Metabolic Syndrome

Med Clin N Am 91 (2007) xiii–xv
Preface
Robert T. Yanagisawa, MD Derek LeRoith, MD, PhD

Guest Editors
Why a comprehensive review on metabolic syndrome so many years after

recognition of this disorder as a worldwide epidemic? As we gain better un-
derstanding of underlying mechanisms of the syndrome, tailored treatment
modalities based on pathophysiology are of paramount importance to
general internists, endocrinologists, cardiologists, hepatologists, ob-gyns,
and everyone involved in the care of patients with metabolic syndrome.
Pi-Sunyer opens this issue of the Medical Clinics of North America with
a critical review of the various definitions of metabolic syndrome that have
been proposed by different organizations. Definitions have come closer over
the years, and we hope to arrive at one unifying criterion that best helps to
identify patients at risk for diabetes and coronary heart disease. Controversy
over the validity of the syndrome itself is eloquently discussed by Brietzke.
He speculates that insulin resistance may be ‘‘yet another spoke in the wheel
of syndrome,’’ but the actual ‘‘hub may prove to be the visceral adipocyte
producing the inflammatory adipokines and free fatty acids.’’ He proposes
that ‘‘the toxic fat syndrome’’ may more accurately describe the state of
the proinflammatory condition.
The basic pathophysiologic mechanism underlying insulin resistance is
discussed by Lann and LeRoith. Metabolic syndrome is exacerbated by ab-
normal insulin signaling/secretion, impaired glucose disposal, lipotoxicity,
and proinflammatory cytokines. Early-stage beta cell dysfunction manifests
as a loss of first-phase insulin secretion leading to eventual hyperinsuline-
mia, resulting in insulin receptor downregulation and the insulin-resistant
state. A mixture of various adipokines contributes to the modulation of
0025-7125/07/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.mcna.2007.06.014 medical.theclinics.com
xiv PREFACE
hypertension and to endothelial dysfunction. Primary prevention of meta-

bolic syndrome requires modification of the lifestyle that leads to this toxic
state. Wyatt leads a national patient education program to improve meta-
bolic syndrome through a behavioral modification program. Despite high
recidivism in weight reduction, successful strategies on a national level are
critical to halting this epidemic.
The challenge of preventing type 2 diabetes with metabolic syndrome is
discussed by Tupper and Gopalkrishnan. Treatment targeting the root cause
of insulin resistance would not only delay the onset but might also reduce
the overall risk. Prevention in the true sense, however, may require more
long-term study. For those with type 2 diabetes, Joffe and Yanagisawa
review new strategies that are currently available for achieving glycemic con-
trol while at the same time minimizing weight gain and the complications
associated with it. Incretin mimetics, a new class of diabetes medications,
including the GLP-1 receptor agonists and the DPP-4 enzyme inhibitors,
has some advantages in this regard. These medications may have the added
benefit of causing beta cell neogenesis and proliferation, thus slowing the
progression of beta cell failure that occurs in the type 2 diabetic.
The hepatic manifestation of metabolic syndrome, nonalcoholic fatty
liver disease (NAFLD), is the leading cause of chronic liver disease in the
Western world. Abdelmalek and Diehl discuss whether NAFLD is due to
insulin resistance in the liver or is the result of systemic insulin resistance.
They demonstrate that like adipose tissue, fatty hepatocytes produce circu-
lating factors such as TNF-alpha and IL-6, contributing to systemic insulin
resistance. Polycystic ovarian syndrome (PCOS) is an ovarian manifestation
of metabolic syndrome leading to hyperandrogenism. Magnotti and Futter-
weit review the pathophysiology and clinical manifestations linked to ovar-
ian insulin resistance. Treatment recommendations based on disease
mechanism are discussed for both NAFLD and PCOS.
The definition of metabolic syndrome for risk assessment of cardiovas-
cular morbidity and mortality has been questioned. Obunai, Jani, and Dan-
gas conclude that although compared to a Framingham Risk Score,
diagnosis of metabolic syndrome proved less sensitive at predicting cardio-
vascular events, recognition of metabolic syndrome is complementary to the
calculation of a Framingham Risk Score.
Directly linked to cardiovascular risk and endovascular disease is the dys-
lipidemia triad of increased triglyceride, decreased HDL, and increased
small dense LDL. Smith reviews the mechanism of dyslipidemia with insulin
resistance, in which an increase in free fatty acids leads to a decrease in the
core lipid content of LDL, resulting in smaller particles. Smaller LDLs are
more readily taken into vascular subendothelial space, and increase athero-
genesis. Standard measures of LDL levels may underestimate the risk of
atherogenis in metabolic syndrome.
Suzuki and Homma describe the role of angiotensin-converting enzyme
inhibitors and angiotensin-II receptor blockers for hypertensive patients
PREFACE xv
with metabolic syndrome. The effect on the renin–angiotensin system block-

ade effect goes beyond diabetes mellitus and may be beneficial for patients
with impaired glucose tolerance. Aggressive blood pressure control is rec-
ommended with high-risk metabolic syndrome.
To reduce the underlying cause of metabolic syndrome, weight reduction
with diet and exercise is critical. To enhance such behavioral modification
efforts, Bray offers his critical assessment of the current options and future
hopes for pharmacological therapy. Kini, Herron, and Yanagisawa review
bariatric surgery as an alternative treatment option for high-risk individuals
who fail noninvasive treatment measures. A closer look into the pathophys-
iological mechanism of bariatric surgery as incretin-modulating surgery may
lead to future noninvasive treatment options targeting the root cause of met-
abolic syndrome.
This issue highlights the current understanding of pathophysiology, as
well as ongoing controversies. The authors are to be commended on their
excellent focused reviews of the various aspects of metabolic syndrome. This
comprehensive review will help us tackle the ongoing obesity epidemic with
an evidence-based approach.
Robert T. Yanagisawa, MD
Derek LeRoith, MD, PhD
Division of Endocrinology, Diabetes, and Bone Diseases
Mount Sinai School of Medicine
One Gustave L. Levy Place, Box 1055
New York, NY 10029-6574, USA
E-mail addresses: robert.yanagisawa@mssm.edu;
derek.leroith@mssm.edu
Med Clin N Am 91 (2007) xvii
Editor’s Note
Merskey & Teasel: Problems with Insurance-Based Research on Chronic
Pain, Med Clin North Am 91(2007):31–43.
In the above-referenced article, the authors make several statements re-
garding the Canadian Institute for the Relief of Pain and Disability
(CIRPD), formerly called the Physical Medicine Research Foundation
(PMRF). The authors’ statement that this charity was founded by an indus-
trial firm and two insurance companies is unsupported. The organization
was established under the scientific leadership of the late Dr. John McM.
Mennell in collaboration with Marc White. Its founding board members
represented diverse professions and had no relationship to insurance
companies at the time. The ideas and opinions expressed in Medical Clinics
of North America do not necessarily reflect those of the Publisher. In-
formation about CIRPD and its sponsorship policies can be found at
http://www.cirpd.org.
Med Clin N Am 91 (2007) 1025–1040
The Metabolic Syndrome:

How to Approach Differing Definitions
Xavier Pi-Sunyer, MD, MPHa,b,*
a
Division of Endocrinology, Diabetes, and Nutrition, St. Luke’s-Roosevelt Hospital,
1111 Amsterdam Avenue, Room 1020, New York, NY 10025, USA
b
Columbia University College of Physicians and Surgeons, 630 West 168th Street,
New York, NY 10032, USA
The metabolic syndrome (MetS) consists of a cluster of risk factors that

are predictive of the progression to type 2 diabetes and coronary heart
disease (CHD). Defining the MetS has not been an attempt to declare the
existence of a new disease produced by a single pathogenesis, but rather
to put together risk factors that as a group have a greater predictive ability
for disease than is the case if they are considered separately. It has been
promulgated as a signal to health professionals that patients who present
with such a cluster are at much greater danger of morbidity and mortality
from type 2 diabetes and CHD than are patients who do not. It is therefore
an alert for physicians, particularly primary care physicians, that they
should pay particular attention to these patients and address their risk
factors vigorously to prevent progression to serious disease.
A problem, however, has been that differing definitions of the MetS have
been put forward by various health agencies. These differing definitions have
hampered evaluation of the syndrome as a public health indicator because
epidemiologists, in studying the value of MetS as a predictor of disease,
have used differing definitions. The lack of consensus on the definition
and its value has led numerous physicians to ignore MetS. In fact, a position
paper from authors in the United States and Europe questioning the validity
of MetS has appeared recently [1].
Nevertheless, in all definitions, insulin resistance is an underlying central
component of the MetS. This was the main thrust of the first description of
the syndrome by Reaven in 1988 [2]. The two most widely used definitions in
the past have been those proposed by the National Cholesterol Education
* Division of Endocrinology, Diabetes, and Nutrition, St. Luke’s-Roosevelt Hospital,

1111 Amsterdam Avenue, Room 1020, NY 10025.
E-mail address: fxp1@columbia.edu
1026 PI-SUNYER
Program (NCEP) [3] and by the World Health Organization (WHO) [4], but
varying versions have been put forward by other organizations [5,6]. Also,
a modification of the NCEP [7] has been published, and most recently the
International Diabetes Federation (IDF) has modified the WHO definition
further [8,9] so that it is now much closer to the NCEP’s. There are still some
differences, such as how to treat waist circumference, which the IDF
requires and for which it uses ethnic-specific measures. The variations,
however, are more minor than previously. It is hoped that at some point,
all groups will agree to a uniform definition.
The cluster of risk factors with the thresholds of abnormality put forward
by the NCEP is shown in Box 1. It includes a large waist circumference, high
triglycerides, low high-density lipoprotein (HDL) cholesterol, high blood
pressure, and elevated fasting plasma glucose. Having any three of the
five risk factors is considered diagnostic for MetS. As mentioned above,
since the first definition [3], the NCEP has modified the threshold level of
fasting glucose from greater than 110 to greater than 100 mg/dL [7,10].
The WHO cluster is shown in Box 2, and the IDF cluster is shown in Box 3.
Whereas the NCEP-Metabolic Syndrome (NCEP-MetS) is used more
widely in North America, the WHO (WHO-MetS) appears to have been
more widely accepted in Europe. The IDF definition is more recent and it
is not yet clear how and by whom it will be used; however, it will likely
replace the WHO definition. The NCEP-MetS was developed as a tool to
identify individuals at high cardiovascular risk, taking into account the
evidence of increased obesity and decreased physical activity around the
world, whereas the WHO-MetS and the IDF-MetS were developed more
as efforts to emphasize insulin resistance as a clinical risk paradigm.
What are the differences among the three? The NCEP-MetS and the IDF
definitions have a lower threshold for blood pressure (130/85 versus 140/90)
than the WHO. The NCEP does not require waist circumference to be
elevated on all patients, whereas the WHO and the IDF do. The NCEP
does not require frank glucose intolerance or diabetes, whereas the WHO
Box 1. NCEP Adult Treatment Panel III (ATPIII) Definition of the

metabolic syndromedany three or more of the following:
Waist circumference >102 cm (40 in) in men and >88 cm (35 in) in
women
Serum triglycerides >150 mg/dL (1.7 mmol/L)
HDL cholesterol <40 mg/dL (1.0 mmol/L) in men and <35 mg/dL
(1.3 mmol/L) in women
Blood pressure >130/85 mm Hg
Serum glucose >100 mg/dL (5.6 mmol/L)dchanged in 2004 from
110 mg/dl (6.1 mmol/L)
THE METABOLIC SYNDROME 1027
Box 2. WHO definition of the metabolic syndrome

Glucose intolerance, impaired fasting glucose, diabetes, or
insulin resistance (assessed by clamp studies), plus at least two
of the following criteria:
Waist-to-hip ratio of >0.90 in men and >0.85 in women
Serum triglycerides >150 mg/dL (1.7 mmol/L), or HDL
cholesterol <35 mg/dL (0.9 mmol/L) in men and <39 mg/dL
(1.0 mmol/L) in women
Blood pressure >140/90
Urinary albumin excretion rate >20 ug/min or
albumin:creatinine ratio >30 mg/g
and the IDF do. And finally, the NCEP and the IDF do not require micro-
albuminuria, whereas the WHO does. The NCEP is the simplest definition,
and probably for this reason has been adopted more widely in North Amer-
ica than the WHO. The IDF definition [9] is quite new, so its usefulness is
still unclear.
Do the definitions identify the same individuals? Ford and Files [11],
using the third National Health and Nutrition Examination Survey
(NHANES III) data set [12], found that about the same number of
Box 3. IDF consensus worldwide definition of the metabolic

syndrome
Central obesity (defined as waist circumference >94 cm for
Europid men and >80 cm for Europid women, with ethnicity-
specific values for other groups), plus any two of the following
four factors:
Raised triglyceride level >150 mg/dL (1.7 mmol/L), or specific
treatment for this lipid abnormality
Reduced HDL cholesterol <40 mg/dL (0.9 mmol/L) in males
and <50 mg/dL (1.1 mmol/L) in females, or specific
treatment for this lipid abnormality.
Raised blood pressure: systolic BP>130 or diastolic BP >85
mm Hg, or treatment of previously diagnosed hypertension
Raised fasting plasma glucose (FPG) >100 mg/dL (5.6 mmol/L),
or previously diagnosed type 2 diabetes. If above 5.6 mmol/
L or 100 mg/dL, oral glucose tolerance test (OGTT) strongly
recommended but is not necessary to define presence of
the syndrome.
1028 PI-SUNYER
individuals were identified using the two criteria: 25.1% for WHO-MetS
versus 23.9% for NCEP-MetS; however, 15% to 20% of individuals were
selected with one definition but not with the other. Meigs and colleagues
[13] looked at the same issue in a group of people in San Antonio and in
the Framingham Offspring Study. Depending on the sex or ethnicity of
the populations, the prevalence of MetS varied up to 24% between the
two definitions. Thus, it is clear that the definitions select somewhat different
cohorts of individuals.
A number of studies have evaluated the ability of these clusters to predict
type 2 diabetes and atherosclerotic cardiovascular disease (CVD), and also
cardiovascular and all-cause mortality.
Predictive value for diabetes

There have been a number of studies that have shown the strong predic-
tive value of MetS for diabetes with both the NCEP and the WHO defini-
tion. Because glucose intolerance is embedded in all clusters, and because
such intolerance is a strong predictor of diabetes [14], this association is
to be expected. Also, central obesity, a component of the MetS, has been
well-characterized as enhancing the incidence of diabetes [15–18]. In the
Framingham Offspring Study, the NCEP definition showed a 6.9 fold risk
and the WHO definition a 6.1 fold risk for type 2 diabetes [19]. In the Insulin
Resistance Atherosclerosis Study, the NCEP, WHO, and IDF definitions of
the MetS had similar predictive values for the incidence of type 2 diabetes,
with adjusted odds ratios of 4.1, 3.7, and 3.4, respectively [20]. In the west of
Scotland cohort, MetS was also a significant predictor of type 2 diabetes
[21]. In a study of Pima Indians, both NCEP and WHO definitions predicted
diabetes, but WHO criteria were more specific and more sensitive [22]. This
was also found in a similar study in Finland [23].
Predictive value for cardiovascular disease

Central obesity
Central obesity is a required component of both the WHO and the IDF
definition of MetS. Central obesity has been shown to increase the risk of
CHD on its own [24–28]. This has been shown in both cross-sectional and
longitudinal studies [26,29–34]. The IDF, by describing ethnic-specific
central obesity measures, emphasizes that the risk of an expanded waist
circumference varies, so that in many parts of the world a lower cutoff
than the NCEP cutoff of 40 in (103 cm) for men and 35 in (88 cm) for
women is too high. Whether such criteria will be widely adopted is unclear
at this time, but a consensus conference held in Asia in 2000 strongly recom-
mended it [35], citing adequate evidence.
Hyperglycemia
Hyperglycemia is a continuously ascending independent risk factor for
CHD, with no apparent threshold [36–39]. This has been repeatedly
described in longitudinal studies.
High blood pressure

High blood pressure is associated with an increased risk of CHD [40–42].
Thomas and colleagues [43] reported in a 14-year study that hypertension
was the most important risk factor associated with increased cardiovascular
mortality among overweight and obese persons who have the metabolic
syndrome. A Japanese study also reported that hypertension was the stron-
gest risk factor for the development of carotid plaque [44].
Lipids
The MetS is characterized by the development of a highly atherogenic
lipid profile known as atherogenic dyslipidemia [3,45,46]. This is character-
ized by hypertriglyceridemia, low HDL-cholesterol, increased numbers of
small, dense, triglyceride-enriched low-density lipoprotein (LDL) particles,
increased remnant lipoproteins, and elevated apolipoprotein B concentra-
tions [3,45]. Because the LDL particles are rather low in cholesterol,
LDL-C levels are often normal [47]. In addition, although not usually
measured clinically because of their short half-life and the difficulty of assay,
FFA nevertheless play an important role in the pathophysiology of the
metabolic syndrome. Intermediates of free fatty acid (FFA) metabolism
can inhibit glucose transport activity. FFA are elevated in central obesity,
leading to insulin resistance [48]. The liver is central to the dyslipidemia of
insulin resistance, because it remains more insulin sensitive and increases
synthesis of triglyceride-rich very-low density lipoprotein (VLDL) particles
[49,50]. This results in fasting hypertriglyceridemia, greater postprandial
hyperlipemia, and elevations in triglyceride-rich remnant lipoproteins. All
of these are associated with increased CVD risk.
Combination of the risk factors

A combination of hypertension and obesity on a background of atheroscle-
rotic dyslipidemia and impaired fasting glucose contributes greatly to future
CHD risk. Numerous studies have documented a relationship of the MetS
defined by one of the MetS clusters and CHD [51–61]. An example is the
Botnia Study in Sweden and Finland, which looked at the effect of the presence
versus the absence of the WHO definition in men and women of 35 to 70 years
old [51]. The cardiovascular mortality rate was found to be higher in those who
met the WHO-MetS (1999 criteria) (12.0%) than in those who did not (2.2%).
The follow-up period was relatively short at 6.9 years.
1030 PI-SUNYER
There is quite a range of risk for CHD among the studies, related to
differences in definition, differences in population characteristics, and differ-
ences in length of follow-up. Also, a few studies have not been able to
document increased risk [62–64]. This may be related to short follow-up
time, small number of events, or other factors discussed in those individual
articles.
Few studies, however, have investigated how different the varying defini-
tions are for predicting CVD. And it is such a predictive value that should
determine the relative merit of each given definition.
Recently, the IDF definition was compared with the NCEP definition in
two studies. Sone and colleagues [65] did not find the IDF definition to be
better in predicting cardiovascular disease than the NCEP definition in
the Japanese Diabetes Complication Study. Also, Hanley and colleagues
[20] did not find the IDF definition to be superior to either the NCEP or
the WHO definitions in predicting diabetes in the Insulin Resistance and
Atherosclerosis Study (IRAS) population. They also found that requiring
waist circumference as a necessary part of the definition did not improve
the prediction of diabetes.
The MetS has been shown to be associated with an increasing risk of both
all-cause mortality and CVD morbidity and mortality [66,67]. More recent
evidence has strengthened this impression. The Multiple Risk Factor Inter-
vention Trial has reported that in 35- to 57-year-old survivors (substituting
body mass index [BMI] for waist circumference because waist was not
measured), MetS was associated with a 21%, 49%, and 50% increased
risk of total, CVD, and CHD mortality, respectively [68]. A Danish study
of postmenopausal women, using the NCEP definition, showed a 3.2 fold
increase in fatal CV events in those who have the MetS [69]. The Framing-
ham Offspring study reported a 2.9 fold increased risk of CVD and a 2.5
fold increased risk of CHD [19]. In a European study using the NCEP
definition, persons above 50 years of age with a family history of heart dis-
ease had a sixfold risk for a first CV event and a fourfold risk for myocardial
infarction [70].
The Kuopio Ischemic Heart Disease Risk Factor Study [66] examined the
association of mortality with the NCEP-MetS and WHO-MetS in men in
central Finland who were 42 to 60 years old at baseline. The follow-up
period was for an average of 11.4 years. Both definitions were significantly
associated with CHD mortality, but only WHO-MetS was also associated
with all-cause mortality.
In a Turkish study, Onat and colleagues [53] found that using the NCEP-
MetS, 27% of men and 38.6% of women (n ¼ 2398) from a representative
population had MetS at baseline. After 5 years of follow-up, MetS, adjusted
for age, was an independent predictor of subsequent overall fatal and non-
fatal CHD in both genders, with a relative risk of 1.71.
The San Antonio Heart Study Phase 2 [71] included 2815 individuals from
a general representative population, of whom 2372 (primary prevention
population) had neither diabetes nor a history of CVD at baseline. They were
followed-up for an average of 12.7 years. The prevalence of the NCEP-MetS
in the general population was 25.2%, 23.2%, and 17.5%, whereas WHO-
MetS prevalence was 25.2%, 23.3%, 17.9%, respectively. In the general
population, NCEP predicted both all-cause (RR 1.27) and cardiovascular
(RR 1.63) mortality, whereas WHO predicted only cardiovascular mortality.
Moreover, in the general population, the NCEP-MetS components predict-
ing cardiovascular mortality were impaired fasting glucose, low HDL choles-
terol, and high blood pressure, whereas WHO-MetS components predicting
cardiovascular mortality are insulin/glucose and high blood pressure [71].
Furthermore, similar to diabetes in the Framingham and other studies,
there was evidence that gender modified the ability of both NCEP and
WHO to predict CV mortality, with the definitions being more predictive
in women than in men. Participants who have only NCEP-MetS on average
had higher blood pressure, higher fasting glucose, higher triglycerides, lower
HDL cholesterol, and a higher waist circumference, but lower fasting insulin
and 2-hour glucose levels than participants who have only WHO-MetS.
There was a significant association with cardiovascular mortality in both [71].
In the primary prevention population in San Antonio, results were simi-
lar. But those who had only WHO-MetS were younger, more likely to be
Mexican-American, and had higher fasting glucose levels than those who
had NCEP-MetS. No difference in HDL was found [71].
In the primary prevention population, NCEP-MetS was predictive of car-
diovascular mortality (HR 2.01) but not of all-cause mortality, whereas
WHO-MetS was predictive of neither all-cause nor cardiovascular mortal-
ity. Furthermore, the corresponding NCEP-MetS components predicting
cardiovascular mortality were abdominal obesity and high blood pressure,
whereas high blood pressure was the only WHO-MetS component predict-
ing cardiovascular mortality [71].
In the general population the cardiovascular mortality hazard ratios for
NCEP-MetS were 4.65 and 1.82 for women and men, respectively, whereas
corresponding hazard ratios for WHO-MetS were 2.83 and 1.15. In con-
trast, in the primary prevention population, there was no definitive evidence
of an interaction with gender for either NCEP-MetS or WHO-MetS with
respect to cardiovascular mortality [71].
Further longitudinal studies comparing mortality in those having the
MetS using differing MetS definitions have reported varying results. A large
North American study showed all-cause death to be higher with the WHO
definition than with the NCEP [72], but the Horn study found that the
NCEP definition was better in predicting risk of fatal CVD in men and
non-fatal in women [73]. Also, a Japanese report of diabetic subjects showed
that both the NCEP and WHO definitions predicted mortality better than
the IDF [65].
The question has been asked how an increasing number of risk factors in
an individual affect predictive power. In one study, those who had no risk
1032 PI-SUNYER
factors were compared with those who had one or two, and also to those
who had three or more. Comparing the latter with the two former, hazard
ratios of 2.1 and 3.1 were found [60]. Three other studies reported that
risk increased as number of factors increased [21,71,74].
Are some components more important than others for CVD risk assess-
ment? Hyperglycemia, low HDL, and high blood pressure have been found
to present greater risk in two studies [57,71], whereas a third suggested that
blood pressure and low HDL were most important [74].
Superiority of the Framingham risk score to the NCEP-MetS has been
documented in predicting CHD [75–77], so that some physicians use the
former exclusively to assess their patients. But the important concept for
practitioners to understand from the MetS paradigm is that there are
patients at risk for type 2 diabetes and CHD who can be identified much
earlier than they presently are, and in whom preventative treatment is likely
to be effective in delaying the onset of disease.
Other components of the metabolic syndrome

As persons who have the MetS have been studied, it has become evident
that there are other components that also cluster in many of these patients.
Two important ones are inflammation and procoagulant activity. Although
these have not been included in any organization’s formal definition of the
MetS, in explanatory statements all agree that these are important markers
for diabetes and CVD [78]. In addition, the levels of adiponectin and of
oxidative stress are also thought to be important. Thus a few comments on
these components follow.
Inflammation
Additional cardiometabolic risk factors include an inflammatory state
that can be characterized by an increase in c-reactive protein (CRP).
Adipocytes secrete various bioactive proteins that collectively are called
adipokines or adipocytokines [79–84]. Through their secretory products,
the cells within adipose tissue depots acquire an ability to cross-talk with
each other and with other organs such as brain, liver, and muscle [85].
Adipocytokines, therefore, can exert local, peripheral, and central effects
[86,87]. Some of these bioactive proteins are tumor necrosis factor-a, inter-
leukin-1 and interleukin-6. They contribute to a state of chronic systemic
and local vascular inflammation as the adipose depot grows.
It is important to note that macrophages are a normal component of
adipose tissue, and that their number increases as the adipose depot enlarges
and they are activated [88]. In obese individuals, enlarged, activated adipo-
cytes recruit macrophages that release various factors that promote both
inflammation and insulin resistance [87,89,90].
Coagulation
The MetS is associated with a procoagulable state. Plasminogen activator
inhibitor-1 (PAI-1), which inhibits tissue plasminogen activator, is a regulator
of the endogenous fibrinolytic system [91]. Increased concentrations of PAI-1
retard the production of plasminogen, leading to a decreased clearance of
clots. This can promote the development of thrombi, which can potentiate
myocardial infarction and stroke [92]. Overexpression of PAI-1 in adipose
tissue has been reported in obese mice and obese individuals [85,93], and
elevated PAI-1 concentrations have been detected in the plasma of obese
individuals [85,94]. A close association between enhanced PAI-1 expression
and intra-abdominal tissue distribution also has been reported [79].
Adiponectin
Adiponectin is an adipokine that is secreted from fat cells in greater
amounts when a person is lean, and in lesser amounts when a person is
obese. Low adiponectin levels found in obese patients increase when weight
is lost [95,96]. Because adiponectin improves glucose use and fat oxidation
[97], low levels lead to both hyperglycemia and dyslipidemia. The impor-
tance of adiponectin in the progression of patients who have the MetS to
diabetes mellitus and CVD has not yet been clarified.
Oxidative stress
Oxidative stress may play a role in the pathophysiology of diabetes and
CVD [98,99]. Oxidative stress occurs when the production of oxidation
products, such as reactive oxygen species, exceeds the neutralizing capacity
of the antioxidant systems. The stress can be detected by measuring some of
these products, such as protein carbonyl levels [100], superoxide levels [101],
oxygen radical (ROS) production rates [102], and oxidized low density lipo-
proteins [103]. Sources for ROS production may include a number of paths,
Table 1
Recommendations for a therapeutic lifestyle diet
NCEP ATPIII guidelines
Polyunsaturated fat Up to 10% of total calories
Monounsaturated fat Up to 20% of total calories
Total fat 25%–35% of total calories
Carbohydrate 50%–60% of total calories
Dietary fiber 20–30 g per day
Protein about 15% of total calories
Data from the National Cholesterol Education Program. Third Report of the NCEP Expert
Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult
Treatment Panel III). National Institutes of Health Publication No. 02-5215; September
2002. Available at: www.nhlbi.nih.gov/guidelines/cholesterol/atp3_rpt.htm. Accessed October
17, 2007.
1034 PI-SUNYER
Table 2
Mediterranean diet
30%–40% of total calories
Fat Emphasize olive oil and other monounsaturated fats
Unrefined cereals and products Whole grain bread, pasta, wheat
Vegetables 2–3 servings/day
Fruits 4–6 servings/day
Fish 4–5 servings/week
Nuts O4 servings/week
Wine moderate consumption
Data from the Oldways Preservation Trust. Mediterranean diet pyramid. Available at:
www.oldwayspt.org/med_pyramid.html. Accessed October 17, 2007.
such as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase,

xanthine oxidase, glucose auto-oxidation, inducible nitrous oxide synthase,
and mitochondrial uncoupling. Such oxidative stress can be damaging to
cells and induce insulin resistance and atherosclerosis.
Treatment
The first line of therapy for the MetS should be lifestyle change. This
should aim at decreasing weight and increasing physical activity. A hypo-
caloric low-fat diet is an appropriate approach [104,105] and is described
in Table 1. Other possible approaches are a Mediterranean diet (Table 2)
[106], and in hypertensives, deitary approaches to stop hypertension
(DASH) [107], or the PREMIER diet [108]. Smoking and alcohol should
be addressed. If any of the risk factors are still elevated after lifestyle change
has been tried for a period of a few months, then pharmacological therapy
for all abnormal Met S components, and also for LDL-cholesterol, need to
be instituted until risk factors are appropriately controlled.
Summary
Physicians and other health professionals can look at the MetS as a useful
tool for identifying individuals at risk for type 2 diabetes and CHD. The
syndrome has confused practitioners because of differences in definition.
The definitions have come closer together over the years. But the principles,
no matter what definition, are clear: people who have glucose intolerance,
blood pressure elevation, dyslipidemia, and central obesity are at increased
risk for type 2 diabetes mellitus and CHD. Identifying the presence in a given
patient of the items in the cluster serves as a warning that risk of disease is
increased, and should stimulate the practitioner to institute preventive mea-
sures early, with lifestyle change initially, and pharmacotherapy subse-
quently, as indicated by any continued risk factor presence.
References
[1] Kahn R, Buse J, Ferrannini E, et al. The metabolic syndrome: time for a critical appraisal:
joint statement from the American Diabetes Association and the European Association for
the Study of Diabetes. Diabetes Care 2005;28(9):2289–304.
[2] Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes
1988;37(12):1595–607.
[3] Executive summary of the third report of the National Cholesterol Education Program
(NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol
in adults (Adult Treatment Panel III). JAMA 2001;285(19):2486–97.
[4] Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its
complications. Part 1: diagnosis and classification of diabetes mellitus, provisional report of
a WHO consultation. Diabet Med 1998;15(7):539–53.
[5] Balkau B, Charles MA, Drivsholm T, et al. Frequency of the WHO metabolic syndrome in
European cohorts, and an alternative definition of an insulin resistance syndrome. Diabetes
Metab 2002;28(5):364–76.
[6] Einhorn D, Reaven GM, Cobin RH, et al. American College of Endocrinology position
statement on the insulin resistance syndrome. Endocr Pract 2003;9(3):237–52.
[7] Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the Na-
tional Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation
2004;110(2):227–39.
[8] Alberti KG, Zimmet P, Shaw J. The metabolic syndromeda new worldwide definition.
Lancet 2005;366(9491):1059–62.
[9] International Diabetes Federation. The IDF consensus worldwide definition of the meta-
bolic syndrome. International Diabetes Federation 2005.
[10] Grundy SM, Brewer HB Jr, Cleeman JI, et al. Definition of metabolic syndrome: report of
the National Heart, Lung, and Blood Institute/American Heart Association conference on
scientific issues related to definition. Circulation 2004;109(3):433–8.
[11] Ford ES, Giles WH. A comparison of the prevalence of the metabolic syndrome using two
proposed definitions. Diabetes Care 2003;26(3):575–81.
[12] National Center for Health Statistics. Plan and operation of the third national health and
nutrition examination survey, 1988–1994. [No. 32]. 1994. National Center for Health Sta-
tistics. Vital and Health Statistics Series 1.
[13] Meigs JB, Wilson PW, Nathan DM, et al. Prevalence and characteristics of the metabolic syn-
drome in the San Antonio Heart and Framingham Offspring studies. Diabetes 2003;52(8):2160–7.
[14] Edelstein SL, Knowler WC, Bain RP, et al. Predictors of progression from impaired glucose
tolerance to NIDDM: an analysis of six prospective studies. Diabetes 1997;46(4):701–10.
[15] Ohlson LO, Larsson B, Svärdsudd K, 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(10):1055–8.
[16] Haffner SM, Stern MP, Hazuda HP, et al. Role of obesity and fat distribution in non-insu-
lin-dependent diabetes mellitus in Mexican Americans and non-Hispanic whites. Diabetes
Care 1986;9(2):153–61.
[17] Kaye SA, Folsom AR, Sprafka JM, et al. Increased incidence of diabetes mellitus in relation
to abdominal adiposity in older women. J Clin Epidemiol 1991;44(3):329–34.
[18] Després JP, Nadeau A, Tremblay A, et al. Role of deep abdominal fat in the association
between regional adipose tissue distribution and glucose tolerance in obese women. Diabe-
tes 1989;38(3):304–9.
[19] Wilson PW, D’Agostino RB, Parise H, et al. Metabolic syndrome as a precursor of cardio-
vascular disease and type 2 diabetes mellitus. Circulation 2005;112(20):3066–72.
[20] Hanley AJ, Karter AJ, Williams K, et al. Prediction of type 2 diabetes mellitus with alter-
native definitions of the metabolic syndrome: the Insulin Resistance Atherosclerosis Study.
Circulation 2005;112(24):3713–21.
1036 PI-SUNYER
[21] Sattar N, Gaw A, Scherbakova O, et al. Metabolic syndrome with and without C-reactive
protein as a predictor of coronary heart disease and diabetes in the West of Scotland
Coronary Prevention Study. Circulation 2003;108(4):414–9.
[22] Hanson RL, Imperatore G, Bennett PH, et al. Components of the ‘‘metabolic syndrome’’
and incidence of type 2 diabetes. Diabetes 2002;51(10):3120–7.
[23] Laaksonen DE, Lakka HM, Niskanen LK, et al. Metabolic syndrome and development of
diabetes mellitus: application and validation of recently suggested definitions of the meta-
bolic syndrome in a prospective cohort study. Am J Epidemiol 2002;156(11):1070–7.
[24] Larsson B, Svärdsudd K, Welin L, et al. Abdominal adipose tissue distribution, obesity,
and risk of cardiovascular disease and death: 13 year follow up of participants in the study
of men born in 1913. Br Med J 1984;288:1401–4.
[25] Lapidus L, Bengtsson C, Larsson B, et al. Distribution of adipose tissue and risk of cardio-
vascular disease and death: a 12 year follow up of participants in the population study of
women in Gothenburg, Sweden. Br Med J (Clin Res Ed) 1984;289(6454):1257–61.
[26] Rexrode KM, Carey VJ, Hennekens CH, et al. Abdominal adiposity and coronary heart
disease in women. JAMA 1998;280(21):1843–8.
[27] Lahmann PH, Lissner L, Gullberg B, et al. A prospective study of adiposity and all-cause
mortality: the Malmo Diet and Cancer Study. Obes Res 2002;10(5):361–9.
[28] Gaudet D, Vohl MC, Perron P, et al. Relationships of abdominal obesity and hyperinsuli-
nemia to angiographically assessed coronary artery disease in men with known mutations in
the LDL receptor gene. Circulation 1998;97(9):871–7.
[29] Lakka HM, Lakka TA, Tuomilehto J, et al. Abdominal obesity is associated with increased
risk of acute coronary events in men. Eur Heart J 2002;23(9):706–13.
[30] Yusuf S, Hawken S, Ounpuu S, et al. Effect of potentially modifiable risk factors associated
with myocardial infarction in 52 countries (the INTERHEART study): case-control study.
Lancet 2004;364(9438):937–52.
[31] Hoefle G, Saely CH, Aczel S, et al. Impact of total and central obesity on vascular
mortality in patients undergoing coronary angiography. Int J Obes (Lond) 2005;29(7):
785–91.
[32] Empana JP, Ducimetiere P, Charles MA, et al. Sagittal abdominal diameter and risk of sud-
den death in asymptomatic middle-aged men: the Paris Prospective Study I. Circulation
2004;110(18):2781–5.
[33] Kim SK, Kim HJ, Hur KY, et al. Visceral fat thickness measured by ultrasonography can
estimate not only visceral obesity but also risks of cardiovascular and metabolic diseases.
Am J Clin Nutr 2004;79(4):593–9.
[34] St-Pierre J, Lemieux I, Vohl MC, et al. Contribution of abdominal obesity and hypertrigly-
ceridemia to impaired fasting glucose and coronary artery disease. Am J Cardiol 2002;
90(1):15–8.
[35] World Health Organization. International Association for the Study of Obesity, Interna-
tional Obesity Task Force. The Asia-Pacific perspective: redefining obesity and its treat-
ment. Health Communications Australia Pty Limited; 2000.
[36] Coutinho M, Gerstein HC, Wang Y, et al. The relationship between glucose and incident
cardiovascular events. A metaregression analysis of published data from 20 studies of
95,783 individuals followed for 12.4 years. Diabetes Care 1999;22(2):233–40.
[37] Gerich JE. Clinical significance, pathogenesis, and management of postprandial hypergly-
cemia. Arch Intern Med 2003;163(11):1306–16.
[38] Selvin E, Marinopoulos S, Berkenblit G, et al. Meta-analysis: glycosylated hemoglobin and
cardiovascular disease in diabetes mellitus. Ann Intern Med 2004;141(6):421–31.
[39] Tominaga M, Eguchi H, Manaka H, et al. Impaired glucose tolerance is a risk factor for
cardiovascular disease, but not impaired fasting glucose. The Funagata Diabetes Study.
Diabetes Care 1999;22(6):920–4.
[40] Basile JN. The importance of systolic blood pressure control and cardiovascular disease
prevention. Curr Treat Options Cardiovasc Med 2003;5(4):271–7.
[41] Franklin SS, Larson MG, Khan SA, et al. Does the relation of blood pressure to coronary
heart disease risk change with aging?. The Framingham Heart Study. Circulation 2001;
103(9):1245–9.
[42] Vasan RS, Larson MG, Leip EP, et al. Impact of high-normal blood pressure on the risk of
cardiovascular disease. N Engl J Med 2001;345(18):1291–7.
[43] Thomas F, Bean K, Pannier B, et al. Cardiovascular mortality in overweight subjects: the
key role of associated risk factors. Hypertension 2005;46(4):654–9.
[44] Ishizaka N, Ishizaka Y, Toda E, et al. Hypertension is the most common component of
metabolic syndrome and the greatest contributor to carotid arteriosclerosis in apparently
healthy Japanese individuals. Hypertens Res 2005;28(1):27–34.
[45] Ruotolo G, Howard BV. Dyslipidemia of the metabolic syndrome. Curr Cardiol Rep 2002;
4(6):494–500.
[46] Brunzell JD, Hokanson JE. Dyslipidemia of central obesity and insulin resistance. Diabetes
Care 1999;22(Suppl 3):C10–3.
[47] Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance and type 2 diabetes on
lipoprotein subclass particle size and concentration determined by nuclear magnetic reso-
nance. Diabetes 2003;52(2):453–62.
[48] Jensen MD, Caruso M, Heiling V, et al. Insulin regulation of lipolysis in nondiabetic and
IDDM subjects. Diabetes 1989;38(12):1595–601.
[49] Ginsberg HN, Huang LS. The insulin resistance syndrome: impact on lipoprotein metab-
olism and atherothrombosis. J Cardiovasc Risk 2000;7(5):325–31.
[50] Ginsberg HN. Review: efficacy and mechanisms of action of statins in the treatment of di-
abetic dyslipidemia. J Clin Endocrinol Metab 2006;91(2):383–92.
[51] Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidity and mortality associated
with the metabolic syndrome. Diabetes Care 2001;24(4):683–9.
[52] Kuusisto J, Lempiainen P, Mykkanen L, et al. Insulin resistance syndrome predicts
coronary heart disease events in elderly type 2 diabetic men. Diabetes Care 2001;24(9):
1629–33.
[53] Onat A, Ceyhan K, Basar O, et al. Metabolic syndrome: major impact on coronary risk in
a population with low cholesterol levelsda prospective and cross-sectional evaluation.
Atherosclerosis 2002;165(2):285–92.
[54] Lempiainen P, Mykkanen L, Pyorala K, et al. Insulin resistance syndrome predicts
coronary heart disease events in elderly nondiabetic men. Circulation 1999;100(2):
123–8.
[55] Kekalainen P, Sarlund H, Pyorala K, et al. Hyperinsulinemia cluster predicts the develop-
ment of type 2 diabetes independently of family history of diabetes. Diabetes Care 1999;
22(1):86–92.
[56] Pyorala M, Miettinen H, Halonen P, et al. Insulin resistance syndrome predicts the risk of
coronary heart disease and stroke in healthy middle-aged men: the 22-year follow-up results
of the Helsinki Policemen Study. Arterioscler Thromb Vasc Biol 2000;20(2):538–44.
[57] Alexander CM, Landsman PB, Teutsch SM, et al. NCEP-defined metabolic syndrome, di-
abetes, and prevalence of coronary heart disease among NHANES III participants age 50
years and older. Diabetes 2003;52(5):1210–4.
[58] Katzmarzyk PT, Church TS, Blair SN. Cardiorespiratory fitness attenuates the effects of
the metabolic syndrome on all-cause and cardiovascular disease mortality in men. Arch
Intern Med 2004;164(10):1092–7.
[59] Girman CJ, Rhodes T, Mercuri M, et al. The metabolic syndrome and risk of major coro-
nary events in the Scandinavian Simvastatin Survival Study (4S) and the Air Force/Texas
Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS). Am J Cardiol 2004;
93(2):136–41.
[60] Malik S, Wong ND, Franklin SS, et al. Impact of the metabolic syndrome on mortality
from coronary heart disease, cardiovascular disease, and all causes in United States adults.
Circulation 2004;110(10):1245–50.
1038 PI-SUNYER
[61] Scuteri A, Najjar SS, Morrell CH, et al. The metabolic syndrome in older individuals: prev-
alence and prediction of cardiovascular events: the Cardiovascular Health Study. Diabetes
Care 2005;28(4):882–7.
[62] Bruno G, Merletti F, Biggeri A, et al. Metabolic syndrome as a predictor of all-cause and
cardiovascular mortality in type 2 diabetes: the Casale Monferrato Study. Diabetes Care
2004;27(11):2689–94.
[63] Resnick HE, Jones K, Ruotolo G, et al. Insulin resistance, the metabolic syndrome, and risk
of incident cardiovascular disease in nondiabetic American Indians: the Strong Heart
Study. Diabetes Care 2003;26(3):861–7.
[64] Marroquin OC, Kip KE, Kelley DE, et al. Metabolic syndrome modifies the cardiovascular
risk associated with angiographic coronary artery disease in women: a report from the
Women’s Ischemia Syndrome Evaluation. Circulation 2004;109(6):714–21.
[65] Sone H, Tanaka S, Ishibashi S, et al. The new worldwide definition of metabolic syndrome
is not a better diagnostic predictor of cardiovascular disease in Japanese diabetic patients
than the existing definitions: additional analysis from the Japan Diabetes Complications
Study. Diabetes Care 2006;29(1):145–7.
[66] Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and car-
diovascular disease mortality in middle-aged men. JAMA 2002;288(21):2709–16.
[67] Ford ES. Risks for all-cause mortality, cardiovascular disease, and diabetes associated
with the metabolic syndrome: a summary of the evidence. Diabetes Care 2005;28(7):
1769–78.
[68] Eberly LE, Prineas R, Cohen JD, et al. Metabolic syndrome: risk factor distribution and
18-year mortality in the multiple risk factor intervention trial. Diabetes Care 2006;29(1):
123–30.
[69] Tanko LB, Bagger YZ, Qin G, et al. Enlarged waist combined with elevated triglycerides is
a strong predictor of accelerated atherogenesis and related cardiovascular mortality in post-
menopausal women. Circulation 2005;%19;111(15):1883–90.
[70] Reinhard W, Holmer SR, Fischer M, et al. Association of the metabolic syndrome with
early coronary disease in families with frequent myocardial infarction. Am J Cardiol
2006;97(7):964–7.
[71] Hunt KJ, Resendez RG, Williams K, et al. National Cholesterol Education Program versus
World Health Organization metabolic syndrome in relation to all-cause and cardiovascular
mortality in the San Antonio Heart Study. Circulation 2004;110(10):1251–7.
[72] Nigam A, Bourassa MG, Fortier A, et al. The metabolic syndrome and its components and
the long-term risk of death in patients with coronary heart disease. Am Heart J 2006;151(2):
514–21.
[73] Dekker JM, Girman C, Rhodes T, et al. Metabolic syndrome and 10-year cardiovascular
disease risk in the Hoorn Study. Circulation 2005;112(5):666–73.
[74] McNeill AM, Rosamond WD, Girman CJ, et al. The metabolic syndrome and 11-year risk
of incident cardiovascular disease in the atherosclerosis risk in communities study. Diabetes
Care 2005;28(2):385–90.
[75] Wannamethee SG, Shaper AG, Lennon L, et al. Metabolic syndrome vs Framingham Risk
Score for prediction of coronary heart disease, stroke, and type 2 diabetes mellitus. Arch
Intern Med 2005;165(22):2644–50.
[76] Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk
factor categories. Circulation 1998;97(18):1837–47.
[77] Stern MP, Williams K, Gonzalez-Villalpando C, et al. Does the metabolic syndrome im-
prove identification of individuals at risk of type 2 diabetes and/or cardiovascular disease?
Diabetes Care 2004;27(11):2676–81.
[78] Pi-Sunyer FX, editor. Abdominal adiposity and cardio metabolic risk. Clin Cornerstone
2006;8(Suppl 4):S1–43.
[79] Shimomura I, Funahashi T, Takahashi M, et al. Enhanced expression of PAI-1 in visceral
fat: possible contributor to vascular disease in obesity. Nat Med 1996;2(7):800–3.
[80] Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 2004;
89(6):2548–56.
[81] Havel PJ. Control of energy homeostasis and insulin action by adipocyte hormones: leptin,
acylation stimulating protein, and adiponectin. Curr Opin Lipidol 2002;13(1):51–9.
[82] Mohamed-Ali V, Pinkney JH, Coppack SW. Adipose tissue as an endocrine and paracrine
organ. Int J Obes Relat Metab Disord 1998;22(12):1145–58.
[83] Hotamisligil GS, Arner P, Caro JF, et al. Increased adipose tissue expression of tumor ne-
crosis factor-alpha in human obesity and insulin resistance. J Clin Invest 1995;95(5):
2409–15.
[84] Trayhurn P, Wood IS. Signalling role of adipose tissue: adipokines and inflammation in
obesity. Biochem Soc Trans 2005;33(Pt 5):1078–81.
[85] Ailhaud G. Adipose tissue as a secretory organ: from adipogenesis to the metabolic syn-
drome. C R Biol 2006;329(8):570–7.
[86] Fantuzzi G. Adipose tissue, adipokines, and inflammation. J Allergy Clin Immunol 2005;
115(5):911–9.
[87] Rajala MW, Scherer PE. Minireview: the adipocytedat the crossroads of energy homeosta-
sis, inflammation, and atherosclerosis. Endocrinology 2003;144(9):3765–73.
[88] Nawrocki AR, Scherer PE. The delicate balance between fat and muscle: adipokines in
metabolic disease and musculoskeletal inflammation. Curr Opin Pharmacol 2004;4(3):
281–9.
[89] Lyon CJ, Law RE, Hsueh WA. Minireview: adiposity, inflammation, and atherogenesis.
Endocrinology 2003;144(6):2195–200.
[90] Greenberg AS, Obin MS. Obesity and the role of adipose tissue in inflammation and metab-
olism. Am J Clin Nutr 2006;83(2):461S–5S.
[91] Wiman B, Chmielewska J, Ranby M. Inactivation of tissue plasminogen activator in
plasma. Demonstration of a complex with a new rapid inhibitor. J Biolumin Chemilumin
1984;259(6):3644–7.
[92] Fruhbeck G, Salvador J. Role of adipocytokines in metabolism and disease. Nutr Res 2004;
24:803–26.
[93] Alessi MC, Lijnen HR, Bastelica D, et al. Adipose tissue and atherothrombosis. Pathophy-
siol Haemost Thromb 2003;33(5–6):290–7.
[94] Matsuzawa Y. The metabolic syndrome and adipocytokines. FEBS Lett 2006;580(12):
2917–21.
[95] Arita Y, Kihara S, Ouchi N, et al. Paradoxical decrease of an adipose-specific protein, adi-
ponectin, in obesity. Biochem Biophys Res Commun 1999;257(1):79–83.
[96] Yang WS, Lee WJ, Funahashi T, et al. Weight reduction increases plasma levels of an ad-
ipose-derived anti-inflammatory protein, adiponectin. J Clin Endocrinol Metab 2001;86(8):
3815–9.
[97] Gable DR, Hurel SJ, Humphries SE. Adiponectin and its gene variants as risk factors for
insulin resistance, the metabolic syndrome and cardiovascular disease. Atherosclerosis
2006;188(2):231–44.
[98] Oberley LW. Free radicals and diabetes. Free Radic Biol Med 1988;5(2):113–24.
[99] Rao AV. Lycopene, tomatoes, and the prevention of coronary heart disease. Exp Biol Med
(Maywood) 2002;227(10):908–13.
[100] Saengsirisuwan V, Perez FR, Sloniger JA, et al. Interactions of exercise training and alpha-
lipoic acid on insulin signaling in skeletal muscle of obese Zucker rats. Am J Physiol Endo-
crinol Metab 2004;287(3):E529–36.
[101] Shiuchi T, Iwai M, Li HS, et al. Angiotensin II type-1 receptor blocker valsartan en-
hances insulin sensitivity in skeletal muscles of diabetic mice. Hypertension 2004;43(5):
1003–10.
[102] Ogihara T, Asano T, Katagiri H, et al. Oxidative stress induces insulin resistance by activat-
ing the nuclear factor-kappa B pathway and disrupting normal subcellular distribution of
phosphatidylinositol 3-kinase. Diabetologia 2004;47(5):794–805.
1040 PI-SUNYER
[103] Hansel B, Giral P, Nobecourt E, et al. Metabolic syndrome is associated with elevated
oxidative stress and dysfunctional dense high-density lipoprotein particles displaying
impaired antioxidative activity. J Clin Endocrinol Metab 2004;89(10):4963–71.
[104] Wilson PW, Grundy SM. The metabolic syndrome: practical guide to origins and treat-
ment: Part I. Circulation 2003;108(12):1422–4.
[105] Grundy SM, Hansen B, Smith SC Jr, et al. Clinical management of metabolic syn-
drome: report of the American Heart Association/National Heart, Lung, and Blood Insti-
tute/American Diabetes Association conference on scientific issues related to management.
Circulation 2004;109(4):551–6.
[106] Willett WC, Sacks F, Trichopoulou A, et al. Mediterranean diet pyramid: a cultural model
for healthy eating. Am J Clin Nutr 1995;61(Suppl 6):1402S–6S.
[107] Appel LJ, Moore TJ, Obarzanek E, et al. for the DASH Collaborative Research Group. A
clinical trial of the effects of dietary patterns on blood pressure. N Engl J Med 1997;336:
1117–24.
[108] Svetkey LP, Harsha DW, Vollmer WM, et al. Premier: a clinical trial of comprehensive life-
style modification for blood pressure control: rationale, design and baseline characteristics.
Ann Epidemiol 2003;13(6):462–71.
Med Clin N Am 91 (2007) 1041–1061
Controversy in Diagnosis
and Management of the Metabolic
Syndrome
Stephen A. Brietzke, MD, FACP, FACE
Division of Endocrinology, Department of Internal Medicine, MA406 UMHC,
1 Hospital Drive, Columbia, MO 65212, USA
How clinicians and academicians describe the concept popularly known

as ‘‘the metabolic syndrome’’ is reminiscent of the old tale of blind men
describing an elephant: having only a ‘‘through the glass dimly’’ view of
the big picture, the description is based on the tactile impression of
a body part interpreted as representative of the whole. As the story goes,
one observer describes the elephant as ‘‘like a snake’’ (the tail); another as
‘‘like a tree’’ (the leg); another as ‘‘like a spear’’ (the tusk); and yet another
as ‘‘like a leaf’’ (the ear). To complete the analogy, the cardiologist describ-
ing the metabolic syndrome calls it ‘‘a state of increased coronary heart
disease risk’’; the diabetologist calls it ‘‘prediabetes’’; the endocrinologist
calls it ‘‘insulin resistance with dyslipidemia’’; the hepatologist calls it ‘‘non-
alcoholic fatty liver disease’’; and the nephrologist calls it ‘‘hypertension and
prehypertension.’’ The present controversy centers on this question: is there
a root cause (ie, an ‘‘elephant’’) for all these epiphenomena, or alternatively,
are they truly separate entities that are simply so common as to coexist in
the same patient by chance alone?
Historical perspective: the evolution of the metabolic syndrome

In his 1988 Banting Lecture, Reaven [1] defined a condition he called
‘‘syndrome X,’’ linking hypertension, type 2 diabetes mellitus (DM), and
increased propensity for atherosclerotic events to an underlying cause of
insulin resistance. Perhaps because of confusion with another syndrome x,
The author serves on speakers’ bureaus for Sanofi-Aventis Pharmaceuticals and for
Squibb-Novo.
E-mail address: brietzkes@health.missouri.edu
1042 BRIETZKE
a name given by cardiologists to the phenomenon of angina pectoris associ-

ated with angiographically normal coronary arteries [2], Reaven’s original
name for the syndrome gave way to the metabolic syndrome, and to such
synonyms as ‘‘the insulin resistance syndrome’’ and ‘‘the cardiometabolic
syndrome.’’ Based on data from the 1999 to 2000 National Health and Nu-
trition Examination Survey, the prevalence of the metabolic syndrome in the
United States now exceeds 25% of the adult population, and has increased
since the 1994 National Health and Nutrition Examination Survey [3].
In 2001, aggressive efforts by the American Association of Clinical Endo-
crinologists led to the recognition of the metabolic syndrome as a diagnostic
entity with its own ICD-9 code 277.7. This latter action potentially gave
physicians the ability to diagnose and manage the syndrome as its own entity,
rather than as component diagnoses. The actual impact of having the meta-
bolic syndrome recognized as a medically codable diagnosis seems to be small,
as a 2005 report by Ford [4] attests. His review of two large databases,
National Hospital Discharge Surveys (years 2002 and 2003) and National
Ambulatory Medical Care Survey (year 2002), identified codes for metabolic
syndrome in only 40 of over 400,000 records from the National Hospital
Discharge Surveys, and in only 2 of nearly 29,000 records from the National
Ambulatory Medical Care Survey [4]. Through 2005, although physicians
may have been thinking about the metabolic syndrome in their patients,
they clearly were not using the label diagnosis frequently for billing and coding
purposes.
By contrast, a Medline search in February 2007 identified 724 peer-reviewed
publications retrievable by a key word query ‘‘metabolic syndrome,’’ which was
published in 2002 to 2003. Clearly, although clinicians have been slow to code
patients as ‘‘metabolic syndrome,’’ scholars continue to write about the syn-
drome at a prolific rate. Notably, the journal Diabetes Care named one of its
section headings ‘‘Metabolic Syndrome/Insulin Resistance Syndrome/Pre-
Diabetes’’ in November 2003, an action which then-editor Davidson [5] justified
in an accompanying editorial in which he speculated that treatment of insulin
resistance in the absence of DM would be of benefit to patients. Over the ensuing
2 years, the journal published articles each month under this section heading.
Then, in September 2005, under new editorial leadership Diabetes Care pub-
lished a joint American Diabetes Association/European Association for the
Study of Diabetes (ADA-EASD) position statement calling for a reappraisal
of the metabolic syndrome as a discrete entity [6]. Subsequent to the publication
of that issue, ‘‘metabolic syndrome’’ was dropped as part of a section heading, in
favor of a section heading that was renamed ‘‘Cardiovascular and Metabolic
Risk’’ in January 2006. Despite the reticence of the current editorial staff to em-
brace ‘‘the metabolic syndrome’’ as an entity, the journal continues to publish
articles that use the term ‘‘metabolic syndrome’’ as part of the title.
The ADA-EASD joint statement included a thoughtful and comprehen-
sive review of available literature, and found it wanting with regard to the
following stated concerns: (1) criteria for the diagnosis of the metabolic
CONTROVERSY IN DIAGNOSIS OF THE METABOLIC SYNDROME 1043
syndrome are ‘‘ambiguous or incomplete’’ and ‘‘the rationale for thresholds

are ill-defined’’; (2) the value of including DM in the definition is unclear; (3)
the role of insulin resistance as the cause of the syndrome is ‘‘uncertain’’; (4)
there is no clear basis for including or excluding other risk factors for
cardiovascular disease from the syndrome definition; (5) cardiovascular
disease risk attributable to the metabolic syndrome may be no greater
than that attributable to individual components of the syndrome; and (6)
treatment for the metabolic syndrome is no different than treatment of
component features. Based on these concerns, the position statement ques-
tioned the value of diagnosing the syndrome as ‘‘unclear’’ [6]. There is both
support and partial refutation for each well-stated concern.
In contrast to the advocacy for cautious skepticism by the ADA-EASD,
the cardiology community strongly advocates recognition of the metabolic
syndrome as a predictor of cardiovascular risk. Grundy [7] authored a posi-
tion statement for the American Heart Association and the National Heart,
Lung, and Blood Institute that embraces the metabolic syndrome as a ‘‘con-
stellation of risk factors’’ associated with increased risk for atherosclerotic
disease and type 2 DM, despite having multiple causes. The American Heart
Association and the National Heart, Lung, and Blood Institute joint state-
ment states that ‘‘.recognition of the syndrome in clinical practice is
encouraged for the identification of a multiple-risk-factor condition and
to promote lifestyle therapies that will reduce all of the metabolic risk
factors simultaneously.’’
Blaha and Elasy [8] analyzed the controversy reflected by the divergent
views of the ADA-EASD and American Heart Association and the
National Heart, Lung, and Blood Institute position statements on the
metabolic syndrome as essentially being a matter of perspective. In their
review of 442 publications dealing with the metabolic syndrome, they found
some that used the metabolic syndrome as a study exposure (the clinical
epidemiologic perspective), and others that used it as an outcome (the path-
ophysiologic perspective). In their analysis, they find that the ADA-EASD
position statement aligns with the pathophysiologic perspective, and that
the basis for most of the expressed concerns is the imprecise definition of
the syndrome and incomplete understanding of its pathophysiology. By
contrast, the American Heart Association position statement aligns itself
with the clinical epidemiologic perspective, and finds the recognition of
features of the metabolic syndrome of substantial clinical use in the identi-
fication of patients at high risk for atherosclerotic events.
This article addresses several aspects of the current controversy surround-
ing the metabolic syndrome including (1) definition of the metabolic
syndrome; (2) evidence for and against the use of metabolic syndrome as
a cardiovascular disease risk predictor; (3) evidence as to underlying patho-
physiology; and (4) evidence for treatment of the metabolic syndrome, as
opposed to components of the syndrome, in a risk-reduction strategy, to
prevent type 2 DM or cardiovascular disease.
1044 BRIETZKE
Defining the metabolic syndrome: what is in a name?

