice [6,73]. The Michigan Neuropathy Screening Instrument is
a brief 15-item questionnaire that can be administered to patients as
a screening tool for neuropathy [74]. Other similar symptom scoring systems
have also been described [81]. Simple visual analog or verbal descriptive
scales may be used to follow patients’ responses to treatment of their
neuropathic symptoms [81–83].
Recently, developments of a number of relatively inexpensive devices
allow suitable assessment of somatosensory function, including vibration,
thermal, light-touch, and pain perception [84]. These types of instruments
allow for cutaneous sensory functions to be assessed noninvasively, and
their measurements are correlated with specific neural fiber function. The
most widely used device in clinical practice is the Semmes-Weinstein
monofilament [85,86]. The filament assesses pressure perception when gentle
pressure is applied to the handle sufficient to buckle the nylon filament.
Although filaments of many different sizes are available, it is the one that
exerts 10-g of pressure that is most commonly used to assess pressure
sensation in the diabetic foot. It is also referred to as the ‘‘5.07 mono-
filament,’’ because during calibration the filaments are calibrated to exert
a force measured in grams that is 10  log of the force exerted at the tip:
hence 5.07 exerts 10-g force. A number of cross-sectional studies have been
conducted that assess the sensitivity of the 10-g monofilament to identify
feet at risk of ulceration. Sensitivities vary from 86% to 100% [76,84,87],
although there is no consensus as to how many sites should be tested. The
commonest algorithm recommends four sites per foot, generally the hallux
and metatarsal heads 1, 3, and 5 [86]. There is little advantage gained,
however, from multiple site assessments [84]. There is also no universal
agreement as to what constitutes an abnormal result (ie, 1, 2, 3, or 4
abnormal results from the sites tested). Despite these problems, the 10-g
monofilament is widely used for the clinical assessment of risk for foot
ulceration but one needs to use 1 g or less to detect neuropathy with a high
sensitivity [84]. A final caution on the use of the filaments: Booth and Young
[88] identified that filaments manufactured by certain companies do not
actually buckle at 10-g force. Indeed, several tested filaments buckled at less
than 8 g. In practice, the authors use 25-lb strain fishing line and cut it into
1000 pieces at a total cost of $5 and provide patients with these to test
themselves at home, using them to assist in behavioral modification, and
have reduced the incidence of foot ulcers by more than 50%.
The graduated Rydel-Seifer tuning fork is used in some centers to assess
neuropathy [89]. This fork uses a visual optical illusion to allow the assessor
to determine the intensity of residual vibration on a 0 to 8 scale at the point
of threshold (disappearance of sensation). Liniger et al [89] reported that
results with this instrument correlated well with other QST measures. The
966 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

tactile circumferential discriminator assesses the perception of calibrated


change in the circumference of a probe (a variation of two-point
discrimination). Vileikyte et al [90] reported a 100% sensitivity in the
identification of patients at risk of foot ulceration. Similarly, this device also
demonstrated good agreement with other measures of QST. Neuropen is
a clinical device that assesses pain using a Neurotip at one end of the ‘‘pen’’
with a 10-g monofilament at the other end. This was shown to be a sensitive
device for assessing nerve function when compared with the simplified nerve
disability score [75].
Quantitative autonomic function tests consist of a series of simple, non-
invasive tests for detecting cardiovascular autonomic neuropathy [53,91].
These tests are based on detection of heart rate and blood pressure response
to a series of maneuvers. Specific tests are used in evaluating disordered re-
gulation of gastrointestinal, genitourinary, pseudomotor function, and peri-
pheral skin blood flow induced by autonomic DN [72].
Biopsy of nerve tissue may be helpful for excluding other causes of
neuropathy and in the determination of predominant pathologic changes in
patients with complex clinical findings as a means of dictating choice of
treatment [42,92]. Skin biopsy has some clinical advantages in diagnosis of
small-fiber neuropathies by quantification of PGP 9.5, when all other
measures are negative [22,93]. Diabetes as the cause of neuropathy is
diagnosed by exclusion of various other causes of neuropathy [32,94]. In
patients presenting with painful feet it has now become apparent that they
may have impaired glucose tolerance [95,96] or the dysmetabolic syndrome
[97]. It has also been used to demonstrate the ability to induce nerve
regeneration [98] and correlates with indices of neuropathy relevant to
function of small unmyelinated C fibers [97]. More recently, Quattrini et al
[99] reported the technique of confocal corneal microscopy in the assessment
of DPN. This is a completely noninvasive technique that offers the future
potential of assessing nerve structure in vivo without the need for biopsy.

Nerve conduction studies


Whole nerve electrophysiologic procedures (eg, NCV, F waves, sensory
or motor amplitudes) have emerged as an important method of tracing the
onset and progression of DPN [100]. An appropriate battery of electro-
physiologic tests supports the measurement of the speed of both sensory and
motor conduction, the amplitude of the propagating neural signal, the
density and synchrony of muscle fibers activated by maximal nerve
stimulation, and the integrity of neuromuscular transmission [80,100]. These
are objective, parametric, noninvasive, and highly reliable measures.
Standard procedures, however, such as maximal NCV, reflect only a limited
aspect of neural activity, and then only in a small subset of large-diameter
and heavily myelinated axons. Even in large-diameter fibers, NCV is
insensitive to many pathologic changes known to be associated with
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 967

DPN. A key role for electrophysiologic assessment, however, is to rule out


other causes of neuropathy or to identify neuropathies superimposed on
DPN. Unilateral conditions, such as entrapments, are far more common in
the patients with diabetes [34]. The principal factors that influence the speed
of NCV are (1) the integrity and degree of myelination of the largest
diameter fibers; (2) the mean cross-sectional diameter of the responding
axons; (3) the representative internodal distance in the segment under study;
and (4) the microenvironment at the nodes, including the distribution of ion
channels. Demyelinating conditions affect conduction velocities, whereas
diabetes primarily reduces amplitudes; the finding of a profound reduction
in conduction velocity strongly supports the occurrence in a diabetic patient
of an alternative condition. Indeed, the odds of occurrence of chronic
inflammatory demyelinating polyneuropathy were 11 times higher among
diabetic than nondiabetic patients [46]. NCV is only gradually diminished
by DPN, with estimates of a loss of approximately 0.5 m/s/y [100]. In a
10-year natural history study of 133 patients with newly diagnosed non–in-
sulin-dependent diabetes mellitus, NCV deteriorated in all six nerve seg-
ments evaluated, but the largest deficit was 3.9 m/s for the sural nerve
(ie, 48.3–44.4 m/s); peroneal motor NCV was decreased by 3 m/s over the
same period [17]. A similar slow rate of decline was demonstrated in the
Diabetes Care and Complications Trial (DCCT). A simple rule is that a 1%
fall in hemoglobin A1c improves conduction velocity about 1.3 m/s [101].
There is, however, a strong correlation (r = 0.74; P \.001) between
myelinated fiber density and whole nerve sural amplitude (Fig. 9) [102].

Management of neuropathy
Once neuropathy is diagnosed, therapy can be instituted with the goal of
ameliorating symptoms and preventing the progression of neuropathy.
Successful management of these syndromes must be geared to the individual
pathogenic processes (Fig. 10).

Control of hyperglycemia
Retrospective and prospective studies have suggested a relationship
between hyperglycemia and the development and severity of DN. Pirart
[7] followed 4400 diabetic patients over 25 years and showed an increase in
prevalence of clinically detectable DN from 12% of patients at the time of
diagnosis of diabetes to almost 50% after 25 years. The highest prevalence
occurred in those people with poorest diabetes control. The DCCT research
Group [14] reported significant effects of intensive insulin therapy on
prevention of neuropathy. The prevalence rates for clinical or electrophys-
iologic evidence of neuropathy were reduced by 50% in those treated by
intensive insulin therapy during 5 years. At that stage of the study, only 3%
of the patients in the primary prevention cohort treated by intensive insulin
therapy showed minimal signs of DN, compared with 10% of those treated
968 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Assess NSS, NDS

Signs and/or symptoms of Signs and/or symptoms of Signs and/or symptoms of


AUTONOMIC NEUROPATHY SMALL FIBER DYSFUNCTION LARGE FIBER DYSFUNCTION

QAFT QST EMG, NCV, QST

AUTONOMIC NEUROPATHY SMALL FIBER NEUROPATHY LARGE FIBER NEUROPATHY

Sensory signs predominate, motor Moderate or severe motor signs and


signs mild or absent symptoms

Family Hx, B12, Folate,


Assess distribution of motor deficit
Lyme ulcer, porphyrins,
Heavy metals, Serum
protein electrophoresis,
Immunoelectrophoresis,
Cytotoxins, Search for If proximal and distal If distal
neoplasm, Immunologic
testing

Family Hx, Immunologic test Anti-GM1 Ab

If non-diabetic neuropathy excluded


If non-diabetic neuropathy excluded
Positive Negative

DISTAL SYMMETRIC NEUROPATHY


DISTAL
DIFFUSE MOTOR NEUROPATHY MOTOR
NEUROPATHY

Anti-GM1 associated DISTAL


MOTOR NEUROPATHY

Fig. 9. A diagnostic algorithm for assessment of neurologic deficit and classification of


neuropathic syndrome is given. NSS, neurologic symptom score; NDS, nerve disability score;
QST, quantitative sensory test; QAFT, quantitative autonomic function test; EMG, electro-
myography; NCV, nerve conduction velocity.

by the conventional regime. In the secondary prevention cohort, intensive


insulin therapy significantly reduced the prevalence of clinical neuropathy
by 56% (7% in intensive insulin therapy group versus 16% in conventional
therapy group). The results of the DCCT study support the necessity for
strict glycemic control, but the effect of insulin as a growth factor and
immunomodulator, aside from its metabolic effects, must also be investi-
gated. In the UK Prospective Diabetes Study, control of blood glucose was
associated with improvement in vibration perception [103–105]. In the
recently described Steno trial [106], a reduction of the odds ratio for the
development of autonomic neuropathy to 0.32 was reported. This was
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 969

Diabetes
Dyslipidemia EFA dysmetabolism
Hyperglycemia
Polyol
pathway DAG
flux Free transition metal PKC β PKC
ions Inhibitor
ARIs
Autoxidation AII
AGE formation ET
Endogenous scavengers
Antioxidants NO
Reactive Oxygen Species
α-Lipoic acid PGI 2
Ischemia/ ONOO- EDHF
A-V shunting reperfusion
Vascular
Nerve and ganglion dysfunction
NEUROPATHY blood flow
Endoneurial hypoxia
Cameron et al., Diabetologia, 2001; 44:1973–88

Fig. 10. Management aimed at pathogenetic mechanisms. (From Cameron NE, Eaton SE,
Cotter MA, Tesfaye S. Vascular factors and metabolic interactions in the pathogenesis of
diabetic neuropathy. Diabetologia 2001;44(11):1973–88; with permission.)

a stepwise, progressive study that involved treatment of type 2 diabetes


patients, with hypotensive drugs, including angiotensin-converting enzyme
inhibitors, Ca2þ channel antagonists, hypoglycemic agents, aspirin, hypo-
lipidemic agents, and antioxidants. These findings argue strongly for the
multifactorial nature of neuropathy and for the need to address the multiple
metabolic abnormalities.

Aldose reductase inhibitors


Aldose reductase inhibitors reduce the flux of glucose through the polyol
pathway, inhibiting tissue accumulation of sorbitol and fructose, and
preventing reduction of redox potentials. In a placebo-controlled double-
blind study of tolrestat, 219 diabetic patients with symmetric polyneurop-
athy, as defined by at least one pathologic cardiovascular reflex, were treated
for 1 year [107]. Patients who received tolrestat showed significant improve-
ment in autonomic function tests and in vibration perception, whereas
placebo-treated patients showed deterioration in most of the parameters
measured [108]. It has now been shown that there is a dose-dependent
improvement in nerve fiber density, particularly small unmyelinated nerve
fibers, in a 12-month study of zenarestat [109]. This was accompanied by an
increase in nerve conduction velocity, albeit the changes in NCV occurred at
a dose of the drug that did not change the nerve fiber density [109]. Impaired
cardiac ejection fractions can be improved with zopolrestat [110]. The
promise shown with the newer aldose reductase inhibitors is being exploited
by at least two other companies and research studies are being done on an
970 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

array of new aldose reductase inhibitors. It is also becoming clear that aldose
reductase inhibition may be insufficient in its own right to achieve the
desirable degree of metabolic enhancement in patients with a multitude of
biochemical abnormalities. Combinations of therapy with aldose reductase
inhibitors and antioxidants may become critical if the relentless progress of
DN is to be abated.

a-Lipoic acid
Lipoic acid (1,2-dithiolane-3-pentanoic acid), a derivative of octanoic
acid, is present in food and is also synthesized by the liver. It is a natural
cofactor in the pyruvate dehydrogenase complex where it binds acyl groups
and transfers them from one part of the complex to another. a-Lipoic acid,
which is also known as thioctic acid, has generated considerable interest as
a thiol replenishing and redox modulating agent. It has been shown to be
effective in ameliorating both the somatic and autonomic neuropathies in
diabetes [111–113]. It is currently undergoing extensive trials in the United
States as both an antidiabetic agent and for the treatment of DN.

c-Linolenic acid
Linoleic acid, an essential fatty acid, is metabolized to dihomo-c-linolenic
acid, which serves as an important constituent of neuronal membrane
phospholipids, and also serves as a substrate for prostaglandin E formation,
seemingly important for preservation of nerve blood flow. In diabetes,
conversion of linoleic acid to c-linolenic acid and subsequent metabolites is
impaired, possibly contributing to the pathogenesis of DN [114]. A recent
multicenter double-blind placebo-controlled trial using c-linolenic acid for
1 year demonstrated significant improvements in both clinical measures and
electrophysiologic testing [115].

Aminoguanidine
Animal studies using aminoguanidine, an inhibitor of the formation of
advanced glycosylation end-products, show improvement in nerve conduc-
tion velocity in streptozotocin-induced DN in rats. Controlled clinical trials
to determine its efficacy in humans [116,117] have been discontinued because
of toxicity. There are, however, successors to aminoguanidine that hold
promise for this approach [118].

Human intravenous immunoglobulin


Immune intervention with intravenous immunoglobulin has become
appropriate in some patients with forms of peripheral DN that are
associated with signs of antineuronal autoimmunity [43,51]. Treatment with
immunoglobulin is well tolerated and is considered safe, especially with
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 971

respect to viral transmission [119]. The major toxicity of intravenous


immunoglobulin has been an anaphylactic reaction, but the frequency of
these reactions is now low and confined mainly to patients with immuno-
globulin (usually IgA) deficiency. Patients may experience severe headache
caused by aseptic meningitis, which resolves spontaneously. In some
instances, it may be necessary to combine treatment with prednisone or
azathioprine. Relapses may occur, requiring repeated courses of therapy.

Neurotrophic therapy
There is now considerable evidence in animal models of diabetes that
decreased expression of nerve growth factor and its receptors, trk A, reduces
retrograde axonal transport of nerve growth factor and diminishes support
of small unmyelinated neurons and their neuropeptides, such as substance P
and calcitonin gene-related peptide, both potent vasodilators [120,121].
Furthermore, recombinant human nerve growth factor administration
restores these neuropeptide levels toward normal and prevents the mani-
festations of sensory neuropathy in animals [122]. In a 15-center, double-
blind, placebo-controlled study of the safety and efficacy of recombinant
human nerve growth factor in 250 subjects with symptomatic small-fiber
neuropathy [18], recombinant human nerve growth factor improved the
neurologic impairment score of the lower limbs, and improved small nerve
fiber function cooling threshold (A delta-fibers) and the ability to perceive
heat pain (C fiber) compared with placebo. These results were consistent
with the postulated actions of nerve growth factor on trk A receptors
present on small-fiber neurons. This led to two large multicenter studies
conducted in the United States and the rest of the world. Results of these
two studies were presented at the American Diabetes Association meetings
in June 1999 [19]. Regrettably, recombinant human nerve growth factor was
not found to have beneficial effects over and above placebo. The reason for
this dichotomy has not been resolved, but this has somewhat dampened the
enthusiasm for growth factor therapy of DN.

Management aimed at symptoms


Pain control
Control of pain constitutes one of the most difficult management issues in
DN. In essence, simple measures are tried first. If no distinction is made for
pain syndromes then the numbers needed to treat to reduce pain by 50% are
1.4 for optimal dose tricyclic antidepressants, 1.9 for dextromethorphan, 3.3
for carbamazepine, 3.4 for tramadol, 3.7 for gabapentin, 5.9 for capsaicin,
6.7 for selective serotonin reuptake inhibitors, and 10 for mexiletine [123]. If,
however, pain is divided according to its derivation from different nerve
fiber type (A delta versus C fiber), spinal cord, or cortical, then different
types of pain respond to different therapies (Fig. 11), as described next.
972 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

TCAs Descending fibres


SSRIs
α2 antags 5HT
Tramadol Opioid
α2
Substance P
DRG
Ca channels
Substantia
C-fiber NMDA
Glutamate AMPA gelatinosa
mGlur
Glutamate Capsaicin

Aδ fiber GABA GABAA


GABAB
Na channels
Interneuron
Topiramate
Carbamazepine Dextromethorphan
TCAs Topiramate
Insulin Gabapentin

Fig. 11. Pain derived from different nerve fiber types.

C-fiber pain
Initially, when there is ongoing damage to the nerves, the patient
experiences the pain of the burning, lancinating, dysesthetic type often
accompanied by hyperalgesia and allodynia. Because the peripheral sympa-
thetic nerve fibers are also small unmyelinated C fibers, sympathetic blocking
agents (clonidine) may improve the pain. Loss of sympathetic regulation of
sweat glands and arteriovenous shunt vessels in the foot creates a favorable
environment for bacteria to penetrate, multiply, and wreak havoc with the
foot. These fibers use the neuropeptide substance P as their neurotransmitter,
and depletion of axonal substance P (capsaicin) often leads to amelioration
of the pain. When the destructive forces persist, however, the individual
becomes pain free and develops impaired warm temperature and pain
thresholds. Disappearance of pain in these circumstances should be hailed as
a warning that the neuropathy is progressing.

Capsaicin. Capsaicin is extracted from chili peppers, and a simple cheap


mixture is to add one to three teaspoons of cayenne pepper to a jar of cold
cream and apply to the area of pain. It has high selectivity for a subset of
sensory neurons, which have been identified as unmyelinated C fiber afferent
or thin-myelinated (A delta) fibers. Prolonged application of capsaicin
depletes stores of substance P, and possibly other neurotransmitters, from
sensory nerve endings. This reduces or abolishes the transmission of painful
stimuli from the peripheral nerve fibers to the higher centers [124]. Care
must be taken to avoid eyes and genitals, and gloves must be worn. Because
of capsaicin’s volatility it is safer to cover affected areas with plastic wrap.
There is initial exacerbation of symptoms followed by relief in 2 to 3 weeks.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 973

Clonidine. There is an element of sympathetic-mediated C-fiber type pain


that can be overcome with clonidine (a2-adrenergic agonist) or phentol-
amine. Clonidine can be applied topically [125], but the dose titration may
be more difficult. If clonidine fails, the local anesthetic agent mexiletine
warrants a trial. Unresponsive patients are treated as outlined in Fig. 9.

A delta-fiber pain
A delta-fiber pain is a more deep-seated, dull, and gnawing ache, which
often does not respond to the previously described measures. A number of
different agents have been used for the pain associated with these fibers with
varying success.

Insulin. Continuous intravenous insulin infusion without resort to blood


glucose lowering may be useful in these patients. A response with reduction
of pain usually occurs within 48 hours [68], and the insulin infusion can be
discontinued. If this measure fails there are several medications available
that may abolish the pain.

Nerve blocking. Lidocaine given by slow infusion has been shown to


provide relief of intractable pain for 3 to 21 days. This form of therapy
may be of most use in self-limited forms of neuropathy. If successful,
therapy can be continued with oral mexiletine. These compounds target the
pain caused by hyperexcitability of superficial, free nerve endings [126].

Tramadol and dextromethorphan. There are two possible targeted therapies.


Tramadol is a centrally acting weak opioid analgesic for use in treating
moderate to severe pain. Tramadol was shown to be better than placebo in
a randomized controlled trial [127] of only 6-weeks’ duration, but a sub-
sequent follow-up study suggested that symptomatic relief could be main-
tained for at least 6 months [128]. Side-effects are, however, relatively
common, and are similar to other opioid-like drugs. Another spinal cord
target for pain relief is the excitatory glutaminergic N-methyl-D-aspartate
receptor. Blockade of N-methyl-D-aspartate receptors is believed to be one
mechanism by which dextromethorphan exerts analgesic efficacy [129]. An
accomplished pharmacist can procure a sugar-free solution of dextro-
methorphan.

Antidepressants. Clinical trials have focused on interrupting pain trans-


mission using antidepressant drugs that inhibit the reuptake of norepineph-
rine or serotonin. This central action accentuates the effects of these
neurotransmitters in activation of endogenous pain-inhibitory systems in
the brain that modulate pain-transmission cells in the spinal cord [130]. Side
effects, including dysautonomia and dry mouth, can be troublesome.
Switching to nortriptyline may lessen some of the anticholinergic effects of
amitriptyline.
974 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Selective serotonin reuptake inhibitors inhibit presynaptic reuptake of


serotonin but not norepinephrine. Studies suggest that treatment with
paroxetine (Sindrup) but not fluoxetine (Max) is associated with significant
pain relief. These drugs should, however, be used with caution in diabetic
patients who may be on other medications, because there is a suggestion that
selective serotonin-reuptake inhibitors might increase the risk of upper
gastrointestinal bleeding [131].

Carbamazepine. Several double-blind placebo-controlled studies have dem-


onstrated carbamazepine to be effective in the management of pain in DN
[32]. Toxic side effects may limit its use in some patients. It is very useful,
however, for those patients with lightning or shooting pain.

Phenytoin. Diphenylhydantoin has long been used in the treatment of pain-


ful neuropathies. Double-blind crossover studies do not demonstrate a the-
rapeutic benefit of phenytoin compared with placebo in DN [132]. Also, side
effects mitigate its use in people with diabetes. Its ability to suppress insulin
secretion has resulted in precipitation of hyperosmolar diabetic coma.

Gabapentin. Gabapentin is an effective anticonvulsant whose mechanism is


not well understood, yet holds additional promise as an analgesic agent in
painful neuropathy [133]. In a multicenter study in United States [62],
gabapentin monotherapy seemed to be efficacious for the treatment of pain
and sleep interference associated with diabetic peripheral neuropathy. It also
exhibits positive effects on mood and quality of life [134]. Effective dosing
may require 1800 to 3600 mg/d and this is associated with untoward side
effects. Perhaps one of the most disconcerting features is the weight gain
associated with long-term use.

Lamotrigine. Lamotrigine is an antiepileptic agent with at least two


antinociceptive properties. A randomized placebo-controlled study [135]
confirmed the efficacy of this agent in patients with neuropathic pain.
Titration needs to be inordinately slow, however, to avoid Stevens-Johnson
syndrome and bradycardia has been reported.

Topiramate. Topiramate is a fructose analog that was initially examined


because of its antidiabetic possibilities. Unfortunately, it was first examined
only in normal animals and had no hypoglycemic properties. It has now
undergone extensive testing for epilepsy, migraine, involuntary movements,
central nervous system injury, and neuropathic pain. Unfortunately, the first
two studies used a titration to 400 mg/d, which was associated with fairly
severe central nervous system side effects, which were prohibitive. The
studies failed to establish an effect in diabetic neuropathic pain. A third
study using different end points with specificity for the nature and site of the
pain and recognizing that a side effect of the drug namely, paresthesias, was
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 975

not mistaken for pain was successful [136]. What has emerged from all the
studies is that the drug lowers blood pressure, improves lipid profiles,
decreases insulin resistance, and increases nerve fiber regeneration in the
skin [97]. It has the potential to relieve pain by altering the biology of the
disease and has now been shown to increase intraepidermal nerve fibers.
Further trials are being done. One must start with no more than 15 mg/d,
preferably at night, and then increase the dose only after the patient can
tolerate the drug. A maximum of 200 mg was sufficient to induce nerve fiber
recovery.

Transcutaneous nerve stimulation (electrotherapy), magnetic field therapy,


infrared light, and electrical cord stimulation. Transcutaneous nerve stimu-
lation (electrotherapy) occasionally may be helpful and certainly represents
one of the more benign therapies for painful neuropathy [137]. Care should be
taken to move the electrodes around to identify sensitive areas and obtain
maximal relief. Static magnetic field therapy [138] has been reported to be of
benefit but it is difficult to blind such studies. Similarly, the use of infrared light
has reportedly had benefit but this remains to be proved. A case series of
patients with severe painful neuropathy unresponsive to conventional therapy
suggested efficacy of using an implanted spinal cord stimulator [139]. This
cannot be generally recommended, however, except in very resistant cases
because it is invasive, expensive, and unproved in controlled studies.

Analgesics. Analgesics are rarely of much benefit in the treatment of painful


neuropathy, although they may be of some use on a short-term basis for
some of the self-limited syndromes, such as painful diabetic third nerve
palsy. Use of narcotics in the setting of chronic pain generally is avoided
because of the risk of addiction.

Calcitonin. In a placebo-controlled study, 10 patients with painful DN were


treated with 100 IU of calcitonin per day. About 39% of patients had near-
complete relief of symptom. The improvement was seen after only 2 weeks
of treatment [140].

Management of small-fiber neuropathies


Management of small-fiber neuropathies includes the following:
 Patients must be instructed on foot care with daily foot inspection
 They must have a mirror in the bathroom for inspection of the soles of
the feet
 Providing patients with a monofilament for self-testing reduces ulcers
 All diabetic patients should wear padded socks
 Shoes must fit well with adequate support and must be inspected for the
presence of foreign bodies (eg, nails, pins, teeth) before donning (ie,
examine the feet and the shoes daily)
976 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

 Patients must exercise care with exposure to heat (no falling asleep in
front of fires)
 Emollient creams should be used for the drying and cracking
 After bathing feet should be thoroughly dried and powdered between
the toes
 Nails should be cut transversely, preferably by a podiatrist

Management of large-fiber neuropathies


Patients with large-fiber neuropathies are incoordinate and ataxic. As
a result, they are more likely to fall than nonneuropathic age-matched
people [141]. It has recently been demonstrated that high-intensity strength
training in older people increases muscle strength in a variety of muscles.
More importantly, the strength training resulted in improved coordination
and balance quantifiable with backward tandem walking [142]. It is vital to
embark on a program of strength training and improvement of balance.
Management of large-fiber neuropathies includes the following:
 Gait and strength training
 Pain management as detailed previously
 Orthotics should be fitted with proper shoes for the deformities
 Tendon lengthening for Achilles tendon shortening
 Bisphosphonates may be given for osteopenia
 Surgical reconstruction and full length casting as necessary

Autonomic neuropathies
The autonomic nervous system supplies all organs in the body and
consists of an afferent and an efferent system, with long efferents in the
vagus (cholinergic) and short postganglionic unmyelinated fibers in the
sympathetic system (adrenergic). A third component is the neuropeptidergic
system with its neurotransmitters substance P, vasoactive intestinal poly-
peptide, and calcitonin gene-related peptide among others. Diabetic auto-
nomic neuropathy can cause dysfunction of every part of the body. Diabetic
autonomic neuropathy often goes completely unrecognized by patient and
physician alike because of its insidious onset and protean multiple organ
involvement. Alternatively, the appearance of complex and confusing
symptoms in a single organ system because of diabetic autonomic neurop-
athy may cause profound symptoms and receive intense diagnostic and
therapeutic attention. Subclinical involvement may be widespread, whereas
clinical symptoms and signs may be focused within a single organ. The
organ systems that most often exhibit prominent clinical autonomic signs
and symptoms in diabetes include the ocular pupil, sweat glands, genito-
urinary system, gastrointestinal tract system, adrenal medullary system, and
the cardiovascular system (Box 1).
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 977

Box 1. Clinical manifestations of autonomic neuropathy


Cardiovascular
Tachycardia, exercise intolerance
Cardiac denervation, painless myocardial infarction
Orthostatic hypotension
Heat intolerance
Alterations in skin blood flow
Gastrointestinal
Esophageal dysfunction
Gastroparesis diabeticorum
Diarrhea
Constipation
Fecal incontinence
Genitourinary
Erectile dysfunction
Retrograde ejaculation
Cystopathy
Neurogenic bladder
Sweating disturbances
Areas of symmetrical anhidrosis
Gustatory sweating
Metabolic
Hypoglycemia unawareness
Hypoglycemia unresponsiveness
Pupillary
Decreased diameter of dark adapted pupil
Argyll-Robertson–type pupil

Involvement of the autonomic nervous system can occur as early as the


first year after diagnosis and major manifestations are cardiovascular,
gastrointestinal, and genitourinary system dysfunction [32,143]. Reduced
exercise tolerance, edema, paradoxical supine or nocturnal hypertension, and
intolerance to heat because of defective thermoregulation are a consequence
of autonomic neuropathy. Defective blood flow in the small capillary
circulation is found with decreased responsiveness to mental arithmetic,
cold pressor, hand grip, and heating [55]. The defect is associated with
a reduction in the amplitude of vasomotion [144] and resembles premature
aging [55]. There are differences in the glabrous and hairy skin circulations.
In hairy skin a functional defect is found before the development of
neuropathy [145] and is correctable with antioxidants [146]. The clinical
counterpart is a dry cold skin, loss of sweating, and development of fissures
and cracks that are portals of entry for organisms leading to infectious ulcers
978 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

and gangrenes. Silent myocardial infarction, respiratory failure, amputa-


tions, and sudden death are hazards for diabetic patients with cardiac
autonomic neuropathy [25,147]. It is vitally important to make this diagnosis
early so that appropriate intervention can be instituted [148].
Disturbances in autonomic nervous system may be functional (eg, gas-
troparesis with hyperglycemia and ketoacidosis) or organic wherein nerve
fibers are actually lost. This creates inordinate difficulties in diagnosing,
treating, and prognosticating and establishing true prevalence rates. Tests of
autonomic function generally stimulate entire reflex pathways. Furthermore,
autonomic control for each organ system is usually divided between opposing
sympathetic and parasympathetic innervation, so that heart rate accelera-
tion, for example, may reflect either decreased parasympathetic or increased
sympathetic nervous system stimulation. Because many conditions affect the
autonomic nervous system and autonomic neuropathy is not unique to
diabetes, the diagnosis of diabetic autonomic neuropathy rests with establish-
ing the diagnosis and excluding other causes. The best studied and for which
there are large databases and evidence to support their use in clinical prac-
tice relate to the evaluation of cardiovascular reflexes. In addition, the
evaluation of orthostasis is fairly straightforward and is readily done in
clinical practice as is the establishment of the cause of gastrointestinal
symptoms and erectile dysfunction. The evaluation of pupillary abnormal-
ities, hypoglycemia unawareness and unresponsiveness, neurovascular dys-
function, and sweating disturbances are for the most part done only in
research laboratories, require specialized equipment and familiarity with the
diagnostic procedures, and are best left in the hands of those who have
a special interest in the area.
Tables 4 and 5 present the diagnostic tests applicable to the diagnosis of
cardiovascular autonomic neuropathy. These tests can be used as a surrogate
for the diagnosis of autonomic neuropathy of any system because it is
generally rare to find involvement (although it does occur) of any other
division of the autonomic nervous system in the absence of cardiovascular
autonomic dysfunction. For example, if one entertains the possibility that
the patient has erectile dysfunction caused by autonomic neuropathy, then
before embarking on a sophisticated and expensive evaluation of erectile
status a measure of heart rate and its variability in response to deep
breathing if normal exclude the likelihood that the erectile dysfunction is
a consequence of disease of the autonomic nervous system and the cause
thereof has to be sought elsewhere. Similarly, it is extremely unusual to find
gastroparesis secondary to autonomic neuropathy in a patient with normal
cardiovascular autonomic reflexes (Figs. 12 and 13).

Prevention and reversibility of autonomic neuropathy


It has now become clear that strict glycemic control [15] and a stepwise
progressive management of hyperglycemia, lipids, blood pressure, and use
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 979

Table 4
Differential diagnosis of diabetic autonomic neuropathy
Clinical manifestations Differential diagnosis
Cardiovascular Cardiovascular disorders
Tachycardia, exercise intolerance Idiopathic orthostatic hypotension,
Cardiac denervation, painless multiple system atrophy with
myocardial infarction Parkinsonism, orthostatic tachycardia,
Orthostatic hypotension hyperadrenergic hypotension
Shy-Drager syndrome
Panhypopituitarism
Pheochromocytoma
Hypovolemia
Congestive heart disease
Carcinoid syndrome
Gastrointestinal Gastrointestinal disorders
Esophageal dysfunction Obstruction
Gastroparesis diabeticorum Bezoars
Diarrhea Secretory diarrhea (endocrine tumors)
Constipation Biliary disease
Fecal incontinence Psychogenic vomiting
Medications
Genitourinary Genitourinary
Erectile dysfunction Genital and pelvic surgery
Retrograde ejaculation Atherosclerotic vascular disease
Cystopathy Medications
Neurogenic bladder Alcohol abuse
Neurovascular Other causes of neurovascular dysfunction
Heat intolerance Chagas’ disease
Gustatory sweating Amyloidosis
Dry skin Arsenic
Impaired skin blood flow
Metabolic Metabolic
Hypoglycemia unawareness Other causes of hypoglycemia,
Hypoglycemia unresponsiveness intensive glycemic control and
Hypoglycemiassociated autonomic failure drugs that mask hypoglycemia
Pupillary Pupillary
Decreased diameter of dark adapted pupil Syphilis
Argyll-Robertson–type pupil

of antioxidants [112] and angiotensin-converting enzyme inhibitors [149]


reduce the odds ratio for autonomic neuropathy to 0.32 [106]. It has also
been shown that mortality is a function of loss of beat-to-beat variability
with myocardial infarction. This can be reduced by 33% with acute
administration of insulin [150]. Kendall et al [151] reported that successful
pancreas transplantation improves epinephrine response and normalizes
hypoglycemia symptom recognition in patients with long-standing diabetes
and established autonomic neuropathy. Burger et al [152] showed that
a reversible metabolic component of cardiovascular autonomic neuropathy
exists in patients with early cardiovascular autonomic neuropathy.
980 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Table 5
Diagnostic tests of cardiovascular autonomic neuropathy
Test Method and parameters
Resting heart rate; beat-to-beat >100 beats/min is abnormal. With the patient at rest
heart rate; variation and supine (no overnight coffee or hypoglycemic
episodes), breathing 6 breaths/min, heart rate
monitored by EKG or ANSCORE device, a
difference in heart rate of [15 beats/min is normal
and \10 beats/min is abnormal, R-R inspiration/
R-R expiration [1.17. All indices of HRV are
age-dependent.
Heart rate response to standing During continuous EKG monitoring, the R-R
interval is measured at beats 15 and 30 after
standing. Normally, a tachycardia is followed by
reflex bradycardia. The 30:15 ratio is normally
[1.03.
Heart rate response to The subject forcibly exhales into the mouthpiece of
Valsalva’s maneuver a manometer to 40 mm Hg for 15 seconds during
EKG monitoring. Healthy subjects develop tachy
cardia and peripheral vasoconstriction during
strain and an overshoot bradycardia and rise in
blood pressure with release. The ratio of longest
R-R shortest R-R should be [1.2.
Systolic blood pressure Systolic blood pressure is measured in the supine
response to standing subject. The patient stands and the systolic blood
pressure is measured after 2 min. Normal response
is a fall of \10 mm Hg, borderline is a fall of
10–29 mm Hg, and abnormal is a fall of
[30 mm Hg with symptoms.
Diastolic blood pressure response The subject squeezes a handgrip dynamometer to
to isometric exercise establish a maximum. Grip is then squeezed at
30% maximum for 5 min. The normal response for
diastolic blood pressure is a rise of [16 mm Hg in
the other arm.
EKG QT-QTc intervals The QTc (corrected QT interval on EKG) should
be \440 ms.
Spectral analysis VLF peak # (sympathetic dysfunction)
LF peak# (sympathetic dysfunction)
HF peak # (parasympathetic dysfunction)
LH/HF ratio # (sympathetic imbalance)
Neurovascular flow Using noninvasive laser Doppler measures of
peripheral sympathetic responses to nociception.

Management of autonomic neuropathy


Postural hypotension
The syndrome of postural hypotension is posture-related dizziness and
syncope. Patients who have type 2 diabetes mellitus and orthostatic
hypotension are hypovolemic and have sympathoadrenal insufficiency; both
factors contribute to the pathogenesis of orthostatic hypotension [153].
Postural hypotension in the patient with diabetic autonomic neuropathy can
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 981

Postural Dizziness
Measure Blood Pressure and
heart rate supine and
standing

Diagnosis =Vaso-Vagal
Evaluate Heart Rate Decrease Syncope

Increase or No Change in Heart Rate

No Change or Slight Vestibular System


Evaluate Diastolic Blood Pressure Increase Abnormality

Decrease in Systolic Blood Pressure


(more than 30 mmHg)

Evaluate Plasma Norepinephrine Supraphysiologic Decreased


Response to standing Response Circulatory Volume

Decreased or Normal Response

Impaired Sympathetic Vascular Nervous


System

Fig. 12. The evaluation of postural dizziness in diabetic patients.

present a difficult management problem. Elevating the blood pressure in the


standing position must be balanced against preventing hypertension in the
supine position.

Supportive garments. Whenever possible, attempts should be made to


increase venous return from the periphery using total body stockings. Leg
compression alone is less effective, presumably reflecting the large capacity
of the abdomen relative to the legs [154]. Patients should be instructed to put
them on while lying down and not to remove them until returning to the
supine position.
982 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Presentation
Nausea,vomiting,
bloating, satiety,
brittle diabetes

History of drugs Nutritional and


e.g. opioids, tricyclic Exclude functional psychiatric evaluation
antidepressants, e.g. hyperglycemia, for eating disorders,
anticholinergics, levodopa, ketoacidosis, e.g. anorexia nervosa,
Ca++ antagonists, octreotide electrolyte imbalance bulimia,
endocrine e.g.
hypo and hyperthyroidism,
Addisons.

al Solid phase gastric emptying No


orm rm
al
Abn

Gastroparesis: Psychologic cause

Fig. 13. The evaluation of the patient suspected of gastroparesis.

Drug therapy. Some patients with postural hypotension may benefit from
treatment with 9-flurohydrocortisone (Table 6). Unfortunately, symptoms
do not improve until edema occurs, and there is a significant risk of
developing congestive heart failure and hypertension. If fluorohydrocorti-
sone does not work satisfactorily, various adrenergic agonists and antago-
nists may be used. If the adrenergic receptor status is known, then therapy
can be guided to the appropriate agent. Metoclopramide may be helpful in
patients with dopamine excess or increased sensitivity to dopaminergic
stimulation. Patients with a2-adrenergic receptor excess may respond to the
a2-antagonist yohimbine. Those few patients in whom b-receptors are
increased may be helped with propranolol. a2-Adrenergic receptor deficiency
can be treated with the a2-agonist clonidine, which in this setting may
paradoxically increase blood pressure. One should start with small doses and
gradually increase the dose. If the preceding measures fail, midodrine, a a1-
adrenergic agonist, or dihydroergotamine in combination with caffeine may
help. A particularly refractory form of postural hypotension occurs in some
patients postprandially and may respond to therapy with octreotide given
subcutaneously in the mornings.

Gastropathy
Gastrointestinal motor disorders are frequent and widespread in type 2
diabetic patients regardless of symptoms [155] and there is a poor correlation
Table 6
Pharmacologic treatment of autonomic neuropathy
Clinical status Drug Dosage Side effects

A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999


Orthostatic hypotension 9a fluorohydrocortisone, mineralocorticoid 0.5–2 mg/d Congestive heart failure, hypertension
Clonidine, a2-adrenergic agonist 0.1–0.5 mg, at bedtime Hypotension, sedation, dry mouth
Octreotide, somatostatin analogue 0.1–0.5 lg/kg/d Injection site pain, diarrhea
Gastroparesis Metoclopramiede, D2-receptor antagonist 10 mg, 30–60 min before meal Galactorrhea, extrapyramidal symptoms
diabeticorum and bedtime
Domperidon, D2-receptor antagonist 10–20 mg, 30–60 min before meal Galactorrhea
and bedtime
Erythromycin, motilin receptor agonist 250 mg, 30 min before meals Abdominal cramp, nausea, diarrhea, rash
Levosulpide, D2-receptor antagonist 25 mg, 3 times/d Galactorrhea
Diabetic diarrhea Metranidazole, broad-spectrum antibiotics 250 mg, 3 times/d,
minimum 3 wk
Clonidine, a2-adrenergic agonist 0.1 mg, 2–3 times/d Orthostatic hypotension
Cholestyramine, bile acid sequestrant 4 g, 1–6 times/d
Loperamide, opiate-receptor agonists 2 mg, four times/d Toxic megacolon
Octreotide, somatostatin analogue 50 lg, 3 times/d Aggravate nutrient malabsorption
(at higher doses)
Cystopathy Bethanechol, acetylcholine receptor agonist 10 mg, 4 times/d
Doxazosin, a1-adrenergic antagonist 1–2 mg, 2–3 times/d Hypotension, headache, palpitation
Erectile dysfunction Vardenafil, tadalafil, cGMP type-5 10–50 mg before sexual activity, Hypotension and fatal cardiac event
phosphodiesterase inhibitors only once per day (with nitrate-containing drugs),
headache, flushing, nasal congestion,
dyspepsia, muscoloskeletal pain,
blurred vision

983
984 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

between symptoms and objective evidence of functional or organic defects.


The first step in management of diabetic gastroparesis consists of multiple,
small feedings. The amount of fat should be decreased, because it tends to
delay gastric emptying. Maintenance of glycemic control is important
[156,157]. Metoclopramide may be used. Cisapride and domperidone
[158,159] have been shown to be effective in some patients, although
probably no more so than metoclopramide. Cisapride has, however, been
withdrawn from the market. Erythromycin given as either a liquid or
suppository also may be helpful. Erythromycin acts on the motilin receptor,
‘‘the sweeper of the gut,’’ and shortens gastric emptying time [160]. If
medications fail and severe gastroparesis persists, jejunostomy placement
into normally functioning bowel may be needed.

Enteropathy
Enteropathy involving the small bowel and colon can produce both
chronic constipation and explosive diabetic diarrhea, making treatment of
this particular complication difficult.

Antibiotics. Stasis of bowel contents with bacterial overgrowth may


contribute to the diarrhea. Treatment with broad-spectrum antibiotics is
the mainstay of therapy, including tetracycline or trimethoprim and
sulfamethoxazole. Metronidazole seems to be the most effective and should
be continued for at least 3 weeks.

Cholestyramine. Retention of bile may occur and can be highly irritating to


the gut. Chelation of bile salts with cholestyramine, 4 g three times a day,
mixed with fluid may offer relief of symptoms.

Diphenoxylate plus atropine. Diphenoxylate plus atropine may help to


control the diarrhea. Toxic megacolon can occur, however, and extreme
care should be used.

Diet. Patients with poor digestion may benefit from a gluten-free diet.
Beware of certain fibers in the neuropathic patient that can lead to bezoar
formation because of bowel stasis in gastroparetic or constipated patients.

Cystopathy
In diabetic autonomic neuropathy, the motor function of the bladder is
unimpaired, but afferent fiber damage results in diminished bladder
sensation. The urinary bladder can be enlarged to more than three times its
normal size. Patients are seen with bladders filled to their umbilicus, yet they
feel no discomfort. Loss of bladder sensation occurs with diminished voiding
frequency, and the patient is no longer able to void completely. Conse-
quently, dribbling and overflow incontinence are common complaints. A
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 985

postvoiding residual of greater than 150 mL is diagnostic of cystopathy.


Cystopathy may put the patients at risk for urinary infections.

Treatment of cystopathy. Patients with cystopathy should be instructed to


palpate their bladder and, if they are unable to initiate micturition when
their bladders are full, use Credé’s maneuver (massage or pressure on the
lower portion of abdomen just above the pubic bone) to start the flow of
urine. The principal aim of the treatment should be to improve bladder
emptying and to reduce the risk of urinary tract infection. Parasympatho-
mimetics, such as bethanechol, are sometimes helpful, although frequently
they do not help fully to empty the bladder. Extended sphincter relaxation
can be achieved with an a1-blocker, such as doxazosin [32]. Self-catheter-
ization can be particularly useful in this setting, with the risk of infection
generally being low.

Sexual dysfunction
Erectile dysfunction occurs in 50% to 75% of diabetic men, and it tends to
occur at an earlier age than in the general population. The incidence of
erectile dysfunction in diabetic men aged 20 to 29 years is 9% and increases to
95% by age 70. It may be the presenting symptom of diabetes. More than
50% notice the onset of erectile dysfunction within 10 years of the diagnosis,
but it may precede the other complications of diabetes. The etiology of
erectile dysfunction in diabetes is multifactorial. Neuropathy, vascular
disease, diabetes control, nutrition, endocrine disorders, psychogenic factors,
and drugs used in the treatment of diabetes and its complications play a role
[161,162]. The diagnosis of the cause of erectile dysfunction is made by
a logical stepwise progression [161,162] in all instances. An approach to
therapy has recently been presented to which the reader is referred (Fig. 14)
[161].
A thorough work-up for impotence includes medical and sexual history;
physical and psychologic evaluations; blood test for diabetes and a check of
levels of testosterone, prolactin, and thyroid hormones; test for nocturnal
erections; tests to assess penile, pelvic, and spinal nerve function; and test to
assess penile blood supply and blood pressure. The flow chart provided is
intended as a guide to assist in defining the problem (see Fig. 14).
The health care provider should initiate questions that help distinguish
the various forms of organic erectile dysfunction from those that are
psychogenic in origin. Physical examination must include an evaluation of
the autonomic nervous system, vascular supply, and the hypothalamic-
pituitary-gonadal axis.
Autonomic neuropathy causing erectile dysfunction is almost always
accompanied by loss of ankle jerks and absence or reduction of vibration
sense over the large toes. More direct evidence of impairment of penile
autonomic function can be obtained by demonstrating normal perianal
sensation, assessing the tone of the anal sphincter during a rectal examination,
986 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

Evaluation of Diabetic Patients with Erectile Dysfunction

Drugs; antihypertensive, antidepressive, tranquilizers

Trauma

Sexual development /androgenization


History Sexual function, onset, all partners, morning erections

Autonomic nerve functions

Vascular status

Sexual development
Penis, testes, scrotum, visual fields, breasts, hair
Testosterone, prolactin
Physicial
Somatic and autonomic nerve function
Perianal sensation
Anal wink
Bulbocavernous reflex
Expiration/ inspiration ratio
Trial of oral Vascular status
agents Pulses
Penile / brachial index

Response No response
sildenafil,
vardenafil
and
tadalafil Abnormal NPT Normal

Intracavernosal injection vasodilator Psychogenic

Erection No erection

Neuropathic Vascular

Injections, Vacuum device,


Sildenafil,Vardenafil, prostheses

Fig. 14. The evaluation of the patient with erectile dysfunction.


A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 987

and ascertaining the presence of an anal wink when the area of the skin
adjacent to the anus is stroked or contraction of the anus when the glans penis
is squeezed (ie, the bulbo-cavernosus reflex). These measurements are easily
and quickly done at the bedside and reflect the integrity of sacral para-
sympathetic divisions.
Vascular disease is usually manifested by buttock claudication but may
be caused by stenosis of the internal pudendal artery. A penile-brachial
index of less than 0.7 indicates diminished blood supply. A venous leak
manifests as unresponsiveness to vasodilators and needs to be evaluated by
penile Doppler sonography.
To distinguish psychogenic from organic erectile dysfunction nocturnal
penile tumescence can be done. Normal nocturnal penile tumescence defines
psychogenic erectile dysfunction, and a negative response to vasodilators
implies vascular insufficiency. Application of nocturnal penile tumescence is
not so simple. It is much like having a sphygmomanometer cuff inflate over
the penis many times during the night while one is trying to have a normal
night’s sleep and the rapid eye movement sleep associated with erections.
The individual may have to take the device home and become familiar with
it over several nights before one has a reliable estimate of the failure of
nocturnal penile tumescence.

Treatment of erectile dysfunction. A number of treatment modalities are


available and each treatment has positive and negative effects; patients must
be made aware of both aspects before a therapeutic decision is made. Before
considering any form of treatment, every effort should be made to have the
patient withdraw from alcohol and eliminate smoking. First and foremost,
the patient should be removed, if possible, from drugs that are known to
cause erectile dysfunction. Metabolic control should be optimized.
According to more recent research, relaxation of the corpus cavernosum
smooth-muscle cells is caused by nitric oxide and cGMP, and the ability to
have and maintain an erection depends on nitric oxide and cGMP. Sildenafil,
vardenafil, and tadalafil exert their effect by transiently increasing nitric
oxide and cGMP levels. These compounds are cGMP type-5 phosphodies-
terase inhibitors that enhance blood flow to the corpora cavernosa with
sexual stimulation. Orally taken a 10- to 50-mg tablet is the usual starting
dose, 60 minutes before sexual activity. Lower doses should be considered in
patients with renal failure and hepatic dysfunction. The duration of the drug
effects is 3 to 48 hours. Before they are prescribed, it is important to exclude
ischemic heart disease. They are absolutely contraindicated in patients being
treated with nitroglycerine or other nitrate-containing drugs. Severe hypo-
tension and fatal cardiac events can occur [163].
Direct injection of prostacyclin into the corpus cavernosum induces
satisfactory erections in a significant number of men. Surgical implantation
of a penile prosthesis may be appropriate. The less expensive type of
prosthesis is a semi-rigid, permanently erect type that may be embarrassing
988 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

and uncomfortable for some patients. The inflatable type is three times more
expensive and subject to mechanical failure, but it avoids the embarrassment
caused by other devices.