Since 1999, four consensus panels have issued recommendations for diag-
nosing the metabolic syndrome (or insulin resistance syndrome), with the
overriding rationale being the recognition of a syndrome of increased risk
for cardiovascular disease, as opposed to a state of increased risk for type
2 DM. The recommendations of these different panels offer similar but
unique syndrome-defining features, summarized as follows.
In its Adult Treatment Panel III (ATP-III) report, the US National
Cholesterol Education Program identified the metabolic syndrome as
a ‘‘multiplex risk factor for cardiovascular disease that is deserving of
more clinical attention’’ [9]. The ATP-III recognized six components of
the metabolic syndrome as contributory to markedly increased cardiovascu-
lar disease risk: (1) abdominal (visceral) obesity; (2) dyslipidemia (hypertri-
glyceridemia with low high-density lipoprotein [HDL] cholesterol and large
fluffy low-density lipoprotein [LDL] cholesterol); (3) elevated blood pres-
sure; (4) insulin resistance (with or without abnormal glucose tolerance);
(5) a proinflammatory state (characterized by elevated highly sensitive C-re-
active protein [hsCRP]); and (6) a prothrombotic state (characterized by
hyperfibrinogenemia and elevated serum levels of plasminogen activator
inhibitor-1 [PAI-1]). Clinical criteria for establishing a diagnosis of the met-
abolic syndrome, as set forth by the ATP-III, were meant to be simple and
pragmatic, and include waist circumference greater than 102 cm (O40 in)
for men, and greater than 88 cm (O35 in) for women; serum triglycerides
greater than or equal to 150 mg/dL; HDL cholesterol less than 40 mg/dL
for men, or less than 50 mg/dL for women; blood pressure greater than
or equal to 130/85 mm Hg; and fasting glucose greater than or equal to
100 mg/dL [10].
In similar fashion to the ATP-III, the World Health Organization recog-
nized the metabolic syndrome as a significant contributor to cardiovascular
morbidity and mortality, but offered a different set of criteria for its diagno-
sis. The major criterion for diagnosis, according to the World Health Orga-
nization panel, is insulin resistance, which can be recognized as either type
2 DM or impaired glucose tolerance, or by glucose uptake below the lowest
quartile for the relevant population, using the hyperinsulinemic euglycemic
clamp technique. In addition to the major criterion of insulin resistance, two
additional criteria must be present to diagnose the metabolic syndrome: (1)
either use of antihypertensive medication or blood pressure greater than or
equal to 140/90 mm Hg, (2) plasma triglycerides greater than or equal to 150
mg/dL, (3) HDL cholesterol less than 35 mg/dL for men or less than 39 mg/
dL for women, (4) body mass index greater than 30 kg/m2 or waist-to-hip
ratio greater than 0.9 in men or greater than 0.85 in women, and (5) urinary
albumin:creatinine ratio greater than or equal to 30 mg/g [11].
The American College of Endocrinology has named a metabolic
syndrome equivalent as ‘‘the insulin resistance syndrome,’’ and offers the
following criteria for its diagnosis: (1) serum triglycerides greater than
150 mg/dL; (2) HDL cholesterol less than 40 mg/dL in men or less than
50 mg/dL in women; (3) blood pressure greater than 130/85 mm Hg; and
(4) impaired fasting glucose (fasting glucose 110–125 mg/dL) and 2-hour
post–oral glucose challenge glucose 140 to 200 mg/dL. The American
College of Endocrinology position statement on the insulin resistance syn-
drome calls for the more widespread use of formal oral glucose tolerance
testing as a case-finding tool [12].
Finally, and most currently, the International Diabetes Foundation has
announced criteria for the diagnosis of the metabolic syndrome. The major
criterion as promulgated by the International Diabetes Foundation is waist
circumference (with normals for men and women representing seven differ-
ent ethnic groups) plus any two of the following additional criteria: (1)
serum triglycerides greater than or equal to 150 mg/dL, or active treatment
for hypertriglyceridemia; (2) HDL cholesterol less than 40 mg/dL for men or
less than 50 mg/dL for women; (3) active antihypertensive treatment or sys-
tolic blood pressure greater than or equal to 130 mm Hg or diastolic blood
pressure greater than or equal to 85 mm Hg; and (4) fasting glucose greater
than or equal to 100 mg/dL, or prior diagnosis of type 2 DM. For fasting
glucose greater than or equal to 100 mg/dL, the International Diabetes
Foundation strongly recommends a 2-hour oral glucose tolerance test, but
does not require it for diagnosis of the metabolic syndrome [13].
Based on the lack of a universally agreed standard for diagnosing (and
even naming) the metabolic syndrome, there seems to be valid concern
about the preciseness of the criteria for defining the metabolic syndrome
as a discrete entity. If a single set of criteria could emerge and be agreed
on by the four consensus panels, it would certainly replace some of the
current confusion with a strengthened case for legitimacy of the metabolic
syndrome as a discrete entity.
Does diagnosing the metabolic syndrome improve the ability to estimate

cardiovascular disease risk?
Abbasi and colleagues [14] identified a strong correlation between nine
‘‘traditional’’ coronary heart disease (CHD) risk factors (age, systolic and
diastolic blood pressure, total cholesterol, triglycerides, HDL cholesterol,
and glucose and insulin response to 75-g oral glucose load) and both
body mass index and steady-state plasma glucose (following intravenous
glucose, insulin, and octreotide infusions; a measure of insulin resistance).
In their study of 314 nondiabetic volunteers, they found that the combina-
tion of greater body mass index and greater insulin resistance (higher steady-
state plasma glucose) predicted higher diastolic blood pressure and
triglycerides and lower HDL cholesterol accurately. By inference, clinical
recognition of an insulin-resistant state (eg, the metabolic syndrome) should
predict a state of increased cardiovascular disease risk.
1046 BRIETZKE
Facchini and colleagues [15] studied 208 individuals who were recruited
on the basis of being nonobese (body mass index !30 kg/m2); had no
hypertension or history of CHD; and had normal glycemic response to
the oral glucose tolerance test. These patients were stratified into three
groups based on their degree of insulin resistance, as assessed by steady state
plasma glucose determined 180 minutes into a constant infusion of glucose,
insulin, and somatostatin. After a mean 6.3 years clinical follow-up, prede-
fined morbid end points of hypertension, type 2 DM, CHD, stroke, and can-
cer were most frequent in individuals in the most insulin-resistant tertile (at
least one end point occurred in 36% of that group), whereas no morbid end
points occurred among members of the least insulin-resistant tertile. Base-
line characteristics associated with insulin resistance in this study included
higher blood pressure (123/77 versus 116/69 mm Hg in the least insulin-
resistant tertile); lower HDL cholesterol (46 versus 58 mg/dL); higher
LDL cholesterol (112 versus 101 mg/dL); and higher triglycerides (128 ver-
sus 73 mg/dL). Based on the study by Facchini and colleagues [15], the clin-
ical use of even higher blood pressure and triglycerides, and lower HDL
cholesterol, to diagnose the metabolic syndrome should identify patients
at even higher risk for these morbid outcomes, and the recognition of the
metabolic syndrome should have clinical value in the identification of indi-
viduals at high risk for type 2 DM, hypertension, and atherosclerotic cardio-
vascular events.
Wilson and colleagues [16] followed a cohort of 3323 men and women for
incident cardiovascular disease (CHD, stroke, new claudication, and
congestive heart failure) and CHD over a period of 8 years, and found
that risk of both cardiovascular disease (relative risk, 2.88 for men and
2.25 for women) and CHD (relative risk, 2.54 for men and 1.54 for women)
was increased among subjects meeting ATP-III criteria for the diagnosis of
metabolic syndrome. Similarly, Sundstrom and colleagues [17] followed a co-
hort of 2322 men in a 32-year longitudinal study, and found that using
either ATP-III or World Health Organization criteria for diagnosing meta-
bolic syndrome identified increased cardiovascular mortality (relative risk,
1.59) in subjects with metabolic syndrome.
In a direct comparison of the performance of Framingham Risk Score
versus metabolic syndrome as predictors of CHD and stroke, Wannamethee
and colleagues [18] followed 5128 men aged 40 to 59 for 20 years, and found
that although the Framingham Score was superior, the presence of meta-
bolic syndrome did predict CHD (relative risk, 1.64) and stroke (relative
risk, 1.61) risk, and concluded that metabolic syndrome was a simple clinical
tool of use in forecasting atherosclerotic event risk. Najarian and colleagues
[19] compared the performance of metabolic syndrome (using ATP-III cri-
teria), type 2 DM, or both as stroke risk predictors in 2097 men and women
in the Framingham Offspring Study. Although the greatest risk was associ-
ated with the presence of both DM and metabolic syndrome (relative risk
for stroke, 3.28 versus both conditions absent), metabolic syndrome was
associated with increased stroke risk (relative risk, 2.10). Lorenzo and
colleagues [20] also compared three different definitions of metabolic syn-
drome (ATP-III, International Diabetes Foundation, and World Health
Organization) with Framingham Risk Score in predicting cardiovascular
events (self-reported myocardial infarction, angina, stroke, or coronary re-
vascularization procedure) among 1088 men and 1471 women in the San
Antonio Heart Study over a period of 7.4 years. They found that regardless
of the metabolic syndrome–defining criteria used, the metabolic syndrome
was a significant predictor of incident cardiovascular disease (odds ratio,
9.55 for International Diabetes Foundation criteria, 9.25 for ATP-III crite-
ria, and 6.47 for World Health Organization criteria) in men over age 45,
comparable with a Framingham Risk Score predicting a 10-year CHD
risk of 10% to 20% (odds ratio for Framingham Score, 11.9). Among
women over age 55, metabolic syndrome performed similarly to a Framing-
ham Score predicting a 10-year CHD risk of 5% to 20%: odds ratios for
CHD were 7.72 for the Framingham Score, 4.40 for metabolic syndrome
by International Diabetes Foundation criteria, 4.98 for metabolic syndrome
by ATP-III criteria, and 5.85 by World Health Organization criteria.
Based on these available data, it can be concluded that recognition of the
metabolic syndrome in patients compliments cardiovascular disease risk
estimation afforded by calculation of the Framingham Risk Score, and
may be easier for busy clinicians, in that no actual calculations are required.
The greatest independent value of diagnosing metabolic syndrome seems to
be for patients not currently expressing type 2 DM, and with no established
history of coronary or peripheral vascular disease.
How clearly defined is the pathophysiology of the metabolic syndrome?

One of the very pragmatic concerns expressed by the ADA-EASD posi-
tion statement is the lack of practical clinical measurement tools to establish
the requisite insulin resistance presumed to be the root cause of the meta-
bolic syndrome. Fasting and postprandial glucose and insulin area-under-
curve analysis varies with laboratory assays for insulin (which are not
well-standardized), and the gold standard research-oriented tools for mea-
suring insulin resistance, such as the euglycemic hyperinsulinemic clamp,
quantitative insulin sensitivity index, and homeostasis model assessment
methods, are cumbersome, time-consuming, and expensive enough to be im-
practical for widespread deployment in clinical settings. Alternative
methods of precisely identifying patients to be labeled with the metabolic
syndrome diagnosis must be sought on the basis of clear understanding of
an underlying pathophysiology. A brief review of current understanding is
pertinent to making an argument that identifiers are or should be ‘‘ready
for prime time’’ soon.
Visceral compartment, as opposed to generalized subcutaneous obesity,
has been associated with deranged lipid and glucose homeostasis for
1048 BRIETZKE
many years [21]. An explosion of information over the past decade now
implicates visceral compartment obesity as likely being the source (and
not simply the expression) of hormonal derangements responsible for the
metabolic syndrome. From that perspective, not only are there common
biochemical features associated with prediabetes, type 2 DM, essential
hypertension, and dyslipidemia, but several interventions associated with
favorable manipulation of adipocyte-derived hormones are associated
with improvements in glycemia, blood pressure, and lipid profiles. In this
evolving new paradigm of metabolic syndrome–related pathophysiology,
the adipocyte has been recast from its prior role as an inert storage bin
for triglyceride, to an active endocrine organ responsible for the generation
of hormones and cytokines (both are often grouped under the heading ‘‘adi-
pokines’’) modulating insulin sensitivity, lipid composition, inflammation,
and endothelial function, with obvious implications for blood pressure
regulation and atherogenesis.
One of the contributors to the development of visceral adiposity may be
the activity of the oxidoreductase 11b-hydroxysteroid dehydrogenase type 1,
which converts inert cortisone into bioactive cortisol. Increased expression
of this enzyme, which is present in visceral but not subcutaneous adipose
tissue in humans, might account for regional hypercortisolism and insulin
resistance within the visceral adipose compartment. Heightened 11b-hy-
droxysteroid dehydrogenase type 1 activity in humans has been associated
with central obesity, insulin resistance, hypertension, and dyslipidemia [22].
Among the adipokines, leptin was first identified, and is now appreciated
as principally useful as a marker for whole-body adipose mass (ie, a serum
measure of fatness), and the most well-established physiologic roles of leptin
seem to be induction of satiety in the central nervous system (to limit
feeding), and to signal adequate nutritional status for reproduction, by per-
mitting gonadotropin-releasing hormone release from the hypothalamus
[23]. In the rare syndrome of generalized lipodystrophy, near-complete
loss of subcutaneous and visceral adipose tissue is associated with near-ab-
sence of leptin. This state is associated with excessive lipid accumulation in
liver and muscle tissue, and with severe insulin resistance [23]. Resistin is
another adipocyte-derived polypeptide, so-named because its expression
induces resistance to the effects of insulin on glucose and lipid metabolism,
and has been offered as a possible link between obesity and type 2 DM [24].
Its actual importance to human glucose homeostasis is not yet established,
and there are inconsistencies in observations derived from animal models
and human experiments. Quantitatively, the adipokine most prevalent in
human serum is adiponectin. Known actions of adiponectin include activa-
tion of the AMP kinase in the liver, which impedes gluconeogenesis [22]; the
same action in muscle tissue promotes GLUT-4–mediated glucose uptake
and oxidation by myocytes [25]. Adiponectin also lowers circulating free
fatty acids by augmenting free fatty acid oxidation in muscle tissue [23],
an effect that may be dependent on the activation of the peroxisome
proliferator-activated receptor-a [25]. Adiponectin inhibits tumor necrosis

factor (TNF)-a and interleukin (IL)-6 production within adipose tissue;
importantly, TNF-a and IL-6 [26] have a reciprocal effect to inhibit adipo-
nectin production. Adipose tissue accounts for approximately 30% of circu-
lating IL-6 [27], and IL-6 has been shown to induce a state of whole-body
insulin resistance, which experimentally can be restored to normal by
administration of IL-6 neutralizing antibodies [23]. TNF-a, by contrast,
probably induces insulin resistance mainly indirectly through induction of
a generally inflammatory state, because its effects are not overcome by
administration of neutralizing antibody [23,28]. Adipocyte-derived TNF-a
and IL-6 are released into portal circulation, and have been shown to
promote hepatic production of fibrinogen and CRP, augmenting the
systemic prothrombotic and inflammatory states, which are both contribu-
tory to atherosclerosis [25].
Healthy visceral adipose tissue can be thought of as an adiponectin-dom-
inant microenvironment, associated with low adipocyte-derived TNF-a, IL-
6, and PAI-1. In this state, actions of insulin in hepatocytes, myocytes, and
adipocytes is optimal. In states of expanded visceral adiposity (ie, central
obesity), the adiponectin-dominant environment is converted to an adipo-
nectin-poor, inflammatory cytokine-enriched microenvironment, in which
adipocyte-derived TNF-a, IL-6, and PAI-1 replace adiponectin as the
predominant hormonal products. A state of systemic insulin resistance,
inflammation, and prothrombotic tendency is established. The metabolic
‘‘toggle switch’’ between the adiponectin-rich and inflammatory-prothrom-
botic cytokine-rich state seems to involve the intracellular signal pathway
nuclear factor-kappa beta in endothelial tissue. Translocation of nuclear
factor-kappa beta from cytosol to nucleus amplifies gene transcription of
TNF-a, IL-6, matrix metalloproteinases, and cellular adhesion molecules
by endothelial cells, and results in increased generation of reactive oxygen
species. In the aggregate, the stage is set for heightened platelet and inflam-
matory cell adhesion to vessel walls, recruitment of macrophages, and
increased oxidative stress, increasing the potential for atherosclerotic plaque
formation and rupture. Nuclear factor-kappa beta translocation to the cell
nucleus is facilitated by its dissociation from an inhibitory peptide, Ikb. This
dissociation is mediated by the action of Ikb kinase, which is promoted by
TNF-a, IL-6, and free fatty acids, and inhibited by insulin and adiponectin
[27,29]. When endothelial integrity is disrupted, adiponectin accumulates in
the subendothelial space, suggesting that many of its effects are local as
opposed to systemic [30]. Adiponectin has been shown to increase
endothelial production of nitric oxide, which is a marker of ‘‘healthy’’ endo-
thelial function [31], and to inhibit TNF-a-mediated endothelial synthesis of
intercellular adhesion molecule (ICAM)-1 and vascular cellular adhesion
molecule 1 [32].
Most of the components and adverse outcomes of the metabolic syn-
drome are associated with hypoadiponectinemia. These include obesity
1050 BRIETZKE
[33]; essential hypertension [34,35]; impaired glucose tolerance and prediabe-

tes [36–38]; type 2 DM [23,33,39,40]; dyslipidemia (low HDL cholesterol,
hypertriglyceridemia, and small dense LDL cholesterol particles) [41,42];
acute coronary syndromes, including acute myocardial infarction [43–46];
and stable CHD [37,47]. In each of these syndrome components, insulin
resistance has been demonstrated, and interventions of aerobic exercise,
weight loss [48,49], and therapy with thiazolidinedione (TZD) drugs amelio-
rate insulin resistance, raise serum adiponectin, and lower circulating
markers of inflammation, such as CRP. Because of the reciprocal relation-
ship of plasma adiponectin with hsCRP, TNF-a, IL-6, and PAI-1, it is not
surprising that markers of inflammation are predictive of cardiovascular [50]
and type 2 DM risk [51]. Plasma hsCRP [52] and IL-6 [53] are associated
with future cardiovascular disease, and both have been found inversely
related to serum adiponectin [54]. When CRP, IL-6, and TNF-a have
been measured, their levels in serum have fallen commensurate with the
improvement in insulin resistance [55] and higher serum adiponectin. Adipo-
nectin gene expression has been found reduced in first-degree relatives of
patients with type 2 DM, suggesting a possible role of adiponectin in the
causation of type 2 DM [56], and in nondiabetic subjects, high adiponectin
experimentally predicts a high degree of insulin sensitivity [38,57]. Based on
the strong association of hypoadiponectinemia with disease risk, and its
response to therapies associated with reduced risk of disease, it seems that
measurement of serum adiponectin offers promise as a good, clinically
useful measure of the degree of metabolic syndrome and in monitoring its
response to treatment.
What is the benefit of metabolic syndrome–focused treatment?

One of the principle critiques offered by the joint ADA-EASD position
paper is a putative lack of evidence that treating metabolic syndrome inde-
pendently of DM offers long-term benefits discrete from treatment of DM
itself; appropriate care of the components of the metabolic syndrome may
produce outcomes equivalent to syndrome-based treatment. A counterview
is that there actually are several clinical trials that examine the efficacy of
simple, root cause–oriented lifestyle modifications or pharmacotherapy on
both incident DM and cardiovascular end points. Examination of several
of these trials is pertinent for the clinician interested in advising patients
perceived to be at risk.
Primary prevention of diabetes mellitus

The Diabetes Prevention Program is a multicenter, randomized clinical
trial that enrolled 3234 patients to one of three treatment arms. One arm
was a placebo group; one was a group entered into a lifestyle-modification
program with goals of maintaining at least 150 minutes per week of aerobic
physical activity, and attaining and maintaining a 7% weight loss; and the
third was a group prescribed metformin, 850 mg twice daily, with no
prescribed lifestyle modification. About two thirds of the patients participat-
ing were women, and 45% were of nonwhite ethnicity (African American,
Asian, and Hispanic). Study eligibility was determined by the presence of
impaired glucose tolerance on a 75-g oral glucose tolerance test; the patients
were at known statistically high risk for developing type 2 DM. The study
end point was incident type 2 DM as defined by annual measurement of
fasting glucose and, if fasting glucose was not diagnostic, by plasma glucose
at the 2-hour point on a standard 75-g oral glucose tolerance test. Over
a mean period of follow-up of 2.8 years, incidence of DM was 11 per 100
person-years in the placebo group, 4.8 per 100 person-years in the lifestyle
intervention group, and 7.8 per 100 person-years in the metformin treatment
group. Numbers needed to treat to prevent one case of DM were 6.9 for
lifestyle intervention and 13.9 for metformin [58]. Two studies of similar
design, conducted in Finland [59] and China [60], produced remarkably
similar results with regard to the impact of lifestyle modification on incident
DM. There is a considerable mass of evidence validating an approach of
aggressive implementation of a program of regular low-intensity aerobic
exercise to prevent the onset of type 2 DM. For those unable or unwilling
to exercise regularly, a less effective and more costly approach is to prescribe
metformin. Whether or not metformin plus regular exercise produces
additive benefit with regard to the prevention of DM is entirely speculative
and a trial to examine the question would be of great interest.
The Diabetes Reduction Assessment with Ramipril and Rosiglitazone
Medication trial examined the efficacy of use of a TZD, rosiglitazone, in
preventing type 2 DM. The study enrolled 5269 patients in 21 participating
countries in North America, South America, Europe, India, and Australia.
About 60% were women. Subject eligibility was similar to the Diabetes
Prevention Program, in that either impaired fasting glucose or impaired
glucose tolerance on a 75-g oral glucose tolerance test had to be present at
baseline. Participating patients were counseled as to ‘‘healthy’’ lifestyle
choices, but there was no attempt to enforce dietary and exercise habits. Ran-
domization was to either rosiglitazone, 8 mg daily, or to placebo. At 3 years
postrandomization, 25% of the placebo group had either developed DM or
died, versus 10.6% of the rosiglitazone-treated group (absolute risk reduction,
14.4%; number needed to treat to prevent one case of DM ¼ 6.9 patients).
Mortality and cardiovascular event rates did not differ in the two groups,
except that 0.5% of the rosiglitazone-treated patients developed congestive
heart failure, versus 0.1% in the placebo group (P ¼ .01), consistent with
the known propensity of TZDs to cause fluid retention and occasional conges-
tive heart failure [61]. This trial validates the concept of using a TZD (specif-
ically, rosiglitazone) in the patient at risk for developing type 2 DM (on the
basis of ‘‘metabolic syndrome,’’ ‘‘prediabetes,’’ ‘‘impaired glucose tolerance,’’
in accord with individual preference) to delay or prevent the onset of DM.
1052 BRIETZKE
Nonpharmacologic treatment of hypertension and prehypertension

Two large meta-analyses have documented blood pressure reductions of
4 to 6/2.5 to 3.7 mm Hg in association with regular aerobic exercise [62,63].
In the Seventh Report of the Joint National Committee on Prevention,
Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7),
Chobanian and colleagues [64] recommend weight reduction as needed to
establish normal body mass index of 19 to 25 kg/m2, estimating a reduction
in systolic blood pressure of 5 to 20 mm Hg for every 10 kg weight loss.
Regular aerobic exercise of a recommended minimum equivalence to 30 min-
utes walking at a brisk pace most days of the week is also recommended
with an expected result of 4 to 9 mm Hg reduction in the systolic blood pres-
sure. JNC-7 identified a new category of prehypertension for patients with
blood pressure 120 to 139/80 to 89 mm Hg based on observational rationale
that individuals with blood pressure in this range often have other risk
factors for CHD, and often develop sustained hypertension by age 55 if
not treated. Lifestyle modifications, including diet and regular aerobic exer-
cise with a goal of attaining and maintaining normal body mass index, are
strongly advocated by JNC-7 for management of prehypertension.
Primary and secondary coronary heart disease prevention with statins

Several large clinical trials have shown incontrovertible value of statin
therapy in reducing atherosclerotic event rates in either the primary or sec-
ondary prevention settings in high-risk groups, notably including patients
with type 2 DM. The largest of these trials, the British Heart Protection
Study, randomized 20,536 patients at high risk for events based on either
prior history of coronary or peripheral vascular disease, presence of either
type 1 or type 2 DM, or age greater than or equal to 65 and a history of
treated hypertension, to receive either simvastatin, 40 mg daily, or placebo,
if the total serum cholesterol was at least 135 mg/dL. Combined end points
of fatal and nonfatal coronary events, fatal and nonfatal stroke, and coro-
nary or peripheral revascularization procedures were reduced by 24% in the
simvastatin treatment group (absolute risk reduction was 5.4%; number
needed to treat to prevent a single end point event ¼ 18.5 patients, over a pe-
riod of 6 years). Benefit was conferred on all subgroups analyzed, including
the elderly hypertensives, and the diabetic patients [65].
Another United Kingdom–based study, the Collaborative Atorvastatin
Diabetes Study, randomized 2838 patients aged 40 to 75 in the United King-
dom and Ireland to either placebo or atorvastatin, 10 mg daily. Further study
eligibility requirements included a diagnosis of type 2 DM; baseline LDL
cholesterol less than 160 mg/dL and triglycerides less than 600 mg/dL;
and either retinopathy, an active smoking history, microalbuminuria, or
hypertension. Primary end points were incident acute coronary events, coro-
nary revascularization, or stroke. The study was terminated 2 years earlier
than anticipated based on strongly positive results favoring low-dose
atorvastatin: after a mean 4 years’ follow-up, primary end points occurred in

9% of the placebo-treated patients and in 5.8% of the atorvastatin-treated pa-
tients, for an absolute event reduction of 3.2% (number needed to treat to pre-
vent one end point ¼ 31.2) [66]. Benefit was conferred regardless of the level of
LDL, HDL, or non-HDL cholesterol at the time of randomization.
The Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm
study examined the impact of statin therapy in non-DM patients aged 40
to 79, with hypertension and three other cardiovascular risk factors. Accord-
ingly, many of these patients meet several criteria for metabolic or insulin
resistance syndrome. A total of 10,305 patients were randomized to either
atorvastatin, 10 mg daily, or placebo. Women comprised 18% of the subjects,
and ethnicity was 94% white. Primary end points of the study were nonfatal
myocardial infarction and fatal CHD events. The planned 5-year trial was
terminated at a mean 3.3 years of follow-up because of incontrovertible
benefit favoring the treatment group: 154 end point events occurred in 5137
subjects in the placebo group (3%) versus 100 end point events in 5168 sub-
jects in the atorvastatin group (1.9%), reflecting a 1.1% absolute event reduc-
tion rate over 3.3 years’ follow-up (number needed to treat to prevent one end
point event ¼ 90.9). At baseline, mean serum LDL was 132 mg/dL, HDL was
50 mg/dL, and triglycerides were 150 mg/dL, again attesting to benefit of sta-
tin treatment to high-risk patients with average lipid values [67].
Primary and secondary coronary heart disease prevention with fibrates

The Helsinki Heart Study was a primary CHD prevention trial that
randomized 4081 Finnish men aged 40 to 55 to receive either placebo or
gemfibrozil, a fibrate drug, 600 mg twice daily. Eligible men were selected
on the basis of dyslipidemia; many would probably qualify for a diagnosis
of metabolic syndrome. At the time of randomization, mean serum lipids
were LDL, 189 mg/dL; HDL, 48 mg/dL; and triglycerides, 179 mg/dL. Pri-
mary end points were fatal and nonfatal MI and total cardiovascular death.
Benefit of gemfibrozil was greatest among men with the lowest HDL choles-
terol (!42 mg/dL) and highest triglycerides (O204 mg/dL) (relative risk, 0.6
versus placebo in low-HDL, high-triglyceride subgroup) [68].
The Veterans Affairs Cooperative Studies Program High-Density Lipo-
protein Cholesterol Intervention Trial randomized 2531 patients to examine
the effect of gemfibrozil, 600 mg twice daily, on coronary events in the sec-
ondary prevention setting. Eligibility requirements included age less than 74;
a past history of CHD (which could include prior myocardial infarction,
documented angina, or angiographic evidence of O50% stenosis of one
or more epicardial coronary arteries, or a history of coronary revasculariza-
tion); and serum lipid profile with low HDL (!40 mg/dL) with average
LDL (%140 mg/dL) and triglycerides (%300 mg/dL). Patients (all male)
were randomized to receive either gemfibrozil or placebo, and primary
end points were nonfatal myocardial infarction and coronary death. After
1054 BRIETZKE
a mean 5.1 years of follow-up, an end point occurred in 275 (21.7%) of 1267
patients in the placebo group, and in 219 (17.3%) of 1264 patients in the
gemfibrozil group. The absolute risk reduction of 4.4% favoring gemfibrozil
treatment in this study translates into a number needed to treat (to prevent
one end point) of 23 patients for 5 years. Subgroup analysis of diabetic
patients and nondiabetic patients showed benefit to each subgroup favoring
treatment with the fibrate [69].
Secondary prevention of atherosclerotic events with thiazolidinediones

Another much-publicized TZD trial, the Prospective Pioglitazone Clini-
cal Trial in Macrovascular Events, addressed the important issue of second-
ary prevention of atherosclerotic cardiovascular events in patients already
known to have type 2 DM, with an intervention (TZD therapy) targeting
amelioration of insulin resistance (ie, metabolic syndrome) in these patients.
A total of 5238 patients on treatment for type 2 DM with hemoglobin A1c
greater than 6.5% were randomized either to usual care of DM plus placebo
or pioglitazone titrated to a dose of 45 mg daily. Because this was a second-
ary prevention study, for eligibility subjects further had to have a history of
prior myocardial infarction, stroke, or peripheral arterial disease. Primary
end points were all-cause mortality; nonfatal myocardial infarction (includ-
ing silent myocardial infarction); stroke; acute coronary syndrome; amputa-
tion; or revascularization procedure. Secondary end points were all-cause
mortality; nonfatal myocardial infarction (excluding silent myocardial
infarction); stroke; and cardiovascular death. At 3 years, there was no
statistically significant difference between treatment groups with regard to
primary end points; there was a significant (P ¼ .027) reduced incidence
of secondary end points in pioglitazone-treated patients. For secondary
end points, 358 (13.6%) occurred in 2633 subjects in the placebo group
and 301 (11.5%) occurred in 2605 subjects in the pioglitzaone group [70].
The absolute risk reduction was 2.1% favoring pioglitazone, and the num-
ber needed to treat to prevent one secondary end point was 47.6. The inci-
dence of congestive heart failure requiring hospitalization was 6% in the
pioglitazone group and 4% in the placebo group. The study has been
heavily criticized for the authors’ emphasis on the secondary end point event
reduction and minimization of the lack of efficacy with regard to the pri-
mary end points [71]. Although acknowledging mixed results of the study,
other critiques favor cautious use of pioglitazone in high CHD risk patients
with type 2 DM [72,73]. More work in this area is needed before the ques-
tion can be resolved without ambiguity.
Drug choices in metabolic syndrome

Previous concerns that thiazide diuretics might actually cause DM have
been largely assuaged by recent work [74–76]. Although associated with
increases in serum triglycerides, the overall reduction in cardiovascular
mortality seen in the ALLHAT trial among thiazide users probably justifies
use of these agents among patients with hypertension and at risk for DM
(either on the basis of the metabolic syndrome, or other risk factors) [77].
Other agents may be preferred for early or first-choice use in the patient
with features of the metabolic syndrome who requires treatment for hyperten-
sion. Telmisartan is an angiotensin receptor blocker that seems also to have
TZD-like effects on peroxisome proliferator-activated receptor-g. As com-
pared with other angiotensin receptor blockers, telmisartan lowers plasma
insulin, triglycerides, and CRP, while raising serum adiponectin [78]. Whether
these effects are sustained in long-term clinical use is currently unknown.
In addition to evidence of use in delaying or preventing the onset of type
2 DM, TZD therapy has been shown to have several other salutary effects
on nonglycemic components of the metabolic syndrome. In a trial compar-
ing diet and exercise therapy versus pioglitazone, 30 mg daily, in nondiabetic
obese patients, pioglitazone was associated with a comparable degree of
reduction in insulin resistance as was a 12-kg weight loss [79]. Both troglita-
zone [80] and pioglitazone [81] have been shown to increase LDL particle
size favorably from the small, dense (proatherogenic) toward large, ‘‘fluffy’’
size associated with resistance to oxidation of the LDL particle. There is
evidence that TZDs may also improve blood pressure in hypertensive pa-
tients. In a trial of nondiabetic hypertensive patients treated with their usual
antihypertensive drug therapy plus rosiglitazone, 8 mg daily, the addition of
rosiglitazone was associated with a blood pressure reduction from 138 2/
85 2 mm Hg to 134 2/80 2 mm Hg [82]; pioglitazone has been shown
to have a similar effect [83]. TZDs have been shown to increase plasma adi-
ponectin [83–85]. There is available information from clinical trials to justify
considering use of TZDs in patients with central obesity, dyslipidemia, and
hypertension; further work is needed before TZD treatment can be consid-
ered ‘‘usual care’’ in regard to the prediabetic or nondiabetic patient with
other features of the metabolic syndrome.
Metabolic syndrome–tailored treatment recommendations

Diet and exercise therapy has proved efficacy in the prevention of DM,
possibly lowers cardiovascular risk independent of its effect on DM risk,
is associated with lower systolic and diastolic blood pressure, and favorably
influences several components of the metabolic syndrome. Plasma concen-
tration of the cellular adhesion molecule ICAM-1 has been found highly
predictive of the risk for future myocardial infarction in the Physicians’
Health Study [86], and circulating ICAM-1 can be reduced through weight
reduction [87]. Based on safety, low cost, and multiple health benefits, diet
and exercise can be recommended enthusiastically to most patients as a start-
ing point.
With regard to drug therapy, metformin has been shown in a large ran-
domized controlled trial (the Diabetes Prevention Program) to be associated
1056 BRIETZKE
with reduced incidence of type 2 DM, although inferior to diet and exercise
in that regard. The salutary effects of TZDs on glycemia (reduced risk of
DM in the Diabetes Reduction Assessment with Ramipril and Rosiglitazone
Medication trial), blood pressure, and lipid composition make an agent of
this pharmacologic class a tempting option for initial monotherapy in the
patient with prediabetes (fasting glucose 100–125 mg/dL); prehypertension
(eg, blood pressure 135/84 mm Hg); and mild dyslipidemia (eg, male patient
with HDL cholesterol 39 mg/dL and triglycerides 164 mg/dL). Buffering
that enthusiasm is the high economic cost of TZDs and their association
with long-term weight gain. Further evidence from clinical trials is awaited
with much interest so that the actual long-term risk versus benefit trade-offs
can be well characterized for TZDs in a large-scale use strategy.
Consideration of either a statin drug or a fibrate drug for treatment of
dyslipidemia (low HDL cholesterol and triglycerides O150 mg/dL) or aver-
age LDL cholesterol (O130 mg/dL) is probably justified in nondiabetic
patients with several features of the metabolic syndrome. Combining statin
drugs with diet has been shown to reduce CRP, ICAM-1, and fibrinogen
[88]. Statins tend to reduce circulating markers of inflammation [89], and
may be associated with a reduced risk of developing type 2 DM [90]. Feno-
fibrate has been shown to improve insulin sensitivity and lower hsCRP, and
lower triglycerides in treated patients [91].
Among the many choices of antihypertensive agents for patients with
metabolic syndrome, new information suggests a preferential role for angio-
tensin-receptor blockers. The angiotensin-receptor blocker candasartan has
been associated with reduced circulating ICAM-1 and improved endothelial
function, as represented by reactive-hyperemia-induced forearm blood flow
in patients with hypercholesterolemia [92]. Irbesartan and candasartan (but
not losartan) have been shown to reduce plasma PAI-1 [93].
Metabolic syndrome: a look toward the future

Far more work remains to be done to unravel the tangled web of patho-
physiology responsible for the metabolic syndrome. A speculative view is
that perhaps the future lies beyond insulin resistance, which is to say, rather
than being the ‘‘hub,’’ insulin resistance may turn out to be yet another
‘‘spoke’’ in the wheel of metabolic syndrome. Perhaps the hub will prove
to be the visceral adipocyte as the source of inflammatory, dysglycemia-pro-
moting adipokines (TNF-a, IL-6, resistin) and free fatty acids, and subnor-
mal adiponectin. Each of these findings have been associated with insulin
resistance, endothelial dysfunction, hypercoagulability, hypertension, and
increased cardiovascular disease risk. Although it is premature to change
the working definition of the metabolic syndrome, it is possible that in the
years ahead the syndrome may be redefined as a state of excess visceral
adiposity (possible names for the ‘‘new’’ syndrome might be ‘‘the adipo-in-
flammatory syndrome’’ or ‘‘the toxic fat syndrome’’) characterized by low
adiponectin and elevated adipocyte-derived inflammatory peptides responsi-

ble for metabolic derangements (impaired glucose tolerance, dyslipidemia),
endothelial dysfunction (hypertension), and hypercoagulability, all of which
conspire to produce a state of heightened risk for atherosclerotic disease.
The syndrome will come of age not only when the root cause is identified,
but when treatment targeting the root cause ameliorates the key component
manifestations of DM, hypertension, and dyslipidemia. Until then, whether
the metabolic syndrome is known by one name or by many, clinicians are
best advised to treat those individual components aggressively with both
lifestyle management (diet and exercise) and rational pharmacotherapy
(metformin, TZDs, angiotensin-converting enzyme inhibitors, angiotensin
receptor blockers, statins, and so forth), being especially mindful of those
therapies that favorably modulate several individual syndrome components
concurrently.
References
[1] Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease [review][71
refs]. Diabetes 1988;37(12):1595–607.
[2] Kaski JC, Elliott PM. Angina pectoris and normal coronary arteriograms: clinical presenta-
tion and hemodynamic characteristics [review][48 refs]. Am J Cardiol 1995;76(13):35D–42D.
[3] Ford ES, Giles WH, Mokdad AH. Increasing prevalence of the metabolic syndrome among
U.S. adults. Diabetes Care 2004;27:2444–9.
[4] Ford ES. Rarer than a blue moon: the use of a diagnostic code for the metabolic syndrome in
the U.S. Diabetes Care 2005;28:1808–9.
[5] Davidson MB. Is treatment of insulin resistance beneficial independent of glycemia? Diabe-
tes Care 2003;26:3184–6.
the Study of Diabetes. Diabetes Care 2005;28:2289–304.
[7] Grundy SM. Metabolic syndrome scientific statement by the American Heart Association
and the National Heart, Lung, and Blood Institute. Arterioscler Thromb Vasc Biol 2005;
25:2243–4.
[8] Blaha M, Elasy TA. Clinical use of the metabolic syndrome: why the confusion? Clin Dia-
betes 2006;24:125–31.
[9] Grundy SM, Brewer HB Jr, Cleeman JI, et al. Definition of metabolic syndrome: report of
the National Heart, Lung, and Blood Institute/American Heart Association conference on
scientific issues related to definition [see comment][review][14 refs]. Circulation 2004;
109(3):433–8.
[10] Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on de-
tection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment
Panel III) final report. Circulation 2002;106:3143–421.
complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of
a WHO consultation [see comment]. Diabet Med 1998;15(7):539–53.
[12] Einhorn D, Reaven GM, Cobin RH, et al. American College of Endocrinology position
statement on the insulin resistance syndrome [review][0 refs]. Endocr Pract 2003;9(3):237–52.
[13] Alberti KG, Zimmet P, Shaw J. Metabolic syndromeda new world-wide definition. A Con-
sensus Statement from the International Diabetes Federation [review][80 refs]. Diabet Med
2006;23(5):469–80.
1058 BRIETZKE
[14] Abbasi F, Brown BW, Lamendola C, et al. Relationship between obesity, insulin resistance,
and coronary heart disease risk. J Am Coll Cardiol 2002;40:937–43.
[15] Facchini FS, Hua N, Abbasi F, et al. Insulin resistance as a predictor of age-related diseases.
J Clin Endocrinol Metab 2001;86:3574–8.
[16] Wilson PWF, D’Agostino RB, Parise H, et al. Metabolic syndrome as a precursor of cardio-
vascular disease and type 2 diabetes mellitus. Circulation 2005;112:3066–72.
[17] Sundstrom J, Riserus U, Byberg L, et al. Clinical value of the metabolic syndrome for long
term prediction of total and cardiovascular mortality: prospective, population based cohort
study. BMJ 2006;332:878–82.
[18] Wannamethee SGP, Shaper AGF, Lennon LM, et al. Metabolic syndrome vs Framingham
Risk Score for prediction of coronary heart disease, stroke, and type 2 diabetes mellitus [ar-
ticle]. Arch Intern Med 2005;165:2644–50.
[19] Najarian RMM, Sullivan LMP, Kannel WBM, et al. Metabolic syndrome compared with
type 2 diabetes mellitus as a risk factor for stroke: the Framingham Offspring Study [article].
Arch Intern Med 2006;166:106–11.
[20] Lorenzo C, Williams K, Hunt KJ, et al. The National Cholesterol Education Program-Adult
Treatment Panel III, International Diabetes Federation, and World Health Organization
definitions of the metabolic syndrome as predictors of incident cardiovascular disease and
diabetes. Diabetes Care 2007;30:8–13.
[21] Fujioka S, Matsuzawa Y, Tokunaga K, et al. Contribution of intra-abdominal fat accumu-
lation to the impairment of glucose and lipid metabolism in human obesity. Metabolism
1987;36:54–9.
[22] Ahima RS. Adipose tissue as an endocrine organ. Obesity 2006;14:242S–9S.
[23] Pittas AG, Joseph NA, Greenberg AS. Adipocytokines and insulin resistance. J Clin Endo-
crinol Metab 2004;89:447–52.
[24] Steppan CM, Bailey ST, Bhat S, et al. The hormone resistin links obesity to diabetes. Nature
2001;409:307–12.
[25] Gable DR, Hurel SJ, Humphries SE. Adiponectin and its gene variants as risk factors for
insulin resistance, the metabolic syndrome and cardiovascular disease. Atherosclerosis
2006;188:231–44.
[26] Fasshauer M, Kralisch S, Klier M, et al. Adiponectin gene expression and secretion is inhibited by
interleukin-6 in 3T3-L1 adipocytes. Biochem Biophys Res Commun 2003;301:1045–50.
[27] Tataranni PA, Ortega E. A burning question: does an adipokine-induced activation of the
immune system mediate the effect of overnutrition on type 2 diabetes? Diabetes 2005;54:
917–27.
[28] Koh KK, Han SH, Quon MJ. Inflammatory markers and the metabolic syndrome: insights
from therapeutic interventions [review][84 refs]. J Am Coll Cardiol 2005;46(11):1978–85.
[29] Ouchi N, Kihara S, Arita Y, et al. Adiponectin, an adipocyte-derived plasma protein, in-
hibits endothelial NF-{kappa}B signaling through a cAMP-dependent pathway. Circulation
2000;102:1296–301.
[30] Matsuzawa Y. Therapy insight: adipocytokines in metabolic syndrome and related cardio-
vascular disease [review][57 refs]. Nat Clin Pract Cardiovasc Med 2006;3(1):35–42.
[31] Chen H, Montagnani M, Funahashi T, et al. Adiponectin stimulates production of nitric
oxide in vascular endothelial cells. J Biol Chem 2003;278:45021–6.
[32] Ouchi N, Kihara S, Arita Y, et al. Novel modulator for endothelial adhesion molecules:
adipocyte-derived plasma protein adiponectin. Circulation 1999;100:2473–6.
[33] Weyer C, Funahashi T, Tanaka S, et al. Hypoadiponectinemia in obesity and type 2 diabetes:
close association with insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab
2001;86:1930–5.
[34] Iwashima Y, Katsuya T, Ishikawa K, et al. Hypoadiponectinemia is an independent risk fac-
tor for hypertension. Hypertension 2004;43:1318–23.
[35] Adamczak M, Wiecek A, Funahashi T, et al. Decreased plasma adiponectin concentration in
patients with essential hypertension. Am J Hypertens 2003;16:72–5.
[36] Choi KM, Lee J, Lee KW, et al. The associations between plasma adiponectin, ghrelin levels
and cardiovascular risk factors. Eur J Endocrinol 2004;150:715–8.
[37] Higashiura K, Ura N, Ohata J, et al. Correlations of adiponectin level with insulin resistance
and atherosclerosis in Japanese male populations. Clin Endocrinol 2004;61:753–9.
[38] Tschritter O, Fritsche A, Thamer C, et al. Plasma adiponectin concentrations predict insulin
sensitivity of both glucose and lipid metabolism. Diabetes 2003;52:239–43.
[39] Lindsay RS, Funahashi T, Hanson RL, et al. Adiponectin and development of type 2 diabe-
tes in the Pima Indian population. Lancet 2002;360:57–8.
[40] Daimon M, Oizumi T, Saitoh T, et al. Decreased serum levels of adiponectin are a risk factor
for the progression to type 2 diabetes in the Japanese population: the Funagata Study.
Diabetes Care 2003;26:2015–20.
[41] Hulthe J, Hulten LM, Fagerberg B. Low adipocyte-derived plasma protein adiponectin con-
centrations are associated with the metabolic syndrome and small dense low-density lipopro-
tein particles: atherosclerosis and insulin resistance study. Metabolism 2003;52:1612–4.
[42] Matsubara M, Maruoka S, Katayose S. Decreased plasma adiponectin concentrations in
women with dyslipidemia. J Clin Endocrinol Metab 2002;87:2764–9.
[43] Pischon TM, Girman CJD, Hotamisligil GSMP, et al. Plasma adiponectin levels and risk of
myocardial infarction in men [article]. JAMA 2004;291:1730–7.
[44] Kumada M, Kihara S, Sumitsuji S, et al. Association of hypoadiponectinemia with coronary
artery disease in men. Arterioscler Thromb Vasc Biol 2003;23:85–9.
[45] Zoccali C, Mallamaci F, Tripepi G, et al. Adiponectin, metabolic risk factors, and cardiovas-
cular events among patients with end-stage renal disease. J Am Soc Nephrol 2002;13:134–41.
[46] Schulze MB, Shai I, Rimm EB, et al. Adiponectin and future coronary heart disease events
among men with type 2 diabetes. Diabetes 2005;54:534–9.
[47] Okamoto Y, Kihara S, Funahashi T, et al. Adiponectin: a key adipocytokine in metabolic
syndrome. Clin Sci 2006;110:267–78.
[48] Esposito KM, Pontillo AM, Di Palo C, et al. Effect of weight loss and lifestyle changes on
vascular inflammatory markers in obese women: a randomized trial [article]. JAMA 2003;
289:1799–804.
[49] Yang WS, Lee WJ, Funahashi T, et al. Weight reduction increases plasma levels of an adi-
pose-derived anti-inflammatory protein, adiponectin. J Clin Endocrinol Metab 2001;86:
3815–9.
[50] Duncan BB, Schmidt MI, Pankow JS, et al. Low-grade systemic inflammation and the de-
velopment of type 2 diabetes: the atherosclerosis risk in communities study. Diabetes
2003;52:1799–805.
[51] Hu FB, Meigs JB, Li TY, et al. Inflammatory markers and risk of developing type 2 diabetes
in women. Diabetes 2004;53:693–700.
[52] Torres JL, Ridker PM. Clinical use of high sensitivity C-reactive protein for the prediction of
adverse cardiovascular events [Miscellaneous article]. Curr Opin Cardiol 2003;18:471–8.
[53] Ridker PM, Rifai N, Stampfer MJ, et al. Plasma concentration of interleukin-6 and the risk
of future myocardial infarction among apparently healthy men. Circulation 2000;101:
1767–72.
[54] Ouchi N, Kihara S, Funahashi T, et al. Reciprocal association of C-reactive protein with adi-
ponectin in blood stream and adipose tissue. Circulation 2003;107:671–4.
[55] Dandona P, Weinstock R, Thusu K, et al. Tumor necrosis factor-{alpha} in sera of obese
patients: fall with weight loss. J Clin Endocrinol Metab 1998;83:2907–10.
[56] Lihn AS, Ostergard T, Nyholm B, et al. Adiponectin expression in adipose tissue is reduced
in first-degree relatives of type 2 diabetic patients. Am J Physiol Endocrinol Metab 2003;284:
E443–8.
[57] Stefan N, Vozarova B, Funahashi T, et al. Plasma adiponectin concentration is associated
with skeletal muscle insulin receptor tyrosine phosphorylation, and low plasma concentra-
tion precedes a decrease in whole-body insulin sensitivity in humans. Diabetes 2002;51:
1884–8.
1060 BRIETZKE
[58] Diabetes Prevention Program Research Group. Reduction in the incidence of type 2 diabetes
with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403.
[59] Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by
changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;
344:1343–50.
[60] Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people
with impaired glucose tolerance. The Da Qing IGT and diabetes study. Diabetes Care 1997;
20:537–44.
[61] DREAM Trial Investigators. Effect of rosiglitazone on the frequency of diabetes in patients
with impaired glucose tolerance or impaired fasting glucose: a randomised controlled trial.
Lancet 2006;368:1096–105.
[62] Whelton SP, Chin A, Xin X, et al. Effect of aerobic exercise on blood pressure: a meta-anal-
ysis of randomized, controlled trials [see comment][summary for patients in Ann Intern Med.
2002 Apr 2;136(7):I16; PMID: 11926806]. Ann Intern Med 2002;136(7):493–503.
[63] Kelley GA, Kelley KS, Tran ZV. Walking and resting blood pressure in adults: a meta-anal-
ysis. Prev Med 2001;33(2 Pt 1):120–7.
[64] Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee
on prevention, detection, evaluation, and treatment of high blood pressure [see comment].
Hypertension 2003;42(6):1206–52.
[65] Heart Protection Study Collaborative Group. MRC/BHF heart protection study of choles-
terol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-con-
trolled trial. Lancet 2002;360:7–22.
[66] Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary prevention of cardiovascular
disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study
(CARDS): multicentre randomised placebo-controlled trial. Lancet 2004;364:685–96.
[67] Sever PS, Dahlof B, Poulter NR, et al. Prevention of coronary and stroke events with ator-
vastatin in hypertensive patients who have average or lower-than-average cholesterol con-
centrations, in the Anglo-Scandinavian Cardiac Outcomes TrialdLipid Lowering Arm
(ASCOT-LLA): a multicentre randomised controlled trial [see comment]. Lancet 2003;
361(9364):1149–58.
[68] Manninen V, Tenkanen L, Koskinen P, et al. Joint effects of serum triglyceride and LDL
cholesterol and HDL cholesterol concentrations on coronary heart disease risk in the Hel-
sinki heart study: implications for treatment [see comment]. Circulation 1992;85(1):37–45.
[69] Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the secondary prevention of coro-
nary heart disease in men with low levels of high-density lipoprotein cholesterol. Veterans
Affairs High-Density Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J
Med 1999;341(6):410–8.
[70] Dormandy JA, Charbonnel B, Eckland DJA, et al. Secondary prevention of macrovascular
events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone
Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 2005;366:
1279–89.
[71] Freemantle N. How well does the evidence on pioglitazone back up researchers’ claims for
a reduction in macrovascular events? [Miscellaneous] BMJ 2005;331:836–8.
[72] Fonseca V, Jawa A, Asnani S. The PROactive Study: the glass is half full. J Clin Endocrinol
Metab 2006;91:25–7.
[73] Rizza R, Henry R, Kahn R. Commentary on the results and clinical implications of the PRO-
active study. Diabetes Care 2005;28:2965–7.
[74] Gress TW, Nieto FJ, Shahar E, et al. The atherosclerosis risk in communities study: hyper-
tension and antihypertensive therapy as risk factors for type 2 diabetes mellitus. N Engl J
Med 2000;342:905–12.
[75] Gurwitz JH, Bohn RL, Glynn RJ, et al. Antihypertensive drug therapy and the initiation of
treatment for diabetes mellitus. Ann Intern Med 1993;118:273–8.
[76] Savage PJ, Pressel SL, Curb JD, et al. Influence of long-term, low-dose, diuretic-based, an-
tihypertensive therapy on glucose, lipid, uric acid, and potassium levels in older men and
women with isolated systolic hypertension: the systolic hypertension in the elderly program.
Arch Intern Med 1998;158:741–51.
[77] ALLHAT Officers and Coordinators for the 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) [see comment][erratum
appears in JAMA 2003 Jan 8;289(2):178]. JAMA 2002;288(23):2981–97.
[78] Kurtz TW. New treatment strategies for patients with hypertension and insulin resistance.
Am J Med 2006;119:S24–30.
[79] Shadid S, Jensen MD. Effects of pioglitazone versus diet and exercise on metabolic health
and fat distribution in upper body obesity. Diabetes Care 2003;26:3148–52.
[80] Tack CJ, Smits P, Demacker PN, et al. Troglitazone decreases the proportion of small, dense
LDL and increases the resistance of LDL to oxidation in obese subjects. Diabetes Care 1998;
21:796–9.
[81] Winkler K, Konrad T, Fullert S, et al. Pioglitazone reduces atherogenic dense LDL particles
in nondiabetic patients with arterial hypertension: a double-blind, placebo-controlled study.
[82] Raji A, Seely EW, Bekins SA, et al. Rosiglitazone improves insulin sensitivity and lowers
blood pressure in hypertensive patients. Diabetes Care 2003;26:172–8.
[83] Hirose H, Kawai T, Yamamoto Y, et al. Effects of pioglitazone on metabolic parameters,
body fat distribution, and serum adiponectin levels in Japanese male patients with type 2
diabetes. Metabolism 2002;51:314–7.
[84] Maeda N, Takahashi M, Funahashi T, et al. PPAR{gamma} ligands increase expression
and plasma concentrations of adiponectin, an adipose-derived protein. Diabetes 2001;50:
2094–9.
[85] Yu JG, Javorschi S, Hevener AL, et al. The effect of thiazolidinediones on plasma adiponec-
tin levels in normal, obese, and type 2 diabetic subjects. Diabetes 2002;51:2968–74.
[86] Ridker PM, Hennekens CH, Roitman-Johnson B, et al. Plasma concentration of soluble in-
tercellular adhesion molecule 1 and risks of future myocardial infarction in apparently
healthy men. Lancet 1998;351:88–92.
[87] Ito H, Ohshima A, Inoue M, et al. Weight reduction decreases soluble cellular adhesion mol-
ecules in obese women. Clin Exp Pharmacol Physiol 2002;29:399–404.
[88] Koh KK, Son JW, Ahn JY, et al. Vascular effects of diet and statin in hypercholesterolemic
patients. Int J Cardiol 2004;95:185–91.
[89] Weitz-Schmidt G. Statins as anti-inflammatory agents. Trends Pharmacol Sci 2002;23:
482–7.
[90] Freeman DJ, Norrie J, Sattar N, et al. Pravastatin and the development of diabetes mellitus:
evidence for a protective treatment effect in the West of Scotland coronary prevention study.
Circulation 2001;103:357–62.
[91] Koh KK, Han SH, Quon MJ, et al. Beneficial effects of fenofibrate to improve endothelial
dysfunction and raise adiponectin levels in patients with primary hypertriglyceridemia.
[92] Wassmann S, Hilgers S, Laufs U, et al. Angiotensin II type 1 receptor antagonism improves
hypercholesterolemia-associated endothelial dysfunction. Arterioscler Thromb Vasc Biol
2002;22:1208–12.
[93] Koh KK, Han SH, Chung WJ, et al. Comparison of effects of losartan, irbesartan, and can-
desartan on flow-mediated brachial artery dilation and on inflammatory and thrombolytic
markers in patients with systemic hypertension. Am J Cardiol 2004;93:1432–5.
Med Clin N Am 91 (2007) 1063–1077
Insulin Resistance as the Underlying

Cause for the Metabolic Syndrome
Danielle Lann, MD, Derek LeRoith, MD, PhD*
Division of Endocrinology, Diabetes, and Bone Diseases,
Department of Medicine, Mount Sinai School of Medicine, One Gustave L. Levy Place,
Box 1055, New York, NY 10029-6574, USA
The metabolic syndrome comprises multiple features, including visceral

obesity, hypertension, dyslipidemia, and impaired glucose tolerance. This
constellation of conditions has also become synonymous with insulin resis-
tance syndrome, which may be a more appropriate term, as insulin resistance
is likely a primary link between the components of the metabolic syndrome [1].
Insulin is an anabolic hormone that has the primary action of facilitating glu-
cose uptake in skeletal and myocardial myocytes, hepatocytes, and adipo-
cytes. In addition, insulin also regulates lipolysis and hepatic glucose
production. Insulin resistance can be defined as a condition of decreased re-
sponsiveness of these target tissues to normal levels of circulating insulin,
that is, a state of decreased insulin sensitivity [2]. Insulin resistance arises
from both genetically determined and acquired metabolic defects [3–5]. This
article outlines how abnormal insulin signaling and secretion, impaired glu-
cose disposal, lipotoxicity, and pro-inflammatory cytokines exacerbate insulin
resistance and result in the perturbations of the metabolic syndrome (Fig. 1).
Defective insulin secretion and signaling

Normally, elevated glucose levels stimulate pancreatic b cells to secrete
insulin and decrease glucagon production. This leads to suppression of he-
patic glucose production and increased glucose uptake in muscle, liver, and
adipose tissues. In the state of insulin resistance, b cell dysfunction occurs,
manifesting as a loss of first phase insulin secretion or the lack of immediate
Dr. Derek LeRoith is presently a consultant and speaker, and receives an honorarium
from Merck, Sanofi-Aventis, Pfizer, Takeda, and Novo Nordisk.
* Corresponding author.
E-mail address: derek.leroith@mssm.edu (D. LeRoith).
1064 LANN & LEROITH
Metabolic Syndrome
Defects in IR Signaling
Hypertension
Abnormal Insulin Secretion INSULIN Visceral Obesity
Impaired Glucose Disposal
Lipotoxicity RESISITANCE Dyslipidemia ( TG HDL)
Inflammatory Cytokines Impaired Glucose Tolerance
Fig. 1. The metabolic syndrome. The metabolic syndrome is characterized by visceral obesity,
dyslipidemia, elevated triglycerides (TG), low high-density lipoprotein (HDL), hypertension,
and impaired glucose tolerance. The perturbations of the metabolic syndrome are the result
of abnormal insulin signaling and secretion, impaired glucose disposal, lipotoxicity, and proin-
flammatory cytokines which contribute to and exacerbate the insulin resistant state, where IR is
the insulin receptor.
release of insulin in response to a glucose load [3]. This deficiency of acute

insulin secretion then leads to postprandial hyperglycemia, and to correct
this glucose excess an exaggerated second phase insulin response occurs.
The chronic hyperinsulinemia that ensues down-regulates insulin receptors,
thereby impairing insulin action [6,7]. In fact, there have been experiments
with transgenic mice demonstrating this phenomenon [8]. These mice
overexpress multiple copies of the human insulin gene leading to a two-
to four-fold increase in serum insulin levels. Despite the extra insulin in
the circulation, the mice are hyperglycemic. This study shows that the hyper-
insulinemia in these mice leads to an insulin resistant state secondary to
down-regulation of the insulin receptor. Therefore, hyperinsulinemia can
be viewed as both the cause and the effect of insulin resistance under
different circumstances.
The insulin receptor (IR) complex contains two a-subunits, capable of
binding insulin, and two b-subunits that possess intrinsic tyrosine kinase ac-
tivity. After insulin binding, the IR undergoes autophosphorylation activat-
ing its kinase, which then allows for the phosphorylation of tyrosine residues
on downstream signaling substrates, such as insulin receptor substrates
(IRS-1, IRS-2) and Shc. This leads to activation of the phosphatidylinositol
3-kinase (PI3K) pathway, which regulates metabolic functions, and the mi-
togen-activated protein kinase (MAPK) pathway, which mediates growth
and mitogenesis. Eventually, pathways that impact glucose and lipid metab-
olism and cellular growth and differentiation become stimulated (Fig. 2) [9].
Insulin resistance can be genetically determined. Impaired insulin signaling
can result from mutations in the genes that encode the IR and its downstream
molecules. For example, there is a case report about a family with an
autosomal dominant inherited defect in the gene encoding Akt2, a distal effec-
tor protein in the insulin receptor signaling cascade, which causes severe
insulin resistance [10]. In this family, those affected by the mutation are
nonobese individuals with profound insulin resistance, who develop diabetes
at early ages. In addition to genetic aberrations, negative effectors of insulin
UNDERLYING CAUSE FOR THE METABOLIC SYNDROME 1065
Insulin Receptor
Shc
IRS
IRS
PI3K MAPK
pathway pathway
PI3-kinase MAP-kinase
Akt
ER K
GLUT4
Translocation NO production
ET-1 Growth
and
Mitogenesis
Glucose Uptake Vasodilation Vasoconstriction
Fig. 2. Insulin signaling transduction. ET-1, endolethin-1; GLUT, glucose transporters. (Data
from Kim JA, Montagnani M, Koh KK, et al. Reciprocal relationships between insulin
resistance and endothelial dysfunction: molecular and pathophysiological mechanisms. Circula-
tion 2006;113:1888–904).
action include agents that cause phosphorylation of serine or threonine

residues on either the IR or its substrates [11]. This reduces IR kinase activity
and its ability to effectively phosphorylate tyrosine residues on the down-
stream signaling proteins, inducing insulin resistance. Previous investigators
have shown that incubation of cells with agents that augment serine/threonine
phosphorylation, such as tumor necrosis factor alpha (TNF-a), ceramide, or
sphingomyelinase, causes reduced insulin-induced tyrosine phosphorylation
of the most proximal signaling substrates [12]. The IRS family of signaling
molecules consists of subtypes that seem to have tissue specific roles when
mediating insulin’s actions, with IRS-1 working primarily in skeletal muscle
and IRS-2 in the liver [13]. Knockout mice with mutations in IRS-1 develop
severe insulin resistance in skeletal muscle, and those with mutations in
IRS-2 develop severe insulin resistance in the liver (Fig. 3).
Dysregulation of glucose disposal and production

Glucose transport into cells is mediated by numerous glucose
transporters (GLUT) and sodium-glucose cotransporters [14]. One of the
1066 LANN & LEROITH
INSULIN
INSULIN RECEPTOR
Agents that cause

pY Serine/Threonine
Phosphorylation
pY IRS pY
pY pY
pY
pS
IRS
pS
pS pS
Normal
Insulin Signaling
INSULIN
RESISTANCE
Fig. 3. Insulin resistance: a proposed cellular mechanism. Insulin binding induces autophos-
phorylation of the IR which then leads to tyrosine phosphorylation of its downstream signaling
intermediates and normal insulin signaling. Agents such as free fatty acids (FFA), TNF-a,
ceramide, or sphingomyelinase cause phosphorylation of serine or threonine residues on either
the IR or its substrates. This reduces IR kinase activity and its ability to effectively phosphor-
ylate tyrosine residues on the downstream signaling proteins, resulting in abrogated insulin
signaling and insulin resistance.
most important glucose transporters, GLUT4, is regulated by insulin. In re-

sponse to insulin, GLUT4 is mobilized from intracellular storage vesicles
and fuses to the cellular membrane to internalize glucose [15]. This process
is mediated by PI3K and TC 10 pathways [16]. Clinical studies completed by
Rothman and colleagues [17] and Cline and colleagues [18] reveal that the
glucose transporter GLUT4 is the major rate controlling step in insulin me-
diated glucose uptake and muscle glycogen synthesis. In parallel, animal
studies on mice with targeted disruption of the GLUT4 gene solely in muscle
have shown that this disturbance leads to reduced glucose transport, insulin
resistance, and glucose intolerance [19]. Taken together, these data suggest
that therapies targeting glucose transport should be considered to improve
insulin sensitivity. Some of these therapeutic options are already available.
Included in their multifaceted effects, thiazolidinediones (TZDs) are potent
insulin sensitizers that facilitate glucose transport [20,21]. Treatment of hu-
man skeletal myotubes with metformin results in increased mRNA expres-
sion of GLUT4 [22]. In addition, metformin treated cardiomyocytes have
GLUT4 that reside longer in the cellular membrane because of decreased
endocytosis of the transporter, the result of activation of AMP kinase [23].

Moreover, physical exercise also enhances the expression and translocation
of GLUT4 and improves insulin sensitivity [24,25].
Insulin resistance is also accompanied by increased endogenous glucose
production. The liver is the major site of gluconeogenesis and insulin is known
to suppress gluconeogenic enzymes. Hepatic insulin resistance results in the
inability to suppress enzymes involved in gluconeogenesis [26]. Animal studies
show that genetic disruption of the PI3K pathway decreases Akt phosphory-
lation, which leads to increased gluconeogenesis, impaired glucose metabo-
lism, and hyperinsulinemia [27]. The transcription factor forkhead box O1
(FoxO1) plays a key role hepatic gluconeogenesis. In response to insulin,
FoxO1 is phosphorylated via PI3K pathway and sequestered in the cyto-
plasm. This prevents FoxO1 mediated transcriptional activation of gluconeo-
genic genes [28]. Animal studies clearly demonstrated that FoxO1 knock out
improves insulin resistance and protects from diabetes [29].
Role of free fatty acids

Previous investigators have shown that obesity and elevated free fatty
acids (FFA) levels play a major role in the development of insulin resistance.
In 1963, Randle and colleagues [30] described the glucose-fatty acid cycle,
also known as the Randle cycle, and showed that increased FFA levels in-
hibits glucose uptake and metabolism in rat muscle cells. These investigators
postulated that elevated FFA oxidation increases the production of mito-
chondrial acetyl-CoA, which inhibits pyruvate dehydrogenase (PDH). In
addition, increased acetyl-CoA results in increased citrate levels, which in-
hibits phosphofructokinase and leads to a buildup of glucose-6-phosphate
(G-6-P). Accumulation of G-6-P inhibits hexokinase (HK) II activity and
lastly leads to reduced glucose uptake (Fig. 4).
In contrast, other investigators have suggested an alternative mechanism
for FFA induced insulin resistance. Increased concentrations of intracellular
fatty acid metabolites, fatty acyl Co-A, ceramides, and diacylglycerol, up-
regulates protein kinase C levels which activates the serine/threonine cas-
cade leading to phosphorylation of serine/threonine sites on IRS-1, which
down-regulates the PI3K pathway [31]. Thus, glucose transport and other
downstream insulin signaling events are depressed. Nonesterified fatty acids
have also been shown to induce hepatic and peripheral insulin resistance,
endothelial dysfunction, pancreatic b cell apoptosis, and increased plasmin-
ogen activator inhibitor-1 (Fig. 4B) [32].
Clinical studies have shown that most obese patients have elevated FFAs
[33], and both acute [34] and chronic [35] elevations in FFAs contribute to
insulin resistance. Even in lean healthy subjects there is a dose-dependent
repression of insulin signaling and glucose disposal in skeletal muscle in
response to escalating levels of FFAs [36].
1068 LANN & LEROITH
A
Randle Cycle
citrate
HK PFK PDH
glucose G-6-P pyruvate Acetyl CoA
GLUT4
Fatty Acids
Plasma
Glucose
B
insulin
Alternative Mechanism
pS pY
IRS
pS pS upregulation of
pS serine/threonine
cascade
PI3K
PKC
Acetyl CoA
Diacylglycerol
Ceramides
GLUT4
translocation
Plasma
Glucose Fatty Acids
Fig. 4. Propsed mechanisms for insulin resistance in skeletal muscle. (A) The Randle Cycle. (B)
The alternative mechanism for FFA-induced insulin resistance in skeletal muscle (Adapted from
Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000;106:171–6; with
permission).
Impaired lipid metabolism

Hepatic insulin resistance also leads to up-regulated triglyceride (TG)
synthesis and down-regulated FFA oxidation [37,38]. The hypertriglyceride-
mia observed in the metabolic syndrome is manifested by elevated serum
levels of triacylglycerols. The transport mechanisms for these particles are
either by chylomicrons released from the gut or hepatically produced very
low-density lipoproteins (VLDLs). Manifestations of the dyslipidemia
observed in insulin resistance are elevated TG/VLDL levels and VLDL’s
structural protein apolipoproteinB and reduced levels of high-density
lipoprotein (HDL) and its major structural protein apolipoprotein A-1
(Apo A-I). Additionally, chylomicron formation and association with its

apoliproteins (ApoB-48 and ApoC-I/II/III) are dysregulated in the presence
of insulin resistance. The role of insulin in lipid metabolism is regulated
through different mechanisms, as discussed below.
Normally, as chylomicrons and TG circulate, they acquire ApoC in ad-
dition to other apolipoproteins. ApoC-II is required for lipolysis of chylomi-
cron TG, as it is an activator of lipoprotein lipase (LpL), whereas, ApoC-III
inhibits LpL. Reductions in LpL levels have been observed in type 2 diabetic
and insulin resistant patients [39]. In vitro and in vivo data suggest that in-
sulin resistant states result in increased ApoC-III production, which is insu-
lin mediated, possibly via the transcription factor FoxO1 [40]. Insulin
resistance also leads to increased VLDL production, and thus VLDL and
TG levels, which eventually must compete with chylomicrons for LpL-me-
diated lipolysis [41]. There are three sources of TG that contribute to the
elevated VLDL levels observed in insulin resistant individuals [42]. First,
lipolysis of TG from adipose tissue causes elevated serum fatty acid (FA)
levels, which results in increased FA flux to the liver. Second, because
LpL levels are decreased, there is an inhibition of lipolysis of chylomicrons
and VLDL and TG, which leads to elevations in TG rich remnants that
eventually are delivered to the liver. Third, insulin resistant states lead to
increased hepatic de novo TG synthesis.
HDL and ApoA-I play important roles in cardioprotective and
antiatherogenic processes. HDL exerts these protective effects mainly
through reverse cholesterol transport (RCT). RCT allows for the transfer
of cholesterol from the plasma membrane onto lipid poor HDLs, which
then transport it to the liver for eventual excretion into bile [42]. Insulin
ultimately acts on ApoA-1 promotor via the activation of the transcription
factor Sp-1 by both the MAPK and PI3K pathways [43]. Insulin resistance
reduces HDL and ApoA-I levels, which can be attributed to the increase in
fractional clearance and catabolism of HDL without changing ApoA-I pro-
duction [44,45]. Another possibility for this change is reduced responsive-
ness of the ApoA-I gene to insulin [46]. In vitro experiments simulating
an insulin resistant state demonstrate that exposure of cells to FFA
suppresses insulin-mediated ApoA-I gene expression by inhibiting Sp-1
activity [47]. TNF-a and interleukin (IL)-1b, inflammatory cytokines
elevated in insulin resistance, down-regulate ApoA-I gene expression at
the transcriptional level in a dose-dependent manner [48].
In clinical practice, treating insulin resistant associated dyslipidemia
typically involves using fibrates (PPARa agonists) and niacin. Fibrates
increased LpL and ApoA-I levels and increase hepatic FFA oxidation.
The Veterans Affairs high-density lipoprotein cholesterol intervention trial
demonstrated that treating both diabetic and nondiabetic patients with prior
coronary heart disease with gemfibrozil significantly decreased subsequent
cardiac events [49]. Previously, niacin was generally not considered as a first
line medication for patients with type 2 diabetes and dyslipidemia, as this
1070 LANN & LEROITH
therapy has been shown to exacerbate insulin resistance and increase glucose
levels. Nevertheless, more recent studies indicate that the use of extended
release niacin seems to have a minimal effect on insulin sensitivity in the set-
ting of appropriate monitoring of glycemic control [50]. Additionally, there
is evidence that the TZD pioglitazone (PPARg agonist) lowers both post-
prandial and fasting TG levels by enhanced VLDL and TG lipolysis, albeit
a small effect [51].
Adipose tissue, cytokines, and proinflammatory states

The metabolic syndrome and insulin resistance are strongly associated
with excess adiposity and inflammatory states. Investigators have shown
that depletion of intramyocellular fat stores in skeletal muscle improves in-
sulin sensitivity [52]. Fat tissue can be considered an endocrine organ, as it
secretes hormones and cytokines that affect insulin’s actions [3].
Insulin resistant states are often associated with serine/threonine
phosphorylation of IRS-1, one of the most proximal downstream signaling
proteins in the insulin signaling cascade. There are a number of proinflam-
matory kinases, such as inhibitor of kB kinase b (IKKb) and c-jun N
terminal kinase-1, that phosphorylate IRS-1 at serine 307 [53]. The excess
adipose tissue observed in the metabolic syndrome results in the overpro-
duction of inflammatory cytokines TNF-a, IL-6, resistin, and C-reactive
protein (CRP) [54]. TNF-a also augments serine phosphorylation of IRS
proteins. Additionally, obese rodents have elevated levels of TNF-a and in-
sulin resistance and, upon lowering of TNF-a levels, improvement in insu-
lin sensitivity is observed [55]. Complimentary to these findings, treatment
of human subjects with extremely high doses of salicylates inhibit IKKb ac-
tivity and ameliorates insulin resistance [56]. Resistin, a fat related hor-
mone, is hypersecreted from adipose tissue and impedes insulin action.
Evidenced by animal studies, genetically induced resistin deficiency in
mice improves insulin-mediated glucose uptake into adipocytes and muscle
[57].
Both obese mice and human beings possess adipose tissue which contains
up-regulated and hyperactivated macrophages [58]. Many proinflammatory
cytokines, such as IL-1b, are produced by these macrophages. IL-1b levels
are elevated in patients with type 2 diabetes which is positively correlated
with the metabolic syndrome [59]. Chronic treatment of adipocytes with
IL-1b leads to decreased expression and translocation of GLUT4 in
response to insulin, caused by decreased expression of IRS-1 [60]. In animal
studies, the deletion of IKKb in the myeloid cell lines that eventually
produce macrophages in obese and high fat diet fed mice confers protection
from insulin resistance [61].
Conversely, adiponectin is a potent anti-inflammatory cytokine that is se-
creted by adipocytes and is down-regulated in obese patients [62]. An inverse
relationship exists between adiponectin levels and insulin sensitivity. In
skeletal muscle, adiponectin aids in glucose transport and FFA oxidation,

and in the liver it suppresses glucose production via down-regulation of glu-
coneogenic enzymes [63]. In clinical practice this effect can be augmented by
treating patients with pioglitazone, which significantly increases adiponectin
levels [64].
Detecting a proinflammatory state can be most easily measured through
serum CRP levels. However, there are no medications that specifically target
this proinflammatory state. The drugs that target the derangements of the
metabolic syndrome, such as TZDs, statins, and fibrates, have been shown
to decrease CRP levels [65,66].
Hypertension
Multiple mechanisms have been proposed to explain the link between
hypertension and insulin resistance. Hyperinsulinemia is associated with
adrenergic overactivity, leading to increased cardiac output and urinary cat-
echolamine excretion [67]. Insulin is also a potent antinatriuretic hormone,
causing sodium retention and plasma volume expansion. Previous studies
have shown that obesity leads to increased renal sympathetic activity, which
results in retention of sodium and, in animal studies, renal denervation
reverses this phenomenon [68,69]. The obesity related hormone leptin has
been linked to increased sympathetic activity, elevated blood pressure and
heart rate in animal studies [70]. Increased sympathetic activity also stimu-
lates the renin-angiotensin system (RAS). The relationship between obesity
and elevated blood pressure may also be explained by the activation of the
RAS in adipose tissue. In vivo studies have shown that obese rats have 50%
higher levels of angiotensinogen mRNA expression in adipose tissue when
compared with lean rats [71]. Several large clinical trials, the Heart Out-
comes Prevention Evaluation (HOPE) study (ramipril) and the Losartan In-
tervention For Endpoint (LIFE) reduction in hypertension study (losartan)
have demonstrated the benefit of blocking the RAS with either an angioten-
sin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers
(ARB) and the improvement of insulin sensitivity [72,73]. In addition, other
clinical studies have shown that inhibition of the RAS with either ACEIs or
ARBs results in a significant increase in adiponectin levels, which is
associated with improved insulin sensitivity [74]. Both acute and chronic
administration of nonesterified fatty acids is associated with an increase
in systemic blood pressure of about 30 mmHg and 16 mmHg, respectively
[75,76].
Elevated levels of serum nonesterified fatty acids leads to increased a-ad-
renergic vasoconstriction and enhanced pressor sensitivity. For example, in
one study, increased levels of nonesterified fatty acids reduces the dose of
phenylephrine required to produce 50% of the maximal vasoconstriction
response; however, there was no response to angiotensin II [77].
1072 LANN & LEROITH
Endothelial dysfunction
Although not considered a classic function, insulin does exert physiologic
effects on endothelial and vascular smooth muscle cells. Physiologic levels of
insulin causes release of nitric oxide (NO) from endothelial cells, leading to
vasodilation of peripheral vasculature, which results in the augmentation of
blood flow and glucose disposal in skeletal muscle [78]. Insulin stimulates
NO production from endothelial cells through the PI3K pathway and the
secretion of endolethin-1 (ET-1), a potent vasoconstrictor, through the
MAPK pathway [79]. Insulin also potentiates acetylcholine-induced vasodi-
lation in both normotensive and hypertensive patients. However, in insulin
resistant states, there is an imbalance between the production of NO and
secretion of ET-1, favoring a state of vasoconstriction. The metabolic
derangement leads to defects in the PI3K pathway and, thus, decreased
NO release; however, the MAPK pathway remains unaffected, resulting in
impaired blood flow and disrupted glucose disposal and a further deteriora-
tion of insulin resistance [79]. Hyperinsulinemia not only up-regulates ET-1
gene expression and stimulates ET-1 release, but it also doubles the number
of ETA receptors (receptors with greatest affinity for ET-1) on vascular
smooth muscle cells (VSMC) [80]. Additionally, insulin and ET-1 seem to
act synergistically to increase VSMC proliferation in culture [80].
Adiposity and the resulting elevation in FFA and TNF-a and decrease in
adiponectin, adversely affect endothelial function and promote atherogene-
sis. Elevated FFA levels have been shown to impair insulin-mediated vaso-
dilation via inhibition of PI3K pathway [81]. TNF-a down-regulates mRNA
for endothelial NO synthase, an enzyme required for NO production and
release [81]. In addition, TNF-a exacerbates inflammatory changes in the
vessel wall by activating NF-kB and causing up-regulating the expression
of vascular adhesion molecules and enhancing monocyte adhesion [82].
Adioponectin has potent antiatherogenic effects: inhibition of monocyte
adhesion, inhibition of macrophage transformation to foam cells, and
decreased proliferation of VSMC [83]. Thus, its suppression leads to endo-
thelial dysfunction and increased risk for cardiovascular disease.
Identifying insulin resistance

Besides using guidelines such as ATP III or World Health Organization
criteria for identifying patients with the metabolic syndrome, there are other
surrogate markers of insulin resistance that can be employed in clinical
practice. For instance, triglyceride level greater than 130 mg/dL, triglycer-
ide-to-HDL ratio greater than 3, and serum insulin level greater than
15 mU/mL can be used to assess insulin resistance [84]. In the primary
care setting, Reaven and colleagues suggest that the most practical approach
to identify patients with insulin resistance is using either the serum
triglyceride level or the triglyceride-to-HDL ratio.
Other research tools used to measure insulin sensitivity include the

homeostatic model assessment (HOMA) of insulin resistance, hyperinsuline-
mic-euglycemic clamp, and the steady state plasma glucose (SSPG). The
HOMA can be calculated using fasting insulin and glucose levels (insulin
glucose O 22.5). This equation is the ratio of the insulin sensitivity of
the patient in question to the insulin sensitivity of a normal individual
(with a normal fasting serum insulin of 5 mU/mL, a normal fasting serum
glucose of 4.5 mmol/L, with the product of these two values equaling
22.5, the normalizing factor) [85,86]. Therefore, a HOMA greater than 1
represents an insulin resistant state. The hyperinsulinemic-euglycemic clamp
and steady state plasma glucose are other research tools used in the assess-
ment of insulin sensitivity [85]. Clamp studies measure the mean glucose rate
required to maintain a euglycemic state steady in a constant insulin infusion.
Thus, a greater glucose requirement indicates greater insulin sensitivity. The
SSPG is the glucose level measured in the setting of a fixed infusion of both
insulin and glucose. The lower the glucose level measured, the greater the
degree of insulin sensitivity.
Summary
Classically, the metabolic syndrome is characterized as group of pathol-
ogies, including visceral obesity, hypertension, dyslipidemia, and impaired
glucose tolerance. It is now realized that insulin resistance plays a principal
role in initiating and perpetuating the pathologic manifestations of the
metabolic syndrome. A more in-depth understanding of the basic patho-
physiologic mechanisms underlying insulin resistance may aid clinicians in
treating and possibly delaying or even preventing the onset of the metabolic
syndrome and its complications.
References
[1] Smith DO, LeRoith D. Insulin resistance, pre-diabetes, and the prevention of type 2 diabetes.
Clin Cornerstone 2004;6(2):7–6.
[2] Setsi G. Pathophysiology of insulin resistance. Best Pract Res Clin Endocrinol Metab 2006;
20(4):665–79.
[3] LeRoith D. Beta cell dysfunction and insulin resistance in type 2 diabetes: role of metabolic
and genetic abnormalities. Am J Med 2002;113(6A):3s–11s.
[4] Vauhkonen I, Niskanen L, Vanninen E, et al. 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 1998;101:86–96.
[5] Kashyap S, Belfort R, Gastaldelli A, et al. A sustained increase in plasma free fatty acids
impairs insulin secretion in nondiabetic subjects genetically predisposed to type 2 diabetes.
Diabetes 2003;52:2461–74.
[6] Del Prato S, Leonetti F, Simonson DC, et al. Effect of sustained physiologic hyperinsulinae-
mia and hyperglycaemia on insulin secretion and insulin sensitivity in man. Diabetologia
1994;37:1025–35.
1074 LANN & LEROITH
[7] Roth J, Qiang X, Marban SL, et al. The obesity pandemic: where have we been and where are
we going? Obes Res 2004;(12 Suppl 2):88s–101s.
[8] Marban SL, Roth J. Transgenic hyperinsulinemia: a mouse model of insulin resistance and
glucose intolerance without obesity. In: Shafrir E, editor. Lessons from animal diabetes IV.
6th edition. Boston: Birkhauser; 1996. p. 201–24.
[9] Biddinger SB, Kahn CR. From mice to men: insights into insulin resistance syndromes.
Annu Rev Physiol 2006;68:123–58.
[10] George S, Rochford JJ, Wolfrum C, et al. A family with severe insulin resistance and diabetes
due to a mutation in Akt2. Science 2004;304(5675):1325–8.
[11] Roth RA, Liui F, Chin JE, et al. Biochemical mechanisms of insulin resistance. Horm Res
1994;41(Suppl 2):51–5.
[12] Paz K, Hemi R, LeRoith D, et al. A molecular basis for insulin resistance. Elevated serine/
threonine phosphorylation of IRS-1 and IRS-2 inhibits their binding to the juxtamembrane
region of the insulin receptor and impairs their ability to undergo insulin-induced tyrosine
phosphorylation. J Bio Chem 1997;272(47):29911–8.
[13] Kido Y, Burks DJ, Withers D, et al. Tissue-specific insulin resistance in mice with mutations
in the insulin receptor, IRS-1, and IRS-2. J Clin Invest 2000;105(2):199–205.
[14] Bouche C, Serdy S, Kahn CR, et al. The cellular fate of glucose and its relevance in type 2
diabetes. Endocr Rev 2004;25(5):807–30.
[15] Watson RT, Pessin JE. Intracellular organization of insulin signaling and GLUT4 translo-
cation. Recent Prog Horm Res 2001;56:175–93.
[16] Saltiel AR, Pessin JE. Insulin signaling pathways in time and space. Trends Cell Biol 2002;
12(2):65–71.
[17] Rothman DL, Shulman RG, Shulman GI. 31P nuclear magnetic resonance measurements of
muscle glucose-6-phosphate: evidence for reduced insulin-dependent muscle glucose trans-
port or phosphorylation activity in non-insulin dependent diabetes mellitus. J Clin Invest
1992;89:1069–75.
[18] Cline CW, Petersen KF, Krssak M, 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.
[19] Zisman A, Peroni OD, Abel ED, et al. Targeted disruption of the glucose transporter 4 se-
lectively in muscle causes insulin resistance and glucose intolerance. Nat Med 2000;6:
924–8.
[20] Petersen KF, Krssak M, Inzucchi S, et al. Mechanism of troglitazone action in type 2 diabe-
tes. Diabetes 2000;49:827–31.
[21] Shintani M, Nishimura H, Yonemitsu S, et al. Troglitazone not only increases GLUT4 but
also increases its translocation in rat adipocytes. Diabetes 2001;50:2296–3000.
[22] Al-Kahlili L, Forsgren M, Kannisto K, et al. Enhanced insulin-stimulated glycogen synthe-
sis in response to insulin, metformin or rosiglitazone is associated with increased mRNA ex-
pression of GLUT4 and peroxisomal proliferator activator receptor gamma co-activator 1.
Diabetologia 2005;48(6):1173–9.
[23] Yang J, Holman GD. Long-term metformin treatment stimulates cardiomyocyte glucose
transport through an AMP-activated protein kinase-dependent reduction in GLUT4 endo-
cytosis. Endocrinology 2006;147(6):2728–36.
[24] Goodyear LJ, Kahn BB. Exercise glucose transport, and insulin sensitivity. Annu Rev Med
1998;49:235–61.
[25] Kim HJ, Lee JS, Kim CK. Effect of exercise training on muscle glucose transporter 4 protein
and intramuscular lipid content in elderly men with impaired glucose tolerance. Eur J Appl
Physiol 2004;93(3):353–8.
[26] Cherrington AD. Banting lecture 1997. Control of glucose uptake and release by the liver in
vivo. Diabetes 1999;53:1633–42.
[27] Taniguchi CM, Kondo T, Sajan M, et al. Divergent regulation of hepatic glucose and lipid
metabolism by phophoinositide 3-kinse via Akt and PKCl/z. Cell Metab 2006;3:343–53.
[28] Puigserver P, Rhee J, Donovan J, et al. Insulin-regulated hepatic gluconeogenesis through

FoxO1-PCG-1alpha interaction. Nature 2003;423:550–5.
[29] Accili D, Arden KC. FoxOs at the crossroads of cellular metabolism, differentiation, and
transformation. Cell 2004;117(4):421–6.
[30] Randle PJ, Garland PB, Hales CN, et al. The glucose fatty-acid cycle. Its role in insulin sen-
sitivity and the metabolic disturbances of diabetes mellitus. Lancet 1963;1:785–9.
[31] Shulman GI. Cellular mechanisms of insulin resistance. J Clin Invest 2000;106:171–6.
[32] Sarafidis PA, Bakris GL. Non-esterified fatty acids and blood pressure elevation: a mechanism
for hypertension in subjects with obesity/insulin resistance? J Hum Hypertens 2007;21:12–9.
[33] Reaven GM, Hollenbeck C, Jeng CY, et al. Measurement of plasma glucose, free fatty acid,
lactate and insulin for 24h in patients with NIDDM. Diabetes 1988;37:1020–4.
[34] Boden G, Jadali F, White J, et al. Effects of fat on insulin-stimulated carbohydrate metab-
olism in normal men. J Clin Invest 1991;88:960–6.
[35] Santomauro AT, Boden G, Silva ME, et al. Overnight lowering of free fatty acids with aci-
pimox improves insulin resistance and glucose tolerance in obese diabetic and non-diabetic
subjects. Diabetes 1999;48:1836–41.
[36] Belfort R, Mandarino L, Kashyap S, et al. Dose-response effect of elevated plasma free fatty
acid on Insulin signaling. Diabetes 2005;54:1640–8.
[37] Schwartz JM, Linfoot P, Dare D, et al. Hepatic de novo lipogenesis in normoinsulinemic and
hyperinsulinemic subjects consuming high-fat, low-carbohydrate and low fat, high-carbohy-
drate isoenergetic diets. Am J Clin Nutr 2003;77:43–50.
[38] Iozzo P, Turpeinen AK, Takala T, et al. Defective liver disposal of free fatty acids in patients
with impaired glucose tolerance. J Clin Endocrinol Metab 2004;89:3496–502.
[39] Merkel M, Eckel RH, Goldberg IJ, et al. Lipoprotein lipase: genetics, lipid uptake, and
regulation. J Lipid Res 2002;43:1997–2006.
[40] Altomonte J, Cong L, Harbaran S, et al. Foxo1 mediates insulin action on apoC-III and
triglyceride metabolism. J Clin Invest 2004;114:1493–503.
[41] Bjorkegren J, Packard CJ, Hamsten A, et al. Accumulation of large very low density lipopro-
tein in plasma during intravenous infusion of a chylomicron-like triglyceride emulsion
reflects competition for a common lipolytic pathway. J Lipid Res 1996;37:76–86.
[42] Chahil T, Ginsberg HN. Diabetic dyslipidemia. Endocrinol Metab Clin North Am 2006;
35(3):491–510.
[43] Zheng XL, Matsubara S, Diao C, et al. Activation of apoprotein AI gene expression by pro-
tein kinase A and C through transcription factor, Sp1. J Biol Chem 2000;275:31747–54.
[44] Horowitz BS, Goldberg IJ, Merab J, et al. Increased plasma and renal clearance of exchange-
able pool of apolipoproteinA-I in subjects with low levels of high density lipoprotein choles-
terol. J Clin Invest 1993;91:1743–60.
[45] Vajo Z, Terry JG, Brinton EA. Increased intra-abdominal fat may lower HDL levels by in-
creasing the fractional catabolic rate of Lp A-I in post-menopausal women. Atherosclerosis
2002;152:229–37.
[46] Mooradian AD, Haas MJ, Wong NC. Transcriptional control of apolipoprotein A-I gene
expression in diabetes. Diabetes 2004;53:513–20.
[47] Haas MJ, Horani MH, Wong NC, et al. Induction of the apolipoprotein AI promoter by Sp1
is repressed by saturated fatty acids. Metabolism 2004;53(10):1342–8.
[48] Haas MJ, Horani M, Mreyoud A, et al. Suppression of apoprotein A-I gene expression in
HepG2 cells by TNF-a and IL-1b. Biochim Biophys Acta 2003;1623:120–8.
[49] Rubins HB, Robins SJ, Collins D, et al. Veterans affairs high-density lipoprotein cholesterol
intervention trial study group. Gemfibrozil for the secondary prevention of coronary heart
disease in men with low levels of high-density lipoprotein cholesterol. N Engl J Med 1999;
341:410–8.
[50] Vega GL, Cater NB, Meguro S, et al. Influence of extended-release nicotinic acid on nones-
terified fatty acid flux in the metabolic syndrome with atherogenic dyslipidemia. Am J Car-
diol 2005;95(11):1309–13.
1076 LANN & LEROITH
[51] Al Majali K, Cooper MB, Staels B, et al. The effect of sensitization to insulin with
pioglitazone on fasting and postprandial lipid metabolism, lipoprotein modification by
lipases, and lipid transfer activities in type 2 diabetic patients. Diabetologia 2006;49(3):
527–37.
[52] Greco AV, Mingrone G, Giancaterini A, et al. Insulin resistance in morbid obesity: reversal
with intramyocellular depletion. Diabetes 2002;51:144–51.
[53] De Luca C, Olefsky JM. Stressed out about obesity and insulin resistance. Nat Med 2006;12:
41–2.
[54] Trayhurn P, Wood IS. Adipokines: inflammation and the pleiotropic role of white adipose
tissue. Br J Nutr 2004;92:347–55.
[55] 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.
[56] Hundal RS, Petersen KF, Mayerson AB, et al. Mechanism by which high-dose aspirin
improves glucose metabolism in type 2 diabetes. J Clin Invest 2002;109:1321–6.
[57] Qi Y, Nie Z, Lee YS, et al. Loss of resistin glucose homeostasis in leptin deficiency. Diabetes
2006;55(11):3083–90.
[58] Weisberg SP, McCann D, Desai M, et al. Obesity is associated with macrophage accumula-
tion in adipose tissue. J Clin Invest 2003;112(12):1796–808.
[59] Salmenniemi U, Ruotsalainen E, Pihlajamaki J, et al. Multiple abnormalities in glucose and
energy metabolism and coordinated changes in levels of adiponectin, cytokines, and adhe-
sion molecules in subjects with metabolic syndrome. Circulation 2004;110:3842–8.
[60] Jager J, Gremeaux T, Cormont M, et al. Interleukin-1b-induced insulin resistance in adipo-
cytes through down-regulation of insulin receptor substrate-1 expression. Endocrinology
2007;148(1):241–51.
[61] Arkan MC, Hevener AL, Greten FR, et al. IKK-beta links inflammation to obesity-induced
insulin resistance. Nat Med 2005;11:191–8.
[62] Arita Y, Kihara S, Ouchi N, et al. Paradoxical decrease of an adipose-specific protein,
adiponectin, in obesity. Biochem Biophys Res Commun 1999;257:79–83.
[63] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365:1415–28.
[64] Gastaldelli A, Miyazaki Y, Mahankali A, et al. The effect of pioglitazone on the liver: role of
adiponectin. Diabetes Care 2006;29(10):2275–81.
[65] Jialal I, Stein D, Balis D, et al. Effect of hydroxymethyl glutaryl coenzyme a reductase inhib-
itor therapy on high sensitive C-reactive protein levels. Circulation 2001;103:1933–5.
[66] Nesto R. C-reactive protein, its role in inflammation, type 2 diabetes and cardiovascular
disease, and the effects of insulin sensitizing treatment with thiazolidinediones. Diabet
Med 2004;21:810–87.
[67] Ferrannini E. Is insulin resistance the cause of the metabolic syndrome? Ann Med 2006;38:
42–51.
[68] Carlson SH, Shelton J, White CR, et al. Elevated sympathetic nerve activity contributes to
hypertension and salt sensitivity in diabetic obese zucker rats. Hypertension 2000;35:403–8.
[69] Kassab S, Kato T, Wilkins FC, et al. Renal denervation attenuates the sodium retention and
hypertention associated with obesity. Hypertension 1995;25:893–7.
[70] Aizawa-Abe M, Ogawa Y, Masuzaki H, et al. Pathophysiological role of leptin in obesity
related hypertention. J Clin Invest 2000;105:1243–52.
[71] Jones BH, Standridge MK, Taylor JW, et al. Angiotensinogen gene expression in adipose
tissue: analysis of obese models and hormonal and nutritional control. Am J Physiol 1997;
273:R236–42.
[72] Yusuf S, Sleight P, Pogue J, et al. Effects of an angiotensin-converting enzyme inhibitor,
ramipril, on cardiovascular event in high-risk patients. The heart outcomes prevention
evaluation study investigators. N Engl J Med 2000;342:145–53.
[73] Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the
losartan intervention for endpoint reduction in hypertension study (LIFE): a randomized
trial against atenolol. Lancet 2002;359(9311):995–1003.
[74] Furuhashi M, Ura N, Higashiura K, et al. Blockade of the renin-angiotensin system

increases adiponectin concentrations in patients with essential hypertension. Hypertension
2003;42:76–81.
[75] Bulow J, Madsen J, Hojgaard L. Reversibility of the effects on local circulation of high lipid
concentrations in blood. Scand J Clin Lab Invest 1990;50:291–6.
[76] Grekin RJ, Vollmer AP, Sider RS. Pressor effects of portal venous oleate infusion: a pro-
posed mechanism for obesity hypertension. Hypertension 1995;26:193–8.
[77] Stepniakowski KT, Goodfriend TL, Egan BM. Fatty acids enhance vascular a-adrenergic
sensitivity. Hypertension 1995;25:774–8.
[78] Baron AD, Clark MG. Role of blood flow in the regulation of muscle glucose uptake. Annu
Rev Nutr 1997;17:487–99.
[79] Kim JA, Montagnani M, Koh KK, et al. Reciprocal relationships between insulin resistance
and endothelial dysfunction: molecular and pathophysiological mechanisms. Circulation
2006;113:1888–904.
[80] Frank HJ, Levin ER, Hu RM, et al. Insulin stimulates endothelin binding and action on
cultured vascular smooth muscle cells. Endocrinology 1993;133:1092–7.
[81] Steinberg HO, Baron AD. Vascular function, insulin resistance, and fatty acids. Diabetolo-
gia 2002;45(5):623–34.
[82] Eck SL, Perkins SL, Carr DP, et al. Inhibition of phorbol ester-induced cellular adhesion by
competitive binding of NF-kappaB in vivo. Mol Cell Biol 1993;13(10):6530–6.
[83] Ouchi N, Kihara S, Arita Y, et al. Adiponectin, an adipocyte-derived plasma protein,
inhibits endothelial NF-kappaB signaling through a cAMP-dependent pathway. Circulation
2000;102(11):1296–301.
[84] McLaughlin T, Abbasi F, Cheal K, et al. Use of metabolic markers to identify overweight
individuals who are insulin resistant. Ann Intern Med 2003;139:802–9.
[85] Bloomgarden ZT. Measures of insulin sensitivity. Clin Lab Med 2006;26:611–33.
[86] Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model assessment: insulin
resistance and beta cell function from fasting plasma glucose and insulin concentrations in
man. Diabetologia 1985;28:412–9.
Med Clin N Am 91 (2007) 1079–1089
America on the Move

Victoria A. Catenacci, MD, Holly R. Wyatt, MD*
Division of Endocrinology, Metabolism and Diabetes, Center for Human Nutrition, University
of Colorado at Denver and Health Sciences Center,
4200 E. Ninth Avenue, Box C263, Denver, CO 80262, USA
The problem we are facing

The current epidemic of obesity has arisen from a gradual weight gain
that appears to have affected all segments of the population over the past
several decades [1,2]. Using longitudinal data from the Coronary Artery
Risk Development in Young Adults study and cross-sectional data from
the National Health and Nutrition Examination Surveys, the average weight
gain of Americans has been estimated to be 1 to 2 pounds per year [3–7].
Some race sex groups, such as African American women, are gaining weight
much more rapidly. Weight maintenance has occurred only in a small mi-
nority of individuals in all race sex groups, except for normal weight white
women, where 51% maintained their baseline weight [3]. The trend of grad-
ual weight gain appears to be occurring in populations in other countries as
well. Brown and colleagues [8] found that middle-aged Australian women
are gaining an average of 0.5 kg (approximately 1.1 lbs) per year. Children
are also not immune to excessive weight gain and are gaining more weight
than appropriate for healthy linear growth [9].
As a result of this consistent, gradual weight gain, the prevalence of ex-
cessive weight gain and obesity continues to increase. As of 2004 in the
United States, 32.2% of adults were obese and 17.1% of children and ado-
lescents were overweight [1]. Overweight or obesity in adults has reached an
astonishing 66.3% of the United States population, reflecting the continued
and steady increases in body weight over time [1]. It is estimated that child-
hood obesity has tripled in the last 40 years [1,10]. Unfortunately, we can
now say the majority of Americans are currently at an unhealthy body
weight that needs some type of management from health care providers.
This work was supported by NIH grant DK 42549.

E-mail address: holly.wyatt@uchsc.edu (H.R. Wyatt).
1080 CATENACCI & WYATT
Individuals maintaining a healthy weight have become the minority. Regret-

tably, the increase in body weight shows little sign of slowing. At this current
rate of weight gain it has been estimated that 73% of the adult population
will be overweight or obese by 2008 [4].
These increases in the prevalence of overweight and obesity are concern-
ing primarily because excessive body fat is linked with a long and growing
list of comorbidities and adverse health complications. Increasing rates of
overweight and obesity translate directly into increases in type 2 diabetes,
heart disease, high blood pressure, sleep apnea, and certain cancers
[11,12]. Increasing body fat, especially visceral fat, is also associated with
increases in the prevalence of the metabolic syndrome [13]. Overweight is
associated with significant health problems in the pediatric population
and is an important early risk factor for much of adult morbidity and mor-
tality [14]. Because of these associations, the increases in overweight and
obesity are a major public health crisis for the United States and the world
[15].
Understanding the small change approach to weight management

While there is agreement about the urgent need to address the epidemic
of obesity, there is not agreement about how best to do this. Past efforts
to produce and maintain weight loss have not been overly successful, either
in obese adults or children [16–18]. One of the reasons for the lack of long-
term success may be related to the fact that substantial weight loss requires
substantial lifestyle changes. Big lifestyle behavioral changes require big
efforts. In the past, despite significant obesity educational initiatives and an
activity and obesity awareness campaign, Americans have had little success
at sustaining this degree of lifestyle change for long periods of time at
a population level.
The recent midcourse review of the Healthy People 2010 initiative illus-
trates this unfortunate truth. Healthy People 2010 provides a vision for
achieving improved health for all Americans. It was developed through a na-
tional process to identify a set of 10-year health objectives to achieve during
the first decade of the 21st century. The midcourse review assesses the status
of the national objectives [19]. The Healthy People 2010 weight goal was for
60% of American adults to be at a healthy body weight by 2010. This rep-
resented an 18-percentage point increase from the value of 42% recorded in
1994. Unfortunately, the trend at the midcourse review has been in the op-
posite direction, with now only 34% of the population at a healthy body
weight, a decrease rather then increase of 8-percentage points (Fig. 1) [20].
Trends in behaviors, such as moderate physical activity, are not much
better. Despite the message that most individuals should engage in 30 min-
utes of activity most days of the week, there has been very little if any in-
crease in the number of individuals successfully changing this behavior
between 1999 to 2002 [21,22]. While the goal for Healthy People 2010 is
AMERICA ON THE MOVE 1081
2010 Healthy
People Target
1988-1994
1999-2000
Total
White
Black
Mexican
American
Female
Male
0 10 20 30 40 50 60
Percent
Fig. 1. Healthy People 2010 Midcourse Review, Adults at a Healthy Weight 1988–94 to
1999–2000. Data are for ages 20 years and over, age adjusted to the 2000 standard population.
Healthy weight is defined as having a body mass index higher than 18.5 but lower than 25.0.
(Adapted from U.S. Department Of Health and Human Services. Centers for Disease Control
and Prevention. National Center for Health Statistics, Hyattsville, MD. Available at: http://
www.cdc.gov/nchs/about/otheract/hpdata2010/focusareas/fa19-nutrition.htm.)
for 50% of the population to engage in this behavior, only 32% report en-
gaging in 30 minutes-a-day of activity, and there has been no increase over
time in moving toward a goal that is critical for maintaining a reduction in
body weight [22].
An alternative population strategy to consider is to prevent weight gain
and to stabilize the weight of the population as a first step, not necessarily
to reduce it [4,23]. Keeping rates of overweight and obesity from increasing
further would be a marked improvement over the current situation [1,19].
Many experts believe we may have more success in preventing obesity
than in reversing it once it is established. The concept is that smaller behav-
ior changes are likely to be more achievable in the population, as opposed to
the larger changes required in achieving permanent weight loss.
Preventing excessive weight gain in children and adults may be our best
chance of stopping the obesity epidemic and should be a major public health
goal. Furthermore, this goal can be accomplished by promoting small,
achievable changes in diet and physical activity [24,25].
America on the Move

To translate the research about causes of weight gain in the population
into action, America on the Move (AOM) was created as an initiative by
the nonprofit America on the Move Foundation [23]. AOM aims to prevent
weight gain in adults and prevent excessive weight gain in children through
the promotion of small increases in steps and small decreases in energy
intake [23]. It appears that most of the weight gain seen in the population is
caused primarily by sustained, slightly positive energy balance [4,9]. In fact,
it has been calculated that the weight gain could be prevented in 90% of the
adult population by modifying energy balance (some combination of de-
creased energy intake and increased energy expenditure) by only 100 kcal/
day [4]. The goal of AOM is to alter energy balance by at least 100 calories
per day to prevent weight gain. Rather that set specific guidelines for how
many calories to consume and how much physical activity to attain,
AOM’s strategy is to help people make small decreases in energy intake
and small increases in physical activity, starting where they are right now.
Other organizations provide specific recommendations for diet and physical
activity; AOM aims to move people from where they are now to a healthier
lifestyle in small steps. The program is designed to get people started and
build on small progressive successes.
The AOM goal to modify energy balance by at least 100 kcal per day can
be accomplished by taking 2,000 steps more per day and eating 100 kcal less
a day. Participants are encouraged to purchase a pedometer and determine
usual steps per day before starting the program, so that the goal of increas-
ing steps by 2,000 can be individualized. This differs from other programs
that recommend a specific number of steps per day (ie, 10,000). Setting an
initial goal of 10,000 steps per day may not be achievable or sustainable
for many sedentary people. AOM encourages small, realistic incremental
changes that fit into any lifestyle.
America on the Move has developed tools to aid people in making small
lifestyle changes and offers these tools free to participants. The tools include
online resources, interactive tools, community support, and events to sup-
port and encourage these two small behavioral changes. Tools are available
for individuals, schools, worksites, and health care providers. All the tools
and programs are free and available via the Web and can be accessed at
http://aom.americaonthemove.org.
AOM tools for individuals

Individuals who want to participate in the program can visit the Web and
register at no charge. This allows them to download tip sheets on 100 ways
to cut 100 calories and add 2,000 steps, and to access other health informa-
tion. Step conversion charts are available that take minutes of nonstep ac-
tivities, such as swimming or biking, and converts minutes of these
activities into steps. Step counter instructions and logs are also available.
Participants can track their behavior changes (increases in steps and achiev-
ing 100-calorie reductions in energy intake) online. For example, they can
track their progress in steps as they walk virtual trails across the United
States. They can participate in 6-week challenges and even join virtual walk-
ing groups. Information that participants provide, including body weight,
demographic information, and behavior changes are stored in a database
that allows tracking of success of the individual or group. Nearly one mil-
lion people have engaged with AOM online as individuals. In addition,
a cell phone version of the AOM program is being piloted with National Ur-
ban League to test how effective this method could be to reach populations
that use cell phones more than computers.
AOM tools for groups

Groups such as worksites, churches, and neighbors can join AOM. Cur-
rently, AOM has over 10,000 groups registered, and over half are worksites.
These reach an audience of over 5,000,000. Group leaders set up their group
and receive a unique group registration code to share with potential group
members. Group members then register individually using the group registra-
tion code unique to the group. This allows participants to check on the entire
group’s progress, as well as view progress of individuals who opt in to the ‘‘My
Challenge Buddies’’ feature. Group leaders can gauge success, recognize, and
reward participant efforts through access to a special administrative Web site.
A coordinator starter kit is available that provides tips for how to help groups
engage AOM. This site allows administrators to monitor group progress in
‘‘real time’’ and generate reports at the completion of each group challenge.
The ‘‘Custom Group’’ is a fee-for-service option and requires a business
agreement and minimum of 6 weeks for AOM to set-up a custom site before
launch. Pricing is based on the eligible population, up to 15,000 individuals.
Custom group participation offers additional tools, such a logo cobranded
throughout the site, customized content to reinforce your group’s messages
and news, real-time participation, and more in depth outcome reports to
help gauge success, tools, tips, and techniques for promoting the program
to your entire organization, as well as online support for your program ad-
ministrator and individual users.
AOM tools for schools

America on the Move also has several tools available for schools. The
Balance First program is designed to start elementary and middle school
students down the path to a healthy lifestyle. This standards-based program
teaches children about the importance of balancing physical activity and
food intake. Downloadable lesson plans and program materials make learn-
ing about getting healthy easy and fun. AOM’s Web site for Kids is a great
way to put energy balance concepts into practice. Teachers can register their
classes and ask students to join, using a unique group registration code.
Kids can then track their daily physical activity by steps or minutes, and en-
joy monitoring their progress along interesting online trails. High schools
and colleges can challenge faculty, staff, and students to make healthy life-
style changes with AOM’s online program for participants 16 and older.
Convenient tracking tools provide immediate feedback for individuals.
AOM tools for health care providers

America on the Move has a Healthcare Professional (HCP) Toolkit that
provides downloadable materials that physicians can use with patients to
help them understand the importance of energy balance, the link between
obesity and health concerns, and the long-term benefits of making wise
food and activity choices.
To access the HCP Toolkit, health care providers need to become a reg-
istered user by going to the AOM Web site [23]. Healthcare providers are
given information to give to their patients to get them started making small
lifestyle changes. Health care providers can also download (at no cost)
AOM’s fact sheets, brochures, signs, and posters for their practice or clinical
setting. The HCP Toolkit also includes guidance for staff on the importance
of becoming role models for patients, increasing knowledge of obesity and
health outcomes, and the need for patient support and follow-up. Table 1
lists resources available in the HCP Toolkit, divided into practice and pa-
tient tools. In addition, AOM offers a brief slide presentation ready to use
in patient education sessions, community workshops, and other speaking
engagements. The slides explain the importance of weight gain prevention
Table 1
Contents of the HealthCare providers toolkit
Practice resources and tools Patient resources and take-home materials
Welcome Letter Quick Tips: Using a Step Counter
A brief letter from Dr. Jim Hill, co-founder Advice for patients interested in purchasing
of America on the Move and using a step counter
How to Use this HCP Toolkit Patient Brochures
Overview of the tools in the HCP Toolkit Brochures tailored to the patient’s level of
readiness for change around healthful
eating and active living
10 Tips for Getting Your Practice On the Move 100 Ways to Cut 100 Calories
Guide to creating an environment that Simple ways for patients to make healthful
supports AOM’s small step approach to eating a part of every day
healthful eating and active living
Guide to Buying a Quality Step Counter 100 Ways to Add 2,000 Steps
Advice for health care professionals interested Simple ways for patients and to add physical
in purchasing and using a step counter activity to each day
5 Steps to Intervention (the 5A’s) and Fast Facts on Weight and Health
Motivation (the 5R’s) Brief overviews present the importance of
Reminder of the basic models used to create maintaining a healthy weight and the link
patient behavior change between weight and some diseases
Patient Readiness Assessment Eat Smart Tip Sheets
Quick reference to assess patient health and Concise information to help patients make
readiness for lifestyle changes smart food choices
Patient Contract and Prescription for Better Tracking Your Progress with America on the
Health Pad Move
Tools to help motivate and document patient Tracking sheet for patients unable to access
behavior change America on the Move’s online tools
and the basic principles of America on the Move. A series of educational

handouts for audience members is also provided on the Web.
How does AOM reach the general population?