Female sexual dysfunction. Women with diabetes mellitus may experience


decreased sexual desire and more pain on sexual intercourse, but they are at
risk of decreased sexual arousal, with inadequate lubrication [164]. Di-
agnosis of female sexual dysfunction using vaginal plethysmography to
measure lubrication and vaginal flushing has not been well established.

Sweating disturbances
Hyperhidrosis of the upper body, often related to eating (gustatory
sweating), and anhidrosis of the lower body is a characteristic feature of
autonomic neuropathy. Gustatory sweating accompanies the ingestion of
certain foods, particularly spicy foods, and cheeses. Gustatory sweating is
more common than previously believed and topically applied glycopyrrolate
(antimuscarinic compound) is very effective treatment in reducing both the
severity and frequency [165,166]. Symptomatic relief can be obtained by
avoiding the specific inciting food. Loss of lower body sweating can cause
dry, brittle skin that cracks easily, predisposing one to ulcer formation that
can lead to loss of the limb. Special attention must be paid to foot care
(Table 7).

Metabolic dysfunction
Hypoglycemia unawareness
Blood glucose concentration is normally maintained during starvation or
increased insulin action by an asymptomatic parasympathetic response with
bradycardia and mild hypotension, followed by a sympathetic response with
glucagon and epinephrine secretion for short-term glucose counterregula-
tion and growth hormone and cortisol in long-term regulation. Blood
glucose concentration is normally maintained during starvation or increased
insulin action by an asymptomatic parasympathetic response with brady-
cardia and mild hypotension, followed by a sympathetic response with
glucagon and epinephrine secretion for short-term glucose counterregula-
tion and growth hormone and cortisol in long-term regulation. The release
of catecholamine alerts the patient to take the required measures to prevent
coma caused by low blood glucose. The absence of warning signs of
impending neuroglycopenia is known as ‘‘hypoglycemic unawareness.’’ The
failure of glucose counterregulation can be confirmed by the absence of
glucagon and epinephrine responses to hypoglycemia induced by a standard,
controlled dose of insulin [167].
In patients with type 1 diabetes mellitus, the glucagon response is
impaired with diabetes duration of 1 to 5 years, and after 14 to 31 years
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 989

Table 7
Evidenced-based summary of major recommendations for evaluation of diabetic autonomic
neuropathy
Recommendations Evidence grading
Baseline determination of HRV should be performed E
for individuals with type 2 diabetes
Baseline determination of HRV should be performed within E
5 years of diagnosis for those with type 1 diabetes
HRV tests should be repeated annually in type 1 and type E
2 individuals
HRV should be performed before developing an exercise B
program for individuals with diabetes
Preoperative HRV should be performed when planning C
the anesthetic management of diabetic patients
Asymptomatic individuals found to have cardiac E
autonomic dysfunction should undergo additional
cardiac evaluation, particularly if additional cardiovascular
risk factors are present
Testing of cardiac autonomic function after a C
myocardial infraction can provide risk stratification,
identifying a subgroup of patients who are at high
risk for cardiovascular death
Testing of HRV can be used to indicate the presence of B
autonomic neuropathy in patients with symptoms that
may derive from autonomic neuropathy (eg,
erectile dysfunction, gastroparesis, and orthostasis)
Abbreviation: HRV, heart rate variability.

of diabetes, the glucagon response is almost undetectable. It is not present in


those with autonomic neuropathy. A syndrome of hypoglycemic autonomic
failure occurs, however, with intensification of diabetes control and repeated
episodes of hypoglycemia. The exact mechanism is not understood, but it
does represent a real barrier to physiologic glycemic control. In the absence
of severe autonomic dysfunction, hypoglycemic awareness associated with
hypoglycemia at least is in part reversible.
Patients with hypoglycemia unawareness and unresponsiveness pose
a significant management problem for the physician. Although autonomic
neuropathy may improve with intensive therapy and normalization of blood
glucose, there is a risk to the patient, who may become hypoglycemic without
being aware of it and who cannot mount a counterregulatory response. It is
the authors’ recommendation that if a pump is used, boluses of smaller than
calculated amounts should be used, and if intensive conventional therapy is
used, long-acting insulin with very small boluses should be given. In general,
normal glucose and hemoglobin A1 levels should not be goals in these patients
to avoid the possibility of hypoglycemia [48].
Further complicating management of some diabetic patients is the
development of a functional autonomic insufficiency associated with in-
tensive insulin treatment, which resembles autonomic neuropathy in all
990 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

relevant aspects. In these instances, it is prudent to relax therapy, as for the


patient with bona fide autonomic neuropathy. If hypoglycemia occurs in
these patients at a certain glucose level, it takes a lower glucose level to
trigger the same symptoms in the next 24 to 48 hours. Avoidance of hypo-
glycemia for a few days results in recovery of the adrenergic response.

Diabetic neuropathies: prospects for the future


Management of DN encompasses a wide variety of therapies. Treatment
must be individualized in a manner that addresses the particular manifes-
tation and underlying pathogenesis of each patient’s unique clinical presen-
tation, without subjecting the patient to untoward medication effects. There
are new areas being explored to enhance blood flow by vasa nervorum, such
as the prostacyclin analogue beraprost; blockade of thromboxane A2; and
drugs that normalize Na/K-ATPase activity, such as cilostazol, a potent
phosphodiesterase inhibitor, and a-lipoic acid. These, however, have not
reached the clinical area.

Summary
Diabetic neuropathy is a common complication of diabetes that often is
associated with considerable morbidity and mortality. The epidemiology
and natural history of DN is clouded with uncertainty, largely because of
confusion regarding the definition and measurement of this disorder.
The recent resurgence of interest in the vascular hypothesis, oxidative
stress, the neurotrophic hypothesis, and the possibility of the role of
autoimmunity has opened up new avenues of investigation for therapeutic
intervention. Paralleling an increased understanding of the pathogenesis of
DN, there must be refinements in the ability to measure quantitatively the
different types of defects that occur in this disorder, so that appropriate
therapies can be targeted to specific fiber types. These tests must be validated
and standardized to allow comparability between studies and a more
meaningful interpretation of study results. The ability to manage success-
fully the many different manifestations of DN depends ultimately on success
in uncovering the pathogenic processes underlying this disorder.

References
[1] Vinik AI, Mitchell BD, Leichter SB, Wagner AL, O’Brian JT, Georges LP. 1994.
Epidemiology of the complications of diabetes. In: Leslie RDG, Robbins DC, editors.
Diabetes: clinical science in practice. Cambridge, United Kingdom: Cambridge University
Press; 1994. p. 221–87.
[2] Dyck PJ, Kratz KM, Karnes JL, Litchy WJ, Klein R, Pach JM, et al. The prevalence by
staged severity of various types of diabetic neuropathy, retinopathy, and nephropathy in
a population-based cohort: The Rochester Diabetic Neuropathy Study. Neurology 1993;
43:817–24.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 991

[3] EURODIAB IDDM. Microvascular and acute complications in IDDM patients. The
EURODIAB IDDM. Diabetologia 1994;37:278–85.
[4] Holzer SE, Camerota A, Martens L, Cuerdon T, Crystal-Peters J, Zagari M. Costs and
duration of care for lower extremity ulcers in patients with diabetes. Clin Ther 1998;20:
169–81.
[5] Caputo GM, Cavanagh PR, Ulbrecht JS, Gibbons GW, Karchmer AW. Assessment and
management of foot disease in patients with diabetes. N Engl J Med 1994;331:854–60.
[6] Young MJ, Boulton AJM, MacLeod AF, Williams DRR, Sonksen PH. A multicentre
study of the prevalence of diabetic peripheral neuropathy in the United Kingdom hospital
clinic population. Diabetologia 1993;36:1–5.
[7] Pirart J. Diabetes mellitus and its degenerative complications: a prospective study of 4,400
patients observed between 1947 and 1973 (3rd and last part). Diabetes Metab 1977;3:
245–56.
[8] Vinik A. Diabetic neuropathy: pathogenesis and therapy. Am J Med 1999;107:17S–26S.
[9] Armstrong DG, Lavery LA, Harkless LB. Validation of a diabetic wound classification
system: the contribution of depth, infection, and ischemia to risk of amputation. Diabetes
Care 1998;21:855–9.
[10] Levitt NS, Stansberry KB, Wychanck S, Vinik AI. Natural progression of autonomic
neuropathy and autonomic function tests in a cohort of IDDM. Diabetes Care 1996;19:
751–4.
[11] Rathmann W, Ziegler D, Jahnke M, Haastert B, Gries FA. Mortality in diabetic patients
with cardiovascular autonomic neuropathy. Diabet Med 1993;10:820–4.
[12] Consensus Statement. Report and recommendations of the San Antonio conference on
diabetic neuropathy. American Diabetes Association American Academy of Neurology.
Diabetes Care 1988;11:592–7.
[13] Watkins PJ. Progression of diabetic autonomic neuropathy. Diabet Med 1993;10(Suppl
2):77S–8S.
[14] DCCT Research Group. The effect of intensive treatment of diabetes on the develop-
ment and progression of long-term complications in insulin-dependent diabetes mellitus.
N Engl J Med 1993;329:977–86.
[15] DCCT Research Group. The effect of intensive diabetes therapy on the development and
progression of neuropathy. Ann Intern Med 1995;122:561–8.
[16] Ziegler D, Cicmir I, Mayer P, Wiefels K, Gries FA. Somatic and autonomic nerve
function during the first year after diagnosis of type 1 (insulin-dependent) diabetes.
Diabetes Res 1988;7:123–7.
[17] Partanen J, Niskanen L, Lehtinen J, Mervaala E, Siitonen O, Uusitupa M. Natural
history of peripheral neuropathy in patients with non-insulin-dependent diabetes mellitus.
N Engl J Med 1995;333:89–94.
[18] Apfel SC, Kessler JA, Adornato BT, Litchy WJ, Sanders C, Rask CA, et al. Recombinant
human nerve growth factor in the treatment of diabetic polyneuropathy. Neurology 1998;
51:695–702.
[19] Vinik AI. Treatment of diabetic polyneuropathy (DPN) with recombinant human nerve
growth factor (rhNGF) [abstract]. Diabetes 1999;48(Suppl 1):A54–5.
[20] Dyck PJ, Kratz KM, Lehman KA, Karnes JL, Melton LJ, O’Brien PC, et al. The
Rochester Diabetic Neuropathy Study: design, criteria for types of neuropathy, selection
bias, and reproducibility of neuropathic tests. Neurology 1991;41:799–807.
[21] Oh SJ. Clinical electromyelography: nerve conduction studies. In: Oh SJ, editor. Nerve
conduction in polyneuropathies. Baltimore: Williams & Wilkins; 1993. p. 579–91.
[22] Kennedy WR, Wendelschafer-Crabb G, Johnson T. Quantitation of epidermal nerves in
diabetic neuropathy. Neurology 1996;47:1042–8.
[23] Herrmann DN, Griffin JW, Hauer P, Cornblath DR, McArthur JC. Epidermal nerve
fiber density and sural nerve morphometry in peripheral neuropathies. Neurology 1999;
53:1634–40.
992 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

[24] Karamitsos DT, Didangelos TP, Athyros VG, Kontopoulos AG. The natural history of
recently diagnosed autonomic neuropathy over a period of 2 years. Diabetes Res Clin
Pract 1998;42:55–63.
[25] Ziegler D. Diabetic cardiovascular autonomic neuropathy: prognosis, diagnosis and
treatment. Diabetes Metab Rev 1994;10:339–83.
[26] Sima AAF, Sugimoto K. Experimental diabetic neuropathy: an update. Diabetologia
1999;42:773–88.
[27] Zochodne DW. Diabetic neuropathies: features and mechanisms. Brain Pathol 1999;9:
369–91.
[28] Greene DA, Stevens MJ, Obrosova I, Feldman EL. Glucose-induced oxidative
stress and programmed cell death in diabetic neuropathy. Eur J Pharmacol 1999;375:
217–23.
[29] Dyck PJ, Karnes JL, O’Brien PC, Litchy WJ, Low PA, Melton LJ III. The Rochester
Diabetic Neuropathy Study: reassessment of tests and criteria for diagnosis and staged
severity. Neurology 1992;42:1164–70.
[30] Dyck PJ. Detection, characterization and staging of polyneuropathy: assessed in diabetes.
Muscle Nerve 1988;11:21–32.
[31] Herman WH, Kennedy L, for the GOAL A1C Study Group. Physician perception of
neuropathy in a large type 2 diabetes population (GOAL A1C study) confirms
underdiagnosis of neuropathy in everyday clinical practice. Diabetologia 2003;46(Suppl
2):A71.
[32] Vinik AI, Holland MT, LeBeau JM, Liuzzi FJ, Stansberry KB, Colen LB. Diabetic
neuropathies. Diabetes Care 1992;15:1926–75.
[33] Karpitskaya Y, Novak CB, Mackinnon SE. Prevalence of smoking, obesity, diabetes
mellitus and thyroid disease in patients with carpal tunnel syndrome. Am Plast Surg 2002;
48:269–73.
[34] Perkins B, Olaleye D, Bril V. Carpal tunnel syndrome in patients with diabetic poly-
neuropathy. Diabetes Care 2002;25:565–9.
[35] Chaudhuri KR, Davidson AR, Morris IM. Limited joint mobility and carpal tunnel
syndrome in insulin dependent diabetes. Br J Rheumatol 1989;28:191–4.
[36] Wilbourn AJ. Diabetic entrapment and compression neuropathies. In: Dyck PJ, Thomas
PK, editors. Diabetic neuropathy. Philadelphia: WB Saunders; 1999. p. 481–508.
[37] Dawson DM. Entrapment neuropathies of the upper extremities. N Engl J Med 1993;329:
2013–8.
[38] Llewelyn JG, Thomas PK, King RH. Epineurial microvasculitis in proximal diabetic
neuropathy. J Neurol 1998;245:159–65.
[39] Vinik AI, Pittenger GL, Milicevic Z, Knezevic-Cuca J. Autoimmune mechanisms in the
pathogenesis of diabetic neuropathy. In: Eisenbarth RG, editor. Molecular mechanisms
of endocrine and organ specific autoimmunity. Georgetown: Landes Company; 1998.
p. 217–51.
[40] Steck AJ, Kappos L. Gangliosides and autoimmune neuropathies: classification and
clinical aspects of autoimmune neuropathies. J Neurol Neurosurg Psychiatry 1994;57:
26–8.
[41] Sander HW, Chokroverty S. Diabetic amyotrophy: current concepts. Semin Neurol 1996;
16:173–8.
[42] Said G, Goulon-Goreau C, Lacroix C, Moulonguet A. Nerve biopsy findings in different
patterns of proximal diabetic neuropathy. Ann Neurol 1994;35:559–69.
[43] Krendel DA, Costigan DA, Hopkins LC. Successful treatment of neuropathies in patients
with diabetes mellitus. Arch Neurol 1995;52:1053–61.
[44] Britland ST, Young RJ, Sharma AK, Clarke BF. Acute and remitting painful diabetic
polyneuropathy: a comparison of peripheral nerve fibre pathology. Pain 1992;48:361–70.
[45] Stewart JD, McKelvey R, Durcan L, Carpenter S, Karpati G. Chronic inflammatory
demyelinating polyneuropathy (CIDP). J Neurol Sci 1996;142:59–64.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 993

[46] Sharma K, Cross J, Farronay O, Ayyar D, Sheber R, Bradley W. Demyelinating


neuropathy in diabetes mellitus. Arch Neurol 2002;59:758–65.
[47] Vinik AI. Diabetic neuropathy, mobility and balance. Geriatric Times 2003;4:13–5.
[48] Vinik A. Diagnosis and management of diabetic neuropathy. Clin Geriatr Med 1999;15:
293–319.
[49] Milicevic Z, Pittenger GL, Stansberry KB, Vinik AI. Raised anti-ganglioside GM1
antibody (GM1 Ab) titers in a subset of patients with distal symmetric polyneuropathy
(DSPN) [abstract]. Diabetes 1997;46:125A.
[50] Milicevic Z, Newlon PG, Pittenger GL, Stansberry KB, Vinik AI. Anti-ganglioside GM1
antibody and distal symmetric ‘‘diabetic polyneuropathy’’ with dominant motor features.
Diabetologia 1997;40:1364–5.
[51] Barada A, Reljanovic M, Milicevic Z, Ljubic S, Car N, Benko B, et al. Proximal diabetic
neuropathy-response to immunotherapy [abstract]. Diabetes 1999;48(Suppl 1):A148.
[52] Bird SJ, Brown MJ. The clinical spectrum of diabetic neuropathy. Semin Neurol 1996;16:
115–22.
[53] Hanson PH, Schumaker P, Debugne TH, Clerin M. Evaluation of somatic and
autonomic small fibers neuropathy in diabetes. Am J Phys Med Rehabil 1992;71:44–7.
[54] McArthur JC, Stocks EA, Hauer P, Cornblath DR, Griffin JW. Epidermal nerve fiber
density: normative reference range and diagnostic efficiency. Arch Neurol 1998;55:
1513–20.
[55] Stansberry KB, Hill MA, Shapiro SA, McNitt PM, Bhatt BA, Vinik AI. Impairment of
peripheral blood flow responses in diabetes resembles an enhanced aging effect. Diabetes
Care 1997;20:1711–6.
[56] Low P, Dotson R. Symptom treatment of painful neuropathy. JAMA 1998;280:1863–4.
[57] Vinik AI, Park TS, Stansberry KB, Pittenger GL. Diabetic neuropathies. Diabetologia
2000;43:957–73.
[58] Krause S, Backonja M. Development of a neuropathic pain questionnaire. Clin J Pain
2003;19(5):306–14.
[59] Galer BS. Neuropathic pain of peripheral origin: advances in pharmacologic treatment.
Neurology 1995;45:S17–25.
[60] Vinik EJ, Stansberry KB, Zarrabi L, Witherspoon CAG, McNitt PM, Vinik AI.
Development of a sensitive, specific quality of life inventory for peripheral neuropathy
[abstract]. Diabetes 2000;49:A819.
[61] Vinik E, Stansberry K, Vinik A. Neuropathy Quality of Life Tool (Norfolk QOL-DN) in
a large clinical trial: comparison with symptom scores (NSS), nerve impairment (NISS),
quantitative sensory (QST), autonomic (QAFT) tests and electrophysiology. Diabetes
2003;52(Suppl 1).
[62] Backonja M, Beydoun A, Edwards KR, Schwartz SL, Fonseca V, Hes M, et al.
Gabapentin for the symptomatic treatment of painful neuropathy in patients with
diabetes mellitus. JAMA 1998;280:1831–6.
[63] Vinik A, Tesfaye S, Zhang D, Bastyr E. LY333531 treatment improves diabetic peripheral
neuropathy (DPN) with symptoms. Diabetes 2002;51(Suppl 2):A79.
[64] Litchy W, Dyck P, Tesfaye S, Zhang D, Bastyr E, the MBBQ Study Group. Diabetic
peripheral neuropathy (DPN) assessed by neurological examination (NE) and composite
scores (CS) is improved with LY333531 treatment. Diabetes 2002;51:A197.
[65] Tesfaye S, Malik R, Harris N, Jakubowski JJ, Mody C, Rennie IG, et al. Arterio-venous
shunting and proliferating new vessels in acute painful neuropathy of rapid glycaemic
control (insulin neuritis). Diabetologia 1996;39:329–35.
[66] Van Heel DA, Levitt NS, Winter TA. Diabetic neuropathic cachexia: the importance of
positive recognition and early nutritional support. Int J Clin Pract 1998;52:591–2.
[67] Holland NR, Crawford TO, Hauer P, Cornblath DR, Griffin JW, McArthur JC. Small-
fiber sensory neuropathies: clinical course and neuropathology of idiopathic cases. Ann
Neurol 1998;44:47–59.
994 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

[68] Said G, Bigo A, Ameri A, Gayno JP, Elgrably F, Chanson P, et al. Uncommon early-
onset neuropathy in diabetic patients. J Neurol 1998;245:61–8.
[69] Vinik AI, Mitchell B. Clinical aspects of diabetic neuropathies. Diabetes Metab Rev 1988;
4:223–53.
[70] Ziegler D, Hanefeld M, Ruhnau KJ, Meissner HP, Lobisch M, Schutte K, et al. Treatment
of symptomatic diabetic peripheral neuropathy with the anti-oxidant alpha-lipoic acid: a
3-week multicentre randomized controlled trial (ALADIN Study). Diabetologia 1995;
38:1425–33.
[71] Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N, Greene DA. A practical
two-step quantitative clinical and electrophysiological assessment for the diagnosis and
staging of diabetic neuropathy. Diabetes Care 1994;17:1281–9.
[72] Vinik AI, Newlon P, Milicevic Z, McNitt P, Stansberry KB. Diabetic neuropathies: an
overview of clinical aspects. In: LeRoith D, Taylor SI, Olefsky JM, editors. Diabetes
mellitus: a fundamental and clinical text. Philadelphia: Lippincott-Raven; 1996. p. 737–51.
[73] Paisley AN, Abbott CA, van Schie CHM, Boulton AJM. A comparison of the Neuropen
against standard quantitative sensory threshold measures for assessing peripheral nerve
function. Diabet Med 2002;19;400–5.
[74] Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N, Greene DA. A practical
two-step quantitative clinical and electrophysiological assessment for the diagnosis and
staging of diabetic neuropathy. Diabetes Care 1994;17:1281–9.
[75] Apfel SC, Asbury A, Bril V, Burns T, Campbell J, Chalk C, et al. Positive neuropathic
sensory symptoms as endpoints in diabetic neuropathy trials [abstract]. J Neurol Sci 2001;
189:3–5.
[76] Armstrong DG, Lavery LA, Vela SA, Quebedeaux TL, Fleischli JG. Choosing a practical
screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern
Med 1998;158:289–92.
[77] Franse LV, Valk GD, Dekker JH, Heine RJ, Van Eijk JTM. Numbness of the feet is
a poor indicator for polyneuropathy in type 2 diabetic patients. Diabetes Care 2000;17:
105–10.
[78] Dyck PJ. Severity and staging of diabetic polyneuropathy. In: Textbook of diabetic
neuropathy. Stuttgart: Thieme; 2002. p. 170–5.
[79] Shy ME, Frohman EM, So YT, Arezzo JC, Cornblath DC, Giuliani MJ, et al.
Quantitative sensory testing. Neurology 2003;602:898–906.
[80] Arezzo JC, Zotova E. Electrophysiologic measures of diabetic neuropathy: mechanism
and meaning. Int Rev Neurobiol 2002;50:229–55.
[81] Meijer JW, Smit AJ, Sondersen EV, Groothoff JW, Eisma WH, Links TP. Symptom
scoring systems to diagnose distal polyneuropathy in diabetes: the Diabetic Neuropathy
Symptom Score. Diabet Med 2002;19:962–5.
[82] Scott LA, Tesfaye S, Rajbhandari S, Harris N, Platter MA, Dent M, et al. Large fiber
involvement precedes small-fiber disease in the development of human diabetic neuro-
pathy [abstract]. Diabetes 2002;19:87.
[83] Ziegler D, Mayer P, Gries F. Evaluation of thermal, pain, and vibratory sensation
thresholds in newly diagnosed type I diabetic subjects. J Neurol 1998;51:1420–4.
[84] Vinik AI, Suwanwalaikorn S, Stansberry KB, Holland MT, McNitt PM, Colen LE.
Quantitative measurement of cutaneous perception in diabetic neuropathy. Muscle Nerve
1995;18:574–84.
[85] Valk GD, de Sonnaville JJ, van Houtum WH, Heine RJ, van Eijk JT, Bouter LM, et al.
The assessment of diabetic polyneuropathy in daily clinical practice: reproducibility and
validity of Semmes Weinstein monofilaments examination and clinical neurological
examination. Muscle Nerve 1997;20:116–8.
[86] Mayfield JA, Sugarman JR. The use of Semmes-Weinstein monofilament and other
threshold tests for preventing foot ulceration and amputation in people with diabetes.
J Fam Pract 2000;49:517–29.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 995

[87] Kumar D. Diabetic peripheral neuropathy: effectiveness of electrotherapy and amitripty-


line for symptomatic relief. Diabetes Care 1998;21:1322–5.
[88] Booth J, Young MJ. Differences in the performance of commercially available 10-g
monofilaments. Diabetes Care 2000;23:984–8.
[89] Liniger C, Albeanu A, Bloise D, Assal JP. The tuning fork revisited. Diabet Med 1990;7:
859–64.
[90] Vileikyte L, Hutchings G, Hollis S, Boulton AJM. The tactile circumferential
discriminator: a new simple screening device to identify diabetic patients at risk of foot
ulceration. Diabetes Care 1997;20:623–6.
[91] Ducher M, Thivolet C, Cerutti C, Laville M, Gustin MP, Paultre CZ, et al. Noninvasive
exploration of cardiac autonomic neuropathy. Diabetes Care 1999;22:388–93.
[92] Jaradeh SS, Prieto TE, Lobeck LJ. Progressive polyradiculoneuropathy in diabetes:
correlation of variables and clinical outcome after immunotherapy. J Neurol Neurosurg
Psychiatry 1999;67:607–12.
[93] Periquet MI, Novak V, Collins MP, Nagaraja HN, Erdem S, Nash SM, et al. Painful
sensory neuropathy: prospective evaluation using skin biopsy. Neurology 1999;53:1641–7.
[94] Krendel DA, Zacharias A, Younger DS. Autoimmune diabetic neuropathy. Neurol Clin
1997;15:959–71.
[95] Smith A, Ramachandran P, Tripp S, Singleton J. Epidermal nerve innervation in impaired
glucose tolerance and diabetes-associated neuropathy. Neurology 2001;57:1701–4.
[96] Sumner C, Sheth S, Griffin J, Cornblath D, Polydefkis M. The spectrum of neuropathy in
diabetes and impaired glucose tolerance. Neurology 2003;60:108–11.
[97] Vinik A, Pittenger G, Anderson A, Stansberry K, McNear E, Barlow P. Topiramate
improves C-fiber neuropathy and features of the dysmetabolic syndrome in type 2
diabetes. Diabetes 2003;52(Suppl 1):A130.
[98] Vinik AI. Neuropathy could be managed with topiramate. Endocrine Today 2003;1:1–8.
[99] Quattrini C, Jeriorska M, Gawkrodger D, Tesfaye S, Malik R. Dermal nerve depletion
and angiogenesis in diabetic neuropathy. Diabetologia 2003;46(Suppl 2):A72.
[100] Arezzo JC. The use of electrophysiology for the assessment of diabetic neuropathy.
Neurosci Res Comm 1997;21:13–22.
[101] Amthor KF, Dahl-Jorgensen K, Berg TJ, Heier MS, Sandvik L, Aagenaes O, et al. The
effect of 8 years of strict glycaemic control on peripheral nerve function in IDDM pa-
tients: the Oslo Study. Diabetologia 1994;37:579–84.
[102] Veves A, Malik RA, Lye RH, Masson EA, Sharma AK, Schady W, et al. The relationship
between sural nerve morphometric findings and measures of peripheral nerve function in
mild diabetic neuropathy. Diabet Med 1991;8:917–21.
[103] UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose
control with metformin on complications in overweight patients with type 2 diabetes
(UKPDS 34). Lancet 1998;352:854–65.
[104] UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with
sulphonylureas or insulin compared with conventional treatment and risk of complica-
tions in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.
[105] UK Prospective Diabetes Study Group. Tight blood pressure control and risk of
macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ
1998;317:703–13.
[106] Gaede P, Vedel P, Parving HH, Pedersen O. Intensified multifactorial intervention in
patients with type 2 diabetes mellitus and microalbuminuria: the Steno type 2 randomised
study. Lancet 1999;353:617–22.
[107] Boulton AJM, Levin S, Comstock J. A multicentre trial of the aldose reductase inhibitor,
tolrestat, in patients with symptomatic diabetic neuropathy. Diabetologia 1990;33:431–7.
[108] Didangelos TP, Karamitsos DT, Athyros VG, Kourtoglou GI. Effect of aldose reductase
inhibition on cardiovascular reflex tests in patients with definite diabetic autonomic
neuropathy. J Diabetes Complications 1998;12:201–7.
996 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

[109] Greene DA, Arezzo JC, Brown MB. Effect of aldose reductase inhibition on nerve con-
duction and morphometry in diabetic neuropathy. Zenarestat Study Group. Neurology
1999;53:580–91.
[110] Johnson BF, Law G, Nesto R, Pfeifer M, Slater W, Vinik A, et al. Aldose reductase
inhibitor zopolrestat improves systolic function in diabetics [abstract]. Diabetes 1999;
48(Suppl 1):A133.
[111] Ziegler D, Schatz H, Conrad F, Gries FA, Ulrich H, Reichel G. Effects of treatment with
the antioxidant alpha-lipoic acid on cardiac autonomic neuropathy in NIDDM patients:
a 4-month randomized controlled multicenter trial (DEKAN Study). Deutsche Kardiale
Autonome Neuropathie. Diabetes Care 1997;20:369–73.
[112] Ziegler D, Gries FA. Alpha-lipoic acid in the treatment of diabetic peripheral and cardiac
autonomic neuropathy. Diabetes 1997;46(Suppl 2):S62–6.
[113] Ziegler D, Hanefeld M, Ruhnau KJ, Hasche H, Lobisch M, Schutte K, et al. Treatment
of symptomatic diabetic polyneuropathy with the antioxidant alpha-lipoic acid: a
7-month multicenter randomized controlled trial (ALADIN III Study). ALADIN III
Study Group. Alpha-Lipoic Acid in Diabetic Neuropathy. Diabetes Care 1999;22:
1296–301.
[114] Jamal GA. The use of gamma linolenic acid in the prevention and treatment of diabetic
neuropathy. Diabet Med 1994;11:145–9.
[115] Keen H, Payan J, Allawi J, Walker J, Jamal GA, Weir AI, et al. Treatment of diabetic
neuropathy with c-linolenic acid. Diabetes Care 1993;16:8–15.
[116] Miyauchi Y, Shikama H, Takasu T, Okamiya H, Umeda M, Hirasaki E, et al. Slowing of
peripheral motor nerve conduction was ameliorated by aminoguanidine in streptozocin-
induced diabetic rats. Eur J Endocrinol 1996;134:467–73.
[117] Schmidt RE, Dorsey DA, Beaudet LN, Reiser KM, Williamson JR, Tilton RG. Effect of
aminoguanidine on the frequency of neuroaxonal dystrophy in the superior mesenteric
sympathetic autonomic ganglia of rats with streptozotocin-induced diabetes. Diabetes
1996;45:284–90.
[118] Nargi SE, Colen LB, Liuzzi F, Al-Abed Y, Vinik AI. PTB treatment restores joint mobility
in a new model of diabetic cheirothropathy [abstract]. Diabetes 1999;48(Suppl 1):A17.
[119] Suez D. Intravenous immunoglobulin therapy: indication, potential side effects and
treatment guidelines. J Intraven Nurs 1995;18:178–90.
[120] Diemel LT, Stevens JC, Willars GB, Tomlinson DR. Depletion of substance P and
calcitonin gene-related peptide in sciatic nerve of rats with experimental diabetes: effects
of insulin and aldose reductase inhibition. Neurosci Lett 1992;137:253–6.
[121] Tomlinson DR, Fernyhough P, Diemel LT. Neurotrophins and peripheral neuropathy.
Philos Trans R Soc Lond B Biol Sci 1996;351:455–62.
[122] Apfel SC, Kessler JA. Neurotropic factors in the therapy of peripheral neuropathy.
Baillieres Clinical Neuropathy 1995;4:593–606.
[123] Sindrup SH, Jensen TS. Efficacy of pharmacological treatments of neuropathic pain: an
update and effect related to mechanism of drug action. Pain 1999;83:389–400.
[124] Rains C, Bryson HM. Topical capsaicin: a review of its pharmacological properties and
therapeutic potential in post-herpetic neuralgia, diabetic neuropathy and osteoarthritis.
Drugs Aging 1995;7:317–28.
[125] Bays-Smith MG, Max MB, Muir J, Kingman A. Transdermal clonidine compared to
placebo in painful diabetic neuropathy using a two-stage Ôenriched enrollmentÕ design.
Pain 1995;60:267–74.
[126] Jarvis B, Coukell AJ. Mexiletine: a review of its therapeutic use in painful diabetic
neuropathy. Drugs 1998;56:691–707.
[127] Harati Y, Gooch C, Swenson M, Edelman S, Greene D, Raskin P, et al. Double-blind
randomized trial of tramadol for the treatment of the pain of diabetic neuropathy.
Neurology 1998;50:1842–6.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 997

[128] Harati Y, Gooch C, Swenson M, Edelman SV, Greene D, Raskin P, et al. Maintenance of
the long-term effectiveness of tramadol in treatment of the pain of diabetic neuropathy.
J Diabetes Complications 2000;14:65–70.
[129] Nelson KA, Park KM, Robinovitz E, Tsigos C, Max MB. High-dose oral dextro-
methorphan versus placebo in painful diabetic neuropathy and postherpetic neuralgia.
Neurology 1997;48:1212–8.
[130] Max M, Lynch S, Muir J. Effects of desipramine, amitryptylline and fluoxetine on pain in
diabetic neuropathy. N Engl J Med 1992;326:1250–6.
[131] de Abajo FJ, Rodriguez LA, Montero D. Association between selective serotonin
reuptake inhibitors and upper gastrointestinal bleeding: population based case control
study. BMJ 1999;319:1106–9.
[132] McQuay H, Carroll D, Jadad AR, Wiffen P, Moore A. Anticonvulsant drugs for
management of pain: a systematic review. BMJ 1995;311:1047–52.
[133] Gorson KC, Schott C, Herman R, Ropper AH, Rand WM. Gabapentin in the treatment
of painful diabetic neuropathy: a placebo controlled, double blind, crossover trial.
J Neurol Neurosurg Psychiatry 1999;66:251–2.
[134] Vinik A, Fonseca V, LaMoreaux L, Hes M, Koto E. Neurontin (gabapentin, GBP)
improves quality of life (QOL) in patients with painful diabetic peripheral neuropathy
[abstract]. Diabetes 1998;47(Suppl 1):A374.
[135] Eisenberg E, Lurie Y, Braker C, Daoud D, Ishay A. Lamotrigine reduces painful diabetic
neuropathy: a randomized, controlled study. Neurology 2001;57:505–9.
[136] Vinik A, Hewitt D, Xiang J. Topiramate in the treatment of painful diabetic neuropathy:
results from a multicenter, randomized, double-blind, placebo-controlled trial [abstract].
Neurology 2003;60(Suppl 1):A154–5.
[137] Somers DL, Somers MF. Treatment of neuropathic pain in a patient with diabetic
neuropathy using transcutaneous electrical nerve stimulation applied to the skin of the
lumbar region. Phys Ther 1999;79:767–75.
[138] Weintraub MI, Wolfe GI, Barohn RA, Cole SP, Parry GJ, Hayat G, et al. Static magnetic
field therapy for symptomatic diabetic neuropathy: a randomized, double-blind, placebo-
controlled trial. Arch Phys Med Rehabil 2003;84:736–46.
[139] Tesfaye S, Watt J, Benbow SJ, Pang KA, Miles J, MacFarlane IA. Electrical spinal-cord
stimulation for painful diabetic peripheral neuropathy. Lancet 1996;348:1698–701.
[140] Zieleniewski W. Calcitonin nasal spray for painful diabetic neuropathy. Lancet 1990;336:
449.
[141] Cavanagh PR, Derr JA, Ulbrecht JS, Maser RE, Orchard TJ. Problems with gait and
posture in neuropathic patients with insulin-dependent diabetes mellitus. Diabet Med
1992;9:469–74.
[142] Nelson ME, Fiatarone MA, Morganti CM, Trice I, Greenberg RA, Evans WJ. Effects of
high-intensity strength training on multiple risk factors for osteoporotic fractures:
a randomized controlled trial. JAMA 1994;272:1909–14.
[143] Zola BE, Vinik AI. Effects of autonomic neuropathy associated with diabetes mellitus on
cardiovascular function. Coron Artery Dis 1992;3:33–41.
[144] Stansberry KB, Shapiro SA, Hill MA, McNitt PM, Meyer MD, Vinik AI. Impaired
peripheral vasomotion in diabetes. Diabetes Care 1996;19:715–21.
[145] Stansberry KB, Peppard HR, Babyak LM, Popp G, McNitt PM, Vinik AI. Primary
nociceptive afferents mediate the blood flow dysfunction in non-glabrous (hairy) skin of
type 2 diabetes. Diabetes Care 1999;22:1549–54.
[146] Haak ES, Usadel KH, Kohleisen M, Yilmaz A, Kusterer K, Haak T. The effect of alpha-
lipoic on the neurovascular reflex arc in patients with diabetic neuropathy assessed by
capillary microscopy. Microvasc Res 1999;58:28–34.
[147] Valensi P. Diabetic autonomic neuropathy: what are the risks? Diabetes Metab 1998;24:
66–72.
998 A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999

[148] Mancia G, Paleari F, Parati G. Early diagnosis of diabetic autonomic neuropathy:


present and future approaches. Diabetologia 1997;40:482–4.
[149] Athyros VG, Didangelos TP, Karamitsos DT, Papageorgiou AA, Boudoulas H,
Kontopoulos AG. Long-term effect of converting enzyme inhibition on circadian
sympathetic and parasympathetic modulation in patients with diabetic autonomic
neuropathy. Acta Cardiol 1998;53:201–9.
[150] Malmberg K, Norhammar A, Wedel H, Ryden L. Glycometabolic state at admission:
important risk marker of mortality in conventionally treated patients with diabetes mellitus
and acute myocardial infarction: long-term results from the Diabetes and Insulin-Glucose
Infusion in Acute Myocardial Infarction (DIGAMI) study. Circulation 1999;99:2626–32.
[151] Kendall DM, Rooney DP, Smets YF, Salazar Bolding L, Robertson RP. Pancreas
transplantation restores epinephrine response and symptom recognition during hypogly-
cemia in patients with long-standing type I diabetes and autonomic neuropathy. Diabetes
1997;46:249–57.
[152] Burger AJ, Weinrauch LA, D’Elia JA, Aronson D. Effects of glycemic control on heart
rate variability in type I diabetic patients with cardiac autonomic neuropathy. Am J
Cardiol 1999;84:687–91.
[153] Laederach-Hofmann K, Weidmann P, Ferrari P. Hypovolemia contributes to the patho-
genesis of orthostatic hypotension in patients with diabetes mellitus. Am J Med 1999;
106:50–8.
[154] Denq JC, Opfer-Gehrking TL, Giuliani M, Felten J, Convertino VA, Low PA. Efficacy of
compression of different capacitance beds in the amelioration of orthostatic hypotension.
Clin Auton Res 1997;7:321–6.
[155] Annese V, Bassotti G, Caruso N, De Cosmo S, Gabbrielli A, Modoni S, et al.
Gastrointestinal motor dysfunction, symptoms, and neuropathy in noninsulin-dependent
(type 2) diabetes mellitus. J Clin Gastroenterol 1999;29:171–7.
[156] Melga P, Mansi C, Ciuchi E, Giusti R, Sciaba L, Prando R. Chronic administration of
levosulpiride and glycemic control in IDDM patients with gastroparesis. Diabetes Care
1997;20:55–8.
[157] Stacher G, Schernthaner G, Francesconi M, Kopp HP, Bergmann H, Stacher-Janotta G,
et al. Cisapride versus Placebo for 8 weeks on glycemic control and gastric emptying in
insulin-dependent diabetes: a double blind cross-over trial. J Clin Endocrinol Metab 1999;
84:2357–62.
[158] Barone JA. Domperidone: a peripherally acting dopamine2-receptor antagonist. Ann
Pharmacother 1999;33:429–40.
[159] Silvers D, Kipnes M, Broadstone V, Patterson D, Quigley EM, McCallum R, et al.
Domperidone in the management of symptoms of diabetic gastroparesis: efficacy,
tolerability, and quality-of-life outcomes in a multicenter controlled trial. DOM-USA-5
Study Group. Clin Ther 1998;20:438–53.
[160] Erbas T, Varoglu E, Erbas B, Tastekin G, Akalin S. Comparison of metoclopramide and
erythromycin in the treatment of diabetic gastroparesis. Diabetes Care 1993;16:1511–4.
[161] Vinik AI, Richardson D. Erectile dysfunction in diabetes. Diabetes Reviews 1998;6:
16–33.
[162] Vinik AI, Richardson D. Erectile dysfunction in diabetes: pills for penile failure. Clinica
Diabetes 1998;16:108–19.
[163] Rendell MS, Rajfer J, Wicker PA, Smith MD. Sildenafil Diabetes Study Group. Sildenafil
for treatment of erectile dysfunction in men with diabetes: a randomized controlled trial.
JAMA 1999;281:421–6.
[164] Enzlin P, Mathieu C, Vanderschueren D, Demyttenaere K. Diabetes mellitus and female
sexuality: a review of 25 years’ research. Diabet Med 1998;15:809–15.
[165] Shaw JE, Abbott CA, Tindle K, Hollis S, Boulton AJ. A randomised controlled trial of
topical glycopyrrolate, the first specific treatment for diabetic gustatory sweating.
Diabetologia 1997;40:299–301.
A.I. Vinik, A. Mehrabyan / Med Clin N Am 88 (2004) 947–999 999

[166] Shaw JE, Parker R, Hollis S, Gokal R, Boulton AJ. Gustatory sweating in diabetes
mellitus. Diabet Med 1996;13:1033–7.
[167] Meyer C, Hering BJ, Grossmann R, Brandhorst H, Brandhorst D, Gerich J, et al.
Improved glucose counterregulation and autonomic symptoms after intraportal islet
transplants alone in patients with long-standing type I diabetes mellitus. Transplantation
1998;66:233–40.
Med Clin N Am 88 (2004) 1001–1036

Diabetic nephropathy and retinopathy


Ali Jawa, MD, Juanita Kcomt, MD,
Vivian A. Fonseca, MD*
Section of Endocrinology, Department of Medicine, Tulane University Health Sciences Center,
SL-53, 1430 Tulane Avenue, New Orleans, LA 70112–2699, USA

Diabetic nephropathy (DN) and diabetic retinopathy (DR) are arguably


the two most dreaded complications of diabetes. Together they contribute
to serious morbidity and mortality. As they progress to end-stage renal dis-
ease (ESRD) and blindness, they impose enormous medical, economic, and
social costs on both the patient and the health care system. Because
nephropathy and retinopathy are frequently linked in patients, this article
reviews their common and individual aspects of pathophysiology, clinical
features, and management.
Diabetic nephropathy is a clinical syndrome characterized by persistent
albuminuria, arterial blood pressure elevation, a relentless decline in
glomerular filtration rate (GFR), and a high risk of cardiovascular
morbidity and mortality. This major life-threatening complication develops
in approximately 20% to 40% of type 1 and less than 20% of type 2 diabetic
patients [1]. DN is the leading known cause of ESRD in the United States,
accounting for an estimated 28,000 new cases of ESRD per year [1].
Retinopathy is a serious microvascular complication of diabetes mellitus
and the leading cause of blindness in adults less than 65 years of age. It is
estimated that about 5.5 million adult patients with diabetes have DR.
About 50,000 new cases of blindness occur per year, out of which 50% are
caused by diabetes and most caused by DR [2].
Both ESRD and blindness are preventable through early detection and
treatment. This article discusses the pathophysiology of these disorders and
strategies to prevent these late complications in patients with diabetes.
Improved understanding of pathophysiology has the potential to lead to
novel medical therapies in the future.

* Corresponding author.
E-mail address: vfonseca@tulane.edu (V.A. Fonseca).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.012
1002 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

Historical perspective
In 1936, Kimmelstiel and Wilson described the renal histology at autopsy
in eight cases of which seven had diabetes, together with hypertension,
albuminuria, edema, and renal failure and had the characteristic nodular
lesions of diabetes mellitus. These findings were extended by other workers,
[3,4] who confirmed the existence of a specific histopathology of the kidney
in diabetes mellitus. Diffuse glomerulosclerosis was subsequently described
and distinguished from the nodular form of Kimmelstiel and Wilson [5].
The introduction of percutaneous renal biopsies and electron microscopy
in the 1950s rapidly led to a greater understanding of the disease. Studies
confirmed that the earliest changes in the kidney in diabetes consisted of the
accumulation of basement membrane–like material in the mesangium
together with basement membrane thickening [6].
The most striking advances in DR relates to its treatment with retinal
photocoagulation. The first use of photocoagulation in humans for retinal
photocoagulation using the xenon arc lamp was by Meyer-Schwickerath in
1946. Several large clinical trials have helped define the epidemiology,
natural history, and management strategies in DR [7–16].

Epidemiology
The epidemiology of DN has been best studied in patients with type 1
diabetes, because the time of clinical onset is usually known. Approximately
25% to 45% of these patients develop clinically evident disease during their
lifetime [17–19]. The peak time to development of nephropathy in type 1
diabetes is between 10 and 15 years after the onset of disease. Importantly,
patients who do not develop proteinuria after 20 to 25 years of diabetes have
a very low subsequent risk of developing overt renal disease of only about 1%
per year [17]. In patients with type 2 diabetes, the prevalence of progressive
renal disease has previously been reported to be lower. Nephropathy develops
in up to 50% of type 2 diabetic Pima Indians 20 years after diagnosis of
type 2 diabetes, however, and 15% have progressed to ESRD by this time
[20,21]. Importantly, proteinuria is a risk factor for cardiovascular disease
and it is possible that previous studies underestimate the prevalence of DN
in type 2 diabetes because many patients died of cardiovascular disease
before developing ESRD.
Recent data suggest that the risk of nephropathy is equivalent in the two
types of diabetes. Evidence in support of this hypothesis in one report were
the observations that the time to proteinuria from the onset of diabetes and
the time to ESRD from the onset of proteinuria were similar in type 1 and
type 2 disease [22].
Diabetic retinopathy is more prevalent among patients with type 1
diabetes than type 2. Within 5 and 10 years of diagnosis, about 58% and
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1003

80%, respectively, have retinopathy. After 15 to 20 years of disease, more


than 90% have some kind of retinopathy and approximately 60% have
proliferative retinopathy. After greater than or equal to 20 years 99% have
retinopathy and 53% have proliferative retinopathy. In comparison, more
than 25% of patients with type 2 diabetes have retinopathy within 2 years of
diagnosis. Sixty percent have some retinopathy and 5% have proliferative
retinopathy greater than or equal to 20 years after diagnosis, far less than
type 1 diabetes [23].

Natural history and pathophysiology


Diabetic nephropathy
The natural history of DN is complex, linked closely with the
pathophysiology, and many changes in the kidney are currently not
detectable in clinical practice. The course of DN is slow and fortunately
modifiable by interventions used in clinical practice. Mogensen et al [24]
propose a five-stage sequence for renal involvement (Table 1).

Stage 1: glomerular hyperfiltration and renomegaly


Even with good or fair glucose control, the GFR remains above control
levels in 25% to 40% of patients. In this subgroup of hyperfiltering patients,
reductions in the GFR and clinical nephropathy eventually develop at
a much greater rate than in control patients with diabetes with normal GFR.
Renal size and GFR are raised in newly diagnosed patients [25]. This raised
GFR was shown to correlate closely with an increased glomerular capillary
filtration surface area [26]. It has been suggested that patients with diabetes

Table 1
Stages of diabetic nephropathy
Stage Characteristics
Normoalbuminuria Normal albumin excretion rate (AER \20 lg/min)
Microalbuminuria Increased albumin excretion rate (AER 20–200 lg/min)
Incipient diabetic Persistent microalbuminuria (in at least two of three
nephropathy collections over 6 mo)  hyperfiltration; blood
pressure elevation
Early overt diabetic Clinical-grade proteinuria (AER [200 lg/min in
nephropathy two of three collections within 6 mo or
dipstick-positive proteinuria); hypertension
Advanced diabetic Progressive proteinuria; hypertension; declining
nephropathy glomerular filtration rate (decreased creatinine
clearance, increased blood urea nitrogen and creatinine)
End-stage renal disease Uremia; nephrotic syndrome; need for renal
replacement therapy (transplantation or dialysis)
Abbreviation: AER, albumin excretion rate in a timed specimen.
1004 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

with the highest GFR early in the course of their disease may be those most
likely to develop DN [27,28]. There have been no prospective studies,
however, demonstrating that patients with hyperfiltration progress to
chronic renal failure at a greater or faster fashion than patients without
hyperfiltration.