AOM reaches the public through its multiple partners, sponsors, and com-
munication channels. Partners include the Young Men’s Christian Associa-
ton or YMCAs of America, National Urban League, National Council of
LaRaza, Public Broadcasting System, and the Girl Scouts. Communication
channels include print, radio and television media, via the Web, mobile,
and 1-800 numbers. In addition, AOM partners with a variety of manufac-
turers and retailers that can bring the AOM message into the community in
places where people live, learn, work, and play. AOM actively promotes mes-
sages to the public about the small change strategy through these channels
and has generated over one billion media impressions since its launch in 2003.
The effectiveness of AOM

The America on the Move Foundation is dedicated to rigorous evalua-
tion of the effectiveness of AOM. The feasibility and overall effectiveness
of the AOM message and strategy has been evaluated in individuals, fami-
lies, overweight and at risk for overweight children, and in schools, work-
sites, and church behavioral settings.
The feasibility of the AOM message

The concepts that the AOM message to increase physical activity by
2,000 steps/day results in significant increases in physical activity, and that
the message to reduce energy intake by 100 kcal/day results in a significant
decrease in total energy intake, have both been evaluated [26,27]. For the
AOM message to increase activity by 2,000 steps/day, a nonrandomized in-
tervention trial in the state of Colorado was implemented to evaluate the
feasibility of an intervention designed to use walking in steps per day as
a target for increasing physical activity. Worksites and churches were used
as a recruitment source and behavioral settings for individual employees
and church members to participate. The intervention was to increase walk-
ing by 2,000 step/day using electronic step counters over a 14-week period of
time and to this end it was effective [26]. In the worksite population, average
steps increased steadily throughout the intervention period, from 7,669 plus
or minus 143 (mean standard error for the mean) steps per day (n ¼ 450)
at baseline to 10,417 plus or minus 275 steps per day (n ¼ 270) at week 14. In
the church groups, average steps per day also increased steadily throughout
the intervention period, from 5,896 plus or minus 291 steps per day
(n ¼ 202) at baseline to 8,471 plus or minus 448 steps per day (n ¼ 99) at
week 14. Even though the groups started at very different baselines both
were successful in increasing steps by 2,000 step per day. Based on this
real world study in two separate behavioral settings, steps per day appears
to be a good target for use in interventions to increase physical activity. It
also supports the use of step counters in setting and monitoring individual
physical activity goals.
The AOM message to reduce energy intake by 100 kcal/day has also been
evaluated [27]. The intent of this feasibility study was to determine whether
the recommendation to reduce energy intake by about 100 kcal/day actually
resulted in lower daily energy intake. In this study participants maintained
a diet diary to record 7 days of their usual dietary intake during a baseline
week. Participants were then given tips for reducing energy intake by about
100 kcal/day and tips for increasing walking. They were asked to make
changes to achieve these goals for a 2-week intervention period. Subjects
reported eating 2,273 kcal/day during the baseline week and 1,859 kcal/day
during the intervention week (P!.01). In addition, subjects reported
consuming significantly less (P!.01) of each macronutrient (protein, fat,
carbohydrate) during the intervention week. During the baseline week,
significantly more sugared sodas (P!.01) were consumed in comparison
to the intervention week. The results indicate that recommendations to re-
duce energy intake by about 100 kcal/day combined with tips for accom-
plishing this can be an effective strategy to reduce total daily energy
intake. The mean daily intake during the AOM intervention week was
reduced by about 300 kcal/day, suggesting that the small change recommen-
dation was feasible, easy to understand, and easy to implement.
The effectiveness of the AOM message

The overall effectiveness of AOM in preventing excessive weight gain has
been evaluated in two different studies using a family based delivery ap-
proach to increase daily steps and decrease dietary intake to reduce weight
gain in children and adults. In the first study, 105 families were evaluated
with at least one 8- to 12-year-old child who was at risk for overweight or
overweight [24]. Eighty-two families were randomly assigned to receive
the AOM intervention and 23 families served as a control. The 13-week
AOM intervention consisted of an increase in daily steps of 2,000 steps/
day and the consumption of two servings of cereal a day as breakfast and
a snack to decrease intake. Substituting cereal for the usual breakfast and
snack choices in these families was felt to be an easy way to result in a small
decrease in caloric intake. Study families were taught about the America on
the Move small changes philosophy, and were provided with step counters
and tracking forms. The intervention was successful in increasing walking
and had a positive effect on the percentage body mass index or BMI-for-
age and percentage fat for overweight or at-risk-for-overweight children.
A positive effect was also seen in the BMI of the parent, however the effect
was mostly seen in girls and mothers. This study, based on the AOM pro-
gram, provided the first evidence to support the positive impact of simple
dietary changes and small increases in physical activity on the prevention of

excessive weight gain in children and general weight gain in adults.
The second more extensive AOM family study evaluated families over
a longer 6-month intervention [25]. The AOM approach for weight gain pre-
vention was evaluated in families with at least one child (7 to 14 years old)
who was overweight or at risk for overweight. These children were the pri-
mary targets of the intervention, with parents as the secondary targets. Fam-
ilies were randomly assigned to either the AOM group or the self-monitor
only group. Those families assigned to the AOM group (n ¼ 116) were
asked to make two small lifestyle changes: (1) use step counters (pedome-
ters) to increase walking by 2,000 steps/day above baseline; and (2) eliminate
100 kcals/day from their typical diet, with an emphasis on replacing typical
dietary sugar with a noncaloric sweetener. Families were taught about the
America on the Move small changes philosophy and were provided with
step counters, tracking forms, nutritional tips for cutting calories, and sim-
ple incentives. Those families assigned to the self monitor group (n ¼ 102)
were asked to use pedometers to record physical activity, but not to make
changes in diet or physical activity patterns. Over a 6-month period, both
groups of children showed significant decreases in BMI-for-age. However,
the AOM group, as compared with the self monitor group, had a signi-
ficantly greater percentage of target children who maintained or reduced
BMI for age (P!.05) and correspondingly, a significantly lower proportion
who increased BMI for age (Fig. 2). There was no significant weight gain
32.6
Increase %
BMI
47.2
AOM Target Kids

SM Target Kids
Maintain or 67.4
Reduce %
BMI 52.8
0 20 40 60 80
Percentage
Fig. 2. The Families on the Move (FOM) intervention group represented by the dark bars had
a significantly greater percentage of target children who maintained or reduced BMI for age
(P!.05) over the 6-month period. A significantly lower proportion of FOM children increased
BMI for age as compared with the self-monitoring (SM) group represented in the gray bars.
(Data from Rodearmel SJ, Wyatt HR, Stroebele N, et al. Small changes in dietary sugar and
physical activity as an approach to preventing excessive weight gain: the America on the
Move family study. Pediatrics 2007;120:e869–79.)
over the 6-month intervention in parents in either group. Both these studies
support the concept that the small changes approach advocated by America
on the Move holds promise for addressing childhood obesity by preventing
excess weight gain in families.
AOM funding
To provide information free of charge to consumer, the America on the
Move Foundation seeks donations and sponsorships to cover the cost of
developing and disseminating the AOM tools and program. The America
on the Move Foundation has received funding from government agencies,
other foundations, and from the private sector.
Summary
Promoting small achievable behavioral changes to patients offers a solu-
tion to stop weight gain and prevent new cases of the metabolic syndrome.
The America on the Move program was developed using the small change
approach to prevent weight gain in adults and prevent excessive weight
gain in children. A change in energy balance of 100 kcal could prevent
weight gain in the majority of individuals. AOM achieves the desirable small
change in energy balance through the promotion of small increases in steps
and small decreases in energy intake, using a participant’s personal baseline.
America on the Move has developed tools to aid people in making small life-
style changes that could decrease intake or increase expenditure by 100 kcal
and offers these tools free to participants. The tools include online resources,
interactive tools, community support, and events to support and encourage
these two small behavioral changes. Tools are available for individuals,
schools, worksites and health care providers. All the tools and programs
are free and available via the Web and can be accessed at http://aom.
americaonthemove.org.
References
[1] Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United
States, 1999–2004. JAMA 2006;295:1549–55.
[2] Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and trends in obesity among US adults,
1999–2000. JAMA 2002;288(14):1723–7.
[3] Lewis CE, Jacobs DR, McCreath H, et al. Weight gain continues in the 1990s: 10 year trends
in weight and overweight from the CARDIA study. Am J Epidemiol 2000;151:1172–81.
[4] Hill JO, Wyatt HR, Reed GW, et al. Obesity and the environment: where do we go from
here? Science 2003;299:853–5.
[5] Centers for Disease Control, National Center for Health Statistics. National health and
nutrition examination survey. Available at: http://www.cdc.gov/nchs/nhanes.htm. Accessed
March 2007.
[6] Williamson DF, Kahn HS, Remington PL, et al. The 10-year incidence of overweight and
major weight gain in US adults. Arch Intern Med 1990;150:665–72.
[7] Mokdad AH, Bowman BA, Ford ES, et al. The continuing epidemics of obesity and diabetes
in the United States. JAMA 2001;286:1195–200.
[8] Brown WJ, Williams L, Ford JH, et al. Identifying the energy gap: magnitude and determi-
nants of 5-year weight gain in midage women. Obes Res 2005;13(8):1431–41.
[9] Wang YC, Gortmaker SL, Sobol AM, et al. Estimating the energy gap among U.S. children:
a counterfactual approach. Pediatrics 2006;118(6):e1721–33.
[10] Ogden CL, Flegal KM, Carroll MD, et al. Prevalence and trends in overweight among US
children and adolescents, 1999–2000. JAMA 2002;288:1728–32.
[11] National Institutes of Health, National Heart Lung and Blood Institute. Clinical guidelines
on the identification, evaluation and treatment of overweight and obesity in adults: the evi-
dence report. Obes Res 1998;6(S2):51S–210S.
[12] Field AE, Coakley EH, Must A, et al. Impact of overweight on the risk of developing com-
mon chronic disease during a 10 yr period. Arch Intern Med 2001;161:1581–6.
US adults. Diabetes Care 2004;27:2444–9.
[14] Whitaker RC, Wyatt JA, Pepe MS, et al. Predicting obesity in young adulthood from child-
hood and parental obesity. N Engl J Med 1997;337:869–73.
[15] World Health Organization. Obesity: preventing and managing the global epidemic.
Geneva: World Health Organization; 1999.
[16] Wadden TA, Osei S. The treatment of obesity: an overview. In: Wadden TA, Stunkard AJ,
editors. Handbook of obesity treatment. New York: Guilford Press; 2002. 11. p. 229–248.
[17] Womble LG, Wang SS, Wadden TA. Commercial and self-help weight loss programs. In:
Wadden, Stunkard, editors. Handbook of obesity treatment. New York: Guilford Press;
2002. p. 395–415.
[18] Barlow S, Trowbridge F, Klish W, et al. Treatment of child and adolescent obesity: reports
from pediatricians, pediatric nurse practitioners, and registered dietitians. Pediatrics 2002;
110(S1):229–35.
[19] U.S. Department of Health and Human Services. Midcourse review healthy people 2010.
Available at: http://www.healthypeople.gov. Accessed March 2007.
[20] U.S. Department of Health and Human Services. Centers for Disease Control and Prevention.
National Center for Health Statistics, Hyattsville (MD). Available at: http://www.cdc.gov/
nchs/about/otheract/hpdata2010/focusareas/fa19-nutrition.htm. Accessed March 2007.
[21] US Department of Health and Human Services, Centers for Disease Control and Prevention,
National Center for Chronic Disease Prevention and Health Promotion. Physical activity
and health: a report of the surgeon general. Atlanta (GA): Centers for Disease Control
and Prevention; 1996.
[22] U.S. Department of Health and Human Services. Centers for Disease Control and Preven-
tion. National Center for Health Statistics, Hyattsville (MD). Available at: http://www.cdc.
gov/nchs/about/otheract/hpdata2010/focusareas/fa22-paf.htm. Accessed March 2007.
[23] AOM Foundation. America on the move. Available at: http://www.americaonthemove.org.
Accessed March 2007.
[24] Rodearmel SJ, Wyatt HR, Barry MJ, et al. A family-based approach to preventing excessive
weight gain. Obesity (Silver Spring) 2006;14:1392–401.
[25] Rodearmel SJ, Wyatt HR, Stroebele N, et al. Small changes in dietary sugar and physical
activity as an approach to preventing excessive weight gain: the America on the Move family
study. Pediatrics, in press.
[26] Wyatt HR, Peters JC, Reed GW, et al. Using electronic step counters to increase lifestyle
physical activity: Colorado on the move. J Phys Act Health 2004;1:181–90.
[27] Stroebele N, Stuht J, Catenacci V, et al. A small-changes approach to reducing energy intake.
Obesity 2006;14(9):A106.
Med Clin N Am 91 (2007) 1091–1105
Prevention of Diabetes Development

in Those with the Metabolic Syndrome
Traci Tupper, MD, Geetha Gopalakrishnan, MD*
Warren Alpert Medical School of Brown University, 1 Hoppin Street,
Suite 200, Providence, RI 02860, USA
The term metabolic syndrome is used to describe individuals who have

insulin resistance with a range of abnormalities in glucose levels (normal, im-
paired fasting glucose, impaired glucose tolerance, or overt diabetes), central
obesity, hypertension, elevated triglycerides, and low high-density lipoprotein
(HDL) cholesterol [1]. However, the specifics of the diagnostic criteria are
debated.
In the first criteria proposed by the World Health Organization in 1998, the
definition centered around the idea that these individuals are insulin-resistant
with either hyperinsulinemia, a fasting plasma glucose R110 mg/dL or a
2-hour post-glucose level R200 mg/dL (ie, after 75 gram oral glucose toler-
ance test). To meet criteria for metabolic syndrome, individuals also needed to
have at least two of the following: central obesity with male waist to hip ratio
O0.9 and female waist to hip ratio O0.5 and/or BMI O30 kg/m2; dyslipide-
mia defined as triglyceride R150 mg/dL or HDL !35 mg/dL; and blood pres-
sure R160/90 or receive treatment for hypertension [2].
In 2001, the Adult Treatment Panel (ATP) III published its guideline de-
fining metabolic syndrome [3–4]. This group focused their definition on risk
factors promoting cardiovascular disease rather than insulin resistance. In
their 2005 updated guideline, they lowered the fasting plasma glucose
threshold to 100 mg/dL and added population specific waist circumference
guidelines. Their current criteria for the diagnosis of metabolic syndrome is
three of the five following characteristics: (1) elevated waist circumference
(R102 cm in white, black and Hispanic men and R90 cm in Asian men,
R88 cm in white, black and Hispanic women and R80 cm in Asian
women); (2) triglycerides O150 mg/dL or treatment for elevated triglycer-
ides; (3) HDL !40 mg/dL in men and !50 mg/dL in women or treatment
E-mail address: ggopala@lifespan.org (G. Gopalakrishnan).
1092 TUPPER & GOPALAKRISHNAN
for low HDL; (4) blood pressure R130/85 or treatment for elevated blood
pressure; and (5) fasting plasma glucose R100 mg/dL or treatment for ele-
vated glucose. Other organizations including the International Diabetes
Federation (IDF) and American College of Endocrinology Task Force
have supported variations of the above diagnostic criteria. Regardless of
the focus of each group, impaired or abnormal glucose tolerance is an im-
portant component in defining metabolic syndrome in all published guide-
lines [5–7].
Fasting plasma glucose (FPG) level R126 mg/dL and 2-hour post-glu-
cose (PG) level R200 mg/dL correlate with the development of microvascu-
lar complications and therefore, these parameters have been chosen for the
diagnosis of diabetes. Meeting the following criteria on two occasions de-
fines diabetes: FPG R126 mg/dL; 2-h PG R200 mg/dL; or random PG
R200 mg/dL in the presence of symptoms. Individuals at high risk for de-
veloping diabetes can be classified to have impaired glucose tolerance
(IGT) or impaired fasting glucose (IFG). IGT is defined as a 2-h PG
R140 mg/dL and !200 mg/dL. IFG was initially defined as a fasting glu-
cose R110 mg/dL but less than 126 mg/dL. In 2003, the lower limit was
changed from 110 mg/dL to 100 mg/dL in order to optimize the sensitivity
and specificity of predicting future diabetes [8–9].
Several prospective observational studies have demonstrated metabolic
syndrome to be a risk factor for the development of type 2 diabetes. In non-
diabetics, metabolic syndrome was associated with a higher incidence of di-
abetes in both the Beaver Dam Study (OR 33.67; 95% CI 7.93-142.96) and
the West of Scotland Coronary Prevention Study (HR 24.4; 95% CI 7.53-
79.6) [10,11]. In a study of 890 nondiabetic Pima Indians, the relative risk
for incident diabetes was higher in individuals diagnosed with metabolic
syndrome as defined by the WHO criteria 3.58 (95% CI 2.56-5.00) com-
pared to the ATPIII criteria 2.09 (95% CI 1.49-2.92) [12]. This difference
may emphasize the importance of insulin resistance, the key component of
the WHO criteria, in the development of type 2 diabetes.
The focus of this chapter is on the prevention of type 2 diabetes in met-
abolic syndrome. Most studies reviewed in this section do not involve a study
population with a diagnosis of metabolic syndrome. However, factors such
as weight and blood glucose that play an important role in the diagnosis of
metabolic syndrome are defined. This article reviews available data regard-
ing the impact of lifestyle modification and drug therapies on the progres-
sion to diabetes in high risk individuals, such as those with hypertension,
dyslipidemia, obesity, and prediabetes (ie, IFG or IGT).
Lifestyle intervention
Genetic factors contribute to the development of insulin resistance and
impaired insulin secretion in type 2 diabetes. However, environmental and
behavioral factors influence the final outcome. In this section, the impact
PREVENTION OF DIABETES 1093
of lifestyle modifications such as diet, exercise, alcohol and smoking cessa-

tion on the progression to diabetes in high risk individuals such as those
with prediabetes and obesity will be reviewed.
Weight loss
Weight reduction can improve glucose tolerance and prevent progression
to type 2 diabetes [13–16]. In the Finnish Diabetes Prevention Study, 522 in-
dividuals with impaired glucose tolerance were assigned either to an inter-
vention group that received counseling on weight reduction, diet, and
exercise or to a control group [16]. The mean BMI was 33.2 kg/m2 at base-
line. Over a 2-year period, the intervention group lost 3.5 kg and the control
group lost 0.8 kg (P!.001). Intervention was associated with a 58% reduc-
tion in the incidence of diabetes after 4 years. Subjects who did not develop
diabetes at the conclusion of this study were followed for an additional 3
years without any intervention [17]. After the 7-year follow-up, the cumula-
tive incidence of diabetes was 43% lower in the intervention group. The de-
crease in the incidence of diabetes correlated with lifestyle changes, and
these changes were maintained in the intervention group even after counsel-
ing was discontinued.
The Diabetes Prevention Program (DPP) also demonstrated the benefit
of lifestyle intervention [13]. 3234 obese subjects with impaired glucose tol-
erance were randomly assigned to receive lifestyle intervention (intensive
diet and 150 minutes of physical activity per week to achieve a 7% weight
reduction goal), pharmacological intervention (Metformin 850 mg BID),
or placebo [18]. The mean BMI of the study participants was 34 kg/m2.
The lifestyle intervention group lost 7% (6.8 kg) of weight in the first year
and maintained the loss for the 3-year duration of the study. The incidence
of diabetes was lower with lifestyle intervention than with either metformin
or placebo. A 58% reduction in the incidence of diabetes was noted with life-
style intervention compared with placebo. This risk reduction correlated with
the success of achieving weight loss goals. In the lifestyle intervention group,
a weight loss of 1 kg reduced the risk of diabetes by 16% [19]. Greater im-
provements in insulin sensitivity and beta-cell function were seen with lifestyle
intervention than with either metformin or placebo. These improvements may
have contributed to greater diabetes risk reduction with lifestyle intervention
[20].
The lifestyle intervention employed in the DPP included 16 sessions of core
curriculum of self-management strategies, a supervised physical activity pro-
gram, and lifestyle coaches to help achieve and maintain weight loss goals. En-
rolling seven people into this intensive intervention program is estimated to
prevent one case of diabetes in three years. However, the cost-effectiveness
of the DPP intensive lifestyle intervention program can be debated. One study
using the Markov model deduced that DPP lifestyle intervention may delay
the development of diabetes by 11 years, reduce the absolute incidence of
diabetes by 20% and improve survival by 0.5 years compared to placebo [21].
The cost per quality-adjusted life-year (QALY) was lower with lifestyle inter-
vention ($1100) than metformin therapy ($31,300). However, another study
using the Archimedes model estimated the cost/QALY over 30 years for the
DPP lifestyle intervention to be $62,600, for metformin to be $35,400, and
to delay lifestyle intervention until the onset of diabetes to be $24,500 [22]. Ac-
cording to the assumptions made in this model, DPP lifestyle intervention
would reduce the chance of getting diabetes from 72% to 61% in high risk in-
dividuals. Most experts agree that lifestyle modification is critical for diabetes
prevention. However, cost may limit national implementation of lifestyle in-
tervention programs like the DPP. Finding less expensive methods to achieve
and maintain weight loss goals is crucial for diabetes prevention.
Exercise
Independent of weight loss, exercise may reduce the risk of diabetes. In
a prospective study of 5990 men followed for over 10 years, 3.3% developed
type 2 diabetes [23]. Regular exercise was independently associated with
a lower incidence of diabetes. In fact, a 500 kcal increase in weekly energy
expenditure reduced the risk for diabetes by 6%. In this study, individuals
with the highest risk for diabetes (ie, high body mass index, hypertension,
or family history of diabetes) benefited most from exercise.
A Finnish study of 897 non-diabetic men followed for an average of 4.2
years also corroborated the above findings [24]. In this study, men who par-
ticipated in moderate physical activity (at least 40 minutes per week) had
a lower risk of developing diabetes than men who did not participate in
these activities (odds ratio 0.44; 95% CI 0.22–0.88). This risk reduction
was independent of age, baseline glucose levels, BMI, triglyceride levels,
family history, and alcohol intake. A subset analysis of high-risk men
who were overweight, hypertensive, or had a positive family history for di-
abetes found a 64% reduction in diabetes risk with moderate physical activ-
ity. Greater reductions in diabetes risk were noted with greater levels of
physical activity in this study [25].
In patients with IGT, the benefits of exercise are similar [26,27]. One hun-
dred eighty-one men with IGT who participated in a 6-year diet-plus-regu-
lar-exercise program were compared with a reference group of 79 men who
had similar baseline characteristics but who were not enrolled in the pro-
gram [26]. After 5 years, glucose tolerance had improved and the incidence
of diabetes was lower in the exercise group compared with controls (11%
versus 29%). Similar results were noted in the DPP, which combined diet
and exercise [13].
Diet
The risk of developing diabetes may also be affected by the composition
of the diet. Different types of food have different effects on postprandial
gylcemia, and this is referred to as the glycemic index. Glucose load takes
into account the carbohydrate content of the food as well as its glycemic in-
dex. In addition to glycemic index and glucose load, fiber content and type
of fatdsuch as saturated, monounsaturated, and polyunsaturateddmay
also impact the risk of developing diabetes.
In the Nurses’ Health Study, a diet high in fiber and polyunsaturated fat,
and low in trans fat and glycemic load, was associated with a lower risk of
developing diabetes in women [28]. In fact, the risk of developing diabetes
increased with increasing glycemic index in this study. This diet had a greater
impact on reducing diabetes risks among minorities (relative risk [RR] 0.54;
CI 0.39–0.73) than among whites (RR 0.77; CI 0.72–0.84). These results
were also confirmed in men who participated in the Health Professionals
Study. This study evaluated the effect of a Western diet characterized by
high consumption of red meat, processed meat, high-fat dairy products,
sweets, and desserts versus a healthier diet characterized by higher consump-
tion of vegetables, fruit, fish, poultry, and whole grains [29,30]. A Western
diet was associated with an increased risk of developing diabetes (RR 1.6,
CI 1.3–1.9), and this risk was even greater in men with a BMI of 30 kg/
m2 (RR 11.2; CI 8.07–15.6).
Intake of cereal fibers may independently improve insulin sensitivity [31].
In a prospective cohort study of nearly 25,000 men and women, high intake
of cereal fiber lowered the risk of diabetes (RR 0.72; 95% CI 0.56-0.93), and
this association was corroborated by a meta-analysis of nine cohort studies
(RR 0.67; 95% CI 0.62-0.72) [32]. Other food products, such as nuts and
dairy, have also been associated with a reduced risk of developing diabetes
in observational studies [33,34].
Coffee
Studies have shown an inverse relationship between the consumption of
coffee and the risk of developing diabetes. Meta-analysis of nine cohort
studies found that the risk of diabetes was lower with greater coffee con-
sumption [35]. In fact, the relative risk was 0.65 (95% CI 0.54-0.78) for those
who consumed greater than 6 cups and 0.72 (95% CI 0.62-0.83) for those
who consumed 4 to 6 cups compared to those who consumed less than 2
cups of coffee daily. Even one cup per day reduced diabetes risk compared
to non-drinkers in the Nurses Health Study (RR 0.87; 95% CI 0.73-1.03)
[36]. This association has also been shown in individuals with IGT who
are current (RR 0.31; 95% CI 0.11-0.87) and past coffee (RR 0.36; 95%
CI 0.16-0.83) consumers [37]. Similar inverse associations are also seen
with decaffeinated coffee and green teas [38].
Alcohol
Moderate drinkers have improved insulin sensitivity and lower plasma
insulin levels compared with nondrinkers [39]. The risk of developing
diabetes is lower with moderate alcohol consumption. The relative risk for
diabetes was 0.61 (95% CI 0.44–0.91) in men who drink two to three alco-
holic beverages per day and 0.4 (95% CI 0.3–0.6) in women who drink one
or more alcoholic beverages per day compared with nondrinkers [40,41]. A
meta-analysis of 15 cohort studies also found a decreased risk for diabetes
with light to moderate alcohol consumption but not heavy [42].
Smoking
Smoking has been shown to increase blood glucose after OGTT, impair
insulin sensitivity, and increase abdominal fat distribution resulting in
greater waist-to-hip ratio [43–45]. The risk of developing diabetes is greater
in smokers versus nonsmokers. This risk increases with quantity and dura-
tion of smoking. In the Physicians Health Study, the relative risk of devel-
oping diabetes as a result of smoking 1 pack per day was 1.7 (95% CI 1.3 to
2.3), smoking less than 1 pack per day was 1.5 (95% CI 1.0 to 2.2), and hav-
ing a past history of smoking was 1.1 (95% CI 1.0 to 1.4) compared to never
smoking [46]. The risk of diabetes increases with total pack years of use
(p for trend !0.001) and with passive or secondhand smoking (RR 1.35,
95% CI 1.06 to 1.71) [46,47]. Smoking cessation also increases diabetes
risk as a result of weight gain, especially in the initial five years [48]. How-
ever, after five years, the risk of developing diabetes decreases with smoking
cessation and reverts to that of never smokers after 20 years.
Pharmacologic therapy
When lifestyle intervention fails, pharmacologic therapies may be helpful
in delaying or preventing diabetes in high-risk individuals. The effect of oral
hypoglycemic agents, antiobesity agents, antilipid agents, antihypertensive
agents, and estrogen on the development of type 2 diabetes is reviewed in
this section [49].
Biguanides
Insulin resistance and beta-cell dysfunction is critical to the development of
type 2 diabetes. Metformin has been shown to improve insulin sensitivity and
preserve beta-cell function [20]. Several studies have looked at the effect of
metformin on the prevention of diabetes [50–52]. Most of these studies were
underpowered and therefore found no significant reduction in the rate of pro-
gression to diabetes with metformin therapy. However, different conclusions
can be drawn from randomized controlled trials with larger sample sizes.
The Biguanides and Prevention of Risks in Obesity (BIGPRO1) study
randomized 324 men with metabolic syndrome (ie, high waist-to-hip ratio)
to receive metformin or placebo [52]. After 1 year, individuals in the metfor-
min arm had greater weight loss, lower fasting blood glucose, and lower
fasting plasma insulin concentrations. In the DPP, 2155 individuals with

IGT were randomized to receive metformin (850 mg twice a day) or placebo
[13]. After an average follow-up of 2.8 years, a 31% reduction in the inci-
dence of diabetes was noted in the metformin-treated group compared
with placebo. To prevent one case of diabetes, 14 individuals with IGT
need to receive metformin therapy. The cost-effectiveness may vary with
age and associated risk factors, since the greatest impact of metformin ther-
apy was primarily observed in individuals under 60 years old and those with
a BMI greater than 35 kg/m2 [21].
At the conclusion of DPP, study participants who did not develop diabe-
tes underwent a second OGTT after a medication washout period of 1 to 2
weeks [53]. Approximately 80% of eligible subjects participated. The inci-
dence of diabetes increased in both the metformin (25.2%–30.6%) and
placebo (33.4%–36.7%) arms after an average washout period of 11 days.
A combined analysis of data from the original DPP and this washout study
showed a persistent 25% reduction in the incidence of diabetes with metfor-
min therapy compared with placebo. Although metformin seems to delay
the progression of diabetes in high risk individuals on therapy, conclusions
about long-term prevention cannot be made based on available data. More
studies with longer follow-up periods are needed to draw any definitive
conclusions.
Thiazolidinediones
Thiazolidinediones have been shown to enhance insulin sensitivity, re-
duce insulin secretory demand, and preserve beta cell function in diabetes.
Only a few randomized control studies evaluate the efficacy of these agents
in preventing type 2 diabetes.
In the TRIPOD trial (Troglitazone in Prevention of Diabetes), 266 His-
panic women with previous gestational diabetes were randomized to trogli-
tazone (400 mg/day) or placebo [54]. After median follow-up of 30 months,
a 55% reduction in the incidence of diabetes was seen in the treatment arm.
Troglitazone improved insulin sensitivity and preserved beta cell function.
The metabolic parameters that are altered with troglitazone therapy persist
long after drug discontinuation. In this study, the annual incidence of dia-
betes remained significantly lower with troglitazone therapy than placebo,
even after an 8-month drug washout period (3.1% vs. 21.2%). A limitation
of this study was the 33% attrition rate.
Women who completed TRIPOD and who did not develop diabetes were
offered participation in PIOPD (The Pioglitazone in Prevention of Diabetes)
study [55]. 89 of the 95 eligible participants enrolled in this study, and 65
completed all study visits. After a three-year drug treatment followed by
a six-month washout, pioglitazone stopped the decline in beta-cell function
in the placebo arm of the TRIPOD and maintained beta-cell function in the
troglitazone arm of TRIPOD. This suggests that the benefits of these drugs
may be a class effect. The Actos Now for Prevention of Diabetes (ACT-
NOW) is evaluating the effect of pioglitazone in the prevention of diabetes
in 600 subjects with IGT. The results of this study should expand our under-
standing of this class of agents even further.
The class effect of thiazolidinediones treatment is further corroborated by
a cohort study of 172 subjects with IGT treated with troglitazone 400 mg
daily followed by pioglitazone 30 mg or rosiglitazone 4 mg daily versus
an untreated comparison group [56]. These agents reduced risk for diabetes
by 89% in high risk individuals. However, their side effect profile may limit
their use in diabetes prevention.
In the DREAM trial (Diabetes Reduction Assessment with Ramipril
and Rosiglitazone Medication), 5269 subjects with impaired fasting glucose
(IFG) or impaired glucose tolerance (IGT) received rosiglitazone (8 mg
daily) or placebo [57]. After a 3-year follow-up, a 62% reduction in the
incidence of diabetes was seen in the treatment arm. Data on the incidence
of diabetes after a drug washout period are not yet available. Although dif-
ferences in mortality rates were not seen between the two treatment arms,
rosiglitazone was associated with the development of heart failure (0.5%)
compared to placebo (0.1%). In the DPP, study participants randomized
to receive troglitazone were less likely to develop diabetes compared to
placebo [58]. This arm was discontinued due to liver safety concerns,
and this drug is no longer available in the United States or United
Kingdom.
Despite impressive risk reduction with thiazolidinediones treatment, drug
treatment is not recommended at this time for the prevention of diabetes.
Studies assessing long-term diabetes incidence, adverse events, and cost-
effectiveness are still needed.
Alpha-glucosidase inhibitors
Acarbose significantly reduces postprandial hyperglycemia. Although
this reduction is not associated with a delay in progression to frank diabe-
tes in individuals with early diabetes, it does reduce the rate of diabetes de-
velopment in individuals with IFG or IGT [59,60]. Once diabetes develops,
the reversibility of beta-cell failure with acarbose therapy seems to be
limited.
In the Study to Prevent Non-Insulin-Dependent Diabetes Mellitus
(STOP-NIDDM), 1429 subjects with impaired glucose tolerance were ran-
domly assigned to acarbose (100 mg three times daily) or placebo [60]. In
this intention to treat analysis, the incidence of diabetes was lower with acar-
bose therapy (32%) compared with placebo (42%) after a 3-year follow-up
period (relative hazard [RH] 0.75, 95% CI 0.63–0.90). Drug-induced gastro-
intestinal complaints such as flatulence were responsible for an early drop-
out rate of 25% in the treatment group. After a 3-month washout, 15%
developed diabetes with acarbose and 10.5% with placebo. Furthermore,
in this study acarbose treatment was also associated with reduced cardiovas-
cular events (RH 0.51, 95% CI 0.2–0.95).
Sulfonylureas
Sulfonylureas are insulin secreatagogues. Subjects with IGT were ran-
domly assigned to tolbutamide or placebo in two studies [61,62]. Neither
study detected a statistically significant decrease in the incidence of diabetes
with treatment. These studies were underpowered to detect a difference.
Orlistat
Orlistat is an antiobesity medication. In the XENDOS (Xenical in the
Prevention of Diabetes in Obese Subjects) study, 3305 obese subjects with
IGT were randomized to orlistat (360 mg) or placebo [63]. A 2.8-kg weight
reduction and a 37% reduction in the incidence of diabetes were noted with
orlistat treatment compared with placebo. The dropout rate was 57%, with
91% of the treatment group reporting gastrointestinal side effects compared
with 65% of the placebo group. However, a pooled analysis of three ran-
domized controlled trials involving 642 obese subjects did not report a signif-
icant reduction in type 2 diabetes development with orlistat versus placebo
[64].
Inhibition of angiotensin II
The exact role of these agents in diabetes prevention has not been fully
elucidated, however, secondary analysis of several randomized control trials
report angiotensin converting enzyme (ACE) inhibitors and angiotensin II
receptor blockers (ARBs) significantly reduce the incidence of type 2 diabe-
tes [65–69]. Meta-analysis of 12 secondary analysis trials involving ACE in-
hibitors and ARBs showed a 25% decrease in the incidence of diabetes with
these agents (14.3 versus 17.4 cases per 1000 patient years) [70]. These results
were corroborated by a second meta-analysis of 11 studies involving patients
with hypertension, coronary disease, or heart failure. In this analysis, ACE
inhibitors and ARBs significantly reduced (0.78, 95% CI 0.73-0.831) the risk
of type 2 diabetes irrespective of the indication for the use of these agents
[71]. In both meta-analysis, diabetes was not the primary end point for
the studies included for analysis. The DREAM trial, however, evaluated
the effect of ramapril on the incidence of diabetes as a primary outcome
[72]. The trial’s 5269 subjects with IFG or IGT received ramipril (15 mg/
d) or placebo. After 3 years, ramapril did not decrease the incidence of di-
abetes (RR 0.91, 95% CI 0.81–1.03). Ramipril did lower plasma glucose 2
hours after a glucose load, but it did not improve fasting glucose.
The discrepancy in findings between the DREAM trial and previous
studies may be explained by the fact that diabetes was not the primary
end point in the prior studies. Therefore, participants may not have been ad-
equately screened for diabetes at baseline or at the study conclusion, and
this may have contributed to some of the differences.
Antilipid agents
Antilipid agents are not routinely recommended for the prevention of di-
abetes; however, high risk individuals such as those with metabolic syn-
drome may receive treatment with these agents for their hyperlipidemia.
Post hoc analysis of the West of Scotland Coronary Prevention Study (WO-
SCOPS) found that the incidence of diabetes was significantly lower with
pravastatin treatment (RR 0.70, 95% CI 0.50–0.99). However, analysis of
other studies involving statins did not find a statistically significant reduc-
tion in the incidence of diabetes [73–76].
In the BIP (Bezafibrate Infraction Prevention) trial, 303 subjects with
IGT were randomized to benzafibrate or placebo [77]. Treatment was asso-
ciated with a 30% reduction in the incidence of diabetes in a post hoc
analysis.
Estrogen therapy
The Heart and Estrogen/Progestin Replacement Study (HERS) evalu-
ated the effect of combined estrogen and progestin therapy (conjugated
estrogen 0.625 mg with medroxyprogesterone 2.5 mg daily) or placebo on
recurrent cardiac events in postmenopausal women with established coro-
nary heart disease [78]. In a post hoc analysis for the development of type
2 diabetes, 2029 women with normal or impaired glucose tolerance were
evaluated. After an average follow-up of 4.1 years, the incidence of type 2
diabetes was 6.2% in the treatment group compared with 9.5% in the pla-
cebo group (RH 0.6; 95% CI 0.5–0.9). To prevent one case of diabetes, 30
postmenopausal need to be treated with hormone replacement.
The Women’s Health Initiative reported similar findings [79]. Self-report
of treatment with insulin or oral hypoglycemic medications was obtained
in the 15,641 postmenopausal women who participated in this study. The
incidence of treated diabetes was 3.5% in the hormone-treated group (conju-
gated estrogen 0.625 mg with medroxyprogesterone 2.5 mg daily) and 4.2%
in the placebo group (RH 0.79, 95% CI 0.7–0.9) after a mean follow-up of 5.6
years. However, postmenopausal hormone replacement therapy increased
the risk for stoke, heart disease, and breast cancer in this study, and these
side effects limit the use of this agent for the prevention of diabetes.
Summary
Over the last several years, our knowledge of diabetes and metabolic syn-
drome has grown tremendously. The definitions for prediabetes and
diabetes as well as metabolic syndrome have been modified numerous times

and will continue to be re-evaluated as we learn more about their underlying
pathophysiology. Preliminary data reviewed in this article support the
notion that lifestyle intervention and pharmacologic therapies may influence
progression to diabetes. However, it is important to keep in mind that many
of these studies were secondary analyses of cohort studies and only a few
were randomized controlled trials. All were limited by the duration of fol-
low-up, and therefore the long-term implications for diabetes prevention,
cost-effectiveness, and adverse outcome are unknown. Furthermore, conclu-
sions about causality need to be made with caution, especially if randomized
controlled data are not available. More studies are needed to clarify the
cost-effectiveness of available disease prevention strategies and to develop
new cost-effective therapeutic options.
References
[1] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365(9468):
1415–28.
[2] Alberti K, Zimmet P. Definition, diagnosis and classification of diabetes mellitus and its
complications. Part 1: Diagnosis and classification of diabetes mellitus provisional report
of a WHO consultation. Diabet Med 1998;15(7):539–53.
[3] Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults.
Executive summary of the third report of the National Cholesterol Education Program
(NCEP) expert panel on detection, evaluation and treatment of high blood cholesterol in
adults (Adult Treatment Panel III). JAMA 2001;285(19):2486–97.
[4] Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syn-
drome: an American Heart Association/National Heart, Lung and Blood Institute scientific
statement. Circulation 2005;112(17):2735–52.
[5] American College of Endocrinology Task Force on the Insulin Resistance Syndrome. Amer-
ican College of Endocrinology position statement on the insulin resistance syndrome.
Endocr Pract 2003;9(3):236–52.
[6] Alberti KG, Zimmet P, Shaw J. The metabolic syndrome: a new worldwide definition. Lan-
cet 2005;366(9491):1059–62.
[7] Balkau B, Charles MA. Comment on the provisional report from WHO consultation. Diabet
Med 1999;16(5):442–3.
[8] The expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care
1997;20(7):1183–97.
[9] The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Follow-up
report on the diagnosis of diabetes mellitus. Diabetes Care 2003;26(11):3160–7.
[10] Hanson RL, Imperatore G, Bennett PH, et al. Components of the ‘‘metabolic syndrome’’
and incidence of type 2 diabetes. Diabetes 2002;51:3120–7.
[11] Klein BE, Klein R, Lee KE. Components of the metabolic syndrome and risk of cardiovas-
cular disease and diabetes in Beaver Dam. Diabetes Care 2002;25:1790–8.
protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coro-
nary Prevention Study. Circulation 2003;108:414–20.
[13] Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2
diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346(6):393–403.
[14] Long SD, O’Brien KO, MacDonald KD Jr, et al. Weight loss in severely obese subjects pre-
vents the progression of impaired glucose tolerance to Type II diabetes. A longitudinal inter-
ventional study. Diabetes Care 1994;17(5):372–5.
[15] Wing RR, Venditti E, Jakicic JM, et al. Lifestyle intervention in overweight individuals with
family history of diabetes. Diabetes Care 1998;21(3):350–9.
344(18):1343–50.
[17] Lindstrom J, Ilanne-Parikka P, Peltonen M, et al. Sustained reduction in the incidence of
type 2 diabetes by lifestyle intervention: follow-up of the Finnish Diabetes Prevention Study.
Lancet 2006;368(9548):1673–9.
[18] The Diabetes Prevention Program (DPP): description of lifestyle intervention. Diabetes Care
2002;259(12):2165–71.
[19] Hamman RF, Wing RR, Edelstein SL, et al. Effect of weight loss with lifestyle intervention
on risk of diabetes. Diabetes Care 2006;29(9):2102–7.
[20] Kitabchi AE, Temprosa M, Knowler WC, et al. Role of insulin secretion and sensitivity in
the evolution of type 2 diabetes in the Diabetes Prevention Program: effects of lifestyle inter-
vention and metformin. Diabetes 2005;54(8):2404–14.
[21] Herman WH, Hoerger TJ, Brandle M, et al. The cost-effectiveness of lifestyle modification or
metformin in preventing type 2 diabetes in adults with impaired glucose tolerance. Ann
Intern Med 2005;142(5):323–32.
[22] Eddy DM, Schlessinger L, Kahn R. Clinical outcomes and cost-effectiveness of strategies for
managing people at high risk for diabetes. Ann Intern Med 2005;143(4):251–64.
[23] Helmrich SP, Ragland DR, Leung RW, et al. Physical activity and reduced occurrence of
non-insulin-dependent diabetes mellitus. N Engl J Med 1991;325(3):147–52.
[24] Lynch J, Helmrich SP, Lakka TA, et al. Moderately intense physical activities and high levels
of cardiorespiratory fitness reduce the risk of non-insulin-dependent diabetes mellitus in
middle-aged men. Arch Intern Med 1996;156(12):1307–14.
[25] Hu FB, Sigal RJ, Rich-Edwards JW, et al. Walking compared with vigorous physical
activity and risk of type 2 diabetes in women: a prospective study. JAMA 1999;
282(15):1433–9.
[26] Eriksson KF, Lindgarde F. Prevention of type 2 (non-insulin-dependent) diabetes mellitus
by diet and physical exercise: the 6-year Malmo feasibility study. Diabetologia 1991;
34(12):891–8.
[27] Hu G, Lindstrom J, Valle TT, et al. Physical activity, body mass index, and risk of type 2
diabetes in patients with normal or impaired glucose regulation. Arch Intern Med 2004;
164(8):892–6.
[28] Shai I, Jiang R, Manson JE, et al. Ethnicity, obesity, and risk of type 2 diabetes in women:
a 20-year follow-up study. Diabetes Care 2006;29(7):1585–90.
[29] van Dam RM, Rimm EB, Willett WC, et al. Dietary patterns and risk for type 2 diabetes
mellitus in U.S. men. Ann Intern Med 2002;136(3):201–9.
[30] van Dam RM, Willett WC, Rimm EB, et al. Dietary fat and meat intake in relation to risk of
type 2 diabetes in men. Diabetes Care 2002;25(3):417–24.
[31] Weickert MO, Mohlig M, Schofl C, et al. Cereal fiber improves whole-body insulin sensitiv-
ity in overweight and obese women. Diabetes Care 2006;29(4):775–80.
[32] Matthias BS, Schulz M, Heidemann C, et al. Fiber and magnesium Intake and Incidence of
Type 2 Diabetes. Arch Intern Med 2007;167:956–65.
[33] Pereira MA, Jacobs DR Jr, Van Horn L, et al. Dairy consumption, obesity, and the insulin
resistance syndrome in young adults: The CARDIA Study. JAMA 2002;287(16):2081–9.
[34] Jaing R, Manson JE, Stampfer MJ, et al. Nuts and peanut butter consumption and risk of
type 2 diabetes in women. JAMA 2002;345:790–6.
[35] van Dam RM, Hu FB. Coffee consumption and risk of type 2 diabetes: a systematic review.
JAMA 2005;294(1):97–104.
[36] van Dam RM, Willett WC, Manson JE, et al. Coffee, caffeine, and risk of type 2 diabetes:
a prospective cohort study in younger and middle-aged U.S. women. Diabetes Care 2006;
29(2):398–403.
[37] Smith B, Wingard DL, Smith TC, et al. Does coffee consumption reduce the risk of type 2
diabetes in individuals with impaired glucose? Diabetes Care 2006;29(11):2385–90.
[38] Iso H, Date C, Wakai K, et al. The relationship between green tea and total caffeine intake
and risk for self-reported type 2 diabetes among Japanese adults. Ann Intern Med 2006;
144(8):554–62.
[39] Davies MJ, Baer DJ, Judd JT, et al. Effects of moderate alcohol intake on fasting insulin and
glucose concentrations and insulin sensitivity in postmenopausal women: a randomized con-
trolled trial. JAMA 2002;287(19):2559–62.
[40] Rimm EB, Chan J, Stampfer MJ, et al. Prospective study of cigarette smoking, alcohol use,
and the risk of diabetes in men. BMJ 1995;310(6979):555–9.
[41] Stampfer MJ, Colditz GA, Willett WC, et al. A prospective study of moderate alcohol drink-
ing and risk of diabetes in women. Am J Epidemiol 1988;128(3):549–58.
[42] Koppes LL, Dekker JM, Hendriks HF, et al. Moderate alcohol consumption lowers the risk
of type 2 diabetes: a meta-analysis of prospective observational studies. Diabetes Care 2005;
28(3):719–25.
[43] Janzon L, Berntorp K, Hanson M, et al. Glucose tolerance and smoking: a population study
of oral and intravenous glucose tolerance tests in middle-aged men. Diabetologia 1983;25(2):
86–8.
[44] Frati AC, Iniestra F, Ariza CR. Acute effect of cigarette smoking on glucose tolerance and
other cardiovascular risk factors. Diabetes Care 1996;19(2):112–8.
[45] Shimokata H, Muller DC, Andres R. Studies in the distribution of body fat III: Effects of
cigarette smoking. JAMA 1989;261(8):1169–73.
[46] Houston TK, Person SD, Pletcher MJ, et al. Active and passive smoking and development of
glucose intolerance among young adults in a propective cohort: CARDIA study. BMJ 2006;
332(7549):1064–9.
[47] Manson JE, Ajani UA, Liu S, et al. A prospective study of cigarette smoking and the inci-
dence of diabetes mellitus among US male physicians. Am J Med 2000;109(7):538–42.
[48] Wannamethee SG, Shaper AG, Perry IJ. Smoking as a modifiable risk factor for type 2
diabetes in middle-aged men. Diabetes Care 2001;24(9):1590–5.
[49] Padwal R, Majumdar SR, Johnson JA, et al. A systematic review of drug therapy to delay or
prevent type 2 diabetes. Diabetes Care 2005;28(3):736–44.
[50] Yang W, Lin L, Qi J. The preventive effect of acarbose and metformin on the IGT population
from becoming diabetes mellitus: a 3-year multicentre prospective study. Chinese Journal
of Endocrinology and Metabolism 2001;17:131–6.
[51] Li CL, Pan CY, Lu JM, et al. Effect of metformin on patients with impaired glucose toler-
ance. Diabet Med 1999;16(6):477–81.
[52] Fontbonne A, Charles MA, Juhan-Vague I, et al. The effect of metformin on the meta-
bolic abnormalities associated with upper-body fat distribution. Diabetes Care 1996;19(9):
920–6.
[53] Effects of withdrawal from metformin on the development of diabetes in the Diabetes Pre-
vention Program. Diabetes Care 2003;26(4):977–80.
[54] Buchanan TA, Xiang AH, Peters RK, et al. Preservation of pancreatic beta-cell function and
prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk
Hispanic women. Diabetes 2002;51(9):2796–803.
[55] Xiang AH, Peters RK, Kjos SL, et al. Effect of pioglitazone on pancreatic beta-cell function
and diabetes risk in Hispanic women with prior gestational diabetes. Diabetes 2006;55(2):
517–22.
[56] Durbin RJ. Thiazolidinedione therapy in the prevention/delay of type 2 diabetes in patients
with impaired glucose tolerance and insulin resistance. Diabetes Obes Metab 2004;6(4):
280–5.
[57] Gerstein HC, Yusuf S, Bosch J, et al. Effect of rosiglitazone on the frequency of diabetes in
patients with impaired glucose tolerance or impaired fasting glucose: a randomised con-
trolled trial. Lancet 2006;368(9541):1096–105.
[58] Knowler WC, Hamman RF, Edelstein SL, et al. The DPP research group: prevention of type
2 diabetes with troglitazone in the Diabetes Prevention Program. Diabetes 2005;54(4):
1150–6.
[59] Kirkman MS, Shankar RR, Shankar S, et al. Treating postprandial hyperglycemia does not
appear to delay progression of early type 2 diabetes: the Early Diabetes Intervention Pro-
gram. Diabetes Care 2006;29(9):2095–101.
[60] Chiasson JL, Josse RG, Gomis R, et al. Acarbose for prevention of type 2 diabetes mellitus:
the STOP-NIDDM randomised trial. Lancet 2002;359(9323):2072–7.
[61] Sartor G, Schersten B, Carlstrom S, et al. Ten-year follow-up of subjects with impaired glu-
cose tolerance: prevention of diabetes by tolbutamide and diet regulation. Diabetes 1980;
29(1):41–9.
[62] Keen H, Jarrett RJ, Ward JD, et al. Borderline diabetics and their response to tolbutamide.
Adv Metab Disord 1973;2(Suppl 2):521–31.
[63] Torgerson JS, Hauptman J, Boldrin MN, et al. Xenical in the prevention of diabetes in obese
subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle changes
for the prevention of type 2 diabetes in obese patients. Diabetes Care 2004;27(1):155–61.
[64] Heymsfield SB, Segal KR, Hauptman J, et al. Effects of weight loss with orlistat on glucose
tolerance and progression to type 2 diabetes in obese adults. Arch Intern Med 2000;160(9):
1321–6.
[65] Yusuf S, Gerstein H, Hoogwerf B, et al. Ramipril and the development of diabetes. JAMA
2001;286(15):1882–5.
[66] 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(23):
2981–97.
[67] Dahlof B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the
Losartan Intervention for Endpoint reduction in hypertension study (LIFE): a randomised
trial against atenolol. Lancet 2002;359(9311):995–1003.
[68] Yusuf S, Ostergren JB, Gerstein HC, et al. Effects of candesartan on the development of
a new diagnosis of diabetes mellitus in patients with heart failure. Circulation 2005;112(1):
48–53.
[69] Vermes E, Ducharme A, Bourassa MG, et al. Enalapril reduces the incidence of diabetes in
patients with chronic heart failure: insight from the Studies of Left Ventricular Dysfunction
(SOLVD). Circulation 2003;107(9):1291–6.
[70] Abuissa H, Jones PG, Marso SP, et al. Angiotensin-converting enzyme inhibitors or Angio-
tensin receptor blockers for prevention of type 2 diabetes: a meta-analysis of randomized
clinical trials. J Am Coll Cardiol 2005;46(5):821–6.
[71] Gillespie EL, White CM, Kardas M, et al. The impact of ACE inhibitors or angiotensin II
type 1 receptor blockers on the development of new-onset type 2 diabetes. Diabetes Care
2005;28(9):2261–6.
[72] Bosch J, Yusuf S, Gerstein HC, et al. Effect of ramipril on the incidence of diabetes. N Engl
J Med 2006;355(15):1551–62.
[73] Freeman DJ, Norrie J, Sattar N, et al. Pravastatin and the development of diabetes mellitus:
evidence for a protective treatment effect in the West of Scotland Coronary Prevention
Study. Circulation 2001;103(3):357–62.
[74] Collins R, Armitage J, Parish S, et al, Heart Protection Study Collaborative Group. MRC/
BHF Heart Protection Study of cholesterol-lowering with simvastatin in 5963 people with
diabetes: a randomized placebo-controlled trial. Lancet 2003;361(9374):2005–16.
[75] Keech A, Colquhoun D, Best J, et al. Secondary prevention of cardiovascular events with
long-term pravastatin in patients with diabetes or impaired fasting glucose. Diabetes Care
2003;26(10):2713–21.
[76] Sever PS, Dahlof B, Poulter NR, et al. Prevention of coronary and stroke events with
atorvastatin in hypertensive patients who have average or lower-than-average cholesterol
concentrations, in the Anglo-Scandinavian Cardiac Outcomes Trial–Lipid Lowering

Arm (ASCOT-LLA): a multicentre randomised controlled trial. Drugs 2004;64(Suppl
2):43–60.
[77] Tenenbaum A, Motro M, Fisman EZ, et al. Peroxisome proliferators-activated receptor
ligand Bezafibrate for prevention of type 2 diabetes mellitus in patients with coronary artery
disease. Circulation 2004;109(18):2197–202.
[78] Kanaya AM, Herrington D, Vittinghoff E, et al. Glycemic effects of postmenopausal
hormone therapy: the heart and estrogen/progestin replacement study. A randomized, dou-
ble-blind, placebo-controlled trial. Ann Intern Med 2003;138(1):1–9.
[79] Margolis KL, Bonds DE, Rodabough RJ, et al. Effect of estrogen plus progestin on the
incidence of diabetes in postmenopausal women: results from the Women’s Health Initiative
Hormone Trial. Diabetologia 2004;47(7):1175–87.
Med Clin N Am 91 (2007) 1107–1123
Metabolic Syndrome and Type 2

Diabetes: Can We Stop the Weight Gain
with Diabetes?
Denise Joffe, MD, Robert T. Yanagisawa, MD*
Division of Endocrinology, Diabetes, and Bone Diseases, Mount Sinai School of Medicine,
One Gustave L. Levy Place, Box 1055, New York, NY 10029–6574, USA
Many patients with type 2 diabetes also have the metabolic syndrome
with its cardinal features of central adiposity, insulin resistance, dyslipide-
mia, and hypertension. This puts these patients at increased risk for cardio-
vascular morbidity and mortality. Although there is strong evidence for the
importance of tight glycemic control in minimizing the microvascular com-
plications of diabetes, many of the current therapies used for optimizing gly-
cemic control also cause weight gain. With this treatment-induced weight
gain, there is a risk of worsening the patient’s insulin resistance. Physicians
need to be aware of this vicious cycle in their overweight type 2 diabetic pa-
tients. This article reviews the strategies that are currently available to
achieve glycemic control while at the same time minimizing weight gain
and the associated complications. These strategies include lifestyle modifica-
tions (diet and exercise); use of metformin; a-glucosidase inhibitors; and
a new class of diabetes medications including the glucagons-like peptide
(GLP-1) receptor agonists (exenatide) and the dipeptidyl-peptidase-IV
(DPP-IV) enzyme inhibitors (sitagliptin, vildagliptin).
Metabolic syndrome and type 2 diabetes

The metabolic syndrome is an insulin-resistant state that is defined by
a group of cardiovascular risk factors that together confer an increased
risk for both cardiovascular disease and type 2 diabetes. Multiple organiza-
tions including the World Health Organization, the National Cholesterol
Education Program Adult panel III, which was recently revised by the
E-mail address: robert.yanagisawa@mssm.edu (R.T. Yanagisawa).
1108 JOFFE & YANAGISAWA
American Heart Association and the National Heart, Lung, and Blood In-
stitute, and the International Diabetes Federation have developed various
criteria to meet their definition of the metabolic syndrome. All of the defini-
tions include obesity, elevated triglycerides, decreased high-density lipopro-
tein cholesterol, hypertension, and impaired fasting glucose or diabetes [1–3].
By definition, a high proportion of patients with type 2 diabetes also have
the metabolic syndrome. In the Verona Diabetes Complications Study, 946
type 2 diabetic patients were assessed for the presence of metabolic syn-
drome and cardiovascular morbidity and mortality over a mean follow-up
period of 4.5 years. A total of 92.3% of this population was found to
have metabolic syndrome according to the World Health Organization cri-
teria. At the start of the study, 31.7% of patients had cardiovascular disease
that was more prevalent among those subjects who also had the metabolic
syndrome (32.9% versus 17.8%; P ¼ .0005). Among subjects who had no
cardiovascular disease at the start of the study, cardiovascular disease events
(myocardial infarction, cerebrovascular accident, peripheral vascular dis-
ease) were significantly increased in patients with the metabolic syndrome
as compared with those patients without (19.9% versus 3.9%; P !.001).
Overall, the presence of metabolic syndrome was associated with an almost
fivefold increase in cardiovascular disease risk among diabetic patients [4].
In a similar study of type 2 diabetes in Finland and Sweden, among 1697
subjects with type 2 diabetes, the metabolic syndrome according to World
Health Organization criteria was observed in 78% of women and 84% of
men. The risk for coronary heart disease and stroke was increased threefold
in those subjects who met the definition of metabolic syndrome. Cardiovas-
cular mortality was also increased (12% versus 2.2%; P ! .001) in those
subjects with metabolic syndrome compared with those subjects without [5].
In a recently published cross-sectional survey of Italian diabetic patients
(N ¼ 8497), the presence of metabolic syndrome according to the American
Heart Association and National Heart Lung Blood Institute criteria was
69.5% among type 2 diabetics. The presence of metabolic syndrome was
present in a significantly larger proportion of type 2 diabetics according
to the International Diabetes Federation criteria (77.6%). This most likely
reflects a lower diagnostic threshold for abdominal obesity by the Interna-
tional Diabetes Federation. Either definition of metabolic syndrome was
an independent statistical indicator of the presence of both microvascular
complications (retinopathy, neuropathy, and nephropathy) and macrovas-
cular complications (cardiovascular disease) [6].
Both the UK Prospective Diabetes Study (UKPDS) and the Diabetes
Control and Complications Trial Research Group have shown that im-
proved glycemic control is associated with a decrease in the rates of diabetic
complications including retinopathy, nephropathy, and neuropathy. In
these trials, a hemoglobin (Hb) A1c of approximately 7% was associated
with fewer complications [7,8]. Based on these results, the American Diabe-
tes Association recommends a HbA1c goal of less than 7% in general, and as
METABOLIC SYNDROME AND TYPE 2 DIABETES 1109
close to normal in individual patients (!6%), avoiding hypoglycemia [9].

Both these trials showed that patients in the intensively treated groups
(HbA1c 7%) had higher rates of hypoglycemia and significantly more weight
gain. In the UKPDS study, patients in the intensively treated group had
a mean weight gain of 2.9 kg [8].
Health care providers need to strike a balance between obtaining tight
glycemic control to avoid both the microvascular and macrovascular com-
plications of diabetes, while also minimizing the weight gain, particularly
in the type 2 diabetic patients. Weight gain puts the patient at risk for wors-
ening of the metabolic syndrome (increased insulin resistance, worsening of
lipid profile, increased blood pressure), increasing the patient’s risk of car-
diovascular morbidity and mortality and the microvascular complications
of diabetes. Many of the medications used to treat diabetes including insu-
lin, sulfonylureas, meglitinides, and thiazolidinediones (TZDs) promote
weight gain. Recently, multiple approaches are available to the physician
to achieve glycemic control while at the same time minimizing weight gain
and the associated complications. These strategies include lifestyle modifica-
tions (diet and exercise); use of metformin; a-glucosidase inhibitors; and
a new class of diabetes medications including the GLP-1 receptor agonists
(exenatide) and the DPP-IV enzyme inhibitors (sitagliptin, vildagliptin).
Lifestyle modification to limit weight gain in diabetics

Lifestyle interventions including diet, exercise, and behavioral modifica-
tion are important components in the long-term management of type 2 di-
abetics. A total of 80% to 90% of persons with type 2 diabetes are
overweight [10]. Moderate weight loss (5%–10% of total body weight) is as-
sociated with a decrease in plasma glucose, a reduction in fasting hyperinsu-
linemia, reduced hepatic gluconeogenesis, and decreased insulin resistance.
This reduction in insulin resistance can mean a decrease in the need for
the diabetic medications (insulin, sulfonylureas, and so forth) that can con-
tribute to increased weight gain (see next section). In addition to the im-
provement in glycemic control, weight loss leads to marked improvements
in blood pressure and dyslipidemia that can lead to decreased cardiovascu-
lar morbidity and mortality [11,12]. In a recent Korean study of 58 patients
with type 2 diabetes, subjects were randomized to either a 16-week intensive
lifestyle modification program and subsequent monthly meetings for 6
months, or to basic dietary education and usual care. The intensive lifestyle
modification program included a dietary plan, physical activity of moderate
intensity for at least 150 minutes per week, and the use of food diaries. At
the end of 6 months, the intensive group lost 2 kg versus a gain of 0.2 kg
in the usual care group. There were significant decreases in the intensive
group versus the usual care group in HbA1c; systolic blood pressure; and ca-
rotid intima-media thickness (marker for future coronary heart disease or
stroke) [13].
Because many type 2 diabetics are on medications that cause weight gain
to maintain glycemic control, weight loss can be difficult. In a 16-week study
of a behavioral weight loss program, comparing nondiabetic obese women
with obese diabetic women (matched for age and weight), the initial weight
loss at 16 weeks was similar between the nondiabetic and diabetic women,
respectively (7.4 versus 6.4 kg). At 1 year of follow-up, however, the diabetic
women gained on average 5.4 kg versus 1 kg in the nondiabetic women [14].
Exercise has been shown to reduce blood glucose levels by increasing insulin
sensitivity, reducing central obesity, and improving blood pressure and
lipids. Although exercise alone does not often cause significant weight
loss, it plays an important role in the maintenance of weight loss [11]. In
a study of type 2 diabetic obese patients enrolled in a weight loss program,
those patients who continued to exercise the most at 1 year had the greatest
weight loss and improvement in glycemic control [15].
In 2004, Norris and colleagues [16] published a meta-analysis study look-
ing at the long-term effectiveness of lifestyle and behavioral weight loss in-
terventions in adults with type 2 diabetes. To be included in the analysis,
studies had to be randomized controlled trials with a minimum of 1-year fol-
low-up, examining weight loss or weight loss strategies using one or more
lifestyle interventions including diet, exercise, or behavioral changes.
Twenty-two studies were included with a total of 4659 participants. The
pooled weight loss for any type of intervention versus usual care among
585 subjects was 1.7 kg (3.1% of baseline bodyweight among 511 subjects).
Among 126 patients who underwent a behavioral intervention and increased
physical activity, those subjects who received a very low calorie diet lost 3 kg
more than those on a low-calorie diet. Among 53 patients who received
identical behavioral and dietary interventions, those who participated in
more intensive exercise programs lost 3.9 kg more than those who partici-
pated in less intense activity or no activity at all. HbA1c levels generally
corresponded to changes in weight. Overall, this analysis showed that
low-calorie and very low calorie diets and behavioral modification combined
with exercise can lead to a clinically meaningful weight loss in overweight
type 2 diabetics, even those who are expected to gain weight.
Weight gain with intensive diabetes control

The UKPDS was a prospective observational study looking at 3867
newly diagnosed patients with type 2 diabetes who were randomized to in-
tensive therapy with a sulfonylurea (chlorpropamide, glibenclamide, or gli-
pizide) or with insulin versus conventional therapy with diet modification.
Over 10 years, the average HbA1c was 7% in the intensive group versus
7.9% in the conventional group. Compared with the conventional group,
the risk in the intensive group was 12% lower for any diabetes-related
end point (sudden death, death from hypoglycemia or hyperglycemia, fatal
or nonfatal myocardial infarction, angina, heart failure, stroke, renal failure,

amputation, vitreous hemorrhage, retinopathy requiring photocoagulation,
blindness in one eye, or cataract extraction); 10% lower for any diabetes-re-
lated death (death from myocardial infarction, stroke, peripheral vascular
disease, renal disease, hyperglycemia or hypoglycemia, and sudden death);
and 6% lower for all-cause mortality. Most of the risk reduction seen in
any of the diabetes-related end points was caused by a 25% risk reduction
in microvascular complications. Weight gain was significantly higher in the
intensive group (mean, 2.9 kg) than in the conventional group. Patients who
were given insulin had a greater gain in weight (4 kg) versus those patients
who were assigned to sulfonylurea, chlorpropamide (2.6 kg), or glibencla-
mide (1.7 kg) [8].
Current medications for the management of type 2 diabetes: effects on

glycemic control and weight
In a recent observational study of 9456 type 2 diabetic patients who were
members of the Kaiser Permanente North-West health system, the 1-year
weight gain associated with the initiation of various diabetic regimens was
quantified. Those patients on insulin had a 1-year weight gain of 3.3 kg
[17]. There are multiple mechanisms for this weight increase in patients
treated with insulin. Insulin is an anabolic hormone that promotes lipogen-
esis, stimulation of triglyceride synthesis, and adipocyte differentiation. It
also promotes muscle synthesis [18]. Other factors that lead to weight gain
include the increased caloric intake by patients to prevent or treat episodes
of hypoglycemia. Weight gain is also associated with improvement in blood
glucose control with a subsequent decrease in glycosuria, hence a decrease in
energy expenditure. If this decrease in energy expenditure is not accompa-
nied by a decrease in caloric intake, weight gain occurs. Insulin also causes
sodium and water retention by a direct effect on the distal tubule in the
kidney [19–21].
In the previously mentioned Kaiser Permanente study, those patients on
a sulfonylurea as monotherapy gained on average 1.8 kg over 1 year, similar
to the weight gain seen in the UKPDS study [8,17]. Sulfonylureas work by
binding to specific sulfonylurea receptors on the pancreatic beta cell, which
causes insulin release. As monotherapy, they can be expected to cause an
HbA1c reduction of approximately 1% to 2%. Because they work by in-
creasing insulin secretion, the mechanism of weight gain seen with this class
of drugs is similar to that seen with insulin [19].
Repaglinide and nateglinide stimulate beta cell insulin secretion by inhi-
bition of the ATP-dependent potassium channels on the pancreatic beta cell
[22]. They work rapidly and with a much shorter duration than traditional
sulfonylureas. They are given right before eating and mimic the first-phase
insulin secretion response that many type 2 diabetic patients do not
demonstrate [19]. In a 1-year prospective randomized study of type 2

diabetic patients who were given either glyburide or repaglinide as
monotherapy, repaglinide was shown to decrease HbA1c by 1.3% in treat-
ment-naive patients. An increase in weight gain of 2.4 kg was observed in
the repaglinide group versus 3.64 kg in the glyburide group [23]. In another
16-week clinical trial in type 2 diabetic patients, patients were randomly as-
signed to receive repaglinide versus nateglinide. The repaglinide group had
a decrease in HbA1c of 1.57% versus 1.04% in the nateglinide group. The
mean weight gain at the end of the study was 1.8 kg in the repaglinide group
compared with 0.7 kg in the nateglinide group [22].
Rosiglitazone and pioglitazone are the two TZDs currently used in the
United States as monotherapy or combination therapy with metformin
and sulfonylureas to treat type 2 diabetes. These medications are insulin sen-
sitizers that primarily work by binding to the peroxisome proliferator-acti-
vated receptor gamma enhancing insulin-mediated glucose uptake into liver,
adipose tissue, and skeletal muscle [19,24,25]. This increase in insulin sensi-
tivity occurs despite significant weight gain that is seen when either rosigli-
tazone or pioglitazone are used as monotherapy. In a 26-week trial of 493
patients with type 2 diabetes, subjects were assigned to receive either rosigli-
tazone (4 or 8 mg) or placebo. Rosiglitazone, 4 and 8 mg, decreased HbA1c
relative to placebo by 1.2% and 1.5%, respectively. There was a dose-depen-
dent increase in weight at the end of the study. The 4-mg group gained
a mean of 1.6 kg, whereas the 8-mg group gained a mean of 3.5 kg. Despite
this increase in weight gain, there was also a significant increase in insulin
sensitivity as measured by the homeostasis model of assessment. Insulin re-
sistance was decreased by 16% in the 4-mg group and 24.6% in the 8-mg
group. In addition, an increase in both low-density and high-density lipo-
protein cholesterol was observed [26]. In another 26-week randomized, dou-
ble-blind, placebo-controlled study evaluating pioglitazone, 408 type 2
diabetics were randomized to receive either placebo or pioglitazone (7.5,
15, 30, or 45 mg). Patients treated with 15 or 30 or 45 mg had significant
decreases in HbA1c that were dose dependent compared with placebo rang-
ing from 1% to 1.6%. As with rosiglitazone, there was a dose-dependent
increase in weight gain compared with baseline ranging from 1.3 to 2.8 kg.
Pioglitazone-treated patients showed significant decreases in both triglycer-
ides and increases in high-density lipoprotein cholesterol [27]. The weight
gain seen with the TZDs is associated with a redistribution of body fat
that is beneficial in terms of promoting insulin sensitivity and preventing
the body habitus typical of the metabolic syndrome. TZDs decrease visceral
fat, which is thought to be responsible for insulin resistance, by releasing
free fatty acids that blunt insulin sensitivity and increase proinflammatory
markers, such as interleukin-6 and plasminogen activator inhibitor-1. Al-
though visceral fat is decreased, subcutaneous fat is increased, which is
thought to be less of a cardiac risk [24,26]. Some of the weight gain seen
with TZDs is from fluid retention, which may be caused by increased
concentrations of vascular permeability factors or endothelial-mediated va-

sodilation. This fluid retention leads to expanded plasma volume and depen-
dent edema [19,24,25].
The a-glucosidase inhibitors (acarbose, miglitol) work by competitively
inhibiting the a-glucosidase enzymes at the brush border of the small intes-
tine. This prevents the enzymes from breaking down oligosaccharides and
disaccharides into monosaccharides before absorption. This results in a de-
lay of the absorption of glucose further down the intestinal tract. This leads
to alterations in the intestinal hormone secretion that overall reduces post-
prandial insulin concentrations through the decreased rise in postprandial
glucose levels. The main adverse effects of this drug are gastrointestinal re-
lated to the fermentation of the nondigested carbohydrates in the colon lead-
ing to flatulence, diarrhea, and abdominal discomfort [19,28]. a-Glucosidase
inhibitors seem to have a weight neutral effect, or in some studies they have
been shown to cause a small amount of weight loss in the range of 0.5 kg
over a period of 1 to 3 years. HbA1c was decreased by 0.4% to 0.8% [29–32].
Metformin is recommended for use in diabetics, particularly obese pa-
tients, because of its favorable effects on weight and its properties as an in-
sulin sensitizer by increasing the peripheral uptake of insulin and decreasing
hepatic gluconeogenesis. Metformin decreases HbA1c by 1% to 1.5%. In
a study of 753 overweight type 2 diabetics in the UKPDS Study, those pa-
tients who received metformin compared with those who received a sulfonyl-
urea or insulin had lower all-cause mortality, and lower incidences of
myocardial infarction or stroke. The patients receiving metformin had no
significant increase in weight, whereas those receiving sulfonylureas or insu-
lin did have significant weight gain [33]. Although some studies using met-
formin actually show a decrease in weight when using metformin, others
show that metformin is weight neutral and is helpful in avoiding excess
weight gain when used alone or in combination with other antihyperglyce-
mics [17,34–38]. The weight loss seen in these studies ranges from 0.6 to
2.7 kg [17,34,37]. The reduction in weight gain seen with metformin is
most likely caused by multiple mechanisms including a decrease in insulin
levels; gastrointestinal side effects, such as anorexia and abdominal discom-
fort; and avoidance of hypoglycemia and decreased hunger (Table 1) [19].
Use of combination therapies to improve glycemic control

and limit weight gain
Although the evidence shows that most of the medications used to treat
diabetes cause weight gain, with the exception of metformin and a-glucosi-
dase inhibitors, various combinations of therapies are used to limit this
weight gain and improve glycemic control at the same time. Although insu-
lin clearly has benefits in the early treatment of type 2 diabetes, by decreas-
ing glucotoxicity and preserving beta cell function, weight gain is a negative
Table 1
Summary of diabetes mellitus medications and their effects on weight and HbA1C
Class of diabetes Effect on Average change in kg % change
mellitus medication weight (specified time period) in A1C
Insulin [17,57] Weight gain 3.3 (1 y) 1–2
1.8 (26 wk)
Sulfonylureas [8,17,22] Weight gain 1.8–3.64 (1 y) 1–2
Repaglinide [21,22] Weight gain 2.4 (1 y) 1.3–1.57
1.8 (16 wk)
Nateglinide [21] Weight gain 0.7 (16 wk) 1.04
Thiazolidinediones Weight gain 1.3–3.5 (26 wk) 1–1.6
[17,25,26]
5 (1 y)
a-Glucosidase inhibitors Weight loss/weight 0.5 loss (1–3 y) 0.4–0.8
[28–31] neutral
Metformin [17,32–37] Weight loss/weight 0.6–2.7 loss 1–1.5
neutral
Exenatide [53–57] Weight loss 0.9–2.3 loss (26 wk) 0.40–1.3
5.3 loss (82 wk)
Dipeptidyl-peptidase-IV Weight neutral 0.1–0.7 loss (24 wk) 0.57–1.52
inhibitors [58–60]
side effect. Many trials have shown the benefit of combining insulin with
metformin, decreasing the dose of insulin needed and at the same time min-
imizing the weight increase [35,36,39–42]. In one randomized, double-blind,
placebo-controlled study, 43 patients with poorly controlled type 2 diabetes
on insulin were randomized to receive metformin versus placebo in combi-
nation with insulin for 6 months. HbA1c decreased by 2.5% in the metfor-
min group versus 1.6% in the placebo group. For patients who received
placebo, insulin doses increased 29% more than in the metformin group.
In the metformin group, insulin doses decreased slightly. Patients in the pla-
cebo group gained on average 3.2 kg versus 0.5 kg in the metformin group.
Metformin in combination with sulfonylurea has also been shown to im-
prove glycemic control better than either agent alone, and the combination
limits the weight gain seen with sulfonylureas. In a double-blind, placebo-
controlled, randomized study of 144 type 2 diabetics poorly controlled
with diet alone, patients were randomly assigned to receive glyburide alone,
metformin alone, or the combination. The combination of metformin and
glyburide led to less treatment failure than either medication alone (36%
versus 25%). There was no significant weight gain in the metformin or com-
bination groups, whereas there was on average a 2.8 kg weight increase in
the glyburide group [38]. Metformin and a TZD work synergistically to in-
crease insulin sensitivity and cause less of a weight gain than seen with
a TZD as monotherapy. In a study of 348 inadequately controlled type 2 di-
abetics on maximal metformin therapy, patients were randomized to receive
placebo versus rosiglitazone, 4 mg, versus rosiglitazone, 8 mg, in addition to
metformin. The patients receiving rosiglitazone had a dose-dependent
decrease in HbA1c of 1% to 1.2%. There was also a dose-dependent increase

in insulin sensitivity in the combination groups. There was an average
weight gain of 1.9 kg in the rosiglitazone-metformin groups that is less
than the typical weight gain seen in patients on rosiglitazone alone [43].
In another multicenter randomized, double-blind, placebo-controlled study,
468 type 2 diabetic patients were randomized in a 32-week study to receive
fixed-dose rosiglitazone-metformin (2 mg/500 mg titrated to 8 mg/2000 mg)
versus rosiglitazone or metformin alone. The combination group had signif-
icant decreases in HbA1c (2.3%) versus metformin (1.8%) versus rosigli-
tazone (1.6%). No significant weight gain was seen in the combination or
metformin groups [44]. Although the combination of a TZD and insulin re-
sults in the lowering of daily insulin dosing by increasing insulin sensitivity,
it leads to significant weight gain ranging from 3.6 to 5.3 kg [45,46].
The incretins: a new class of drugs to limit the weight gain in diabetics
A new class of medications has been developed and recently approved by
the Food and Drug Administration for the management of type 2 diabetes.
The currently approved drugs that are being used clinically include exena-
tide (GLP-1 agonist) and sitagliptin phosphate (DPP-IV inhibitor). Al-
though traditional medications used in the treatment of type 2 diabetes
work by either stimulating insulin secretion, suppressing hepatic glucose re-
duction, or increasing insulin sensitivity, none of these treatments can fully
prevent the eventual beta cell failure that is typically seen as diabetes prog-
resses. Although insulin has become the treatment of choice as this beta cell
failure occurs, the negative side effects, such as hypoglycemia and weight
gain, are evident. The newer medications, exenatide and sitagliptin, were de-
veloped to enhance the actions of GLP-1, a gut-derived incretin hormone
that is lacking in type 2 diabetics [47,48].
The ‘‘incretin effect’’ was first observed and published in the 1960s. It was
demonstrated that when subjects were given an equivalent oral glucose load
to an intravenous glucose infusion, the oral glucose caused a significantly
higher insulin response [49]. This led to the discovery of the incretins, which
are defined as gut-derived hormones that enhance insulin secretion in a glu-
cose-dependent manner. The first such hormone that was discovered was
glucose-dependent insulinotropic peptide (GIP), which is a protein hormone
that is both produced and secreted by the K cells in the duodenum and je-
junum. The other hormone that was later discovered was GLP-1, which is
produced and secreted by the L cell found in the distal ileum and colon.
GIP and GLP-1 levels increase as the gut is stimulated on ingestion of
food. Although both GIP and GLP-1 levels rise when a meal is ingested,
they have very short half-lives (less than 2 minutes) because of degradation
by DPP-IV [47,48].
GLP-1 contributes to glucose homeostasis in a number of ways. It stim-
ulates glucose-dependent insulin secretion, contributing to the regulation of
postprandial glucose levels. It also suppresses glucagon in a glucose-depen-

dent manner. In addition, GLP-1 has direct effects on the gut, causing a de-
lay in gastric emptying. GLP-1 also acts centrally at the level of the
hypothalamus causing increased satiety, decreased appetite, and decreased
food and water intake. Animal models and cell lines also show that GLP-
1 stimulates insulin gene transcription and increases beta cell mass
(Fig. 1) [47,48].
Studies have shown that type 2 diabetics have reduced levels of GLP-1. In
a study by Vilsboll and coworkers [50] in 2001, 12 type 2 diabetic patients
and 12 healthy subjects matched by age and body mass index were given
a test meal and plasma levels of GIP and GLP-1 were monitored. GIP levels
were minimally reduced in the type 2 diabetic patients compared with
healthy subjects, but GLP-1 levels were significantly reduced at 75, 90,
and 120 minutes after ingestion in the diabetics. The type 2 diabetic patients
Fig. 1. Physiology of GLP-1 secretion and action on GLP-1 receptors in different organs and
tissues. (From Drucker DJ, Nauck MA. The incretin system: glucagons-like peptide-1 receptor
agonists and dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet 2006;368:1696–705;
with permission.)
also had an abnormal physiologic insulin response with abnormally low

levels of C peptide and insulin at this time.
Exenatide is a GLP-1 agonist that is synthetically derived from the natu-
rally occurring GLP-1 agonist, exendin-4. Exendin-4 was discovered in the
salivary gland venom of the Gila monster, a lizard that is found in the south-
west United States. Exenatide is defined as an incretin mimetic because it
mimics the biologic actions of GLP-1. Exenatide is not degraded by DPP-
IV, so it has a longer half-life (60–90 minutes) with increased levels lasting
4 to 6 hours after subcutaneous injection. Exenatide improves both fasting
and postprandial glucose control by a number of mechanisms including glu-
cose-dependent insulin secretion, restoration of the first- and second-phase
insulin response, suppression of glucagon in a glucose-dependent manner,
delayed gastric emptying, and decreased food intake [47,48,51,52]. Delayed
gastric emptying slows the passage of food from the stomach to the intes-
tine, delaying the entry of glucose into the circulation. This slowing of gas-
tric emptying also leads to the sensation of fullness that signals satiety,
which in turn leads to reduced food intake. Exenatide causes weight loss
and increased insulin sensitivity [51]. In addition, exenatide has been shown
in animal models to promote beta cell proliferation and improve function by
increasing the transcription of genes involved in insulin secretion [47].
Exenatide has been studied as an add-on therapy for diabetic patients
with inadequate glycemic control on a sulfonylurea, metformin, or both
[53]. A 30-week triple-blind placebo-controlled study was completed by
272 patients on metformin with an average HbA1c of 8.2 and body mass in-
dex of 34.2. They were randomized to receive either placebo or 5 mg subcu-
taneously twice daily for 4 weeks, followed by an additional 26 weeks of
placebo versus 5 or 10 mg twice daily of exenatide. At the end of the trial,
HbA1c was decreased by 0.78% in the 10-mg group and 0.40% in the 5-mg
group. The placebo group had an increase in HbA1c of 0.08%. An HbA1c
less than 7% was achieved in 46% of the 10-mg group, 32% in the 5-mg
group, and 13% in the placebo group. The exenatide-treated patients had
a significant dose-dependent progressive weight loss of 2.8 kg in the 10-mg
group and 1.6 kg in the 5-mg group versus the placebo group. Adverse effects
included nausea, diarrhea, vomiting, and hypoglycemia [54]. A similar trial
with the same design was performed in 377 patients on sulfonylureas with
average HbA1c of 8.6% and body mass index of 33. HbA1c changes from
baseline were as follows: 0.86% in the 10-mg group, 0.46% in the 5-mg
group, and þ0.12% in the placebo group. An HbA1c less than 7% was
achieved in 41% of the 10-mg group, 33% in the 5-mg group, and 9% in
the placebo group. The exenatide-treated patients had a significant dose-de-
pendent progressive weight loss of 1.6 kg in the 10-mg group and a nonsignif-
icant loss of 0.9 kg in the 5-mg arm. The weight loss in the sulfonylurea
patients was not as great as those treated with metformin. Dose-dependent
nausea was a common adverse effect that was seen with initiation of ther-
apy, but the incidence decreased over time. There was a higher incidence
of mild to moderate hypoglycemia, which suggested that this may have been
caused by the sulfonylurea [55]. A study of similar design was performed in
733 subjects inadequately controlled on sulfonylurea and metformin. The
patients randomized to exenatide, 5 or 10 mg twice daily, were continued
on maximal metformin doses and were randomized to receive either maxi-
mally effective doses of sulfonylurea or minimal doses. HbA1c changes
from baseline were as follows: 0.8% in the 10-mg group, 0.6% in the
5-mg group, and þ0.2% in the placebo group. Of the 550 patients complet-
ing the trial at 30 weeks, an HbA1c less than 7% was achieved in 34% of the
10-mg group, 27% in the 5-mg group, and 9% in the placebo group. The ex-
enatide-treated patients had a significant progressive weight loss of 1.6 kg in
the 10-mg group, 1.6 mg in the 5-mg group, and a nonsignificant loss of 0.9 kg
in the placebo group. Mild or moderate nausea was again the most common
adverse event. The incidence of mild to moderate hypoglycemia was 28% in
the 10-mg group, 19% in the 5-mg group, and 13% in the placebo group. The
incidence of hypoglycemia was lower in the minimally treated versus maxi-
mally treated patients. These data suggest that the sulfonylurea dose should
be reduced when initiating exenatide treatment to minimize hypoglycemia
[56]. In an uncontrolled open-label extension of the exenatide-metformin
trial, HbA1c was decreased by 1.3% at 82 weeks in the exenatide-treated pa-
tients. A total of 59% of patients reached an HbA1c less than 7% at 82
weeks. There was a progressive reduction in weight at 82 weeks of 5.3 kg.
Overall, exenatide was well tolerated and showed a progressive reduction
in weight and a durable reduction in HbA1c over 82 weeks [57]. Another
clinical trial randomized 551 patients with type 2 diabetes suboptimally con-
trolled on sulfonylurea and metformin to receive either exenatide, 10 mg sub-
cutaneously twice a day, or glargine insulin titrated to maintain a fasting
glucose level less than 100 mg/dL. At week 26, both therapies reduced
HbA1c by 1.11%. As expected, exenatide reduced postprandial glucose ex-
cursions more than glargine, but glargine reduced fasting glucose levels
more than exenatide. There was a significant difference in weight between
the two groups at week 26: 2.3 kg in the exenatide group and þ1.8 kg
in the glargine group. Nausea, vomiting, and diarrhea were more common
in the exenatide-treated patients. Although exenatide and glargine had sim-
ilar effects on glycemic control, exenatide had the added advantage of pro-
moting weight loss [58].
Exenatide is available in prefilled pens of 5 or 10 mg. The recommended
starting dose is 5 mg, to be given subcutaneously before breakfast and be-
fore dinner. The dose can be increased to 10 mg after 1 month as tolerated.
If the patient is on a sulfonylurea, the dose should be reduced before initi-
ation. It is not recommended for patients with a creatinine clearance less
than 30 mL/min [53].
Sitagliptin is currently the only Food and Drug Administration approved
oral DPP-IV inhibitor available for clinical use. The recommended daily
dose is 100 mg by mouth. Sitagliptin is a potent, competitive, reversible
inhibitor of the DPP-IV enzyme that degrades both GIP and GLP-1, aug-
menting their levels. A single dose has been shown to inhibit plasma
DPP-IV levels by over 80% over a 24-hour period. Sitagliptin has similar
mechanisms of action to exenatide including the glucose-dependent stimula-
tion of insulin secretion, suppression of glucagon postprandially, beta cell
proliferation, and the prevention of beta cell apoptosis. In contrast to exe-
natide, sitagliptin does not delay gastric emptying or cause the weight loss
seen with exenatide [48].
Sitagliptin has been studied as monotherapy and in combination with
metformin or pioglitazone. In a randomized, double-blind, placebo-con-
trolled study, 741 patients with type 2 diabetes and average HbA1c of 8%
were randomized to receive 100 or 200 mg of sitagliptin or placebo over
a 24-week period. Those patients in the 100- or 200-mg arms had significant
decreases in HbA1c compared with placebo: 0.79% and 0.94%, respec-
tively. Reductions in HbA1c were greatest among those patients with base-
line HbA1c greater than 9% versus those with baseline HbA1c less than 8%.
HbA1c reduction ranged from 0.57% to 1.52% in the 100-mg group and
from 0.65% to 1.50% in the 200-mg group. Insulin resistance was de-
creased in patients on sitagliptin. No meaningful body weight changes
from baseline were observed in the 100-mg group (0.2 kg) or 200-mg group
(0.1 kg) [59].
In another double-blind placebo-controlled study, 701 type 2 diabetics
on metformin (O1500 mg) with average HbA1c of 8% were randomized
to receive placebo or 100 mg of sitagliptin in addition to metformin. Pa-
tients who did not meet certain goals for glycemic control were also started
on pioglitazone as ‘‘rescue therapy.’’ At week 24, sitagliptin led to a signif-
icant reduction in HbA1c, 0.65% compared with placebo. There were sig-
nificant improvements in fasting glucose, postprandial glucose, fasting
insulin and fasting C peptide, postmeal insulin and C peptide levels, and in-
sulin resistance. A larger proportion of patients in the sitagliptin group
achieved an HbA1c of less than 7% with sitagliptin (47%) versus 18.3%
in the placebo group. There were small but significant decreases in body
weight from baseline in both groups (0.6 and 0.7 kg), but between
groups there was no significant difference. Sitagliptin was well tolerated
with no significant increase in gastrointestinal side effects compared with
metformin treatment alone, and there was no significant increase in hypo-
glycemia [60]. In another double- blind, placebo-controlled study, patients
who were on pioglitazone, 30 mg or 45 mg, as monotherapy were random-
ized to receive sitagliptin, 100 mg, daily or placebo. Those patients who
received sitagliptin had significant reductions in HbA1c compared with pla-
cebo. The proportion of patients with HbA1c less than 7% at the end of the
study was 45.4% in the sitagliptin group versus 23% in the placebo group.
There was no significant difference in weight gain between the two groups:
þ1.8 kg in the sitagliptin group versus þ1.5 kg in the placebo group. There
was no significant difference between the two groups in terms of
hypoglycemia or gastrointestinal symptoms including nausea, vomiting, or

diarrhea [61].
Exenatide is recommended for type 2 diabetic patients who have failed on
one or two oral hypoglycemic agents. Sitagliptin can be used as monother-
apy or as add-on therapy to metformin or a TZD. There does not seem to be
significant differences in the HbA1c lowering of these two classes of drugs.
The significant difference between sitagliptin and exenatide is in the weight
loss that is observed. Patients on exenatide lose significantly more weight,
whereas sitagliptin seems to have a weight neutral effect (see Table 1).
This difference is probably caused by the delayed gastric emptying and in-
creased nausea seen with exenatide use and not sitagliptin use. This differ-
ence is caused by the higher plasma levels of GLP-1 and GIP seen after
exenatide use versus a more modest increase in levels after a dose of sitaglip-
tin [48]. A long-acting exenatide preparation that would be given once
a week is awaiting Food and Drug Administration approval. In addition,
vildagliptin, a DPP-IV inhibitor, is also pending approval.
Summary
Obesity is a major component of the metabolic syndrome, and is also
present in 80% to 90% of type 2 diabetic patients. The increased insulin re-
sistance, dyslipidemia, and hypertension that so often are seen in these pa-
tients lead to increased cardiovascular morbidity and mortality. Treatment
of the metabolic syndrome and type 2 diabetes must include lifestyle modi-
fication, glycemic control, and modification of risk factors. A moderate
weight loss of 5% to 10% can lead to improved glycemic control, increased
insulin sensitivity, improvement in lipids, and a lowering of blood pressure.
It has been well established that tight glycemic control with HbA1c less than
7% leads to significant decreases in microvascular complications of diabetes
and also possibly a decrease in macrovascular complications. Tight glycemic
control often leads to weight gain, however, making even modest weight loss
in type 2 diabetics very difficult. Patients often require insulin, secreta-
gogues, or TZDs, all of which can contribute to increased weight gain in
type 2 diabetics. Medications, such as metformin and acarbose, have been
used as monotherapy or in combination with other medications to limit
weight gain. For the first time in years, there is now available a new class
of medications, the incretins (exenatide and sitagliptin) for the treatment
of diabetes. These medications do not cause the weight gain seen in older
medications, and may actually cause significant weight loss. As seen in an-
imal and cell models, these medications may also have the added benefit of
causing beta cell neogenesis and proliferation, slowing the progression of
beta cell failure that occurs in the type 2 diabetic. The physician needs to
strike a balance in their obese diabetics by striving for optimal glycemic con-
trol and at the same time minimizing weight gain.
References
complications. Part 1: diagnosis and classification of diabetes mellitus: provisional report of
a WHO consultation. Diabet Med 1998;15:539–53.
[2] National Cholesterol Education Program (NCEP). Expert panel on detection, evaluation,
and treatment of high blood cholesterol in adults (adult treatment panel III): executive sum-
mary of the third report of the National Cholesterol Education Program (NCEP) Expert
Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult
Treatment Panel III). JAMA 2001;285:2486–97.
[3] Alberti KG, Zimmet P, Shaw J, IDF Epidemiology Task Force Consensus Group. The met-
abolic syndrome: a new worldwide definition. Lancet 2005;366:1059–62.
[4] Bonora E, Targher G, Formentini F, et al. The metabolic syndrome is an independent pre-
dictor of cardiovascular disease in type 2 diabetic subjects. Prospective data from the Verona
Diabetes Complications Study. Diabet Med 2003;21:52–8.
with the metabolic syndrome. Diabetes Care 2001;24:683–9.
[6] The Metascreen Writing Committee. The metabolic syndrome is a risk indicator of micro-
vascular and macrovascular complications in diabetes. Diabetes Care 2006;29:2701–7.
[7] The Diabetes Control and Complications Trial Research Group. The effect of intensive
treatment of diabetes on the development of and progression of long term complications
in insulin dependent diabetes mellitus. N Engl J Med 1993;329:977–86.
[8] UK Prospective Diabetes Study (UKPDS) Group. Intensive blood glucose control with
sulfonylureas or insulin compared with conventional treatment and risk of complications
in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–53.
[9] American Diabetes Association. Standards of medical care in diabetes-2006. Diabetes Care
2006;29(Suppl 1):s4–42.
[10] Gerstein HC, Haynes RB. Evidence-based diabetes care. Ontario (Canada): BC Decker,
Inc.; 2000. p. 252–76.
[11] Williams KV, Kelley DE. Metabolic consequences of weight loss on glucose metabolism and
insulin action in type 2 diabetes. Diabetes Obes Metab 2000;2:121–9.
[12] Klein S, Sheard NF, Pi-Sunyer X, et al. Weight management through lifestyle modification
for the prevention and management of type 2 diabetes: rationale and strategies. Diabetes
Care 2004;27(8):2067–73.
[13] Kim SH, Lee SJ, Kang ES, et al. Effects of lifestyle modification on metabolic parameters
and carotid intima-media thickness in patients with type 2 diabetes mellitus. Metabolism
2006;55:1053–9.
[14] Guare JC, Wing RR, Grant A. Comparison of obese NIDDM and non-diabetic women:
short and long term weight loss. Obes Res 1995;3:329–35.
[15] Wing RR, Epstein LH, Paternostro-Bayles M, et al. Exercise in a behavioral weight control
program for obese patients with type 2 diabetes. Diabetologia 1988;31:902–9.
[16] Norris SL, Zhang X, Avenell A, et al. Long term effectiveness of lifestyle and behavioral
weight loss interventions in adults with type 2 diabetes: a meta-analysis. Am J Med 2004;
117:762–74.
[17] Nichols GA, Gomez-Caminero A. Weight changes following the initiation of new anti-
hyperglycemic therapies. Diabetes Obes Metab 2007;9(1):96–102.
[18] Flier JS. Obesity and insulin resistance. J Clin Invest 2000;106(4):473–81.
[19] Purnell JQ, Weyer C. Weight effect of current and experimental drugs for diabetes mellitus.
Treat Endocrinol 2003;2(1):33–47.
[20] Carver C. Insulin treatment and the problem of weight gain in type 2 diabetes. Diabetes Educ
2006;32(6):910–7.
[21] Eldor R, Stern E, Milicevic Z, et al. Early use of insulin in type 2 diabetes. Diabetes Res Clin
Pract 2005;68S1:S30–5.
[22] Rosenstock J, Hassman DR, Madder RD, et al. Repaglinide versus nateglinide monother-
apy. Diabetes Care 2004;27:1265–70.
[23] Marbury T, Huang WC, Strange P, et al. Repaglinide vs. glyburide: a one year comparison
trial. Diabetes Res Clin Pract 1999;43(3):155–66.
[24] Vasudevan AR, Balasubramanyam A. Thiazolidinediones: a review of their mechanisms of
insulin sensitization, therapeutic potential, clinical efficacy and tolerability. Diabetes Tech-
nol Ther 2004;6(6):850–63.
[25] Yki-Jarvinen H. Thiazolidinediones. N Engl J Med 2004;351:1106–18.
[26] Lebovitz HE, Dole JF, Patwardhan R, et al. Rosiglitazone monotherapy is effective in
patients with type 2 diabetes. J Clin Endocrinol Metab 2001;86(1):280–8.
[27] Aronoff S, Rosenblatt S, Braithwaite S, et al. Pioglitazone hydrochloride monotherapy im-
proves glycemic control in the treatment of patients with type 2 diabetes: a 6 month random-
ized placebo-controlled dose-response study. The Pioglitazone 001 Study Group. Diabetes
Care 2000;23(11):1605–11.
[28] Krentz AJ, Bailey CJ. Oral antidiabetic agents, current role in type 2 diabetes mellitus. Drugs
2005;65(3):385–411.
[29] Hanefield M, Fischer S, Schulze J, et al. Therapeutic potentials of acarbose as first line drug
in NIDDM insufficiency treated with diet alone. Diabetes Care 1991;14(8):732–7.
[30] Hotta N, Kakuta H, Sano T, et al. Long-term effect of acarbose on glycemic control in non-
insulin dependent diabetes mellitus: a placebo-controlled double blind study. Diabet Med
1993;10(2):134–8.
[31] Wolever TM, Chiasson JL, Josse RG, et al. Small weight loss on long-term acarbose therapy
with no change in dietary pattern or nutrient intake of individuals with NIDDM. Int J Obes
Relat Metab Disord 1997;21(9):756–63.
[32] Holman RR, Cull CA, Turner RC. A randomized double-blind trial of acarbose in type 2
diabetes shows improved glycemic control over 3 years (UK prospective diabetes study
44). Diabetes Care 1999;22(6):960–4.
[33] 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.
[34] DeFronzo RA, Goodman AM. Efficacy of metformin in patients with NIDDM. NEJM
1995;333(9):541–9.
[35] Relimpio F, Pumar A, Losada F, et al. Adding metformin versus insulin dose increase in in-
sulin-treated but poorly controlled type 2 diabetes mellitus: an open label randomized trial.
Diabet Med 1998;15(12):997–1002.
[36] Aviles-Santa L, Sindig J, Raskin P. Effects of metformin in patients with poorly controlled,
insulin-treated type 2 diabetes mellitus: a randomized, double-blind placebo controlled trial.
Ann Intern Med 1999;131(3):182–8.
[37] Stumvoli M, Nurjhan N, Perriello G, et al. Metabolic effects of metformin in non-insulin
dependent diabetes mellitus. NEJM 1995;333:550–4.
[38] Hermann LS, Schersten B, Bitzen PO, et al. Therapeutic comparison of metformin and sul-
fonylurea, alone and in various combinations: a double blind controlled study. Diabetes
Care 1994;17(10):1100–9.
[39] Chaudhuri A, Tomar R, Mohanty P, et al. The combination of insulin and metformin in
treatment of non-insulin dependent diabetes mellitus. Endocr Pract 1998;4(5):259–67.
[40] Ponssen H, Elte JWF, Lehert P, et al. Combined metformin and insulin therapy for patients
with type 2 diabetes mellitus. Clin Ther 2000;22(6):709–18.
[41] Wulffele MG, Kooy A, Lehert P, et al. Combination of insulin and metformin in the treat-
ment of type 2 diabetes. Diabetes Care 2002;25:2133–40.
[42] Johnson M, Krosnick A, Carson P, et al. A retrospective chart review of uncontrolled use of
metformin as an add-on therapy in type 2 diabetes. Clin Ther 20(4):691–8.
[43] Fonseca V, Rosenstock J, Patwardhan R, et al. Effect of metformin and rosiglitazone com-
bination therapy in patients with type 2 diabetes mellitus: a randomized controlled trial.
JAMA 2000;283(13):1695–702.
[44] Rosenstock J, Rood J, Cobitz A, et al. Initial treatment with rosiglitazone/metformin fixed-
dose combination therapy compared with either rosiglitazone or metformin in patients with
uncontrolled type 2 diabetes. Diabetes Obes Metab 2006;8(6):650–60.
[45] Actos (pioglitazone hydrochloride tablets). Osaka: Takeda Chemical Industries LTD.; 2002.
[data on file].
[46] Raskin P, Rendell M, Riddle MC, et al. A randomized trial of rosiglitazone therapy in pa-
tients with inadequately controlled insulin-treated type 2 diabetes. Diabetes Care 2001;
24(7):1226–32.
[47] Kendall DM, Kim D, Maggs D. Incretin mimetics and dipeptidyl peptidase-IV inhibitors:
a review of emerging therapies for type 2 diabetes. Diabetes Technol Ther 2006;8(3):385–96.
[48] Drucker DJ, Nauck MA. The incretin system: glucagon-like peptide-1 receptor agonists and
dipeptidyl peptidase-4 inhibitors in type 2 diabetes. Lancet 2006;368:1696–705.
[49] Perley M, Kipnis DM. Plasma insulin responses to oral and intravenous glucose: studies in
normal and diabetic patients. J Clin Invest 1967;46:1954–62.
[50] Vilsboll T, Krarup T, Deacon C, et al. Reduced post-prandial concentrations of intact bio-
logically active glucagons-like peptide-1 in type 2 diabetic patients. Diabetes 2001;50:609–13.
[51] Odegard PS, Setter SM, Iltz J. Update in the pharmacologic treatment of diabetes mellitus.
Diabetes Educ 2006;32(5):693–711.
[52] Iltz J, Baker DE, Setter SM, et al. Exenatide: an incretin mimetic for the treatment of type 2
diabetes mellitus. Clin Ther 28(5):652–65.
[53] Exenatide for type 2 diabetes. The Medical Letter 2006;47(1210):45–6.
[54] DeFronzo RA, Ratner RE, Han J, et al. Effects of exenatide (exendin-4) on glycemic control
and weight over 30 weeks in metformin-treated patients with type 2 diabetes. Diabetes Care
2005;28:1092–100.
[55] Buse JB, Henry RR, Han J, et al. Effects of exenatide (exendin-4) on glycemic control over 30
weeks in sulfonylurea-treated patients with type 2 diabetes. Diabetes Care 2004;27:2628–35.
[56] Kendall D, Riddle MC, Rosenstock J, et al. Effects of exenatide (exendin-4) on glycemic con-
trol over 30 weeks in patients with type 2 diabetes treated with metformin and sulfonylurea.
[57] Ratner RE, Maggs D, Nielsen LL, et al. Long term effects of exenatide therapy over 82 weeks
on glycemic control and weight in over-weight metformin treated patients with diabetes mel-
litus. Diabetes Obes Metab 2006;8:419–28.
[58] Heine RJ, Van Gaal LF, Johns D, et al. Exenatide versus insulin glargine in patients with
suboptimally controlled type 2 diabetes. Ann Intern Med 2005;143:559–69.
[59] Aschner P, Kipnes MS, Lunceford JK, et al. Effect of the dipeptidyl peptidase-4 inhibitor
sitagliptin as monotherapy on glycemic control in patients with type 2 diabetes. Diabetes
Care 2006;29:2632–7.
[60] Charbonnel B, Karasik A, Liu J, et al. Efficacy and safety of the dipeptidyl peptidase-4 in-
hibitor sitagliptin added to ongoing metformin therapy in patients with type 2 diabetes inad-
equately controlled with metformin alone. Diabetes Care 2006;29(12):2638–43.
[61] Rosenstock J, Brazg R, Andryuk PJ, et al. Efficacy and safety of the dipeptidyl peptidase-4
inhibitor sitagliptin added to ongoing pioglitazone therapy in patients with type 2 diabetes:
a 24 week multicenter, randomized, double-blind, placebo-controlled, parallel-group study.
Clinical Theraputics 2006;28:1556–67.
Med Clin N Am 91 (2007) 1125–1149
Nonalcoholic Fatty Liver Disease

as a Complication of Insulin Resistance
Manal F. Abdelmalek, MD, MPH*,
Anna Mae Diehl, MD
Division of Gastroenterology, Hepatology, and Nutrition, Duke University Medical Center,
P.O. Box 3913, Durham, NC 27710, USA
Nonalcoholic fatty liver disease (NAFLD) spans a spectrum of hepatic

pathology from hepatic steatosis (Fig. 1A) at the most clinically benign end,
through an intermediate lesiondnonalcoholic steatohepatitis (NASH)
(Fig. 1B), to cirrhosis (Fig. 1C) at the opposite end of the disease spectrum
[1–4]. Liver cell injury in NASH is often accompanied by spotty lobular infil-
tration with inflammatory cells, and sometimes with fibrosis. Hepatocyte
injury and inflammatory cell infiltration generally are worse in the perivenous
areas of the liver than in the periportal areas. Interestingly, cirrhotic livers that
result from NAFLD may have relatively little steatosis or inflammation. Thus,
based on histologic findings and negative testing for other causes of liver
disease, many of these cases were once considered ‘‘cryptogenic’’ [5]. Although
the pathogenesis is actively being investigated, NAFLD is currently consid-
ered the hepatic manifestation of the metabolic syndrome [6]. To formulate
an approach to diagnosis and treatment, it is first important to understand
the epidemiology, natural history, pathogenesis, and clinical features of
NAFLD.
Epidemiology
Triglyceride accumulation in the liver (hepatic steatosis), the hallmark of
NAFLD, can be identified by abdominal imaging techniques, providing
Dr. Abdelmalek is supported by the National Institute of Diabetes and Digestive and
Kidney Diseases of the National Institutes of Health (NIH) grant no. 5K23-DK062116.
Dr. Diehl is an investigator in the NIH funded Nonalcoholic Steatohepatitis Clinical
Research Network (5UO1-DK061713) and receives funding from the NIH grant no.
5RO1-DK053792.
E-mail address: manal.abdelmalek@duke.edu (M.F. Abdelmalek).
1126 ABDELMALEK & DIEHL
Fig. 1. Photomicrographs of (A) macrovesicular steatosis, (B) steatohepatitis with pericellular

fibrosis, and (C) macronodular cirrhosis that demonstrate the range of NAFLD. (Courtesy of
M Gottfried, MD, Durham, NC, and C Liu, MD, PhD, Gainesville, FL.)
a fairly convenient noninvasive means to estimate the prevalence of fatty

liver in the general population. A recent proton-nuclear magnetic resonance
(NMR) spectroscopy study of over 2000 adults in one urban area in the
United States suggests that values for liver triglyceride content are 5.5%
or lower in healthy, nonobese individuals [7]. If values above this cutoff
are considered to reflect abnormal hepatic accumulation of fat, about a third
of the general population has hepatic steatosis. Other work suggests that the
overall combined prevalence of chronic hepatitis C, hepatitis B, and alcohol-
associated liver disease accounts for fewer than 20% of serum alanine ami-
notransferase (ALT) elevations in the general United States population [8].
Hence, NAFLD is currently the leading cause of chronic liver disease in the
United States. Certain subgroups have an even higher prevalence of
NAFLD than the general population [9]. The risk of NAFLD is increased
in individuals who have type 2 diabetes [10,11], with abdominal sonography
demonstrating that at least half have hepatic steatosis [12]. Other popula-
tion-based studies, such as the Third National Health and Nutrition Exam-
ination Survey (NHANES III), [8,13] and the Dallas Heart Study [7],
NONALCOHOLIC FATTY LIVER DISEASE 1127
confirm that type 2 diabetes is highly correlated with NAFLD. The likeli-
hood of having NAFLD is also directly proportional to body weight [1].
Therefore, it is not surprising that most patients undergoing bariatric sur-
gery for morbid obesity have NAFLD (37%–93%), NASH (26%–44%),
and rarely, bridging fibrosis or cirrhosis (2%–9%), demonstrated by intra-
operative liver biopsy [14–16].
Natural history and prognosis

Although emerging evidence suggests that NAFLD is likely to be one of
the most common causes of cirrhosis in the United States and several other
countries, skepticism persists about its clinical significance. Recent studies
are helping to dispel the belief that NAFLD is a benign disease. In the pop-
ulation-based study cited earlier [17], individuals with a diagnosis of
NAFLD had a higher overall rate of mortality compared with age-, gender-
and body mass index (BMI)-matched controls who did not have NAFLD.
Liver disease contributed significantly to this difference, being the third lead-
ing cause of death in NAFLD patients, compared with the thirteenth cause
of death in those without this diagnosis. Overall, about 5% of patients who
have NAFLD develop cirrhosis over an average of a 7-year period, with
1.7% dying of complications of liver cirrhosis [17]. Although the likelihood
of developing cirrhosis is low, the high prevalence and chronic nature of
NAFLD translates to a significant health burden for the general
community.
One relatively unique feature that distinguishes patients who have cirrho-
sis caused by NAFLD from those who have cirrhosis caused by other liver
diseases is that patients who have NAFLD-related cirrhosis are more likely
to have other, serious end-organ complications of the metabolic syndrome,
particularly type 2 diabetes, cardiovascular disease, and non-liver cancers.
Indeed, as in other individuals who have the chronic metabolic syndrome,
the two leading causes of death are cardiovascular disease and cancer in
individuals who have NAFLD. There is growing evidence that NAFLD
contributes to the pathogenesis of the systemic insulin resistance syndrome.
Thus, even in the absence of cirrhosis, NAFLD probably plays some role in
the two leading causes of mortality that eventually ensue from long-term
consequences of the metabolic syndrome.
Although relatively few long-term follow-up studies of patients who have
NAFLD have been reported, the accumulating data about liver-specific out-
comes are very consistent [17–22]. First, it is apparent that in NAFLD, as in
other liver diseases, liver-specific morbidity and mortality are largely
restricted to individuals who develop cirrhosis. Second, as in other liver dis-
eases, patients who have NAFLD-related cirrhosis are more likely to have
other, serious end-organ complications of the metabolic syndrome, particu-
larly type 2 diabetes, cardiovascular disease [23], and non-liver cancers.
Third, patients who have advanced fibrosis caused by NAFLD are at risk
for hepatocellular cancer (5-year cumulative incidence as high as 20%)
and death from liver disease (10-year liver-related mortality rates as high
as 11%). Fourth, like many other causes of chronic hepatitis, NAFLD often
recurs after liver transplantation [24].
Clinical diagnosis and evaluation

Clinical manifestations
NAFLD is the hepatic manifestation of the insulin resistance (metabolic)
syndrome. Risk factors underlying ‘‘primary’’ NAFLD include obesity
[25,26], type 2 diabetes mellitus (DM) [12], dyslipidemia, hypertriglyceride-
mia [27], history of cyclic weight gain and loss [28], and hypertension [29].
Each of these conditions also conveys a risk for cardiovascular disease.
Thus, treatment of patients who have NAFLD should aim to identify and
treat associated metabolic factors such as obesity, glucose intolerance, dys-
lipidemia, and hypertension. Although the frequent association of NAFLD
with the metabolic syndrome is well-known [30,31], the metabolic syndrome
is now recognized as a strong predictor of the presence of NAFLD [32,33]
and progressive fibrosis. The worldwide increase in obesity and the
metabolic syndrome [34] has paralleled the worldwide increase in NAFLD
[34–38].
Before making a diagnosis of ‘‘primary’’ NAFLD, secondary causes of
NAFLD should also be considered and excluded when evaluating a patient
who has elevated serum aminotransferases (Box 1). Use of alcohol and med-
ications that cause hepatic steatosis (Box 2) should also be ascertained with
a detailed medical history. Finally, the presence of NAFLD should also be
considered in those who have persistent elevation of serum ALT levels for
which another cause cannot be found.
Patients who have NAFLD are generally asymptomatic, although some
report fatigue or right upper quadrant abdominal discomfort. The right up-
per quadrant discomfort may be mistaken for irritable bowel syndrome or
gallbladder disease. Hepatomegaly is commonly noted on physical examina-
tion. Patients who have NAFLD are subject to health problems directly re-
lating to the carriage of excess adipose tissue (ie, arthritis, nonspecific bodily
aches and pains, sleep disturbance, dyspnea on mild exertion, excessive
sweating, social stigmatization and discrimination) that may contribute to
low quality of life and depression. Patients who have advanced liver disease
may present with manifestations of muscle wasting, jaundice, gastrointesti-
nal bleeding, or ascites. By definition, most patients do not consume a signif-
icant amount of alcohol. Although is debatable what constitutes
‘‘significant’’ alcohol intake, consumption of fewer than 2 drinks (20 g)
per day in men and 1 drink (10 g)/day in women has been suggested as a
cutoff [3].
Box 1. Secondary causes of nonalcoholic fatty liver disease

Inborn errors of metabolism
Abetalipoproteinemia
Andersen’s disease
Cholesterol ester storage disease
Familial hepatosteatosis
Galactosemia
Glycogen storage disease
Hereditary fructose intolerance
Hypobetalipoproteinemia/dysbetalipoproteinemia
Lipoatrophy or lipodystrophy
Mauriac syndrome
Refsum’s syndrome
Schwachman syndrome
Systemic carnitine deficiency
Tyrosinemia
Weber-Christian syndrome
Wilson’s disease
Nutritional
Gastrointestinal surgery for morbid obesity
Severe starvation or cachexia
Kwashiorkor and marasmus
Refeeding syndrome
Total parenteral nutrition
Protein malnutrition
Infection
HIV Infection
Hepatitis C viral Infection (genotype 3)
Bacillus cereus toxin
Bacterial overgrowth
Miscellaneous
Inflammatory bowel disease
Acute fatty liver of pregnancy
Short bowel syndrome
Wolman’s disease
Environmental hepatotoxin exposure
Aside from abnormalities in serum glucose, cholesterol, and triglycerides,

abnormal liver enzymes are frequently noted in patients who have NASH;
however, liver enzymes are insensitive for the detection of NAFLD, and
may be normal in up to 78% of patients. Among patients who have normal
Box 2. Medications known to cause nonalcoholic fatty liver

disease
Retroviral agents
Zidovidine
Didanosine
Fialuridine
Antibiotics
Azaserine
Bleomycine
Puromycin
Tetracycline
Cytotoxic/cytostatic drugs
Azacitidine
Azauridine
L-asparginase
Methotrexate
Other drugs
Amiodarone
Aspirin
Calcium channel blocker
Cocaine
Ethyl bromide
Glucocorticoids
Heavy metals
Hydrazine
Hypoglycin
Orotate
Perhexiline maleate
Synthetic estrogens
Tamoxifen
Valproic acid
ALT levels, the full histological spectrum of disease may be present [39].
When present, liver enzyme elevations are generally restricted to ALT and
aspartate aminotransferase (AST). Glutamyltransferase and alkaline phos-
phatase may or may not be mildly elevated. Elevations of AST and ALT
are typically within 1.5 times the upper limit of normal levels, and rarely ex-
ceed 10 times the upper limit of normal [40]. Such elevations should prompt
further evaluation for acute hepatitis or acute on chronic liver injury.
Hyperglycemia (caused by the association with diabetes) is present in one
third of patients. Hyperlipidemia (usually triglycerides) is present in
approximately 20% to 25% of patients. Immunoglobulin A level may be el-

evated in about 25% of patients. Antinuclear antibody titer is positive in
about one third of patients, and does not necessarily represent the presence
of a coexistent autoimmune condition. Abnormal iron indices are common,
but generally do not indicate genetic hemochromatosis. Clinical evaluation
should focus on the presence of metabolic risk factors, as well as risk factors
for advanced hepatic fibrosis such as age greater than 45 years, diabetes, el-
evated BMI, an AST/ALT ratio greater than one, or thrombocytopenia [40].
Hepatic imaging
Once ongoing alcohol use (O20 g/day) and other common causes of liver
disease are excluded by clinical and laboratory evaluation, the liver is usu-
ally imaged by sonography, CT scan, or MRI. These modalities can be
used to determine the presence of biliary tract disease and focal liver disease,
which may be responsible for elevation of liver enzyme levels. Magnetic res-
onance spectroscopy has the advantage over the other commonly used im-
aging modalities of sonography, CT, and MRI for the assessment of
hepatic steatosis, because it is quantitative [41]; however, none of these
diagnostic studies distinguish between fatty liver, NASH, and NASH with
fibrosis, and therefore cannot be used to make these distinctions [42].
Although sonography is slightly more sensitive for detecting hepatic steato-
sis, CT scan is more specific but more expensive. Sufficient data on the com-
parative assessment of these tests on which to base a recommendation,
including their cost and predictive values, are lacking. Hence, a recommen-
dation about the use of one modality versus another cannot be made at this
time. It is, however, common practice to use either sonography or CT scan
in the diagnostic evaluation of presumptive NAFLD.
Histology and role of liver biopsy

The diagnosis of NASH, as opposed to benign steatosis, and its grade and
stage can only be made with precision by a liver biopsy [43]. A grading and
staging scheme (Box 3) was recently published [44] and is currently being
validated by the NASH Clinical Research Network. Hepatocyte injury and
inflammatory cell inflammatory cell infiltration generally are worse in the peri-
venous than in the periportal areas of the liver. The pattern of fibrosis is typi-
cally pericellular and sinusoidal (dubbed ‘‘chicken-wire’’ fibrosis). As cirrhosis
evolves, fibrous septa bridge portal and perivenous areas (see Fig. 1B).
Histologic endpoints are important in therapeutic trials and for transla-
tional research efforts focused on defining the pathophysiology of this condi-
tion. The role of liver biopsy in NAFLD in routine clinical care has been
debated. A liver biopsy can confirm the diagnosis of NASH, discern the pres-
ence of fibrosis or cirrhosis, and exclude other forms of chronic liver disease
[45,46]. A liver biopsy does carry a small risk of morbidity and mortality,
Box 3. Proposed grading and staging of NAFLD

Benign steatosis
Mild: <33% steatosis
Moderate: ‚33% and •66% steatosis
Severe: >66% steatosis
Nonalcoholic stevatohepatitis
Grade 1, mild
Steatosis: predominantly macrovesicular, ranges from less than
33% to up to 66% of the lobules
Ballooning: occasionally observed; zone 3 hepatocytes
Lobular inflammation: scattered and mild acute (polymorphs) and
chronic inflammation (mononuclear cells)
Portal inflammation: none or mild
Grade 2, moderate
Steatosis: any degree, usually mixed macrovesicular and
microvesicular
Ballooning: present in zone 3
Lobular inflammation: polymorphs may be noted associated with
ballooned hepatocytes, or pericellular fibrosis; greater than
mild chronic inflammation
Portal inflammation: none, mild to moderate
Grade 3, severe (florid steatohepatitis)
Steatosis: usually >66% (zone 3 or panacinar); commonly mixed
steatosis
Ballooning: predominantly zone 3; marked
Lobular inflammation: scattered acute and chronic inflammation;
polymorphs may appear concentrated in zone 3 areas of
ballooning and perisinusoidal fibrosis
Portal inflammation: mild or moderate; not predominant or
marked
Staging (requires Masson’s trichrome or equivalent stain):
a separate portal-based process with sparing of zone 3 has
been proposed, but remains to be established or refuted
Fibrosis stage
Stage 1: zone 3 perivenular, perisinusoidal, or pericellular fibrosis;
focal or extensive
Stage 2: as for stage 1 plus focal or extensive portal fibrosis
Stage 3: bridging fibrosis, focal or extensive
Stage 4: cirrhosis with or without residual perisinusoidal fibrosis
Modified from Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation
of a histological scoring system for nonalcoholic fatty liver disease. Hepatology
2005;41(6):1315; with permission.
and is expensive. In the absence of US Food and Drug Administration (FDA)-

approved therapies for NAFLD, with the exception of those patients who
have advanced fibrosis (stage F3–4), knowing the histology may not alter clin-
ical care of individual patients. Therefore, the decision to perform a biopsy in-
volves an assessment of the specific clinical circumstances in a given individual
who is suspected to have NAFLD. The cost and risks of the biopsy are gener-
ally weighed against the value of the information obtained from the biopsy in
estimating prognosis and guiding future management decisions. A reasonable
approach to a patient who is suspected to have NAFLD is clinical practice is to
consider a liver biopsy in those patients at increased risk for having fibrosis.
Such patients are those who have the metabolic syndrome, and those patients
who have persistently elevated liver enzymes despite optimal management of
associated metabolic conditions such as obesity, diabetes, or hyperlipidemia.
Obesity, older age, diabetes, and an AST:ALT ratio greater than 1 have been
associated with advanced fibrosis [40,47]. Assuming no absolute contraindica-
tions to the procedure, a liver biopsy should also be considered in patients who
have clinical features suggestive of advanced liver disease (ie, mild thrombocy-
topenia, hypoalbuminemia, hyperbilirubinemia, or altered liver contour on
imaging). A diagnostic algorithm is outlined in Fig. 2.
Pathogenesis: nonalcoholic fatty liver disease as a complication of insulin

resistance
Over the last decade, considerable progress has been made in delineating
the mechanisms that cause steatosis and steatohepatitis [48]; however, it
Fig. 2. Diagnostic approach to evaluation of NAFLD. IRS, insulin resistance syndrome;