Stage 2: early glomerular lesions


There is mild thickening of glomerular basement membrane 18 to 24
months after the onset of type 1 diabetes and may be pronounced after 3.5
to 5 years [29]. Exercise-induced microalbuminuria is the only clinical
evidence of renal involvement during this stage, which may extend from 4 or
5 to 15 years following the diagnosis of diabetes. Alteration in the molecular
structure of components of the glomerulus and its basement membrane have
also been suggested as primary pathogenetic mechanisms. Glycosylation of
the basement membrane has been shown to occur and may result in the
increased filtration of proteins [30]. A reduction in the negative charge of the
basement membrane secondary to a degree in sialic acid and sulphated
proteoglycans has been suggested as the basis for the proteinuria of DN.
The repulsive electrostatic interaction with negatively charged plasma
proteins, such as albumin, is reduced and so increased filtration of albumin
may occur [31]. Vigstrup and Mogensen [32] published an article about
microalbuminuria predicting proliferative DR. Mogensen [33] pointed
out the importance of microalbuminuria as a predictor of clinical dia-
betic nephropathy (DN). Finally the same author proposed the five-stage
sequence for renal involvement in type 1 diabetes mellitus [24].

Stage 3: incipient diabetic nephropathy—stage of microalbuminuria


Microalbuminuria, defined by a daily urinary albumin excretion rate
of 20 to 200 lg/min, predicts renal functional deterioration and a poor
outcome [34]. It is also associated with vascular damage in other organs.

Stage 4: clinical nephropathy—macroalbuminuria, falling glomerular


filtration rate
The incidence of macroalbuminuria peaks in patients who have had
diabetes for 15 to 20 years. Without intervention, the GFR in macro-
albuminuric patients with type 1 diabetes falls at about 1 mL/min/mo [34,35].
Nephrotic syndrome is also very common.

Stage 5: end-stage renal disease


End-stage renal disease and its multiple complications and comorbid
conditions occur after 20 to 30 years of diabetes in 30% to 40% of patients
with type 1 diabetes. Uremic symptoms and signs are manifested at creatinine
clearances that are higher than that in nondiabetic persons, and renal
replacement therapy in suboptimal-treated individuals is needed within 2 to
3 years of the onset of nephrotic syndrome.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1005

Diabetic retinopathy
In contrast, the natural history of DR is not as clearly defined, although
the condition can be easily classified clinically (Table 2). The early stage of
DR is characterized by loss of pericytes around capillaries in the retina. This
is followed by development of weakness in the capillary wall that leads
to capillary aneurysm formation (microaneurysm) and fluid leakage from
capillaries as their walls become more permeable. Impaired vascular
function gradually develops leading to areas of ischemia and infarction.
In response to these changes local growth factors are secreted that con-
tribute to new vessel proliferation.

Pathology
There are three major histologic changes in the glomeruli in DN: (1)
mesangial expansion, (2) glomerular basement membrane thickening, and
(3) glomerular sclerosis (Figs. 1–3) [36]. Glomerular sclerosis may have
a nodular appearance called the ‘‘Kimmelstiel-Wilson lesion’’ and is often
associated with hyaline deposits in the glomerular arterioles reflecting the
insudation of plasma proteins, such as fibrin, immunoglobulins, and
complement into the vascular wall [36,37].
The mesangial expansion and glomerulosclerosis do not always develop
in parallel, suggesting that they may have somewhat different pathogenesis
[37]. Mesangial expansion may be directly induced by hyperglycemia, by
increased matrix production, or glycosylation of matrix proteins. In vitro
studies have demonstrated that hyperglycemia stimulates mesangial cell
matrix formation [37,38].
Nonproliferative DR is characterized by structural abnormalities of the
retinal vessels (capillaries primarily, although venules and arterioles also

Table 2
Classification of diabetic retinopathy
Classification Characteristics Impact on vision
Background (nonproliferative) Microaneurysms; venous None
retinopathy dilatation; hemorrhages; exudates
Background retinopathy Macular edema May impair vision
with maculopathy
Proliferative retinopathy Neovascularization Vision already
(pathognomonic feature); affected at this stage
fibrous proliferation;
preretinal hemorrhage;
vitreous hemorrhage
Advanced diabetic Vitreous opacities (hemorrhage and Vision already
eye disease fibrous tissues) Retinal detachment affected at this stage
Involutional retinopathy Residual scarring from previously Vision already
active proliferative retinopathy affected at this stage
1006 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

Fig. 1. Normal glomerulus (Hematoxylin-eosin stain, 100).

may be involved); varying degrees of retinal nonperfusion; retinal edema;


lipid exudates; and intraretinal hemorrhages (Figs. 4–6). Proliferative DR
may include any of the previously mentioned changes with additional
findings of optic disc or retinal or iris neovascularization. Tractional or
retinal neovascularization may cause vitreous hemorrhages (Fig. 7). Vascu-
lar endothelial growth factor (VEGF) is a potent angiogenic and mitogenic
molecule; increased VEGF is present in the retina of diabetic patients. It acts
as a permeability factor and is implicated in increased amounts of vascular
leakage and in the initiation of tumor angiogenesis [39–41].

Fig. 2. Diffuse and nodular glomerulosclerosis (Kimmelstiel-Wilson lesion; Periodic acid Schiff
stain, 40).
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1007

Fig. 3. Advanced diabetic nephropathy with diffuse and nodular mesangial expansion and
hyaline thickening of arterioles (Periodic acid Schiff stain, 100).

Pathogenesis
The pathogenesis of both DN and DR is complex and multiple factors
are involved in the process. Major risk factors for DN include the following:
Hypertension
Poor glycemic control
Ethnicity (African American, Mexican American, and Pima Indians)
Genetic susceptibility
Increased glomerular filtration rate
Increased plasma prorenin activity
Increased sodium-lithium and sodium-hydrogen countertransport
Furthermore, there is a complex interaction of these factors and the
presence of one may exacerbate the effects of another factor. First outlined
are factors that are common to both conditions and then discussed are
factors that may be specific to any one of the complications (Fig. 8).

Fig. 4. Normal fundus.


1008 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

Fig. 5. Microaneurysms.

Abnormalities related to hyperglycemia


Several epidemiologic and large prospective clinical studies have shown
a strong association between glycemic control and diabetic microvascular
complications. Glycemia-related vascular damage has been hypothesized to
be mediated through various biochemical pathways including the hexos-
amine pathway, the advanced glycation end-product formation pathway,
and the diacylglycerol (DAG)–protein kinase C (PKC) pathway. All seem to
be caused by overproduction of superoxide by the mitochondrial electron-
transport chain. The superoxide partially inhibits the glycolytic enzyme
glyceraldehydes phosphate dehydrogenase, thereby diverting upstream
metabolites from glycolysis into the four major glucose-driven signaling
pathways causing hyperglycemic damage (Fig. 9) [40,42].
Glycosylation end-products, oxidative stress, and protein kinase C
Increased activation of the DAG-PKC signal transduction pathway has
been identified in vascular tissues from diabetic animals, and in vascular cells
exposed to elevated glucose. Vascular abnormalities associated with glucose-
induced PKC activation leading to increased synthesis of DAG include
altered vascular blood flow, extracellular matrix deposition, basement
membrane thickening, increased permeability, and neovascularization [43].
Recent studies have yielded clues that may link hyperglycemia, pericyte
death, and DR. Apoptosis of retinal capillary pericytes and, to a much lesser

Fig. 6. Nonproliferative retinopathy with dot and blot hemorrhages.


A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1009

Fig. 7. Neovascularization (proliferative retinopathy).

extent, retinal capillary endothelial cells, has been demonstrated in humans


who have early DR, but not in nondiabetic control patients [44,45]. Pericyte
glutathione content, a basic defense against peroxidation, becomes depleted
in high-glucose conditions [46]. During periods of glucose fluctuation, genes
that encode products that regulate pericyte survival and death are up-
regulated [47].
It seems that a high concentration of glucose followed by a large glucose
fluctuation may be a death signal for retinal capillary pericytes. This
apoptosis leads to a cascade of events that result first in background DR,
and then with more extensive loss of pericytes and damage to endothelial
cells and with a release of more factors (eg, transforming growth factor
[TGF]-b), into the retinal milieu, an induction of distant phenomena occurs
(eg, proliferative changes of the venular endothelial cells, proliferative
diabetic retinopathy [PDR]).

Glucose

glycation Urinary protein


AGEs

=angiotensin
AT 1 receptor

Increased Efferent
glomerular arteriolar
pressure constriction

Ang I l
Ang I l

Fig. 8. Pathologic processes leading to glomerular injury and proteinuria. AGE, advanced
glycation end-product; Ang, angiotensin.
1010
+ +
NADPH NADP NAD NADH
Glucose
Sorbitol Fructose
Polyol pathway

A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036


Glucose-6-P

GFAT
Glucosamine-6-P UDP - GlcNAc
Fructose-6-P
Gln Glu
Hexosamine pathway

Pentose-5-phosphates
TK
+ +
Erythrose-4-phosphate Thiamine NADH NAD
DHAP α-Glycerol - P DAG PKC
Diacylglycerol pathway
Glyceraldehyde-3-P

+ Methylglyoxal AGEs
NAD
GAPDH O2 AGE pathway
NADH

1, 3-Diphosphoglycerate
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1011

Glycosylation of tissue proteins also may contribute to the development


of DN and other microvascular complications. In chronic hyperglycemia,
some of the excess glucose combines with free amino acids on circulating or
tissue proteins. This nonenzymatic process initially forms reversible early
glycosylation products and later irreversible advanced glycosylation end-
products by an Amadori rearrangement.
Circulating advanced glycation end-product levels are increased in
diabetics, particularly those with renal insufficiency, because advanced
glycosylation end-products are normally excreted in the urine [48]. The
net effect is tissue accumulation of advanced glycosylation end-products, in
part by crosslinking with collagen, which can contribute to the associated
renal and microvascular complications. Activation of cytokines may be
another factor involved in the matrix accumulation in DN [49–51]. As an
example, hyperglycemia increases the expression of TGF-b in the glomeruli
and of matrix proteins specifically stimulated by this cytokine [51,52]. TGF-
b may contribute to both the cellular hypertrophy and enhanced collagen
synthesis that are seen in DN [53]. The administration of an angiotensin-
converting enzyme (ACE) inhibitor to patients who have type 1 diabetes and
nephropathy lowers serum concentrations of TGF-b [54]. An inverse
correlation has been found between decreased TGF-b levels and renopro-
tection, as determined by changes in the glomerular filtration over time.
It has been proposed that activation of PKC by hyperglycemia contributes
to the renal disease and other vascular complications of diabetes [55]. Before
the discussion of its role, it is helpful to review the biochemistry and action of
protein kinases. The reversible phosphorylation of proteins is the principal
means of governing protein activity within cells. Protein kinases belong to
a large family of enzymes that contain a similar 250–amino acid catalytic
(kinase) domain but differ according to the amino acids on either side of the
kinase domain. PKC activity is increased in glomeruli, retina, aorta, and
heart of diabetic animals. This elevation in activity is probably caused by
enhanced de novo synthesis of DAG, a major endogenous activator of PKC.
A role for PKC in the pathogenesis of DN is suggested by the results of
several animal experiments. First, PKC is activated in glomeruli isolated
from diabetic rats [56]. Second, activated PKC (especially the activated beta
isoform) in glomerular epithelial cells of induced diabetic rats participate in

=
Fig. 9. Potential mechanism by which hyperglycemia can cause tissue damage. Hyperglycemia-
induced mitochondrial superoxide overproduction partially inhibits the glycolytic enzyme
GAPDH, thereby diverting upstream metabolites from glycolysis into glucose-driven signaling
pathways of glucose overuse. AGE, advanced glycation end-product; DAG, diacylglycerol;
DHAP, dihydroxyacetone phosphate; GAPDH, glyceraldehyde phosphate dehydrogenase;
GFAT, glutamine fructose-6-phosphate amidotransferase; PKC, protein kinase C; UDP, uridin
diphosphate. (Adapted from Hammes HP, Du X, Edelstein D, Taguchi T, Matsumura T, Ju Q,
et al. Benfotiamine blocks three major pathways of hyperglycemic damage and prevents
experimental diabetic retinopathy. Nat Med 2003;9:294–9; with permission.)
1012 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

the glycated albumin-induced stimulation of basement membrane type IV


collagen production by glomerular endothelial cells [57].
Therapies aimed at lower PKC activity are in development. As an
example, treatment of diabetic rats with d-a-tocopherol, which inhibits
PKC activation, prevents glomerular hyperfiltration and minimizes the
development of proteinuria (2.4 versus 9.1 mg/d in control diabetic rats,
versus 1.2 mg/d in nondiabetic rats) [58,59]. Preferential activation of the
PKC beta isoform by elevated glucose is reported to occur in a variety of
vascular tissues. This has led to the development of LY333531, a PKC beta
isoform specific inhibitor, which has shown potential in animal models to be
an orally effective and nontoxic therapy able to produce significant improve-
ments in DR, DN, neuropathy, and cardiac dysfunction [43]. Additionally,
the antioxidant vitamin E has been identified as an inhibitor of the DAG-
PKC pathway, and shows promise in reducing vascular complications in
animal models of diabetes. Given the overwhelming evidence indicating a role
for PKC activation in contributing to the development of diabetic vascular
complications, pharmacologic therapies that can modulate this pathway,
particularly with PKC isoform selectivity, show great promise for treatment
of vascular complications, even in the presence of hyperglycemia [58–61].
Aldose reductase
Aldose reductase is an enzyme that converts sugars into their respective
alcohols. For example, glucose is converted into sorbitol and galactose is
converted into galactitol. Because sorbitol and galactitol cannot easily diffuse
out of cells, their intracellular concentration increases. Osmotic forces then
cause water to diffuse into the cell, which results in electrolyte imbalance. The
resultant damage to lens epithelial cells, which have a high concentration of
aldose reductase, is responsible for the cataracts seen in children, in
experimental animals with galactosemia, and in animals with experimental
diabetes mellitus [62]. In addition, secondary metabolic changes in the target
tissue, such as depletion of myoinositol, lead to tissue damage. Because aldose
reductase also is found in high concentration in retinal pericytes and Schwann
cells, some investigators suggest that DR and DN may be caused by aldose
reductase–mediated damage. Strong support for this theory is that aldose
reductase inhibitors inhibit both cataract formation and pericyte loss in
experimental animals. Despite these theoretic benefits, however, clinical trials
have failed to show a reduction in the incidence of DR or of DN by aldose
reductase inhibitors, possibly because an effective aldose reductase inhibitor
that has few systemic side effects has yet to be developed [62].

Abnormalities independent of hyperglycemia


Growth factors
Interest in the role of growth factors, independent of glycemia, in DN
and DR stem from the fact that growth of vascular and matrix tissue is an
important component of the pathology.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1013

In the kidney, growth hormone has been implicated in the early stage
of hypertrophy and hyperfiltration. As discussed later, other growth factors
stimulated by angiotensin II may play a role in the increase in intra-
glomerular matrix hypertrophy. Angiotensin II itself has direct and potent
cellular growth-promoting actions [63]. In addition, it stimulates production
of important growth factors, such as TGF-b. The latter plays a key role in
extracellular matrix formation in the mesangium of the kidney.
Activation of the renin-angiotensin system also leads to a selective
constriction of the efferent arteriole (compared with the afferent one). This
leads to an increase in intraglomerular pressure, an important contributor to
renal damage. Indeed, the selective dilatation of the efferent arterioles has
been suggested as a major factor in the greatly beneficial effects of ACE
inhibitors and angiotensin receptor blockers in DN.
Furthermore, there is a decreased prevalence and possibly a regression of
DR in patients with growth hormone deficiency. Growth hormone may play
a causative or at least an important supportive role in the development and
progression of diabetic vascular complications. Poulsen noted reversal of
florid DR in a woman who had postpartum hemorrhagic necrosis of the
pituitary gland (panhypopituitarism). More recently, growth hormone
deficiency was found to be somewhat protective against retinopathy [64].
Administration of insulin-like growth factor also has been associated with
retinal changes, although these are not entirely specific for DR.
It has been recently recognized that vasoproliferative factors, released by
the retina itself, retinal vessels, or the retinal pigment epithelium, which may
induce neovascularization. Vascular endothelial growth factor, which
inhibits the growth of retinal endothelial cells in vitro, has recently been
implicated in DR [65] and has also been found to be increased in the vitreous
fluid of patients with DR [66].

Risk factors and clinical predictors of diabetic nephropathy and


retinopathy
Glycemic control
Diabetic nephropathy is more likely to develop in patients with poor
glycemic control. Patients with type 1 diabetes whose hemoglobin A1c
concentration is maintained below 8.1% are at much lower risk for renal
disease [21]. Randomized clinical trials have confirmed the predictive value
of poor control compared with good control in determining the risk of
nephropathy and retinopathy. The United Kingdom Prospective Diabetes
Study (UKPDS) of patients with type 2 diabetes found that fewer patients
treated with intensive versus conventional therapy had progression of
microalbuminuria (27% versus 39%) and proteinuria (7% versus 13%)
over 15 years of follow-up [67]. The Diabetes Control and Complications
Trial (DCCT) showed that combined, intensive therapy reduced the
1014 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

occurrence of microalbuminuria (urinary albumin excretion of 40 mg


per 24 hours) by 39% and that of albuminuria (urinary albumin excretion of
300 mg per 24 hours) by 54% [68].
The DCCT showed 76% reduction in the rate of development of any
retinopathy and an 80% reduction in progression of established retinopathy
versus those with conventional control. These risk reductions, achieved
at a median hemoglobin A1c level difference of 9.1% for conventional
treatment versus 7.3% for intensive treatment, have been maintained
through 7 years of continued follow-up in the Epidemiology of Diabetes
Interventions and Complications study, even though the difference in mean
hemoglobin A1c levels of the two former randomized treatment groups was
only 0.4% 1 year after conclusion of the DCCT (8.3% in the former
conventional treatment group versus 7.9% in the former intensive treatment
group); continued to narrow; and became statistically nonsignificant by 5
years (8.1% versus 8.2%, P = .09). The further rate of progression of
complications from their levels at the end of the DCCT remains less in the
former intensive treatment group. The benefits of 6.5 years of intensive
treatment extend well beyond the period of its most intensive implementa-
tion. Intensive treatment should be started as soon as is safely possible after
the onset of type 1 diabetes mellitus and maintained thereafter, aiming for
a practicable target hemoglobin A1c level of 7% or less [66,69,70].
Wisconsin epidemiologic study of DR was a population-based study in
southern Wisconsin conducted to determine the prevalence and severity of
DR and associated risk variables. It showed a positive correlation between
severity of retinopathy and high levels of glycosylated hemoglobin after 10
years of diabetes [10].
For patients with advanced retinopathy, however, even the most rigorous
control of blood glucose may not prevent progression. Even patients
attaining normoglycemia by pancreatic transplantation continue to show
progression of retinopathy [71].

Duration of disease
In various randomized controlled trials, the total duration of disease has
been found to be the strongest predictor of development and progression of
DR [72]. In the Wisconsin epidemiologic study of DR, prevalence in
younger-onset patients with diabetes was 8%, 25%, 60%, and 80% at 3, 5,
10, and 15 years after diagnosis, respectively [10].

Blood pressure
Some research indicates that elevated systolic blood pressure is a moder-
ate risk factor for both DN and DR, more so for the latter [72]. In UKPDS
trial tight blood pressure control was shown to cause 34% reduction in
progression of retinopathy and a 47% reduced risk of deterioration in visual
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1015

acuity of three lines in association with a 10/5 mm Hg reduction in blood


pressure [13]. The appropriate Blood Pressure Control in Diabetes Trial was
conducted to determine whether intensive blood pressure control (diastolic
blood pressure goal of 75 mm Hg) offers additional benefit over moderate
control (diastolic blood pressure between 80 and 89 mm Hg). Intensive
therapy showed a lower incidence of deaths; however, there was no
difference with regards to progression of DR [73].

Genetics
The increased synthesis of angiotensin II plays an important role in
initiation and progression of DN by affecting hemodynamic and non-
hemodynamic mechanisms [74]. Studies have shown that an inversion (I)-
deletion (D) polymorphism of the ACE gene (ACE/ID) is associated with
the level of circulating ACE and with an increased risk of coronary heart
disease in nondiabetic patients [75]. Yoshida et al [76] followed 168
proteinuric patients with type 2 diabetes for 10 years and found in an
analysis of the clinical course of the three ACE genotypes that most patients
with the DD genotypes (95%) progressed to ESRD within 10 years. Two
recent studies have confirmed that the D allele has a deleterious effect on
renal function in patients with type 2 diabetes [77,78].
Several studies have shown that the likelihood of developing DN is
markedly increased in patients with a diabetic sibling or parent who has
DN; these observations have been made in both type 1 and type 2 diabetes
[79–81]. One report, for example, evaluated Pima Indians in which two
successive generations had type 2 diabetes [79]. The likelihood of the
offspring developing overt proteinuria was 14% if neither parent had
proteinuria, 23% if one parent had proteinuria, and 46% if both parents
had proteinuria.
One component of the genetic risk may be the ACE gene genotype. In
patients with type 2 diabetes, the DD polymorphism has been associated
with an increased risk for the development of DN, more severe proteinuria,
a greater likelihood of progressive renal failure, and enhanced mortality on
dialysis [76,82,83]. A critical review of 19 studies examining a possible link
between the ACE gene genotype and DN failed to confirm an association
among whites with either type 1 or type 2 diabetes, but could not exclude
a possible association in Asians [84]. Unfortunately, because of poor
methodology, no definite conclusions could be drawn.
An enhanced risk may also be caused by inheritance of one allele of the
aldose reductase gene, the rate-limiting enzyme for the polyol pathway. In
a controlled study of patients with type 1 diabetes, homozygosity for the Z-2
allele was significantly associated with an odds ratio of 5.25 for the presence
of nephropathy [85].
Many patients with salt-sensitive essential hypertension have an elevation
in red cell sodium-lithium countertransport; increased sodium-hydrogen
1016 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

exchange has also been linked to the development of hypertension. These


abnormalities are thought to be markers for enhanced sodium transport at
sites that might induce a rise in blood pressure, such as the kidney or
vascular smooth muscle. Some studies have suggested that type 1 diabetics
with nephropathy have higher rates of sodium-hydrogen exchange and
red cell sodium-lithium countertransport than those without renal disease
[86–88]. Sodium-hydrogen exchange activity is concordant among type 1
diabetic siblings, suggesting that this activity is genetically determined [89].

Glomerular filtration rate


Approximately half of patients with type 1 diabetes of less than 5 years
duration have an elevated GFR that is 25% to 50% above normal. Those
patients with glomerular hyperfiltration seem to be at increased risk for
diabetic renal disease [90,91]. In one prospective study, for example, patients
with type 1 diabetes and a GFR above 125 mL/min had a risk of developing
microalbuminuria within 8 years of approximately 50% versus only 5% in
patients with a lower GFR that was similar to that seen in nondiabetics [90].
The glomerular hyperfiltration in type 1 diabetics is typically associated
with glomerular hypertrophy and increased renal size [92]. The association
between these hemodynamic and structural changes and the development of
DN may be related both to intraglomerular hypertension (which drives the
hyperfiltration) and to glomerular hypertrophy (which also increases wall
stress). Therapy aimed at reversing these changes (with aggressive control of
plasma glucose concentration early in the course of the disease [92], dietary
protein restriction, and antihypertensive therapy) may slow the rate of
progression of the renal disease.
The findings in type 2 diabetes are somewhat different. Up to 45% of
affected patients initially have a GFR that is more than 2 standard
deviations above age-matched nondiabetic and obese controls [93,94]. The
degree of hyperfiltration (averaging 117 to 133 mL/min), however, is less
than that seen in type 1 diabetics. Type 2 diabetics are also older and more
likely to have atherosclerotic vascular disease, which limits increases in both
glomerular filtration and glomerular size [95].

Ethnicity
The incidence and severity of DN are increased in blacks (threefold to
sixfold compared with whites), Mexican-Americans, and Pima Indians with
type 2 diabetes [20,96,97]. This observation in such genetically disparate
populations suggests a primary role for socioeconomic factors, such as diet,
poor control of hyperglycemia, hypertension, and obesity [98].
As an example, there seems to be an important association between
hypertension and disease progression in black patients with type 2 diabetes.
A cross-sectional study found that GFR was normal in patients who were
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1017

normotensive; in comparison, hypertension was associated with a decline in


renal function, particularly if it developed after the onset of diabetes and
the patient had been diabetic for more than 10 years [99]. It is not clear,
however, if the hypertension worsened the renal disease or was simply
a marker for more severe renal involvement.
The importance of genetic influences, however, in the racial propensity to
DN cannot be excluded. Even when adjustments are made for the increased
incidence of hypertension and lower socioeconomic status in blacks, there
still seems to be as much as a 4.8-fold increase in the risk of ESRD caused by
DN in blacks [97]. This seems to occur only in type 2 diabetes, with no
increase in risk seen with type 1 diabetes.

Relationship between diabetic nephropathy and retinopathy


Patients with nephropathy and type 1 diabetes almost always have other
signs of diabetic microvascular disease, such as retinopathy and neuropathy
[18,19]. By the time advanced retinopathy has occurred, there are histologic
changes in the glomeruli and increased protein excretion that is at least in
the microalbuminuric range [100]. Renal disease as evidenced by pro-
teinuria, elevated blood urea nitrogen, and elevated blood creatinine is an
excellent predictor of the presence of retinopathy [101]. Even patients who
have microalbuminuria are at high risk of developing retinopathy [102].
Similarly, 35% of patients symptomatic for retinopathy have proteinuria,
elevated blood urea nitrogen, or elevated creatinine.
The relationship between DN and DR is less predictable in type 2
diabetes. In one study of 35 patients with diabetes and significant pro-
teinuria (>300 mg/d), 27 (77%) were found to have DN [103]. DR was
present in 15 (56%) of the 27 and in 0 of the 8 individuals without DN,
thereby resulting in a sensitivity and specificity of 40% and 100%,
respectively. Further analysis of some of these patients plus additional type
2 diabetics with proteinuria found that, among those without retinopathy,
approximately 30% did not have DN on renal biopsy [104].
Type 2 diabetics with marked proteinuria and retinopathy most likely
have DN, whereas those without retinopathy have a high incidence of
nondiabetic glomerular disease.

Pregnancy
In women who begin a pregnancy without retinopathy, the risk of
developing nonproliferative DR is about 10%. Further, those with non-
proliferative DR at the onset of pregnancy and those who have or who
develop systemic hypertension tend to show progression, with increased
hemorrhages, cotton wool spots, and macular edema [105]. Fortunately,
usually some regression occurs after delivery. About 4% of pregnant
women who have nonproliferative DR progress to PDR. Those with
untreated PDR at the onset of pregnancy frequently do poorly unless they
are treated using panretinal photocoagulation. Finally, patients who have
1018 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

previously had successfully treated PDR usually do not worsen during


pregnancy. Women who begin pregnancy with poorly controlled diabetes,
however, but who then are suddenly brought under strict control frequently
have severe, rapid worsening of their retinopathy and do not always recover
after delivery [105].

Other renal diseases


Proteinuria in diabetes mellitus is occasionally caused by a glomerular
disease other than DN. As examples, membranous nephropathy, minimal
change disease, IgA nephropathy, Henoch-Schönlein purpura, thin base-
ment membrane disease, and a proliferative glomerulonephritis have all
been described [19,95,104,106–113].
Major clinical clues suggesting nondiabetic kidney disease include onset
of proteinuria less than 5 years from the documented onset of diabetes,
acute onset of renal disease, presence of an active urine sediment containing
red cells and cellular casts, and the absence of DR or DN. Lack of
retinopathy in type 2 diabetes does not preclude DN, however, which
remains the most likely diagnosis [19,103,104].

Diagnosis
Diagnosis of diabetic nephropathy
Guidelines for systematic screening have been developed because patients
with nephropathy are often asymptomatic and because a number of effective
intervention strategies can slow disease progression (Table 3).
Screening for DN needs to be a routine component of diabetes care.
The American Diabetes Association recommends yearly screening for
individuals with type 2 diabetes and yearly screening for those with type 1
diabetes after 5 years’ duration of disease (but not before puberty). Several
screening techniques are available: the albumin:creatinine ratio from
a random spot urine collection, a 24-hour urine collection for albuminuria
and creatinine (allowing calculation of creatinine clearance as well), or
a timed (eg, overnight or 3- to 4-hour) urine collection are all acceptable.
Positive results need to be confirmed with a second measurement because of
the high variability in albumin excretion in people with diabetes. Use of
urine dipsticks for microalbuminuria (on fresh morning specimens) is
reasonable for initial screens, but these findings are semiquantitative, and
positive tests should be followed-up by a 24-hour or timed urine collec-
tion. Microalbuminuria is considered to be present if urinary albumin
excretion is 30 to 300 mg per 24 hours (equivalent to 20–200 lg/min on
a timed specimen or 30–300 mg/g creatinine on a random sample) (Fig. 10).
Short-term hyperglycemia, exercise, urinary tract infections, marked hy-
pertension, heart failure, and acute febrile illness can cause transient
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1019

Table 3
Interpretation of urinary albumin excretion (based on different assessment methods)
Urine albumin: Morning urine
Urinary AER Urinary AER creatinine ratio albumin
Finding (mug/min) (mg/24 hr) (mg/g) concentration (mg/L)
Normal \20 \30 \30 \30
Microalbuminuria 20–200 30–300 30–300 30–300
Macroalbuminuria >200 >300 >300 >300
Abbreviation: AER, albumin excretion rate in a timed specimen.
From American Diabetes Recommendations 2003. Diabetic nephropathy. Diabetes Care
2003;26(Suppl 1); with permission from The American Diabetes Association.

elevations in urinary albumin excretion. There is also marked day-to-day


variability in albumin excretion, so at least two to three collections done in
a 3- to 6-month period should show elevated levels before a patient is
designated as having microalbuminuria and treatment is initiated.

Detection of diabetic retinopathy


Guidelines for systematic screening have been developed because patients
with retinopathy are often asymptomatic, and photocoagulation treatment
is more effective in reducing visual loss when applied at specific, frequently
asymptomatic stages of retinopathy. Regular dilated eye examinations are
effective in detecting and treating vision-threatening DR.
Current guidelines suggest that diabetic patients have an initial dilated
and comprehensive eye examination by an ophthalmologist shortly after the
diagnosis of diabetes is made in patients with type 2 diabetes, and within 3
to 5 years after the onset of type 1 diabetes (but not before age 10 years)
[114]. Any patient with visual symptoms or abnormalities should be referred
for ophthalmologic evaluation.
Subsequent examinations for both type 1 and type 2 diabetic patients
should be repeated annually by an ophthalmologist who is experienced in
diagnosing the presence of DR and is knowledgeable about its management
[114]. More frequent examinations are needed especially in patients who have
progressive disease and severe nonproliferative disease. Women with diabetes
should have a comprehensive eye examination when planning pregnancy and
during the first trimester of pregnancy and should be followed closely during
pregnancy.
Patients with any level of macular edema, severe nonproliferative
retinopathy, or any proliferative retinopathy require the prompt care of an
ophthalmologist who is knowledgeable and experienced in the management
and treatment of DR.
The most sensitive way to detect retinopathy is by fundus photography
through a dilated pupil, involving seven stereoscopic 30-degree standard
fields. Proper fundus photography requires a photographer skilled in
1020 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

Test for microalbuminuria

No
+ for albumin

Yes
Condition that may invalidate
urine albumin excretion?

Yes No

No Treat and/or wait until


resolved. Repeat test
+ for protein?

Yes

Repeat microalbuminuria
test twice within 3-6 month period.

Rescreen No
2 of 3 tests positive?
in one year
Yes

Microalbuminuria, begin treatment

Fig. 10. Screening for microalbuminuria. (From Molith M, Franz M, Keane W, Megensen CE,
Parring H, Steffes M. Clinical practice recommendations: diabetic nephropathy. Diabet Care
2003;26(Suppl 1):S96; with permission.)

obtaining the rigorously defined and technically challenging photographic


fields of appropriate quality plus a reader skilled in the interpretation of the
photographs. If either of these components is not available or does not meet
the defined standards, a dilated ophthalmic examination by an experienced
ophthalmologist is recommended. Such examinations should be performed
by ophthalmologists because the difficulty in assessing DR is such that in one
study the rates of serious errors in assessment were 52% for general
internists, 50% for medical residents, and 33% for diabetologists but were
only 9% for general ophthalmologists and 0% for retinal specialists [115].
New screening technologies include dynamic light scattering [116], Raman
spectroscopy [117], autofluorescence imaging [118], and Doppler flowmetry
[119]. All use laser light to measure various molecular structures and
physiologies to detect abnormalities before the advanced stages of histopa-
thology. These can aid in earlier diagnosis and treatment.
Teleopthalmology is an emerging new tool for diagnosis. Gomez-Ulla
et al [120] obtained retinal images and sent by internet for grading by an
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1021

ophthalmologist to a remote reference center. The results were compared


with findings by an ophthalmologist on direct eye examination and found to
have 100% agreement.

Microalbuminuria and proteinuria


Microalbuminuria and proteinuria are part of the spectrum of clinical
manifestation of DN. Microalbuminuria is defined as the presence of
urinary albumin above the normal but below the detectable range with the
conventional urine dipstick methodology. It is recognized as a complication
of diabetes because of changes in the kidney secondary to hyperglycemia. In
diabetic patients, this consists of a urinary albumin excretion rate of 20 to
200 lg/min (30–300 lg/mg of creatinine on a spot urine sample or 30–200
mg per 24 hours), because rates within this range have been shown to predict
the progression of DN. In contrast the terms ‘‘proteinuria,’’ ‘‘albuminuria,’’
and ‘‘overt nephropathy’’ are used when the urine dipstick is positive or the
albumin excretion is greater than 200 mg per 24 hours. The difference
between microalbuminuria and overt proteinuria is essentially a matter of
degree. Microalbuminuria may be transient or reversible in its early stages,
however, whereas proteinuria usually progresses over a variable period to
ESRD.
Several studies have demonstrated that microalbuminuria is a risk factor
for cardiovascular events [121–126]. Recent data suggest that it may occur
even in nondiabetics, may be a precursor of cardiovascular disease, and may
be related to insulin resistance [127–130]. Microalbuminuria may precede
and predict the development of type 2 diabetes [131], and the progression of
microalbuminuria is associated with a worsening prognosis for cardiovas-
cular disease risk [124].
Albuminuria clusters with other cardiovascular disease risk factors,
particularly dyslipidemia, left ventricular hypertrophy, and the absence of
nocturnal drops in both systolic and diastolic blood pressures. Elevated
systolic blood pressure is a significant determining factor in the development
of microalbuminuria and the progression of albuminuria in type 2 diabetes.
Data also suggest that microalbuminuria reflects increased leakage of
albumin across the endothelial barrier and is a clinically easily measurable
indicator of endothelial integrity. Indeed, patients with microalbuminuria
are likely to have several biochemical and functional abnormalities of
endothelial function.

Clinical disease progression


The likelihood of progression from microalbuminuria to overt nephrop-
athy (positive urine dipstick for protein) is determined by the type and
duration of diabetes. In type 1 diabetics, clinical renal involvement begins 10
to 15 years after the diagnosis of diabetes; patients without proteinuria after
20 years have a risk of developing overt renal disease of 1% per year [132].
1022 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

Patients who progress are more likely to have higher hemoglobin A1c values
and a higher blood pressure than nonprogressors [133,134]. A retrospective
study measured both albumin excretion and glycemic control in 1613 patients
with type 1 diabetes [134]. The risk of having microalbuminuria increased
abruptly at hemoglobin A1c value above 8.1%. In a prospective report,
multivariate analysis of 1134 patients with type 1 diabetes found that higher
values for hemoglobin A1c were independent risk factors for progression to
microalbuminuria [135]. Progressors also had higher systolic and diastolic
blood pressures (123/75 mm Hg versus 118/73 mm Hg for nonprogressors).
Progression from microalbuminuria to overt nephropathy within a 10-year
period occurs in 20% to 40% of white patients with type 2 diabetes [136,137].
Risk factors contributing to progression include hyperglycemia, hyperten-
sion, and cigarette smoking. Other studies of Pima Indians and Israeli patients
with type 2 diabetes have found a 4- to 5-year rate of progression to overt
proteinuria of 37% to 42% [138,139]. A 4-year follow-up of 34 patients with
overt proteinuria found a mean loss in GFR of 0.93 mL/min/mo, a rate similar
to that observed in patients with type 1 diabetes [138].

Management
Prevention
Strong clinical trial data suggest that both DN and DR can be prevented
by good glycemic control. In addition to glycemic control data suggest
that good blood pressure control may also decrease the onset of DR and
DN. Furthermore, the development of microalbuminuria is delayed by
ACE inhibitors [140,141].
The DCCT, a randomized, multicenter, controlled clinical trial, demon-
strated that intensive treatment of type 1 diabetes-decreased the progression
of nephropathy and retinopathy. The incidence of microalbuminuria was
significantly reduced by 39% in three combined cohorts, by 34% in the
primary-prevention cohort, and by 43% in the secondary-intervention
cohort [68].
The UKPDS, a randomized, multicenter, controlled clinical trial, demon-
strated that a policy of intensive treatment with goal of meticulous glycemic
control could decrease complications of type 2 diabetes. Patients assigned to
the intensive policy A1c of 7% had a significant 25% risk reduction in
microvascular end points (P \ .01) compared with those in the conventional
policy A1c of 7.9%. At 9, 12, and 15 years follow-up the risk reduction in the
appearance of microalbuminuria was 24%, 33%, and 30%, respectively [67].
The benefit of antihypertensive therapy with an ACE inhibitor in type 1
diabetes can be demonstrated early in the course of the disease when
microalbuminuria is the only clinical manifestation. In one study, the
administration of an ACE inhibitor to normotensive type 1 diabetics
with microalbuminuria decreased both albumin excretion and at 2 years
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1023

progression of disease when compared with patients treated on placebo


[142,143]. The Heart Outcomes Prevention Evaluation (HOPE) study, an
international randomized trial, was designed to evaluate the effects of the
ACE inhibitor ramipril and vitamin E in patients at high risk for cardiovas-
cular events. Ramipril use was associated with a significant 25% reduction in
risk for the composite end point of myocardial infarction, stroke, or
cardiovascular death after a median follow-up period of 4.5 years. The
Microalbuminuria, Cardiovascular, and Renal Outcomes in HOPE, a sub-
study in this patient population, showed that ramipril treatment was
associated with a decreased risk of development of overt nephropathy [144].

Treatment of diabetic nephropathy


Studies of blood pressure control and the evolution and progression of
DN have focused on ACE inhibitors or more recently on angiotensin
receptor blockers. The value of ACE inhibitors in patients with established
DN was demonstrated in a landmark study with captopril [145]. Four
hundred nine patients with overt proteinuria and creatinine less than or equal
to 2.5 were randomized to therapy with either captopril or placebo. With
equivalent blood pressure control, patients treated with captopril had
a slower rate of increase in creatinine concentration and a lesser likelihood
of progressing to ESRD or death [145,146]. Beneficial response to captopril,
which was seen in both hypertensive and normotensive subjects, is consistent
with smaller studies, which suggested that antihypertensive therapy, partic-
ularly with an ACE inhibitor, slowed the rate of progression in DN [141,147].
Captopril treatment was associated with a 50% reduction in the risk of the
combined end points of death, dialysis, and transplantation that was
independent of blood pressure.
There has been less information on the effect of antihypertensive therapy
with ACE inhibitors in patients with nephropathy caused by type 2 diabetes,
although a similar benefit seems to be present. More data from large clinical
trials are available on the efficacy of angiotensin receptor blockers. In the
UKPDS each 10 mm Hg reduction in systolic pressure was associated with
a 12% risk reduction in diabetic complications (P \ .001); the lowest risk
occurred at a systolic pressure below 120 mm Hg. There was no difference
between captopril and atenolol in progression of complications [148].
Similar results were found in the ALLHAT study [149]. HOPE, however,
showed decreased proteinuria with ACE inhibitors [144].
Brenner et al [150] assessed the role of the angiotensin II receptor antagonist
losartan in 1513 patients with type 2 diabetes and nephropathy in a random-
ized, double-blind study comparing losartan with placebo, both taken in
addition to conventional antihypertensive treatment over a mean of 3.4 years.
Losartan significantly reduced the incidence of a doubling of the serum
creatinine concentration (risk reduction, 25%) and ESRD (risk reduction,
28%) but had no effect on the rate of death. The benefit exceeded that
1024 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

attributable to changes in blood pressure. The composite of morbidity and


mortality from cardiovascular causes was similar in the two groups, although
the rate of first hospitalization for heart failure was significantly lower with
losartan (risk reduction, 32%). The level of proteinuria declined by 35% with
losartan. Losartan conferred significant renal benefits in patients with type 2
diabetes and nephropathy, and it was generally well tolerated.
In another large study, the angiotensin II receptor blocker irbesartan was
effective in protecting against the progression of nephropathy caused by
type 2 diabetes. This protection was independent of the reduction in blood
pressure it causes [151]. Irbesartan was also shown to be renoprotective
independently of its blood pressure lowering effect in patients with type 2
diabetes and microalbuminuria, slowing the progression to overt proteinuria
[152].
The CALM study was conducted to assess and compare the effects of
candesartan or lisinopril, or both, on blood pressure and urinary albumin
excretion in patients with microalbuminuria, hypertension, and type 2
diabetes. In this prospective, randomized, double-blind study there was a 4-
week placebo run in period and 12 weeks monotherapy with candesartan
or lisinopril followed by 12 weeks monotherapy or combination treatment.
At 24 weeks the mean reduction in diastolic blood pressure with combina-
tion treatment (16.3 mm Hg, 13.6–18.9 mm Hg, P \ . 001) was significantly
greater than that with candesartan (10.4 mm Hg, 7.7–13.1 mm Hg,
P \ .001) or lisinopril (mean 10.7 mm Hg, 8–13.5 mm Hg, P \ .001).
Furthermore, the reduction in urinary albumin:creatinine ratio with combi-
nation treatment (50%, 36%–61%, P \ .001) was greater than with
candesartan (24%, 0%–43%, P = .05) and lisinopril (39%, 20%–54%,
P \ .001). In conclusion, combination treatment was found to be well
tolerated and more effective in reducing blood pressure [153]. The primary
aim of the Fosinopril Versus Amlodipine Cardiovascular Events Random-
ized Trial was to compare the effects of fosinopril and amlodipine on serum
lipids and diabetes control in non–insulin-dependent diabetes mellitus
patients with hypertension. A total of 380 hypertensive diabetics were ran-
domly assigned to open-label fosinopril (20 mg/d) or amlodipine (10 mg/d)
and followed for up to 3.5 years. If blood pressure was not controlled, the
other study drug was added. Both treatments were effective in lowering
blood pressure [154]. Combination therapy is frequently used in clinical
practice for optimal blood pressure control.
Once the plasma creatinine is elevated indicating low clearance the
prognosis worsens significantly. Patients at this stage of nephropathy are
difficult to manage because of the presence of other complications including
cardiovascular disease. In addition, the pharmacokinetics of insulin and
other medications change because of the decreased kidney metabolism and
clearance. Metformin is contraindicated and thiazolidinediones become
difficult to use because of fluid retention. Both hypoglycemia and hyper-
glycemia are frequent.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1025

The management of ESRD caused by DN is beyond the scope of this


article. Data suggest that early referral to a specialist experienced in
managing such patients improves outcomes [155]. As the disease progresses
patients should be prepared for dialysis. In general, patients have better
outcomes with peritoneal than hemodialysis and better outcomes with
transplantation than dialysis. For patients with type 1 diabetes a combined
kidney pancreas transplant, apart from eliminating the need for insulin,
leads to better outcomes than kidney transplantation alone.

Treatment of diabetic retinopathy


Medical therapy
The success of laser photocoagulation in treating DR is well established
and medical therapy has very little role other than a supportive one.
Nevertheless, several clinical trails have been performed to develop medical
therapy. Although aspirin inhibits platelet secretion and aggregation, it does
not influence the progression of retinopathy, affect visual acuity, or
influence the incidence of vitreous hemorrhages [156]. The Ticlopidine
Microangiopathy of Diabetes Study Group in France examined the effect of
ticlopidine, an inhibitor of adenosine diphosphate–induced platelet aggre-
gation, showing that was associated with a sevenfold decrease in micro-
aneurysm count during 3 years of follow-up compared with placebo in
insulin-treated patient with no benefit in non–insulin-treated patients [157].
This is only one study showing a statistical benefit, however, but it was
not performed long enough to show a clinical benefit. It is not routinely
prescribed in United States.

Surgical therapy
Several multicenter, prospective, randomized, controlled studies have
demonstrated that intervention with laser photocoagulation surgery or
vitrectomy may preserve vision in certain patients with DR. These studies
are discussed next.

Panretinal photocoagulation
Panretinal photocoagulation is the treatment of choice for high-risk
retinopathy. The Diabetic Retinopathy Study first established the benefit in
treatment of eyes with high-risk criteria with proliferative retinopathy. Laser
panretinal photocoagulation significantly reduced the likelihood that an eye
with high-risk characteristics progresses to severe visual loss, up to greater
than a 50% reduction in visual loss [7]. Eyes with high-risk characteristics
are defined as those with neovascularization of the disk greater than half the
disk area, those with any neovascularization of the disk and vitreous
hemorrhage, or those with neovascularization elsewhere greater than half
the disc area and vitreous or preretinal hemorrhage.
1026 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

The mechanism of action of panretinal photocoagulation is still unknown.


Some investigators have hypothesized that panretinal photocoagulation
decreases the production of vasoproliferative factors by eliminating some
of the hypoxic retina or by stimulating the release of antiangiogenic factors
from the retinal pigment epithelium [158]. An alternative hypothesis is that
chronic hypoxia stimulates neovascularization by causing vessel dilatation
resulting in endothelial cell proliferation. By thinning the retina, panretinal
photocoagulation increases oxygenation of the remaining retina as it enables
an increased diffusion of oxygen from the choroid and so decreases
vasodilatation [159]. Finally, others suggest that panretinal photocoagulation
leads to an increase in vasoinhibitors by stimulating the retinal pigment
epithelium to produce inhibitors of vasoproliferation [160].
The number of burns necessary to achieve these goals has not been
established. Some retinal specialists believe that there is no upper limit to the
total number of burns and that treatment should be continued until regression
occurs [161]. The only prospective, controlled study, however, found that eyes
that received supplementary treatment had no difference in reduction in risk
factors or better visual acuity than did those that only received standard
panretinal photocoagulation [162]. About two thirds of eyes with high-risk
characteristics that receive panretinal photocoagulation have regression of
their high-risk characteristics by 3 months after treatment.

Photocoagulation for treatment of macular edema


Patients who have macular edema have the best prognosis for improved
vision if they have circinate retinopathy of recent duration or focal, well-
defined leaking areas and good capillary perfusion surrounding the foveal
avascular zone. Patients with an especially poor prognosis have dense lipid
exudates in the center of the foveola. Other poor prognostic signs include
diffuse edema with multiple leaking areas, extensive central capillary
nonperfusion, increased blood pressure, and cystoid macular edema [163].
Nevertheless, the Early Treatment Diabetic Retinopathy Study Research
Group (ETDRS) showed that even eyes with these adverse findings
benefited from treatment when compared with control eyes [164].
Patz et al [163] was the first to show that argon laser photocoagulation
decreases or stabilizes macular edema. Later, the ETDRS confirmed these
results. The ETDRS defined clinically significant macular edema as retinal
thickening involving the center of the macula, or hard exudates within 500
lm of the center of the macula or an area of macular edema greater than one
disk area but within one disc diameter of the center of the macula.
The treatment strategy is to treat all leaking microaneurysms further than
500 lm from the center of the macula and to place a grid of 100 to 200 lm burns
in areas of diffuse capillary leakage and in areas of capillary nonperfusion.
After 3 years of follow-up, 15% of eyes with clinically significant macular
edema had doubling of the visual angle as opposed to 32% of untreated
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1027

control eyes [164]. The ETDRS also showed that panretinal photocoagulation
should not be given to eyes with clinically significant macular edema unless
high-risk criteria are present [165]. An acceptable alternative treatment to the
ETDRS strategy is a grid treatment [166].
Panretinal photocoagulation has significant complications: it often causes
decreased visual acuity by increasing macular edema [167] or by causing
macular pucker. Fortunately, the edema frequently regresses spontaneously
over 6 months, but the visual field is usually moderately, yet permanently,
decreased. Color vision and dark adaptation, which are often already
impaired, are worsened by panretinal photocoagulation [168]. For this
reason, panretinal photocoagulation is not recommended for patients with
background DR, who should nevertheless be followed-up closely to detect
any disease progression.

Vitrectomy
Vitrectomy, introduced by Machemer et al [169], plays a vital role in the
management of severe complications of DR. The major indications are
nonclearing vitreous hemorrhage, macular involving or threatening traction
retinal detachment, and combined traction-rhegmatogenous retinal detach-
ment. Less common indications are macular edema with a thickened and taut
posterior hyaloid [170], macular heterotopia, epiretinal membrane, severe
preretinal macular hemorrhage, and neovascular glaucoma with cloudy
media.

Summary
There has been much progress in the understanding of the pathogenesis and
pathophysiology of DN and DR. This has resulted in significant advances in
treatment. In particular the blockade of the renin-angiotensin system for DN
and laser photocoagulation for DR has resulted in decreasing long-term
morbidity. Nevertheless, the impact of these complications remains significant
and clinicians should remain vigilant. Regular screening as recommended by
guidelines and prompt institution of treatment lead to further reductions in
morbidity and mortality.

References
[1] Skyler JS. Microvascular complications: retinopathy and nephropathy. Endocrinol
Metab Clin North Am 2001;30:833–56.
[2] Prevent Blindness America. R&B legend Gladys Knight sings praises of early detection
and management of diabetes. Schaumberg (IL): Prevent Blindness America; 2003.
[3] Allen AC. So called intercapillary glomerulosclerosis, a lesion associated with diabetes
mellitus. Arch Pathol 1941;32:33–51.
[4] Bell ET. Renal lesions in diabetes mellitus. Am J Pathol 1942;18:744–5.
[5] Bell ET. Renal diseases. London: Henry Kimpton Publishers; 1946.
1028 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

[6] Kimmelstiel P. Studies on renal biopsy specimens with the aid of electron microscopy. 1.
Glomeruli in diabetes. Am J Clin Pathol 1962;38:270–9.
[7] The Diabetic Retinopathy Study Research Group. Four risk factors for severe visual loss
in diabetic retinopathy. The third report from the Diabetic Retinopathy Study. Arch
Ophthalmol 1979;97:654–5.
[8] Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for
diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1.
Arch Ophthalmol 1985;103:1796–806.
[9] The Diabetic Retinopathy Vitrectomy Study Research Group. Two-year course of visual
acuity in severe proliferative diabetic retinopathy with conventional management.
Diabetic Retinopathy Vitrectomy Study (DRVS) report number 1. Ophthalmology
1985;92:492–502.
[10] Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin epidemiologic
study of diabetic retinopathy. II. Prevalence and risk of diabetic retinopathy when age at
diagnosis is less than 30 years. Arch Ophthalmol 1984;102:520–6.
[11] The Diabetes Control and Complications Trial Research Group. The relationship of
glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in
the diabetes control and complications trial. Diabetes 1995;44:968–83.
[12] Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, Motoyoshi S, et al. Intensive
insulin therapy prevents the progression of diabetic microvascular complications in
Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective
6-year study. Diabetes Res Clin Pract 1995;28:103–17.
[13] UK Prospective Diabetes Study Group. Tight blood pressure control and risk of
macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ
1998;317:703–13.
[14] Reichard P, Nilsson BY, Rosenqvist U. The effect of long-term intensified insulin treatment
on the development of microvascular complications of diabetes mellitus. N Engl J Med
1993;329:304–9.
[15] Hypertension in Diabetes Study (HDS). I. Prevalence of hypertension in newly presenting
type 2 diabetic patients and the association with risk factors for cardiovascular and
diabetic complications. J Hypertens 1993;11:309–17.
[16] Chaturvedi N, Sjolie AK, Stephenson JM, Abrahamian H, Keipes M, Castellarin A, et al.
Effect of lisinopril on progression of retinopathy in normotensive people with type 1
diabetes. The EUCLID Study Group. EURODIAB Controlled Trial of Lisinopril in
Insulin-Dependent Diabetes Mellitus. Lancet 1998;351:28–31.
[17] Ismail N, Becker B, Strzelczyk P, Ritz E. Renal disease and hypertension in non–insulin-
dependent diabetes mellitus. Kidney Int 1999;55:1–28.
[18] Orchard TJ, Dorman JS, Maser RE, Becker DJ, Drash AL, Ellis D, et al. Prevalence of
complications in IDDM by sex and duration. Pittsburgh Epidemiology of Diabetes
Complications Study II. Diabetes 1990;39:1116–24.
[19] Parving HH, Hommel E, Mathiesen E, Skott P, Edsberg B, Bahnsen M, et al. Prevalence
of microalbuminuria, arterial hypertension, retinopathy and neuropathy in patients with
insulin dependent diabetes. BMJ 1988;296:156–60.
[20] Nelson RG, Knowler WC, Pettitt DJ, Saad MF, Bennett PH. Diabetic kidney disease in
Pima Indians. Diabetes Care 1993;16:335–41.
[21] Bojestig M, Arnqvist HJ, Hermansson G, Karlberg BE, Ludvigsson J. Declining
incidence of nephropathy in insulin-dependent diabetes mellitus. N Engl J Med 1994;330:
15–28.
[22] Rossing P, Rossing K, Jacobsen P, Parving HH. Unchanged incidence of diabetic
nephropathy in IDDM patients. Diabetes 1995;44:739–43.
[23] Klein R. Vision disorders in diabetes. In: Diabetes in America. Bethesda: National
Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases;
1995. p. 293.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1029

[24] Mogensen CE, Christensen CK, Vittinghus E. The stages in diabetic renal disease, with
emphasis on the stage of incipient diabetic nephropathy. Diabetes 1983;32(Suppl 2):
64–78.
[25] Mogensen CE, Andersen MJ. Increased kidney size and glomerular filtration rate in
untreated juvenile diabetes: normalization by insulin-treatment. Diabetologia 1975;11:
221–4.
[26] Kroustrup JP, Gundersen HJ, Osterby R. Glomerular size and structure in diabetes
mellitus. III. Early enlargement of the capillary surface. Diabetologia 1977;13:207–10.
[27] Hostetter TH, Rennke HG, Brenner BM. The case for intrarenal hypertension in the
initiation and progression of diabetic and other glomerulopathies. Am J Med 1982;72:
375–80.
[28] Mogensen CE, Christensen CK. Predicting diabetic nephropathy in insulin-dependent
patients. N Engl J Med 1984;311:89–93.
[29] Osterby R, Gundersen HJ. Glomerular size and structure in diabetes mellitus. I. Early
abnormalities. Diabetologia 1975;11:225–9.
[30] Schober E, Pollak A, Coradello H, Lubec G. Glycosylation of glomerular basement
membrane in type 1 (insulin-dependent) diabetic children. Diabetologia 1982;23:485–7.
[31] Cohen MP, Wu VY, Surma ML. Non-collagen protein and proteoglycan in renal
glomerular basement membrane. Biochim Biophys Acta 1981;678:322–8.
[32] Vigstrup J, Mogensen CE. Proliferative diabetic retinopathy: at risk patients identified by
early detection of microalbuminuria. Acta Ophthalmol 1985;63:530–4.
[33] Mogensen CE. Microalbuminuria as a predictor of clinical diabetic nephropathy. Kidney
Int 1987;31:673–89.
[34] Mogensen CE, Hansen KW, Osterby R, Damsgaard EM. Blood pressure elevation versus
abnormal albuminuria in the genesis and prediction of renal disease in diabetes. Diabetes
Care 1992;15:1192–204.
[35] Mogensen CE. Long-term antihypertensive treatment inhibiting progression of diabetic
nephropathy. BMJ 1982;285:685–8.
[36] Fioretto P, Steffes MW, Brown DM, Mauer SM. An overview of renal pathology in
insulin-dependent diabetes mellitus in relationship to altered glomerular hemodynamics.
Am J Kidney Dis 1992;20:549–58.
[37] Harris RD, Steffes MW, Bilous RW, Sutherland DE, Mauer SM. Global glomerular
sclerosis and glomerular arteriolar hyalinosis in insulin dependent diabetes. Kidney Int
1991;40:107–14.
[38] Heilig CW, Concepcion LA, Riser BL, Freytag SO, Zhu M, Cortes P. Overexpression of
glucose transporters in rat mesangial cells cultured in a normal glucose milieu mimics the
diabetic phenotype. J Clin Invest 1995;96:1802–14.
[39] Chakrabarti S, Cukiernik M, Hileeto D, Evans T, Chen S. Role of vasoactive factors in
the pathogenesis of early changes in diabetic retinopathy. Diabetes Metab Res Rev 2000;16:
393–407.
[40] Hammes HP, Du X, Edelstein D, Taguchi T, Matsumura T, Ju Q, et al. Benfotiamine
blocks three major pathways of hyperglycemic damage and prevents experimental
diabetic retinopathy. Nat Med 2003;9:294–9.
[41] Neely KA, Quillen DA, Schachat AP, Gardner TW, Blankenship GW. Diabetic
retinopathy. Med Clin North Am 1998;82:847–76.
[42] Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature
2001;414:813–20.
[43] Way KJ, Katai N, King GL. Protein kinase C and the development of diabetic vascular
complications. Diabet Med 2001;18:945–59.
[44] Li W, Yanoff M, Liu X, Ye X. Retinal capillary pericyte apoptosis in early human
diabetic retinopathy. Chin Med J (Engl) 1997;110:659–63.
[45] Mizutani M, Kern TS, Lorenzi M. Accelerated death of retinal microvascular cells in
human and experimental diabetic retinopathy. J Clin Invest 1996;97:2883–90.
1030 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

[46] Meister A, Anderson ME. Glutathione. Annu Rev Biochem 1983;52:711–60.


[47] Li W, Liu X, Yanoff M, Cohen S, Ye X. Cultured retinal capillary pericytes die by apoptosis
after an abrupt fluctuation from high to low glucose levels: a comparative study with retinal
capillary endothelial cells. Diabetologia 1996;39:537–47.
[48] Heilig CW, Concepcion LA, Riser BL, Freytag SO, Zhu M, Cortes P. Overexpression of
glucose transporters in rat mesangial cells cultured in a normal glucose milieu mimics the
diabetic phenotype. J Clin Invest 1995;96:1802–14.
[49] De Vriese AS, Tilton RG, Elger M, Stephan CC, Kriz W, Lameire NH. Antibodies
against vascular endothelial growth factor improve early renal dysfunction in
experimental diabetes. J Am Soc Nephrol 2001;12:993–1000.
[50] Utimura R, Fujihara CK, Mattar AL, Malheiros DM, De LN, Zatz R. Mycophenolate
mofetil prevents the development of glomerular injury in experimental diabetes. Kidney
Int 2003;63:209–16.
[51] Wolf G. Molecular mechanisms of renal hypertrophy: role of p27Kip1. Kidney Int 1999;
56:1262–5.
[52] Sharma K, Ziyadeh FN. Hyperglycemia and diabetic kidney disease: the case for
transforming growth factor-beta as a key mediator. Diabetes 1995;44:1139–46.
[53] Border WA, Noble NA. Evidence that TGF-beta should be a therapeutic target in
diabetic nephropathy. Kidney Int 1998;54:1390–1.
[54] Sharma K, Eltayeb BO, McGowan TA, Dunn SR, Alzahabi B, Rohde R, et al. Captopril-
induced reduction of serum levels of transforming growth factor-beta1 correlates with
long-term renoprotection in insulin-dependent diabetic patients. Am J Kidney Dis 1999;
34:818–23.
[55] Cooper ME. Pathogenesis, prevention, and treatment of diabetic nephropathy. Lancet
1998;352:213–9.
[56] Craven PA, DeRubertis FR. Protein kinase C is activated in glomeruli from
streptozotocin diabetic rats: possible mediation by glucose. J Clin Invest 1989;83:
1667–75.
[57] Chen S, Cohen MP, Lautenslager GT, Shearman CW, Ziyadeh FN. Glycated albumin
stimulates TGF-beta 1 production and protein kinase C activity in glomerular endothelial
cells. Kidney Int 2001;59:673–81.
[58] Koya D, King GL. Protein kinase C activation and the development of diabetic
complications. Diabetes 1998;47:859–66.
[59] Koya D, Jirousek MR, Lin YW, Ishii H, Kuboki K, King GL. Characterization of
protein kinase C beta isoform activation on the gene expression of transforming growth
factor-beta, extracellular matrix components, and prostanoids in the glomeruli of diabetic
rats. J Clin Invest 1997;100:115–26.
[60] Koya D, Lee IK, Ishii H, Kanoh H, King GL. Prevention of glomerular dysfunction in
diabetic rats by treatment with d-alpha-tocopherol. J Am Soc Nephrol 1997;8:426–35.
[61] Ishii H, Jirousek MR, Koya D, Takagi C, Xia P, Clermont A, et al. Amelioration of vascular
dysfunctions in diabetic rats by an oral PKC beta inhibitor. Science 1996;272:728–31.
[62] Frank RN. The aldose reductase controversy. Diabetes 1994;43:169–72.
[63] Whiteside CI, Thompson J. The role of angiotensin-II in progressive diabetic
glomerulopathy in the rat. Endocrinology 1989;125:1932–40.
[64] Alzaid AA, Dinneen SF, Melton LJ III, Rizza RA. The role of growth hormone in the
development of diabetic retinopathy. Diabetes Care 1994;17:531–4.
[65] Aiello LP, Avery RL, Arrigg PG, Keyt BA, Jampel HD, Shah ST, et al. Vascular
endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other
retinal disorders. N Engl J Med 1994;331:1480–7.
[66] Mitamura Y, Tashimo A, Nakamura Y, Tagawa H, Ohtsuka K, Mizue Y, et al. Vitreous
levels of placenta growth factor and vascular endothelial growth factor in patients with
proliferative diabetic retinopathy. Diabetes Care 2002;25:2352.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1031

[67] UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with
sulphonylureas or insulin compared with conventional treatment and risk of complica-
tions in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.
[68] The Diabetes Control and Complications Trial Research Group. The effect of intensive
treatment of diabetes on the development and progression of long-term complications in
insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.
[69] Writing Team for Diabetes Control and Complications Trial/Epidemiology of Diabetic
Intervensions and Complications Research Group. Effect of intensive therapy on the
microvascular complications of type 1 diabetes mellitus. JAMA 2002;287:2563–9.
[70] Epidemiology of Diabetes Interventions and Complications (EDIC). Design, implementa-
tion, and preliminary results of a long-term follow-up of the Diabetes Control and
Complications Trial cohort. Diabetes Care 1999;22:99–111.
[71] American Diabetes Association. Diabetic retinopathy. Diabetes Care 2000;23(Suppl 1):
S73–6.
[72] Klein R, Klein BE, Moss SE. Epidemiology of proliferative diabetic retinopathy.
Diabetes Care 1992;15:1875–91.
[73] Estacio RO, Jeffers BW, Gifford N, Schrier RW. Effect of blood pressure control on
diabetic microvascular complications in patients with hypertension and type 2 diabetes.
Diabetes Care 2000;23(Suppl 2):B54–64.
[74] Parving HH, Osterby R, Ritz E. Diabetic nephropathy. In: Brenner BM, editor. The
kidney. Philadelphia: WB Saunders; 2000. p. 1731.
[75] Cambien F, Poirier O, Lecerf L, Evans A, Cambou JP, Arveiler D, et al. Deletion
polymorphism in the gene for angiotensin-converting enzyme is a potent risk factor for
myocardial infarction. Nature 1992;359:641–4.
[76] Yoshida H, Kuriyama S, Atsumi Y, Tomonari H, Mitarai T, Hamaguchi A, et al.
Angiotensin I converting enzyme gene polymorphism in non-insulin dependent diabetes
mellitus. Kidney Int 1996;50:657–64.
[77] Schmidt S, Ritz E. Angiotensin I converting enzyme gene polymorphism and diabetic
nephropathy in type II diabetes. Nephrol Dial Transplant 1997;12(Suppl 2):37–41.
[78] Schmidt S, Strojek K, Grzeszczak W, Bergis K, Ritz E. Excess of DD homozygotes in
haemodialysed patients with type II diabetes. The Diabetic Nephropathy Study Group.
Nephrol Dial Transplant 1997;12:427–9.
[79] Pettitt DJ, Saad MF, Bennett PH, Nelson RG, Knowler WC. Familial predisposition to
renal disease in two generations of Pima Indians with type 2 (non-insulin-dependent)
diabetes mellitus. Diabetologia 1990;33:438–43.
[80] Satko SG, Langefeld CD, Daeihagh P, Bowden DW, Rich SS, Freedman BI.
Nephropathy in siblings of African Americans with overt type 2 diabetic nephropathy.
Am J Kidney Dis 2002;40:489–94.
[81] Trevisan R, Viberti G. Genetic factors in the development of diabetic nephropathy. J Lab
Clin Med 1995;126:342–9.
[82] Jeffers BW, Estacio RO, Raynolds MV, Schrier RW. Angiotensin-converting enzyme
gene polymorphism in non-insulin dependent diabetes mellitus and its relationship with
diabetic nephropathy. Kidney Int 1997;52:473–7.
[83] Kuramoto N, Iizuka T, Ito H, Yagui K, Omura M, Nozaki O, et al. Effect of ACE gene
on diabetic nephropathy in NIDDM patients with insulin resistance. Am J Kidney Dis
1999;33:276–81.
[84] Kunz R, Bork JP, Fritsche L, Ringel J, Sharma AM. Association between the
angiotensin-converting enzyme-insertion/deletion polymorphism and diabetic nephropa-
thy: a methodologic appraisal and systematic review. J Am Soc Nephrol 1998;9:1653–63.
[85] Shah VO, Scavini M, Nikolic J, Sun Y, Vai S, Griffith JK, et al. Z-2 microsatellite allele is
linked to increased expression of the aldose reductase gene in diabetic nephropathy. J Clin
Endocrinol Metab 1998;83:2886–91.
1032 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

[86] Ritz E, Orth SR. Nephropathy in patients with type 2 diabetes mellitus. N Engl J Med
1999;341:1127–33.
[87] Earle K, Viberti GC. Familial, hemodynamic and metabolic factors in the predisposition
to diabetic kidney disease. Kidney Int 1994;45:434–7.
[88] Barzilay J, Warram JH, Bak M, Laffel LM, Canessa M, Krolewski AS. Predisposition to
hypertension: risk factor for nephropathy and hypertension in IDDM. Kidney Int 1992;
41:723–30.
[89] Trevisan R, Fioretto P, Barbosa J, Mauer M. Insulin-dependent diabetic sibling pairs are
concordant for sodium-hydrogen antiport activity. Kidney Int 1999;55:2383–9.
[90] Rudberg S, Persson B, Dahlquist G. Increased glomerular filtration rate as a predictor of
diabetic nephropathy: an 8-year prospective study. Kidney Int 1992;41:822–8.
[91] Mogensen CE. Prediction of clinical diabetic nephropathy in IDDM patients: alternatives
to microalbuminuria? Diabetes 1990;39:761–7.
[92] Tuttle KR, Bruton JL, Perusek MC, Lancaster JL, Kopp DT, DeFronzo RA. Effect of
strict glycemic control on renal hemodynamic response to amino acids and renal
enlargement in insulin-dependent diabetes mellitus. N Engl J Med 1991;324:1626–32.
[93] Vora JP, Dolben J, Dean JD, Thomas D, Williams JD, Owens DR, et al. Renal
hemodynamics in newly presenting non-insulin dependent diabetes mellitus. Kidney Int
1992;41:829–35.
[94] Nowack R, Raum E, Blum W, Ritz E. Renal hemodynamics in recent-onset type II diabetes.
Am J Kidney Dis 1992;20:342–7.
[95] Gambara V, Mecca G, Remuzzi G, Bertani T. Heterogeneous nature of renal lesions in
type II diabetes. J Am Soc Nephrol 1993;3:1458–66.
[96] Smith SR, Svetkey LP, Dennis VW. Racial differences in the incidence and progression of
renal diseases. Kidney Int 1991;40:815–22.
[97] Brancati FL, Whittle JC, Whelton PK, Seidler AJ, Klag MJ. The excess incidence of
diabetic end-stage renal disease among blacks: a population-based study of potential
explanatory factors. JAMA 1992;268:3079–84.
[98] Kohler KA, McClellan WM, Ziemer DC, Kleinbaum DG, Boring JR. Risk factors for
microalbuminuria in black Americans with newly diagnosed type 2 diabetes. Am J
Kidney Dis 2000;36:903–13.
[99] Chaiken RL, Palmisano J, Norton ME, Banerji MA, Bard M, Sachimechi I, et al.
Interaction of hypertension and diabetes on renal function in black NIDDM subjects.
Kidney Int 1995;47:1697–702.
[100] Ng LL, Davies JE, Siczkowski M, Sweeney FP, Quinn PA, Krolewski B, et al. Abnormal
Na þ /H þ antiporter phenotype and turnover of immortalized lymphoblasts from type
1 diabetic patients with nephropathy. J Clin Invest 1994;93:2750–7.
[101] The ACE Inhibitors in Diabetic Nephropathy Trialist Group. Should all patients with
type 1 diabetes mellitus and microalbuminuria receive angiotensin-converting enzyme
inhibitors? A meta-analysis of individual patient data. Ann Intern Med 2001;134:370–9.
[102] American Diabetes Association. Diabetic nephropathy. Diabetes Care 2000;23(Suppl 1):
S69–72.
[103] Parving HH, Gall MA, Skott P, Jorgensen HE, Lokkegaard H, Jorgensen F, et al.
Prevalence and causes of albuminuria in non-insulin-dependent diabetic patients. Kidney
Int 1992;41:758–62.
[104] Christensen PK, Larsen S, Horn T, Olsen S, Parving HH. Causes of albuminuria in
patients with type 2 diabetes without diabetic retinopathy. Kidney Int 2000;58:1719–31.
[105] Rosenn B, Miodovnik M, Kranias G, Khoury J, Combs CA, Mimouni F, et al.
Progression of diabetic retinopathy in pregnancy: association with hypertension in
pregnancy. Am J Obstet Gynecol 1992;166:1214–8.
[106] Mazzucco G, Bertani T, Fortunato M, Bernardi M, Leutner M, Boldorini R, et al.
Different patterns of renal damage in type 2 diabetes mellitus: a multicentric study on 393
biopsies. Am J Kidney Dis 2002;39:713–20.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1033

[107] Kasinath BS, Mujais SK, Spargo BH, Katz AI. Nondiabetic renal disease in patients with
diabetes mellitus. Am J Med 1983;75:613–7.
[108] Furuta T, Seino J, Saito T, Sato H, Agatsuma J, Ootaka T, et al. Insulin deposits in
membranous nephropathy associated with diabetes mellitus. Clin Nephrol 1992;37:
65–9.
[109] Gans RO, Ueda Y, Ito S, Kohli R, Min I, Shafi M, et al. The occurrence of IgA-nephropathy
in patients with diabetes mellitus may not be coincidental: a report of five cases. Am J
Kidney Dis 1992;20:255–60.
[110] Padilla B, Weiss M, Kant KS. Henoch-Schönlein purpura in a patient with diabetic
nephropathy: case report and a review of the literature. Am J Kidney Dis 1992;20:
191–4.
[111] Mak SK, Gwi E, Chan KW, Wong PN, Lo KY, Lee KF, et al. Clinical predictors of non-
diabetic renal disease in patients with non-insulin dependent diabetes mellitus. Nephrol
Dial Transplant 1997;12:2588–91.
[112] Matsumae T, Fukusaki M, Sakata N, Takebayashi S, Naito S. Thin glomerular basement
membrane in diabetic patients with urinary abnormalities. Clin Nephrol 1994;42:221–6.
[113] Olsen S. Identification of non-diabetic glomerular disease in renal biopsies from diabetics:
a dilemma. Nephrol Dial Transplant 1999;14:1846–9.
[114] American Diabetes Association. Standards of medical care for patients with diabetes
mellitus. Diabetes Care 2003;26(Suppl 1):S33–50.
[115] Sussman EJ, Tsiaras WG, Soper KA. Diagnosis of diabetic eye disease. JAMA 1982;247:
3231–4.
[116] Sebag J, Ansari RR, Dunker S, Suh KI. Dynamic light scattering of diabetic vitreopathy.
Diabetes Technol Ther 1999;1:169–76.
[117] Katz A, Kruger EF, Minko G, Liu CH, Rosen RB, Alfano RR. Detection of glutamate in
the eye by Raman spectroscopy. J Biomed Opt 2003;8:167–72.
[118] Van Schaik HJ, Coppens J, Van den Berg TJ, Van Best JA. Autofluorescence distribution
along the corneal axis in diabetic and healthy humans. Exp Eye Res 1999;69:505–10.
[119] Michelson G, Welzenbach J, Pal I, Harazny J. Functional imaging of the retinal micro-
vasculature by scanning laser Doppler flowmetry. Int Ophthalmol 2001;23:327–35.
[120] Gomez-Ulla F, Fernandez MI, Gonzalez F, Rey P, Rodriguez M, Rodriguez-Cid MJ,
et al. Digital retinal images and teleophthalmology for detecting and grading diabetic
retinopathy. Diabetes Care 2002;25:1384–9.
[121] Borch-Johnsen K, Feldt-Rasmussen B, Strandgaard S, Schroll M, Jensen JS. Urinary
albumin excretion: an independent predictor of ischemic heart disease. Arterioscler
Thromb Vasc Biol 1999;19:1992–7.
[122] Deckert T, Kofoed-Enevoldsen A, Norgaard K, Borch-Johnsen K, Feldt-Rasmussen B,
Jensen T. Microalbuminuria: implications for micro- and macrovascular disease.
Diabetes Care 1992;15:1181–91.
[123] Gerstein HC, Mann JF, Yi Q, Zinman B, Dinneen SF, Hoogwerf B, et al. Albuminuria
and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic
individuals. JAMA 2001;286:421–6.
[124] Spoelstra-de Man AM, Brouwer CB, Stehouwer CD, Smulders YM. Rapid progression
of albumin excretion is an independent predictor of cardiovascular mortality in patients
with type 2 diabetes and microalbuminuria. Diabetes Care 2001;24:2097–101.
[125] Tuttle KR, Puhlman ME, Cooney SK, Short R. Urinary albumin and insulin as
predictors of coronary artery disease: an angiographic study. Am J Kidney Dis 1999;34:
918–25.
[126] Winocour PH, Harland JO, Millar JP, Laker MF, Alberti KG. Microalbuminuria and
associated cardiovascular risk factors in the community. Atherosclerosis 1992;93:71–81.
[127] Mykkanen L, Zaccaro DJ, Wagenknecht LE, Robbins DC, Gabriel M, Haffner SM.
Microalbuminuria is associated with insulin resistance in nondiabetic subjects: the insulin
resistance atherosclerosis study. Diabetes 1998;47:793–800.
1034 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

[128] Pinkney JH, Denver AE, Mohamed-Ali V, Foster C, Yudkin JS. Insulin resistance in
non-insulin-dependent diabetes mellitus is associated with microalbuminuria indepen-
dently of ambulatory blood pressure. J Diabetes Complications 1995;9:230–3.
[129] Wasada T, Katsumori K, Saeki A, Saito S, Omori Y. Urinary albumin excretion rate is
related to insulin resistance in normotensive subjects with impaired glucose tolerance.
Diabetes Res Clin Pract 1997;34:157–62.
[130] Yudkin JS. Hyperinsulinaemia, insulin resistance, microalbuminuria and the risk of
coronary heart disease. Ann Med 1996;28:433–8.
[131] Mykkanen L, Haffner SM, Kuusisto J, Pyorala K, Laakso M. Microalbuminuria
precedes the development of NIDDM. Diabetes 1994;43:552–7.
[132] Krolewski AS, Warram JH, Rand LI, Kahn CR. Epidemiologic approach to the
etiology of type I diabetes mellitus and its complications. N Engl J Med 1987;317:
1390–8.
[133] Almdal T, Norgaard K, Feldt-Rasmussen B, Deckert T. The predictive value of micro-
albuminuria in IDDM: a five-year follow-up study. Diabetes Care 1994;17:120–5.
[134] Krolewski AS, Laffel LM, Krolewski M, Quinn M, Warram JH. Glycosylated
hemoglobin and the risk of microalbuminuria in patients with insulin-dependent diabetes
mellitus. N Engl J Med 1995;332:1251–5.
[135] Chaturvedi N, Bandinelli S, Mangili R, Penno G, Rottiers RE, Fuller JH. Micro-
albuminuria in type 1 diabetes: rates, risk factors and glycemic threshold. Kidney Int
2001;60:219–27.
[136] Mogensen CE. Microalbuminuria predicts clinical proteinuria and early mortality in
maturity-onset diabetes. N Engl J Med 1984;310:356–60.
[137] Klein R, Klein BE, Moss SE, Cruickshanks KJ. Ten-year incidence of gross proteinuria in
people with diabetes. Diabetes 1995;44:916–23.
[138] Nelson RG, Bennett PH, Beck GJ, Tan M, Knowler WC, Mitch WE, et al. Development
and progression of renal disease in Pima Indians with non-insulin-dependent diabetes
mellitus. Diabetic Renal Disease Study Group. N Engl J Med 1996;335:1636–42.
[139] Ravid M, Savin H, Jutrin I, Bental T, Katz B, Lishner M. Long-term stabilizing effect of
angiotensin-converting enzyme inhibition on plasma creatinine and on proteinuria in
normotensive type II diabetic patients. Ann Intern Med 1993;118:577–81.
[140] Mogensen CE, Hansen KW, Pedersen MM, Christensen CK. Renal factors influencing
blood pressure threshold and choice of treatment for hypertension in IDDM. Diabetes
Care 1991;14(Suppl 4):13–26.
[141] Parving HH, Hommel E, Jensen BR, Hansen HP. Long-term beneficial effect of ACE
inhibition on diabetic nephropathy in normotensive type 1 diabetic patients. Kidney Int
2001;60:228–34.
[142] The Microalbuminuria Captopril Study Group. Captopril reduces the risk of
nephropathy in IDDM patients with microalbuminuria. Diabetologia 1996;39:587–93.
[143] Viberti G, Mogensen CE, Groop LC, Pauls JF. Effect of captopril on progression to
clinical proteinuria in patients with insulin-dependent diabetes mellitus and micro-
albuminuria. European Microalbuminuria Captopril Study Group. JAMA 1994;271:
275–9.
[144] Gerstein HC. Reduction of cardiovascular events and microvascular complications in
diabetes with ACE inhibitor treatment: HOPE and MICRO-HOPE. Diabetes Metab Res
Rev 2002;18(Suppl 3):S82–5.
[145] Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of angiotensin-converting-
enzyme inhibition on diabetic nephropathy. The Collaborative Study Group. N Engl J
Med 1993;329:1456–62.
[146] Hebert LA, Bain RP, Verme D, Cattran D, Whittier FC, Tolchin N, et al. Remission of
nephrotic range proteinuria in type I diabetes. Collaborative Study Group. Kidney Int
1994;46:1688–93.
A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036 1035

[147] Kasiske BL, Kalil RS, Ma JZ, Liao M, Keane WF. Effect of antihypertensive therapy on
the kidney in patients with diabetes: a meta-regression analysis. Ann Intern Med 1993;
118:129–38.
[148] Adler AI, Stratton IM, Neil HA, Yudkin JS, Matthews DR, Cull CA, et al. Association
of systolic blood pressure with macrovascular and microvascular complications of type 2
diabetes (UKPDS 36): prospective observational study. BMJ 2000;321:412–9.
[149] ALLHAT Officers and Coordinators for ALLHAT Collaborative Research Group.
Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting
enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-
Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA 2002;288:
2981–97.
[150] Brenner BM, Cooper ME, Zeeuw DD, Grunfeld JP, Keane WF, Kurokawa K, et al. The
losartan renal protection study: rationale, study design and baseline characteristics of
RENAAL (Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist
Losartan). J Renin Angiotensin Aldosterone Syst 2000;1:328–35.
[151] Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects
of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and
nephropathy. N Engl J Med 2001;345:861–9.
[152] Parving HH, Lehnert H, Brochner-Mortensen J, Gomis R, Andersen S, Arner P. Effect of
irbesartan on the development of diabetic nephropathy in patients with type 2 diabetes.
Ugeskr Laeger 2001;163:5519–24.
[153] Mogensen CE, Neldam S, Tikkanen I, Oren S, Viskoper R, Watts RW, et al. Randomised
controlled trial of dual blockade of renin-angiotensin system in patients with
hypertension, microalbuminuria, and non-insulin dependent diabetes: the candesartan
and lisinopril microalbuminuria (CALM) study. BMJ 2000;321:1440–4.
[154] Tatti P, Pahor M, Byington RP, Di Mauro P, Guarisco R, Strollo G, et al. Outcome
results of the Fosinopril Versus Amlodipine Cardiovascular Events Randomized Trial
(FACET) in patients with hypertension and NIDDM. Diabetes Care 1998;21:597–603.
[155] Levinsky NG. Specialist evaluation in chronic kidney disease: too little, too late. Ann
Intern Med 2002;137:542–3.
[156] Early Treatment Diabetic Retinopathy Study Research Group. Effects of aspirin
treatment on diabetic retinopathy. ETDRS report number 8. Ophthalmology 1991;
98(Suppl 5):757–65.
[157] The TIMAD Study Group. Ticlopidine treatment reduces the progression of non-
proliferative diabetic retinopathy. Arch Ophthalmol 1990;108:1577–83.
[158] Shinoda K, Ishida S, Kawashima S, Wakabayashi T, Uchita M, Matsuzaki T, et al.
Clinical factors related to the aqueous levels of vascular endothelial growth factor and
hepatocyte growth factor in proliferative diabetic retinopathy. Curr Eye Res 2000;21:
655–61.
[159] Stefansson E, Machemer R, de Juan E Jr, McCuen BW, Peterson J. Retinal oxygenation
and laser treatment in patients with diabetic retinopathy. Am J Ophthalmol 1992;113:
36–8.
[160] Marshall J, Mellerio J. Laser irradiation of retinal tissue. Brit Med Bull 1970;26(2):
156–60.
[161] Reddy VM, Zamora RL, Olk RJ. Quantitation of retinal ablation in proliferative diabetic
retinopathy. Am J Ophthalmol 1995;119:760–6.
[162] Doft BH, Metz DJ, Kelsey SF. Augmentation laser for proliferative diabetic retinopathy
that fails to respond to initial panretinal photocoagulation. Ophthalmology 1992;99:
1728–34.
[163] Patz A, Schatz H, Berkow JW, Gittelsohn AM, Ticho U. Macular edema: an overlooked
complication of diabetic retinopathy. Trans Am Acad Ophthalmol Otolaryngol 1973;77:
OP34–42.
1036 A. Jawa et al / Med Clin N Am 88 (2004) 1001–1036

[164] Early Treatment Diabetic Retinopathy Study Research Group. Focal photocoagulation
treatment of diabetic macular edema: relationship of treatment effect to fluorescein
angiographic and other retinal characteristics at baseline. ETDRS report no. 19. Arch
Ophthalmol 1995;113:1144–55.
[165] Early Treatment Diabetic Retinopathy Study Research Group. Early photocoagulation
for diabetic retinopathy. ETDRS report number 9. Ophthalmology 1991;98(Suppl 5):
766–85.
[166] Striph GG, Hart WM Jr, Olk RJ. Modified grid laser photocoagulation for diabetic
macular edema: the effect on the central visual field. Ophthalmology 1988;95:1673–9.
[167] Ferris FL III, Podgor MJ, Davis MD. Macular edema in diabetic retinopathy study
patients. Diabetic Retinopathy Study Report number 12. Ophthalmology 1987;94:
754–60.
[168] Pender PM, Benson WE, Compton H, Cox GB. The effects of panretinal photocoag-
ulation on dark adaptation in diabetics with proliferative retinopathy. Ophthalmology
1981;88:635–8.
[169] Machemer R, Parel JM, Buettner H. A new concept for vitreous surgery. I.
Instrumentation. Am J Ophthalmol 1972;73:1–7.
[170] Lewis H, Abrams GW, Blumenkranz MS, Campo RV. Vitrectomy for diabetic macular
traction and edema associated with posterior hyaloidal traction. Ophthalmology 1992;99:
753–9.
Med Clin N Am 88 (2004) 1037–1061

Ischemic heart disease and congestive


heart failure in diabetic patients
W.H. Wilson Tang, MDa,*, Anjli Maroo, MDa,
James B. Young, MDb
a
Department of Cardiovascular Medicine, Cleveland Clinic Foundation
9500 Euclid Avenue, Desk F25, Cleveland, OH 44195, USA
b
Department of Medicine, Cleveland Clinic Foundation, 9500 Euclid Avenue,
Desk F25, Cleveland, OH 44195, USA

Diabetes mellitus is a well-established independent risk factor for cardio-


vascular disease (CVD) in both men and women. The risk of CVD is increased
2 to 4-fold in patients with diabetes, and CVD contributes to more than 50%
of deaths in this patient population [1]. The worldwide prevalence of diabetes
is expected to double over the next 10 years, and concomitantly, the number of
diabetic patients with ischemic heart disease or heart failure is expected to
increase. Patients with impaired glucose tolerance or insulin resistance, which
usually precede overt type 2 diabetes by several years, will add to the growing
burden of cardiovascular disease, particularly in the younger population.
Furthermore, diabetic patients with CVD suffer a worse long-term prognosis
than their nondiabetic counterparts. In fact, the long-term mortality of
diabetic patients without previously diagnosed CVD approaches that of
nondiabetic individuals with a history of myocardial infarction (MI) (Fig. 1)
[2].
Despite the complex metabolic interactions linking diabetes and CVD,
these two disease processes are frequently diagnosed and managed separately
by physicians in different subspecialties. This article summarizes our current
understanding of ischemic heart disease and heart failure (HF) in patients with
diabetes mellitus, highlighting gaps in our knowledge about the relationship
between diabetes and CVD. Special consideration will be given to new
strategies to treat the adverse effects of abnormal glucose metabolism on the
cardiovascular system.

* Corresponding author.
E-mail address: tangw@ccf.org (W.H.W. Tang).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.008
1038 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

100

80
Survival (%)

60

40
Nondiabetic subjects without prior MI
Diabetic subjects without prior MI
Nondiabetic subjects with prior MI
20 Diabetic subjects with prior MI

0
0 1 2 3 4 5 6 7 8
Year

Fig. 1. Kaplan–Meier estimates of the probability of death from coronary heart disease in 1059
subjects with type 2 diabetes and 1378 nondiabetic subjects with and without previous
myocardial infarction. I bars indicate 95% confidence intervals. Thin line, nondiabetic subject
without prior MI; medium line, diabetic subjects without prior MI; gray line, nondiabetic
subjects with prior MI; heavy line, diabetic subjects with prior MI. (From Haffner et al.
Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic
subjects with and without prior myocardial infarction. N Engl J Med 1998;339:229, Ó
Massachusetts Medical Society. All rights reserved; with permission.)

Pathophysiology
Ischemic heart disease results from progressive or unstable coronary
atherosclerosis. Atherosclerotic coronary artery disease in diabetic patients
is often a diffuse process, affecting proximal and distal coronary segments.
Diabetic patients frequently suffer from microvascular coronary disease
and a reduction in vasodilatory reserve. Hyperinsulinemia and the insulin
resistance syndrome are believed to have several adverse metabolic
consequences that may promote atherogenesis [3–5]. For example, hyper-
insulinemia may impair endothelial function by inhibition of nitric oxide
synthesis and increased production of endothelin-1 [6]. Advanced glycosyla-
tion end products and elevated levels of free fatty acids, which are products of
the hyperinsulinemic and hyperglycemic states, increase the production of
reactive oxygen species, leading to increased oxidative stress and inflamma-
tion [7,8]. In addition, hyperglycemia is associated with an increase in
chemoattractant cytokines and cell adhesion molecules such as E-selectin,
vascular cell adhesion molecule-1, and intracellular cell adhesion molecule [9].
Diabetic patients are subject to a range of serum lipid abnormalities,
including elevated levels of total cholesterol, very-low-density lipoprotein
cholesterol, and triglycerides and reduced levels of high-density lipoprotein
(HDL) cholesterol [10,11]. In diabetic patients, diminished activity of
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1039

lipoprotein lipase leads to an accumulation of small, dense low-density


lipoprotein (LDL) particles, which are prone to oxidative modification and
uptake into macrophage-derived foam cells. These lipid abnormalities
contribute to vascular injury and the development and progression of
atherosclerotic lesions. Diabetes is associated with multiple perturbations of
coagulation factors, platelet function, and inflammatory mediators that
contribute to atherosclerotic plaque rupture and thrombosis formation
(Box 1).
Levels of the procoagulants fibrinogen, tissue factor pathway inhibitor
(TFPI), plasminogen activator inhibitor-1, and factor VII are increased [12].
Levels of fasting insulin are inversely proportional to endogenous tissue
plasminogen activator activity, resulting in an impairment of endogenous
fibrinolysis in diabetic patients [13]. Platelets in diabetic individuals demon-
strate increased aggregation and activation, with increased production of the
potent platelet activator, thromboxane A2, and upregulation of the glycopro-
tein (GP) IIb/IIIa receptor [14–16]. Finally, elevated markers of inflammation,
such as high sensitivity C-reactive protein and interleukin-6 have been
associated with the development of both diabetes and atherosclerosis,
although a common causal relationship remains to be proven [17,18].

Detection of asymptomatic ischemic heart disease in diabetic patients


Despite an apparent lack of symptoms, the prevalence of transient
asymptomatic ischemic episodes, or ‘‘silent’’ ischemia, approaches 10% to
15% in diabetic individuals, compared with 1% to 4% in their nondiabetic
counterparts [19–21]. In some individuals, ischemic episodes are consistently
asymptomatic; however, more commonly, ischemic episodes are an admix-
ture of symptomatic and asymptomatic events.
Our understanding of the mechanisms underlying silent ischemia is poor.
Several theories have been proposed, including altered thresholds of pain
sensitivity, autonomic neuropathy leading to sympathetic denervation,
higher production of b-endorphins, and increased production of anti-
inflammatory cytokines [22–25]. Primary abnormalities of coronary blood

Box 1. Selected thrombogenic risk factors in diabetes


 Endothelial dysfunction
 Platelet abnormalities
Increased primary and secondary aggregation responses
Increased release of a granule contents
Increased thromboxane A2 production
 Coagulation abnormalities
Increased plasminogen activator inhibitor (PAI-1)
Increased fibrinogen
1040 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

flow reserve may mediate silent ischemic episodes, which occur primarily at
rest or with minimal exertion [26].
The decision to perform screening in asymptomatic diabetic individuals is
a difficult one because a lack of symptoms in the diabetic patient is not
reassuring. Indeed, multiple studies [27,28] have shown that asymptomatic
ischemia predicts multivessel coronary disease, increased adverse clinical
outcomes, and poor survival. Nevertheless, there is a wide variability in risk
profiles for CVD in diabetic patients. Because efforts to screen all diabetic
individuals for CVD are not cost-effective, the American Diabetes Associ-
ation and the American College of Cardiology/American Heart Association
(ACC/AHA) have recommended screening diabetic patients at high risk for
CVD who are about to embark on a moderate- to high-intensity exercise
program. [29,30] ‘‘High risk’’ features are shown in Box 2.
There are multiple modalities available for CVD screening. The reported
accuracies of these tests in diabetic patients are shown in Table 1. The test most
commonly used to screen for ischemia is the exercise treadmill test. The
exercise treadmill test is easy to perform, relatively inexpensive, and is capable
of generating several types of prognostic information, including ischemic ST–
T wave abnormalities, exercise capacity, and heart rate recovery. The exercise
treadmill test, however, is unable to localize ischemia and has diminished
accuracy in the setting of baseline electrocardiograph abnormalities such as
left ventricular hypertrophy, digoxin effect, resting ST-segment abnormalities,
conduction abnormalities, and ventricular-paced rhythms.
Stress myocardial perfusion imaging (eg, stress single photon emission
computed tomography [SPECT]) can be used for risk stratification and
diagnosis of CVD in patients with diabetes when the standard exercise
treadmill test is inadequate. Several studies [31,32] have shown that SPECT
has similar prognostic value in patients with and without diabetes. In
symptomatic patients with diabetes, abnormalities on stress SPECT imaging
independently predict the subsequent occurrence of cardiac death and MI
[33]. Although data on the use of SPECT imaging in asymptomatic patients
with diabetes are limited, preliminary studies have demonstrated that

Box 2. Factors that increase risk for cardiovascular disease


in diabetic patients
 Age greater than 35 years old
 Type II diabetes of more than 10 years’ duration
 Type I diabetes of more than 15 years’ duration
 Any major cardiac risk factor
 Microvascular disease (proliferative retinopathy, nephropathy,
microalbuminuria)
 Peripheral vascular disease
 Autonomic neuropathy
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1041

Table 1
Characteristics of tests used for noninvasive assessment for CVD in diabetic patients
Symptomatic Positive Negative
patients Sensitivity Specificity predictive predictive
Test n included? (%) (%) value (%) value (%)
Exercise tolerance 190 Yes 47 81 85 41
testing [141]
SPECT [31] 138 Yes 86 56 89 48
Dobutamine stress 52 Yes 82 54 84 50
echocardiography
[142]

abnormalities on stress SPECT imaging predict poor cardiovascular outcome


[34]. A randomized multicenter study, the Detection of Ischemia in Asymp-
tomatic Diabetics Trial (DIAD), is underway to help identify a high-risk
group of patients with diabetes who may benefit from screening with SPECT
stress imaging [35]. In addition, stress echocardiography has recently been
shown to provide incremental prognostic information for risk stratification of
diabetic patients with suspected CVD, compared with clinical and exercise test
variables [36]; however, the application of stress echocardiography for
screening asymptomatic patients with diabetes awaits further study.
Electron beam computed tomography (EBCT) has emerged as a new
technology to detect and quantitate coronary artery calcium deposits. The
presence of coronary artery calcification, a product of atherosclerotic plaque
formation, is always abnormal. EBCT allows measurement of the coronary
calcium area and density and the calculation of a ‘‘coronary calcium score,’’
which serves as a semiquantitative measure of coronary plaque burden. The
current role of EBCT for screening asymptomatic diabetic individuals,
however, is controversial. The prevalence of coronary artery calcium in
asymptomatic diabetic adults is higher than in their nondiabetic counterparts
[37]. A clear association has not been demonstrated between coronary calcium
and future cardiovascular events in diabetic individuals. Currently, the ACC/
AHA guidelines do not recommend general screening with EBCT because it is
unclear whether EBCT will provide incremental predictive value over
traditional risk models, such as the Framingham risk score [38].

Diagnostic testing for the evaluation of symptomatic ischemic heart disease


The evaluation of angina in diabetic patients begins with a careful history,
physical examination, and electrocardiogram. Further risk stratification can
be performed using the stress test modalities described in the previous section.
Theoretically, the pretest probability of disease in symptomatic diabetic
patients should be higher than the pretest probability of disease in asymp-
tomatic diabetic patients and in symptomatic nondiabetic patients.
1042 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

Coronary angiography is considered the gold standard test for the


diagnosis of CVD and is recommended in patients with poorly controlled or
new-onset angina or in patients with abnormal or high-risk noninvasive test
results. In addition, coronary angiography is indicated in diabetic patients
with multiple risk factors and a high clinical suspicion of coronary artery
disease, despite ‘‘normal’’ noninvasive test results [39]. Diabetic patients with
renal insufficiency warrant special consideration before angiography. The
risks of progressive renal failure and dialysis should be clearly explained to the
patient. Measures to prevent renal failure include adequate peri-procedural
hydration, peri-procedural administration of N-acetylcysteine, sparing use of
low osmolar, nonionic contrast, and the use of biplane imaging. Diabetic
patients who take metformin are at risk for lactic acidosis after catheterization
[40,41]. Metformin should be withheld on the day prior to and 2 days after the
procedure to reduce the risk of lactic acidosis. Insulin and oral hypoglycemic
medications should be withheld on the morning of the procedure to prevent
hypoglycemia.