AFLD, alcoholic fatty liver disease; *, cutoff for women is O10 g/day.
remains less evident why only a minority of individuals who have these
‘‘early’’ stages of fatty liver disease progress to cirrhosis or develop liver can-
cer. In the early stages of NAFLD, fat accumulates within hepatocytes when
mechanisms that promote lipid disposal (lipoprotein secretion and fatty acid
oxidation) cannot keep pace with mechanisms that promote lipid uptake
and biosynthesis. Increased hepatocyte accumulation of triglyceride (TG),
the hallmark of NAFLD, is closely linked to insulin resistance. Insulin resis-
tance increases lipolysis of peripheral adipose tissue, with resultant increased
fat influx into the liver in the form of free fatty acids (FFA). Furthermore,
hyperinsulinemia promotes de novo TG synthesis within the liver and in-
hibits free fatty acid oxidation. Hepatic TG accumulation results from alter-
ations of factors (hepatic and systemic) that control the balance between
hepatic lipid input (ie, uptake and synthesis), and output (ie, oxidation);
however, the mechanisms driving TG accumulation differ among individ-
uals and within a given individual at different points in time. In turn, these
differences in steatosis pathogenesis prompt heterogeneous compensatory
responses that exert different outcomes on hepatocyte viability. Studies of
different rodent models of NAFLD support the concept that liver injury
in NAFLD results from hepatocyte accumulation of FFA and the associ-
ated cellular stress responses that develop as fatty cells attempt to dispose
of the excess lipids [48].
Three of the best-characterized factors that modulate the evolution of
fatty liver disease are fatty acids, tumor necrosis factor alpha (TNFa),
and adiponectin [49–51]. In addition to modulating lipid homeostasis, var-
ious adipocytokines also regulate carbohydrate metabolism by influencing
cellular sensitivity to insulin. Fatty acids routinely traffic between the liver
and adipose tissues that are important sources of TNFa and adiponectin
[52]. Interestingly, the latter two proteins regulate fatty acid turnover within
hepatocytes. Adiponectin generally reduces lipid accumulation within hepa-
tocytes by inhibiting fatty acid import and increasing fatty acid oxidation
and export. Adiponectin is also a potent insulin-sensitizing agent [53].
TNFa antagonizes the actions of adiponectin, and thereby promotes hepa-
tocyte steatosis and insulin resistance [53]. Moreover, the severity of
NAFLD-related liver damage parallels the severity of insulin resistance,
with cirrhosis and liver-related mortality being greatest in patients who
have DM [54,55]. Despite the apparent robustness of this clinical correla-
tion, it has been difficult to understand whether the association between
NAFLD and DM reflects the negative outcomes of inhibiting insulin actions
in the liver itself, or NAFLD results from loss of insulin activity in extra-he-
patic tissues. If hepatic (rather than systemic) insulin resistance is in fact the
driving force for NAFLD, this might explain why systemic insulin resistance
is not a prerequisite for hepatic steatosis [55]. On the other hand, it does not
clarify why many humans and experimental animals who have NAFLD
exhibit both increased rates of hepatocyte lipid synthesis (suggestive of
enhanced hepatic insulin activity), and hyperinsulinemia with excessive
hepatic glucose output (typical of hepatic insulin resistance). The latter find-
ings suggest that differential inhibition of various insulin-regulated signaling
pathways within hepatocytes might be necessary for NAFLD pathogenesis.
The confusing role of insulin resistance in NAFLD pathogenesis/progres-
sion has not been resolved by studies of animal models of NAFLD. Indeed,
such work only underscores the complexity of this relationship. Similar to
many obese humans, leptin-deficient (ob/ob) and leptin-resistant (db/db)
murine models of NAFLD exhibit hyperinsulinemia, hyperglycemia, and
evidence of both systemic and hepatic insulin resistance [56]. Treatment
with insulin sensitizing agents has been beneficial, albeit inconsistently. On
the other hand, methionine and choline deficient (MCD) diet-fed mice de-
velop NASH and progressive hepatic fibrosis despite diet-induced decreases
in serum insulin and glucose, suggestive of enhanced systemic sensitivity to
insulin. Nevertheless, insulin-sensitizing agents also seem to improve MCD
diet-induced liver disease [57]. Recent evidence of reduced insulin receptor
and insulin receptor substrate 2 (IRS2) tyrosine phosphorylation in the
livers of MCD diet-fed mice supports the possibility that these animals
have hepatic insulin resistance [58]. On the other hand, mice who have he-
patocyte-specific activation of protein kinase B/Akt, a prototypical down-
stream target of insulin signaling, develop severe NASH, progressive
fibrosis, and eventual hepatocellular carcinoma [59], suggesting that certain
insulin-initiated signals might actually promote (rather than prevent) liver
damage. At this point, no unifying hypothesis has emerged to reconcile these
seemingly contradictory results.
Efforts to equate DM with extremely severe insulin resistance contribute to
the confusion. The flaw in such logic is demonstrated by recent data from the
Diabetes Prevention Program (DPP). In that large population of insulin-resis-
tant adults, hyposecretion of insulin was a better predictor of subsequent DM
than the initial severity of insulin resistance [60]. This finding led the DPP in-
vestigators to speculate that insulin resistance ‘‘unmasked’’ individuals who
have an inherently reduced capacity for pancreatic beta cell hyperplasia.
The latter individuals eventually became overtly diabetic. This insight raises
the intriguing possibility that cirrhosis, which strongly correlates with DM,
might result from an inadequate hepatic hyperplastic response to similar met-
abolic stress. The latter is a particularly attractive concept because hepatocyte
hyperplasia is necessary to replace dying fatty hepatocytes, and progressive
liver damage results when regeneration cannot keep pace with injury.
Furthermore, situations that increase TNFa relative to adiponectin pro-
mote hepatic steatosis and insulin resistance [49]. TNFa also increases mito-
chondrial generation of reactive oxygen species (ROS), promotes hepatocyte
apoptosis, and recruits inflammatory cells to the liver. Hence, protracted ex-
posure to TNFa generates oxidative and apoptotic stress that sometimes
overwhelms antioxidant and antiapoptotic defenses, leading to steatohepa-
titis [61]. Studies in mouse models of NASH, as well as mice who have eth-
anol-induced steatohepatitis, prove that over-production of TNFa relative
to adiponectin causes steatohepatitis, because treatments that inhibit TNFa

or that increase adiponectin improve steatohepatitis in all of these models
[48]. In addition, studies in humans who have NASH demonstrate that
the relative risk of developing steatohepatitis correlates with increases in
TNFa or decreases in adiponectin levels [62].
Given strong experimental and clinical evidence that unopposed TNFa
activity promotes steatosis and steatohepatitis, it is interesting that there
is now compelling evidence that the simple accumulation of fatty acids
within hepatocytes is sufficient to trigger these cells to produce TNFa
[63]. Fatty acids induce signaling in hepatocytes that activates kinases,
such as inhibitor kappa kinase (IKK) beta that, in turn, activate the nuclear
factor-kappa B (NF-kB) transcription factor, driving hepatocyte synthesis
of TNFa and interleukin (IL)-6 [63,64]. Recent studies in transgenic mice
who have hepatocyte-specific overexpression of IKK-beta demonstrate
that hepatocyte-derived IL-6 is responsible for systemic insulin-resistance
[64]. Therefore, like adipose tissue, fatty livers (and specifically, fatty hepa-
tocytes) also make soluble factors that circulate to distant tissues and con-
tribute to systemic insulin resistance (ie, the metabolic syndrome). These
mechanisms of liver injury are likely similar in obese and nonobese individ-
uals who develop NASH [65]. In support of this concept, recent data suggest
that gut bacteria of some nonobese individuals might promote excessive he-
patic accumulation of fatty acids, as well as exposure to other bacterial fac-
tors (eg, lipopolysaccharide or other Toll-like receptor agonists) that trigger
hepatic TNFa production. As in obese individuals, increased TNFa would
antagonize adiponectin activity, and promote steatosis, steaohepatitis, and
insulin resistance [66].
It is generally believed that progression from fatty liver disease to
cirrhosis is predominantly dictated by the severity of oxidant stress and
consequent necroinflammation that occurs in individuals who have steato-
hepatitis [67,68]; however, findings in animal models of steatohepatitis
cast some doubt on this assumption, because mice that develop severe stea-
tohepatitis do not uniformly progress to cirrhosis [69]. In fact, progression
to cirrhosis is also poorly predicted by the gravity of the injurious insult
in human fatty liver disease. For example, although there is no doubt that
alcohol is hepatotoxic, most lifelong heavy drinkers do not become cirrhotic
[70]. Similarly, although obesity clearly increases exposure to fat-derived in-
flammatory mediators, some morbidly obese individuals have normal livers
at the time of gastric bypass surgery [71].
These apparent paradoxes might be explained by the failure to acknowl-
edge that liver damage is determined by the adequacy of liver repair mech-
anisms as well as the severity of a particular noxious insult. Individuals who
are ‘‘poor repairers’’ suffer more net liver damage for any given level of in-
jury than those who are ‘‘average repairers,’’ whereas those who are ‘‘super
repairers’’ may survive relatively unscathed, with little evidence of liver dam-
age despite a significant noxious exposure. Viewed from this perspective,
individuals who merely develop steatosis despite constant bombardment

with inflammatory factors might be ‘‘super repairers,’’ whereas those who
develop steatohepatitis have only ‘‘average’’ repair capabilities, and the
minority who have ‘‘poor’’ repair abilities develop cirrhosis.
Indeed, the possibility that differences in repair responses might contribute
to liver disease outcome merits consideration in fatty liver disease because this
condition is often associated with obesity, and adipose tissue is an important
source of various mediators that modulate wound-healing responses [49–51].
Indeed, hepatic stellate cells (HSC) express receptors for several of the adi-
pose-derived factors that modulate HSC activation, including leptin, angio-
tensin, adiponectin, and norepinephrine [72]. Studies in mice demonstrate
that leptin, angiotensin, and norepinephrine promote HSC proliferation, up-
regulate HSC expression of pro-fibrogenic cytokines such as transforming
growth factor-beta (TGF-b), and induce collagen gene expression [72,73].
Conversely, adiponectin appears to inhibit HSC activation and decrease liver
fibrosis [74]. It is likely that plasminogen activator inhibitor1 (PAI-1) also reg-
ulates HSC because it has been shown to influence fibrosis in other tissues [75].
In summary, studies of animal models and patients who have fatty liver
disease suggest that the early-intermediate stages of this condition (ie, stea-
tosis and steatohepatitis) are caused by excessive exposure to fatty acids and
inflammatory cytokines that induce hepatocyte steatosis, threaten hepato-
cyte viability, and promote hepatic and systemic insulin resistance. Resul-
tant increases in the rate of liver cell death trigger repair responses. The
latter are modulated by various factors that regulate the activation of he-
patic stellate cells. In some individuals, the net effect of this process is ‘‘un-
healthy’’ repair, with resultant cirrhosis. More research is needed to clarify
the molecular basis for inter-individual differences in repair responses that
are triggered by chronic fatty liver injury. Improved understanding of
such pathobiology should enhance identification of individuals who are at
greatest risk for developing cirrhosis, as well as the development of effective
treatments to abort disease progression.
Potential management strategies

The objective of treating NAFLD is to prevent disease progression and
long-term complications related to cirrhosis. No practice guidelines for
the management of NAFLD currently exist; however, flexible management
strategies that may be tailored to specific patient circumstances are pro-
posed. Specific recommendations are based on relevant and published infor-
mation. Circumstances in which evidence-based literature does not provide
data are also be specified.
General considerations
NAFLD should be considered in any patient who presents with abnor-
mal liver enzymes or hepatomegaly, who is incidentally noted to have
a ‘‘bright’’ liver on sonogram, or who has any features of the metabolic syn-
drome. Initially, it is important to consider and exclude other causes of
chronic liver disease, because NAFLD may coexist with other conditions
that may require alternative therapies (ie, hepatitis C virus [HCV] infection).
Whether alcohol use should be completely prohibited or diminished to levels
less than 20 g/day, both to define the existence of NAFLD and for counsel-
ing of patients, remains unclear. For now, physicians need to tailor recom-
mendations regarding minimal alcohol consumption to individual patients
depending on patient interview and associated liver histology. Because liver
enzymes are not a sensitive measure for the presence of significant liver dis-
ease, even patients who have normal liver enzymes can have evidence of
NASH or cirrhosis [39]. Therefore, liver biopsy, which remains the ‘‘gold-
standard’’ for resolution of histologic abnormalities, should be considered
in those persons at risk for fibrosis. In the future, magnetic resonance spec-
troscopy may prove to be the optimal means of quantifying visceral and pe-
riphery adiposity. A schematic diagram of a proposed therapeutic approach
is outlined in Fig. 3.
Weight management
Diet and exercise continue to be the cornerstone of therapeutic interven-
tions. Because obesity consistently has been associated with NAFLD, weight
reduction has been one of the initial forms of treatment interventions. Medical
interventions include diet and exercise, but preferably a combination of the
Fig. 3. Approach to treatment of NAFLD. IGC, impaired glycemic control; HTN, hyperten-
sion; HCC, hepatocellular carcinoma.
two. Most of the studies have suggested that weight loss can be associated with
biochemical improvement [76–78]. Rarely, rapid weight loss may be associ-
ated with worsening inflammation [79]. Patients who are overweight (body
mass index O25 kg/m2) and have NAFLD should be considered for a weight
loss program. A target of 10% of baseline weight is often used as an initial goal
of weight loss [80]; however, even attaining as little as a 3% decrease in weight
may enhance glucose tolerance and improve post-exercise insulin sensitivity
[81]. Weight loss should proceed at a rate of 1 to 2 pounds a week [82]. Dietary
recommendations generally include both caloric restriction and a decrease in
saturated fats as well as total fats to 30% or less of total calories [83,84]. Cur-
rently there are no data to support or refute the value of decreasing saturated
fats and increasing the fiber content of diet on NAFLD. On the other hand,
decreased carbohydrate consumption may decrease insulin secretion [85,86].
Diet modifications are usually accompanied by a recommendation to exercise
regularly. Both intermittent as well as daily exercise can help achieve weight
loss and improve insulin sensitivity, even in the physically fit and fat individual
[87]. Routine exercise alters substrate use in skeletal muscle and improves in-
sulin sensitivity [88–90]. Unfortunately, despite earnest efforts, only about one
third of patients are effective in achieving target levels of exercise [91–93].
Obese persons may be even more resistant to weight loss [94,95]. Studies of
diet and exercise therapy reveal improved biochemical parameters but vari-
able changes in histology[77,79,96–98]. Further research is needed to substan-
tiate the benefits of weight loss on hepatic histopathology.
For patients who have a body mass index greater than 35 kg/m2 and
NAFLD, more aggressive weight management, such as bariatric surgery,
should be considered in individuals who do not have evidence of portal hy-
pertension. The current bariatric surgical procedures can be divided into
two categories: restrictive or malabsorptive surgeries. Restrictive surgeries
(ie, vertical banded gastroplasty or laproscopic gastric band procedures) de-
crease the capacity of the stomach and the amount of food consumption.
Malabsorptive surgeries (ie, Roux-en-Y gastric bypass) bypass a large por-
tion of small intestine, thus preventing absorption of fats and nutrients. The
decision to perform this surgery should take into consideration the morbid-
ity and mortality associated with the procedure, as well as the risk of devel-
oping subacute nonalcoholic steatohepatitis and liver failure during rapid
weight loss; however, several studies have reported beneficial effects of bari-
atric surgery, with improvements in diabetes, hyperlipidemia, and hyperten-
sion [99], as well as steatosis and NASH [100]. Whether bariatric surgery
also improves [101–103] or worsens fibrosis[97,104] remains unclear.
Insulin sensitizing agents for the treatment of nonalcoholic

fatty liver disease
Current pharmacologic treatment approaches are guided by knowledge
of NAFLD pathogenesis. To date, treatment approaches have been tailored
to improving hepatic insulin resistance or inhibiting the inflammatory re-

sponse responsible for chronic hepatitis. Multiple pharmacologic agents
have been tested for the treatment of NASH. Insulin-sensitizing agents
are particularly attractive therapeutic agents for NASH because they im-
prove peripheral and hepatic insulin resistance (Table 1). Metformin, a bi-
guanide antihyperglycemic agent, has been evaluated in animals as well as
patients who have NAFLD. In animal models of fatty liver, metformin im-
proved hepatic steatosis and decreased hepatic tumor necrosis factor (TNF)
and lipogenic transcription factors [105]. Several small pilot studies of up to
6 months duration using metformin at doses of 1.0 to 1.5 g/day have showed
improvement in ALT levels compared with baseline [106–108]. Other pilot
studies, using metformin doses of 1.5 to 2.0 g/day noted improvement in
liver biochemistry, improved insulin resistance, and decreased echogenicity
of the liver by ultrasound imaging [109], however, change in liver histology
was variable [110,111].
The thiazolidinediones are promising therapies for NASH because these
peroxisome proliferator-activated receptor (PPARg) agonists inhibit inflam-
mation, alter skeletal muscle glucose uptake, decrease central adiposity, pro-
mote adipocyte differentiation, and alter mitochondrial mass and
thermogenesis. PPARg agonists are also attractive therapeutic agents because
of their anti-fibrotic properties. Troglitazone showed promising results in a pi-
lot trial before being removed from the market because of idiosyncratic liver
toxicity [112]. Although the PPARg agonists appear to be safer, they are
not currently US Food and Drug Administration (FDA)-approved for use
in patients who have chronic liver disease. Two well-designed pilot studies us-
ing rosiglitazone (4 mg twice daily) [113] and pioglitazone (30 mg daily) [114]
showed improvement in liver aminotransferases, radiologic, or histologic end-
points [114–116]. Pioglitazone, but not rosiglitazone, was associated with im-
provement in the overall fibrosis stage. A randomized trial of 20 nondiabetic
patients who had NASH comparing pioglitazone (30 mg daily) plus vitamin E
(400 IU daily) with vitamin E (400 IU daily) alone, found that both groups im-
proved hepatic steatosis grade, although the improvement was greater with
pioglitazone [115]; however, two thirds of patients experienced weight gain
and the biochemical improvement reversed after discontinuation of treat-
ment. Interpretation of these studies without a placebo group is difficult be-
cause ALT levels, hepatic steatosis, and inflammation on liver biopsy are
insensitive measures of disease activity, and may improve over time as fibrosis
progresses. In a recent placebo-controlled trial that compared diet plus piogli-
tazone (n ¼ 26) with diet plus placebo (n ¼ 21), pioglitazone improved glyce-
mic control and glucose tolerance (P !.001), normalized liver
aminotransferase levels as it decreased plasma AST levels (40% versus 21%,
P ¼ .04), decreased ALT levels (58% versus 34%, P ! .001), decreased hepatic
fat content (54% versus 0%, P !.001), and increased hepatic insulin sensitiv-
ity (48% versus 14%, P ¼ .008) [116]. Administration of pioglitazone
improved the histologic features of steatosis (P ¼ .003), ballooning necrosis
Table 1
Insulin sensitizing agents in the treatment of NAFLD
Reference Therapy N Study type Duration Liver biochemistry Liver histology
Marchesini et al [106] Metformin 14 Open label 4 mo Improved NA
Magalotti et al [107] Metformin 11 Open label 6 mo Improved NA
Schwimmer et al [108] Metformin 10 Open label 6 mo Improved NA
Uygun et al [109] Metformin 34 RCT, open label 6 mo Improved Improved (S)
NONALCOHOLIC FATTY LIVER DISEASE

Nair et al [110] Metformin 15 Open label 12 mo Improved Variable
Bugianesi et al [111] Metformin versus vit 55a Open label, randomized 12 mo Improved Improveda (S, I, F)
E versus diet
Caldwell et al [112] Troglitazoneb 10 Open label % 6 mo Improved Improved (I)
Neushwander-Tetri Rosiglitazone 30 Open label 48 wk Improved Improved (I, F)c
et al [113]
Promrat et al [114] Pioglitazone 18 Open label 12 mo Improved Improved (S, I, F)
Bajaj et al [125] Pioglitazone 11 Open label 4 mo Not done NAd
Belfort et al [116] Pioglitazone 26 RCT 12 mo Improved Improved (S, I)d
Sanyal et al [115] Vit E versus vit 21 RCT, open label 6 mo Not reported Pioglitazone
E þ pioglitazone improved (S, I)
This table is not a comprehensive list of all clinical studies in patients with NAFLD. But is only intended to reflect only those trials with O10 adult pa-
tients, R4 months duration, and with histologic endpoints.
a
55 of 110 patients enrolled were assigned to metformin. Of these, 17 of 55 patients underwent biopsy to assess change in liver histology.
b
Drug removed from the market because of idiosyncratic hepatotoxicity.
c
Improvement seen in zone 3 fibrosis but no improvement noted in overall fibrosis stage.
d
Magnetic resonance spectroscopy noted improvement in hepatic steatosis.
Abbreviations: F, fibrosis; I, inflammation; mo, months; NA, not available; RCT, randomized controlled trial; S, steatosis; V it E, vitamin E; wk, weeks.
1141
(P ¼ .02), and necroinflammation (P ¼ .001) compared with placebo [116].

The reduction in fibrosis, however, did not differ significantly from that in
the placebo group [116]. Despite these promising results, potential hepatotox-
icity in the setting of liver disease remains a concern. Large randomized
placebo-controlled trials are currently underway to further evaluate these
encouraging data.
Because hypertriglyceridemia and low high-density lipoprotein (HDL)
cholesterol levels are a manifestation of insulin resistance, and occur com-
monly among patients who have NAFLD, several investigators have used
lipid lowering agents (ie, statins, fibrates) to treat NAFLD (Table 2). Unfor-
tunately, these agents have not had a proven benefit [117]. Two pilot studies
using atorvastatin showed improvement in biochemical [118] and histologic
parameters in a small sample of patients who had NAFLD [119,120]; how-
ever, other reports have suggested no significant histologic differences be-
tween controls and patients who had NAFLD using various statin drugs
[121]. In any case, it does not appear that patients who have NAFLD are
at increased risk of hepatotoxicity with use of these agents [122,123]. Two
small studies have also examined the fenofibrates, one 12 month trial of clo-
fibrate (2 g daily) [124] showed no improvement in liver enzymes or histol-
ogy, whereas gemfibrazole (600 mg daily) [117] improved liver enzymes after
4 weeks compared with no treatment. The role of resin binding agents has
not yet been investigated. Therefore, no recommendation about the safety
or efficacy of lipid-lowering agents in the management of NAFLD can be
made at this time.
The role of pharmacologic agents to induce weight loss or improve oxi-
dant stress in patients who have NAFLD has not been well-studied. The
role of antioxidants (eg, vitamin C, vitamin E, betaine, S-adenosylmethio-
nine, and the like) in improving insulin resistance and the histologic features
of NAFLD remains uncertain. Given the small sample size, the randomized,
placebo-controlled trials that have been performed for NAFLD have not
shown convincing shown evidence of histologic improvement. Therefore,
before recommending these (or other) agents as generalized treatments for
NAFLD, larger, prospective controlled trials are needed to confirm their
safety, efficacy, and impact on the natural history of disease progression.
Monitoring of patients who have nonalcoholic fatty liver disease

Regardless of the treatment approach selected, all patients who have
NAFLD merit routine follow-up to monitor for evidence of disease progres-
sion. Routine monitoring by liver enzymes alone is insufficient, given their
lack of sensitivity for detecting inflammation or fibrosis. In addition, it
may take several decades for a patient at risk for disease progression to
develop complications of cirrhosis. In view of the insensitivity of existing
noninvasive tests for detecting progression of fibrosis, a periodic (every 7–10
years) liver biopsy has been suggested for patients who have NAFLD, as for
Table 2
Lipid lowering agents in the treatment of NAFLD
Study Liver Liver
Reference Therapy N type Duration biochemistry histology
a
Gomez-Dominguez Atorvastatin 22 Open label 12 mo Improved Not available
et al [118]
Kiyici et al [119] Atorvastatin 27 Open label 6 mo Improved Not available
Hatzitolios et al [120] Atorvastatin 28 Open label 6 mo Improved Not available
Rallidis et al [121] Provastatin 5 Open label 6 mo Improved Improved (I)
Laurin et al [124] Clofibrate 16 Open label 12 mo No change No change
Basaranoglu Gemfibrazil 46 RCT 1 mo Improved Not available
et al [117]
a
Improvement in hepatic steatosis as assessed by ultrasonography.
patients who have other types chronic liver disease. Studies of noninvasive
fibrosis biomarkers are ongoing to determine if they might be helpful in
monitoring fibrosis progression/regression, obviating the need for repeat
biopsy in some individuals.
Given the intimate association of NAFLD with insulin resistance, a bian-
nual physical examination and blood work to monitor for complications of
the metabolic syndrome should be considered. General surveillance, coun-
seling, and early intervention to control weight gain, hyperglycemia, hyper-
lipidemia, and hyperuricemia are warranted. If a decline in liver function
(eg, bilirubin, albumin, prothromin time) or significant drop in platelet
count (ie, marker of portal hypertension) is noted during follow-up, imaging
of the abdomen should be performed to assess the liver contour and pres-
ence of splenomegaly, intra-abdominal varices or ascites. Once cirrhosis is
diagnosed by imaging or liver biopsy, screening, prophylaxis, or manage-
ment of varices, ascites, encephalopathy, and hepatocellular carcinoma is
initiated. Assuming no contraindications exist, such complications warrant
consideration of liver transplantation and early referral to a liver transplant
center. In patients who have any form of chronic liver disease, most hepatol-
ogists elect to vaccinate patients against hepatitis A and B to protect the pa-
tient from superinfection and the associated risk of acute on chronic liver
injury. Readers should familiarize themselves with the screening, prevention,
and palliation of complications related to cirrhosis, because that discussion
is beyond the scope of this article.
Summary
Largely because of epidemic obesity and diabetes, NAFLD is now ac-
knowledged as the most common cause of chronic liver disease in the West-
ern world. As such, physicians across all specialties will be confronted with
the evaluation and management of liver disease as a complication of insulin
resistance. It is clear that NAFLD leads to liver-related morbidity and
mortality in a subset of people; however, a better understanding of the nat-

ural history for NAFLD will permit better identification of those patients at
risk for fibrosis progression. Although diagnosis and evaluation are rela-
tively straightforward, many issues remain unresolved. Large series of
well-characterized patients will need to be followed to better establish the
natural history and to clearly define the associated morbidity and mortality
caused by this chronic condition. Furthermore, noninvasive means of dis-
cerning risk factors for fibrosis progression will allow clinicians and investi-
gators to select out patients at high risk for disease progression for more
comprehensive diagnostic evaluations, follow-up, and treatment interven-
tions. Currently, treatment is limited to weight loss, exercise, and treatment
of individual features of the metabolic syndrome. Effective pharmacologic
therapies to improve steatohepatitis are currently being investigated, and
several promising agents are on the horizon.
References
[1] Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221–31.
[2] Sanyal AJ. AGA technical review on nonalcoholic fatty liver disease. Gastroenterology
2002;123:1705–25.
[3] Neuschwander-Tetri BA, Caldwell SH. Nonalcoholic steatohepatitis: summary of an
AASLD single topic conference. Hepatology 2003;37:1202–19.
[4] Teli MR, James OF, Burt AD, et al. The natural history of nonalcoholic fatty liver: a
follow-up study. Hepatology 1995;22:1714–9.
[5] Brunt EM. Pathology of nonalcoholic steatohepatitis. Hepatol Res 2005;33(2):68–71.
[6] Marchesini G, Forlani G. NASH: from liver disease to metabolic disorders and back to
clinical hepatology. Hepatology 2002;35:497–9.
[7] Browning JD, Szczepaniak LS, Dobbins R, et al. Prevalence of hepatic steatosis in an urban
population in the United States: impact of ethnicity. Hepatology 2004;40:1387–95.
[8] Clark JM, Brancati FL, Diehl AM. The prevalence and etiology of elevated aminotransfer-
ase levels in the United States. Am J Gastroenterol 2003;98(5):960–7.
[9] Weston SR, Leyden W, Murphy R, et al. Racial and ethnic distribution of nonalcoholic
fatty liver in persons with newly diagnosed chronic liver disease. Hepatology 2005;41:
372–9.
[10] Poonawala A, Nair SP, Thuluvath PJ. Prevalence of obesity and diabetes in patients with
cryptogenic cirrhosis: a case-control study. Hepatology 2000;32:689–92.
[11] Clark JM, Diehl AM. Hepatic steatosis and type 2 diabetes mellitus. Curr Diab Rep 2002;
2(3):210–5.
[12] Gupte P, Amarapurkar D, Agal S, et al. Non-alcoholic steatohepatitis in type 2 diabetes
mellitus. J Gastroenterol Hepatol 2004;19:854–8.
[13] Ioannou GN, Boyko ES, Less SP. The prevalence and predictors of elevated serum amino-
transferase activity in the United States in 1999–2002. Am J Gastroenterol 2006;101(1):
76–82.
[14] Boza C, Riquelme A, Ibanez L, et al. Predictors of nonalcoholic steatohepatitis (NASH) in
obese patients undergoing gastric bypass. Obes Surg 2005;15:1148–53.
[15] Srivastava S, Younossi ZM. Morbid obesity, nonalcoholic fatty liver disease, and weight
loss surgery. Hepatology 2005;42:490–2.
[16] Oliveira CP, Faintuch J, Rascovski A, et al. Lipid peroxidation in bariatric candidates with
non-alcoholic fatty liver disease (NAFLD)dpreliminary findings. Obes Surg 2005;15:
502–5.
[17] Adams LA, Lymp JF, St Sauver J, et al. The natural history of nonalcoholic fatty liver dis-
ease: a population-based cohort study. Gastroenterology 2005;129(1):113–21.
[18] Hashimoto E, Yatsuji S, Kaneda H, et al. The characteristics and natural history of Japa-
nese patients with nonalcoholic fatty liver disease. Hepatol Res 2005;33:72–6.
[19] Bugianesi E, Leone N, Vanni E, et al. Expanding the natural history of nonalcoholic stea-
tohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology
2002;123:134–40.
[20] Powell EE, Cooksley WG, Hanson R, et al. The natural history of nonalcoholic steatohe-
patitis: a follow-up study of forty-two patients for up to 21 years. Hepatology 1990;11:
74–80.
[21] Matteoni CA, Younossi ZM, Gramlich T, et al. Nonalcoholic fatty liver disease: a spectrum
of clinical and pathological severity. Gastroenterology 1999;116:1413–9.
[22] Ratziu V, Bonyhay L, Di Martino V, et al. Survival, liver failure, and hepatocellular carci-
noma in obesity-related cryptogenic cirrhosis. Hepatology 2002;35:1485–93.
[23] Jepsen P, Vilstrup H, Mellemkjaer L, et al. Prognosis of patients with a diagnosis of fatty
liverda registry-based cohort study. Hepatogastroenterology 2003;50:2101–4.
[24] Contos MJ, Cales W, Sterling RK, et al. Development of nonalcoholic fatty liver disease
after orthotopic liver transplantation for cryptogenic cirrhosis. Liver Transpl 2001;7:
363–73.
[25] Silverman JF, O’Brien KF, Long S, et al. Liver pathology in morbidly obese patients with
and without diabetes. Am J Gastroenterol 1990;85:1349–55.
[26] Andersen T, Christoffersen P, Gluud C. The liver in consecutive patients with morbid obe-
sity: a clinical, morphological, and biochemical study. Int J Obes 1984;8:107–15.
[27] Assy N, Kaita K, Mymin D, et al. Fatty infiltration of liver in hyperlipidemic patients. Dig
Dis Sci 2000;45:1929–34.
[28] Berkovitz D. Metabolic changes associated with obesity before and after weight reduction.
JAMA 1964;187:399–403.
[29] Donati G, Stagni B, Piscaglia F, et al. Increased prevalence of fatty liver in arterial hyper-
tensive patients with normal liver enzymes: role of insulin resistance. Gut 2004;53:1020–3.
[30] Marchesini G, Natale S, Manini R, et al. Review article: the treatment of fatty liver disease
associated with the metabolic syndrome. Aliment Pharmacol Ther 2005;22(Suppl 2):37–9.
[31] Reaven GM. Banting Lecture 1988. Role of insulin resistance in human disease. Diabetes
1988;37:1595–607.
U.S. adults. Diabetes Care 2004;27:2444–9.
[33] Hamaguchi M, Kojima T, Takeda N, et al. The metabolic syndrome as a predictor of non-
alcoholic fatty liver disease. Ann Intern Med 2005;143(10):722–8.
[34] World Health Organization Western Pacific Region, International Association for the
Study of Obesity/International Obesity Task Force. The Asia-Pacific perspective: Redefin-
ing obesity and its treatment. Melbourne (Australia). Health Communications Australia;
2000.
[35] Bellentani S, Saccoccio G, Masutti F, et al. Prevalence of and risk factors for hepatic stea-
tosis in Northern Italy. Ann Intern Med 2000;132:112–7.
[36] Akbar DH, Kawther AH. Nonalcoholic fatty liver disease in Saudi type 2 diabetic subjects
attending a medical outpatient clinic: prevalence and general characteristics [Letter]. Dia-
betes Care 2003;26:3351–2.
[37] Chitturi S, Farrell GC, George J. Non-alcoholic steatohepatitis in the Asia-Pacific region:
future shock? J Gastroenterol Hepatol 2004;19:368–74.
[38] Kojima S, Watanabe N, Numata M, et al. Increase in the prevalence of fatty liver in
Japan over the past 12 years: analysis of clinical background. J Gastroenterol 2003;38:
954–61.
[39] Mofrad P, Contos MJ, Haque M, et al. Clinical and histologic spectrum of non-alcoholic
fatty liver disease associated with normal ALT values. Hepatology 2003;37(6):1286–92.
[40] Angulo P, Keach JC, Batts KP, et al. Independent predictors of liver fibrosis in patients with
non-alcoholic steatohepatitis. Hepatology 1999;30:1356–62.
[41] Szczepaniak LS, Nurenberg P, Leonard D, et al. Magnetic resonance spectroscopy to mea-
sure hepatic triglyceride content: prevalence of hepatic steatosis in the general population.
Am J Physiol Endocrinol Metab 2005;288:E462–8.
[42] Joy D, Thava VR, Scott BB. Diagnosis of fatty liver disease: is biopsy necessary? Eur J Gas-
troenterol Hepatol 2003;15:539–43.
[43] Saadeh S, Younossi ZM, Remer EM, et al. The utility of radiological imaging in non-alco-
holic fatty liver disease. Gastroenterology 2002;123:745–50.
[44] Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring
system for nonalcoholic fatty liver disease. Hepatology 2005;41(6):1313–21.
[45] Van Ness MM, Diehl AM. Is liver biopsy useful in the evaluation of patients with chron-
ically elevated liver enzymes? Ann Intern Med 1989;111:473–8.
[46] Adams LA, Lindor KD, Angulo P. The prevalence of autoantibodies and autoimmune
hepatitis in patients with non-alcoholic fatty liver disease. Am J Gastroenterol 2004;99:
1316–20.
[47] Younossi ZM, Gramlich T, Matteoni CA, et al. Nonalcoholic fatty liver disease in patients
with type 2 diabetes. Clin Gastroenterol Hepatol 2004;2(3):262–5.
[48] Diehl AM. Lessons from animal models of NASH. Hepatol Res 2005;33:138–44.
[49] Chaldakov GN, Stankulov IS, Hristova M, et al. Adipobiology of disease: adipokines and
adipokine-targeted pharmacology. Curr Pharm Des 2003;9:1023–31.
[50] Klaus S. Adipose tissue as a regulator of energy balance. Curr Drug Targets 2004;5:241–50.
[51] Rajala MW, Scherer PE. Minireview: the adipocytedat the crossroads of energy homeosta-
sis, inflammation, and atherosclerosis. Endocrinology 2003;144:3765–73.
[52] Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 2004;
89:2548–56.
[53] Arner P. The adipocyte in insulin resistance: key molecules and the impact of the thiazoli-
dinediones. Trends Endocrinol Metab 2003;14:137–45.
[54] Farrell GC, Larter CZ. Nonalcoholic fatty liver disease: from steatosis to cirrhosis. Hepa-
tology 2006;43(2 Suppl 1):S99–112.
[55] Tilg H, Hotamisligil GS. Nonalcoholic fatty liver disease: cytokine-adipokine interplay and
regulation of insulin resistance. Gastroenterology 2006;131:934–45.
[56] Nomura F, Ohnishi K, Suzuki T, et al. Fatty liver in rats induced by excessive intake of a nu-
tritionally adequate liquid diet. Int J Obes 1987;11(6):603–8.
[57] Assy N, Grozovski M, Bersudsky I, et al. Effect of insulin-sensitizing agents in combination
with ezetimibe, and valsartan in rats with non-alcoholic fatty liver disease. World J Gastro-
enterol 2006;12:4369–76.
[58] Schattenberg JM, Wang Y, Singh R, et al. Hepatocyte CYP2E1 overexpression and
steatohepatitis lead to impaired hepatic insulin signaling. J Biol Chem 2005;280:
9887–94.
[59] Alexia C, Fourmatgeat P, Delautier D, et al. Insulin-like growth factor-I stimulates H4II rat
hepatoma cell proliferation: dominant role of PI-30 K/Akt signaling. Exp Cell Res 2006;312:
1142–52.
[60] Florez JC, Jablonski KA, Bayley N, et al. TCF7L2 polymorphisms and progression to di-
abetes in the Diabetes Prevention Program. N Engl J Med 2006;355:241–50.
[61] Pessayre D, Mansouri A, Fromenty B. Nonalcoholic steatosis and steatohepatitis. V. Mi-
tochondrial dysfunction in steatohepatitis. Am J Physiol Gastrointest Liver Physiol 2002;
282:G193–9.
[62] Hui JM, Hodge A, Farrell GC, et al. Beyond insulin resistance in NASH: TNF-alpha or
adiponectin? Hepatology 2004;40:46–54.
[63] Feldstein AE, Werneburg NW, Canbay A, et al. Free fatty acids promote hepatic lipotox-
icity by stimulating TNF-alpha expression via a lysosomal pathway. Hepatology 2004;40:
185–94.
[64] Arkan MC, Hevener AL, Greten FR, et al. IKK-beta links inflammation to obesity-
induced insulin resistance. Nat Med 2005;11:191–8.
[65] Garg A. Lipodystrophies. Am J Med 2000;108:143–52.
[66] Backhed F, Ding H, Wang T, et al. The gut microbiota as an environmental factor that
regulates fat storage. Proc Natl Acad Sci USA 2004;101:15718–23.
[67] Chitturi S, Farrell GC. Etiopathogenesis of nonalcoholic steatohepatitis. Semin Liver Dis
2001;21:27–41.
[68] Green RM. Hepatic metabolism and not simply the metabolic syndrome. Hepatology 2003;
38:14–7.
[69] Koteish A, Diehl AM. Animal models of steatosis. Semin Liver Dis 2001;21:89–104.
[70] Tsukamoto H, Gaal K, French SW. Insights into the pathogenesis of alcoholic liver necro-
sis and fibrosis: status report. Hepatology 1990;12:599–608.
[71] Mottin CC, Moretto M, Padoin AV, et al. Histological behavior of hepatic steatosis in
morbidly obese patients after weight loss induced by bariatric surgery. Obes Surg 2005;
15:788–93.
[72] Bataller R, Brenner DA. Liver fibrosis. J Clin Invest 2005;115:209–18.
[73] Friedman SL. Stellate cells: a moving target in hepatic fibrogenesis. Hepatology 2004;40:
1041–3.
[74] Kamada Y, Tamura S, Kiso S, et al. Enhanced carbon tetrachloride-induced liver fibrosis in
mice lacking adiponectin. Gastroenterology 2003;125:1796–807.
[75] Lijnen HR. Pleiotropic functions of plasminogen activator inhibitor-1. J Thromb Haemost
2005;3:35–45.
[76] Fujikawa K, Ohata K, Honda T, et al. Nonalcoholic steatohepatitis with improved hepatic
fibrosis after weight reduction. Intern Med 2004;43(4):289–94.
[77] Ueno T, Sugawara H, Sujaku K, et al. Therapeutic effects of restricted diet and exercise in
obese patients with fatty liver. J Hepatol 1997;27(1):103–7.
[78] Brolin RE. Bariatric surgery and long-term control of morbid obesity. JAMA 2002;288:
2973–6.
[79] Andersen T, Gluud C, Franzmann MB, et al. Hepatic effects of dietary weight loss in
morbidly obese subjects. J Hepatol 1991;12:224–9.
[80] Goldstein DJ. Beneficial health effects of modest weight loss. Int J Obes Relat Metab
Disord 1992;16:397–415.
[81] O’Leary VB, Marchetti CM, Krishnan RK, et al. Exercise-induced reversal of insulin resis-
tance in obese elderly is associated with reduced visceral fat. J Appl Physiol 2006;100(5):
1584–9.
[82] Anonymous. Executive summary of the clinical guidelines on the identification, evaluation,
and treatment of overweight and obesity in adults. Arch Intern Med 1998;158:1855–67.
[83] Muzio F, Mondazzi L, Sommariva D, et al. Long-term effects of low-calorie diet on the
metabolic syndrome in obese nondiabetic patients. Diabetes Care 2005;28(6):1485–6.
[84] Minehira K, Tappy L. Dietary and lifestyle interventions in the management of the meta-
bolic syndrome: present status and future perspective. Eur J Clin Nutr 2002;56(12):1264–9.
[85] Viviani GL, Carta G, Berri F, et al. Effects of normoglycemia after a low carbohydrate diet
in NIDDM. Insulin secretion and effectiveness. Minerva Endocrinol 1984;9(2):229–32.
[86] Arora SK, McFarlane SI. The case for low carbohydrate diets in diabetes management.
Nutr Metab (Lond) 2005;2:16–25.
[87] Niskanen L, Uusitupa M, Sarlund H, et al. The effects of weight loss on insulin sensitivity,
skeletal muscle composition and capillary density in obese non-diabetic subjects. Int J Obes
Relat Metab Disord 1996;20:154–60.
[88] Kriska A. Can a physically active lifestyle prevent type 2 diabetes? Exerc Sport Sci Rev
2003;31:132–7.
[89] Duncan GE, Perri MG, Theriaque DW, et al. Exercise training, without weight loss,
increases insulin sensitivity and postheparin plasma lipase activity in previously sedentary
adults. Diabetes Care 2003;26:557–62.
[90] Uusitupa M, Lindi V, Louheranta A, et al. Long-term improvement in insulin sensitivity by

changing lifestyles of people with impaired glucose tolerance: 4-year results from the Finn-
ish Diabetes Prevention Study. Diabetes 2003;52:2532–8.
[91] Lean ME. Is long-term weight loss possible? Br J Nutr 2000;83(Suppl 1):S103–11.
[92] Eden KB, Orleans CT, Mulrow CD, et al. Does counseling by clinicians improve physical
activity? A summary of the evidence for the U.S. Preventive Services Task Force. Ann
Intern Med 2002;137:208–15.
[93] Jones NL, Killian KJ. Exercise limitation in health and disease. N Engl J Med 2000;343:
632–41.
[94] van Baak MA. Exercise training and substrate utilization in obesity. Int J Obes Relat Metab
Disord 1999;23:S11–7.
[95] Hoppeler H. Skeletal muscle substrate metabolism. Int J Obes Relat Metab Disord 1999;23:
S7–10.
[96] Palmer M, Schaffner F. Effect of weight reduction on hepatic abnormalities in overweight
patients. Gastroenterology 1990;99:1408–13.
[97] Drenick EJ, Simmons F, Murphy JF. Effect on hepatic morphology of treatment of
obesity by fasting, reducing diets and small bowel bypass. N Engl J Med 1970;282:
829–34.
[98] Vajro P, Fontanella A, Perna C, et al. Persistent hyperaminotransferasemia resolving after
weight reduction in obese children. J Pediatr 1994;125:239–41.
[99] Sjostrom L, Lindroos AK, Peltonen M, et al. Lifestyle, diabetes, and cardiovascular risk
factors 10 years after bariatric surgery. N Engl J Med 2004;351(26):2683–93.
[100] Dixon JB, Bhathal PS, Hughes NR, et al. Nonalcoholic fatty liver disease: improvement in
liver histological analysis with weight loss. Hepatology 2004;39(6):1647–54.
[101] Blackburn GL, Mun EC. Effects of weight loss surgeries on liver disease. Semin Liver Dis
2004;24(4):371–9.
[102] Klein S, Mittendorfer B, Eagon C, et al. Gastric bypass surgery improves metabolic and
hepatic abnormalities associated with nonalcoholic fatty liver disease. Gastroenterology
2006;130:1564–72.
[103] Mathurin P, Gonzalez F, Kerdraon O, et al. The evolution of severe steatosis after bariatric
surgery is related to insulin resistance. Gastroenterology 2006;130(6):1617–24.
[104] Kral JG, THung SN, Biron S, et al. Effects of surgical treatment of the metabolic syndrome
on liver fibrosis and cirrhosis. Surgery 2004;135:48–58.
[105] Lin HZ, Yang SQ, Chuckaree C, et al. Metformin reverses fatty liver disease in obese, lep-
tin-deficient mice. Nat Med 2000;6:998–1003.
[106] Marchesini G, Brizi M, Bianchi G, et al. Metformin in non-alcoholic steatohepatitis. Lan-
cet 2001;358:893–4.
[107] Magalotti D, Marchesini G, Ramilli S, et al. Splanchnic haemodynamics in non-alcoholic
fatty liver disease: effect of a dietary/pharmacological treatment. A pilot study. Dig Liver
Dis 2004;36:406–11.
[108] Schwimmer JB, Middleton MS, Deutsch R, et al. A phase 2 clinical trial of metformin as
a treatment for non-diabetic paediatric nonalcoholic steatohepatitis. Aliment Pharmacol
Ther 2005;21:871–9.
[109] Uygun A, Kadayifci A, Isik AT, et al. Metformin in the treatment of patients with nonal-
coholic steatohepatitis. Aliment Pharmacol Ther 2004;19:537–44.
[110] Nair S, Diehl AM, Wiseman M, et al. Metformin in the treatment of non-alcoholic steato-
hepatitis: a pilot open label trial. Aliment Pharmacol Ther 2004;20(1):23–8.
[111] Bugianesi E, Gentilcore E, Manini R, et al. A randomized controlled trial of metformin ver-
sus vitamin E or prescriptive diet in nonalcoholic fatty liver disease. Am J Gastroenterol
2005;100(5):1082–90.
[112] Caldwell SH, Hespenheide EE, Redick JA, et al. A pilot study of a thiazolidinedione, tro-
glitazone, in nonalcoholic steatohepatitis. Am J Gastroenterol 2001;96(2):519–25.
[113] Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, et al. Improved nonalcoholic steato-
hepatitis after 48 weeks of treatment with the PPAR-gamma ligand rosiglitazone. Hepatol-
ogy 2003;38(4):1008–17.
[114] Promrat K, Lutchman G, Uwaifo GI, et al. A pilot study of pioglitazone treatment for non-
alcoholic steatohepatitis. Hepatology 2004;39(1):188–96.
[115] Sanyal AJ, Mofrad PS, Contos MJ, et al. A pilot study of vitamin E versus vitamin E and
pioglitazone for the treatment of nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol
2004;2(12):1107–15.
[116] Belfort R, Harrison SA, Brown K, et al. A placebo-controlled trial of pioglitazone in sub-
jects with nonalcoholic steatohepatitis. N Engl J Med 2006;355(22):2297–307.
[117] Basaranoglu M, Acbay O, Sonsuz A. A controlled trial of gemfibrozil in the treatment of
patients with nonalcoholic steatohepatitis. J Hepatol 1999;31(2):384.
[118] Gomez-Dominguez E, Gisbert JP, Moreno-Monteagudo JA, et al. A pilot study of atorvas-
tatin treatment in dyslipemid, non-alcoholic fatty liver patients. Aliment Pharmacol Ther
2006;23(11):1643–7.
[119] Kiyici M, Gulten M, Gurel S, et al. Ursodeoxycholic acid and atorvastatin in the treatment
of nonalcoholic steatohepatitis. Can J Gastroenterol 2003;17(12):713–8.
[120] Hatzitolios A, Savopoulos C, Lazaraki G, et al. Efficacy of omega-3 fatty acids, atorvasta-
tin, and orlistat in non-alcoholic fatty liver disease with dyslipidemia. Indian J Gastroen-
terol 2004;23:131–4.
[121] Rallidis LS, Drakoulis CK, Parasi AS. Provastatin in patients with nonalcoholic steatohe-
patitis: results of a pilot study. Atherosclerosis 2004;174:193–6.
[122] Nair S, Wiseman M. HMG-CoA reductase inhibitors in nonalcoholic fatty liver disease: is
potential hepatotoxicity an issue in these patients? A case-control study based on histology
[Abstract]. Hepatology 2002;36:409A.
[123] Chalasani N, Aljadhey H, Kesterson J, et al. Patients with elevated liver enzymes are not at
higher risk for statin hepatotoxicity. Gastroenterology 2004;126(5):1287–92.
[124] Laurin J, Lindor KD, Crippin JS, et al. Ursodeoxycholic acid or clofibrate in the treatment
of nonalcohol induced steatohepatitis: a pilot study. Hepatology 1996;23:1464–7.
[125] Bajaj M, Suramornkul S, Pratipanowatr T, et al. Pioglitazone reduces hepatic fat content
and augments splanchnic glucose uptake in patients with type 2 diabetes. Diabetes 2003;
52:1364–70.
Med Clin N Am 91 (2007) 1151–1168
Obesity and the Polycystic Ovary

Syndrome
Michael Magnotti, MD, Walter Futterweit, MD*
Mount Sinai School of Medicine, New York, NY 10029, USA
Polycystic ovary syndrome (PCOS) is the most common metabolic ab-

normality in reproductive-aged women, occurring globally in 6% to 10%
of this population [1–3]. Identification of such women is most important
in that they constitute the largest group of women at risk for the develop-
ment of cardiovascular disease (CVD) and type 2 diabetes mellitus
(T2DM). It is estimated that more than 50% of these women are undiag-
nosed. The fact that it likely affects 4 to 5 million women in the United
States, with distinct metabolic and cardiovascular risk factors and associ-
ated features of metabolic syndrome (MS) (also known as ‘‘insulin resis-
tance syndrome’’), makes diagnosis and treatment at an early stage
imperative for proactive and aggressive management. The past decade has
demonstrated that some of the features of MS, namely, insulin resistance
and compensatory hyperinsulinism, are frequently present in women with
PCOS. The phenotype associated with this syndrome is not restricted to
bothersome cosmetic symptoms in reproductive-aged women, but includes
a cluster of findings present in these women that may have profound health
implications later in their life.
Recognition of the syndrome affords the physician not only an opportu-
nity to reduce the emotional impact of skin manifestations, such as acne,
hirsutism, androgenic alopecia, and infertility, but also to attenuate the
likely later metabolic and CVD complications. The cardinal features of
the syndrome are the presence of hyperandrogenism and hyperinsulinemia,
which often vary over time [4]. Moreover, obesity is often associated with
PCOS [4]. Surprisingly few studies have addressed the frequency of PCOS
in unselected overweight or obese women. In one such study of 113 over-
weight and obese premenopausal women from Spain, however, the
E-mail address: wfutt@ix.netcom.com (W. Futterweit).
1152 MAGNOTTI & FUTTERWEIT
prevalence of PCOS was 28%, compared with the 5.5% prevalence of PCOS
in Spanish lean women [5].
The presence of obesity further contributes to the metabolic derange-
ments of the syndrome. It is well established that obesity is commonly asso-
ciated with insulin resistance. The obese woman with PCOS exhibits not
only the intrinsic insulin resistance of the syndrome but also that related
to the presence of excess adiposity. This is emphasized by the beneficial ef-
fects of a hypocaloric diet [6], exercise [7], and insulin-sensitizing agents [8,9]
on metabolic and reproductive features of the syndrome.
The occurrence of obesity in conjunction with the skin and hair manifes-
tations of the syndrome can have a further negative effect on self-esteem and
self-image [10]. The fear of social rejection makes some women reclusive and
may retard their development of social skills and confidence. Such women
are usually helped considerably once an accurate diagnosis and treatment
of the underlying endocrine and ovarian disorders is obtained. Modification
of lifestyle and correction of the underlying pathophysiologic condition can
help ameliorate psychologic dysfunction.
Diagnostic criteria for polycystic ovary syndrome

Defining the syndrome (phenotype) has been a matter of debate. Until
several years ago, most investigators used an adopted consensus of PCOS
experts at the 1990 National Institute of Child Health and Human Develop-
ment conference [11], which consisted of (1) the presence of oligoamenor-
rhea; (2) clinical and biochemical features of hyperandrogenism; and (3)
the exclusion of other disease states that may mimic PCOS, such as nonclas-
sical congenital adrenal hyperplasia, virilizing syndromes of the ovary or ad-
renal, and Cushing’s syndrome. In 2003, a newer definition of the
syndrome’s phenotype was adopted at the Rotterdam PCOS Consensus
Group Conference [12] to allow for the inclusion of women with regular
menses (present in about 25% of women with PCOS) who manifest the
morphologic changes of polycystic ovaries on ultrasonography [12]. The lat-
ter was not accepted as a diagnostic criterion in the aforementioned 1990
conference. Finally, a recent Phenotype Task Force of the Androgen Excess
Society published their Guidelines stressing the importance of hyperandro-
genism as the cardinal feature of PCOS [13], while accepting the major
guidelines of the 1990 Conference consensus report.
There is a subset of women with PCOS who have regular cycles. This may
vary from 25% to 40% of patients with the syndrome and is mainly com-
prised of women with generally regular menses associated with episodes
of irregularity (oligo-ovulation) [14,15]. In addition, a history of regular
or almost regular cycles does not prove that ovulation has occurred. Docu-
mented ovulation in a menstrual cycle requires biochemical confirmation of
elevated serum progesterone in the luteal phase.
OBESITY AND THE POLYCYSTIC OVARY SYNDROME 1153
Pathogenesis of polycystic ovary syndrome

An understanding of the pathogenesis of PCOS substantiates the concept
that it seems to be a multigenic disorder. Clinical studies support an in-
creased frequency of PCOS in first-degree relatives of affected women as
suggested by a higher prevalence of glucose intolerance, insulin resistance,
and hyperandrogenemia [16–18]. Many candidate genes for PCOS have
been proposed. A number of these genes may alter the hypothalamic-pitui-
tary-ovarian axis and others may play a major role in the pathogenesis of
insulin resistance, and the associated phenotype of this heterogeneous disor-
der [19]. Recent studies have demonstrated a candidate gene near the insulin
receptor gene locus [20,21]. In addition, several alternative hypotheses have
been proposed for the pathogenesis of PCOS [22], including
1. Hypothalamic-pituitary abnormalities that result in gonadotropin-
releasing hormone and luteinizing hormone dysfunction.
2. A primary enzymatic defect in ovarian or combined ovarian and adrenal
steroidogenesis.
3. A metabolic disorder characterized by insulin resistance in conjunction
with compensatory hyperinsulinemia that exerts adverse effects on the
hypothalamus, pituitary, ovaries, and adrenal glands.
Clinical features of polycystic ovary syndrome

The manifestations of PCOS are quite varied, and depend on ethnicity,
environmental factors, and the presence and severity of obesity and insulin
resistance. The most common clinical features are listed in Box 1. There are
no pathognomonic characteristics of the syndrome, but the combination of
oligoamenorrhea and clinical or biochemical findings of androgen excess are
Box 1. Clinical manifestations of PCOS

Menstrual dysfunction or infertility
Hirsutism
Acne
Alopecia
Visceral (android, central) fat distribution and obesity
Miscellaneous
Sleep apnea
Acanthosis nigricans and skin tags
Data from Futterweit W, Diamanti-Kandarakis E, Azziz R. Clinical features of the

polycystic ovary syndrome. In: Azziz R, Nestler JE, Dewailly D, editors. Androgen
excess disorders in women: polycystic ovary syndrome and other disorders. 2nd
edition. Totowa (NJ): Humana Press; 2006. p. 155–67.
most commonly noted. The presence of polycystic ovaries as defined on ul-

trasonography is frequently noted, but this alone does not establish the di-
agnosis [23,24]. Weight gain and an inability to lose weight are common
major complaints in women with PCOS. The intensity varies, but a persistent
weight gain of 10 to 15 pounds per year may be noted. As women gain
weight the clinical features of hirsutism, acne, menstrual cyclicity, and ob-
structive sleep apnea frequently worsen. Associated with the weight gain
is increased appetite with a tendency to compulsive ingestion of carbohy-
drate-rich foods. Episodes of postprandial hyperinsulinemia may occur in
mid-afternoon and be associated with symptoms of hypoglycemia.
Because this article reviews the impact, pathophysiology, and associated
risks of obesity and MS in PCOS, the reader may wish to review articles on
other aspects of clinical findings, diagnosis, risks for later complications, ul-
trasonographic morphology of ovaries, and management of women with
PCOS [1,4,14,24–28].
Weight gain and obesity in polycystic ovary syndrome

Weight gain in PCOS often intensifies peripubertally. It may well be an
inciting factor for the development of menstrual dysfunction; dysfunctional
bleeding; and signs of hyperandrogenism, such as hirsutism, acne, and alo-
pecia. The body mass index (BMI) (weight in kilometers per meters squared)
may serve as a modest predictor of oligomenorrhea and associated features
of PCOS [29]. In a study by van Hooff and colleagues [29], both obese and
lean adolescent girls with PCOS were at high risk for developing persistent
oligomenorrhea. In another study of 230 girls aged 16 to 25 years there was
a significantly increased prevalence of elevated BMI in women with PCOS
than in matched controls [24]. Obesity (BMI O30 kg/m2) or being over-
weight (BMI O25 kg/m2) is frequently present in women with PCOS with
a reported wide range of 38% to 88% [27,30]. Sixty percent of women
with PCOS in the United States are obese [14], but the frequency seems to
be lower in other populations [31]. Differences in the prevalence of obesity
vary among different ethnic groups, suggesting that both environmental
and genetic factors are of importance in determining individual
susceptibility.
As in normal women, the development of obesity in PCOS is associated
with increased insulin resistance. In addition, most of the PCOS population
has a greater degree of insulin resistance than age- and BMI-matched con-
trols [32]. This disparity is amplified in those women with PCOS who are
also obese, leading to a higher degree of insulin resistance than either con-
dition alone, whereas lean PCOS subjects tend to have a lesser degree of
insulin resistance [32–34].
Many women with PCOS may relate a rather sudden onset of weight gain
over 6 to 18 months, which further induces symptoms of menstrual
dysfunction with infertility, acne, and hirsutism. The weight gain is usually
associated with an increase in carbohydrate craving and postprandial symp-
toms of hypoglycemia. The latter include lack of concentration, hunger,
sweats, headaches, tremulousness, poor concentration, irritability, and
sleepiness caused by an exaggerated response of insulin following a carbohy-
drate-rich meal. A central (abdominal, visceral) distribution of fat (Fig. 1) is
often present in both obese and lean women with PCOS. Obesity may be
a major factor that exacerbates the symptoms and signs of PCOS rendering
them more susceptible to developing T2DM, cardiovascular events, and re-
productive abnormalities. The evidence that PCOS is associated with in-
creased risk for CVD is steadily mounting.
Obesity also seems to play a pathogenetic role in the development of
PCOS [35]. Obesity is frequently associated with insulin resistance, which
acts synergistically with luteinizing hormone to intensify ovarian hyperan-
drogenism, decrease hepatic production of sex hormone-binding globulin
(SHBG), and reduce insulin-like growth factor binding protein-1 [36]. More-
over, obesity reinforces the genetic predisposition to hormonal anovulatory
disorders in PCOS. Metabolic complications including impaired glucose tol-
erance (IGT) leading to T2DM (in approximately 40% of subjects with
obesity), hypertension, dyslipidemia, and possible development of estro-
gen-dependent tumors (eg, endometrial carcinoma) are more common in
obese patients with PCOS than in lean women with PCOS. Weight loss in
Fig. 1. Central fat distribution in a 31-year-old woman with polycystic ovary syndrome.
patients with hyperandrogenism, with or without the clinical presence of

PCOS, should be the first therapeutic option because it decreases androgen
levels, increases SHBG, and may restore ovulation. The increase in insulin
sensitivity from as little as a 7% reduction in body weight can restore fertil-
ity, and decrease hirsutism in some women with androgen excess [6,37,38].
Investigators have also demonstrated that abnormalities in 17,20-lyase ac-
tivity diminish in parallel to the reduction of hyperinsulinemia in women
with PCOS [39].
Visceral (central, android) fat distribution and obesity

A central (abdominal, visceral, android) distribution of fat is often pres-
ent in both obese and lean women with PCOS (see Fig. 1). Abdominal obe-
sity may be clinically defined as a waist circumference of more than 88 cm
(O35 in) in women, and more than 102 cm (O40 in) in men. In Asian
women a waist circumference of more than 30 in may indicate increased vis-
ceral fat distribution. A major initial pathogenetic factor in the develop-
ment of MS (see later) seems to be visceral adiposity leading to insulin
resistance. The link to insulin resistance may partially be explained by the
frequently associated increased levels of free fatty acids that can inhibit
the insulin signaling mechanism, leading to decreased glucose transport to
skeletal muscle [40]. It is important to note that visceral adiposity does
not necessarily correlate with the BMI. Visceral fat in nonobese women
with PCOS contributes significantly to high serum triglyceride and fasting
insulin levels [41]. As visceral adiposity increases, the insulin sensitivity de-
clines [42,43].
The metabolic and cardiovascular risks in PCOS seem to correlate with
abdominal obesity. Altered parameters of inflammation with insulin resis-
tance and visceral obesity are often present in PCOS [44].
Acanthosis nigricans and skin tags

Acanthosis nigricans with or without skin tags are cutaneous markers of
hyperinsulinism [45,46]. In the absence of a paraneoplastic syndrome, acan-
thosis nigricans is an epiphenomenon of dermal hyperplasia, which may be
seen on the nape of the neck, axillae, groin, inner lips of vulvae, periumbil-
ical, and inframammary areas, and on the dorsum of fingers and knuckles.
They appear as brownish gray, velvety, or verrucous hyperparakeratotic
pigmented areas, more commonly seen in obese women of Hispanic or Af-
rican-American descent (Figs. 2 and 3), although they may also be noted oc-
casionally in obese and lean (BMI !25 kg/m2) white populations. Skin tags
are infrequent in most women before the age of 40 years and, if present be-
fore that age, suggest the presence of hyperinsulinism.
Fig. 2. Acanthosis nigricans and hirsutism in a 29-year-old patient with polycystic ovary
syndrome.
Polycystic ovary syndrome and metabolic syndrome