Medical therapy for ischemic heart disease in the diabetic patient


Medications play an integral role in the management of anginal symptoms
and the prevention of progression of atherosclerosis (Table 2). Nitrates are
first-line antianginal agents that decrease myocardial oxygen demand by
reducing preload and afterload and increase myocardial oxygen supply by
vasodilatation of the coronary arteries. Chronic therapy is associated with
nitrate tolerance, a phenomenon that may be increased in patients with
diabetes because of impaired endothelium-dependent vasodilation [42,43].
For this reason, many patients observe a 12- to 14-hour nitrate-free period.
Aspirin is an antiplatelet agent that irreversibly inhibits cyclooxygenase,
resulting in inhibition of thromboxane synthesis and platelet aggregation. An
aspirin (81–325 mg/d) is the recommended therapy for secondary prevention
of CVD in all diabetic patients with a history of angina, MI, vascular
revascularization procedure, stroke or transient ischemic attack, peripheral
vascular disease, or claudication [44]. This recommendation is based on two
large meta-analyses of major secondary prevention trials, performed by the
Antithrombotic Trialists’ Collaboration [45,46]. These trials, which included
4502 patients with diabetes, demonstrated prevention of 38 vascular events for
every 1000 diabetic patients treated with aspirin. In addition, the American
Diabetes Association recommends low-dose aspirin therapy for primary
prevention of CVD in all diabetic patients at high risk for CVD. Factors that
confer high risk include: family history of coronary heart disease, cigarette
smoking, hypertension, weight greater than 120% of ideal body weight,
microalbuminuria or macro albuminuria, dyslipidemia (total cholesterol
200 mg/dL, LDL cholesterol 100 mg/dL, HDL cholesterol 55 mg/dL
in women and 45 mg/dL in men, or triglyceride level 200) [44].
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1043

Table 2
Medical therapies for ischemic heart disease
Drug Starting dosage Major side effects
Nitrates
Sublingual 0.3–0.4 mg Headache, lightheadedness,
Oral 10 mg 3 times/d/ flushing, orthostasis
30 mg 4 times/da
Transdermal 0.2 mg/h
b-Blockers
b1 Selective 25 mg 4 times/d Bradycardia, atrioventricular
Metoprolol succinate 25 mg 2 times/d block, heart failure, fatigue,
Metoprolol tartrate 25 mg 4 times/d depression, erectile dysfunction,
Atenolol then 2 times/d exacerbation of claudication,
Nonselective 40 mg 2 times/d bronchospasm, increased insulin
Propranolol then 4 times/d induced hypoglycemia
Calcium channel blockers
Verapamil 80 mg 4 times/d/ Bradycardia, atrioventricular
Amlodipine 240 mg 4 times/da block, heart failure, flushing,
Nifedipine 5 mg 4 times/d headache, constipation,
Diltiazem 20 mg 4 times/d pedal edema
60 mg 4 times/d/
240 mg 4 times/da
ACE inhibitors
Captopril 6.25–12.5 mg 3 times/d Hypotension, renal insufficiency,
Enalapril 2.5–5 mg 4 times/d hyperkalemia, cough,
Lisinopril 2.5–5 mg 4 times/d angioneurotic edema,
Ramipril 1.25–2.5 mg 4 times/d anaphylaxis
Aspirin 81–325 mg 4 times/d Gastrointestinal ulcers, renal
dysfunction, bronchospasm,
rash
Clopidogrel 75 mg 4 times/d Gastrointestinal ulcers,
rash, thrombocytopenia,
throbocytopenic
thrombotic purpura (rare)
a
Sustained release preparation.

The adenosine diphosphate (ADP) receptor antagonists ticlopidine and


clopidogrel are thienopyridine agents that irreversibly block the binding of
ADP to platelet type 2 purinergic receptors, thereby inhibiting ADP-
induced platelet aggregation. Clopidogrel, 75 mg daily, is an acceptable
alternative for patients with hypersensitivity or intolerance to aspirin. There
is evidence that clopidogrel may be beneficial for prevention of recurrent
CVD events in diabetic patients with a history of CVD. Clopidogrel was
compared with aspirin for secondary prevention of CVD events in
a randomized, controlled trial, the Clopidogrel versus Aspirin in Patients
at Risk of Ischemic Events (CAPRIE) study [47]. Overall, clopidogrel use
was associated with an 8.7% relative risk reduction in the composite
endpoint of vascular death, MI, or ischemic stroke. In the diabetes substudy
1044 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

of this trial, 3866 patients with diabetes were randomized to treatment with
either clopidogrel or aspirin. After multivariable adjustment, clopidogrel
was associated with a 13.1% relative risk reduction in the composite
endpoint compared with aspirin (P = 0.032) [48].
Clopidogrel may also benefit patients presenting with acute ischemic
heart disease (unstable angina or non–Q-wave MI). In the Clopidogrel in
Unstable Angina to Prevent Recurrent Events (CURE) trial, the composite
outcome of CV death, nonfatal MI, or stroke occurred in 9.3% of patients
treated with a combination of clopidogrel plus aspirin, compared with
11.4% of patients treated with aspirin alone, after a mean of 9 months’
follow-up (P  0.001). A trend toward benefit with the combination therapy
was maintained in the subgroup analysis of the 2840 diabetic patients in the
trial (14.2% event rate in the combination group versus 16.7% in the group
taking aspirin alone, P = not significant) [49]. As a result, updated guide-
lines from the ACC/AHA task force have recommended that clopidogrel be
added to aspirin early in the treatment of patients presenting with unstable
angina or non–Q-wave MI [50]. In clinical practice, the addition of
clopidogrel is often deferred until after the patient’s cardiologist has chosen
a revascularization strategy (percutaneous coronary intervention [PCI]
versus coronary artery bypass graft surgery [CABG] versus medical ther-
apy). Often, clopidogrel is not administered to patients who are scheduled to
undergo CABG because of the recommendation that CABG be delayed for
a minimum of 5 days after the last dose to prevent complications from
bleeding. This recommendation is controversial, and several trials eval-
uating the optimal timing and dosage of clopidogrel administration are
underway.
GP IIb/IIIa inhibitors are intravenous antiplatelet drugs used during
medical therapy for acute coronary syndromes or as adjunctive therapy
during PCI. In the general population, the use of GP IIb/IIIa inhibitors has
been associated with a reduction of up to 65 acute ischemic events per 1000
patients treated during elective and urgent PCI and with 15 to 32 events per
1000 patients treated for acute coronary syndromes [51]. One agent
particularly, abciximab, has been associated with the long-term benefit of
reduced mortality [52]. Several contemporary trials in PCI have shown that
treatment with GP IIb/IIIa inhibitors is associated with a reduction in the
combined 30-day endpoint of death, MI, and urgent revascularization in
both diabetic and nondiabetic subpopulations [53]. It is presently unclear if
the magnitude of benefit varies among the three available GP IIb/IIIa
inhibitors. The effect of GP IIb/IIIa inhibitors on 6-month target vessel
revascularization (TVR), a surrogate measure of clinical restenosis, is less
certain. Rates of TVR varied markedly between trials and within the
diabetic and nondiabetic subgroups. A substantial body of evidence has
demonstrated that treatment with abciximab during PCI is associated with
a marked reduction in 1-year mortality [52,54,55], a benefit that is
particularly apparent for diabetic patients [56]. The benefit of GP IIb/IIIa
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1045

inhibitors in non–ST-elevated MI has been demonstrated in a meta-analysis


of six large randomized controlled trials [57]. In the 6458 diabetic patients
enrolled in these trials, treatment with GP IIb/IIIa inhibitors was associated
with a reduction in 30-day mortality from 6.2% to 4.6% (P = 0.007). Thus,
the available data support the use of GP IIb/IIIa inhibitors in diabetic
patients during PCI and during treatment of non–ST-elevated MI.
b-Blockers (BBs) are powerful antianginal drugs that reduce myocardial
demand by decreasing heart rate and cardiac contractility. BBs, however, have
often been withheld in diabetic patients because of fears that these agents
worsen insulin resistance, attenuate the warning symptoms of hypoglycemia,
and worsen dyslipidemia. As experience with the use of BBs in patients with
diabetes has accrued, these adverse effects have not proven to be common
[58,59]. In the Bezafibrate Infarction Prevention (BIP) study, total mortality
during a 3-year follow-up period was substantially reduced (7.8% versus
14.0%, BB versus no BB, respectively) in diabetic patients who received BBs,
compared with those who did not. This reduction was demonstrated in
diabetic patients with angina but without previous MI (3.9% versus 9.8%, BB
versus no BB, respectively), as well as in diabetic patients with previous MI
(9.7% versus 15.4%, BB versus no BB, respectively) [60]. Several landmark
studies have demonstrated a reduction in reinfarction and an improvement in
survival in patients who have already suffered an MI [61,62]. Despite
convincing evidence that diabetic patients derive substantial benefit from b
blockade, data from the National Cooperative Cardiovascular Project [63]
have shown that diabetic patients are less likely to be discharged with a BB
than nondiabetic patients following acute MI (odds ratio [OR] 0.93, 95%
confidence interval [CI] 0.88–0.99). Thus, further efforts to increase use of
these medications in diabetic patients are warranted.
Angiotensin converting enzyme (ACE) inhibitors have a well-established
role in the early therapy of diabetic patients who have suffered an MI. One
of the largest studies to address this question was the Gruppo Italiano per lo
Studio della Sopravvivenza nell’Infarcto Miocardio (GISSI-3) trial [64].
This trial included 2790 diabetic patients in an open-label study of lisinopril,
initiated during the first 24 hours of an MI. In the entire cohort
(n = 18,895), treatment with lisinopril was associated with an 11% overall
reduction in 6-week mortality compared with controls (6.3% versus 7.1%,
OR 0.88, 95% CI 0.79–0.99). In the diabetic subgroup, treatment with
lisinopril was associated with a 30% reduction in 6-week mortality (8.7%
versus 12.4%, OR 0.68, 95% CI 0.53–0.86), which was maintained at 6-
month follow-up [65]. A recent systematic overview of four major
randomized trials of early ACE inhibitor treatment following acute MI
showed that early ACE inhibitor therapy was associated with a 7.1% 30-day
mortality, compared with 7.6% in control subjects (7% proportional
reduction; 95% CI 2%–11%) [66]. Diabetic patients showed a trend toward
improved survival with ACE inhibitor therapy (30-day mortality, 10.3% in
those treated with an ACE inhibitor versus 12.0% in control subjects). The
1046 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

absolute benefits of ACE inhibitor treatment were greater for diabetic versus
nondiabetic patients (17.3 lives saved per 1000 versus 3.2 live saved per 1000,
respectively).
ACE inhibitor therapy is also beneficial in diabetic patients with at least one
cardiovascular risk factor but no history of cardiovascular disease. The Heart
Outcomes Prevention Evaluation (HOPE) study and the Microalbuminuria,
Cardiovascular, and Renal Outcomes (MICRO) HOPE substudy demon-
strated that ramipril therapy was associated with a 25% reduction in the
combined endpoint of MI, stroke, or cardiovascular death (95% CI 12–36,
P = 0.0004) [67]. This benefit was seen irrespective of a history of cardiovas-
cular disease. Additionally, recent results from the Efficacy of Perindopril in
Reduction of Cardiovascular Events among Patients with Stable Coronary
Artery Disease (EUROPA) study [68] demonstrate the benefit of ACE
inhibitor therapy in diabetic patients with stable CVD.
Calcium channel blockers (CCB) induce coronary and peripheral vasodi-
latation, decrease heart rate, and reduce cardiac contractility. As a group,
CCBs are effective antianginal agents. Although CCBs are generally used in
hypertensive diabetic patients who require multiple drugs for blood pressure
control, CCBs may also be used for therapy of angina in the presence of
contraindications or significant side effects to BBs [69]. In addition, CCBs may
be added to BBs or nitrates if these drugs provide inadequate symptom relief.
The dihydropyridine class of drugs (eg, amlodipine) produces more peripheral
vasodilatation and less negative chronotropic and inotropic effects than CCBs
such as diltiazem and verapamil. Use of short-acting CCBs (eg, nifedipine) is
discouraged because of data suggesting an increased risk of MI [70,71]. On the
other hand, CCBs are the treatment of choice for Prinzmetal’s (or variant)
angina.
In addition to pharmacotherapy for angina, diabetic patients with ischemic
heart disease should undergo intensive cardiac risk factor modification to
prevent progression of atherosclerosis. These therapies include treatment of
hypertension (target blood pressure 130/85 mm Hg), lipid-lowering thera-
pies (target LDL cholesterol 100 mg/dl, triglycerides 150 mg/dl, HDL 40
mg/dl), smoking cessation, glycemic control (target hemoglobin A1c 6.5%),
and weight control. For patients with chronic stable angina that is refractory
to conventional antianginal medications and CVD that is not amenable to
further surgical or percutaneous revascularization, enhanced external coun-
terpulsation is a mechanical therapeutic option that may provide symptom
relief [72].

Coronary revascularization
Despite maximal medical therapy, many diabetics require coronary artery
revascularization, using methods such as percutaneous transluminal coronary
angioplasty (PTCA), PCI with coronary stenting, or CABG for treatment of
ischemic heart disease. It is clear that diabetic patients have a higher risk
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1047

adverse outcomes following either percutaneous or surgical revascularization


than nondiabetic patients [73–75]. According to the National Heart, Lung,
and Blood Institute’s 1985–1986 PTCA registry, despite similar procedural
success rates of balloon angioplasty, diabetic patients had a higher rate of in-
hospital death, MI, and emergent CABG, compared with nondiabetic patients
(11% versus 6.7%, OR 1.74, P \ 0.001) and a lower rate of survival at 9-year
follow-up (67.4% versus 83.9%, P \ 0.0001) [75]. In pooled analyses of PCI
with coronary stenting, diabetic patients were found to have higher rates of
death, MI, or repeat revascularization at 6 months or 1-year follow-up (21.1%
versus 15.3%, OR 1.5%, 95%CI 1.3–1.7), compared with their nondiabetic
counterparts [76]. Similarly, in a surgical series of CABG [74], diabetic patients
had a higher incidence of postoperative death (3.9% versus 1.6%, P  0.05)
and a lower 10-year survival (50% versus 71%, P  0.05), compared with
nondiabetic patients [74].
The advent of coronary stent implantation has significantly improved the
results of PCI, specifically by reducing the rate of restenosis and by
decreasing the need for TVR [77–79]. In diabetic patients, although coronary
stenting has improved outcomes compared with PTCA, diabetes remains an
independent risk factor for adverse events, and diabetic patients continue to
have a poorer prognosis than nondiabetic patients [80]. Intracoronary
radiation may be used to treat in-stent restenosis in both diabetics and
nondiabetics. Drug-eluting stents (eg, sirolimus and paclitaxel-eluting stents)
have been associated with a marked reduction in the development of
restenosis [81]. The preliminary results of the Sirolimus-Eluting Stent in
Native Coronary Lesions (SIRIUS) trial demonstrated a significant im-
provement with the use of sirolimus-coated stents compared with bare metal
stents, with respect to in-segment restenosis (8.9% versus 36.3%) and target
lesion revascularization (4.1% versus 16.6%) [82]. In the recently presented
TAXUS IV trial, the use of a paclitaxel-coated stent was associated with
a reduction of target lesion revascularization from 11.3% to 3.0% (P \
0.0001) and of in-stent binary restenosis from 24.4% to 5.5%, compared with
bare metal stents. The composite endpoint of cardiac death, MI, or TVR was
also reduced from 15.0% to 8.5% (P = 0.0002) [83]. The effects of drug-
eluting stents in the diabetic subpopulation are currently being evaluated.
The optimal method of coronary revascularization (PTCA versus PCI with
stenting versus CABG) has been a matter of long-standing debate. Initial data
from two large randomized trials demonstrated a survival benefit in diabetic
patients treated with CABG rather than PTCA for multivessel CVD [84,85].
This finding was supported by data from the landmark Bypass Angioplasty
Revascularization Investigation (BARI) study [86]. In this study, the 5- and
7-year survival rates were similar between PTCA and CABG in the overall
study population. In the diabetic subgroup (n = 353), however, 5- and 7-year
survival rates were higher in the group assigned to CABG, compared with
PTCA (5-year survival, 80.6% versus 65.5% [P = 0.003]; 7-year survival,
76.4% versus 55.7% [P=0.0011]) [87,88]. The results of the BARI
1048 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

randomized trial were not seen in the BARI registry, raising some questions
about generalizing the results to diabetic patients outside of controlled clinical
trials [89].
Two major trials, the Arterial Revascularization Therapies Study (ARTS)
and Stent or Surgery (SOS) study, were designed to compare PCI with
coronary stenting versus CABG [90,91]. Both trials found no significant
difference in death or MI between the PCI and CABG groups in long-term
follow-up; however, the need for repeat TVR was increased in the PCI groups
compared with the CABG groups (21% versus 3.8% in ARTS, P \ 0.001).
The diabetic patients in ARTS (n = 208) had a higher 1-year event-free
survival with CABG, compared with PCI (84.4% versus 63.4%, P \ 0.001)
[92]. Most of this difference resulted from the increased need for repeat TVR
following PCI. Despite these data favoring selection of CABG in diabetic
patients, advances in adjunctive therapies during PCI and the recent in-
troduction of drug-eluting stents leave the question about the optimal
revascularization strategy for diabetic patients unanswered. Further studies
are ongoing to help guide clinical decision making. In addition, the BARI 2D
study has been designed to evaluate strategies for glycemic control during
revascularization [93].

Heart failure in diabetic patients: epidemiology


HF is common in the diabetic population, with an overall prevalence of
12%, based on community-based studies [94]. Diabetes is present in 24% to
28% of patients enrolled in large-scale clinical trials of HF therapies [95–97].
Furthermore, diabetic patients constitute 25% to 30% of all patients
hospitalized for HF [98,99]. Overall, the estimated incidence of developing
symptomatic HF is 3.3% per 100 person-years [94]. This risk is increased
2.4-fold in diabetic men and 5-fold in diabetic women, compared with
matched controls, independent of coexisting hypertension or ischemic heart
disease [100]. According to the ACC/AHA HF guidelines, the presence of
diabetes mellitus is regarded as stage A HF (patients at risk of HF) or stage
B HF (patients with structural abnormalities such as left ventricular
hypertrophy without overt cardiac dysfunction) [101].
Conversely, there is some evidence that HF may be an independent risk
factor for the development of diabetes. In the BIP study, over a 7.7-year
follow-up period, diabetes developed in 13% of patients with New York Heart
Association (NYHA) class I HF, 15% with NYHA class II, and 20% with
NYHA class III (P for trend = 0.05) [102]. In an elderly cohort of patients
followed for 3 years, HF predicted development of noninsulin dependent
diabetes independently of age, sex, family history of diabetes, body mass
index, waist/hip ratio, systolic and diastolic blood pressure, and therapy for
HF (OR = 1.4, 95% CI = 1.1–1.8) [103]. The association between HF and
diabetes in this study became stronger as HF severity increased. Interestingly,
the use of drugs such as losartan and carvedilol has been associated with
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1049

a significant reduction in the incidence of new-onset diabetes mellitus in


patients with HF, when compared with placebo [104,105]. Therefore, in
addition to regular screening for diabetes (minimum every 3 years), strategies
to prevent diabetes in patients with HF should be used whenever possible.
Although the evidence is controversial, most experts acknowledge the
existence of a distinct diabetic cardiomyopathy. Morphologic features
associated with diabetic cardiomyopathy include myocyte hypertrophy,
interstitial fibrosis, intramyocardial microangiopathy, and infiltration with
periodic acid-Schiff-positive materials [106,107]. Many of these histologic
findings, however, are nonspecific and are present in other forms of non-
ischemic cardiomyopathies. Population-based studies, such as the Framing-
ham Heart Study and Strong Heart Study, have shown that in diabetic
patients, characteristic echocardiographic structural and functional abnor-
malities such as increased left ventricular mass, wall thickness, and measures
of diastolic dysfunction can precede overt HF by many years [108,109].
Recently, investigators have raised the possibility that plasma B-type
natriuretic peptide (BNP) may be helpful in detecting early left ventricular
dysfunction in diabetic patients [110], but this hypothesis requires further
testing, and currently, plasma BNP is not recommended as a screening tool.
The incidence of new-onset HF following acute coronary syndromes is at
least 2 times higher in diabetic versus nondiabetic patients [111,112]. Heart
failure may occur in up to 50% of diabetic patients who suffer an acute MI
[113]. In addition, diabetic patients in the GUSTO IIb trial had increased
mortality after MI, independent of the presence of ST elevation [114]. Stress-
induced hyperglycemia during acute MI has been consistently shown to
increase mortality, even in patients without a previous of diabetes [115]. The
exact mechanisms underlying progression to HF in diabetic patients with
acute MI are unclear. The development of HF may be independent of the
initial degree of myocardial damage and of long-term ventricular remodel-
ing. In the Multicenter Investigation of the Limitation of Infarct Size
(MILIS) study, diabetic patients with smaller infarct sizes (lower peak
creatine phosphokinase levels, smaller areas under the curve of serial
creatine phosphokinase measurements, and fewer new Q waves on serial
electrocardiography) still had a worse prognosis compared with their
nondiabetic peers (4-year cardiac mortality 25.9% versus 14.5%). In the
Survival and Ventricular Enlargement (SAVE) echocardiographic substudy,
the degree of left ventricular dilatation 2 years after MI was actually greater
in nondiabetic compared with diabetic patients [116]. Furthermore, the
development of left ventricular dilatation 2 years after MI did not predict
subsequent HF in diabetic patients, even after multivariable adjustment.

Management of heart failure in patients with diabetes


Although the management of chronic heart failure in patients with
diabetes mellitus follows the same general approach used to treat these two
1050 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

syndromes individually, special considerations should be given when using


some HF drugs in the diabetic population (Table 3). Every effort should be
made to maximize myocardial blood flow (either by percutaneous or surgical
revascularization) in the failing heart to reverse the underlying, hibernating
myocardium. Diuretic medications have been the mainstay of therapy for
volume overload. Loop diuretics, the most commonly used diuretics, can be
highly effective in both diabetic and nondiabetic patients; however, careful
monitoring of serum electrolytes and renal function is required. Anecdotal
observations have demonstrated that high-dose thiazide diuretics may impair
insulin sensitivity [117]. These diuretics should be avoided unless clinically
required. Use of spironolactone, a nonselective aldosterone antagonist and
a weak diuretic, in patients with severe HF and left ventricular ejection
fraction (LVEF) less than 35% conferred a 30% reduction in the risk of
death, a 35% reduction in hospitalization for worsening HF, and a significant
improvement in the symptoms of HF in the Randomized Aldactone
Evaluation Study (RALES) [118]. Unlike other diuretics, spironolactone
does not impair glucose tolerance. A selective aldosterone antagonist,
eplerenone, was recently shown to reduce death from cardiovascular causes
versus placebo in patients with post-MI HF (relative risk, 0.83, 95% CI 0.72–
0.94, P = 0.005) [119]. Although they demonstrate potential benefits in
diabetic patients with advanced HF, these two drugs may substantially
increase the risk of developing hyperkalemia and renal insufficiency and
should be used with caution in the diabetic population [120].
ACE inhibitors (or angiotensin-II receptor antagonists for ACE-inhibitor
intolerant patients) should be given at target doses to every diabetic patient
with HF unless contraindicated. The evidence supporting their use was
discussed previously. It should be noted that in addition to conferring
a mortality reduction benefit in those with LV dysfunction, ACE inhibitors
may increase insulin sensitivity in diabetic patients by affecting the early
steps of insulin signaling [121].
Angiotensin II receptor blocker (ARB) drugs act selectively at the
angiotensin II type (AT)1 receptor. Data on the benefit of ARBs in patients
with HF have conflicted. In the Reduction in Endpoints in NIDDM with the
Angiotensin II Antagonist Losartan (RENAAL) trial, losartan use was not
associated with a significant reduction in all-cause or cardiovascular mortality
[122]. In the Valsartan Heart Failure Trial (Val-HeFT), there was no
significant difference in overall mortality in diabetic patients treated with
valsartan versus placebo [96]. In contrast, in the recently published Cande-
sartan in Heart Failure Assessment of Reduction in Mortality and Morbidity
(CHARM) trials, candesartan use was associated with a reduction of less than
40% in the composite endpoint of cardiovascular death or hospital admission
for HF versus placebo in patients with LVEF who were not receiving ACE
inhibitors because of previous intolerance (unadjusted hazard ratio 0.77, 95%
CI 0.67–0.89) and less than 40% in patients with LVEF who were receiving
concurrent therapy with ACE inhibitors (unadjusted hazard ratio 0.85, 95%
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1051

Table 3
Common heart failure drugs with special considerations for diabetic patients
Drug class Start dose Target Special considerations
(examples) (mg) dose (mg) in diabetics
ACE inhibitors
Captopril 6.25–12.5 50 3 times/d Indicated for all HF patients
Enalapril 2.5–5 10 2 times/d unless contraindicated (#
Lisinopril 2.5–5 10–20 4 times/d blood pressure, "potassium,
Ramipril 1.25–2.5 5 2 times/d "creatinine (use hydralazine/
isosorbide dinitrate if
creatinine 3 mg/dL,
angioedema/cough)
b-Blockers
Carvedilol 3.125–6.25 25 2 times/d Indicated for all systolic HF
Metoprolol 12.5–25 100 4 times/d patients unless contraindicated
succinate (#pulse rate, #blood pressure,
Bisoprolol 2.5–5 20 4 times/d heart block, reactive airway
disease)
Spironolactone 12.5–25 N/A Indicated for advanced systolic HF
(NYHA class III-IV), need to
closely watch for "potassium,
"creatinine in diabetic patients;
no need for up-titration
Hydralazine/ 25 3 times/d 100 3 times/d Indicated for ACE-inhibitor/ARB-
isosorbide 10 4 times/d 40 4 times/d intolerant patients and those
dinitrate with advanced renal insufficiency
Diuretics
Furosemide 20–40 Titrate to Indicated for symptomatic relief
Bumetanide 1–2 euvolemia from fluid retention; thiazides
Torsemide 1–10 (but not loop diuretics) may
Metolazone 2.5–5 attenuate insulin sensitivity
Digoxin 0.125 N/A Indicated for advanced HF to
prevent morbidity, particularly
with concomitant atrial
fibrillation; watch for toxicity
especially with amiodarone and
renal insufficiency; prefer a lower
dose (0.125 4 times/d or 4 times
every other day) especially in
elderly and in women; no need
for up-titration
ARBs
Losartan 25 50 4 times/d Indicated for ACE-inhibitor
Valsartan 80 160 4 times/d intolerant HF patients; appears
Candesartan 4 32 4 times/d to be beneficial when added to
ACE-inhibitor in HF patients
(CHARM) and possibly in CHF
patients with preserved LVEF
1052 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

CI 0.75–0.96) [123,124]. Further analyses of the diabetic patients in these


studies are pending, although preliminary evaluation of the nondiabetic
patients in the CHARM trial demonstrated a reduction in new-onset
diagnoses of diabetes in patients treated with candesartan. Overall, there
has not been a consistent class-specific mortality benefit associated with
ARBs. Nevertheless, ARBs definitely should be prescribed to patients who
cannot tolerate ACE inhibitors. Furthermore, there may be a role for adding
candesartan to ACE inhibitor therapy in patients who are able to tolerate the
combination.
BBs have been well tolerated and clearly effective in diabetic subgroups in
large-scale HF trials. Historically, BBs were contraindicated in HF patients
in general and in diabetic patients particularly because of adverse effects on
insulin sensitivity and lipid profiles. Several contemporary studies, however,
have firmly established the beneficial role of BBs in diabetic patients with
HF, with respect to survival, improvement in LVEF, and reversal of left
ventricular remodeling [125]. Moreover, the use of metoprolol succinate did
not result in an increased incidence of diabetes, worsening glycemic control,
or masking of hypoglycemic symptoms in the Metoprolol CR/XL Ran-
domized Intervention Trial in Congestive Heart Failure (MERIT-HF) [126].
The nonselective BB carvedilol is also well tolerated in diabetic patients
[127], and in a recent randomized, controlled trial, the Carvedilol Or
Metoprolol European Trial (COMET), all-cause mortality was lower with
the use of carvedilol (25 mg twice per day), compared with metoprolol
tartrate (50 mg twice a day) in patients with LVEF less than and NYHA
Class II-IV HF [97]. Thus, BB therapy should be attempted in all diabetic
patients with chronic stable HF.
Digoxin should be limited to patients with symptomatic congestive HF or
those with concomitant rapid atrial fibrillation. Although digoxin is effective
in reducing symptoms of HF and decreasing the number of HF hospital-
izations, it has not been shown to confer any mortality benefit, and digoxin
toxicity is common in diabetic patients with concurrent renal insufficiency
[128].
Given the multiple drug categories that are available for the treatment of
chronic HF, care must be taken to avoid polypharmacy and drug-drug
interactions. In the diabetic patient particularly, physicians must exercise
vigilance to prevent disturbances in electrolytes and renal function and to
maximize patient compliance with drug regimens. Although several drug
classes can confer benefit in the diabetic patient with HF, tolerability of
prescribed regimens is crucial to treating this multisystem disorder.

Glycemic control in diabetic patients with cardiovascular diseases


Metabolic interventions are exciting new therapies that are particularly
attractive for patients with diabetes. Metabolic intervention may improve
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1053

myocardial performance in multiple settings, including stable ischemia, acute


coronary syndromes, and chronic HF. In ischemia, metabolic perturbations
may blunt the compensatory response of the diabetic myocardium. Perturba-
tions include increased circulating plasma insulin, glucose, and fatty acid
levels, in addition to decreased rates of glucose uptake, glycolysis, and
myocardial pyruvate oxidation [129]. Therefore, one of the principal
objectives of metabolic interventions is to decrease myocardial uptake and
oxidation of free fatty acids and to increase myocardial uptake and use of
glucose and lactate [130].
The Swedish Diabetes and Insulin-Glucose Infusion in Acute Myocardial
Infarction (DIGAMI) study tested this concept by randomizing patients with
suspected acute myocardial infarction to an infusion of glucose-insulin-
potassium (GIK) plus standard medical therapy versus standard therapy
alone [131]. The DIGAMI protocol used an insulin and glucose infusion for
tight glycemic control for at least 24 hours during the acute event. This
regimen was followed by treatment with subcutaneous insulin for the
subsequent 3 months. At 1-year follow-up, mortality in the treatment group
was 18.6%, compared with 26.1% in the control group, representing a 29%
decrease in mortality with the use of GIK infusion. These results were
maintained during long-term follow-up and confirmed in a recent meta-
analysis of all trials on GIK infusion during acute MI [132].
In patients with chronic stable angina, two new classes of metabolic agents,
the 3-ketoacyl-coenzyme A thiolase (3-KAT) inhibitors and the partial fatty
acid oxidation (pFOX) inhibitors, have been used to treat symptoms of
ischemic heart disease. Trimetazidine (a prototype 3-KAT inhibitor) and
ranolazine (a prototype pFOX inhibitor) have been shown to improve exercise
tolerance and alleviate anginal symptoms [133–135], but neither drug is
currently available in the United States.
Tight glycemic control is an essential component of the management of HF
and ischemia. The UK Prospective Diabetes 35 study [136] showed that every
1% decrement in hemoglobin A1c level was associated with a 14% and 16%
relative reduction in the risk of developing MI and HF, respectively. Many
diabetic patients rely on oral hypoglycemic drugs, taken either alone or in
combination with insulin, to maintain glycemic control; however, there is
a paucity of information regarding the safety and efficacy or newer oral
hypoglycemic drugs in diabetic patients with HF. Both the biguanides (eg,
metformin) and the thiazolidinediones (TZDs) (eg, rosiglitazone and piogli-
tazone) are regarded as ‘‘relatively contraindicated’’ in diabetic patients with
HF. Cases of fatal lactic acidosis were reported in patients with HF and
diabetes who were treated with biguanides [137]. As a result, the use of
metformin is relatively contraindicated in patients who have renal insuffi-
ciency (serum creatinine 1.5 mg/dL), baseline liver function test abnormal-
ities, are elderly (age  80 years old), or who are taking medications for HF.
There is also a concern that the use of thiazolidinediones may cause overt HF.
Fluid retention, weight gain, and pulmonary edema have been reported in
1054 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

patients using TZDs [138–140]. The mechanism underlying this phenomenon


is unclear, but TZD-related fluid retention is largely restricted to the periphery
and is usually reversible with drug withdrawal [141]. These observations have
led to the hypothesis that TZDs may cause altered vascular permeability
leading to fluid retention. Because TZDs have multiple beneficial effects in
diabetic patients, including improvement in glycemic control, lipid profile,
and endothelial function, as well as amelioration of albuminuria and diabetic
nephropathy, further clarification of the relationships among TZDs, fluid
retention, and congestive HF will help physicians maximize the use of these
drugs in appropriate patients [142].

Summary
Ischemic heart disease and HF are major contributors to morbidity and
mortality in patients with diabetes mellitus. Although we have made great
progress in our understanding of the pathophysiology of cardiovascular
disease and diabetes mellitus, cooperation between cardiologists, diabetol-
ogists, and internists is needed to optimize and balance therapies for these
disorders.

References
[1] Laakso M, Lehto S. Epidemiology of risk factors for cardiovascular disease in diabetes
and impaired glucose tolerance. Atherosclerosis 1998;137(Suppl):S65–73.
[2] Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M, et al. Mortality from
coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with
and without prior myocardial infarction. N Engl J Med 1998;339:229–34.
[3] Ruige JB, Assendelft WJ, Dekker JM, et al. Insulin and risk of cardiovascular disease:
a meta-analysis. Circulation 1998;97:996–1001.
[4] Stout RW. Insulin and atheroma. 20-yr perspective. Diabetes Care 1990;13:631–54.
[5] Despres JP, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S, et al.
Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med
1996;334:952–7.
[6] Arcaro G, Cretti A, Balzano S, Lechi A, Muggeo M, Bonora E, et al. Insulin causes
endothelial dysfunction in humans: sites and mechanisms. Circulation 2002;105:576–82.
[7] Bierhaus A, Hofmann MA, Ziegler R, Nawroth PP, et al. AGEs and their interaction
with AGE-receptors in vascular disease and diabetes mellitus. I. The AGE concept.
Cardiovasc Res 1998;37:586–600.
[8] Evans JL, Goldfine ID, Maddux BA, Grodsky GM, et al. Are oxidative stress-activated
signaling pathways mediators of insulin resistance and beta-cell dysfunction? Diabetes
2003;52:1–8.
[9] Marfella R, Esposito K, Giunta R, Coppola G, De Angelis L, Farzati B, et al. Circulating
adhesion molecules in humans: role of hyperglycemia and hyperinsulinemia. Circulation
2000;101:2247–51.
[10] Laakso M. Lipids and lipoproteins as risk factors for coronary heart disease in
non-insulin-dependent diabetes mellitus. Ann Med 1996;28:341–5.
[11] Syvanne M, Taskinen MR. Lipids and lipoproteins as coronary risk factors in
non-insulin-dependent diabetes mellitus. Lancet 1997;350(Suppl 1):SI20–3.
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1055

[12] Matsuda T, Morishita E, Jokaji H, Asakura H, Saito M, Yoshida T, et al. Mechanism on


disorders of coagulation and fibrinolysis in diabetes. Diabetes 1996;45(Suppl 3):S109–10.
[13] Eliasson M, Asplund K, Evrin PE, Lindahl B, Lundblad D. Hyperinsulinemia predicts
low tissue plasminogen activator activity in a healthy population: the Northern Sweden
MONICA Study. Metabolism 1994;43:1579–86.
[14] Tschoepe D, Roesen P, Schwippert B, Gries FA. Platelets in diabetes: the role in the
hemostatic regulation in atherosclerosis. Semin Thromb Hemost 1993;19:122–8.
[15] Davi G, Catalano I, Averna M, Notarbartolo A, Strano A, Ciabattoni G, et al.
Thromboxane biosynthesis and platelet function in type II diabetes mellitus. N Engl J
Med 1990;322:1769–74.
[16] Knobler H, Savion N, Shenkman B, Kotev-Emeth S, Varon D. Shear-induced platelet
adhesion and aggregation on subendothelium are increased in diabetic patients. Thromb
Res 1998;90:181–90.
[17] Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM, et al. C-reactive protein,
interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 2001;286:327–34.
[18] Biondi-Zoccai GG, Abbate A, Liuzzo G, Biasucci LM. Atherothrombosis, inflammation,
and diabetes. J Am Coll Cardiol 2003;41:1071–7.
[19] Milan Study on Atherosclerosis and Diabetes (MiSAD) Group. Prevalence of un-
recognized silent myocardial ischemia and its association with atherosclerotic risk factors
in noninsulin-dependent diabetes mellitus. Am J Cardiol 1997;79:134–9.
[20] Koistinen MJ. Prevalence of asymptomatic myocardial ischaemia in diabetic subjects.
BMJ 1990;301:92–5.
[21] Naka M, Hiramatsu K, Aizawa T, Momose A, Yoshizawa K, Shigematsu S, et al. Silent
myocardial ischemia in patients with non-insulin-dependent diabetes mellitus as judged by
treadmill exercise testing and coronary angiography. Am Heart J 1992;123:46–53.
[22] Droste C, Meyer-Blankenburg H, Greenlee MW, Roskamm H, et al. Effect of physical
exercise on pain thresholds and plasma beta-endorphins in patients with silent and
symptomatic myocardial ischaemia. Eur Heart J 1988;(9 Suppl):25–33.
[23] Marchant B, Umachandran V, Wilkinson P, Medback S, Kopelman PG, Timmis AD,
et al. Reexamination of the role of endogenous opiates in silent myocardial ischemia.
J Am Coll Cardiol 1994;23:645–51.
[24] Koistinen MJ, Airaksinen KE, Huikuri HV, Pirttiaho H, Linnaluoto MK, Ikaheimo MJ,
et al. Asymptomatic coronary artery disease in diabetes: associated with autonomic
neuropathy? Acta Diabetol 1992;28:199–202.
[25] Zarich S, Waxman S, Freeman RT, Mittleman M, Hegarty P, Nesto RW, et al. Effect of
autonomic nervous system dysfunction on the circadian pattern of myocardial ischemia in
diabetes mellitus. J Am Coll Cardiol 1994;24:956–62.
[26] Chierchia S, Brunelli C, Simonetti I, Lazzari M, Maseri A. Sequence of events in angina
at rest: primary reduction in coronary flow. Circulation 1980;61:759–68.
[27] Deedwania PC, Carbajal EV. Silent ischemia during daily life is an independent predictor
of mortality in stable angina. Circulation 1990;81:748–56.
[28] Rogers WJ, Bourassa MG, Andrews TC, Bertolet BD, Blumenthal RS, Chaitman BR,
et al. Asymptomatic Cardiac Ischemia Pilot (ACIP) study: outcome at 1 year for patients
with asymptomatic cardiac ischemia randomized to medical therapy or revascularization.
The ACIP Investigators. J Am Coll Cardiol 1995;26:594–605.
[29] American Diabetes Association. Diabetes mellitus and exercise. Diabetes Care 1997;20:
1908–12.
[30] Gibbons RJ, Balady GJ, Timothy Bricker J, Chaitman BR, Fletcher GF, Froelicher VF,
et al. ACC/AHA 2002 guideline update for exercise testing: summary article. A report of
the American College of Cardiology/American Heart Association Task Force on Practice
Guidelines (Committee to Update the 1997 Exercise Testing Guidelines). J Am Coll
Cardiol 2002;40:1531–40.
1056 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

[31] Kang X, Berman DS, Lewin HC, Cohen I, Friedman JD, Germano G, et al. Incremental
prognostic value of myocardial perfusion single photon emission computed tomography
in patients with diabetes mellitus. Am Heart J 1999;138:1025–32.
[32] Kang X, Berman DS, Lewin H, Miranda R, Erel J, Friedman JD, et al. Comparative
ability of myocardial perfusion single-photon emission computed tomography to detect
coronary artery disease in patients with and without diabetes mellitus. Am Heart J 1999;
137:949–57.
[33] Giri S, Shaw LJ, Murthy DR, Travin MI, Miller DD, Hachamovitch R, et al. Impact
of diabetes on the risk stratification using stress single-photon emission computed
tomography myocardial perfusion imaging in patients with symptoms suggestive of
coronary artery disease. Circulation 2002;105:32–40.
[34] De Lorenzo A, Lima RS, Siqueira-Filho AG, Pantoja MR. Prevalence and prognostic
value of perfusion defects detected by stress technetium-99m sestamibi myocardial
perfusion single-photon emission computed tomography in asymptomatic patients with
diabetes mellitus and no known coronary artery disease. Am J Cardiol 2002;90:827–32.
[35] Wackers FJ, Zaret BL. Detection of myocardial ischemia in patients with diabetes
mellitus. Circulation 2002;105:5–7.
[36] Elhendy A, Arruda AM, Mahoney DW, Pellikka PA. Prognostic stratification of diabetic
patients by exercise echocardiography. J Am Coll Cardiol 2001;37:1551–7.
[37] Hoff JA, Quinn L, Sevrukov A, Lipton RB, Daviglus M, Garside DB, et al. The
prevalence of coronary artery calcium among diabetic individuals without known
coronary artery disease. J Am Coll Cardiol 2003;41:1008–12.
[38] Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB.
Prediction of coronary heart disease using risk factor categories. Circulation 1998;97:
1837–42.
[39] Scanlon PJ, Faxon DP, Audet AM, Carabello B, Dehmer GJ, Eagte KA, et al. ACC/
AHA guidelines for coronary angiography. A report of the American College of
Cardiology/American Heart Association Task Force on practice guidelines (Committee
on Coronary Angiography). Developed in collaboration with the Society for Cardiac
Angiography and Interventions. J Am Coll Cardiol 1999;33:1756–824.
[40] Lebacq EG, Tirzmalis A. Metformin and lactic acidosis. Lancet 1972;1:314–5.
[41] Misbin RI, Green L, Stadel BV, Gueriguian JL, Gubbi A, Fleming GA. Lactic acidosis in
patients with diabetes treated with metformin. N Engl J Med 1998;338:265–6.
[42] McVeigh G, Brennan G, Hayes R, Johnston D, et al. Primary nitrate tolerance in diabetes
mellitus. Diabetologia 1994;37:115–7.
[43] McVeigh GE, Morgan DR, Allen P, Trimble M, Hamilton P, Dixon LJ, et al. Early
vascular abnormalities and de novo nitrate tolerance in diabetes mellitus. Diabetes Obes
Metab 2002;4:336–41.
[44] Colwell JA. Aspirin therapy in diabetes. Diabetes Care 2003;26(Suppl 1):S87–8.
[45] Antiplatelet Trialists’ Collaboration. Collaborative overview of randomized trials of
antiplatelet therapy—I: prevention of death, myocardial infarction, and stroke by
prolonged antiplatelet therapy in various categories of patients. BMJ 1994;308:81–106.
[46] Antiplatelet Trialists’ Collaboration. Collaborative meta-analysis of randomised trials of
antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk
patients. BMJ 2002;324:71–86.
[47] CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin
in patients at risk of ischaemic events (CAPRIE). Lancet 1996;348:1329–39.
[48] Bhatt DL, Marso SP, Hirsch AT, Ringleb PA, Hacke W, Topel EJ, et al. Amplified
benefit of clopidogrel versus aspirin in patients with diabetes mellitus. Am J Cardiol 2002;
90:625–8.
[49] Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. Effects of clopidogrel
in addition to aspirin in patients with acute coronary syndromes without ST-segment
elevation. N Engl J Med 2001;345:494–502.
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1057

[50] Braunwald E, Antman EM, Beasley JW, Califf RM, Cheitlin MD, Hochman JS, et al.
ACC/AHA 2002 guideline update for the management of patients with unstable angina
and non-ST-segment elevation myocardial infarction–summary article: a report of the
American College of Cardiology/American Heart Association task force on practice
guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll
Cardiol 2002;40:1366–74.
[51] Lincoff AM, Califf RM, Topol EJ. Platelet glycoprotein IIb/IIIa receptor blockade in
coronary artery disease. J Am Coll Cardiol 2000;35:1103–15.
[52] Anderson KM, Califf RM, Stone GW, Neumann FJ, Montalescot G, Miller DP, et al. Long-
term mortality benefit with abciximab in patients undergoing percutaneous coronary
intervention. J Am Coll Cardiol 2001;37:2059–65.
[53] Lincoff AM. Important triad in cardiovascular medicine: diabetes, coronary intervention,
and platelet glycoprotein IIb/IIIa receptor blockade. Circulation 2003;107:1556–9.
[54] Topol EJ, Mark DB, Lincoff AM, Cohen E, Burton J, Kleiman N, et al. Outcomes at 1 year
and economic implications of platelet glycoprotein IIb/IIIa blockade in patients undergoing
coronary stenting: results from a multicentre randomised trial. EPISTENT Investigators.
Evaluation of Platelet IIb/IIIa Inhibitor for Stenting. Lancet 1999;354:2019–24.
[55] Topol EJ, Lincoff AM, Kereiakes DJ, Kleirnan NS, Cohen EA, Ferguson JJ, et al. Multi-
year follow-up of abciximab therapy in three randomized, placebo-controlled trials of
percutaneous coronary revascularization. Am J Med 2002;113:1–6.
[56] Kereiakes DJ, Lincoff AM, Anderson KM, Achenbach R, Patel K, Barnathan E, et al.
Abciximab survival advantage following percutaneous coronary intervention is predicted
by clinical risk profile. Am J Cardiol 2002;90:628–30.
[57] Boersma E, Harrington RA, Moliterno DJ, White H, Theroux P, Van de Werf F, et al.
Platelet glycoprotein IIb/IIIa inhibitors in acute coronary syndromes: a meta-analysis of
all major randomised clinical trials. Lancet 2002;359:189–98.
[58] Majumdar SR. Beta-blockers for the treatment of hypertension in patients with diabetes:
exploring the contraindication myth. Cardiovasc Drugs Ther 1999;13:435–9.
[59] Marengo C, Marena S, Renzetti A, Mossino M, Pagano G, et al. Beta-blockers in
hypertensive non-insulin-dependent diabetics: comparison between penbutolol and pro-
pranolol on metabolic control and response to insulin-induced hypoglycemia. Acta
Diabetol Lat 1988;25:141–8.
[60] Jonas M, Reicher-Reiss H, Boyko V, Shotan A, Mandelzweig L, Goldbourt U, et al.
Usefulness of beta-blocker therapy in patients with non-insulin-dependent diabetes
mellitus and coronary artery disease. Bezafibrate Infarction Prevention (BIP) Study
Group. Am J Cardiol 1996;77:1273–7.
[61] Gundersen T. Secondary prevention after myocardial infarction: subgroup analysis of
patients at risk in the Norwegian Timolol Multicenter Study. Clin Cardiol 1985;8:253–65.
[62] Malmberg K, Herlitz J, Hjalmarson A, Ryden L. Effects of metoprolol on mortality and
late infarction in diabetics with suspected acute myocardial infarction. Retrospective data
from two large studies. Eur Heart J 1989;10:423–8.
[63] Krumholz HM, Radford MJ, Wang Y, Chen J, Heiat A, Marciniak TA. National use and
effectiveness of beta-blockers for the treatment of elderly patients after acute myocardial
infarction. National Cooperative Cardiovascular Project. Jama 1998;280:623–9.
[64] Gruppo Italiano per lo Studio della Sopravvivenza nell’infarto Miocardico. GISSI-3: effects
of lisinopril and transdermal glyceryl trinitrate singly and together on 6-week mortality and
ventricular function after acute myocardial infarction. Lancet 1994;343:1115–22.
[65] Zuanetti G, Latini R, Maggioni AP, Franzosi M, Santoro L, Tagnoni G. Effect of the
ACE inhibitor lisinopril on mortality in diabetic patients with acute myocardial
infarction: data from the GISSI-3 study. Circulation 1997;96:4239–45.
[66] ACE Inhibitor Myocardial Infarction Collaborative Group. Circulation. Indications for
ACE inhibitors in the early treatment of acute myocardial infarction: systematic overview
of individual data from 100,000 patients in randomized trials. 1998;97:2202–12.
1058 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

[67] Heart Outcomes Prevention Evaluation Study Investigators. Effects of ramipril on


cardiovascular and microvascular outcomes in people with diabetes mellitus: results of the
HOPE study and MICRO-HOPE substudy. Lancet 2000;355:253–9.
[68] Fox KM. Efficacy of perindopril in reduction of cardiovascular events among patients
with stable coronary artery disease: randomised, double-blind, placebo-controlled,
multicentre trial (the EUROPA study). Lancet 2003;362:782.
[69] Messerli FH. Evolution of calcium antagonists: past, present, and future. Clin Cardiol
2003;26:II12–6.
[70] Furberg CD, Psaty BM, Meyer JV. Nifedipine. Dose-related increase in mortality in
patients with coronary heart disease. Circulation 1995;92:1326–31.
[71] Furberg CD, Psaty BM. Corrections to the nifedipine meta-analysis. Circulation 1996;93:
1475–6.
[72] Linnemeier G, Rutter MK, Barsness G, Kennard ED, Nesto RW. Enhanced External
Counterpulsation for the relief of angina in patients with diabetes: safety, efficacy and 1-
year clinical outcomes. Am Heart J 2003;146:453–8.
[73] Stein B, Weintraub WS, Gebhart SP, Cohen-Bernstein CL, Grosswald R, Liberman
HA, et al. Influence of diabetes mellitus on early and late outcome after percutaneous
transluminal coronary angioplasty. Circulation 1995;91:979–89.
[74] Thourani VH, Weintraub WS, Stein B, Gebhart SS, Craver JM, Jones EL, et al. Influence
of diabetes mellitus on early and late outcome after coronary artery bypass grafting. Ann
Thorac Surg 1999;67:1045–52.
[75] Kip KE, Faxon DP, Detre KM, Yeh W, Kelsey SF, Currier JW. Coronary angioplasty in
diabetic patients. The National Heart, Lung, and Blood Institute Percutaneous
Transluminal Coronary Angioplasty Registry. Circulation 1996;94:1818–25.
[76] Mak KH, Faxon DP. Clinical studies on coronary revascularization in patients with
type 2 diabetes. Eur Heart J 2003;24:1087–103.
[77] Van Belle E, Bauters C, Hubert E, Bodart JC, Abolmaali K, Meurice T, et al. Restenosis
rates in diabetic patients: a comparison of coronary stenting and balloon angioplasty in
native coronary vessels. Circulation 1997;96:1454–60.
[78] Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, et al. A randomized
comparison of coronary-stent placement and balloon angioplasty in the treatment of
coronary artery disease. Stent Restenosis Study Investigators. N Engl J Med 1994;331:
496–501.
[79] Mehilli J, Kastrati A, Dirschinger J, Dotzer F, Pache J, Hausleiter J, et al. Comparison of
stenting with balloon angioplasty for lesions of small coronary vessels in patients with
diabetes mellitus. Am J Med 2002;112:13–8.
[80] Abizaid A, Kornowski R, Mintz GS, Hong MK, Abizaid AS, Mehran R, et al. The
influence of diabetes mellitus on acute and late clinical outcomes following coronary stent
implantation. J Am Coll Cardiol 1998;32:584–9.
[81] Morice MC, Serruys PW, Sousa JE, Fajadet J, Ben Hayashi E, Perin M, et al. A
randomized comparison of a sirolimus-eluting stent with a standard stent for coronary
revascularization. N Engl J Med 2002;346:1773–80.
[82] Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, et al. One-year
clinical results with the slow-release, polymer-based, paclitaxel-eluting TAXUS stent: the
TAXUS-IV trial. Circulation 2004;109:1942–7.
[83] TAXUS IV Angiographic Study. Presented by Stephen G. Ellis at Transcatheter
Cardiovascular Therapeutics (TCT) [2003]. Washington DC, 23 September, 2003.
[84] Kurbaan AS, Bowker TJ, Ilsley CD, Sigwart U, Rickards AF. Difference in the mortality
of the CABRI diabetic and nondiabetic populations and its relation to coronary artery
disease and the revascularization mode. Am J Cardiol 2001;87:947–50.
[85] King SB 3rd, Kosinski AS, Guyton RA, Lembo NJ, Weintraub WS, et al. Eight-year
mortality in the Emory Angioplasty versus Surgery Trial (EAST). J Am Coll Cardiol
2000;35:1116–21.
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1059