MS is defined as a cluster of abnormalities that are more likely to occur in
patients with insulin resistance, particularly those who are obese. It identifies
patients at increased risk for T2DM and CVD. The prevalence of insulin re-
sistance and compensatory hyperinsulinemia is increased in women with
PCOS. Data vary, but one report indicates that 75% of women with
PCOS fulfill the criteria for MS [47]. Recently, a report noted that adoles-
cents with PCOS have an increased risk of MS with hyperandrogenemia
that is independent of obesity or insulin resistance [48]. Not all patients
with insulin resistance have any or all of the features present in MS. Fur-
thermore, clusters of the latter can develop in the absence of insulin resis-
tance. Nevertheless, the likelihood that a woman with PCOS has insulin
resistance leading to hyperinsulinemia is so substantial that all women
with PCOS should undergo diagnostic testing for the components of MS.
The prevalence of MS in women with PCOS under the age of 20 has been
reported as high as 23% [49]. It is not surprising that with advancing age
(30–39-year-old group) this same study reported that the rate increased to
53% [49]. This suggests that diligent follow-up of women with PCOS at
all ages should be maintained for the cluster of features comprising MS [50].
Both obesity and physical inactivity are major factors that work synergis-
tically with the inherent postreceptor defect and lead to insulin resistance in
women with PCOS [32]. Obesity increases the likelihood of occurrence of in-
sulin resistance. Although not all obese subjects are insulin resistant, obesity
is frequently associated. Most obese and nonobese oligomenorrheic women
with PCOS have insulin resistance [32,51].
Fig. 3. Acanthosis nigricans and skin tags in a 37-year-old Hispanic woman with polycystic
ovary syndrome.
Insulin resistance is difficult to assess accurately in a clinical setting with-

out the gold-standard of hyperinsulinemic-euglycemic clamp studies. This
particularly applies in the setting of basal insulin levels in the normal or
high-normal range and in the nonobese subgroup of women with PCOS.
Methods that measure only fasting glucose and insulin levels (including ho-
meostasis model assessment and the quantitative insulin sensitivity check in-
dex) may be inaccurate in assessing insulin sensitivity in these women [52]. It
is not unlikely that data may underestimate the frequency of insulin resis-
tance in PCOS subjects with borderline normal fasting glucose and insulin
levels. Perhaps other hormonal and genetic factors, and ethnicity, may influ-
ence the degree of insulin resistance and contribute to conflicting reports of
varying prevalence of insulin resistance in women with PCOS. Because ac-
curate assessment of insulin sensitivity is difficult in a clinical practice set-
ting, it is prudent to regard all obese women with PCOS as likely to have
insulin resistance and at risk for MS, and assume that most nonobese
women with PCOS have some if not most of the features of MS [51].
Polycystic ovary syndrome and the risk for type 2 diabetes mellitus
Epidemiologic studies indicate a high rate of IGT of 31% to 35% and
a 7.5% rate of T2DM, at the initial evaluation of young women with
PCOS. These studies comprised primarily obese and overweight subjects

with PCOS [53,54]. Comparative rates of IGT and T2DM in the 20- to
44-year-old United States female population have been reported as 7.8%
for IGT and 1% for undiagnosed diabetes [54]. A large population study
showed that normal fasting blood glucose with IGT identifies subjects
with an adverse CVD risk factor and possible risk of sudden death [55].
Other risk factors in large population studies include increasing age, BMI,
and visceral body fat distribution affecting conversion to IGT [55]. Data re-
lating to conversion of IGT to T2DM (5–10 fold) and those with normogly-
cemia and normal glucose tolerance to IGT and T2DM vary in different
PCOS populations [53,54,56] and the presence of first-degree relatives
with T2DM increases the risk. The need for retesting at frequent intervals
is necessary because the fasting blood glucose and glycohemoglobin levels
usually are normal even in IGT and are a poor guide to indicating conver-
sion of normal glucose tolerance to IGT, and T2DM in PCOS [55,57]. In
a recent controlled study of changes in glucose tolerance in a 3-year time
span involving 71women with PCOS (mean age, 28 years; mean BMI, 27
kg/m2) and normal controls, the conversion rate of those with IGT to
T2DM was 2.2% per year [58]. The data demonstrated, however, a pro-
nounced conversion rate of 16% per year to IGT in the women with
PCOS and normal glucose tolerance. This report is likely to be followed
by further prospective studies of other PCOS women with normal glucose
intolerance and their conversion rate to IGT and T2DM.
In a study of women attending a diabetic clinic in an academic center,
27% of the premenopausal women had PCOS [59]. An 82% prevalence of
ultrasonographic morphology of premenopausal women at a T2DM clinic
in the United Kingdom has also been recorded [60].
Adipokines, obesity, and metabolic syndrome

In recent years, research has demonstrated that adipose tissue secretes
a variety of endocrinologically active substances, known collectively as ‘‘adi-
pokines.’’ These substances have a wide range of effects that are slowly being
elucidated. They have been closely linked to such processes as regulation of
appetite, insulin resistance, and immune modulation, and may play an
important role in the pathogenesis of various disease states.
The first adipokine to be discovered was leptin, the product of the ob
gene. It was noted that mice deficient in this protein became morbidly obese
and developed severe insulin resistance and type 2 diabetes. In humans, lep-
tin is produced largely by adipocytes, but also in smaller quantities by a va-
riety of other tissues. Leptin is involved in the regulation of appetite and
body weight, acting as a signal to the hypothalamus to decrease appetite
and increase energy expenditure. Levels of leptin are proportional to total
body fat content and are higher in females than males [61,62]. Because of
this, any gain in body fat causes a rise in leptin levels, signaling the hypo-
thalamus to decrease food intake and increase energy expenditure, thereby
promoting weight loss. If intact, this mechanism should ensure the mainte-
nance of normal body weight, but given the current obesity epidemic, there
must be more to the story. What happens to leptin in obesity?
As expected, leptin levels are generally very high in obese patients. De-
spite these elevated levels, however, obese patients do not increase energy
expenditure or decrease appetite in response. They seem to be resistant to
the actions of leptin on the hypothalamus, but the mechanism of this ‘‘leptin
resistance’’ remains unclear. Attempts at using leptin as a treatment for obe-
sity have also been largely unsuccessful; a randomized controlled trial of this
demonstrated efficacy in only a small subset of obese patients [63].
Adiponectin is an adipokine that seems to be closely related to a number
of the components of MS. It is produced exclusively by adipocytes, in
greater quantities by subcutaneous than visceral adipose cells [64]. As
with leptin, levels are higher in females than males, but unlike leptin, levels
of adiponectin are generally inversely related to the degree of obesity. Over-
all, adiponectin is a beneficial hormone that acts to decrease insulin resis-
tance and inhibit atherogenesis [65]. Its concentrations are decreased in
patients with obesity; states of insulin resistance (eg, PCOS); T2DM; and
macrovascular disease. In addition, adiponectin levels correlate with insulin
sensitivity (by glucose disposal rate) [64], and low levels may be predictive of
future diabetes [66]. Adiponectin-deficient knockout mice fed a fat- and car-
bohydrate-rich diet develop insulin resistance, hyperglycemia, and have
a higher degree of atherogenesis than normal controls [64]. These effects
can be reversed by administration of adiponectin. A small number of hu-
mans with genetic adiponectin deficiencies have also been found and they
exhibit many of the components of MS [64].
In mice, resistin is an adipokine that is produced almost exclusively by
adipose cells. There is evidence that increased levels of resistin are associated
with insulin-resistant states (in mice) [67] and it seems likely that resistin is
involved in the pathogenesis of insulin resistance in these animals. In addi-
tion, there is considerable evidence that administration of exogenous resistin
can cause insulin resistance in mice [65]. Despite this, resistin does not seem
to be associated with states of obesity in these animals [67]. These findings
initially generated a great deal of excitement, but it turns out that in humans
most studies demonstrate that resistin is not produced by mature adipose
cells and has a relatively minor role in the pathogenesis of insulin resistance
[65]. Its main action seems to be as a proinflammatory agent.
Role of adipokines in polycystic ovary syndrome

PCOS confers insulin resistance independently of BMI. Several studies
have looked at adipokine levels in patients with PCOS to determine whether
they may be involved in its pathogenesis. A case control study of 64 women

with PCOS demonstrated that these women had significantly lower adipo-
nectin levels than age- and BMI-matched nonhyperandrogenic controls (in-
dependently of BMI) [68]. Using a multiple regression analysis, the authors
demonstrated that free testosterone levels, waist-to-hip ratio, and age were
the major determinants of adiponectin concentrations in these women.
The finding that adiponectin levels are lower in patients with PCOS than
age- and BMI-matched controls was also confirmed in several other studies
[69,70]. In addition, adiponectin levels show a strong inverse correlation
with central adiposity [69]. Interestingly, several studies have also demon-
strated a positive correlation between free testosterone and adiponectin
levels [68,69]. This remains somewhat perplexing given the higher levels of
adiponectin found in females than males (despite lower free testosterone
levels in females) and the inverse relationship between PCOS and adiponec-
tin (despite higher androgen levels in these patients). The impact of this re-
lationship is likely to be minimal in patients with PCOS, however, because it
is overshadowed by the inverse correlation between adiponectin and both
central obesity and PCOS itself.
Despite the inverse correlation seen between insulin resistance and adipo-
nectin levels in control patients, most studies have not been able to demon-
strate a similar correlation in patients with PCOS. This may simply reflect
an inability accurately to measure insulin resistance, or it may suggest
that adiponectin is not involved in the pathogenesis of insulin resistance
in patients with PCOS. Conversely, it may indicate the involvement of mul-
tiple pathways in the pathogenesis of insulin resistance in the syndrome.
Levels of leptin correlate well with the amount of body fat in both pa-
tients with PCOS and controls. There is no significant difference in leptin
concentrations between these two groups after controlling for BMI
[69,70]. It seems unlikely that leptin has a significant role in the pathogenesis
of PCOS. Resistin levels were found to be higher, however, in patients with
PCOS than age- and BMI-matched controls [70]. Given the poor relation
between resistin and insulin resistance in humans, however, it seems unlikely
that this is responsible for the increased insulin resistance in these patients.
Nevertheless, it may play a contributory role in the proinflammatory state
associated with MS.
Ghrelin is another adipokine that is involved in the regulation of food in-
take and body weight. It is the endogenous ligand of the growth hormone
releasing factor receptor in the central nervous system, but its receptors
are also present in the periphery, including on the ovaries [71]. As with adi-
ponectin, its levels are reduced in patients with PCOS (independent of body
weight) and obesity [69]. There is conflicting evidence on ghrelin’s involve-
ment in insulin resistance and the pathogenesis of PCOS itself.
It is clear that adiponectin and resistin levels are altered in patients with
PCOS, but it is unclear whether they are involved in the pathophysiology of
the syndrome. Of all the adipokines presently known, adiponectin is most
heavily correlated with PCOS (independently of BMI) and preliminary evi-

dence suggests that it may be involved in its pathogenesis. Resistin is also
positively correlated with PCOS, but its role in humans needs to be further
defined. Finally, leptin correlates well only with BMI. Given the relatively
recent recognition of the endocrine functions of adipose tissue and ghrelin,
it is likely that time will bring the discovery of new adipokines that may help
to explain better the complex relationships between obesity, PCOS, and MS.
Treatment
The treatment of PCOS must be individualized, taking into account the
patient’s major manifestations of the syndrome and her treatment goals. Be-
cause a thorough discussion of the treatment of PCOS requires its own
article, this section briefly explores the major treatment options available.
Table 1 lists the most commonly used agents in the treatment of PCOS.
In the case of the obese patient with PCOS, the first step in treatment
should be an attempt at weight loss. As little as a 7% weight loss can lead
to a significant reduction in hyperandrogenism and a restoration of fertility
Table 1
Major agents used in the treatment of the PCOS
Drug Name Common Brands Effective On Major Side Effects
Biguanides Glucophage Insulin resistance Gastrointestinal upset
Metformin Glumetza Metabolic syndrome (can be minimized
Riomet Ovulation induction by titrating dose
Fortamet Hyperandrogenic over a few weeks)
symptoms
Amenorrhea
Thiazolidenediones Avandia Insulin resistance Weight gain
Rosiglitazone Actos Metabolic syndrome Edema
Pioglitazone Ovulation induction Congestive heart
Hyperandrogenic failure (in
symptoms susceptible
Amenorrhea patients)
Oral contraceptive Yasmin/Yaz Hyperandrogenic Thromboembolism
agents Ortho-Cyclen/ symptoms (especially in
Ethinyl estradiol Tri-Cyclen Amenorrhea smokers O35)
plus drospirenone Desogen/ Hyperkalemia
Ethinyl estradiol Mircette (with drospirenone)
plus norgestimate
Ethinyl estradiol
plus desogestrel
Antiandrogens Aldactone Hyperandrogenic Hyperkalemia
Spironolactone symptoms Teratogenic
5a-Reductase Proscar Hyperandrogenic Teratogenic
inhibitors symptoms
Finasteride
[6,37]. A similar degree of weight loss can also bring about an improvement
in insulin resistance, lipid profiles, and T2DM. Although it is often difficult
to achieve and maintain, weight loss can be extremely gratifying for the pa-
tient, improving self-image and relieving many (or all) of the manifestations
of the syndrome.
An insulin-sensitizing agent, such as the biguanide metformin, can target
many of the problems seen in the obese PCOS patient. Metformin’s effects
are mainly on the liver, lowering insulin resistance and decreasing gluconeo-
genesis. It has obvious beneficial effects on the IGT or overt T2DM that is
often seen in obese patients with the syndrome. In addition, the resulting
increase in insulin sensitivity triggers a decrease in circulating insulin levels,
leading to a decrease in ovarian androgen production and an increase in he-
patic SHBG production. This combination leads to lower free androgen
levels, and can result in the resumption of spontaneous ovulation and regu-
lar menses. Because of this, metformin (with or without clomiphene citrate)
is often used to induce ovulation in women with PCOS who desire fertility
[72].
Although it remains controversial, a recent literature review of six pro-
spective trials demonstrated that adding metformin to patients who are re-
sistant to clomiphene citrate significantly increases the odds of ovulation
[73]. Because the proposed target of metformin is insulin resistance, it seems
logical that it should be most effective in patients with the highest degree of
resistance. A recent review of the evidence, however, showed that it is diffi-
cult to use measures of insulin resistance to predict which women will
respond to metformin [74], although this may be related to the present
lack of a reliable, noninvasive measure of insulin resistance.
Furthermore, women with PCOS have a significantly higher rate of mis-
carriage than normal controls [75]. There is evidence that metformin may
help to reduce the spontaneous abortion rate in these women if continued
through the first trimester or even the entire pregnancy [76]. Although this
approach is often used in practice, the safety of metformin in pregnancy
has not been validated with large prospective trials, so its routine use
throughout pregnancy cannot be recommended at this time. Because met-
formin targets one of the key pathologies present in PCOS (insulin resis-
tance), its use should be considered in most women with the syndrome
(especially those who also have MS).
In patients with PCOS who do not desire pregnancy, it is important to
remember that some form of birth control should be part of the treatment
regimen. This is because treatments, such as weight loss or insulin sensi-
tizers, can often result in resumption of fertility and unwanted pregnancies.
The combination oral contraceptive pill (OCP) is often used in this situation
both to prevent unwanted pregnancies and improve the symptoms of the
disorder. The exogenous estrogen in the OCP feeds back on the pituitary,
leading to decreased luteinizing hormone production and decreased ovarian
androgen production. In addition, estrogen causes increased SHBG
production by the liver, leading to further decreases in free androgen levels.

The progestin component is important to protect the endometrium from the
effects of unopposed estrogen. Because many of the synthetic progestins
used in OCPs can have proandrogenic effects, however, one must take
care in selecting an appropriate OCP for the patient with PCOS. The pro-
gestin drospirenone has some antiandrogen effects and should theoretically
be the ideal progestin for a mild to moderately hyperandrogenic patient.
Preliminary data suggest efficacy in reducing androgen levels and hirsutism
in many PCOS patients [77,78]. Parenthetically, any form of estrogen may
further increase insulin resistance and increase the risk of thromboembolism
and should be used only when deemed necessary [79].
Antiandrogens can also be effective in reducing many of the cosmetic ef-
fects of hyperandrogenism. The potassium-sparing diuretic spironolactone
is the most commonly used antiandrogen in the United States, although cy-
proterone acetate is also available in many other countries. It is important to
note that spironolactone is teratogenic and must be used with an effective
form of contraception, even in patients with reduced fertility. Because of
this, it is commonly given together with an OCP containing a progestin,
such as drospirenone, to combine the antiandrogen effects of both agents
and provide effective birth control. Careful monitoring of potassium levels
is especially important with this combination, although it is generally well
tolerated. The 5a-reductase inhibitor finasteride may also be useful in the
treatment of hirsutism, although it is often less effective than other antian-
drogens [80]. It too must be given together with an effective form of contra-
ception because of teratogenicity.
Summary
PCOS is extremely common among reproductive-aged women, but often
goes undiagnosed. Although considerable controversy exists as to the pre-
cise criteria for defining the syndrome, it is generally agreed on that hyper-
androgenism is a cardinal feature and that polycystic ovarian morphology
on ultrasonography is not required for the diagnosis. PCOS is often associ-
ated with obesity and it imparts a degree of insulin resistance in excess of
that conferred by obesity alone. It is associated with MS and carries a greatly
increased risk of IGT and T2DM and cardiovascular risks. Treatment of
PCOS may provide relief of cosmetic problems and depression by improving
patient self-esteem. In addition, because of its association with MS, T2DM,
and CVD, its recognition and treatment can potentially be life saving.
A number of treatment options exist, but an attempt at weight loss
should be the first step in the obese patient. Insulin sensitizers, such as met-
formin, or a thiazolidinedione, such as pioglitazone or rosiglitazone, can di-
rectly target the insulin resistance associated with the syndrome and may
lead to a restoration of fertility. OCPs and antiandrogens are useful in
women who do not desire pregnancy and have a preponderance of hyperan-

drogenic symptoms.
References
[1] Azziz R, Woods KS, Reyna R, et al. The prevalence and features of the polycystic ovary syn-
drome in an unselected population. J Clin Endocrinol Metab 2004;89:2745–9.
[2] Hart R, Hickey M, Franks S. Definitions, prevalence and symptoms of polycystic ovaries
and polycystic ovary syndrome. Best Pract Res Clin Obstet Gynaecol 2004;18:671–83.
[3] Asuncion M, Calvo RM, San Millan JL, et al. A prospective study of the prevalence of the
polycystic ovary syndrome in unselected caucasian women from Spain. J Clin Endocrinol
Metab 2000;85:2434–8.
[4] Franks S. Polycystic ovary syndrome. N Engl J Med 1995;333:853–61.
[5] Alvarez-Blasco F, Botella-Carretero JI, San Milan JL, et al. Prevalence and characteristics
of the polycystic ovary syndrome in overweight and obese women. Arch Intern Med 2006;
166:2081–6.
[6] Pasquali R, Antenucci D, Casimirri F, et al. Clinical and hormonal characteristics of obese
amenorrheic hyperandrogenic women before and after weight loss. J Clin Endocrinol Metab
1989;68:173–9.
[7] Moran LJ, Brinkworth G, Noakes M, et al. Effects of lifestyle modification in polycystic
ovarian syndrome. Reprod Biomed Online 2006;12:569–78.
[8] Moghetti F, Castello R, Negri C, et al. Metformin effects on clinical features, endocrine and
metabolic profiles, and insulin sensitivity in polycystic ovary syndrome: a randomized, dou-
ble-blind, placebo-controlled 6-month trial, followed by open, long-term clinical evaluation.
J Clin Endocrinol Metab 2000;85:139–46.
[9] Baillargeon JP, Iuorno MJ, Nestler JE. Insulin sensitizers for polycystic ovary syndrome.
Clin Obstet Gynecol 2003;46:325–40.
[10] Kitzinger K, Willmott J. ‘‘The thief of womanhood’’: women’s experience of polycystic ovar-
ian syndrome. Soc Sci Med 2002;54:349–61.
[11] Zawadsky JK, Dunaif A, et al. Polycystic ovary syndrome. In: Dunaif A, Givens JR, Hasel-
tine FP, editors. Diagnostic criteria for polycystic ovary syndrome: towards a rational
approach. Boston: Blackwell Scientific Publications; 1992. p. 377–84.
[12] Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003
consensus on diagnostic criteria and long-term health risks related to polycystic ovary syn-
drome. Fertil Steril 2004;81:19–25.
[13] Azziz R, Carmina E, Dewailly D, et al. Position statement: criteria for defining polycystic
ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Soci-
ety guideline. J Clin Endocrinol Metab 2006;91:4237–45.
[14] Azziz R, Sanchez LA, Knochenhauer ES, et al. Androgen excess in women: experience with
over 1000 consecutive patients. J Clin Endocrinol Metab 2004;89:453–62.
[15] Carmina E, Lobo RA. Polycystic ovaries in hirsute women with normal menses. Am J Med
2001;111:602–6.
[16] Kahsar-Miller MD, Nixon C, Boots LR, et al. Prevalence of polycystic ovary syndrome
(PCOS) in first-degree relatives of patients with PCOS. Fertil Steril 2001;75:35–8.
[17] Yildiz BO, Yarali H, Oguz H, et al. Glucose intolerance, insulin resistance, and hyperandro-
genemia in first degree relatives of women with polycystic ovary syndrome. J Clin Endocrinol
Metab 2003;88:2031–6.
[18] Sam S, Legro RS, Essah P, et al. Evidence for metabolic and reproductive phenotypes in
mothers of women with polycystic ovary syndrome. Proc Natl Acad Sci USA 2006;103:
7030–5.
[19] Escobar-Morreale HF, Luque-Ramirez M, San Millan JL. The molecular-genetic basis of
functional hyperandrogenism and the polycystic ovary syndrome. Endocr Rev 2005;26:
251–82.
[20] Urbanek M, Woodroffe A, Ewens KG, et al. Candidate gene region for polycystic ovary syn-
drome on chromosome 19p13.2. J Clin Endocrinol Metab 2005;90:6623–9.
[21] Tucci S, Futterweit W, Concepcion ES, et al. Evidence for association of polycystic ovary
syndrome in caucasian women with a marker at the insulin receptor gene locus. J Clin En-
docrinol Metab 2001;86:446–9.
[22] Futterweit W. Pathophysiology of polycystic ovarian syndrome. In: Redmond GP, editor.
Androgenic disorders. New York: Raven Press; 1995. p. 77–166.
[23] Polson DW, Adams J, Wadsworth J, et al. Polycystic ovaries: a common finding in normal
women. Lancet 1988;1:870–2.
[24] Michelmore KE, Balen AH, Douglas DB, et al. Polycystic ovaries and associated clinical and
biochemical features in young women. Clin Endocrinol (Oxf) 1999;51:779–86.
[25] Futterweit W, Diamanti-Kandarakis E, Azziz R. Clinical features of the polycystic ovary
syndrome. In: Azziz R, Nestler JE, Dewailly D, editors. Androgen excess disorders in
women: polycystic ovary syndrome and other disorders. 2nd edition. Totowa (NJ): Humana
Press Inc.; 2006. p. 155–67.
[26] Ehrmann DA. Polycystic ovary syndrome. N Engl J Med 2005;352:1223–36.
[27] Balen AH, Conway GS, Kaltsas G, et al. Polycystic ovary syndrome: the spectrum of the dis-
order in 1741 patients. Hum Reprod 1995;10:2107–11.
[28] Yeh HC, Futterweit W, Thornton JC. Polycystic ovarian disease: US features in 104
patients. Radiology 1987;163:11–6.
[29] van Hooff MH, Voorhorst FJ, Kapstein MB, et al. Predictive value of menstrual cycle
pattern, body mass index, hormone levels and polycystic ovaries at age 15 years for oligo-
amenorrhoea at age 18 years. Hum Reprod 2004;19:383–92.
[30] Legro RS. The genetics of obesity: lessons for polycystic ovary syndrome. Ann N Y Acad Sci
2000;900:193–202.
[31] Carmina E, Legro RS, Stamets K, et al. Difference in body weight between American and
Italian women with polycystic ovary syndrome: influence of the diet. Hum Reprod 2003;
18:2289–93.
[32] Dunaif A, Segal KR, Futterweit W, et al. Profound peripheral insulin resistance, indepen-
dent of obesity, in polycystic ovary syndrome. Diabetes 1989;38:1165–74.
[33] Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implica-
tions for pathogenesis. Endocr Rev 1997;18:774–800.
[34] Venkatesan AM, Dunaif A, Corbould A. Insulin resistance in polycystic ovary syndrome:
progress and paradoxes. Recent Prog Horm Res 2001;56:295–308.
[35] Gambeneri A, Pelusi C, Vicennati V, et al. Obesity and the polycystic ovary syndrome. Int
J Obes Relat Metab Disord 2002;26:883–9.
[36] Futterweit W, et al. An endocrine approach to obesity. In: Simopoulos AP, VanItallie TB,
Gullo SP, editors. Obesity: new directions in assessment and management. New York:
Charles Press; 1994. p. 96–121.
[37] Kiddy DS, Hamilton-Fairley D, Bush A, et al. Improvement in endocrine and ovarian func-
tion during dietary treatment of obese women with polycystic ovary syndrome. Clin Endo-
crinol (Oxf) 1992;36:105–11.
[38] Harborne L, Fleming R, Lyall H, et al. Metformin or antiandrogen in the treatment of hir-
sutism in polycystic ovary syndrome. J Clin Endocrinol Metab 2003;88:4116–23.
[39] Jakubowicz DJ, Nestler JE. 17 alpha-hydroxyprogesterone responses to leuprolide and
serum androgens in obese women with and without polycystic ovary syndrome after dietary
weight loss. J Clin Endocrinol Metab 1997;82:556–60.
[40] Boden G, Shulman GI. Free fatty acids in obesity and type 2 diabetes defining their role in the
development of insulin resistance and b-cell dysfunction. Eur J Clin Invest 2002;32(Suppl 3):
14–23.
[41] Vildirim B, Sadir N, Kaleli B. Relation of intra-abdominal fat distribution to metabolic
disorders in nonobese patients with polycystic ovary syndrome. Fertil Steril 2003;79:
1358–64.
[42] Carey DG, Jenkins AB, Campbell LV, et al. Abdominal fat and insulin resistance in normal
and overweight women: direct measurements reveal a strong relationship in subjects at both
low and high risk of IDDM. Diabetes 1996;45:633–8.
[43] DeNino WF, Tchernof A, Dionne IJ, et al. Contribution of abdominal adiposity to age-
related differences in insulin sensitivity and plasma lipids in healthy nonobese women. Dia-
betes Care 2001;24:925–32.
[44] Puder JJ, Varga S, Kraenzlin M, et al. Central fat excess in polycystic ovary syndrome:
relation to low-grade inflammation and insulin resistance. J Clin Endocrinol Metab 2005;
90:6014–21.
[45] Hermanns-Le T, Scheen A, Pierard GE. Acanthosis nigricans associated with insulin resis-
tance: pathophysiology and management. Am J Clin Dermatol 2004;5:199–203.
[46] Dunaif A, Green G, Phelps RG, et al. Acanthosis nigricans, insulin action, and hyperandro-
genism: clinical, histological and biochemical findings. J Clin Endocrinol Metab 1991;73:
590–5.
[47] Glueck CJ, Papanna R, Wang P, et al. Incidence and treatment of the metabolic syndrome in
newly referred women with confirmed polycystic ovarian syndrome. Metabolism 2003;52:
908–15.
[48] Coviello AD, Legro RS, Dunaif A. Adolescent girls with polycystic ovary syndrome have an
increased risk of the metabolic syndrome associated with increased androgen levels indepen-
dent of obesity and insulin resistance. J Clin Endocrinol Metab 2006;91:492–7.
[49] Apridonidze T, Essah PA, Iuorno MJ, et al. Prevalence and characteristics of the metabolic
syndrome in women with polycystic ovary syndrome. J Clin Endocrinol 2005;90:1929–35.
[50] Cobin RH, Futterweit W, Nestler JE, et al. American association of clinical endocrinologists
position statement on metabolic and cardiovascular consequences of polycystic ovary syn-
drome. Polycystic Ovary Writing Committee. Endocr Pract 2005;11:125–34.
[51] Nestler JE. Insulin resistance syndrome and polycystic ovary syndrome. Endocr Pract 2003;
9(Suppl 2):86–9.
[52] Diamanti-Kandarakis E, Kouli C, Alexandraki K, et al. Failure of mathematical indices to
accurately assess insulin resistance in lean, overweight, or obese women with polycystic
ovary syndrome. J Clin Endocrinol Metab 2004;89:1273–6.
[53] Ehrmann DA, Barnes RB, Rosenfeld RL, et al. Prevalence of impaired glucose tolerance and
diabetes in women with polycystic ovary syndrome. Diabetes Care 1999;22:141–6.
[54] Legro RS, Kunselman AR, Dodson WC, et al. Prevalence and predictors of risk for type 2
diabetes mellitus and impaired glucose tolerance in polycystic ovary syndrome: a prospective,
controlled study in 264 affected women. J Clin Endocrinol Metab 1999;84:165–9.
[55] Tominaga M, Eguchi H, Manaka H, et al. Impaired glucose tolerance is a risk factor for car-
diovascular disease, but not impaired fasting glucose. The Funagata Diabetes Study. Diabe-
tes Care 1999;22:920–4.
[56] Norman RJ, Masters L, Milner CR, et al. Relative risk of conversion from normoglycemia to
impaired glucose tolerance of non-insulin dependent diabetes mellitus in polycystic ovarian
syndrome. Hum Reprod 2001;16:1995–8.
[57] Golland IM, Vaughan Williams CA, Shalet SM, et al. Lack of predictive value of HbA1C for
impaired glucose tolerance in polycystic ovary syndrome. Gynecol Endocrinol 1989;3:
229–35.
[58] Legro RS, Gnatuk CL, Kunselman AR, et al. Changes in glucose tolerance over time in
women with polycystic ovary syndrome: a controlled study. J Clin Endocrinol Metab
2005;90:3236–42.
[59] Peppard HR, Marfori J, Iuorno MJ, et al. Prevalence of polycystic ovary syndrome among
premenopausal women with type 2 diabetes mellitus. Diabetes Care 2001;24:1050–2.
[60] Conn JJ, Jacobs HS, Conway GS. The prevalence of polycystic ovaries in women with type 2
diabetes mellitus. Clin Endocrinol (Oxf) 2000;52:81–6.
[61] Considine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive-leptin concentrations
in normal weight and obese humans. N Engl J Med 1996;334:292–5.
[62] Havel PJ, Kasim-Karakas S, Dubuc GR, et al. Gender differences in plasma leptin concen-
trations. Nat Med 1996;2:949–50.
[63] Heymsfield SB, Greenberg AS, Fujioka K, et al. Recombinant leptin for weight loss in obese
and lean adults: a randomized, controlled, dose-escalation trial. JAMA 1999;283:1567–8.
[64] Matsuzawa Y, Funahashi T, Kihara S, et al. Adiponectin and metabolic syndrome. Arterios-
cler Thromb Vasc Biol 2004;24:29–33.
[65] Koerner A, Kratzsch J, Kiess W. Adipocytokines: leptindthe classical, resistindthe contro-
versial, adiponectindthe promising, and more to come. Best Pract Res Clin Endocrinol
Metab 2005;19:525–46.
[66] Yamamoto Y, Hirose H, Saito I, et al. Adiponectin, an adipocyte-derived protein, predicts
future insulin resistance: two-year follow-up study in Japanese population. J Clin Endocrinol
Metab 2004;89:87–90.
[67] Rea R, Donnelly R. Resistin: an adipocyte-derived hormone. Has it a role in diabetes and
obesity? Diabetes Obes Metab 2004;6:163–70.
[68] Escobar-Morreale HF, Villuendas G, Botella-Carretero JI, et al. Adiponectin and resistin in
PCOS: a clinical, biochemical and molecular genetic study. Hum Reprod 2006;21:2257–65.
[69] Glintborg D, Andersen M, Hagen C, et al. Evaluation of metabolic risk markers in polycystic
ovary syndrome (PCOS). Adiponectin, ghrelin, leptin and body composition in hirsute
PCOS patients and controls. Eur J Endocrinol 2006;155:337–45.
[70] Carmina E, Orio F, Palomba S, et al. Evidence for altered adipocyte function in polycystic
ovary syndrome. Eur J Endocrinol 2005;152:389–94.
[71] Skommer J, Katulski K, Poreba E, et al. Gherlin expression in women with polycystic ovary
syndrome: a preliminary study. Eur J Gynaecol Oncol 2005;26:553–6.
[72] Neveu N, Granger L, St-Michel P, et al. Comparison of clomiphene citrate, metformin, or
the combination of both for first-line ovulation induction and achievement of pregnancy
in 154 women with polycystic ovary syndrome. Fertil Steril 2007;87:113–20.
[73] Liu KE, Tataryn IV, Sagle M. Use of metformin for ovulation induction in women who have
polycystic ovary syndrome with or without evidence of insulin resistance. J Obstet Gynaecol
Can 2006;28:595–9.
[74] Siebert TI, Kruger TF, Steyn DW, et al. Is the addition of metformin efficacious in the treat-
ment of clomiphene citrate-resistant patients with polycystic ovary syndrome? A structured
literature review. Fertil Steril 2006;86:1432–7.
[75] Homburg R. Pregnancy complications in PCOS. Best Pract Res Clin Endocrinol Metab
2006;20:281–92.
[76] Khattab S, Mohsen IA, Foutouh IA, et al. Metformin reduces abortion in pregnant women
with polycystic ovary syndrome. Gynecol Endocrinol 2006;22:680–4.
[77] Guido M, Romualdi D, Giuliani M, et al. Drospirenone for the treatment of hirsute women
with polycystic ovary syndrome: a clinical, endocrinological, metabolic pilot study. J Clin
Endocrinol Metab 2004;89:2817–23.
[78] Batukan C, Muderris II. Efficacy of a new oral contraceptive containing drospirenone and
ethinyl estradiol in the long-term treatment of hirsutism. Fertil Steril 2006;85:436–40.
[79] Shulman LP. Oral contraceptives: risks. Obstet Gynecol Clin North Am 2000;27:695–704.
[80] Azziz R. The evaluation and management of hirsutism. Obstet Gynecol 2003;101:995–1007.
Med Clin N Am 91 (2007) 1169–1184
Cardiovascular Morbidity and Mortality

of the Metabolic Syndrome
Kotaro Obunai, MDa,*, Sonal Jani, MDb,
George D. Dangas, MD, PhDa
a
Division of Cardiology, Department of Medicine, Columbia University College of Physicians
and Surgeons, Columbia University Medical Center, 161 Fort Washington Avenue,
HIP 5th Floor, New York, NY 10032, USA
b
Division of Cardiology, Department of Medicine, Beth Israel Medical Center, 5th floor,
Baird Hall, First Avenue at 16th Street, New York, NY 10003, USA
Cardiovascular disease (CVD), particularly coronary heart disease

(CHD), remains a major health concern in the developed world. More
than 15 million adults have CHD in the United States, and it kills one
out of every five people [1]. It affects equal numbers of men and women,
and has a high prevalence among minority populations. In 2004, approxi-
mately one third of deaths from CVD occurred prematurely, and over
147,000 Americans killed by CVD were under age 65. Despite the advances
in both reperfusion therapy and post myocardial infarction (MI) manage-
ment, approximately 38% of the people who experience a coronary event
in a given year will die from it. Primary prevention is obligatory; it is the
most affordable and feasible strategy currently available to us.
The metabolic syndrome (MetS) refers to a constellation of interrelated
cardiac risk factors that appear to directly promote development of athero-
sclerotic cardiovascular disease (ASCVD) [2–5]. It receives widespread
attention as obesity becomes a crisis of epidemic proportions in the United
States. This syndrome occurs in approximately one fourth of all American
adults [6]. Escalating rates of both obesity and diabetes among the younger
population, caused by discernible changes in lifestyle, diet, and environ-
ment, continue to fuel the cardiovascular epidemic. In this article, the
authors review the characteristic pathogenesis of MetS, and the evidence
from previous observational studies for its association with CVD. Contro-
versy exists as to whether the MetS actually reflects a unique underlying
pathological process. Many feel that it carries no incremental risk above
E-mail address: ko2182@columbia.edu (K. Obunai).
1170 OBUNAI et al
and beyond its independent risk factor components [7]. Such debates raise
uncertainty about the clinical value of diagnosing MetS and its role in pre-
dicting future cardiovascular events. These arguments and the clinical impli-
cations of the MetS are also reviewed.
Metabolic risk factors and cardiovascular disease

Central obesity and insulin resistance are the common underlying denom-
inators of MetS, and are thought to manifest as several metabolic risk fac-
tors, including atherogenic dyslipidemia, elevated plasma glucose, elevated
blood pressure, proinflammatory state, and prothrombotic state [2–5,8].
These metabolic risk factors are believed to have a direct effect on athero-
sclerotic disease.
Dyslipidemia
The dyslipidemia found in individuals who have the MetS consists of an
aggregate of lipoprotein abnormalities that include: elevated serum triglyc-
eride and apolipoprotein B (Apo B), low concentration of high-density lipo-
protein (HDL) cholesterol, and increased number of small dense low-density
lipoprotein (LDL) particles and small HDL particles. All of these abnormal-
ities are independently atherogenic [3,9–11].
Hyperglycemia
Metabolic abnormalities observed in diabetics, including hyperglycemia,
insulin resistance, and dyslipidemia, alter normal arterial function and ren-
der arteries susceptible to atherosclerosis [8]. The postulated mechanisms for
this include: endothelial dysfunction with decreased bioavailability of nitric
oxide [12,13], inflammation and proliferation of smooth muscle cells [14,15],
platelet dysfunction [16], and a decrease in endogenous anticoagulation as-
sociated with increased tissue factor and plasma coagulation factors [17,18].
Inflammation and prothrombotic markers

Individuals who have characteristics of MetS commonly manifest a proin-
flammatory and prothrombotic state. This is evidence by elevated levels of
C-reactive protein [19], interleukin-6, and plasminogen activator inhibitor-
1 (PAI-1) [20]. Studies have shown a linear relation between the number
of the MetS components and the level of C-reactive protein [6]. Inflamma-
tory and prothrombotic markers are known to be associated with an in-
creased risk of subsequent CVD [21].
Risk of cardiovascular disease associated with the metabolic syndrome

The data to support MetS as a risk factor for the subsequent develop-
ment of CVD are outlined below. Once diabetes emerges, cardiovascular
CARDIOVASCULAR MORBIDITY AND MORTALITY OF THE METS 1171
risk increases even further [22]. Summaries of relevant studies described in

this article are shown in Table 1.
Part of the difficulty in establishing the diagnosis of MetS is that it lacks
a uniformly accepted definition. There are three currently recognized criteria
sets, as reviewed separately in this issue. The existence of several diagnostic
criteria for the syndrome and the use of different outcomes can lead to dif-
ficulty when comparing data from separate studies. Although the World
Health Organization (WHO) criteria are more widely accepted and used
in Europe, the National Cholesterol Education Program Adult Treatment
Panel III (NCEP ATP-III) criteria for the MetS are viewed as simpler and
are more commonly used in the United States [3,23]. In most long-term fol-
low-up studies, either the NCEP ATP-III definition or the WHO criteria
were used. The International Diabetes Federation (IDF) criteria were de-
fined more recently, and there have been fewer studies testing its predictive
value for CVD compared with other criteria [5].
Impact on development of cardiovascular disease

and major coronary events
A total of 12,089 individuals who did not have diabetes or CVD at base-
line were followed for an average of 11 years in the Atherosclerosis Risk In
Communities (ARIC) study [24]. Men and women who had the MetS were
1.5 to 2 times more likely to develop CHD than individuals who did not
have MetS after adjustment for age, smoking status, LDL cholesterol,
and race. Among subjects in the placebo arms of the secondary prevention
trial (Scandinavian Simvastatin Survival Study: 4S) and the primary preven-
tion trial (the Air Force/Texas Coronary Atherosclerosis Prevention Study),
subjects who had MetS were 1.5 and 1.4 times, respectively, more likely to
have major coronary events (MCE) than those who did not have the syn-
drome during the follow-up [25].
Impact on cardiovascular disease/coronary heart disease mortality

and overall mortality
In the Kuopio Ischemic Heart Disease Risk Factor Study, 1209 Finnish
men were followed up for a median of 11.4 years [26]. Subjects who had
diabetes or history of CVD at enrollment were excluded from the analysis.
After adjusting for variables such as age, smoking status, family history of
CHD, and LDL cholesterol, subjects who had MetS were at increased risk
for CVD and CHD mortality (relative risk [RR] 2.52 and 4.16, respectively).
They were also at increased risk for all-cause mortality (RR 2.02) when com-
pared with those who did not have the syndrome.
In the Aerobics Center Longitudinal Study, performed in the United
States, 19,223 men who had no history of CHD or stroke at the time of
baseline examination were followed for a median of 17 years [27]. Again,
after adjustment for variables such as age, year of examination, smoking
1172
Table 1
Characteristics of Cohort studies of metabolic syndrome and incident cardiovascular disease and death
Study name, Age (mean), Follow-up MetS
Study author, year (country) Sample size sex years criteria Outcomes RR
McNeill, 2005 [24] ARIC (US) 12,089, DM and 54, men 43% 11 NCEP MI, fatal CHD M 1.46 (1.23–1.74)
CHD excluded coronary W 2.05 (1.59–2.64)
revascularization
Ischemic stroke M 1.42 (0.96–2.11)
OBUNAI
W 1.96 (1.28–3.00)
Girman, 2004 [25] 4S (US) 1991, DM excluded, 59, men 81% 5.4 Modified Major coronary 1.5 (1.2–1.8)
et al
CHD included NCEP events (fatal and
AFCAPS/TexCAPS 3188, DM and CHD 58, men 85% 5 Modified non-fatal MI, 1.4 (1.04–1.9)
(US) excluded NCEP sudden cardiac
death, and
unstable angina)
Lakka, 2002 [26] Kuopio Study 1,209, DM and CHD 52, men 100% 11.4 NCEP All-cause mortality 2.02 (1.14–3.59)
Kurl, 2006 [30] (Finland) excluded CVD mortality 2.52 (1.10–5.78)
1,131, stroke, DM, CHD mortality 4.16 (1.60–10.8)
CHD excluded All stroke 2.05 (1.03–4.11)
Ischemic stroke 2.41 (1.12–5.32)
Katzmarzyk, 2004 ACLS (US) 19,223, DM 43, men 100% 10.2 NCEP All-cause mortality 1.29 (1.05–1.57)
[27] included, CHD CVD mortality 1.89 (1.36–2.60)
excluded
Malik, 2004 [28] NHANES II (US) 6255, DM and CHD 50, men 46% 13.3 Modified All-cause mortality 1.40 (1.19–1.66)
included NCEP CVD mortality 1.82 (1.40–2.37)
CHD mortality 2.02 (1.42–2.89)
Ford 2004 [31] NHANES II (US) 2431, DM and CHD 50, men 46% 13.5 Modified All-cause mortality 1.15 (0.92–1.45)
included NCEP CVD mortality 1.37 (1.02–1.85)
CHD mortality 1.29 (0.92–1.82)
Mortality from 1.68 (0.86–3.27)
stroke
Levantesi, 2005 [29] GISSI-Prevenzione 10,384, with MI 59, men 85% 3.5 Modified All-cause mortality 1.29 (1.10–1.51)
Trial (Italy) !3months NCEP CV events (CV 1.23 (1.06–1.42)
CARDIOVASCULAR MORBIDITY AND MORTALITY OF THE METS

death, nonfatal
MI, nonfatal stroke)
Kip 2004 [32] WISE (US) 780, DM included, 58, women 100% 3.5 NCEP All-cause mortlality 2.01 (1.26–3.20)
CHD excluded MACE 1.88 (1.38–2.57)
Hunt 2004 [33] San Antonio Heart 2815, DM and CHD 43, men 43% 12.7 NCEP All-cause mortality 1.47 (1.13–1.92)
Study (US) included CVD mortality 2.53 (1.74–3.67)
M1.82 (1.14–2.91)
W4.65 (2.35–9.21)
Butler 2006 [40] Health ABC (US) 3035, DM and CHD 73.6, men 49% 6 NCEP Coronary events 1.56 (1.28–1.91)
included (hospitalization
for MI, angina,
revascularization)
MI 1.51 (1.12–2.05)
Hospitalization for 1.49 (1.10–2.00)
CHF
Sattar 2003 [60] WOSCOPS (UK) 6447, DM and CHD 55, men 100% 4.9 Modified Non-fatal MI or 1.30 (1.00–1.67)
excluded NCEP CHD death
Abbreviations: DM, diabetes mellitus; M, men; MACE, major adverse cardiac events; NCEP, National Cholesterol Education Program; RR, relative risk;
W, women.
1173
1174 OBUNAI et al
status, alcohol consumption, and parental CVD, the relative risks of all-
cause and CVD mortality were 1.29 and 1.89, respectively, for men who
had the MetS compared with healthy men.
The mortality impact of the MetS has also been compared with the mor-
tality impact of pre-existing CVD. A prospective cohort study from the Sec-
ond National Health and Nutrition Examination Survey (NHANES II)
followed 6255 subjects between the ages of 30 to 75 years, of whom 54%
were female. This was representative of 64 million adults in the United
States. Subjects were followed for a mean of 13.3 years [28]. Compared
with those who had neither MetS nor prior CVD, hazard ratios (HRs) for
CHD mortality were 2.02 for those who had MetS and 4.19 for those
who had pre-existing CVD. For CVD mortality, HRs were 1.82 and 3.14,
respectively; for overall mortality, HRs were 1.40 and 1.87, respectively.
When subjects who had no metabolic risk factors were used as the reference
group, the HR for CHD mortality was an impressive 3.51 for those who had
the MetS.
Impact on mortality and major coronary events in individuals

who have prior myocardial infarction
Type 2 diabetes mellitus is a well-recognized risk factor for cardiovascu-
lar events and mortality in patients who have pre-existing CHD. The clinical
impact of MetS in patients who had prior-MI was studied using GISSI-Pre-
venzione Trial data [29]. Although not as strong as the presence of diabetes,
presence of MetS was a significant predictor of death or cardiovascular
events, with RR 1.29 and 1.23, respectively.
Stroke
In the ARIC study, 216 incident ischemic stroke events occurred during
the follow-up of 12,089 individuals [24]. The RR for the events was 1.96 in
women and 1.42 in men. Other cohort studies also demonstrated an elevated
risk for ischemic stroke (HR ¼ 2.41, Kuopio Ischemic Heart Disease Risk Fac-
tor Study) [30] and its subsequent mortality (HR ¼ 1.68, NHANES II) [31].
Women
An analysis of the cardiovascular risk of MetS in women was performed
using data from the Women’s Ischemia Syndrome Evaluation (WISE) Study
[32]. Seven hundred and eighty women who were referred for coronary
angiography were classified by body mass index (BMI) and the presence
or absence of the MetS, and then followed up for an average of 3.5 years.
This analysis showed that the MetS, but not BMI alone, was strongly asso-
ciated with angiographic coronary artery disease (CAD) and conferred an
approximate twofold adjusted risk of death and major adverse cardiac
events (MACE).
Also, note that in the previously mentioned ARIC study, the risk of CHD
associated with the MetS was significantly higher in women (HRs ¼ 2.55)
than in men (HRs ¼ 1.51) (P ¼ .0006) [24]. In the San Antonio Heart Study,
the cardiovascular mortality risk in subjects who had MetS was also
shown to be significantly higher in women than in men. HRs for NCEP-MetS
were 4.65 and 1.82 (P ¼ .03), whereas HRs for WHO-MetS were 2.83 and 1.15
(P ¼ .02), respectively [33]. The gender differences seen in cardiovascular
mortality were only significant in individuals who had both MetS and type 2
diabetes. The subgroup analysis of two recently published meta-analyses
indicate that the MetS might be a stronger risk factor for CVD in women
than in men, with RR 2.10 versus 1.57 [34], and RR 2.63 versus 1.98 [35],
respectively.
Gender-specific differences exist not only in the diagnosis and prognosis
of CHD in women, but also in the pathophysiology of CHD [36]. High
levels of inflammation and the decrease in estrogen associated with meno-
pause promote a wide variety of changes associated with atherosclerosis,
most notably decreases in arterial compliance and endothelial function.
Estrogen deficiency with anovulatory cycles may lead to similar changes
even in premenopausal women. As women are known to have relatively
smaller arterial lumens and lower rates of successful revascularization, the
development of vascular dysfunction becomes especially concerning. Sex
hormones have also been implicated in the atherosclerotic process. Their
detrimental role on vascular wall function, plaque deposition, and impaired
flow reserve needs to be better understood. Furthermore, the impact of
diabetes as a risk factor is more potent in women [37,38]. In women who
have MetS, chronic hyperglycemia coupled with high insulin levels further
reduces coronary vasodilator function.
The observations regarding women and the MetS are particularly impor-
tant in light of the public health impact of CHD on this population. The life-
time risk of dying from heart disease for a woman is one in three. It was
noted with the recently updated guidelines on prevention of CHD in women
that health care professionals should focus on women’s lifetime heart dis-
ease risk and not just on short-term risk [39]. As discussed later, this is an
area where determination of the MetS may be superior to calculation of
Framingham risk score, which focuses primarily on 10-year risk. The guide-
lines emphasize that prevalence of CHD in women is such that nearly all
women should be considered at risk for atherosclerosis.
Elderly adults
The majority of data regarding the MetS and cardiovascular outcome are
derived from populations consisting of middle-aged and younger subjects;
however, the impact of the MetS on cardiovascular morbidity and mortality
has been assessed in the elderly population in the Health, Aging, and Body
Composition Study (Health ABC) cohort [40]. A total of 3035 individuals,
1176 OBUNAI et al
aged 70 to 79 years, were followed for 6 years. Subjects who had MetS were
at a significantly higher risk for MCE (HR 1.56), MI (HR 1.51), and hospi-
talization caused by heart failure (HR 1.49); however, no significant differ-
ence in mortality was observed.
Aortic stenosis
The development and progression of calcific aortic stenosis are linked to
various traditional risk factors for vascular atherosclerosis [41,42]. Further-
more, calcific valvular lesions, such as those seen in patients who have aortic
stenosis, share many histological similarities with atherosclerosis [43,44].
Echocardiographic progression of aortic stenosis and clinical outcomes, in-
cluding death and need for aortic replacement, were evaluated in a retrospec-
tive study of 105 patients who had moderate to severe aortic stenosis [45].
During the follow-up of 28 months, patients who had the MetS had more
rapid disease progression and a significantly lower 3-year event-free survival
(44% versus 69%t) compared with those who did not have the MetS [45].
Heart failure
The relationship between obesity and the risk of heart failure (HF) is
well-established [46]. The potential mechanisms underlying this correlation
include alterations in cardiac loading conditions associated with increases
in blood volume, cardiac output, blood pressure, and peripheral vascular
resistance. The presence of a sleep-disordered breathing pattern, which is
typically seen in obese individuals, is also important [47]. This long-term
hemodynamic overload may result in hypertrophy and remodeling of both
left and right ventricles. Previous studies have also shown that insulin resis-
tance may be associated with both left ventricular systolic and diastolic
dysfunction and left ventricular remodeling [48–50].
Recently published data suggest an association between the MetS and
HF, potentially because of a direct myocardial effect in addition to its
atherogenic effects [51]. In one study, a total of 2314 50-year-old Swedish
men free from HF, myocardial infarction, and valvular heart disease at
baseline were followed until the age of 70. Metabolic syndrome was a signifi-
cant predictor of HF (HR ¼ 1.80) independent of established etiologies of
HF, including an interim myocardial infarction, hypertension, and obesity.
The authors concluded that higher incidence of HF among individuals who
have MetS could be caused by numerous plausible direct myocardial toxic
effects of insulin resistance and its accompanying hyperinsulinemia.
Meta-analysis
A meta-analysis of 21 prospective cohort studies using either NCEP
ATP-III or WHO diagnostic criteria for the MetS was recently published
[34]. Individuals who had the MetS, compared with those who did not,
had an increased mortality from all causes (RR 1.35; 95% confidence inter-
val [CI], 1.17–1.56) and CVD (RR 1.74; 95% CI, 1.29–2.35), as well as an
increased incidence of CVD (RR 1.53; 95% CI, 1.26–1.87), CHD
(RR 1.52; 95% CI, 1.37–1.69), and stroke (RR 1.76; 95% CI, 1.37–2.25).
Subgroup analysis also indicated that the MetS might be a stronger risk fac-
tor for CVD in women than in men. The relative risk of CVD associated
with the MetS was higher in studies that used the WHO definition compared
with studies that used the NCEP ATP III definition. Another meta-analysis
including 37 studies and 172,573 individuals [35] showed that the MetS had
an RR of cardiovascular events and death of 1.78 (95% CI, 1.58-2.0). Again,
the association between MetS and cardiovascular mortality was stronger in
women (RR ¼ 2.63 versus 1.98 P ¼ .09) [35].
Compare with the sum of its parts

Questions have been raised as to whether the risk for CHD associated
with MetS is greater than just the sum of its individual components. A pro-
spective cohort study of 3589 British women who were free of CHD at base-
line showed that the magnitude of association between CVD and the
individual components of the syndrome was similar to or even larger than
that for the syndrome [52]. Similar conclusions were drawn from several
other studies, including ARIC [24] and WOSCOPS (West of Scotland Cor-
onary Prevention Study) [53]. These results have led some experts in the field
to question the clinical and predictive value of defining the syndrome [7].
Comparison with Framingham-based risk scores

The Framingham Risk Score (FRS) has been derived from the Framing-
ham Heart Study, and is the most commonly recommended tool in the
United States for global risk assessment in the primary prevention setting
[37,54]. Short-term absolute risk, defined as the probability of a person de-
veloping a hard CHD endpoint (MI or cardiac death) in the next 10 years,
can be estimated by calculating the number of Framingham points assigned
to each graded risk (Table 2). The predictive value of the FRS has been val-
idated in multiple ethnically and geographically diverse cohorts [55]. The
score can be calculated based on non-fasting blood samples in most individ-
uals (whereas most MetS criteria require sampling of fasting blood) because
cholesterol and HDL cholesterol are not measurably altered in the non-
fasting state and triglyceride is not considered for the risk calculation. At
least three recent studies have compared the short-term predictive value of
the MetS with that of the FRS [24,56,57]. Using receiver operator character-
istic curves, all three studies demonstrated that risk prediction by FRS is
superior to the diagnosis of MetS.
The inferiority of MetS criteria in prediction of short-term cardiovascular
events has been explained through several postulates [7,58–60]. First, MetS
criteria do not include several major risk factors, such as age, gender,
1178 OBUNAI et al
Table 2
Risk factors for CHD, their association with Framingham risk scoring system and MetS diag-
nostic criteria, and target risk factors for lifestyle and pharmacological interventions
Framingham Metabolic Treatment
Risk Score syndrome strategies
Major Risk Factors
Elevated blood pressure X X Lifestyle/Pharm
Diabetes mellitus X X Lifestyle/Pharm
Elevated total and LDL cholesterol X Lifestyle/Pharm
Low HDL cholesterol X X Lifestyle/Pharm
Cigarette smoking X Lifestyle
Advancing age X
Male sex X
Family history of premature ASCVD
Emerging Risk Factors
Elevated trigliycerides X Lifestyle/Pharm
Elevated lipoprotein(a)
Small LDL/HDL particles
Inflammatory markers
Prothrombotic factors
Other Risk Factors
Obesity X Lifestyle
Physical inactivity Lifestyle
Atherogenic diet Lifestyle
Abbreviations: Lifestyle, lifestyle intervention; Pharm, pharmacological intervention.
smoking status, or total cholesterol, which greatly enhance risk prediction

and are included in the FRS (see Table 2). Second, the linear association
between the severity of risk factors and CVD risk is not considered in the
dichotomous nature of MetS criteria. Finally, in several unique components
of the MetS, incremental risk prediction beyond the usual risk factors of
FRS has not been proven in prospective studies. Indeed, the Framingham
investigators found little or no increase in the predictive power of FRS
for cardiovascular events by adding abdominal obesity, triglycerides, or
fasting glucose, which are more accurate predictors of type 2 diabetes
[61]. These less conventional risk factors may have long-term importance,
but require further data to test their validity in predicting vascular events.
Risk assessment for future development of atherosclerotic cardiovascular

disease in individuals who have the metabolic syndrome
To reduce the lifetime risk of ASCVD, all individuals found to have the
MetS deserve a comprehensive global risk assessment and long-term moni-
toring [2]. Individuals who have clinical ASCVD (eg, CHD, stroke, or peri-
pheral artery disease [PAD]) or who have diabetes belong in a high-risk
category and should be treated more aggressively with risk factor-targeted
pharmacotherapy, in addition to lifestyle intervention to significantly reduce
morbidity and mortality. Recommendations from ATP-III and the 2006

American College of Cardiology (ACC)/American Heart Association
(AHA) guidelines on secondary prevention should be followed in this risk
category [4,62,63].
For MetS patients who do not have CHD or its equivalents, global risks
need to be assessed on an individual basis to tailor the intensity of preven-
tive therapy [4]. It should be remembered, however, that detecting MetS is
only one part of overall risk assessment for ASCVD, and it is not a robust
tool to estimate short-term (eg, 10-year) risk for ASCVD [2]. This is largely
because of the lack of major risk factors included in its diagnostic criteria, as
discussed earlier, whereas standard risk-prediction algorithms do include
them (see Table 2). Thus, short-term risk assessment among individuals
who have no previous history of ASCVD is best performed with risk equa-
tions such as FRS [64]. For use in daily clinical practice, the computer or
personal digital assistant-based and Web-based FRS calculators are avail-
able (http://www.nhlbi.nih.gov/guidelines/cholesterol/index.htm). This as-
sessment triages patients into three risk categories, based on 10-year
absolute risk for a hard CHD: high risk (10-year risk O20%), intermediate
risk (10-year risk 10% to 20%), or lower risk (10-year risk !10%) [4,37].
Recommendations for the management of individual risk factors, from
ACC/AHA [65], American Diabetes Association [66], ATP III [4], and
The Seventh Report of the Joint National Committee (JNC VII) [67] are
available to match the intensity of therapy to the risk hazard.
Clinical implications of diagnosing the metabolic syndrome