[86] The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Compar-


ison of coronary bypass surgery with angioplasty in patients with multivessel disease.
N Engl J Med 1996;335:217–25.
[87] The BARI Investigators. Influence of diabetes on 5-year mortality and morbidity in
a randomized trial comparing CABG and PTCA in patients with multivessel disease: the
Bypass Angioplasty Revascularization Investigation (BARI). Circulation 1997;96:1761–9.
[88] The BARI Investigators. Seven-year outcome in the Bypass Angioplasty Revascularization
Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000;35:1122–9.
[89] Feit F, Brooks MM, Sopko G, Keller NM, Rosen A, Krone R, et al. Long-term clinical
outcome in the Bypass Angioplasty Revascularization Investigation Registry: comparison
with the randomized trial. BARI Investigators. Circulation 2000;101:2795–802.
[90] SoS Investigators. Coronary artery bypass surgery versus percutaneous coronary
intervention with stent implantation in patients with multivessel coronary artery disease
(the Stent or Surgery trial): a randomised controlled trial. Lancet 2002;360:965–70.
[91] Serruys PW, Unger F, Sousa JE, Jatene A, Bennier HJ, Schonberger JP, et al.
Comparison of coronary-artery bypass surgery and stenting for the treatment of multi-
vessel disease. N Engl J Med 2001;344:1117–24.
[92] Abizaid A, Costa MA, Centemero M, Abizaid AS, Legrand VM, Limet RV, et al. Clinical
and economic impact of diabetes mellitus on percutaneous and surgical treatment of
multivessel coronary disease patients: insights from the Arterial Revascularization
Therapy Study (ARTS) trial. Circulation 2001;104:533–8.
[93] Sobel BE, Frye R, Detre KM. Burgeoning dilemmas in the management of diabetes and
cardiovascular disease: rationale for the Bypass Angioplasty Revascularization In-
vestigation 2 Diabetes (BARI 2D) Trial. Circulation 2003;107:636–42.
[94] Nichols GA, Hillier TA, Erbey JR, Brown JB. Congestive heart failure in type 2 diabetes:
prevalence, incidence, and risk factors. Diabetes Care 2001;24:1614–9.
[95] Pfeffer MA, Swedberg K, Granger CB, et al. Effects of candesartan on mortality and
morbidity in patients with chronic heart failure: the CHARM-Overall programme.
Lancet 2003;362:759–66.
[96] Cohn JN, Tognoni G. A randomized trial of the angiotensin-receptor blocker valsartan in
chronic heart failure. N Engl J Med 2001;345:1667–75.
[97] Poole-Wilson PA, Swedberg K, Cleland JG, Di Lenarda A, Hanrath P, Komajda M,
et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with
chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET):
randomised controlled trial. Lancet 2003;362:7–13.
[98] Polanczyk CA, Rohde LE, Dec GW, Disalvo T. Ten-year trends in hospital care for
congestive heart failure: improved outcomes and increased use of resources. Arch Intern
Med 2000;160:325–32.
[99] Reis SE, Holubkov R, Edmundowicz D, McNamara DM, Zell KA, Detre KM, et al.
Treatment of patients admitted to the hospital with congestive heart failure: specialty-
related disparities in practice patterns and outcomes. J Am Coll Cardiol 1997;30:733–8.
[100] Kannel WB, Hjortland M, Castelli WP. Role of diabetes in congestive heart failure: the
Framingham study. Am J Cardiol 1974;34:29–34.
[101] Hunt SA, Baker DW, Chin MH, Cinquegrani MP, Feldmanmd AM, Francis GS, et al.
ACC/AHA Guidelines for the Evaluation and Management of Chronic Heart Failure in
the Adult: Executive Summary A Report of the American College of Cardiology/
American Heart Association Task Force on Practice Guidelines (Committee to Revise the
1995 Guidelines for the Evaluation and Management of Heart Failure): Developed in
Collaboration With the International Society for Heart and Lung Transplantation;
Endorsed by the Heart Failure Society of America. Circulation 2001;104:2996–3007.
[102] Tenenbaum A, Motro M, Fisman EZ, Leor J, Freimark D, Boyko V, et al. Functional
class in patients with heart failure is associated with the development of diabetes. Am J
Med 2003;114:271–5.
1060 W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061

[103] Amato L, Paolisso G, Cacciatore F, et al. Congestive heart failure predicts the
development of non-insulin-dependent diabetes mellitus in the elderly. The Osservatorio
Geriatrico Regione Campania Group. Diabetes Metab 1997;23:213–8.
[104] Poole-Wilson PA. Clinical Trials Update II (Heart Failure)-The Carvedilol Or
Metoprolol European Trial (COMET). European Society of Cardiology Congress.
Vienna, Austria, 2 September, 2003.
[105] Dahlof B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, Faire U, et al. Cardio-
vascular morbidity and mortality in the Losartan Intervention For Endpoint reduction
in hypertension study (LIFE): a randomised trial against atenolol. Lancet 2002;359:
995–1003.
[106] Hardin NJ. The myocardial and vascular pathology of diabetic cardiomyopathy. Coron
Artery Dis 1996;7:99–108.
[107] van Hoeven KH, Factor SM. A comparison of the pathological spectrum of hypertensive,
diabetic, and hypertensive-diabetic heart disease. Circulation 1990;82:848–55.
[108] Galderisi M, Anderson KM, Wilson PW, Levy D. Echocardiographic evidence for the
existence of a distinct diabetic cardiomyopathy (the Framingham Heart Study). Am J
Cardiol 1991;68:85–9.
[109] Devereux RB, Roman MJ, Paranicas M, et al. Impact of diabetes on cardiac structure
and function: the strong heart study. Circulation 2000;101:2271–6.
[110] Epshteyn V, Morrison K, Krishnaswamy P, et al. Utility of B-type natriuretic peptide
(BNP) as a screen for left ventricular dysfunction in patients with diabetes. Diabetes Care
2003;26:2081–7.
[111] Melchior T, Rask-Madsen C, Torp-Pedersen C, Hildebrandt P, Kober L, Jensen G, et al.
The impact of heart failure on prognosis of diabetic and non-diabetic patients with
myocardial infarction: a 15-year follow-up study. Eur J Heart Fail 2001;3:83.
[112] Donnan PT, Boyle DI, Broomhall J, Hunter K, MacDonald TM, Newton RW, et al.
Prognosis following first acute myocardial infarction in Type 2 diabetes: a comparative
population study. Diabet Med 2002;19:448–55.
[113] Timmis AD. Diabetic heart disease: clinical considerations. Heart 2001;85:463–9.
[114] McGuire DK, Emanuelsson H, Granger CB, Magnus Ohman E, Moliterno DJ, white
HD, et al. Influence of diabetes mellitus on clinical outcomes across the spectrum of acute
coronary syndromes. Findings from the GUSTO-IIb study. GUSTO IIb Investigators.
Eur Heart J 2000;21:1750–8.
[115] Capes SE, Hunt D, Malmberg K, Gerstein H. Stress hyperglycaemia and increased risk of
death after myocardial infarction in patients with and without diabetes: a systematic
overview. Lancet 2000;355:773–8.
[116] Solomon SD, St John Sutton M, Lamas GA, Plappart P, Roulean JL, Skali H, et al.
Ventricular remodeling does not accompany the development of heart failure in diabetic
patients after myocardial infarction. Circulation 2002;106:1251–5.
[117] Harper R, Ennis CN, Heaney AP, Sheridan B, Gormley M, Atkinsen AB, et al. A
comparison of the effects of low- and conventional-dose thiazide diuretic on insulin action
in hypertensive patients with NIDDM. Diabetologia 1995;38:853–9.
[118] Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, et al. The effect of
spironolactone on morbidity and mortality in patients with severe heart failure.
Randomized Aldactone Evaluation Study Investigators. N Engl J Med 1999;341:709–17.
[119] Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, et al. Eplerenone,
a selective aldosterone blocker, in patients with left ventricular dysfunction after
myocardial infarction. N Engl J Med 2003;348:1309–21.
[120] Tang WH, Francis GS. Spironolactone in chronic heart failure:all’s well that ends well.
J Am Coll Cardiol 2003;41:215–6.
[121] McFarlane SI, Kumar A, Sowers JR. Mechanisms by which angiotensin-converting
enzyme inhibitors prevent diabetes and cardiovascular disease. Am J Cardiol 2003;91:
30H–7H.
W.H.W. Tang et al / Med Clin N Am 88 (2004) 1037–1061 1061

[122] Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving HH, et al. Effects
of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and
nephropathy. N Engl J Med 2001;345:861–9.
[123] Granger CB, McMurray JJ, Yusuf S, Held P, Michelson EL, Olottson B, et al. Effects
of candesartan in patients with chronic heart failure and reduced left-ventricular systolic
function intolerant to angiotensin-converting-enzyme inhibitors: the CHARM-Alterna-
tive trial. Lancet 2003;362:772–6.
[124] McMurray JJ, Ostergren J, Swedberg K, Granger CB, Held P, Michelson EL, et al.
Effects of candesartan in patients with chronic heart failure and reduced left-ventricular
systolic function taking angiotensin-converting-enzyme inhibitors: the CHARM-Added
trial. Lancet 2003;362:767–71.
[125] Krum H. Beta-blockers in heart failure. The Ônew wave’ of clinical trials. Drugs 1999;58:203–10.
[126] The MERIT-HF Investigators. Effect of metoprolol CR/XL in chronic heart failure:
Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure
(MERIT-HF). Lancet 1999;353:2001–7.
[127] Krum H, Ninio D, MacDonald P. Baseline predictors of tolerability to carvedilol in
patients with chronic heart failure. Heart 2000;84:615–9.
[128] The Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in
patients with heart failure. N Engl J Med 1997;336:525–33.
[129] Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and
arrhythmias. Lancet 1994;343:155–8.
[130] Marzilli M. Management of ischaemic heart disease in diabetic patients–is there a role for
cardiac metabolic agents? Curr Med Res Opin 2001;17:153–8.
[131] Malmberg K, Ryden L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, et al. Randomized
trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic
patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year.
J Am Coll Cardiol 1995;26:57–65.
[132] Fath-Ordoubadi F, Beatt KJ. Glucose-insulin-potassium therapy for treatment of acute
myocardial infarction: an overview of randomized placebo-controlled trials. Circulation
1997;96:1152–6.
[133] Stanley WC. Partial fatty acid oxidation inhibitors for stable angina. Expert Opin Investig
Drugs 2002;11:615–29.
[134] Szwed H, Hradec J, Preda I. Anti-ischaemic efficacy and tolerability of trimetazidine
administered to patients with angina pectoris: results of three studies. Coron Artery Dis
2001;12(Suppl 1):S25–8.
[135] Szwed H, Sadowski Z, Pachocki R, Domzal-Bochenska M, Szymczak K, Sydlowski Z,
et al. The antiischemic effects and tolerability of trimetazidine in coronary diabetic
patients. A substudy from TRIMPOL-1. Cardiovasc Drugs Ther 1999;13:217–22.
[136] Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, et al. Association
of glycaemia with macrovascular and microvascular complications of type 2 diabetes
(UKPDS 35): prospective observational study. BMJ 2000;321:405–12.
[137] Sulkin TV, Bosman D, Krentz AJ. Contraindications to metformin therapy in patients
with NIDDM. Diabetes Care 1997;20:925–8.
[138] Niemeyer NV, Janney LM. Thiazolidinedione-induced edema. Pharmacotherapy 2002;22:
924–9.
[139] Gorson DM. Significant weight gain with rezulin therapy. Arch Intern Med 1999;159:99.
[140] Inoue K, Sano H. Troglitazone-induced pulmonary edema. Arch Intern Med 2000;160:
871–2.
[141] Tang WH, Francis GS, Hoogwerf BJ, Young JB, et al. Fluid retention after initiation of
thiazolidinedione therapy in diabetic patients with established chronic heart failure. J Am
Coll Cardiol 2003;41:1394–8.
[142] Diamant M, Heine RJ. Thiazolidinediones in type 2 diabetes mellitus: current clinical
evidence. Drugs 2003;63:1373–405.
Med Clin N Am 88 (2004) 1063–1084

Acute hyperglycemic crisis in the elderly


Jason L. Gaglia, MDa,b, Jennifer Wyckoff, MDa,b,
Martin J. Abrahamson, MDa,b,*
a
Joslin Diabetes Center, Beth Israel Deaconess Medical Center,
1 Joslin Place, Boston, MA 02215, USA
b
Harvard Medical School, Boston, MA, USA

Acute hyperglycemic crises may be characterized as hyperglycemic


hyperosmolar syndrome (HHS), diabetic ketoacidosis (DKA), or have
elements of both. DKA is distinguished by the presence of a metabolic
acidosis caused by excess ketone production, whereas HHS is characterized
by profound hyperosmolality with serum glucose concentrations that are
usually higher than those seen in pure DKA. In a retrospective review of 613
patients, Wachtel et al [1] noted considerable overlap among these entities
with 22% of patients having DKA only, 45% having HHS only, and 33%
having features of both. This understanding is particularly important when
treating older individuals, who are more likely to present with mixed
disturbances. In the series by Wachtel et al [1], nearly a third of those
patients who presented with mixed acidosis and hyperosmolality were older
than 60 years of age.
Although HHS is noted to have a higher mortality than DKA, this may
represent the demographics of the presenting populations with mortality
rates that rise with increasing age. When mortality is stratified by age,
there is no significant difference based on category of hyperglycemic
emergency. Mortality rates for HHS range from 10% for those under age
75, to 19% in those age 75 to 84, and 35% in those over 84 years of age
[2]. Similar outcomes have been reported for DKA with mortality rates of
8% for patients 60 to 69 years of age, 27% for patients 70 to 79 years of
age, and 33% for patients greater than 79 years of age [3]. In a recent
multivariate analysis by MacIsaac et al [4], only age (and not category of
hyperglycemic emergency) was found to be a significant predictor of
mortality.

* Corresponding author. Joslin Diabetes Center, Beth Israel Deaconess Medical Center,
1 Joslin Place, Boston, MA 02215.
E-mail address: martin.abrahamson@joslin.harvard.edu (M.J. Abrahamson).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.010
1064 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

Precipitating causes
The commonest precipitating cause of hyperglycemic crises is infection,
which accounts for 20% to 55% of cases, with the most common infections
being pneumonia and urinary tract infections [5]. Omission of diabetes
therapy or inadequate treatment has been recognized as another common
cause of hyperglycemic crises, accounting for 20% to 40% of cases [5]. Up
to 40% of older patients who present with hyperglycemic crises do not have
a previous diagnosis of diabetes [6], with the strongest risk factor in the
general population for undiagnosed diabetes being increasing age [7]. Other
acute medical illnesses that may contribute as precipitating factors include
cerebral vascular accidents, myocardial infarctions, pancreatitis, trauma,
and alcohol intoxication; combined, these account for about 10% of all
cases (Box 1) [5]. Various medications including diuretics, b-blockers,

Box 1. Common hyperglycemia


Omission of diabetes therapy
Infections
Pneumonia
Urinary tract infection
Sepsis
Abscess
Vascular events
Cerebral vascular accident
Myocardial infarction
Acute pulmonary embolus
Mesenteric thrombus
Trauma, burns, subdural hematoma
Heat stroke
Gastrointestinal events
Acute pancreatitis
Acute cholecystitis
Intestinal obstruction
Iatrogenic
Peritoneal dialysis
Total parenteral nutrition
Endocrine disorders
Acromegaly
Thyrotoxicosis
Cushing’s syndrome
Pheochromocytoma
Glucagonoma
Drugs and medications
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1065

phenytoin, steroids, pentamidine, sympathomimetic agents, and antipsy-


chotics have been associated with the development of hyperglycemic crises
(Box 2) [5]. Particularly common in the geriatric population is diuretic use.
In the series by Wachtel et al [2], 19% of the patients with HHS were on
thiazide diuretics and 19% were taking furosemide [2].

Clinical presentation
The symptoms of uncontrolled diabetes often precede acute metabolic
decompensation. Symptoms may include malaise, blurred vision, polyuria,
polydipsia, and weight loss. Typically, DKA evolves rapidly over a period of
hours, whereas HHS tends to evolve over days leading to the more severe
hyperosmolality. Dehydration may be worsened by concomitant use of
diuretics [2]. Typical signs of dehydration include dry mucous membranes,
decreased skin turgor, hypotension, and tachycardia. Guidelines have been
established for the determination of volume status based on physical
examination and orthostasis (Table 1). In elderly patients, however, it may
be difficult to judge skin turgor, and patients with long-standing neuropathy
may have impaired response to intravascular volume depletion. A fruity odor
on the breath indicates the presence of acetone from ketogenesis. Kussmaul’s

Box 2. Drugs and medications reported to precipitate


hyperglycemic crises
Steroids
Sympathomimetic agents (dobutamine, terbutaline)
Diuretics (thiazides, ethacrynic acid)
b-Blockers (propranolol)
Calcium-channel blockers
H2 blockers (cimetidine)
Phenytoin
Antipsychotics (chlorpromazine, loxapine, clozapine,
olanzapine, risperidone)
Interferon-a
Ribavirin
Protease inhibitors
Pentamidine
L-asparagine
FK506
Alcohol
Cocaine
Ecstasy
1066 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

Table 1
Clinical assessment of dehydration
Clinical finding % dehydration Estimated volume
Decreased tissue turgor 5 1L
Orthostatic change in
pulse alone 10 2L
Orthostatic change in
pulse and blood pressure
([15/10 mm Hg) 15–20 3–4 L
Supine hypotension or sepsis >20 >4 L

respirations may occur to compensate for metabolic acidosis. Abdominal


pain, nausea, and vomiting are more common in those with DKA than HHS
[8]. This may be related to the increased production of prostaglandins I2 and
E2 by adipose tissue in the absence of insulin. Similar symptoms occur when
prostaglandin I2 is infused into normal subjects [9]. Hypertonicity and
hyperglycemia may also worsen gastroparesis. Lethargy, obtundation, or
coma may be present. In general, the level of consciousness correlates with
osmolality rather than acidemia [10]. Mental status changes and coma are
more frequent in the hyperosmolar state. In some patients with HHS, focal
neurologic signs or seizures may be the dominant clinical finding. Elderly
patients seem to be at higher risk for developing hyperosmolality, because
there is a significant positive correlation with age (P \ .0001) [4]. Impaired
thirst sensation and decreased ability to obtain fluids in the elderly may
contribute to their risk for dehydration.
Although the most common precipitating event is infection, many
patients are normothermic or hypothermic at presentation. This finding is
thought to be caused in part by the skin vasodilation that accompanies the
metabolic acidosis and also by low fuel-substrate availability in the setting
of relative insulinopenia [11,12]. In one case series, hyperglycemia repre-
sented 11.8% of all admissions for severe accidental hypothermia and was
a more common cause than hypothyroidism (8%) [12]. All patients should
be evaluated for infectious precipitants, even in the absence of fever.

Pathophysiology
In all hyperglycemic crises, insulin deficiency, absolute or relative, against
a background of increased insulin resistance is the principle underlying
defect. Insulin levels are not adequate to maintain normal serum glucose
and suppress ketogenesis. Hyperglycemia itself can further compromise
insulin secretory capacity and increase insulin resistance causing a vicious
cycle of worsening hyperglycemia and decreasing insulin production [13]. In
elderly individuals with decreased baseline secretory reserve, this effect may
be quite pronounced.
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1067

The severity of insulin resistance depends on a number of factors,


including the underlying pathophysiology of the diabetes, carbohydrate
intake of the patient, levels of counterregulatory hormones present, and
stress of the precipitating event. Cortisol, glucagon, growth hormone, and
epinephrine all increase insulin resistance [14–18]. These counterregulatory
hormones increase the gluconeogenesis and glycogenolysis and decrease use
of glucose by down-regulating glycolysis and glycogen synthesis and in the
setting of limited insulin secretory capacity exacerbate hyperglycemia. If the
insulin deficiency is severe enough, increased ketogenesis may lead to
ketoacidosis.
The acidosis resulting from excessive ketogenesis distinguishes DKA
from HHS. Insulin normally suppresses ketogenesis. In a state of relative
insulin deficiency, ketogenesis occurs when free fatty acids are metabolized
into ketones [19]. Acetoacetate, b-hydroxybutyrate, and acetone make up
the three ketone bodies produced. During normal metabolism, the ratio of
b-hydroxybutyrate to acetoacetate is 1:1, with acetone being present in
relatively small quantities. Ketones trigger insulin release from the pancreas.
The insulin, in turn, down-regulates the overproduction of ketones. In
insulin deficiency, the levels of all three ketone bodies can increase
dramatically, and the ratio of b-hydroxybutyrate to acetoacetate changes
to as much as 10:1. b-hydroxybutyrate and acetoacetate are the keto acids
responsible for the acidosis of DKA. Acetone does not cause acidosis and is
harmlessly excreted in the lungs, causing the ‘‘fruity odor.’’
Hyperglycemia causes an osmotic diuresis and dehydration. Although the
osmotic contribution of glucose is only 5 mOsm/L for every 90 mg/dL (as
opposed to the 2 mOsm/L for each 1 mEq/L increase in serum sodium) even
with moderate elevation in glucose a mixed hypertonicity may occur. Serum
hyperosmolality develops as relatively hypo-osmolar urine is excreted.
Ketones lost in the urine are osmotically active; to maintain electrical
neutrality cations, such as sodium, potassium, calcium, and magnesium, are
excreted with the ketones. This high solute excretion further impairs water
reabsorption.
When hypovolemia becomes severe, the glomerular filtration falls leading
to decreased renal losses of glucose and worsening hyperglycemia and
hyperosmolality. In elderly patients with impaired glomerular filtration rates
there is a decrease in glucosuria, which may lead to more severe
hyperglycemia [20]. Insulin deficiency itself promotes further loss of water
because insulin may enhance reabsorption of water and sodium from the
renal tubules (Fig. 1) [21].

Laboratory findings
Initial evaluation of elderly patients presenting with hyperglycemic crisis
should include plasma glucose, electrolytes, bicarbonate, blood urea
nitrogen (BUN), creatinine, creatinine kinase, phosphate, urine and serum
1068 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

Insulin Deficiency and Increased Counter-Regulatory Hormones

Adipose Tissue Liver Muscle

Increased Lipolysis Increased glucose production Increased Proteolysis


and decreased glucose
uptake
Gluconeogenesis Gluconeogenesis

Increased Free Fatty Acids Hyperglycemia Increased Amino


Acids

Increased Ketogenesis Glycosuria

Ketonemia and Ketonuria Osmotic Diuresis

Ketoacidosis with Decreased Loss of Electrolytes and


Alkali Reserve Dehydration
Decreased GFR
Decreased urinary (impaired renal function)
clearance of
ketones Hyperosmolality Lactic Acid Production
Decreased Tissue
Perfusion

Pure Diabetic Ketoacidosis Pure Hyperosmolar State

Fig. 1. Insulin deficiency, relative or absolute, leads to increased hepatic glucose production
and decreased peripheral use of glucose, which results in hyperglycemia, osmotic diuresis, and
dehydration. In severe insulin deficiency, ketone body production is augmented leading to
ketonemia and eventual ketoacidosis. GFR, glomerular filtration rate.

ketones, calculation of the anion gap, arterial blood gas, complete blood
count with differential, urinalysis, pregnancy test in all women of re-
productive age, and electrocardiogram. Measurement of hemoglobin A1C
may provide useful information about the underlying degree of metabolic
control. If clinically indicated, components of an infectious work-up
including culture of blood, urine, and throat and a chest radiograph and
possible abdominal imaging should be performed. If pneumonia is suspected
without evidence of infiltrate on the admission radiograph, chest radiogra-
phy should be repeated after adequate hydration. If myocardial infarction is
suspected, cardiac enzymes should be evaluated.
Leukocytosis with elevated granulocytes is commonly seen and white cell
counts of 12 to 20 K are not unusual. Mild leukocytosis by itself does not
imply infection, but instead is a response to stress and dehydration. White
cell counts above 30 K, however, are highly suggestive of infection.
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1069

Similarly, hemoglobin and hematocrit are also frequently elevated by


dehydration with a falsely elevated mean corpuscular volume secondary
to the relative hypotonicity of cell-counter dilution fluid (339 mOsm/L).
The serum creatinine at the time of initial presentation is frequently
elevated. This should be followed closely with treatment because it may
represent intrinsic renal disease, prerenal azotemia in the setting of
hypovolemia, or false elevation associated with elevated serum ketones.
Both acetoacetate and bilirubin interfere with the measurement of creatinine
using the alkaline picrate (Jaffe) assay [22]. Treatment leads to resolution of
ketosis and restoration of intravascular volume with a concomitant decrease
in creatinine except in the case of intrinsic renal disease. Patients with renal
disease require extra caution because if they are oliguric or anuric, they may
become volume overloaded during treatment and are more prone to life-
threatening electrolyte abnormalities. This is especially true in the geriatric
population, which has a much higher prevalence of underlying renal disease.
In one series by Malone et al [3], more than 50% of patients over the age of
65 who presented with DKA had underlying renal disease (more than twice
the rate in those less than 65).
A variety of serum enzymes may be elevated in hyperglycemic crises
including hepatic transaminases, lactate dehydrogenase, creatinine kinase,
amylase, and lipase. Hypercholesterolemia and hypertriglyceridemia are
often present and some of these enzyme elevations may represent tri-
glyceride effects [23]. The clinical significance of these findings is discussed
elsewhere in this article.

Treatment of hyperglycemic crises


The therapeutic goals for the treatment of hyperglycemic crisis are based
on correcting the underlying pathophysiologic abnormalities. These include
volume resuscitation, improving circulatory volume and tissue perfusion,
correcting plasma osmolality and electrolyte imbalances, and inhibiting
ketogenesis. These steps usually lead to correction of acid base disturbances
and hyperglycemia. The precipitating events must be identified and treated,
and appropriate steps taken to prevent recurrence.

Fluids
Fluid replacement is one of the mainstays of treatment. Fluid re-
placement lowers blood sugars independent of insulin; reduces the levels of
counter-regulatory hormones (which improves insulin sensitivity); and
restores intravascular volume. Although not recommended, it is possible
to treat HHS without the administration of insulin, but with fluids alone
[20,24,25]. This approach is discouraged, because there have been case
reports of patients becoming acidemic if insulin is withheld [20]. Fluid
1070 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

replacement by itself greatly decreases serum glucose concentrations


through dilution (24%–36% decline) and improved renal perfusion, which
increases glucosuria (29%–76% decline in glucose) [20]. The preferred
initial fluid is normal saline (0.9% NaCl), which is hypotonic relative to
the patient’s serum osmolality and remains within the intravascular
compartment. The initial rate of infusion depends on the underlying car-
diac status and the degree of volume depletion; normally 15 to 20 mL/kg/h
(1–1.5 L in an average adult) is given in the first hour to expand rapidly
the intravascular space. Calculations of total water deficits based on
serum sodium concentration must be interpreted with caution because
hyperglycemia and hyperlipidemia may cause pseudohyponatremia in the
setting of osmotic flux of water from the intracellular to the extracellular
space (Box 3). It is best to use corrected serum sodium and physical
examination to determine degree of dehydration (see Table 1), with the
caveat that patients with neuropathy may have blunted responses. If the
diuresis has been going on for several days, the dehydration may be so
severe that the serum sodium is elevated with the corrected sodium being
even higher. Once the corrected sodium has normalized, most clinicians
change to treatment with half normal saline. The rate of fluid replacement
should also take into account ongoing urinary losses with a goal of
correcting fluid deficits in 24 hours and serum osmolality corrected at a rate
of approximately 3 mOsm/kg/h (Table 2). Once serum glucose has dropped
to less than 250 mg/dL, 5% dextrose is frequently added to the fluids and
the insulin infusion rate adjusted to maintain blood glucose in the 120 to
180 mg/dL range.

Box 3. Commonly used calculations in the evaluation of severe


hyperglycemia
Calculation of the anion gap
Gap = [Naþ]  [Cl þ HCO3]

Calculation of serum osmolality


Osmolality = 2[Naþ] þ (glucose in mg/dL)/18 þ (BUN in mg/dL)/2.8 þ EtOH/4.6

Calculation of effective serum osmolality


Effective osmolality = 2[Naþ] þ (glucose in mg/dL)/18

Corrected serum sodium


Corrected Na = [Naþ ] þ 1.6(glucose in mg/dL  100)/100 þ .002(triglycerides in mg/dL)

Total body water deficit


Total body water deficit (L) = 0.6 * (bodyweight in kg) * (1  140/[Naþ])

Urea is freely permeable and does not contribute to effective


osmolality. Some calculations of osmolality include
potassium in addition to sodium.
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1071

Table 2
Suggested replacement fluids
Time Volume
1st h 1–2 L normal saline
2nd h 1 L normal saline
3rd h 500 mL to 1 L normal saline
4th h 500 mL to 1 L normal or half normal saline
5th h 500 mL to 1 L normal or half normal saline
Total first 5 h 3.5–5 L
6th–12th h 250–500 mL/h half normal saline
Administer normal saline or colloid as indicated to maintain hemodynamic stability. Once
corrected sodium approaches normal consider changing to half normal saline (usually about
4 h). Change to D5 half normal saline when blood glucose reaches 250 mg/dL. Patients with
a prolonged antecedent period may require additional fluid replacement and caution must be
used with patients who are elderly, with a history of congestive heart failure, or renal failure.

Insulin therapy
Before the introduction of insulin in 1922, diabetic coma was almost
uniformly fatal. The primary goals of insulin therapy are to reverse
ketogenesis (if present); suppress lipolysis; and inhibit hepatic gluconeogen-
esis. Insulin also decreases the effective plasma osmolality by increasing
cellular permeability to glucose. There are several cautions, however, to the
use of insulin therapy in hyperglycemic crises. Insulin therapy should not be
initiated in patients with hypotension and severe hyperglycemia until
volume resuscitation has occurred. Insulin enhances intracellular transport
of glucose with a resultant fluid shift from the extracellular to the
intracellular compartment. As much as 2 to 3 L may shift out of the
intravascular compartment [26], increasing the risk of hypovolemic shock
and thromboembolism in the setting of inadequate volume resuscitation.
Because insulin also mediates the re-entry of potassium into the intracellular
compartment, insulin administration can lead to life-threatening hypokale-
mia, and should also be withheld in patients who are hypokalemic at
presentation (potassium less than 3.3) until potassium replacement therapy
is given (potassium greater than 3.5) [27]. Insulin therapy in these situations
may result in arrhythmias, impaired cardiovascular function, or muscle
weakness and respiratory failure. Ultimately, it is the adequacy of fluid and
electrolyte repletion that determines the effectiveness of insulin therapy.
During hyperglycemic crises there is usually profound insulin resistance
requiring supraphysiologic doses of insulin to prevent ketogenesis, gluco-
neogenesis, and lipolysis [28,29] and resolution of hyperglycemia and
hyperosmolality [30]. Intravenous administration is the preferred initial
route of administration because it leads to a more rapid initial fall in glucose
and ketones when compared with subcutaneous or intramuscular adminis-
tration [31]. After the first 2 hours of treatment, the rate of decline of
ketonemia and hyperglycemia is comparable between all three routes of
1072 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

administration [31]. Intravenous administration offers the advantages,


however, in that it can be titrated rapidly to avoid hypoglycemia and
is delivered more reliably in the setting of poor tissue perfusion or
dehydration.
Before initiation of intravenous insulin infusion, it is suggested that about
50 mL of the insulin-saline solution be allowed to run through the tubing to
minimize subsequent insulin binding to the tubing during therapy. An initial
loading dose of 0.15 U/kg (usually 10 U in adults) of regular insulin is
followed by a continuous infusion at a rate of 0.1 U/kg/h (usually 5–10 U/h
in adults) [32]. This results in supraphysiologic serum insulin concentrations,
which are adequate to suppress ketogenesis, gluconeogenesis, and lipolysis.
As sensitivity to insulin decreases with increasing age, older patients tend to
require more insulin during the acute management of hyperglycemia [3]. The
rate of intravenous insulin infusion must be reassessed and adjusted
frequently based on the rate of change of serum glucose concentrations.
Insulin should not be stopped (unless hypoglycemia occurs, in which case it
is stopped temporarily and restarted when patients are euglycemic again);
rather, dextrose should be added to the intravenous fluids to maintain
a blood glucose level between 120 and 180 mg/dL (Box 4).
In centers where it is difficult to administer continuous intravenous
insulin, other routes of administration may be used. Even in this setting, it is
still preferable that the initial dose be given intravenously (0.15 U/kg)
followed by subcutaneous or intramuscular insulin at 0.1 U/kg/h, titrated to
decrease glucose by approximately 100 mg/dL/h. When regular insulin is
used intramuscular administration leads to faster peak insulin levels and less
risk of hypoglycemia when compared with subcutaneous delivery. With the
newer rapid-acting insulin analogues subcutaneous administration also
leads to a rapid peak but without the risk of myonecrosis inherent to
intramuscular administration. The efficacy of subcutaneous rapid-acting
insulin has been found to be comparable with regular intravenous insulin in
this setting [33]. The authors suggest administration of rapid-acting insulin,
such as lispro or aspart, subcutaneously every 2 to 4 hours if intravenous
administration is not available. As previously noted, subcutaneous admin-
istration should be avoided in patients with evidence of hypotension or
severe hypovolemia because the absorption is less reliable. Repeated
frequent administration of insulin should be avoided especially in the
setting of impaired renal function because administration at shorter
intervals may lead to a cumulative effect with subsequent hypoglycemia.
In patients with normal renal function, the biologic effects of the rapid-
acting insulin analogues last 3 to 4 hours.
Once the anion gap has closed, the bicarbonate is above 18, and the
patient stable and able to eat, the transition to a standard subcutaneous
insulin regimen should be considered. A frequent mistake is discontinuation
of the insulin infusion prematurely, before conversion to subcutaneous
insulin, leading to rebound acidosis. Because regular insulin administered
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1073

Box 4. Insulin management guidelines


Regular insulin, 10 U intravenously STAT or 0.1 U/kg/h
Start regular insulin infusion at 5 U/h or 0.1 U/kg/h
Increase by 1 U/h every 1 to 2 hours if less than 10% drop in
glucose or no improvement in acid-base status
Decrease insulin by 1 to 2 U/h when glucose less than 250 mg/dL
or progressive improvement
Do not decrease insulin infusion to less than 1 U/h
Add D5 to IV fluids to maintain glucose between 120 and
180 mg/dL
Maintain plasma glucose between 120 and 180 mg/dL
If plasma glucose less than 60 mg/dL, stop insulin infusion for
30 minutes to 1 hour and then restart infusion
If plasma glucose dropping consistently less than 100 mg/dL,
change to intravenous fluids to D10 to maintain glucose 120 to
180 mg/dL, while on insulin infusion
Once patient can eat and anion gap is resolving, consider change
to subcutaneous insulin
Overlap short-acting insulin with continuous insulin infusion by
at least 30 minutes
Consider initiation of long-acting insulin at same time, usually
easiest at breakfast or dinner
For patients previously managed on insulin: return to prior dose
or re-evaluate insulin regimen before returning to prior dose
For new insulin patients, consider a short- and long-acting insulin
regimen
Avoid transition to oral therapy until acute glucotoxicity has
resolved and renal function returned to normal

Adapted from Abrahamson M, Beaser R, Ramachandiran C, Grossman S,


Laffel L, Rosenzweig J. Guidelines for management of hyperglycemic
emergencies. Boston: Beth Israel Deaconess Medical Center and Joslin
Diabetes Center; 2002; with permission.

intravenously has a half-life of only 4 to 5 minutes and subcutaneous regular


insulin requires 15 to 60 minutes for onset of action, short- (regular) or
rapid-acting (lispro or aspart) subcutaneous insulin administration should
be overlapped by at least 30 minutes with the intravenous infusion. Short- or
rapid-acting insulin should not be withheld even if the patient is euglycemic
at the time of transition to a subcutaneous regimen. In addition,
administration of intermediate- (neutral protamine Hagedorn [NPH] or
lente) or long-acting (ultralente or glargine) insulin as part of the initial
insulin regimen is also recommended to help minimize the effects of any
1074 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

accidental insulin omission later in the day. For patients in whom total
insulin requirements are not known, an initial dose of 0.5 to 1 U/kg/d total
is usually given in a mixed regimen including both short- or rapid-acting and
intermediate- or long-acting insulins.

Bicarbonate
Although the use of alkali therapy may seem an attractive method to
correct the metabolic acidosis seen in DKA, it is usually not necessary
because the metabolic acidosis tends to correct with insulin administration
as protons are consumed and bicarbonate is regenerated during ketoanion
metabolism. The administration of bicarbonate may actually lead to
rebound alkalosis, worsened hypokalemia, paradoxical central nervous
system acidosis, increase in intracellular acidosis, and prolongation of
ketosis. Joslin [34] was one of the first to recognize the pitfalls of bicarbonate
therapy writing in 1917 that, ‘‘The dangers attendant upon the use of alkalis
in the treatment of acid intoxication far outweigh their advantages.’’ Joslin’s
avoidance of bicarbonate therapy may have contributed to the decreased
mortality seen in his treatment protocol compared with other major clinical
centers. When studied in a randomized prospective manner, there is no
benefit to bicarbonate therapy in patients with a pH between 6.9 and 7.14
[35,36]. The groups reported with pH less than 7 have been very small,
however, and there have been no prospective randomized studies evaluating
the use of bicarbonate when the arterial pH is less than 6.9. Severe acidemia
has been associated with vascular refractoriness to adrenergic action, central
nervous system depression, and impaired myocardial contractility. Bicar-
bonate therapy is reserved for those patients with clinical manifestations of
acidemia or a pH less than 7. Usually 1 to 2 ampules of sodium bicarbonate
(50–100 mEq) is added to a liter of half normal saline to make a nearly
isotonic solution, which is administered over a period of 1 to 2 hours. If
potassium is also low, 10 to 20 mEq of potassium chloride may be added.
The venous pH should be rechecked 30 minutes after administration and
treatment repeated if the corrected pH is below 7.

Potassium
In 1946 Holler [37] first recognized the importance of potassium reple-
tion in the treatment of hyperglycemic crises. Before his publication, it was
not uncommon for patients to die after apparent treatment and resolution
of diabetic coma. One such description taken from a 1944 report is as
follows: ‘‘...after considerable hours of therapy and improvement, they
suddenly fail rapidly and die. Analyses show that the serum sodium,
chloride, carbon dioxide and the blood sugar were satisfactory. The urine
sugar and ketones had cleared up’’ [38]. Potassium is important for muscle
function, and hypokalemia has been associated with arrhythmias, cardiac
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1075

arrest, cardiovascular collapse, muscle weakness, and respiratory failure.


Potassium is predominantly an intracellular cation; however, increased
extracellular hyperosmolality secondary to hyperglycemia causes a shift of
water and potassium from the intracellular to the extracellular compart-
ment. This potassium shift is enhanced by insulin deficiency, acidosis, and
accelerated breakdown of intracellular proteins. Osmotic diuresis, second-
ary hyperaldosteronism, and urinary ketoanion excretion (in the form of
potassium salts) leads to increased urinary potassium losses. With the
administration of fluids and insulin there is typically a rapid decline in
plasma potassium concentration as potassium shifts back into the cells. To
prevent hypokalemia, potassium supplementation is begun if the serum
potassium concentration is less than 5.5 mEq/L. Potassium should initially
be withheld in patients with anuria, acute renal failure, or potassium above
5.5 mEq/L. Usually 20 to 30 mEq/h of potassium is needed to maintain
plasma potassium levels between 4 and 5 (see Table 3 for guidelines).
Although some authors use potassium chloride alone, others recommend
administration of one third of the potassium replacement as potassium
phosphate to avoid excessive chloride administration and to prevent severe
hypophosphatemia [5].

Phosphate
Total body phosphate depletion is common in DKA and HHS. Like
potassium, phosphate shifts from the intracellular to the extracellular
compartment in hyperglycemic crises and serum levels of phosphate at
presentation are typically normal or elevated. Osmotic diuresis, however,
leads to enhanced urinary phosphate losses and total body depletion.
During insulin therapy phosphate reenters the intracellular compartment
with resultant hypophosphatemia. Because of the prolonged duration of
symptoms in HHS and frequent comorbid conditions, phosphate levels
are often lower than those seen in DKA. Potential complications of
severe hypophosphatemia include decreased cardiac output, respiratory
muscle weakness, rhabdomyolysis, central nervous system depression, sei-
zures and coma, acute renal failure, and hemolytic anemia [39]. In theory,
phosphate depletion may also contribute to decreased concentrations of

Table 3
Guidelines for potassium replacement
Serum K (mEq/L) Additional K in IVFs
\3.5 40 mEq/L
3.5–4.5 20 mEq/L
4.5–5.5 10 mEq/L
>5.5 Stop K infusion
Do not administer K if K is [5.5 or patient is anuric.
Abbreviation: IVFs, intravenous fluids.
1076 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

2,3-diphosphoglycerate leading to a shift of the oxygen dissociation curve to


the left and decreasing tissue oxygen delivery. In practice, however, no effect
has been seen on oxygen availability with higher 2,3-diphosphoglycerate
levels after phosphate repletion compared with controls with DKA [40].
Because intravenous phosphate replacement may lead to symptomatic
hypocalcemia and hypomagnesemia [41,42] and has not been proved
beneficial in randomized controlled trials [40,43,44] the degree of phosphate
replacement and type of phosphate therapy remain controversial. The
authors’ approach is to reserve phosphate replacement for those with severe
hypophosphatemia of 1.5 mg/dL or less, normal serum calcium concentra-
tion, and normal renal function. The use of small amounts of potassium
phosphate intravenously seems to be safe in this setting (30–60 mM
potassium phosphate over 12–24 hours). Serum calcium, magnesium, and
phosphate levels should all be monitored during phosphate infusion. Oral
phosphate replacement is preferable to intravenous and should be com-
menced as soon as possible.

Complications
Pancreatitis
Nonspecific elevations of amylase and lipase may occur in both DKA
and HHS. In one recent series, nonspecific elevations in amylase and lipase
occurred in 16% to 25% of cases of DKA. Serum amylase and lipase may be
elevated to more than three times normal without clinical or CT evidence of
actual pancreatitis [45]. The cause of nonspecific hyperamylasemia in
hyperglycemic crises is thought to be multifactorial including salivary
hyperamylasemia, reduced renal clearance of amylase, and increased
leakage from the pancreas. Much less is known about the causes of
hyperlipasemia but it is postulated that the reduced glomerular filtration
seen in hyperglycemic crises may result in decreased renal clearance and that
nonpancreatic lipases may be measured in the current assay generating false-
positive results [45].
Coexisting acute pancreatitis, however, may occur in 10% to 15% of
cases of DKA [46]. Acute pancreatitis in hyperglycemic crises is often
associated with transient severe hypertriglyceridemia. Insulin deficiency
promotes lipolysis, which releases free fatty acids. With the inhibition of
lipoprotein lipase in peripheral tissues and increased delivery of free fatty
acids to the liver, severe hypertriglyceridemia may ensue. In a recent series
by Nair et al [46] of 100 consecutive patients admitted with DKA, 22 had
mild hypertriglyceridemia (greater than 500 mg/dL) and 8 had severe
hypertriglyceridemia (greater than 1000 mg/dL). Half of the patients with
severe hypertriglyceridemia were diagnosed with pancreatitis based on CT
findings. In these patients, the hypertriglyceridemia was transient with
triglyceride levels decreasing to less than 300 mg/dL with resolution of
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1077

ketosis. Acute pancreatitis was seen more often in patients with more severe
metabolic acidosis (pH 7.15 versus 7.31, P = .0001) and higher serum glucose
(934 mg/dL versus 714 mg/dL, P = .02) at the time of presentation [46].

Rhabdomyolysis
Rhabdomyolysis is a clinical and biochemical syndrome resulting from
the destruction of myocytes and leakage of their cellular contents into the
plasma. It is a frequently unrecognized feature of hyperglycemic crises with
an incidence of 17% in one large series (defined as creatinine kinase levels
greater than 1000 IU/L) [47]. Patients who presented with rhabdomyolysis
had significantly higher blood glucose levels and serum osmolality on
admission. Patients with rhabdomyolysis also had evidence of decreased
renal function with significantly higher concentration of BUN, creatinine,
and b2-microglobulin [47,48]. In a series of 265 patients with hyperglycemic
crises, 44 of which had evidence of rhabdomyolysis, the mortality within 1
week of presentation was significantly higher in those with rhabdomyolysis
(38.5% for DKA, 35.5% for HHS) than those without (9.7% for DKA,
26.7% for HHS, P \ .05) [47]. In this series by Wang et al [47] there was not
a significant age-related difference in the incidence of rhabdomyolysis. The
mean age was higher in those with ketoacidosis and rhabdomyolysis who
died, however, compared with those who recovered (mean age 62.6  6
versus 40  8 years old) [47]. Given the positive correlations between age,
osmolality, and increased mortality [4] it is particularly important to identify
and treat those patients in the geriatric population who present with
rhabdomyolysis as a component of their hyperglycemic emergency.

Thromboembolism
Although there are no data demonstrating the safety or efficacy of low-
dose or low-molecular-weight heparin in hyperglycemic crises, thrombopro-
phylaxis should be considered for all patients without a contraindication to
low-level anticoagulation. Diabetes itself is associated with various pro-
thrombotic abnormalities including vascular endothelial dysfunction, eleva-
tion of coagulation activation markers and clotting factors, increase in
plasminogen activator inhibitor type 1, decrease in the anticoagulant protein
C, and platelet hyperactivity (reviewed in [49]). In hyperglycemic crises, such
as HSS or DKA, the combination of severe dehydration, increased blood
viscosity, low cardiac output, and stimulation of prothrombotic mediators
by hyperglycemia can lead to a hypercoagulable state and thromboembo-
lism. A number of case reports and retrospective reviews from the 1970s
suggest an increased incidence of vascular occlusive disease including
cerebral infarct, myocardial infarction, pulmonary embolism, mesenteric
thrombosis, and disseminated intravascular coagulation in HHS and DKA
[50–52]. With the introduction of prophylactic anticoagulation (twice daily
low-dose subcutaneous heparin) and aggressive early hydration the incidence
1078 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

of these complications has fallen dramatically in recent years [1,53]. In the


United Kingdom (where full intravenous heparin anticoagulation has been
frequently used in the treatment of HSS) [54] there is now a shift toward
prophylactic subcutaneous heparin dosing, because the risk of bleeding from
full-dose anticoagulation is believed to be greater than the risk of vascular
thrombosis with current treatment protocols [55].

Hyperchloremic metabolic acidosis


Hyperchloremic metabolic acidosis is only present in approximately 10%
of patients admitted with DKA at presentation; however, hyperchloremia
develops in most patients by 4 to 8 hours after initiation of therapy [56]. This
occurs because chloride losses are less than sodium losses during treatment
of DKA with excretion of ketoanions as sodium salts, and increased
chloride reabsorption caused by decreased availability of bicarbonate in the
proximal tubule [56,57]. Because initial replacement fluids usually contain
154 mmol/L of sodium and chloride (approximately 54 mmol/L in excess of
the 100 mmol/L of chloride in the serum) relative hyperchloremia occurs
during treatment. This is usually of no clinical consequence and gradually
corrects over the next 24 to 48 hours with enhanced renal acid excretion.
Normalization of the anion gap with persistent mild reduction of bi-
carbonate is not an unexpected finding during initial treatment.

Cerebral edema
Asymptomatic cerebral edema is very common in hyperglycemic emer-
gencies and is usually present before treatment is begun [58]. Clinically
apparent cerebral edema is extremely rare in adults but occurs in
approximately 1% of children with DKA and is associated with a mortality
rate of approximately 20% [59]. It had been previously hypothesized that
cerebral edema is caused by the accumulation of idiogenic osmolytes in
brain cells exposed to hyperosmolar conditions and that the rapid decrease
in extracellular osmolality during treatment resulted in brain swelling. More
recent studies do not indicate an association between cerebral edema and the
change in serum glucose concentration during therapy or the rate of fluid or
sodium administration. Instead, cerebral edema may be related to brain
ischemia. Cerebral edema has been associated with lower partial pressures of
arterial carbon dioxide, higher BUN concentrations, slow rise in serum
sodium with treatment, and treatment with bicarbonate. Hypocapnia
(which causes cerebral vasoconstriction), dehydration (which decreases
brain perfusion), and bicarbonate therapy (which can cause central nervous
hypoxia) are all expected to increase brain ischemia and lead to worsening
vasogenic edema several hours after initiation of therapy and reperfusion.
This may also help explain why cerebral edema is more common in children
than adults, because children’s brains have higher oxygen requirements and
are more susceptible to ischemia [59]. A slow rise in serum sodium with
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1079

headache and rapid deterioration in level of consciousness should arouse


clinical suspicion for cerebral edema. In the geriatric population, other
potential causes of changes in level of consciousness should also be
evaluated. If the diagnosis is confirmed, treatment with hyperosmolar
therapy, such as mannitol, should be started immediately. Recommended
starting doses are 0.5 to 2 g/kg body weight, repeated as needed [60].

Acute gastric dilation


Acute gastric dilation caused by hypertonicity-induced gastroparesis is an
uncommon but potentially fatal complication of hyperglycemic emergen-
cies. There is an increased risk of gastrointestinal bleeding in this setting
[61]. Decompression with a nasogastric tube and possible use of pro-
phylactic gastric acid–reducing agents, such as proton pump inhibitors, may
be required. As the hypertonicity is corrected, this condition should remit.

Adult respiratory distress syndrome


Although rare, pulmonary edema or adult respiratory distress syndrome
may occur during treatment of hyperglycemic crises. By increasing left atrial
pressure and decreasing osmotic pressure, relative hypotonic fluid admin-
istration may lead to capillary leak and pulmonary edema. The development
of an increased A-a gradient, dyspnea, hypoxemia, rales, or infiltrates
during resuscitation should lead the clinician to suspect the development of
adult respiratory distress syndrome [62]. PaO2 and A-a gradient should be
monitored closely in these patients. Because crystalloid infusion may be
a major factor in the development of adult respiratory distress syndrome,
some authors favor the addition of colloid administration for treatment of
hypotension unresponsive to crystalloid replacement to lower fluid intake in
these patients [5].

Monitoring the patient


In the setting of severe dehydration central access and invasive
monitoring should be considered during resuscitation. ICU admission
may be necessary if there is hemodynamic instability, decreased level of
consciousness, inability to protect the airway, presence of abdominal
distention or succussion splash, or if insulin infusion or frequent monitoring
cannot be performed in a general ward. With institution of insulin therapy
the patient should be monitored closely for hypotension, hypoglycemia, and
hypokalemia. Glucose should be checked hourly and electrolytes and acid-
base status checked every 2 to 4 hours as indicated (Fig. 2). Venous pH may
be used if there is no need to measure arterial PO2 or PCO2. The venous pH is
usually 0.03 units less than the arterial pH. Although useful at time of
presentation, following plasma or urine ketones frequently as measured by
1080 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

Time
Vitals/Examination
Pulse
Blood Pressure
Respirations
Temperature
Mental Status
Laboratories
Na
K
Cl
HCO3
Cr
Ca
PO4
Mg
Anion Gap
Calculated Effective Osms
Fluids
Normal Saline
1/2 normal saline

D5, 1/2 normal saline


Repletion
K
PO4
Diabetes
Glucose
Insulin (U/hr)
Output
Urine (ml/hr)

Fig. 2. Sample flow sheet to aid in the management of hyperglycemic crises.

the nitroprusside reaction is not helpful in the management of DKA. This is


because the nitroprusside reaction does not measure the principal ketone
body, b-hydroxybutyrate. During correction of DKA, b-hydroxybutyrate is
oxidized to acetoacetate, which is measured. Urinary ketones during the
recovery phase of DKA can actually increase because of the increase in
acetoacetate. The bedside measurement of serum b-hydroxybutyrate, which
has recently become available, may prove useful in this setting given the
limitation of urine measurements [19].

Prevention
Unfortunately, one third of diabetes cases (usually type 2) are un-
diagnosed [63]. Members of ethnic minorities including African Americans,
Hispanic Americans, Native Americans, Asian Americans, and Pacific
Islanders are particularly at risk [64]. Twenty percent to 40% of patients
who present with hyperglycemic crises do not have a prior history of insulin
use [3] or diabetes [2,6] and the early signs and symptoms of hyperglycemia
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1081

are often unnoticed. The strongest risk factor for undiagnosed diabetes is
increasing age with the odds increasing by 1.05 for each year increase in age
[7] and almost one third of those admitted to hospital with hyperosmolarity
come from nursing homes [2]. Wide implementation of the screening
guidelines proposed by the Expert Committee on the Diagnosis and
Classification of Diabetes Mellitus [64] would essentially eliminate these
undiagnosed cases of diabetes [7].
The geriatric population is at particular risk for developing hyperglyce-
mic crises with the development of diabetes. With increasing age, insulin
secretory reserve, insulin sensitivity, and thirst mechanisms decrease. The
elderly are particularly vulnerable to hyperglycemia and dehydration, the
key components of hyperglycemic emergencies. As mortality increases with
increasing age, it is particularly important to identify patients at risk within
the geriatric population. If recognized early, hyperglycemia can frequently
be treated in the outpatient setting even with moderate or large ketonuria,
provided patients can take fluids, monitor blood glucose frequently, and
follow standard ‘‘sick day rules’’ [65]. In the Memphis Chronic Disease
Program, hospital days for diabetes-related admissions were reduced by
over 60% after institution of an outpatient monitoring program [66,67].
With increased diabetes surveillance and aggressive early treatment of
hyperglycemia and its complications, morbidity and mortality from acute
diabetic crises in the geriatric population can be greatly reduced.

References
[1] Wachtel TJ, Tetu-Mouradjian LM, Goldman DL, Ellis SE, O’Sullivan PS. Hyper-
osmolarity and acidosis in diabetes mellitus: a three-year experience in Rhode Island.
J Gen Intern Med 1991;6:495–502.
[2] Wachtel TJ, Silliman RA, Lamberton P. Prognostic factors in the diabetic hyperosmolar
state. J Am Geriatr Soc 1987;35:737–41.
[3] Malone ML, Gennis V, Goodwin JS. Characteristics of diabetic ketoacidosis in older
versus younger adults. J Am Geriatr Soc 1992;40:1100–4.
[4] MacIsaac RJ, Lee LY, McNeil KJ, Tsalamandris C, Jerums G. Influence of age on the
presentation and outcome of acidotic and hyperosmolar diabetic emergencies. Intern Med
J 2002;32:379–85.
[5] Kitabchi AE, Umpierrez GE, Murphy MB, Barrett EJ, Kreisberg RA, Malone JI, et al.
Management of hyperglycemic crises in patients with diabetes. Diabetes Care 2001;24:
131–53.
[6] Gale EA, Dornan TL, Tattersall RB. Severely uncontrolled diabetes in the over-fifties.
Diabetologia 1981;21:25–8.
[7] Dallo FJ, Weller SC. Effectiveness of diabetes mellitus screening recommendations. Proc
Natl Acad Sci U S A 2003;100:10574–9.
[8] Umpierrez G, Freire AX. Abdominal pain in patients with hyperglycemic crises. J Crit
Care 2002;17:63–7.
[9] Axelrod L, Shulman GI, Blackshear PJ, Bornstein W, Roussell AM, Aoki TT. Plasma level
of 13, 14-dihydro-15-keto-PGE2 in patients with diabetic ketoacidosis and in normal
fasting subjects. Diabetes 1986;35:1004–10.
1082 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

[10] Fulop M, Rosenblatt A, Kreitzer SM, Gerstenhaber B. Hyperosmolar nature of diabetic


coma. Diabetes 1975;24:594–9.
[11] Matz R. Hypothermia in diabetic acidosis. Hormones 1972;3:36–41.
[12] Gale EA, Tattersall RB. Hypothermia: a complication of diabetic ketoacidosis. BMJ 1978;
2:1387–9.
[13] Rossetti L, Giaccari A, DeFronzo RA. Glucose toxicity. Diabetes Care 1990;13:610–30.
[14] Schade DS, Eaton RP, Standefer J. Glucocorticoid regulation of plasma ketone body
concentration in insulin deficient man. J Clin Endocrinol Metab 1977;44:1069–79.
[15] Alberti KG, Hockaday TD. Diabetic coma: serum growth hormone before and during
treatment. Diabetologia 1973;9:13–9.
[16] Muller WA, Faloona GR, Unger RH. Hyperglucagonemia in diabetic ketoacidosis: its
prevalence and significance. Am J Med 1973;54:52–7.
[17] Christensen NJ. Plasma norepinephrine and epinephrine in untreated diabetics, during
fasting and after insulin administration. Diabetes 1974;23:1–8.
[18] Barnes AJ, Bloom SR, Goerge K, Alberti GM, Smythe P, Alford FP, et al. Ketoacidosis in
pancreatectomized man. N Engl J Med 1977;296:1250–3.
[19] Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of
monitoring to diabetes. Diabetes Metab Res Rev 1999;15:412–26.
[20] West ML, Marsden PA, Singer GG, Halperin ML. Quantitative analysis of glucose loss
during acute therapy for hyperglycemic hyperosmolar syndrome. Diabetes Care 1986;9:
465–71.
[21] Skott P, Hother-Nielsen O, Bruun NE, Giese J, Nielsen MD, Beck-Nielsen H, et al. Effects
of insulin on kidney function and sodium excretion in healthy subjects. Diabetologia 1989;
32:694–9.
[22] Lolekha PH, Taksinamanee R. Evaluation of serum creatinine measurement by the direct
acidification method for errors contributed by non-creatinine chromogens. Clin Chim Acta
1980;107:97–104.
[23] Lorber D. Nonketotic hypertonicity in diabetes mellitus. Med Clin North Am 1995;79:
39–52.
[24] Przasnyski EJ, Fariss BL. Hyperosmolar hyperglycemic non-ketotic syndrome treated
without insulin: case reports. Mil Med 1978;143:556–7.
[25] Worthley LI. Hyperosmolar coma treated with intravenous sterile water: a study of three
cases. Arch Intern Med 1986;146:945–7.
[26] Metz S. Little risk of hyperosmolar coma following hyperglycemia during cardiopulmo-
nary bypass. Anesthesiology 1991;75:912–3.
[27] Abramson E, Arky R. Diabetic acidosis with initial hypokalemia: therapeutic implications.
JAMA 1966;196:401–3.
[28] Ginsberg HN. Investigation of insulin resistance during diabetic ketoacidosis: role
of counterregulatory substances and effect of insulin therapy. Metabolism 1977;26:
1135–46.
[29] Owen OE, Trapp VE, Reichard GA Jr, Mozzoli MA, Smith R, Boden G. Effects of
therapy on the nature and quantity of fuels oxidized during diabetic ketoacidosis. Diabetes
1980;29:365–72.
[30] Rosenthal NR, Barrett EJ. An assessment of insulin action in hyperosmolar hyperglyce-
mic nonketotic diabetic patients. J Clin Endocrinol Metab 1985;60:607–10.
[31] Fisher JN, Shahshahani MN, Kitabchi AE. Diabetic ketoacidosis: low-dose insulin
therapy by various routes. N Engl J Med 1977;297:238–41.
[32] Harrower AD. Treatment of diabetic ketoacidosis by direct addition of insulin to
intravenous infusion: a comparison of ‘‘high dose’’ and ‘‘low dose’’ techniques. Br J Clin
Pract 1979;33:85–6.
[33] Umpierrez GE, Latif K, Cuervo R, Karabell A, Freire AX, Kitabchi AE. Subcutaneous
aspart insulin: a safe and cost effective treatment of diabetic ketoacidosis. Presented at the
ADA 63rd Scientific Sessions, New Orleans, 13–17 June 2003.
J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084 1083

[34] Joslin EP. The treatment of diabetes mellitus with observations on the disease based upon
thirteen hundred cases. 2nd edition. Philadelphia: Lea and Febiger; 1917.
[35] Morris LR, Murphy MB, Kitabchi AE. Bicarbonate therapy in severe diabetic
ketoacidosis. Ann Intern Med 1986;105:836–40.
[36] Gamba G, Oseguera J, Castrejon M, Gomez-Perez FJ. Bicarbonate therapy in severe
diabetic ketoacidosis: a double blind, randomized, placebo controlled trial. Rev Invest Clin
1991;43:234–8.
[37] Holler JW. Potassium deficiency occurring during the treatment of diabetic acidosis.
JAMA 1946;131:1186–9.
[38] Buttler AM. Case records of the Massachusetts General Hospital: weekly clinicopatho-
logical exercises. Case 30451. N Engl J Med 1944;231:657–61.
[39] Knochel JP. Hypophosphatemia. Clin Nephrol 1977;7:131–7.
[40] Gibby OM, Veale KE, Hayes TM, Jones JG, Wardrop CA. Oxygen availability from the
blood and the effect of phosphate replacement on erythrocyte 2, 3-diphosphoglycerate and
haemoglobin-oxygen affinity in diabetic ketoacidosis. Diabetologia 1978;15:381–5.
[41] Zipf WB, Bacon GE, Spencer ML, Kelch RP, Hopwood NJ, Hawker CD. Hypocalcemia,
hypomagnesemia, and transient hypoparathyroidism during therapy with potassium
phosphate in diabetic ketoacidosis. Diabetes Care 1979;2:265–8.
[42] Winter RJ, Harris CJ, Phillips LS, Green OC. Diabetic ketoacidosis: induction of
hypocalcemia and hypomagnesemia by phosphate therapy. Am J Med 1979;67:897–900.
[43] Wilson HK, Keuer SP, Lea AS, Boyd AE III, Eknoyan G. Phosphate therapy in diabetic
ketoacidosis. Arch Intern Med 1982;142:517–20.
[44] Fisher JN, Kitabchi AE. A randomized study of phosphate therapy in the treatment of
diabetic ketoacidosis. J Clin Endocrinol Metab 1983;57:177–80.
[45] Yadav D, Nair S, Norkus EP, Pitchumoni CS. Nonspecific hyperamylasemia and
hyperlipasemia in diabetic ketoacidosis: incidence and correlation with biochemical
abnormalities. Am J Gastroenterol 2000;95:3123–8.
[46] Nair S, Yadav D, Pitchumoni CS. Association of diabetic ketoacidosis and acute
pancreatitis: observations in 100 consecutive episodes of DKA. Am J Gastroenterol 2000;
95:2795–800.
[47] Wang LM, Tsai ST, Ho LT, Hu SC, Lee CH. Rhabdomyolysis in diabetic emergencies.
Diabetes Res Clin Pract 1994;26:209–14.
[48] Moller-Petersen J, Andersen PT, Hjorne N, Ditzel J. Nontraumatic rhabdomyolysis during
diabetic ketoacidosis. Diabetologia 1986;29:229–34.
[49] Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15:
44–54.
[50] Arieff AI, Carroll HJ. Nonketotic hyperosmolar coma with hyperglycemia: clinical
features, pathophysiology, renal function, acid-base balance, plasma-cerebrospinal fluid
equilibria and the effects of therapy in 37 cases. Medicine (Baltimore) 1972;51:73–94.
[51] Whelton MJ, Walde D, Havard CW. Hyperosmolar non-ketotic diabetic coma: with
particular reference to vascular complications. BMJ 1971;1:85–6.
[52] Tchertkoff V, Nayak SV, Kamath C, Salomon MI. Hyperosmolar nonketotic diabetic
coma: vascular complications. J Am Geriatr Soc 1974;22:462–6.
[53] Pinies JA, Cairo G, Gaztambide S, Vazquez JA. Course and prognosis of 132 patients with
diabetic non ketotic hyperosmolar state. Diabetes Metab 1994;20:43–8.
[54] Hope RA, editor. Diabetic emergencies. Oxford handbook of clinical medicine. 4th
edition. Oxford: Oxford University Press; 1998. p. 784.
[55] Kian K, Eiger G. Anticoagulant therapy in hyperosmolar non-ketotic diabetic coma.
Diabet Med 2003;20:603.
[56] Adrogue HJ, Wilson H, Boyd AE III, Suki WN, Eknoyan G. Plasma acid-base patterns in
diabetic ketoacidosis. N Engl J Med 1982;307:1603–10.
[57] Oh MS, Banerji MA, Carroll HJ. The mechanism of hyperchloremic acidosis during the
recovery phase of diabetic ketoacidosis. Diabetes 1981;30:310–3.
1084 J.L. Gaglia et al / Med Clin N Am 88 (2004) 1063–1084

[58] Hoffman WH, Steinhart CM, el Gammal T, Steele S, Cuadrado AR, Morse PK. Cranial
CT in children and adolescents with diabetic ketoacidosis. Am J Neuroradiol 1988;9:
733–9.
[59] Glaser N, Barnett P, McCaslin I, Nelson D, Trainor J, Louie J, et al. Risk factors for
cerebral edema in children with diabetic ketoacidosis. The Pediatric Emergency Medicine
Collaborative Research Committee of the American Academy of Pediatrics. N Engl J Med
2001;344:264–9.
[60] Bello FA, Sotos JF. Cerebral oedema in diabetic ketoacidosis in children. Lancet 1990;
336:64.
[61] Hirsch ML. Gastric hemorrhage in diabetic coma. Diabetes 1960;9:94–6.
[62] Carroll P, Matz R. Adult respiratory distress syndrome complicating severely uncontrolled
diabetes mellitus: report of nine cases and a review of the literature. Diabetes Care 1982;5:
574–80.
[63] Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, et al.
Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in US
adults. The Third National Health and Nutrition Examination Survey, 1988–1994.
Diabetes Care 1998;21:518–24.
[64] Report of the expert committee on the diagnosis and classification of diabetes mellitus.
Diabetes Care 2003;26:S5–20.
[65] Travaglini MT, Garg SK, Chase HP. Use of insulin lispro in the outpatient management of
ketonuria. Arch Pediatr Adolesc Med 1998;152:672–5.
[66] Runyan JW Jr, Zwaag RV, Joyner MB, Miller ST. The Memphis diabetes continuing care
program. Diabetes Care 1980;3:382–6.
[67] Runyan JW Jr. The Memphis chronic disease program: comparisons in outcome and the
nurse’s extended role. JAMA 1975;231:264–7.
Med Clin N Am 88 (2004) 1085–1105

In-hospital management of type 2


diabetes mellitus
Lillian F. Lien, MD*, M. Angelyn Bethel, MD,
Mark N. Feinglos, MD, CM
Division of Endocrinology, Metabolism, and Nutrition, Department of Medicine,
Duke University Medical Center, Box 3921, Durham, NC 27710, USA

The economic burden of type 2 diabetes mellitus and inpatient


hospitalization
The detrimental effects of diabetes mellitus on quality of life have been
well described, but recently, an increasing economic burden has been
identified. Hospitalization costs for the diabetic population represent
a significant economic impact, and with the increasing prevalence of type
2 diabetes, the importance of minimizing hospital costs is paramount.
Studies such as the 1998 pan-European multicenter study, Cost of Diabetes
in Europe—type 2 (CODE-2) [1], for example found that the costs for 777
Swedish patients with type 2 diabetes represented approximately 6% of the
national total health expenditures in 1997. Furthermore, hospitalization-
related costs were 42% of the total direct costs, and 25% of the patients
were hospitalized at any given time during a 1-year period [1]. Similar
findings have been described in the United States. Bhattacharyya and Else
[2] studied principal cost drivers in patients with type 2 diabetes in
a managed care setting in Honolulu, Hawaii, in 1995 and found that
‘‘hospitalized patients, on average, incur an almost 424% higher medical
cost compared with patients with no experience with inpatient care.’’
O’Brien et al [3] studied acute care inpatient profiles in discharge databases
from five American states from 1994 to 1995. The average cost per event for
a cardiovascular complication such as acute myocardial infarction was
$27,630, of which 60% was the result of acute care hospitalization.

Dr. Feinglos has received research support from Aventis Pharmaceuticals, Glaxo-Smith-
Kline, Pfizer Pharmaceuticals, Novartis, and Hoffman La-Roche.
* Corresponding author.
E-mail address: lien0002@mc.duke.edu (L.F. Lien).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.002
1086 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

Admitting diagnoses leading to hospitalization in patients with type 2


diabetes
The association between type 2 diabetes and cardiovascular disease is
well established [4,5]. Erkens et al [6] compiled data from patients with type
2 diabetes over several years throughout six cities in the Netherlands. The
risks of cardiovascular drug use and cardiovascular hospitalization were
markedly increased in the diabetic patients at the time of initiation of oral
antidiabetic therapy versus matched nondiabetic subjects (relative risks 1.28
and 1.54, respectively). In the Scotland database, Donnan et al [7] described
a group of patients with type 2, type 1, or no diabetes in 1995. Finished
consultant episodes detailing inpatient hospital time spent under the care of
a particular specialist were recorded for each patient. Cardiovascular disease
was responsible for 27% of the finished consultant episodes in type 2 dia-
betic patients, whereas the corresponding percentages were 8% and 12% in
the type 1 and nondiabetic cohorts, respectively. Additionally, 6% of
patients with type 2 diabetes were hospitalized for neurologic complications,
compared with 2% in type 1 diabetic patients and 2% in nondiabetic
patients. Levetan and Magee [4] note that there are considerable data linking
unspecified diabetes to the incidence, morbidity, and mortality of stroke.
Moss et al [8] also studied risk factors for hospitalization in the
Wisconsin Epidemiologic Study of Diabetic Retinopathy cohort. In the
group of participants with ‘‘older-onset’’ diabetes, self-reported reasons for
hospitalizations showed a distribution that included heart disease, 7.8%,
glycemic control, 3%, stroke, 1.8%, amputation, 1.5%, and renal compli-
cations 0.9%. Baseline characteristics of this group also revealed that 39.4%
of patients with elevated glycosylated hemoglobin of 10.5% to 18.3% had
been hospitalized within the year before the survey, compared with only
20.1% of patients with glycosylated hemoglobin of 5.0% to 7.9%. Higher
systolic blood pressure in patients with type 2 diabetes was also associated
with increased risk of hospitalization.

Impact of type 2 diabetes on the patient’s hospital course


In the Scotland database, the median length of stay was significantly
increased in patients who had been diagnosed with type 2 diabetes;
nondiabetic patients were hospitalized for 3 days, patients with type 1
diabetes for 3 days, and patients with type 2 diabetes were hospitalized for 7
days (P \ 0.0001) [7]. The mean length of stay in the Swedish CODE-2
study was even longer at 14.8 days [1].
Diabetes has also been shown to affect infection rate and mortality.
Yellin et al [9] showed that in a group of 10,863 patients who had cardio-
vascular surgery with median sternotomy, 280 patients developed deep
sternal infection, and a subset of 15 patients experienced major bleeding; the
mortality rate for the subset of patients with major bleeding was 53.3%. In
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1087

the authors’ analysis of predisposing risk factors, 13 of the 15 patients with


major bleeding had underlying comorbidities, and nine had type 2 diabetes.
Likewise, Berger et al [10] showed a significant increase in mortality at
30 days and 1 year for patients with diabetes in a cohort of Medicare
beneficiaries hospitalized for acute myocardial infarction in US nonfederal
medical centers from 1994 to1996. Donnan et al [11] also demonstrated that
patients with type 2 diabetes had a 40% higher death rate after a first acute
myocardial infarction, even after adjustment for factors such as age,
smoking and hypertension status.

Inpatient management: laying the groundwork


Education and nursing care
The care of a hospitalized patient with diabetes can be particularly
challenging for the nursing staff. Many patients with long-standing diabetes
are accomplished at managing their disease in the outpatient setting. The
loss of control during hospitalization resulting from issues of diet, glucose
monitoring, and the medication regimen can be stressful; however, among
patients who do not have a thorough understanding of diabetes manage-
ment, the hospital stay provides an opportunity to improve diabetes edu-
cation. Unfortunately, the rapid turnover of staff nurses creates a dearth of
personnel who are qualified to provide such education. In this setting, the
use of a Certified Diabetes Educator has proven beneficial. Feddersen and
Lockwood [12] showed that in an experimental group having both patient
and nursing staff diabetes education, the mean length of stay for patients on
insulin was shortened (10.1  9.1 vs 11.4  8.5 days, P  0.05). Several
studies have attempted to demonstrate cost savings resulting from the
implementation of specialized diabetes nursing care. A recent trial by Davies
et al [13] randomized patients with type 1 or type 2 diabetes either to control
or to an intervention group in which a diabetes specialist nurse provided
care and advice. The authors found a significant shortening of length of
hospital stay in the intervention group (8.0 days vs 11.0 days, respectively).
Results also showed a significant improvement in knowledge and satisfac-
tion in the intervention group, although no difference was found in quality
of life measures. Levetan et al [14] demonstrated not only the importance of
diabetes education but also the value of a multidisciplinary approach to
diabetes management. Patients seen by the diabetes team had a significantly
lower length of stay than patients seen by an individual consultant or
primary physician. In many institutions, the multidisciplinary team ap-
proach has become the standard of care.

Glycemic goals
Although the diagnosis of diabetes is clearly a predictor of poor outcome,
hyperglycemia has been shown to be an independent marker of in-hospital
1088 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

mortality, even when diabetes is undiagnosed [15]. Glycemic control has


been shown to promote white blood cell function and facilitate wound
healing [4,16,17]. Improvements in hyperglycemia have also reduced
morbidity and mortality. A study of patients with diabetes and acute
myocardial infarction randomized the patients to either an insulin–glucose
infusion or conventional therapy (control). The patients given insulin–
glucose infusion received insulin intravenously (IV) for at least 24 hours,
with the goal of reaching normoglycemia, followed by multidose sub-
cutaneous insulin, whereas control patients received insulin only if it was
‘‘clinically indicated.’’ Overall mortality at 1 year was reduced by 30% in
the insulin infusion group versus the control [18]. Furthermore, a mortal-
ity benefit was still persistent even after a longer-term follow-up period
(average 3.4 years) [19]. Furnary et al demonstrated in multiple studies that
the use of continuous IV insulin infusion decreases the risk of deep sternal
wound infection by 66% and perioperative absolute mortality by 57%
[20–22].
Not only has continuous insulin infusion been shown to reduce
cardiovascular and perioperative mortality, Van den Berghe et al [23]
showed that insulin infusions reduce morbidity and mortality in a general
intensive care setting. Patients in intensive care were randomized to one of
two groups. One group received intensive insulin therapy by IV insulin
infusion initiated for glucose levels exceeding 110 mg/dL, with maintenance
values between 80 and 110 mg/dL. The other group underwent conventional
treatment and received an insulin infusion only for glucose levels greater
than 215 mg/dL, with maintenance values between 180 and 200 mg/dL. The
study showed substantial reductions in intensive care unit mortality, in-
hospital mortality, and morbidity; specifically, intensive care unit mortality
was reduced from 8.0% in the conventional group to 4.6% in the intensive
group (P  0.04). These studies provide a basis for the recommendation of
glycemic goals in the hospitalized patient. For general glycemic goals for
people with diabetes, the American Diabetes Association (ADA) recom-
mends a preprandial plasma glucose level between 90 and 130 mg/dL and
peak postprandial glucose level of less than 180 mg/dL [24]. In a recent
ADA technical review, the Diabetes in Hospitals Writing Committee
recommends a preprandial plasma glucose level of less than 110 mg/dL
and peak postprandial less than 180 mg/dL for hospital management of
hyperglycemia [25].

Management goals
The first goal of inpatient diabetes management is safety. In most cases,
the medical regimen used in the outpatient setting will need to be modified.
Oral hypoglycemic agents may need to be discontinued in favor of scheduled
insulin (see later discussion), and insulin requirements may be markedly
different. Multiple issues may contribute to the difficulty in attaining
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1089

glycemic control in the hospitalized patient. Hirsch et al [17] have


identified variables of erratic insulin absorption, the effects of stress-induced
counter-regulatory hormone responses, change in gut motility, and poor
coordination of timing of insulin administration and meals. Additionally,
medications other than those for treatment of diabetes should be critically
reviewed for incompatibility under conditions of any acute illness. Another
common safety concern is the occurrence of hypoglycemia during the
hospital stay. The possibility of hypoglycemia is increased in the hospital-
ized patient because of acute illness, erratic caloric intake, poor coordi-
nation of insulin dosing with mealtime, and altered consciousness.
Hypoglycemia may also be more difficult to detect in the hospital setting.
Inquiring if patients have ever had documented hypoglycemia or if they
suffer from hypoglycemic unawareness may provide useful information.
Many patients with long-standing diabetes are at particular risk for
unheralded hypoglycemia.
The second inpatient management goal is to achieve control of
hyperglycemia. A detailed description of choices for inpatient diabetes
management is discussed later. Hospital staff should be especially vigilant
for the development of diabetic ketoacidosis (DKA) because the diagnosis
of type 2 diabetes does not preclude this possibility. Many patients with
long-standing type 2 diabetes may have b-cell exhaustion to a degree that
approaches type 1 diabetes. A severe stress in such patients may precipitate
the development of DKA. In chronically ill patients, the development of
DKA may be particularly insidious because it may present without hyper-
glycemia. Management of acute hyperglycemia should also entail a search
for the underlying cause. In addition to insufficient insulin dosing, other
possibilities include infection, dehydration, a cardiac event, medication
effects (ie, steroids, epinephrine), and rebound from a previous hypo-
glycemic episode.

Monitoring
Bedside glucose monitoring should be performed at least four times
daily (ie, before meals and at bedtime for the patient who is eating). A
glucose check at 3 AM can also be useful in patients with fasting hyper-
glycemia because an elevated 3 AM glucose level could indicate insufficient
night time insulin dosing, whereas low 3 AM glucose may be indicative of
an early peak in PM insulin or insufficient caloric intake at bedtime.
Patients with persistent hypoglycemia may require an overall reduction in
insulin dose.
The patient who is assigned to no oral intake of food or liquid (NPO,
‘‘nothing by mouth’’) or to continuous tube feedings should have glucose
levels checked at least every 6 hours. In special circumstances, such as an
unusual bolus tube feeding schedule, the timing of the bedside glucose
checks should be carefully coordinated with the timing of the feedings.
1090 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

Careful monitoring may also detect undiagnosed diabetes. Levetan et al


[26] examined a group of patients with documented plasma glucoses greater
than 200 mg/dL. After examining hospital notes pertaining to this group,
the authors found that two thirds of the inpatients’ daily progress records
failed to refer to the possibility of undiagnosed diabetes.

Inpatient insulin administration


The administration of insulin to patients with type 2 diabetes is the most
common treatment for hyperglycemia in the inpatient setting. The following
recommendations address issues pertaining to insulin monotherapy.

Subcutaneous insulin
Baseline data collection
One of the first steps in determining an appropriate insulin regimen for
an inpatient with type 2 diabetes consists of gathering important base-
line data on the diabetes history, if any. Crucial data items include outpa-
tient insulin or oral agent regimen, dietary habits, usual weight, duration
of diabetes, assessment of home glucose control (frequency of glucose
monitoring, glucose values, and compliance), frequency and severity of
hypoglycemia, and the presence and severity of diabetic complications.
Assessment of outpatient compliance may be particularly valuable; the
stated outpatient regimen may be excessive in patients who are not
regularly taking their medications. For example, a patient who reportedly
takes a total of 40 units of subcutaneous insulin twice daily at home but is,
in truth, noncompliant could suffer from hypoglycemia if this full regimen
is instituted on admission. A history of severe diabetes complications
should suggest the possibility of autonomic neuropathy, which can
complicate inpatient management of blood glucose and hemodynamic
monitoring.

Insulin dosing schedule definitions


A discussion of insulin dosing requires an understanding of common
terminology. ‘‘Standing dose insulin’’ refers to insulin received on a fixed
schedule. ‘‘Sliding scale insulin’’ refers to an insulin dose calculated ac-
cording to the current, measured glucose level and given only on an as-
needed basis. (The use of supplemental insulin is preferred to the use of
sliding scale insulin alone, as discussed later.)

Standing dose insulin regimens


Standing-dose regimens use widely varying types of insulin (Table 1) and
number of daily injections. Conventional regimens generally require
injections once or twice daily. These regimens offer an advantage in
convenience and simplicity but lack flexibility and are limited in their ability
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1091

Table 1
Types of insulin formulations
Insulin formulations Onset to peak Duration Appearance Comments
Ultra-short–acting
Lispro (Humalog) 15 min–1 h 3–4 h Clear Must be taken with food
Aspart (Novolog) 15 min–1 h 3–4 h Clear Must be taken with food
Short Acting
Regular 30 min–4 h 4–6 h Clear
Long Acting
Neutral protamine 1–8 h 12–20 h Cloudy Can be mixed, but draw the
Hagedorn (NPH) short acting insulin firsta
Ultralente 3–10 h 18–24 h Cloudy
Glargine (Lantus) ‘‘No peak’’ 24 h Clearb The only long-acting insulin
that is clearb
NPH dose should be cut by
approximately 20% when
converting to glargine
Cannot be mixed with
short-acting insulin
Combination insulins Long-acting component Short-acting component
70/30 (regular) 70% NPH 30% regular
NovoLog mix 70/30 70% aspart-protamine 30% aspart
suspension
75/25 75% lispro-protamine 25% lispro
suspension
a
The NPH insulin should not be drawn first because contamination of the regular insulin
bottle by a needle containing NPH can alter the kinetics of the regular insulin.
b
Clear.
Data from Refs. [28,32,66].

to achieve tight glycemic control [27,28]. Thus, the concept of regimen


intensification, arises frequently, albeit temporarily, in the inpatient setting.
Various types of insulin are used, but these regimens all require the use
of three to four daily injections to achieve tight glycemic control [27].
Regardless of which style of insulin regimen is chosen, the most important
governing principle is to individualize the regimen to suit the patient’s
unique needs [29].

Determining insulin dose. Calculating a weight-based dosing regimen is


appropriate not only for the patient who is newly diagnosed and being
initiated to insulin therapy but also for the patient who has an established
home regimen that has provided insufficient control in the outpatient
setting. Alternatively, patients with adequate home glucose control will
likely have similar insulin requirements during hospitalization. Patients with
an established history of insulin resistance may require insulin doses in
excess of the amount calculated from weight.
For patients with type 2 diabetes, the total daily insulin dosage can be
estimated at 0.3 to 0.6 units/kg/day. This range reflects varying degrees of
1092 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

insulin sensitivity in patients with type 2 diabetes. Patients who are very
insulin resistant may require dosages as high as 0.6 to 1.0 units/kg/day [30],
whereas a patient who is previously insulin naıı̈ve may be more sensitive and
benefit from a lower starting dosage. After estimating total daily insulin
dose, the next step is to determine the frequency of standing dose insulin
administration. Several options are available.

One injection daily. One daily injection of long-acting insulin is rarely


appropriate for diabetes management in the inpatient setting. This regimen
is typically used in the outpatient setting. Single daily insulin injections
seldom provide adequate glucose control but may be used in combination
with oral hypoglycemic agents.

Two injections daily. For patients who are eating, the estimated daily insulin
dose should be divided into two injections:
 Sixty-seven percent (two thirds) of the total daily dosage is administered
in the morning, before breakfast.
 Of this amount, 67% (two thirds) is given as neutral protamine
Hagedorn (NPH) insulin and 33% (one third) as regular insulin.
 Thirty-three percent (one third) of the total daily dosage is administered
before the evening meal.
 Of this evening amount, the dose is divided into 50% NPH and 50%
regular. Alternatively, in patients on this regimen with persistent
fasting hyperglycemia, the evening amount can be divided into 67%
(two thirds) NPH insulin and 33% (one third) regular insulin.
A premixed preparation of 70% NPH and 30% regular insulin is
available for the patient who has difficulty learning how to mix the insulins.
For the patient who can mix insulin, a ‘‘split-mix’’ regimen can provide
additional flexibility. In this regimen, the ratio of NPH to regular insulin can
be adjusted as needed [28]. For patients who are NPO, all doses of insulin
are typically reduced by 50%.

Three injections daily. Regimens of three injections are not recommended


for patients who are NPO; however, they can be useful in patients who are
eating and experience fasting hyperglycemia:
 Sixty-seven percent (two thirds) of the total daily dosage is administered
before breakfast.
 As in two-injection regimens, this amount should consist of approx-
imately 67% (two thirds) NPH insulin and 33% (one third) regular
insulin.
 Seventeen percent (one sixth) of the total daily dosage is administered as
regular insulin before the evening meal.
 Seventeen percent (one sixth) of the total daily dosage is administered as
NPH before bedtime [28].
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1093

Four injections daily. There are two options for using a four-injection regimen
for the patient who is eating. The first option is to use regular insulin and NPH
insulin:
 Twenty-five percent (one fourth) of the total daily dosage is admin-
istered as regular insulin before breakfast, and again before lunch, and
again before dinner.
 Twenty-five percent (one fourth) of the total daily dosage is admin-
istered as NPH insulin before bedtime.
For optimal efficacy of the pre-meal regular and bedtime NPH intensive
regimens, it is important to consider proper timing of the regular insulin in
relation to meals. The most challenging aspect of this regimen in the
inpatient setting is to ensure the delivery of regular insulin 30 to 45 minutes
before a meal [4,17].
In the patient who is not eating, the total daily insulin dosage is reduced
by 50%. Twenty-five percent of the reduced total daily dosage is given as
regular insulin every 6 hours. NPH should not be used at bedtime in patients
who are NPO for an extended period of time. In the patient who is NPO
only overnight (ie, in preparation for a procedure), however, NPH can be
given at full dose to prevent fasting hyperglycemia.
The second option is to use ultra-short–acting insulin and long-acting,
peakless insulin:
 Seventeen percent (one sixth) of the total daily dosage is administered as
ultra-short acting insulin (eg, lispro or aspart) before breakfast, and
again before lunch, and again before dinner.
 Fifty percent (half) of the total daily dosage is administered before
bedtime as long-acting, peakless insulin (glargine).
An alternative method is to use 0.1 units/kg of ultra-short–acting insulin
at each meal and 0.3 to 0.7 units/kg of long-acting, peakless insulin at
bedtime [29]. Although glargine can be given consistently at any time of the
day, it is commonly given at bedtime, in part to prevent the mistake of
mixing it with a short-acting insulin [31]. Glargine is maintained at an acidic
pH level of 4 to maintain solubility, which is incompatible with other insulin
types [32].
Patients who are not eating should not receive ultra-short–acting insulin.
These patients can either be converted to regular insulin every 6 hours, as
above, or glargine can be used alone. Although there is little evidence to
support the use of glargine alone, when used at an appropriate dose, it
should not cause fasting hypoglycemia.

Continuous subcutaneous insulin infusion


Subcutaneous insulin pump therapy provides a continuous basal dose of
insulin, combined with as-needed bolus doses. In the inpatient setting,
1094 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

continuous subcutaneous insulin infusion can be continued provided the


hospital does not have specific regulations prohibiting patient self admin-
istration of medications. For continuous subcutaneous insulin infusion to
be continued safely, the patient must be alert and oriented and desire the
responsibility of maintaining glucose control. Even when the patient
chooses self-management of diabetes, supervision by nursing staff may be
necessary to maintain documentation of administered doses. If the patient’s
mental status changes during the hospitalization, the insulin regimen should
be changed to one of the nurse-administered injection regimens described
earlier. Close monitoring of the infusion site for signs of infection is
essential [33].

Supplemental insulin
Inappropriate use of sliding scale alone. A common misconception is that
a sliding scale insulin regimen alone is sufficient for diabetes management. A
sliding scale, when used alone, cannot achieve adequate management of
hyperglycemia. These regimens react to the presence of hyperglycemia
instead of acting to prevent the occurrence of hyperglycemia [17,25,29].
Therefore, by definition, patients using this regimen must reach an unaccept-
able level of hyperglycemia before receiving insulin. Sliding scale insulin
given at meal times alone also provides no coverage at night and, therefore,
may result in the development of fasting hyperglycemia. Therefore, a sliding
scale regimen should be used only as a supplemental regimen in conjunction
with a scheduled insulin dosage [17,25,28,29,34].

Calculating supplemental insulin dose. Supplemental insulin should be


administered at meals only (Table 2). The addition of supplemental insulin
at bedtime or 3 AM is discouraged because it may cause hypoglycemia. The
type of insulin used for the supplement should correspond to the type of
short-acting insulin given in the standing dose regimen. A patient receiving
standing dose lispro should also receive lispro in the supplement rather than
regular insulin.

Table 2
Calculating supplemental insulin dose (The interval of the supplement should be 5% of the total
daily scheduled standing dose insulin. The number of dosage units shown in the table are provided
based on a patient receiving 100 units of insulin daily.)
Plasma glucose (mg/dL) Supplemental regular insulin
70–200 Give scheduled insulin
201–250 Add 5 units to scheduled insulin dose
251–300 Add 10 units to scheduled insulin dose
301–350 Add 15 units to scheduled insulin dose
351–400 Add 20 units and notify on-call physician
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1095

Intravenous insulin
Intravenous insulin administration protocols
A number of protocols have been developed to assist with titration of IV
insulin infusion rates. One of the first published protocols was the Portland
protocol described by Furnary et al [20–22]. The need for an IV insulin
infusion protocol that addresses individual characteristics of the patient has
gradually received further attention by other authors as well [25]. Trence
et al [35] published an IV insulin infusion protocol consisting of four
algorithms that take into account the patient’s level of insulin sensitivity.
Similarly, Lien et al [36] developed an IV insulin nomogram that dictates the
rate of change of the insulin infusion according to the rate of change of
glucose levels through the novel method of using a multiplication factor
obtained from an easily accessed format. This allows an even more
individualized approach to patients with varying insulin sensitivity. Imple-
mentation of the new nomogram was found to significantly reduce errors in
the delivery of IV insulin, which also reduced persistent hyperglycemia in
critical care patients [36].

Using intravenous insulin in the inpatient setting


Initiation of the insulin infusion. A patient with DKA always requires an
insulin infusion. Additionally, even in the absence of DKA, a critically ill
patient with unacceptable hyperglycemia for 24 to 48 hours is a candidate
for IV insulin, particularly if the hyperglycemia persists despite increasing
doses of subcutaneous insulin. Many surgical units now also consider IV
insulin to be the treatment of choice in diabetic patients during the
perioperative and postoperative period, based on the data for improved
rates of morbidity and mortality [18,20,23].
Bedside glucose monitoring should be performed hourly for the patient on
IV insulin. Regular insulin is typically used for IV administration. Although
ultra-short–acting types of insulin have been given IV, the efficacy and safety
in this setting is not well studied. Initial insulin infusion rates differ in insulin
resistant and non-insulin resistant states. For the patient in DKA, 0.1 units/
kg/hour is a typical starting rate. For patients not in DKA, insulin at a rate
of 0.025 units/kg/hour is a more appropriate initial infusion rate.

Discontinuation of the insulin infusion. The method of discontinuation of an


insulin infusion is as important as its initiation. One of the most frequently
encountered errors is to discontinue the insulin infusion without adminis-
tration of subcutaneous insulin [36]. Because the half-life of IV insulin is less
than 10 minutes, the first dose of subcutaneous insulin must be given 1 hour
before discontinuing the insulin infusion. Failure to do so allows the rapid
development of hyperglycemia in any patient without sufficient endogenous
insulin secretory capacity. Patients with type 1 diabetes or long-standing
type 2 diabetes are at particular risk for this complication.
1096 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

Several methods are used to transition patients from IV to subcutaneous


insulin; however, these practices have not been studied objectively.
Typically, some fraction of the 24-hour IV insulin requirement is used as
the total daily subcutaneous dosage. Suggestions range from a reduction of
0% to 50%. The choice of insulin dosage depends on several factors: the
level of glucose control in the last 24 hours, the level of physiologic stress in
the last 24 hours, nutritional status, and pre-hospital insulin regimen, if
known. Once the total daily dosage is calculated, it should be divided into
four subcutaneous injections and titrated as described above.
Management of hypoglycemia
Administration of IV dextrose (D50) is the most rapid method of
alleviating hypoglycemia and is appropriate for patients who are un-
conscious, severely symptomatic, or NPO. Patients who are alert and able
to eat should be given 15 g of carbohydrate in a rapidly available form (ie,
one half cup of fruit juice, 4 oz. of nondiet soda, or 3 glucose tablets [glucose
tablet dosages vary depending on the over-the-counter brand. Most contain
approximately 4 g of fast-acting carbohydrate per tablet.]). A common error
is to over-treat hypoglycemia with an excess of carbohydrate. This, in
combination with the counter-regulatory hormone response to hypoglyce-
mia, facilitates subsequent hyperglycemia. After treatment of any hypogly-
cemic episode, frequent bedside glucose monitoring should be continued
until a stable glucose level is assured.

Inpatient use of oral agents


Value of oral agents for inpatients
Oral hypoglycemic agents are another option for inpatient diabetes
control. They can be used alone or in combination with insulin [37]. The
value of oral agents, however, is limited in the inpatient population by several
factors. Severity of illness, planned procedures, and potential complications
should be considered when choosing a diabetes management regimen.
Patients who are too ill to maintain adequate caloric intake should not be
managed with an oral hypoglycemic agent. Likewise, patients who may be
NPO because of illness or planned procedures are not appropriate candidates
for oral agents.
A common error in this population of patients is the discontinuation of
oral agents in the absence of an alternate method for diabetes control. These
patients should instead be converted to a subcutaneous or IV insulin
regimen during hospitalization, as described above. Management with
insulin in these circumstances is safer and has the added benefit of increased
dosing flexibility when caloric intake is erratic.
When oral hypoglycemic therapy is maintained during hospitalization,
particular attention must be given to the potential for complications of
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1097

therapy and adverse outcomes. Each class of hypoglycemic agents carries


a unique risk and benefit profile in this population.

Biguanides
Metformin is currently the only biguanide available for use in the United
States. This agent acts primarily by reducing hepatic glucose output and
increasing peripheral glucose uptake [38]. Metformin is presently considered
a first-line agent for treatment of obese patients with type 2 diabetes because
it does not promote weight gain and may facilitate weight loss. It has the
additional advantage of not causing hypoglycemia.
The most feared complication of metformin is the development of lactic
acidosis. Lactic acidosis can be classified into two types. Type A is typically
caused by marked tissue hypoperfusion with resulting anaerobic production
of lactic acid. In contrast, the mechanism of type B lactic acidosis is less well
understood. Patients with type B lactic acidosis do not exhibit signs of
hypoperfusion. Postulated mechanisms include toxin-induced impairment of
cellular metabolism or regional areas of ischemia [39]. Lactic acidosis
associated with metformin therapy is of the type B and has been attributed
to accumulation of the drug [40].
The actual incidence of lactic acidosis attributable to metformin is
unclear. The use of another biguanide, phenformin, was clearly associated
with the development of lactic acidosis, and this medication was withdrawn
from the market in the late 1970s. Phenformin was estimated to cause an
incidence of lactic acidosis of 40 to 64 cases per 100,000 patient-years. In
contrast, recent examinations of lactic acidosis attributable to metformin
estimate the incidence to be 9 cases per 100,000 patient-years [41].
The incidence of lactic acidosis is presumed to be higher in patients who
have conditions predisposing to hypoxia, including renal insufficiency,
congestive heart failure requiring medical therapy, cardiovascular collapse,
acute myocardial infarction, and septicemia. According to the product
labeling, metformin is contraindicated in these patients; and caution should
be used in patients older than 80, those with hepatic disease, and in those
with chronic obstructive pulmonary disease associated with hypoxemia [42].
However, there are abundant data that demonstrates that metformin is
frequently prescribed for patients who have one or more contraindications
to its use [43–45]. The impact of this prescribing pattern is difficult to
evaluate. In a recent systematic review of 176 prospective and cohort studies
[46] (including 17,156 patients taking metformin and 8943 not taking
metformin), no cases of lactic acidosis were identified. Within the 164
prospective studies included in the systematic review, 156 cases allowed the
inclusion of patients with at least one contraindication, and no adverse
effects were observed; however, there was insufficient information to
estimate the number of participants having a contraindication. These and
other studies have speculated that the association of metformin and lactic
1098 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

acidosis may be coincidental rather than causal, but until the drug is studied
in this population, the safety of metformin in this setting remains unproven.
Concern regarding lactic acidosis and the use of metformin also arises in
conjunction with the use of contrast media. Metformin is excreted unchanged
in the urine, and, in patients with normal renal function, over 90% of
the drug is eliminated within 24 hours. When renal function is impaired,
however, metformin accumulates, theoretically increasing the risk of lactic
acidosis. The administration of contrast media carries the risk of contrast-
induced nephropathy with a peak incidence occurring after approximately 24
to 48 hours. The risk of contrast-induced nephropathy is increased in patients
who are volume-depleted and in those with diabetic nephropathy. Although
the development of lactic acidosis in this setting is a rare complication, it has
been estimated that the incidence of lactic acidosis in diabetic patients
receiving metformin and contrast media is 2 cases per million patients per
year [47]. Current recommendations are that metformin be stopped before
and for 48 hours after contrast administration. Serum creatinine should be
followed closely to detect the development of renal insufficiency. Pretreat-
ment with adequate hydration and N-acetylcysteine has also been shown to
reduce the risk of contrast-induced nephropathy [48–50].

Sulfonylureas
Sulfonylureas act by binding to and closing an ATP-dependent potas-
sium channel (KATP). In the pancreatic b-cell, this binding-closing
mechanism results in sustained membrane depolarization, activation of
voltage-dependent calcium channels, calcium influx, and migration of
insulin-containing vesicles to the cell surface, culminating in insulin release.
The most commonly used sulfonylureas in the United States are glyburide,
glipizide, and glimepiride. In the inpatient setting, the use of sulfonylurea
therapy is limited by the propensity for hypoglycemia. Sulfonylureas should
be discontinued in any patient who is NPO. The kinetics of sulfonylureas
may be changed with hepatic or renal dysfunction; thus, they should be
avoided in patients with significant impairment.
Although the morbidity and mortality benefits of intensive glycemic
control in the acute setting have been demonstrated, particularly among
patients after acute myocardial infarction, the role of sulfonylureas in
achieving glucose control is uncertain. In fact, the debate over the safety of
sulfonylurea therapy is ongoing in patients with concomitant heart disease.
The findings of the University Group Diabetes Program [51] demonstrated
excess cardiovascular mortality in a group treated with tolbutamide,
compared with groups treated with placebo or insulin. However, subsequent
criticism of the University Group Diabetes Program study design and lack
of demonstrable risk in other large studies, such as the United Kingdom
Prospective Diabetes Study [52] revitalized the use of these agents for
diabetes control.
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1099

Recently, the safety of sulfonylureas has again come into question.


Further investigation into the phenomenon of ischemic preconditioning
raises concern regarding direct effects of sulfonylurea therapy on the
myocardium, especially in the setting of acute injury. KATP channels found
in the myocardium are similar, although not identical, to those found in the
b-cell. When myocardium is subjected in vitro to repeated ischemia before
prolonged coronary artery occlusion, it is resistant to infarction. This
‘‘conditioning’’ has been shown to be dependent on the activation of KATP
channels [53]. Recent studies using sequential balloon angioplasty to model
ischemic preconditioning compared treatment with glyburide and glimepir-
ide [54,55]. Glimepiride binds the pancreatic KATP channel with more
affinity than the KATP channel found in myocardium. In patients pretreated
with glimepiride or placebo, chest pain scores and ST segment elevations
were improved after sequential balloon angioplasty, indicating the presence
of ischemic preconditioning. In contrast, glyburide appeared to block
ischemic preconditioning, resulting in persistent chest pain and ST segment
elevation after sequential balloon inflation. There are no studies assessing
the effect of glipizide on ischemic preconditioning.
No studies have been designed or powered to address morbidity or
mortality endpoints related to the use of sulfonylureas in this setting. It
remains unclear whether the surrogate endpoints of chest pain and ST seg-
ment elevation will, in fact, correlate with morbidity or mortality. Therefore,
given the paucity of data available and the suggestion that glyburide may
have unwanted effects, it seems reasonable to avoid glyburide, at least in
patients with active coronary artery disease [56].

Thiazolidinediones
Thiazolidinediones bind to peroxisome proliferator-activator receptors
and enhance peripheral insulin sensitivity through a series of mechanisms
resulting in transcription of insulin-responsive genes [57]. Both pioglitazone
and rosiglitazone are currently available for use in the United States.
Troglitazone, the first thiazolidinedione was removed from the market
because of reports of liver failure. Thiazolidinediones should not be used in
patients with active liver disease or those with transaminases increased more
than 2.5 times normal. If thiazolidinediones are discontinued in the inpatient
setting, effects on insulin sensitivity may persist. Therefore, as the drug effect
wanes, insulin requirements may increase.
The use of thiazolidinediones has been implicated in the exacerbation of
fluid retention, a particular concern in patients with congestive heart failure.
Although both animal and human studies have demonstrated a beneficial
effect of thiazolidinediones on markers of congestive heart failure (ie,
increased cardiac output and stroke volume, decreased peripheral vascular
resistance, decreased left ventricular cavity dilatation), the contribution of
excess edema to worsening heart failure has been more difficult to assess
1100 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

[58,59]. A recent cross-sectional examination of Medicare patients with


heart failure demonstrated that more than 20% of patients with moderately
to severely reduced left ventricular dysfunction were treated with a thiazo-
lidinedione [60]. In one retrospective study of health insurance claims,
patients receiving thiazolidinediones had a 12.4% adjusted risk of de-
veloping heart failure compared with 8.4% in the control group after 36
months of follow-up [61]. The experience in a specialized heart failure clinic,
however, indicates that although 17% of patients using thiazolidinediones
experienced significant fluid retention, the fluid was predominately periph-
eral. Furthermore, fluid retention did not correlate with baseline heart
failure severity or the development of worsening heart failure. Only two of
the 19 patients with worsening edema attributed to thiazolidinedione
therapy had signs of pulmonary edema at the time of drug discontinuation.
All patients experienced resolution of peripheral edema after stopping the
drug [62].
The mechanism of fluid retention caused by thiazolidinediones is poorly
understood, and the clinical benefit of thiazolidinedione therapy in patients
with heart failure remains to be seen. Certainly the initiation of thiazolidi-
nedione therapy is not warranted in patients with poorly compensated heart
failure, and recent initiation of thiazolidinediones should remain in the
differential diagnosis of patients exhibiting worsening heart failure. Al-
though large studies evaluating the safety of thiazolidinedione therapy in
patients with heart failure have not been carried out, recent evidence
indicates that these medications can be used safely and effectively in patients
with stable chronic heart failure if careful monitoring is performed.

Other oral hypoglycemic agents


Other commonly used oral hypoglycemic agents are the meglitinides
(repaglinide, nateglinide) and a-glucosidase inhibitors (acarbose, miglitol).
The clinical value of both these classes in the inpatient setting has not been
studied. Although the potential for hypoglycemia in both classes is limited,
their major role is in modifying postprandial hyperglycemia. Therefore, in
hospitalized patients with erratic or absent food intake, there is no clear role
for these agents. Patients should have these medications discontinued in
favor of a scheduled insulin regimen as described above.

Inpatient management: diet therapy


Appropriate nutrition in the hospital is paramount, not only for those
patients who rely solely on dietary control of their diabetes but also for any
inpatient with diabetes. Although standardized meal patterns based on
exchanges have been used, a variety of alternative systems are also available.
Particularly useful for the inpatient setting is the consistent carbohydrate
diet [25]. In this regimen, the carbohydrate content in each meal is
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1101

reproducible from day to day. However, all three meals should not have
identical carbohydrate content [63]. The consistent carbohydrate approach
also emphasizes the importance of a mixed meal. Meals consisting solely of
carbohydrates, although palatable, may not be tolerated by inactive
inpatients. Carbohydrate should be consumed in a balanced meal with
protein, fat, and fiber.
The patient diagnosed with type 2 diabetes who has been managed by
dietary control requires careful attention in the inpatient setting. The stress
of acute illness may prompt an intensification of treatment that can be
accomplished with either insulin or oral agents, as outlined above.
In patients receiving continuous tube feeding, 25% of the total daily
insulin dose should be administered as regular insulin every 6 hours. In
patients receiving bolus tube feeding, the most important aspect of insulin
administration is coordination of the timing of the insulin with the timing of
the feeding. Collaboration with the inpatient nutritionist is essential. For
example, if a patient is being initiated to bolus feeds, suggest a feeding
schedule of every 3 hours to help correspond better with the insulin
administered on an every 6-hour schedule.

Discharge planning and follow-up after hospitalization


Preparation for discharge is an important component of care for the
inpatient with type 2 diabetes, particularly if significant treatment changes
have been made. The following items should be considered:
 Outpatient regimen. When possible, the final outpatient medication
regimen should be initiated at least 24 hours in advance of discharge.
Because the patterns of food intake, amount of physical activity, and
levels of physiologic stress may change after discharge, it is likely that
the regimen will require adjustment [64]. However, evaluating the
recommended regimen in a controlled environment may help avoid
immediate problems after discharge. This is especially important when
the outpatient insulin regimen differs from the inpatient schedule. If an
oral agent has been initiated, the patient should be carefully educated
about the possibility of hypoglycemia or other common side effects. On
discharge, patients with diabetes should have access to a home glucose
meter, test strips, lancets, and, if applicable, syringes for insulin. If
financial or psychosocial issues are an impediment to adequate diabetes
care, a multidisciplinary approach to outpatient planning, including
a social service consultation should be initiated early in the hospital-
ization.
 Follow-up care. It is important to identify a health care professional
who will assume responsibility for supervising diabetes management.
Ideally, this person should be readily available for consultation, as it is
likely that the diabetes regimen will change when the patient resumes
1102 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

normal activity. Continuity of care is best maintained if this individual


receives information regarding changes made to the diabetes regimen
during hospitalization.
 Education. Any hospitalization is an opportunity to reinforce the
principles of diabetes education [65]. Patients should be reminded of
proper dietary therapy, the risks and benefits of oral agents and insulin,
and proper technique of medication administration. For patients whose
ability to follow the treatment regimen is limited, it is especially
important to identify a responsible caregiver who should be educated in
diabetes management.
Patients taking hypoglycemic medications should be counseled to
maintain an available supply of glucose gel or tablets in the home or car
for treatment of hypoglycemia. For patients at risk of significant hypogly-
cemia, a Glucagon Emergency Kit (consisting of an intramuscular injection
of glucagon for unconscious hypoglycemia) may be useful. However, use of
this device requires that the patient have adequate hepatic function and
a caregiver to administer the injection. All patients taking hypoglycemic
medications should obtain a Medic-Alert bracelet identifying their disease.

Summary
Hospital management of the patient with type 2 diabetes poses many
challenges but also a unique opportunity to improve glycemic control and
patient care. A basic understanding of the goals of inpatient management
and glycemic targets is essential. Proper administration of subcutaneous and
IV insulin, as well as appropriate use or discontinuation of oral hypogly-
cemic agents, can reduce the complexity of a patient’s hospital course and
potentially reduce overall morbidity and mortality.

References
[1] Henriksson F, Agardh C, Berne C, Bolinder J, Lonnqvist F, Stenstrom P, et al. Direct
medical costs for patients with type 2 diabetes in Sweden. J Intern Med 2000;248:387–96.
[2] Bhattacharyya S, Else B. Medical costs of managed care in patients with type 2 diabetes
mellitus. Clin Ther 1999;21(12):2131–42.
[3] O’Brien J, Shomphe L, Kavanagh P, Raggio G, Caro J. Direct medical costs of
complications resulting from type 2 diabetes in the US Diabetes Care 1998;21(7):1122–8.
[4] Levetan C, Magee M. Hospital management of diabetes. Endocrinol Metab Clin North
Am 2000;29(4):745–70.
[5] Jarrett R, Shipley M. Type 2 (non-insulin dependent) diabetes mellitus and cardiovascu-
lar disease – putative association via common antecedents; further evidence from the
Whitehall Study. Diabetologia 1988;31:737–40.
[6] Erkens J, Klungel O, Stolk R, Spoelstra J, Grobbree D, Leufkens H. Cardiovascular
drug use and hospitalizations attributable to type 2 diabetes. Diabetes Care 2001;24(8):
1428–32.
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1103

[7] Donnan P, Leese G, Morris A for the DARTS/MEMO Collaboration. Hospitalizations


for people with type 1 and type 2 diabetes compared with the nondiabetic population of
Tayside, Scotland: a retrospective cohort study of resource use. Diabetes Care 2000;23(12):
1774–9.
[8] Moss S, Klein R, Klein B. Risk factors for hospitalization in people with diabetes. Arch
Intern Med 1999;159:2053–7.
[9] Yellin A, Refaely Y, Paley M, Simansky D. Major bleeding complicating deep sternal
infection after cardiac surgery. J Thorac Cardiovasc Surg 2003;125:554–8.
[10] Berger A, Breall J, Gersh B, Johnson A, Oetgen W, Marciniak T, et al. Effect of diabetes
mellitus and insulin use on survival after acute myocardial infarction in the elderly (the
cooperative cardiovascular project). Am J Cardiol 2001;87:272–7.
[11] Donnan P, Boyle D, Broomhall J, Hunter K, MacDonald T, Newton R, et alfor the
DARTS/MEMO collaboration. Prognosis following first acute myocardial infarction in
type 2 diabetes: a comparative population study. Diabet Med 2002;19:448–55.
[12] Feddersen E, Lockwood D. An inpatient diabetes educator’s impact on length of hospital
stay. Diabetes Educ 1994;20(2):125–8.
[13] Davies M, Dixon S, Currie C, Davis R, Peters J. Evaluation of a hospital diabetes specialist
nursing Service: a randomized controlled trial. Diabet Med 2001;18:301–7.
[14] Levetan C, Salas J, Wilets I, Zumoff B. Impact of endocrine and diabetes team
consultation on hospital length of stay for patients with diabetes. Am J Med 1995;99(1):
22–8.
[15] Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE. Hyperglycemia:
an independent marker of in-hospital mortality in patients with undiagnosed diabetes.
J Clin Endocrinol Metab 2002;87(3):978–82.
[16] Luna B, Feinglos MN. Drug-induced hyperglycemia. JAMA 2001;286:1945–8.
[17] Hirsch I, Paauw D, Brunzell J. Inpatient management of adults with diabetes. Diabetes
Care 1995;18(6):870–8.
[18] Malmberg K, Ryden L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, et al. Randomized
trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic
patients with acute myocardial infarction (DIGAMI Study): effects on mortality at 1 year.
J Am Coll Cardio 1995;26:57–65.
[19] Malmberg K, Norhammar A, Wedel H, Ryden L. Glycometabolic state at admission:
important risk marker of mortality in conventionally treated patients with diabetes mellitus
and acute myocardial infarction: long-term results from the diabetes and insulin-glucose
infusion in acute myocardial infarction (DIGAMI) study. Circulation 1999;99(20):
2626–32.
[20] Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin S, et al. Continuous
insulin infusion reduces mortality in patients with diabetes undergoing coronary artery
bypass grafting. J Thorac Cardiovasc Surg 2003;125:1007–21.
[21] Furnary AP, Zerr KJ, Grunkemeier GL, Starr A. Continuous insulin infusion reduces the
incidence of deep sternal wound infection in diabetic patients after cardiac surgical
procedures. Ann Thorac Surg 1999;67:352–62.
[22] Zerr KJ, Furnary AP, Grunkemeier GL, Bookin S, Kanhere V, Starr A. Glucose control
lowers the risk of wound infection in diabetics after open heart operation. Ann Thorac
Surg 1997;63:356–61.
[23] Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al.
Intensive insulin therapy in critically ill patients. N Engl J Med 2001;345:1359–67.
[24] Anonymous. (A.DA). Summary of recommendations for adults with diabetes mellitus.
Diabetes Care 2003;26(Suppl. 1):S33–50.
[25] Clement S, Braithwaite S, Magee M, Ahmann A, Smith E, Schafer R, et al. Management
of diabetes and hyperglycemia in hospitals. Diabetes Care 2004;27(2):553–91.
[26] Levetan C, Passaro M, Jablonski K, Kass M, Ratner R. Unrecognized diabetes among
hospitalized patients. Diabetes Care 1998;21(2):246–9.
1104 L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105

[27] Belfiore F, Iannello S. Insulin treatment in Type 1 and Type 2 diabetes: practical goals and
algorithms. In: Belfiore F, Mogensen C, editors-in-chief. New Concepts in Diabetes and Its
Treatment. Basil: Karger; 2000. p. 72–89.
[28] American Diabetes Association. Multiple-component insulin regimens. In: Farkas-Hirsch
R, editor. Intensive diabetes management. Alexandria, VA: ADA, Inc; 1995. p. 51–64.
[29] Nathan M, Leahy J. Insulin management of hospitalized diabetic patients. In: Leahy J,
Cefalu W, editors. Insulin therapy. New York: Marcel Dekker, Inc; 2002. p. 153–72.
[30] Nathan D. Insulin treatment of type 2 diabetes mellitus. In: Prote D, Sherwin R, Baron A,
editors. Ellenberg & Rifkin’s diabetes mellitus. 6th edition. New York: McGraw-Hill;
2003. p. 515–22.
[31] Hamann A, Matthaei S, Rosak C, Silvestre L for the HOE901/4007 Study Group. A
randomized clinical trial comparing breakfast, dinner, or bedtime administration of insulin
glargine in patients with type 1 diabetes. Diabetes Care 2003;26(6):1738–44.
[32] Wittlin S, Woehrle H, Gerich J. Insulin pharmacokinetics. In: Leahy J, Cefalu W, editors-in-
chief. Insulin therapy. New York: Marcel Dekker, Inc; 2002. p. 73–85.
[33] American Diabetes Association. Insulin infusion pump therapy. In: Farkas-Hirsch R,
editor. Intensive diabetes management. Alexandria, VA: ADA, Inc; 1995. p. 65–78.
[34] Lilley S, Levine G. Management of hospitalized patients with type 2 diabetes mellitus. Am
Fam Physician 1998;57(5):1079–88.
[35] Trence DL, Kelly JL, Hirsch IB. The rationale and management of hyperglycemia for in-
patients with cardiovascular disease: time for change. J Clin Endocrinol Metab 2003;88:
2430–7.
[36] Lien L, Spratt S, Woods Z, Osborne K, Feinglos M. A new intravenous insulin nomogram
in intensive care units improves management of persistent hyperglycemia. Diabetes 2003;
52(Suppl 1):A125.
[37] Yki-Jarvinen H. Combination therapies with insulin in type 2 diabetes. Diabetes Care
2001;24(4):758–67.
[38] Bailey CJ, Turner RC. Metformin. N Engl J Med 1996;334(9):574–9.
[39] Rose BD, Post TW, Narins RG. Causes of lactic acidosis. In: Rose BD, editor. UpToDate.
Wellesley, MA; 2003.
[40] Feinglos MN, Bethel MA. Treatment of type 2 diabetes mellitus. Med Clin N Am 1998;
82(4):757–90.
[41] Stang M, Wysowski DK, Butler-Jones D. Incidence of lactic acidosis in metformin users.
Diabetes Care 1999;22(6):925–7.
[42] Glucophage (metformin hydrochloride){package insert}. Princeton, NF: Bristol-Myers
Squibb; December 1998.
[43] Masoudi FA, Wang Y, Inzucchi SE, Setaro JF, Havranek EP, Foody JM, et al.
Metformin and thiazolidinedione use in Medicare patients with heart failure. JAMA 2003;
290(1):81–5.
[44] Calabrese AT, Coley KC, DaPos SV, Swanson D, Rao H. Evaluation of prescribing
practices: risk of lactic acidosis with metformin therapy. Arch Intern Med 2002;162(4):
434–7.
[45] Emslie-Smith AM, Boyle DIR, Evans JMM, Sullivan F, Morris AD. Contraindication to
metformin therapy in patients with type 2 diabetes—a population-based study of
adherence to prescribing guidelines. Diabet Med 2001;18(6):483–8.
[46] Salpeter S, Greyber E, Pasternak G, Salpeter E, editors. Risk of fatal and nonfatal lactic
acidosis with metformin use in type 2 diabetes mellitus. Cochrane database of systematic
reviews 2003. Vol 1.
[47] Pond GD, Smyth SH, Roach DJ, Hunter G. Metformin and contrast media: genuine risk
or witch hunt? Radiology 1996;201(3):879–80.
[48] Tadros GM, Mouhayar EN, Akinwande AO, Campbell B, Wood C, Blankenship JA.
Prevention of radiocontrast-induced nephropathy with N-acetylcysteine in patients
undergoing coronary angiography. J Invasive Cardiol 2003;15(6):311–4.
L.F. Lien et al / Med Clin N Am 88 (2004) 1085–1105 1105

[49] Kay J, Chow WH, Chan TM, Lo SK, Kwok OH, Yip A, et al. Acetylcysteine for
prevention of acute deterioration of renal function following elective coronary
angiography and intervention: a randomized controlled trial. JAMA 2003;289(5):553–8.
[50] Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W. Prevention
of radiographic- contrast-agent-induced reductions in renal function by acetylcysteine.
N Engl J Med 2000;343(3):180–4.
[51] University Group Diabetes Program. A study of the effects of hypoglycemic agents on
vascular complications in patients with adult-onset diabetes: mortality results. Diabetes
1970;19:785–830.
[52] Feinglos MN, Bethel MA. Therapy of type 2 diabetes, cardiovascular death, and the
UGDP. Am Heart J 1999;138(Suppl 5):S346–52.
[53] O’Rourke B. Myocardial K-ATP channels in preconditioning. Circ Res 2000;87(10):
845–55.
[54] Lee T-M, Chou T-F. Impairment of myocardial protection in type 2 diabetic patients.
J Clin Endocrinol Metab 2002;88(2):531–7.
[55] Scognamiglio R, Avogaro A, Vigili de Kreutzenberg S, Negut C, Palisi M, Bagolin E, et al.
Effects of treatment with sulfonylurea drugs or insulin on ischemia-induced myocardial
dysfunction in type 2 diabetes. Diabetes 2002;51(3):808–12.
[56] Riddle M. Editorial: Sulfonylureas differ in effects on ischemic preconditioning – is it time
to retire glyburide? J Clin Endocrinol Metab 2003;88(2):528–30.
[57] Saltiel AR, Olefsky JM. Thiazolidinediones in the treatment of insulin resistance and type
II diabetes. Diabetes 1996;45(12):1661–9.
[58] Ogino K, Furuse Y, Uchida K, Shimoyama M, Kimugawa T, Osaki S, et al. Troglitazone
improves cardiac function in patients with congestive heart failure. Cardiovasc Drugs Ther
2002;16(3):215–20.
[59] Shiomi T, Hiroyuki T, Hayashidani S, Suematsu N, Ikeuchi M, Wen J, et al. Pioglitazone,
a peroxisome proliferator-activated receptor-gamma agonist, attenuates left ventricular
remodeling and failure after experimental myocardial infarction. Circulation 2002;106(24):
3126–32.
[60] Masoudi FA, Wang Y, Enzucchi SE, Setaro J, Havranek EP, Foody JM, et al. Metformin
and thiazolidinedione use in medicare patients with heart failure. JAMA 2003;290(1):81–5.
[61] Delea T, Hagiwara M, Edelsberg J, Oster G. Exposure to glitazone antidiabetics and risk
of heart failure among persons with type 2 diabetes: a retrospective population–based
cohort analysis (abstr.). J Am Coll Cardiol 2002;39(Suppl):184A.
[62] Tang WHW, Francis GS, Hoogwerf BJ, Young JB. Fluid retention after initiation of
thiazolidinedione therapy in diabetic patients with established chronic heart failure. J Am
Coll Cardiol 2003;41(8):1394–8.
[63] American Diabetes Association. Translation of the diabetes nutrition recommendations
for health care institutions. Diabetes Care 2003;26(Suppl 1):S70–2.
[64] Surwit R, Schneider M, Feinglos M. Stress and diabetes mellitus. Diabetes Care 1992;15:
1413–22.
[65] Davis E, Midgett L, Gourley C. Teach less, teach better at every opportunity. Diabetes
Educ 1994;20(3):236–40.
[66] NovoLog Mix 70/30 [prescribing information and information pamphlet] Princeton, NJ:
Novo Nordisk Pharmaceuticals, Inc; April 2003.
Med Clin N Am 88 (2004) 1107–1116

Hypoglycemia in type 2 diabetes


Salomon Banarer, MDa, Philip E. Cryer, MDb,*
a
Department of Medicine, University of Louisville, ACB 3rd Floor, Suite 83G11,
550 Jackson Street, Louisville, KY 40209, USA
b
Department of Medicine, Washington University School of Medicine, Campus Box 8127,
660 South Euclid Avenue, St. Louis, MO 63110, USA

The barrier of hypoglycemia


Glycemic control makes a difference for people with type 1 (T1) [1] and 2
(T2) [2,3] diabetes mellitus (DM), but the barrier of hypoglycemia precludes
true glycemic control, that is, maintenance of euglycemia over a lifetime of
diabetes, and thus full realization of its established microvascular and
potential macrovascular benefits. Iatrogenic hypoglycemia is the limiting
factor in the glycemic management of diabetes [4,5], which is true for all
people with diabetes. This article focuses on T2DM treated with insulin
secretagogues or with insulin, which ultimately includes the majority of
people with T2DM. However, even early in the course of T2DM, when
patients are responsive to oral agents and defenses against hypoglycemia
appear to be intact, true euglycemia is rarely maintained over time because
of the barrier of hypoglycemia.
Iatrogenic hypoglycemia causes recurrent and occasionally persistent
physical morbidity or psychosocial morbidity, or both, in most people with
T1DM and many with advanced T2DM; and it is sometimes fatal.
Furthermore, because of the barrier of iatrogenic hypoglycemia, complica-
tions of diabetes can develop or progress despite aggressive glycemic
therapy. For example, in T1DM, retinopathy developed or progressed in
14% of patients during 6.5 years of intensive therapy (compared with 32%
patients during conventional therapy) in the Diabetes Control and
Complications Trial (DCCT) [1]. In T2DM, any microvascular endpoint
was reached in 8% of the patients during intensive therapy after 10 years of

This work was supported in part by NIH grants R37 DK27085, MO1 RR00036, P60
DK20579, and T32 DK07120, and grants from the American Diabetes Association.
* Corresponding author.
E-mail address: pcryer@wustl.edu (P.E. Cryer).

0025-7125/04/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2004.04.003
1108 S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116

follow up (compared with 11% during conventional therapy), as reported in


the United Kingdom Prospective Diabetes Study (UKPDS) [2].
Perhaps because of the barrier of iatrogenic hypoglycemia, aggressive
glycemic therapy has had little impact on the incidence of the macrovascular
complications of diabetes [1–3]. The relationship between mean glycemia
(HbA1C) and microvascular complications is such that it is practical to
improve glycemic control enough to reduce the incidence of retinopathy,
neuropathy, and nephropathy [1–3]. The relationship between mean
glycemia and macrovascular disease, however, appears to be shifted toward
lower mean glycemia [6,7]. Thus, it may not be practical, because of the
barrier of hypoglycemia, to hold glucose levels low enough and long enough
with current therapeutic regimens to reduce the incidence of macrovascular
events in a substantial proportion of people with diabetes.

Frequency of hypoglycemia
During aggressive glycemic therapy, the person with T1DM typically suf-
fers plasma glucose concentrations less that 50 to 60 mg/dL (2.8–3.3 mmol/L)
approximately 10% of the time, symptomatic hypoglycemia approximately
twice per week, and severe, and at least temporarily disabling, hypoglycemia
approximately once per year [4,5]. Based on controlled studies designed to
include treatment to near euglycemia, valid estimates of the frequencies of
these hypoglycemias during aggressive glycemic therapy of T2DM are limited.
Overall, however, the rates are lower in T2DM [4,5].
Representative event rates for severe iatrogenic hypoglycemia (ie, those
episodes that require the assistance of another person) during aggressive
glycemic therapy of T1DM range from 62 [1] through 110 [8] to 170 [9]
episodes per 100 patient-years. Event rates for severe iatrogenic hypogly-
cemia during aggressive glycemic therapy of T2DM with insulin range from
3 [10] through 10 [11] to 73 [9] episodes per 100 patient-years. Based on these
data, it appears that the rates of severe hypoglycemia in T2DM are roughly
10% of those in T1DM, even during aggressive therapy with insulin; rates of
severe hypoglycemia are undoubtedly much lower during treatment with
oral agents.
Unfortunately, hypoglycemia event rates in persons with T2DM have not
been reported in the UKPDS. The cumulative incidence of major hypogly-
cemia (defined as requiring medical assistance or hospital admission) over
6 years was 3.3% in patients taking a sulfonylurea and 11.2% in those taking
insulin [12]. For comparison, severe hypoglycemia (requiring the assistance
of another person) occurred in 65% of persons with T1DM over 6.5 years in
the DCCT [1]. Notably, the UKPDS did report an incidence of major
hypoglycemia in patients using metformin [12]. In theory, biguanides,
thiazolidinediones, and a-glucosidase inhibitors should not cause hypogly-
cemia [5]. Based on data from 5063 patients with T2DM using monotherapy
S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116 1109

with metformin, a sulfonylurea, or insulin, the UKPDS indicates relative


risks (compared with diet alone) of self-reported, incapacitating hypoglyce-
mia of 3, 12 and 55, respectively [13].
The UKPDS findings are that hypoglycemia became progressively more
limiting to glycemic control in patients with advanced T2DM [14]. The
frequencies of severe hypoglycemia have been reported to be similar in
patients with T2DM and T1DM, matched for duration of insulin therapy
[15]. Indeed, a recent population-based study [16] disclosed the striking
finding that hypoglycemia requiring the assistance of emergency medical
personnel occurred as frequently and in the same proportion of patients
with insulin-treated T2DM and T1DM. Given the loss of insulin secretion
over time in T2DM [12], these findings suggest that hypoglycemia becomes
a progressively more frequent clinical problem, approximating that in
T1DM, as patients approach the insulin-deficient end of the spectrum of
T2DM [4,5].

Physiology of glucose counter-regulation and its pathophysiology in diabetes


The prevention or correction of hypoglycemia normally involves both
decrements in insulin and increments in glucose counter-regulatory factors
[17]. There are redundant glucose counter-regulatory (plasma glucose
raising) factors and a hierarchy among these factors. The first defense
against falling plasma glucose concentrations is a decrease in insulin
secretion. Insulin levels fall as plasma glucose declines within the physiologic
postabsorptive plasma glucose concentration range; the estimated arterialized
venous plasma glucose threshold is approximately 81 mg/dL (4.5 mmol/L).
Glucose counter-regulatory hormones are released as plasma glucose falls
just below the physiologic range. Estimated glycemic thresholds are
approximately 68 mg/dL (3.8 mmol/L). Among the counter-regulatory
hormones, glucagon plays a primary role. Thus, glucagon is the second
defense against falling plasma glucose concentrations. Epinephrine, albeit
demonstrably involved, normally is not critical; however, it becomes
critical when glucagon is deficient. Thus, epinephrine is the third defense
against falling plasma glucose concentrations. Cortisol and growth
hormone are involved in the defense against prolonged hypoglycemia
but are not critical to the correction of even prolonged hypoglycemia or to
the prevention of hypoglycemia after an overnight fast. To the extent that
they are involved, glucose autoregulation and neural factors play minor
roles. In summary, decrements in insulin and increments in glucagon and,
absent the latter, increments in epinephrine stand high in the hierarchy of
redundant glucose counter-regulatory factors that normally prevent or
correct hypoglycemia. All three of these mechanisms are impaired in
patients with T1DM and may be reduced in advanced T2DM [4,5].
Iatrogenic hypoglycemia is the result of the interplay of absolute or relative
therapeutic insulin excess and compromised physiologic and behavioral
1110 S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116

defenses against falling plasma glucose concentrations in T1DM [4,5]. As


glucose levels fall, insulin levels do not decrease, glucagon levels do not
increase, and the increase in epinephrine levels is typically attenuated.
(Specifically, the glycemic threshold for epinephrine secretion is shifted to
lower plasma glucose concentrations.) The combination of absent glucagon
and attenuated epinephrine responses causes the clinical syndrome of
‘‘defective glucose counter-regulation.’’ Patients with this syndrome are at
a 25-fold or more increased risk for severe hypoglycemia during aggressive
glycemic therapy compared with patients with absent glucagon but intact
epinephrine responses [4,5]. The attenuated epinephrine response is also
a marker for a reduced sympathoadrenal (sympathetic neural as well as
adrenomedullary) response and, thus, reduced neurogenic (autonomic)
symptom responses. The latter causes the clinical syndrome of ‘‘hypoglycemia
unawareness,’’ or loss of the largely neurogenic warning symptoms that
previously allowed the patient to recognize and self treat episodes of
hypoglycemia. Impaired awareness of hypoglycemia is also associated with
a high frequency of severe hypoglycemia.
Recent antecedent hypoglycemia, including asymptomatic hypoglycemia,
shifts glycemic thresholds for autonomic (including epinephrine) and
symptomatic responses to subsequent hypoglycemia to lower plasma
glucose concentrations in nondiabetic individuals, people with T1DM,
and those with T2DM [4,5,17,18]. It has also been shown to impair glycemic
defense against hyperinsulinemia and to reduce detection of hypoglycemia
in the clinical setting in T1DM [4,5].
The concept of hypoglycemia-associated autonomic failure (HAAF) in
T1DM (Fig. 1) [4,5,18] posits that recent antecedent hypoglycemia causes
both defective glucose counter-regulation (by reducing the epinephrine
response to subsequent hypoglycemia in the setting of an absent glucagon
response) and hypoglycemia unawareness (by reducing the sympathoadre-
nal and the resultant neurogenic symptom responses to subsequent
hypoglycemia) and, thus, a vicious cycle of recurrent iatrogenic hypoglyce-
mia. Perhaps the most compelling support for the clinical impact of HAAF
is the finding that as few as 2 to 3 weeks of scrupulous avoidance of
hypoglycemia reverses hypoglycemia unawareness and improves the re-
duced epinephrine component of defective glucose counter-regulation in
most affected patients (reviewed in references 4 and 5). The mediators and
mechanisms of HAAF are largely unknown but are under active in-
vestigation [4].
It appears that the concept of hypoglycemia-associated autonomic failure
also applies to persons with advanced T2DM [4,5,19]. Given the evidence that
iatrogenic hypoglycemia becomes a progressively more frequent clinical
problem as patients approach the insulin-deficient end of the spectrum of
T2DM [12,14–16], Segel et al [19] found that the glucagon response to
hypoglycemia is virtually absent in T2DM (anti-glutamic acid decarboxylase-
negative) patients who require long-term insulin and show reduced C-peptide
S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116 1111

Hypoglycemia-Associated Autonomic Failure


Absolute Insulin Deficiency

No Type 1 Diabetes No Glucagon

Imperfect Insulin
Replacement

HYPOGLYCEMIA
Hypoglycemia Defective Glucose
Unawareness Counterregulation
Reduced Autonomic (Including
Adrenomedullary) Responses

Symptom Epinephrine

Fig. 1. Hypoglycemia-associated autonomic failure in diabetes. (Modified from Cryer PE.


Iatrogenic hypoglycemia as a cause of hypoglycemia-associated autonomic failure in IDDM:
a vicious cycle. Diabetes 1992;41:255–60; with permission.)

levels (ie, those patients approaching the insulin-deficient end of the spectrum
of T2DM) as it is in T1DM. Furthermore, in these patients and another group
with more insulin-sufficient T2DM, glycemic thresholds for sympathoadrenal
and symptomatic responses to hypoglycemia were shifted to lower plasma
glucose concentrations following hypoglycemia, as they are in T1DM. Thus,
people with advanced T2DM, like those with T1DM, are at risk for developing
hypoglycemia-associated autonomic failure, that is, defective glucose counter-
regulation and hypoglycemia unawareness. This may explain why iatrogenic
hypoglycemia becomes progressively more limiting to glycemic control as
patients approach the insulin-deficient end of the spectrum of T2DM [14–16].
Ultimately, these patients become indistinguishable from those with T1DM,
from a pathophysiologic and therapeutic perspective; this perspective is
illustrated in Fig. 2.

Risk factors for hypoglycemia


The conventional risk factors for hypoglycemia in diabetes are based on
the premise that absolute or relative therapeutic insulin excess, whether
endogenous or exogenous insulin, is the sole determinant of risk (Box 1)
[4,5,20].
Clearly, we must look beyond these conventional risk factors if we are to
understand the pathogenesis of most episodes of severe hypoglycemia.
Iatrogenic hypoglycemia is more appropriately viewed as the result of
the interplay of absolute or relative therapeutic insulin excess and
compromised physiologic and behavioral defenses against falling plasma
1112 S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116

Glucose Insulin Glucagon Epinephrine

Nondiabetic

T1DM No No Attenuated

• Defective Glucose Counterregulation


• Hypoglycemia Unawareness
• Hypoglycemia-Associated Autonomic Failure

T2DM - No - No - Attenuated

Fig. 2. Iatrogenic hypoglycemia, the limiting factor in the glycemic management of diabetes, is
the result of the interplay of absolute or relative therapeutic insulin excess and compromised
physiologic and behavioral defenses against falling plasma glucose concentrations in T1DM
and advanced T2DM.

glucose concentrations in T1DM and advanced T2DM (see Fig. 2) [4,5].


In other words, although marked insulin excess alone can cause
hypoglycemia, the integrity of the glucose counter-regulatory systems
determines whether less marked hyperinsulinemia, which must occur from
time to time because of the pharmacokinetic imperfections of current
regimens, causes an episode of hypoglycemia. Risk factors in the latter

Box 1. Risk factors for iatrogenic hypoglycemia in diabetes


Absolute or relative therapeutic insulin excess
1. Insulin (or other glucose-lowering drug) doses that are
excessive, ill-timed, or of the wrong type
2. Decreased exogenous glucose delivery (missed meals or
snacks, overnight fast, delayed gastric emptying)
3. Increased glucose utilization (exercise)
4. Decreased endogenous glucose production (alcohol)
5. Increased sensitivity to Insulin (late after exercise, middle
of the night, weight loss, increased fitness, drugs that
increase insulin sensitivity, improved glycemic control)
6. Decreased insulin clearance (renal failure)
Compromised physiologic and behavioral defenses against
falling plasma glucose concentrations
1. Insulin deficiency
2. History of severe hypoglycemia, or hypoglycemia
unawareness, or both
3. Aggressive glycemic therapy per se (lower HbA1C, lower
glycemic goals)
S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116 1113

category that are well established in T1DM [4,5,21,22] and are relevant to
T2DM, include: (1) insulin deficiency; (2) a history of severe hypoglyce-
mia, hypoglycemia unawareness, or both; and (3) aggressive glycemic
therapy per se, as evinced by lower HbA1C levels or lower glycemic goals.
These are clinical surrogates of the key features of HAAF in diabetes
[4,5,17,18]. Insulin deficiency indicates that insulin levels will not decrease
and predicts accurately that glucagon levels will not increase as plasma
glucose concentrations fall. A history of hypoglycemia, or of hypoglyce-
mia unawareness or aggressive glycemic therapy per se, which imply
recent antecedent hypoglycemia, are proximate causes of HAAF.

Hypoglycemia risk reduction


Reducing the risk of iatrogenic hypoglycemia [5] involves: (1) addressing
the issue of hypoglycemia and thus acknowledging the problem; (2) applying
the principles of aggressive glycemic therapy; and (3) practicing hypoglyce-
mia risk reduction.
Patient education is a key principle of aggressive glycemic therapy. Given
professional guidance and ongoing support, glycemic control is accom-
plished by a knowledgeable and motivated person and can be sustained
despite progression of the insulin deficiency and increased complexity of the
glycemic regimen. Additional principles include frequent self-monitoring of
blood glucose, the use of flexible insulin dosages and other drug regimens,
and selection of individualized glycemic goals.
Generally, hypoglycemia risk reduction involves considering each of the
conventional risk factors, making the appropriate regimen adjustments, and
considering the risk factors indicative of compromised physiologic and
behavioral defenses against falling plasma glucose concentrations (see Box
1). In a patient with a history of hypoglycemia unawareness (which implies
episodes of hypoglycemia whether or not episodes have been documented),
a 2- to 3-week period of scrupulous avoidance of hypoglycemia is advisable;
its efficacy can be assessed by the return of symptoms.
Theoretically, monotherapeutic treatment of T2DM using a biguanide,
thiazolidinedione, or a-glucosidase inhibitor should not cause hypoglycemia.
Patients responsive to these drugs must have endogenous insulin secretion,
and insulin secretion should decrease appropriately as plasma glucose
concentrations fall. Nonetheless, hypoglycemia, including the level of
severity that requires medical attention or admission to hospital, has been
reported during treatment with metformin [12]. Also, thiazolidinediones and
metformin increase the risk of hypoglycemia when combined with other
glucose-lowering drugs, and a-glucosidase inhibitors necessitate treatment of
hypoglycemia with glucose rather than complex carbohydrates. Among the
sulfonylureas, hypoglycemia is more frequently reported in patients using
long-acting agents such as chlorpropamide or glyburide (glibenclamide) [5].
1114 S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116

The frequency of hypoglycemia in people with T2DM who use rapid-acting


insulin secretagogues such as repaglinide or nateglinide remains to be
defined, but to the extent these drugs enhance glucose-stimulated insulin
secretion, the frequency would be expected to be low. Ultimately, however,
most people with T2DM require treatment with insulin to achieve near
euglycemia [23].
Insulin regimens that minimize the risk of hypoglycemia have been
reviewed [5]. These regimens include the use of a long-acting basal insulin
analog (eg, glargine or detemir) and a rapid-acting analog (eg, lispro or
aspart) with meals in a basal-bolus insulin regimen. These regimens have
been reported to reduce the frequency of hypoglycemia, particularly
nocturnal hypoglycemia. In a split-mixed insulin regimen, moving the dose
time of neutral protamine Hagedorn (NPH) insulin to bedtime has also been
reported to reduce nocturnal hypoglycemia. Although the use of continuous
subcutaneous insulin infusion is a conceptually attractive approach to the
problem of hypoglycemia, convincing evidence is needed from controlled
studies that it reduces hypoglycemia at comparable degrees of glycemic
control.
Specific issues such as meals, exercise, alcohol, advancing age, and the
common problem of nocturnal hypoglycemia have also been reviewed [5].
Approaches to the problem of nocturnal hypoglycemia include the use of
insulin analogs, bedtime snacks (although their efficacy is largely limited to
the first half of the night), and bedtime treatments intended to produce more
sustained increments in endogenous glucose production (eg, alanine or
terbutaline) or exogenous glucose delivery (eg, uncooked cornstarch or an
a-glucosidase inhibitor) throughout the night. The efficacy of these latter
approaches remains to be established in large controlled clinical trials.

Summary
Iatrogenic hypoglycemia is the limiting factor in the glycemic manage-
ment of diabetes and a barrier to true glycemic control and its established
microvascular and potential macrovascular long-term benefits. Compared
with T1DM, severe hypoglycemia occurs less frequently in T2DM, even
during aggressive glycemic therapy, presumably because of intact glucose
counter-regulatory systems early in the course of T2DM. Iatrogenic
hypoglycemia, however, becomes a progressively more frequent problem,
ultimately approximating that in T1DM, in advanced T2DM because of
compromised physiologic and behavioral defenses against falling plasma
glucose concentrations. These syndromes of defective glucose counter-
regulation and hypoglycemia unawareness and the concept of hypoglyce-
mia-associated autonomic failure are analogous to those that develop early
in the course of T1DM. Caregivers should strive to reduce mean glycemia as
much as can be accomplished safely by practicing hypoglycemia risk
S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116 1115

reduction: addressing the issue, applying the principles of aggressive


glycemic therapy, and considering both the conventional risk factors and
those indicative of compromised glucose counter-regulation. Clearly, people
with diabetes need more physiologic approaches to glycemic control, using
current regimens and those to be developed. Regimens need to be tailored to
the degree of insulin deficiency: absolute in established T1DM, relative early
in the course of T2DM, and progressively more absolute in advanced
T2DM.
The reality or possibility of hypoglycemia should not be used by the
caregiver or the patient as an excuse for poor glycemic control particularly
in view of the growing array of glucose-lowering drugs that can be used to
optimize therapy and achieve the best control possible in a given individual
with T2DM. Nonetheless, better methods such as those that provide plasma
glucose-regulated insulin secretion or replacement are needed for people
with T2DM, as well as those with T1DM, if we are to achieve euglycemia
safely over a lifetime of diabetes.

References
[1] The Diabetes Control and Complications Trial Research Group. The effect of intensive
treatment of diabetes on the development and progression of long-term complications in
insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.
[2] The United Kingdom Prospective Diabetes Study Research Group. Intensive blood-
glucose control with sulphonylureas or insulin compared with conventional treatment and
risk of complications in patients with type 2 diabetes. Lancet 1998;352:837–53.
[3] The United Kingdom Prospective Diabetes Study Research Group. Effect of intensive
blood-glucose control with metformin on complications in overweight patients with type 2
diabetes. Lancet 1998;352:854–65.
[4] Cryer PE. Hypoglycaemia: the limiting factor in the glycaemic management of Type I and
Type II Diabetes. Diabetologia 2002;45:937–48.
[5] Cryer PE, Davis SN, Shamoon H. Hypoglycemia in diabetes. Diabetes Care 2003;26:
1902–12.
[6] Stratton IM, Adler AI, Neil HAW, Matthews DR, Manley SE, Cull CA, et al, on behalf of
the UKPDS Research Group. Association of glycaemia with macrovascular and
microvascular complications of type 2 diabetes: prospective observational study. BMJ
2000;321:405–12.
[7] Khaw K-T, Wareham N, Luben R, Bingham S, Oakes S, Welch A, et al. Glycated
haemoglobin, diabetes, and mortality in men in Norfolk cohort of European Prospective
Investigation of Cancer and Nutrition (EPIC-Norfolk). BMJ 2001;322:1–6.
[8] Reichard P, Berglund B, Britz A, Cars I, Nilsson BY, Rosenqvist U. Intensified conventional
insulin treatment retards the microvascular complications of insulin-dependent diabetes
mellitus (IDDM): the Stockholm Diabetes Intervention Study (SDIS) after 5 years. J Intern
Med 1991;230:101–8.
[9] MacLeod KM, Hepburn DA, Frier BM. Frequency and morbidity of severe hypoglycae-
mia in insulin-treated diabetic patients. Diabet Med 1993;10:238–45.
[10] Abraira C, Colwell JA, Nuttall FQ, Sawin CT, Nagel NJ, Comstock JP, et al. Veterans
Affairs Cooperative Study on glycemic control and complications in type II diabetes:
results of the feasibility trial. Diabetes Care 1995;18:1113–23.
[11] Saudek CD, Duckworth WC, Giobbie-Hurder A, Henderson WG, Henry RR, Kelley DE,
Morgan NA, et al for the Department of Veterans Affairs Implantable Insulin Pump
1116 S. Banarer, P.E. Cryer / Med Clin N Am 88 (2004) 1107–1116

Study Group. Implantable insulin pump vs. multiple-dose insulin for non-insulin-
dependent diabetes mellitus: a randomized clinical trial. JAMA 1996;276:1249–58.
[12] The United Kingdom Prospective Diabetes Study (UKPDS) Group. Overview of 6 years’
therapy of type II diabetes: a progressive disease. Diabetes 1995;44:1249–58.
[13] Cull CA, Wright AD, MacLeod KM, Holman RR. Hypoglycaemia in patients with Type 2
diabetes in the UKPDS [abstract]. Diabetologia 2001;44:A217.
[14] The United Kingdom Prospective Diabetes Study (UKPDS) Research Group. A 6-year,
randomized, controlled trial comparing sulfonylurea, insulin, and metformin therapy in
patients with newly diagnosed type 2 diabetes that could not be controlled with diet
therapy. Ann Intern Med 1998;128:165–75.
[15] Hepburn DA, MacLeod KM, Pell AC, Scougal IJ, Frier BM. Frequency and symptoms of
hypoglycaemia experienced by patients with type 2 diabetes treated with insulin. Diabet
Med 1993;10:231–7.
[16] Leese GP, Wang J, Broomhall J, Kelly P, Marsden A, Morrison W, et al for the DARTS/
MEMO Collaboration. Frequency of severe hypoglycemia requiring emergency treatment
in type 1 and type 2 diabetes. Diabetes Care 2003;26:1176–80.
[17] Cryer PE. The prevention and correction of hypoglycemia. In: Jefferson LS, Cherrington
AD, editors. Handbook of physiology, the endocrine pancreas and regulation of
metabolism. New York: Oxford University Press; 2001. p. 1057–92.
[18] Dagogo-Jack SE, Craft S, Cryer PE. Hypoglycemia-associated autonomic failure in
insulin-dependent diabetes mellitus. J Clin Invest 1993;91:819–28.
[19] Segel SA, Paramore DS, Cryer PE. Hypoglycemia-associated autonomic failure in
advanced type 2 diabetes. Diabetes 2002;51:724–33.
[20] The Diabetes Control and Complications Trial Research Group. Epidemiology of severe
hypoglycemia in the Diabetes Control and Complications Trial. Am J Med 1991;90:450–9.
[21] The Diabetes Control and Complications Trial Research Group. Hypoglycemia in the
Diabetes Control and Complications Trial. Diabetes 1997;46:271–86.
[22] Mühlhauser I, Overmann H, Bender R, Bott U, Berger M. Risk factors of severe
hypoglycaemia in adult patients with type 1 diabetes: a prospective population based
study. Diabetologia 1998;41:1274–82.
[23] Nathan DM. Initial management of glycemia in type 2 diabetes mellitus. N Engl J Med
2002;347:1342–9.

S-ar putea să vă placă și