The surge of public interest in the MetS and the flux of recent publica-
tions on this subject have increased health care providers’ interest and un-
derstanding of the links between obesity, insulin resistance, and CVD.
Although studies examining the short-term predictive value of the MetS
do not support its superiority over the Framingham-based risk prediction
system, the diagnosis of the MetS remains a strong predictor of CVD
risk, and identifies high-risk patients in whom detailed global risk assess-
ment, intensive lifestyle interventions, and potential pharmacotherapy are
warranted.
Furthermore, there are several implications for the clinical use of this en-
tity. First, several limitations of the FRS have been documented. The FRS
tends to underestimate the true burden of risk in certain populations, such
as those who have multiple borderline risk factors. This serves to falsely re-
assure these individuals as low risk [68,69]. Also, it has been recognized that
FRS is not a robust tool for lifetime risk prediction, especially in young in-
dividuals who have risk factors [68,70,71]. For example, let us calculate the
10-year risk for a hard CHD in a 50-year-old obese man who has the follow-
ing risk profile: no type 2 diabetes, no cigarette smoking, systolic blood
1180 OBUNAI et al
pressure of 165 mmHg, serum total cholesterol 260 mg/dL, HDL cholesterol
60 mg/dL, triglyceride 250 mg/dL, and LDL cholesterol 150 mg/dL. His 10
year risk is 8%, whereas his estimated lifetime risk is 69% [68]. In these in-
dividual subsets, care providers may falsely underestimate the true lifetime
risk burden because of relatively low 10-year risk score, and may undertreat
risk factors. Indeed, according to the NCEP ATP-III guidelines for the CHD
prevention [4], lifestyle intervention may not be indicated in this 50-year-old
man, given his relatively low 10-year risk score and serum LDL level.
Any single risk factor can produce cumulative damage and increase mor-
bidity and mortality if left untreated long-term, however. As discussed ear-
lier, the MetS carries an increased relative risk for CVD in the short and
long term. The simple label of MetS, even with low 10-year risk score,
may encourage health care providers to be more aggressive in managing
risk factors.
Second, a diagnosis of the MetS emphasizes the importance of risk factor
modification through lifestyle changes. These interventions should focus on
weight reduction, increased physical activity, an anti-atherogenic diet, and
smoking cessation. Unfortunately, lifestyle intervention is often neglected
in daily practice. Studies demonstrated that increased obesity reduces the
beneficial effect of other risk-reduction strategies [72]. Institution of lifestyle
interventions helps to lessen the impact of each metabolic risk factor on an
individual at every stage, but particularly in early phases, and slows the sub-
sequent development of CVD. As stated earlier, identification of individuals
who have the MetS is aimed at long-term risk reduction primarily through
lifestyle modification, whereas the FRS was developed to identify individ-
uals at increased short-term risk.
Incorporating the diagnosis of the MetS into routine clinical practice
may serve as a ‘‘teachable moment,’’ in which health care providers can
help patients understand their true lifetime risk profile, and motivate them
to adopt lifestyle changes that promote health and longevity.
Summary
CVD remains the most widespread health care problem in the United
States. Recently, the MetS has received an increasing amount of attention
because of the growing prevalence of obesity and its association with heart
disease. Longitudinal observational studies have confirmed that the MetS
is a risk factor for the subsequent development of CVD and mortality.
Evidence also suggests that women who have MetS are more susceptible
to CVD and cardiovascular mortality than men. Identifying patients who
have MetS may enhance the clinical management of individuals who have
various risk factors.
Although the scientific basis for the MetS is still debated, its clinical value
in identifying individuals who have a compilation of risk factors who are at
an increased risk for CVD is undisputed. All patients who have the MetS
should receive a thorough global risk assessment using validated risk calcu-
lation tools such as the FRS. Also, in patients who have the MetS, long-
term management of each risk factor is essential. This is true even in those
individuals who have low FRS, because lifetime risk for CVD with MetS is
higher than expected in untreated patients. Lifestyle intervention is funda-
mental and should be introduced, maintained, and reinforced in all individ-
uals. Pharmacotherapy, targeting individual risk factors, should be guided
by global risk assessment and accepted treatment guidelines.
References
[1] Rosamond W, Flegal K, Friday G, et al, for the American Heart Association Statistics
Committee and Stroke Statistics subcommittee. Heart disease and stroke statisticsd2007
update: a report from the American Heart Association Statistics Committee and Stroke
Statistics Subcommittee. Circulation 2007;115:e69–171.
[2] Grundy SM, Cleeman JI, Daniels SR, et al, American Heart Association. National Heart,
Lung, and Blood Institute. Diagnosis and management of the metabolic syndrome: an
American Heart Association/National Heart, Lung, and Blood Institute scientific statement.
Circulation 2005;112(17):2735–52.
[3] Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults.
Executive summary of the third report of the National Cholesterol Education Program
(NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol
in Adults (Adult Treatment Panel III). JAMA 2001;285:2486–97.
[4] National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation,
and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third re-
port of the National Cholesterol Education Program (NCEP) Expert Panel on Detection,
Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel
III) final report. Circulation 2002;106:3143–421.
[5] Alberti KGMM, Zimmet P, Shaw J, et al, for the IDF Epidemiology Task Force Consensus
Group. The metabolic syndromeda new worldwide definition. Lancet 2005;366:1059–62.
[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(3):356–9.
the Study of Diabetes. Diabetes Care 2005;28:2289–304.
[8] Beckman JA, Creager MA, Libby P. Diabetes and atherosclerosis: epidemiology, patho-
physiology, and management. JAMA 2002;287(19):2570–81.
[9] Carr MC, Brunzell JD. Abdominal obesity and dyslipidemia in the metabolic syndrome:
importance of type 2 diabetes and familial combined hyperlipidemia in coronary artery
disease risk. J Clin Endocrinol Metab 2004;89(6):2601–7.
[10] Carmena R, Duriez P, Fruchart JC. Atherogenic lipoprotein particles in atherosclerosis.
Circulation 2004;109(III):2–7.
[11] Ginsberg HN, Zhang YL, Hernandez-Ono A. Metabolic syndrome: focus on dyslipidemia.
Obesity (Silver Spring) 2006;14(Suppl 1):41S–9S.
[12] Williams SB, Cusco JA, Roddy MA, et al. Impaired nitric oxide-mediated vasodilation in
patients with non–insulin-dependent diabetes mellitus. J Am Coll Cardiol 1996;27:567–74.
[13] Hopfner RL, Gopalakrishnan V. Endothelin: emerging role in diabetic vascular complica-
tions. Diabetologia 1999;42:1383–94.
1182 OBUNAI et al
[14] Nugent AG, McGurk C, Hayes JR, et al. Impaired vasoconstriction to endothelin 1 in
patients with NIDDM. Diabetes 1996;45:105–7.
[15] Suzuki LA, Poot M, Gerrity RG, et al. Diabetes accelerates smooth muscle accumulation in
lesions of atherosclerosis. Diabetes 2001;50:851–60.
[16] Vinik AI, Erbas T, Park TS, et al. Platelet dysfunction in type 2 diabetes. Diabetes Care 2001;
24:1476–85.
[17] Carr ME. Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15:
44–54.
[18] Ceriello A, Giacomello R, Stel G, et al. Hyperglycemia-induced thrombin formation in
diabetes. Diabetes 1995;44:924–8.
[19] Ridker PM, Buring JE, Cook NR, et al. C-reactive protein, the metabolic syndrome, and risk
of incident cardiovascular events: an 8-year follow-up of 14,719 initially healthy American
women. Circulation 2003;107:391–7.
[20] Festa A, D’Agostino R Jr, Tracy RP, et al. Elevated levels of acute-phase proteins and
plasminogen activator inhibitor-1 predict the development of type 2 diabetes: the Insulin
Resistance Atherosclerosis Study. Diabetes 2002;51:1131–7.
[21] Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection
and prevention. Circulation 2003;107:363–9.
[22] Nesto RW. Correlation between cardiovascular disease and diabetes mellitus: current con-
cepts. Am J Med 2004;116(Suppl 5A):11S–22S.
a WHO consultation. Diabet Med 1998;15:539–53.
[24] McNeill AM, Rosamond WD, Girman CJ, et al. The metabolic syndrome and 11-year risk of
incident cardiovascular disease in the Atherosclerosis Risk in Communities Study. Diabetes
Care 2005;28:385–90.
[25] Girman CJ, Rhodes T, Mercuri M, et al. The metabolic syndrome and risk of major coronary
events in the Scandinavian Simvastatin Survival Study (4S) and the Air Force/Texas
Coronary Atherosclerosis Prevention Study (AFCAPS/TexCAPS). Am J Cardiol 2004;93:
136–41.
[26] Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardio-
vascular disease mortality in middle-aged men. JAMA 2002;288:2709–16.
[27] Katzmarzyk PT, Church TS, Blair SN. Cardiorespiratory fitness attenuates the effects of the
metabolic syndrome on all-cause and cardiovascular disease mortality in men. Arch Intern
Med 2004;164:1092–7.
[28] Malik S, Wong D, Franklin SS, et al. Impact of the metabolic syndrome on mortality from
coronary heart disease, cardiovascular disease and all causes in United Stated adults. Circu-
lation 2004;110:1245–50.
[29] Levantesi G, Macchia A, Marfisi R, et al, GISSI-Prevenzione Investigators. Metabolic
syndrome and risk of cardiovascular events after myocardial infarction. J Am Coll Cardiol
2005;46(2):277–83.
[30] Kurl S, Laukkanen JA, Niskanen L, et al. Metabolic syndrome and the risk of stroke in
middle-aged men. Stroke 2006;37(3):806–11.
[31] Ford ES. The metabolic syndrome and mortality from cardiovascular disease and all-cause:
findings from the National Health and Nutrition Examination Survey II Mortality Study.
Atherosclerosis 2004;173:309–14.
[32] Kip KE, Marroquin OC, Kelley DE, et al. Clinical importance of obesity versus the meta-
bolic syndrome in cardiovascular risk in women: a report from the Women’s Ischemia
Syndrome Evaluation (WISE) Study. Circulation 2004;109(6):706–13.
[33] Hunt KJ, Resendez RG, Williams K, et al. National Cholesterol Education Program
versus World Health Organization metabolic syndrome in relation to all-cause and
cardiovascular mortality in the San Antonio Heart Study. Circulation 2004;110:
1251–7.
[34] Galassi A, Reynolds K, He J. Metabolic syndrome and risk of cardiovascular disease:

a meta-analysis. Am J Med 2006;119(10):812–9.
[35] Gami AS, Witt BJ, Howard DE, et al. Metabolic syndrome and risk of incident cardiovas-
cular events and death: a systematic review and meta-analysis of longitudinal studies. J Am
Coll Cardiol 2007;49(4):403–14.
[36] Bairey Merz CN, Shaw LJ, Reis SE, et al, WISE Investigators. Insights from the NHLBI-
sponsored Women’s Ischemia Syndrome Evaluation (WISE) Study: part II: gender differ-
ences in presentation, diagnosis, and outcome with regard to gender-based pathophysiology
of atherosclerosis and macrovascular and microvascular coronary disease. J Am Coll
Cardiol 2006;47(Suppl 3):S21–9.
[37] Wilson PW, D’Agostino RB, Levy D, et al. Prediction of coronary heart disease using risk
factor categories. Circulation 1998;97(18):1837–47.
[38] Huxley R, Barzi F, Woodward M. Excess risk of fatal coronary heart disease associated with
diabetes in men and women: meta-analysis of 37 prospective cohort studies. BMJ 2006;332:
73–8.
[39] Mosca L, Banka CL, Benjamin EJ, et al. Evidence-based guidelines for cardiovascular dis-
ease prevention in women: 2007 update. Circulation 2007;115(11):1481–501.
[40] Butler J, Rodondi N, Zhu Y, et al, Health ABC Study. Metabolic syndrome and the risk of
cardiovascular disease in older adults. J Am Coll Cardiol 2006;47(8):1595–602.
[41] Mohler ER, Sheridan MJ, Nichols R, et al. Development and progression of aortic valve
stenosis: atherosclerosis risk factorsda causal relationship? A clinical morphologic study.
Clin Cardiol 1991;14:995–9.
[42] Peltier M, Trojette F, Sarano ME, et al. Relation between cardiovascular risk factors and
nonrheumatic severe calcific aortic stenosis among patients with a three-cuspid aortic valve.
Am J Cardiol 2003;91(1):97–9.
[43] Rajamannan NM, Gersh B, Bonow RO. Calcific aortic stenosis: from bench to the bedsided
emerging clinical and cellular concepts. Heart 2003;89:801–5.
[44] Olsson M, Thyberg J, Nilsson J. Presence of oxidized low density lipoprotein in nonrheu-
matic stenotic aortic valves. Arterioscler Thromb Vasc Biol 1999;19(5):1218–22.
[45] Briand M, Lemieux I, Dumesnil JG, et al. Metabolic syndrome negatively influences disease
progression and prognosis in aortic stenosis. J Am Coll Cardiol 2006;47(11):2229–36.
[46] Kenchaiah S, Evans JC, Levy D, et al. Obesity and the risk of heart failure. N Engl J Med
2002;347:305–13.
[47] Alpert MA. Obesity cardiomyopathy: pathophysiology and evolution of the clinical syn-
drome. Am J Med Sci 2001;321(4):225–36.
[48] Ärnlöv J, Lind L, Zethelius B, et al. Several factors associated with the insulin resistance
syndrome are predictors of left ventricular systolic dysfunction in a male population after
20 years of follow-up. Am Heart J 2001;142:720–4.
[49] 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.
[50] Rutter MK, Parise H, Benjamin EJ, et al. Impact of glucose intolerance and insulin resis-
tance on cardiac structure and function: sex-related differences in the Framingham Heart
Study. Circulation 2003;107:448–54.
[51] Ingelsson E, Arnlov J, Lind L, et al. Metabolic syndrome and risk for heart failure in middle-
aged men. Heart 2006;92(10):1409–13.
[52] Lawlor DA, Smith GD, Ebrahim S. Does the new International Diabetes Federation defini-
tion of the metabolic syndrome predict CHD any more strongly than older definitions? Find-
ings from the British Women’s Heart and Health Study. Diabetologia 2006;49:41–8.
protein as a predictor of coronary heart disease and diabetes in the West of Scotland
Coronary Prevention Study. Circulation 2003;108:414–9.
[54] Grundy SM, Pasternak R, Greenland P, et al. Assessment of cardiovascular risk by use of
multiple-risk-factor assessment equations: a statement for healthcare professionals from
1184 OBUNAI et al
the American Heart Association and the American College of Cardiology. Circulation 1999;
100(13):1481–92.
[55] D’Agostino RB Sr., Grundy S, Sullivan LM, et al, CHD Risk Prediction Group. Validation
of the Framingham coronary heart disease prediction scores: results of a multiple ethnic
groups investigation. JAMA 2001;286(2):180–7.
[56] Stern MP, Williams K, Gonzalez-Villalpando C, et al. Does the metabolic syndrome
improve identification of individuals at risk of type 2 diabetes and/or cardiovascular disease?
Diabetes Care 2004;27(11):2676–81.
[57] Wannamethee SG, Shaper AG, Lennon L, et al. Metabolic syndrome vs Framingham Risk
Score for prediction of coronary heart disease, stroke, and type 2 diabetes mellitus. Arch
Intern Med 2005;165(22):2644–50.
[58] Greenland P. Critical questions about the metabolic syndrome. Circulation 2005;112:
3675–6.
[59] Meigs JB. Metabolic syndrome: in search of a clinical role. Diabetes Care 2004;27:2761–3.
[60] Sattar N, Forouhi NG. Metabolic syndrome criteria: ready for clinical prime time or work in
progress? Eur Heart J 2005;26:1249–51.
[61] Grundy SM, Brewer HB Jr, Cleeman JI, et al, American Heart Association. National Heart,
Lung, and Blood Institute. Definition of metabolic syndrome: report of the National Heart,
Lung, and Blood Institute/American Heart Association conference on scientific issues
related to definition. Circulation 2004;109(3):433–8.
[62] Smith SC Jr, Allen J, Blair SN, et al. AHA/ACC guidelines for secondary prevention for
patients with coronary and other atherosclerotic vascular disease: 2006 update endorsed
by the National Heart, Lung, and Blood Institute. J Am Coll Cardiol 2006;47:2130–9.
[63] Buse JB, Ginsberg HN, Bakris GL, et al, American Heart Association. American Diabetes
Association. Primary prevention of cardiovascular diseases in people with diabetes mellitus:
a scientific statement from the American Heart Association and the American Diabetes
Association. Circulation 2007;115(1):114–26.
[64] Grundy SM, Balady GJ, Criqui MH, et al. Primary prevention of coronary heart disease:
guidance from Framingham: a statement for healthcare professionals from the AHA Task
Force on Risk Reduction. American Heart Association. Circulation 1998;97(18):1876–87.
[65] Pearson TA, Blair SN, Daniels SR, et al. AHA guidelines for primary prevention of cardio-
vascular disease and stroke: 2002 update: consensus panel guide to comprehensive risk
reduction for adult patients without coronary or other atherosclerotic vascular diseases.
American Heart Association Science Advisory and Coordinating Committee. Circulation
2002;106:388–91.
[66] American Diabetes Association. Standards of medical care in diabetesd2007. Diabetes Care
2007;30:S4–41.
[67] Chobanian AV, Bakris GL, Black HR, 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.
[68] Lloyd-Jones DM, Leip EP, Larson MG, et al. Prediction of lifetime risk for cardiovascular
disease by risk factor burden at 50 years of age. Circulation 2006;113:791–8.
[69] Vasan RS, Sullivan LM, Wilson PW, et al. Relative importance of borderline and elevated
levels of coronary heart disease risk factors. Ann Intern Med 2005;142:393–402.
[70] Vasan RS, D’Agostino RB Sr. Age and time need not and should not be eliminated from the
coronary risk prediction models. Circulation 2005;111:542–5.
[71] Ridker PM, Cook N. Should age and time be eliminated from cardiovascular risk prediction
models? Rationale for the creation of a new national risk detection program. Circulation
2005;111:657–8.
[72] Hu FB, Stampfer MJ, Manson JE, et al. Trends in the incidence of coronary heart disease
and changes in diet and lifestyle in women. N Engl J Med 2000;343:530–7.
Med Clin N Am 91 (2007) 1185–1210
Treatment of the Dyslipidemia

of Insulin Resistance
Donald A. Smith, MD, MPH, FACP
Zena and Michael A. Wiener Cardiovascular Institute, Marie-Josee and
Henry R. Kravis Center for Cardiovascular Health, Mount Sinai School of Medicine,
Box 1014, 1 Gustave Levy Place, New York City, NY 10029–657, USA
The increase in ischemic cardiovascular mortality and morbidity in

persons with insulin resistance has been adequately documented elsewhere
in this issue. Some of the most important risk factors for this increased
mortality and morbidity are the lipid abnormalities, which form two out
of five criteria for defining the metabolic syndrome, a definition focusing
on risk factors for ischemic cardiovascular disease. This article explores
the description, pathophysiology, and treatment of these atherogenic lipid
abnormalities.
Diabetic dyslipidemia
Insulin resistance and type 2 diabetes are associated with what is com-
monly termed the ‘‘dyslipidemic triad’’: an increase in triglycerides, a
decrease in high-density lipoprotein (HDL) cholesterol, and increased
smaller, denser, low-density lipoprotein (LDL) particles, although LDL
cholesterol levels, measured by the Friedewald equation, in the standard
lipid profile are approximately the same. This has now been well docu-
mented by nuclear magnetic resonance (NMR) imaging of plasma lipopro-
teins [1] in the Framingham offspring population [2], the Multi-Ethnic Study
of Atherosclerosis (MESA) population [3], and in a smaller metabolic study
done in South Carolina on insulin-sensitive, resistant, and type 2 diabetic
individuals, the first two groups defined by hyperinsulinemic, hyperglycemic
clamps using glucose disposal rates (Table 1) [4]. Very low density lipopro-
tein (VLDL), LDL, and HDL particles are divided into small, intermediate,
and large sizes. As one goes from insulin sensitive, to insulin resistant, to
E-mail address: donald.smith@mssm.edu
1186 SMITH
Table 1
Conventional lipid profile measurements and nuclear magnetic resonance lipoprotein subclass
particle concentrations in insulin sensitive, insulin resistant, and type 2 diabetic patients [4]
IS IR Diabetes P IR P DM
N ¼ 56 N ¼ 46 N ¼ 46 versus IS versus IS
Standard lipid profile mg/dL (mmol/L)
Total cholesterol 168 (4.3) 183 (4.7) 195 (5) NS 0.004
LDL cholesterol 105 (2.7) 117 (3) 121 (3.1) NS 0.03
HDL cholesterol 43 (1.1) 40 (1) 41 (1) NS NS
Triglycerides 95 (1.07) 133 (1.49) 167 (1.88) 0.03 !0.0001
NMR lipoprotein subclass (nmol/L)
VLDL
Total 79.5 83.8 99.2 NS 0.03
Large 1.7 3.4 4.8 0.03 0.0002
LDL
Total 1201 1435 1593 0.006 !0.0001
Small 366 668 805 0.009 0.0002
Intermediate 238 213 385 NS 0.03
Large 589 534 385 NS 0.006
HDL
Large 6.9 5.3 4.7 0.03 NS
Small 14.2 15.2 17.3 NS 0.01
Abbreviations: DM, diabetes mellitus; HDL, high-density lipoprotein; IR, insulin resistant;
IS, insulin sensitive; LDL, low-density lipoprotein; NMR, nuclear magnetic resonance; VLDL,
very low density lipoprotein.
Only particles with significant differences among two or more groups are included. Signifi-
cant differences in bold type.
diabetic, there are steady increases in large VLDL and total and small LDL
particles. In going from insulin resistant to diabetic, there are significant
increases in total VLDL and intermediate LDL particles, with decreases
in large LDL particles. In the standard lipid profile one sees no change in
HDL cholesterol and increases in triglycerides of 35 mg/dL at each step
from insulin sensitive to resistant to diabetic. LDL cholesterol increases
15% from an insulin-sensitive to a diabetic lipid profile, while LDL particle
number increases 33%. This calls into question the ability of the standard
lipid profile accurately to predict the lipid risk in the insulin-resistant
patient. Similar changes in particle numbers are seen in the MESA [3] and
Framingham Offspring data [2].
As seen in the United Kingdom Prospective Diabetes Study, those who
have progressed from insulin resistance into new-onset adult-onset diabetes
have triglycerides approximately 60 mg/dL higher in both genders and HDL
cholesterol 5 mg/dL lower in men and over 10 mg/dL lower in women than
age-matched nondiabetic individuals [5]. Diabetic men compared with non-
diabetic men have 50% versus 25% small type B LDL particles [6], and
diabetic women compared with nondiabetic women have 36% versus 6%,
respectively [7].
TREATING THE DYSLIPIDEMIA OF INSULIN RESISTANCE 1187
Mechanism of dyslipidemia
In the insulin-resistant state, insulin-dependent lipase in the adipocyte is
less well inhibited by insulin resulting in increased lipolysis and release of
free fatty acids producing elevated serum levels [8,9]. These free fatty acids
are taken up by the hepatocyte, increasing the synthesis of triglycerides.
ApoB100 (the apoprotein found in VLDL, intermediate density lipopro-
teins, and LDL) is produced constantly in the hepatocyte but needs both
triglycerides and cholesterol to prevent its endoplasmic-reticulum-associated
degradation by proteosomes posttranslationally. The excess triglycerides
available with insulin resistance block this apolipoprotein B degradation
resulting in the production and secretion of more apolipoprotein B VLDL
particles and the consequent hypertriglyceridemia seen in insulin-resistant
and type-2-diabetic dyslipidemias.
The increased VLDL particles transfer more triglycerides to LDL and
HDL particles in exchange for cholesterol from LDL and HDL particles
by the enzyme cholesterol ester transfer protein. As the triglyceride-enriched
LDL and HDL particles pass through the liver, hepatic lipase (increased
activity with insulin resistance) hydrolyses the triglycerides to free fatty
acids, which diffuse from the particle, decreasing the core lipid content of
both LDL and HDL resulting in smaller particles. The catabolism of
HDL particles by the kidney is inversely proportional to the size of the
HDL particle, with smaller particles more rapidly catabolized resulting in
a drop in serum HDL cholesterol concentration. The smaller LDL particles
are more rapidly taken up into the vascular subendothelial space, more av-
idly associate with the proteoglycans in the subendothelial space, are more
rapidly oxidized than large LDL, and are ultimately more atherogenic [9,10].
In addition, because the cholesterol carried by LDL particles is carried by
smaller particles, LDL cholesterol as measured by the standard lipid profile
using the Friedewald equation [11] in diabetic populations is not elevated as
it would be if the LDL particle number were the same, and the size of each
LDL particle larger as in the usual noninsulin resistant population. Conse-
quently, standard LDL cholesterol, standardized as a percentile of the Fra-
mingham population, underestimates the population percentile for LDL as
measured by apolipoprotein B, or by LDL particle number using NMR
spectroscopy [12,13]. Standard LDL cholesterol, although measuring cho-
lesterol in other lipoproteins, is primarily a function of LDL particle number
and cholesterol content; and small decreases in the radius can cause large
decreases in particle volume (volume changes by the cube of the radius in
a sphere). An LDL cholesterol concentration of less than 100 mg/dL (twen-
tieth percentile for the Framingham concentration) is equivalent, by twenti-
eth population percentile, to 130 mg/dL of non-HDL cholesterol [14], 1100
nmol/L of LDL particles by NMR spectroscopy [13], or to 85 mg/dL of
apolipoprotein B (Table 2) [15–17]. In examining a population of 1784 adult
diabetic subjects who had concurrent NMR lipoprotein analysis and
1188 SMITH
Table 2
Apolipoprotein B–related lipid goals
Goals (population percentiles)
Test 50th 20th 5th Reference Population Study
LDL cholesterol mg/dL !130 !100 !70 Framingham Offspring [15]
Non-HDL cholesterol !160 !130 !100 National Cholesterol Education
mg/dL Program goals [35]
Apolipoprotein B mg/dL !105 !85 !65 Framingham Offspring [15]
!110 !90 !70 NHANES [16]
Suggested by Grundy SM [17]
Particle number (NMR) !1300 !1000 !700 MESA [3]
nmol/L !1400 !1100 !800 Framingham Offspring [13]
Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; NMR, nuclear
magnetic resonance.
standard lipid profile determinations in the commercial laboratory for

NMR lipid measurements [12], 74% of those with LDL cholesterol under
100 mg/dL by the standard lipid profile (twentieth population percentile)
had LDL particle numbers greater than 1000 nmol/L (twentieth population
percentile) with 33% over 1300 nmol/L (fiftieth population percentile). Of
821 adult diabetic patients with LDL cholesterol under 70 mg/dL (fifth pop-
ulation percentile), 84% had LDL particle numbers greater than 700 nmol/L
(fifth population percentile) with 32% over 1300 nmol/L (fiftieth population
percentile). In treating diabetic or insulin-resistant patients, clinicians may
be fooled that they are reducing LDL risk sufficiently when assuming they
are at goal by LDL cholesterol measured by the standard lipid profile.
Treatment for the dyslipidemia of insulin resistance

Hypoglycemic agents
Generally, hypoglycemic agents have not been given to insulin-resistant
patients with normal glucose levels even though metformin [18], acarbose
[19], and recently rosiglitazone [20] can prevent the onset of type 2 diabetes
by up to 60% using rosiglitazone. It is reasonable to ask what effects such
drugs might have on lipid levels.
Some insulin sensitizers have been shown to improve baseline lipid levels.
In a 6-month study of pioglitazone, 45 mg per day, metformin, 3 g per day,
and gliclazide (a sulfonylurea), 80 mg per day, in 20 type 2 diabetic patients
in each arm [21], there was no significant change in LDL cholesterol or tri-
glycerides, but an increase in HDL cholesterol in the pioglitazone group and
increased large LDL2 to small LDL3 in both the metformin and pioglitazone
groups, but not in the gliclazide group. Another study [22] of 24 weeks in
two groups of almost 400 middle-aged obese, type 2 diabetic patients with
baseline hemoglobin A1c 7.5%, triglycerides 250 mg/dL, HDL cholesterol
39 mg/dL, and LDL cholesterol 108 mg/dL compared the lipid effects of
pioglitazone, 45 mg per day, with rosiglitazone, 4 mg twice daily. Pioglita-

zone lowered triglycerides 52 mg/dL compared with 13 mg/dL with rosigli-
tazone, lowered LDL cholesterol 21 mg/dL compared with 12 mg/dL, and
increased HDL cholesterol 5 mg/dL compared with 2 mg/dL.
Do these beneficial lipid changes using metformin and thiazolidinediones
translate into preventive clinical benefits? Lacking long-term, cardiovascular
prevention studies in purely insulin-resistant patients without diabetes, one
can look to see the preventive effects of these agents in more severely resis-
tant, or diabetic, patients. In the United Kingdom Prospective Diabetes
Study [23], 342 overweight new-onset diabetic patients were randomly
assigned to metformin with 951 randomly assigned to chlorpropramide
(N ¼ 256), glibenclamide (N ¼ 277), or insulin (N ¼ 409) versus conven-
tional dietary therapy and followed for 10.7 years. For macrovascular dis-
ease events combined (myocardial infarction [MI], sudden death, angina,
stroke, and peripheral vascular disease), the metformin group had a relative
risk reduction of 30% (absolute risk reduction over 10 years of 1.4%, num-
ber needed to treat for 10 years ¼ 71) from the conventional treatment
group but did not differ significantly from the other intensive groups. Sim-
ilarly, for MI alone there was a significant 39% decrease compared with the
conventional treatment group, but no significant difference when compared
with the patients treated intensively with insulin or a sulfonylurea.
The Prospective Pioglitazone Clinical Trial in Macrovascular Events
PROactive Study [24] randomized pioglitazone or placebo to ongoing hypo-
glycemic therapy in 5238 obese patients with type 2 diabetes and previous
ischemic macrovascular disease with mean baseline hemoglobin A1c 7.8%,
LDL cholesterol 113 mg/dL, HDL cholesterol 43 mg/dL, and triglycerides
162 mg/dL. In the pioglitazone and placebo groups, respectively; triglycer-
ides were reduced 11% versus 2%, HDL cholesterol raised 19% versus
10%, and LDL cholesterol raised 7% versus 5% (not significant). At 34.5
months, the secondary end point of all-cause mortality, nonfatal MI, and
stroke was reduced by 16% (P ¼ .027) with an absolute risk reduction of
2.1%, number needed to treat for 3 years being 48. When silent MI, acute
coronary syndrome, coronary or peripheral revascularization, and leg am-
putation were included (the primary end point), the 10% relative risk reduc-
tion (2% absolute risk reduction) was not significant (P ¼ .95). These
preventive results of insulin-sensitizing drugs can be compared with the
more dramatic preventive effects of the lipid-altering drugs discussed next.
Statins
LDL are the most common atherogenic lipid particles in the dyslipidemia
of insulin resistance and type 2 diabetes, and 3-hydroxy-3-methylglutaryl
(HMG) coenzyme A reductase inhibitors (statins), which powerfully reduce
LDL, have generally become the initial drugs of choice for treatment. In ad-
dition, the more potent statins reduce triglycerides: simvastatin (12%–18%),
1190 SMITH
atorvastatin (20%–28%), and rosuvastatin (20%–26%) [25]. All statins in-

crease HDL cholesterol from 3% to 7%, with rosuvastatin producing in-
creases from 7.5% to 9.5% [25]. Most clinical trials have focused on
diabetic subgroups and the best that can be done for persons with insulin
resistance but without the hyperglycemia of type 2 diabetes is to explore
these diabetic subgroups and insulin-resistant subgroups where available.
A meta-analysis done for the American College of Physicians on lipid-
lowering therapy in type 2 diabetes (Table 3) [26] included many primary
and secondary prevention trials using statins on a wide variety of patients.
Diabetic subgroup analyses in some of the trials (Heart Protection Study,
primary and secondary prevention arms, CARE, 4S, LIPS) showed statisti-
cally significant preventive effects, whereas analyses in others (AFCAPS/
TexCAPS, PROSPER, ASCOT-LLA, LIPID, and Post-CABG) did not.
The meta-analysis with vastly increased numbers of participants, however,
showed a primary preventive relative risk reduction of 22% (relative risk,
0.78; 95% confidence interval [CI], 0.67–0.89) and absolute risk reduction
of 3% over 4.3 years of treatment. Secondary prevention produced a similar
relative risk reduction of 24% (relative risk, 0.76; 95% CI, 0.59–0.93) but
a doubling of the absolute risk reduction to 7% over 4.9 years of treatment,
showing the relatively more powerful public-health effect in higher-risk
patients over these relatively short testing periods.
The West of Scotland Coronary Prevention Study [27] was one primary
prevention trial where a subanalysis in men with the insulin resistance
syndrome was performed [28]. A total of 6595 men in Scotland age 45 to
64 years and with an average baseline LDL cholesterol of 192 mg/dL
were randomized to pravastatin, 40 mg daily, versus placebo and followed
for 4.9 years. In the total group there was a 31% reduction in MI and cor-
onary heart disease (CHD) death, with a similar 27% reduction in the 1691
men with the metabolic syndrome (using body mass index O28.8 kg/m2
rather than increased waist circumference). Study risk for these events in
the placebo group was 6.2% in those neither with metabolic syndrome
nor diabetes, 10.4% in those with metabolic syndrome and no diabetes,
and 17.6% in those with diabetes. The addition of the metabolic syndrome
was equivalent in predicting risk to being 10 years older, or to being
a smoker. The presence of the metabolic syndrome added independently
to the CHD risk calculated using standard risk factors of age, smoking sta-
tus, systolic blood pressure, and total/HDL cholesterol ratio. When all the
individual components of the metabolic syndrome were included in the mul-
tivariate model (body mass index, fasting glucose, diastolic blood pressure,
and triglycerides), metabolic syndrome no longer independently predicted
risk.
A subsequent major primary prevention trial, the Collaborative Atorvas-
tatin Diabetes Study (CARDS) [29], enrolled only diabetic patients, 40 to 75
years old, with at least one other risk factor (retinopathy, hypertension, cur-
rent smoking, or albuminuria). Baseline LDL cholesterol had to be less than
160 mg/dL, and triglycerides less than 600 mg/dL. Subjects were random-
ized to atorvastatin 10 mg daily versus placebo with a primary end point
of time to first occurrence of an acute CHD event, coronary revasculariza-
tion, or stroke. Baseline LDL cholesterol of 118 mg/dL was lowered in the
atorvastatin group to 84 mg/dL and increased to 124 mg/dL in the placebo
group by year 4. The trial was stopped 2 years early with a 3.9-year 37%
relative risk reduction (relative risk, 0.63; 95% CI, 0.48–0.83) and absolute
risk reduction of 3.7%, similar to the 3% absolute risk reduction at 4.3
years seen in the American College of Physicians meta-analysis presented
previously.
A more recent meta-analysis of LDL-cholesterol lowering by statins in
90,056 participants in 14 primary and secondary, statin and placebo-ran-
domized clinical trials, including CARDS, performed by the Cholesterol
Treatment Trialists’ Collaboration [30] has shown the same 20% reduction
in major cardiovascular events (major coronary event, stroke, or coronary
revascularization) per 1 mmol (approximately 40 mg/dL) reduction in
LDL cholesterol in those with and without diabetes, and importantly in
those with baseline LDL cholesterol levels less than 100 mg/dL (2.6
mmol/L). In absolute terms, 1 mmol/L reduction in LDL cholesterol
(approximately 40 mg/dL) resulted in 4.8% fewer major cardiovascular
events in those with CHD and 2.5% fewer in those without CHD over
a 5-year period. In comparison, the absolute rate reduction in diabetic
patients without CHD in the CARDS trials was 3.7%, consistent with their
increased baseline risk for cardiovascular events.
The Scandinavian Simvastatin Survival Study was a secondary preven-
tion study with a subanalysis of those with the lipid profile of the metabolic
syndrome [31]. It compared simvastatin, 20 to 40 mg daily, versus placebo
for 5.4 years in men and women ages 35 to 70 years with CHD who had to-
tal cholesterol levels between 210 and 310 mg/dL. Overall, there was a 34%
reduction in risk for MI or CHD death with a 9.1% absolute risk reduction.
The quartile of the study population with HDL cholesterol less than 39
(1 mmol/L) and triglycerides greater than 159 mg/dL (1.8 mmol/L), cut-
points very similar to those for the dyslipidemia of the metabolic syndrome,
had a 52% relative risk reduction with a 17% absolute risk reduction. When
those with diabetes were removed from this group, this 17% absolute risk
reduction was the same (see Table 3). This important 4S subanalysis demon-
strated that in this trial statins had an exaggerated effect on risk reduction in
those with the lipid profile of the metabolic syndrome.
Several trials of more intensive versus moderate LDL-lowering have been
published including Treating to New Targets (TNT), Incremental Decrease
in End Points Through Aggressive Lipid-Lowering (IDEAL), Pravastatin or
Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocar-
dial Infarction-22 (PROVE-IT), and Aggrastat-to-Zocor (A-to-Z), the first
two in patients with stable CHD, the latter two in patients with acute cor-
onary syndromes. A meta-analysis of these four trials [32] comparing usual
1192
Table 3
Clinical trials of statin therapy for dyslipidemia in patients with diabetes
Study Cohort Drug N Length (y) Events RRR ARR NNT
Primary prevention
CTT [30] Combineda Statins versus placebo 90,056 5 MI/CHD death 23 1.8 55
Diabetes Statins versus placebo 18,686 5 MI/CHD death 26 1.7 58
CTT Combineda Statins versus placebo 90,056 5 MI/CHD death/CVA/ 21 2.5 40
revascularization
Diabetes Statins versus placebo 18,686 5 MI/CHD death/CVA/ 25 2.7 37
revascularization
ACP [26] Diabetes Statins versus placebo 10,930 (135) 4.3 Mixed primary end points, 22 3 33
(includes gemfibrozil usually MI/CHD death/
versus placebo) CVA/revascularization
CARDS [29] Diabetes (higher risk) Atorvastatin, 10 mg, versus 2838 3.9 MI/CHD death 33 1.6 62
placebo
SMITH
d d MI/CHD death/CVA/ 37 3.7 27
revascularization/
hospitalization for
unstable angina
WOSCOPS [27,28] Combineda Pravastatin, 40 mg, versus 6595 4.9 MI/CHD death 31 2.4 42
placebo
Metabolic syndrome Pravastatin, 40 mg, versus 1691 4.9 MI/CHD death 27 2.7 43
placebo
Secondary prevention
4S [31] Combineda Simvastatin, 20–40 mg, 4444 5.4 MI/CHD death 34 9.1 11
versus placebo
HDL-C !39 þ TG O159 Simvastatin, 20–40 mg, 458 5.4 MI/CHD death 52 17 6
mg/dL, versus placebo
includes DM
HDL-C !39 þ TG O159 Simvastatin, 20–40 mg, 385 5.4 MI/CHD death 49 17.4 6
mg/dL, versus placebo
without DM
CTT [30] Combineda Statins versus placebo 90,056 5 MI/CHD death 23 3 43
Diabetes Statins versus placebo 18,686 5 MI/CHD death NA NA NA
CTT Combineda Statins versus placebo 90,056 5 MI/CHD death/CVA/ 21 4.8 20
revascularization
Diabetes Statins versus placebo 18,686 5 MI/CHD death/CVA/ NA NA NA
revascularization
ACP [26] Diabetes Statins versus placebo 10,930 (135) 4.9 Mixed primary end points, 24 7 14
(includes gemfibrozil usually MI/CHD death/
TREATING THE DYSLIPIDEMIA OF INSULIN RESISTANCE

versus placebo) CVA/revascularization
TNT [62] No diabetes Atorvastatin, 80 versus 10 8500 4.9 MI/CHD death/CVA 18 1.9 53
mg
Diabetes Atorvastatin, 80 versus 10 1501 4.9 MI/CHD death/CVA 25 4 25
mg
Metabolic syndrome Atorvastatin, 80 versus 10 5584 4.9 MI/CHD death/CVA 29 3.5 29
(includes diabetes) mg
Metabolic syndrome Atorvastatin, 80 versus 10 4353 4.9 MI/CHD death/CVA 30 3.4 29
(without diabetes) mg
Meta analysis [32] Combineda
TNT Atorvastatin, 80 mg versus d 2
10 mg
IDEAL Atorvastatin, 80 mg, versus d 2 CHD death/MI 16 1.4 71
simvastatin 40 mg
PROVE-IT Atorvastatin, 80 mg, versus d 4.9 CHD death/MI/ 16 3.5 29
simvastatin, 40 mg revascularization/CVA/
hospitalization for
unstable angina
A-to-Z Simvastatin, 80 mg versus d 4.8
20 mg
(continued on next page)
1193
1194
Table 3 (continued )
Combined primary and secondary prevention
CTT [30] No diabetes Statins versus placebo 71,370 5 MI/CHD death 23 2.4 41
Diabetes 18,686 5 MI/CHD death 22 2.2 45
No diabetes Statins versus placebo 71,370 5 MI/CHD death/CVA/ 21 3.7 27
revascularization
SMITH
Diabetes 18,686 5 MI/CHD death/CVA/ 21 3.6 27
revascularization
a
Includes those with and without diabetes.
Abbreviations: A-to-Z, Aggrastat-to-Zocor; AAR, absolute risk reduction over length of study; ACP, American College of Physicians Meta-analysis;
CARDS, Collaborative Atorvastatin Diabetes Study; CHD, coronary heart disease; CTT, Cholesterol Treatment Trialists; CVA, cerebrovascular
accident; DM, diabetes mellitus; HDL, high-density lipoprotein; IDEAL, Incremental Decrease in End Points Through Aggressive Lipid-lowering; NA,
not reported; NNT, number need to treat; MI, myocardial infarction; PROVE-IT, Pravastatin or Atorvastatin Evaluation and Infection Therapy; RRR, rel-
ative risk reduction, all statistically significant unless otherwise indicated; TNT, Treating to New Targets.
lower-dose statin therapy with highest-dose statin therapy produced a differ-

ence in LDL cholesterol of 101 mg/dL versus 75 mg/dL, a 26% further re-
duction in LDL cholesterol. CHD death and MI were reduced 16% (odds
ratio, 0.84; 95% CI, 0.77–0.91), absolute risk reduction 1.4%. When coro-
nary revascularization, stroke, and hospitalization for unstable angina
were combined with CHD death and MI, the relative risk reduction re-
mained 16% (odds ratio, 0.84; 95% CI, 0.80–0.89), but the absolute risk re-
duction increased to 3.5%. Overall mortality trended downward by 6%, but
there were not enough deaths to show statistical significance. Combining the
25% risk reduction to be expected from a low-moderate-dose statin and the
additional effects of a higher-dose or more powerful statin, cardiovascular
events were lowered approximately 40% (odds ratio odds ratio ¼ 0.75
0.84 ¼ 0.63) or a 1% relative risk reduction for each 1.8 mg/dL reduction
in LDL cholesterol. Each of these trials had diabetic patients, but a subanal-
ysis focusing on them was not performed.
A subanalysis was performed, however, of persons with the metabolic
syndrome in the TNT study [33]. Approximately 20,000 patients with
CHD were treated with 10 versus 80 mg per day of atorvastatin for
a mean follow-up of 4.9 years. Of these, approximately 5600 had the meta-
bolic syndrome (using body mass index O28 kg/m2 to replace increased
waist circumference, which had not been measured). Included in these
5600 insulin-resistant subjects were 22% who already had diabetes. Subjects
with diabetes and the metabolic syndrome had a higher incidence of the
principal end point of MI, CHD death, resuscitated cardiac arrest, and
stroke (11.3%) than those with the metabolic syndrome alone (9.9%),
than did those with neither (7.5%). In those with the metabolic syndrome
but no diabetes, primary end points were reduced 30% (11.6 versus 8.2%)
with high- versus low-dose atorvastatin (hazards ratio, 0.70; 95% CI,
0.57–0.84). LDL cholesterol differences were 99 versus 73 mg/dL, and tri-
glyceride differences 175 versus 148 mg/dL. When examining the total non-
diabetic population, risk reductions with higher-dose atorvastatin started
occurring when a subject had three or more components of the metabolic
syndrome, but not with fewer.
Using such data, an LDL cholesterol less than 70 mg/dL in diabetic
patients with CHD was an ‘‘optional’’ goal in the 2004 National Cholesterol
Education Program (NCEP) update [34] and more recently has been
upgraded to a ‘‘reasonable’’ recommendation in the 2006 guidelines of the
American Heart Association and the American College of Cardiology,
and endorsed by the National Heart, Blood and Lung Institute [35]. The
TNT results imply the same for persons with CHD and insulin resistance.
This was endorsed by the American Heart Association and National Heart,
Lung, and Blood Institute in 2005 in which very high risk patients had an
optional LDL cholesterol goal of less than 70 mg/dL [36]. Qualifying for
such status are persons with CHD and three or more (multiple) risk factors
of the metabolic syndrome.
1196 SMITH
The NCEP guidelines for LDL cholesterol goal in those with diabetes but
without CHD remains at less than 100 mg/dL, a goal also promoted by the
American Heart Association and National Heart, Lung, and Blood Institute
[36]. If a patient with the metabolic syndrome has a greater than 20%
10-year risk for CHD by the Framingham equation, the LDL cholesterol
goal is also less than 100 mg/dL. The reduction of LDL cholesterol from
124 to 84 mg/dL producing primary-preventive effects in the CARDS trial
has been the major clinical evidence supporting this goal.
There are young persons with three criteria for the metabolic syndrome
whose risk may be in the low (!10% 10-year risk) or low-intermediate
range (10%–15% 10-year risk), whose LDL cholesterol goal at this point
should still simply be less than 130 mg/dL (rather than !100 mg/dL).
One can simply think of the presence of the metabolic syndrome in such per-
sons as an additional risk factor equivalent to smoking, or an age 10 years
older, as was demonstrated in the WOSCOPS primary prevention trial, and
decide on an LDL cholesterol goal using NCEP criteria with an additional
risk factor added. For example, a man under 45 years old without hyperten-
sion or a positive family history of early CHD but with three metabolic syn-
drome criteria of an HDL cholesterol less than 40 mg/dL, triglycerides
greater than 150 mg/dL, and a waist circumference greater than 40 in, is
considered to have two criteria for determining LDL cholesterol goal, not
just a low HDL cholesterol but also the metabolic syndrome. His LDL cho-
lesterol goal drops from less than 160 mg/dL using the one NCEP criteria,
to less than 130 mg/dL using one NCEP criteria plus the presence of the
metabolic syndrome.
Fibric acids
Fibric acids affect many aspects of diabetic dyslipidemia [37]. They in-
crease the activity of lipoprotein lipase and suppress inhibition of lipoprotein
lipase by decreasing levels of ApoCIII, its inhibitor, resulting in increased ca-
tabolism of lipoproteins carrying triglycerides, and consequent lowering of
serum triglyceride levels. They increase production of ApoA1 increasing
HDL cholesterol levels and by increasing the activity of adenosine binding
cassette protein receptor 1 (ABCA1) in peripheral cells increase the reverse
cholesterol transport from the periphery to the liver for biliary excretion.
They decrease triglyceride production in the liver, decreasing VLDL produc-
tion with consequent decreases in serum triglycerides and in LDL cholesterol.
By decreasing VLDL production there is less exchange of triglycerides for
cholesterol in LDL and HDL with consequent increase in the sizes of LDL
and HDL, decreased catabolism of HDL, and consequent increases in HDL
cholesterol. Such changes improve lipid profiles of insulin-resistant and type
2 diabetic patients and decrease clinical cardiovascular events.
The Veterans Administration HDL Intervention Trial (see Table 4) was
the most recent trial with the most suggestive results for use of fibric acids
in insulin-resistant diabetic patients [38]. In 1991, 2531 male veterans were

randomly assigned to gemfibrozil, 600 mg twice a day, versus placebo and
followed for 5.1 years. Lipid inclusion criteria were an HDL cholesterol
less than or equal to 40 mg/dL, LDL cholesterol less than or equal to
140, and triglycerides less than or equal to 300. These middle-aged men,
average age 64 years, had a mean body mass index of 29 kg/m2 with a waist
to thigh circumference of 0.96, at the eightieth percentile for men. Not sur-
prisingly, 25% had a history of diabetes and 57% had hypertension. There
was a 22% relative risk reduction in CHD death and MI (P ¼ .006) and
a 4.4% absolute risk reduction in overall subjects. In a subanalysis [39] of
those with baseline diabetes, defined by history or fasting plasma glucose
greater than 126 mg/dL, versus those without diabetes, relative risk reduc-
tion of MI, CHD death, and stroke was 32% versus 18% with absolute
risk reductions of 9.9% versus 3.3%, respectively (Table 4).
A subanalysis of 1732 men [39] without diabetes, and who had measured
baseline fasting insulin levels, were divided into quartiles with increasing
insulin levels indicating increasing insulin resistance. Those in the upper
quartile of fasting insulin concentration, or most insulin resistant, were
the only group with a statistically significant relative decrease in CHD
death, MI, and stroke of 35%, with a trend to increasing preventive effect
from the lowest to the highest quartile of insulin resistance.
This study was reminiscent of the primary prevention Helsinki Heart
Trial using gemfibrozil, 600 mg twice a day, in 4100 men, age 40 to 65,
non-HDL cholesterol greater than 200, without CHD [40]. The overall trial
showed a reduction in CHD death and MI of 34% (relative risk, 0.66; 95%
CI, 0.47–0.93; P !.02). In a subanalysis [41] of those in the placebo group,
those with LDL/HDL ratio greater than 5 plus triglycerides greater than 200
mg/dL had a relative risk of 3.8 for MI and CHD death compared with
those with one or both values lower. It was in this group with triglycerides
greater than 200 mg/dL and LDL/HDL ratios greater than 5 at baseline that
gemfibrozil decreased CHD events by 71%, significantly more than in any
other group. In a group with triglycerides greater than 170 mg/dL and
LDL/HDL ratios greater than 4.5, gemfibrozil decreased CHD events by
53%. In retrospect, these groups were most likely insulin resistant with
higher triglycerides and lower HDL cholesterol levels and helps define those
in whom fibric acids might be most preventive.
The most recent clinical trial of a fibric acid was the Fenofibrate Interven-
tion and Event Lowering in Diabetes (FIELD) study [42]. A total of 9800
type 2 diabetic subjects, age 40 to 75 years, and with total cholesterol 115
to 250 mg/dL and either total cholesterol/HDL cholesterol ratio greater
than or equal to 4, or triglycerides 89 to 443 mg/dL were enrolled. One quar-
ter had a history of CHD, three quarters had no such history. Only 37% had
the two qualifying lipid criteria for the insulin-resistance or metabolic syn-
drome: triglycerides greater than 150 mg/dL, and HDL cholesterol less
than 40 mg/dL in men or less than 50 mg/dL in women. Participants could
1198
Table 4
Clinical trials of fibric acid therapy for dyslipidemia in patients with diabetes
Primary prevention
HHS [40,41] Combineda Gemfibrozil, 1200 mg, 4081 5 MI/CHD death 27 1.7 58
versus placebo
Combineda: TG R200 Gemfibrozil, 1200 mg, 408 5 MI/CHD death 71 9.1 10
mg/dL;LDL/HDL versus placebo
O5
Combineda: TG O170 Gemfibrozil, 1200 mg, 694 5 MI/CHD death 53 NA NA
mg/dLLDL/HDL versus placebo
O4.5
FIELD [42] Diabetes Fenofibrate, 160 mg, 7664 5 MI/CHD death 25 1.9 52
versus placebo MI/CHD death/CVA/ 19 2 50
SMITH
revascularization/
total death
VA-HIT [38,39] Combineda Gemfibrozil, 1200 mg, 2531 5.1 MI/CHD death 22 4.4 23
versus placebo
No diabetes Gemfibrozil, 1200 mg, 1748 5.1 MI/CHD death/CVA 18 3.3 28
versus placebo
Diabetes Gemfibrozil, 1200 mg, 769 5.1 MI/CHD death/CVA 32 9.9 10
versus placebo
No diabetes, highest Gemfibrozil, 1200 mg, 431 5.1 MI/CHD death/CVA 35 NA NA
quartile fasting versus placebo
insulin, most insulin
resistant
FIELD [42] Diabetes Fenofibrate, 160 mg, 2131 5 MI/CHD death þ8 (NS) d d
versus placebo MI/CHD death/CVA/ þ2 (NS) d d
revascularization/
CVD death
BIP [44] Combineda Bezafibrate, 400 mg, 3090 6.2 MI/CHD death 7 (NS) 1.4 (NS) d
versus placebo
TG R200 mg/dL 459 6.2 MI/CHD death 40 7.7 12
Metabolic syndrome, 1470 6.2 MI/CHD death 25 4.3 23
3 components
Metabolic syndrome, 575 6.2 MI only 35 5.6 17
4–5 components
d d Cardiac death 56 6.6 15
TREATING THE DYSLIPIDEMIA OF INSULIN RESISTANCE

Combined primary/secondary prevention
FIELD [42] Diabetes Fenofibrate, 160 mg, 9795 5 MI/CHD death 11 (NS) 9 (Adj, d d
versus placebo P ¼ .27, NS)
Diabetes Fenofibrate, 160 mg, 9795 5 MI/CHD death/CVA/ 11 1.4 71
versus placebo revascularization/
CVD death
Diabetes þ TG O150 Fenofibrate, 160 mg, 3710 5 MI/CHD death/CVA/ 14 (P ¼ .06, NS) d d
mg/dL þ HDL !40 versus placebo revascularization/
mg/dL in men, !50 CVD death
mg/dL in women
a
Abbreviations: AAR, absolute risk reduction over length of trial; Adj, adjusted for use of statins on-trial; BIP, Bezafibrate Infarction Prevention Study;
CHD, coronary heart disease; CVA, cerebrovascular accident; CVD death, cardiovascular death; FIELD, Fenofibrate Intervention and Lowering in Diabetes;
HDL, high-density lipoprotein; HHS, Helsinki Heart Study; LDL, low-density lipoprotein; MI, myocardial infarction; NA, not reported; NNT, number need
to treat; NS, not significant; RRR, relative risk reduction, all statistically significant unless otherwise indicated; TG, triglycerides; VA-HIT, Veterans Affairs
High-Density Lipoprotein Intervention Trial.
1199
1200 SMITH
not be on lipid-altering therapy at baseline, but there was no restriction on

lipid-lowering medications that could be added after the trial started. The
11% reduction in the primary end point of MI and CHD death was not sig-
nificant largely because by the end of the study 36% of those on placebo but
only 19% of those on fenofibrate had electively and independently started
statin therapy. When the end point was first adjusted for the differential
on-trial statin use, the adjusted primary end point was decreased 19%
(P ¼ .01). A later time-dependent, rather than fixed, adjustment for statin
use, however, showed only a nonsignificant 9% reduction in CHD events
(P ¼ .27) [43]. Using the same time-dependent adjustment for statin use,
the original 11% statistically significant decrease in all cardiovascular events
was reduced to a nonsignificant 4% (P ¼ .45). As seen in the Helsinki Trial,
the unadjusted primary end point was reduced slightly more in the group
with triglycerides greater than 150 mg/dL and HDL cholesterol less than
40 mg/dL in men and less than 50 mg/dL in women (14% relative risk re-
duction; P ¼ .06). In the subjects with diabetes by history but without pre-
vious CHD (see Table 4), there was a significant 19% relative risk reduction
in total cardiovascular events and a 2% absolute risk reduction. Surpris-
ingly, no preventive benefit was seen in those with previous CHD, perhaps
because of a greater use of elective statins in the placebo group in these men
at higher risk.
A large clinical trial with a fibrate used in Europe, the Bezafibrate Infarc-
tion Prevention Study [44], was a secondary prevention trial in 3090 subjects
with HDL cholesterol less than or equal to 45 mg/dL, triglycerides less than
or equal to 300 mg/dL, and LDL cholesterol less than or equal to 180 mg/
dL randomized to receive either 400 mg bezafibrate per day or a placebo and
followed for a mean of 6.2 years. The reduction in the primary end point of
MI or sudden death was 7.3% and was not statistically significant. In a post
hoc subanalysis of those with triglycerides greater than or equal to 200 mg/
dL, however, there was a 39.5% reduction in this primary end point (P ¼
.02) suggesting the beneficial effects exist primarily in those with higher tri-
glycerides. In similar subanalyses in those with three or more components of
the metabolic syndrome, and especially in those with four or five compo-
nents, significant relative risk reductions of 25% to 40% occurred (see
Table 4).
Given the reductions in events in the statin-stained FIELD study, the
very positive results in the Veterans Affairs High-Density Lipoprotein Inter-
vention Trial in those with increasing insulin resistance, the remarkable ef-
fects in the Helsinki Trial in those with triglycerides greater than 200 mg/L
and LDL/HDL ratio greater than 5, and the positive results in the Bezafi-
brate Infarction Prevention Study in those with triglycerides 200 mg/L or
over, or with three or more components of the metabolic syndrome, these
studies suggest that there are definite and measurable cardiovascular preven-
tive effects with fibric acids in diabetic patients and those with insulin resis-
tance. The preventive benefits are always larger, and perhaps only become
significant, in those with higher baseline triglycerides and lower HDL cho-
lesterol levels. The sum of these effects in those with triglycerides greater
than 200 mg/dL and especially accompanied by HDL cholesterol of less
than 40 in men, and less than 50 in women, is as dramatic as the effects
of statins in persons with diabetes reviewed previously. Although statins re-
main the first choice in those with insulin resistance or diabetes, in those pa-
tients with the dyslipidemia of the metabolic syndrome with raised
triglycerides and low HDL cholesterol, these studies provide evidence that
treatment with a fibric acid as the first lipid-altering drug in those with tri-
glycerides greater than 200 mg/dL and low HDL cholesterol levels can give
equivalent preventive effects.
Niacin
One obvious focus in treating the dyslipidemia of diabetes is to raise the
lowered HDL cholesterol. Although lifestyle changes, fibric acids, and the
more potent statins raise HDL cholesterol, the most potent pharmacologic
means of increasing HDL cholesterol is niacin, previously called nicotinic
acid or vitamin B3. HDL cholesterol is raised in a dose- and time-dependent
manner with continuous increases seen up to 2 g per day [45] and with in-
termediate-release niacin up to 2.5 g per day [46]. Niacin formulation is
very important with inositol preparations (nonflush niacin) producing no
lipid changes because of inadequacy of absorption [47]. Slow, time-release
niacin produces more increases in hepatic enzymes and less of an increase
in HDL cholesterol or lowering of triglycerides than crystalline niacin
[48]. Intermediate-release niacin has HDL cholesterol effects equivalent to
immediate-release niacin with fewer side effects [49]. In another comparative
study, intermediate niacin at a dose of 1500 mg per day converted 56% of
individuals with small, pattern-B LDL cholesterol to large, pattern-A cho-
lesterol, whereas immediate-release niacin at 3000 mg per day converted
93% of such subjects [50]. Intermediate-release niacin increases HDL cho-
lesterol and decreases LDL cholesterol up to a dose of 2500 mg per day
at which these effects plateau, although triglycerides are lowered further
at higher doses [46]. Niacin in large doses, such as 4500 mg per day, pro-
duces significant increases in glucose levels [51], which is why for many years
its use was cautioned in those with diabetes. More recent studies, however,
show smaller transient elevations in fasting glucose, and generally small or
no increases (þ0.2%–þ0.3%) in hemoglobin A1c when used in 1- to 3-g
quantities per day [52,53]. An occasional patient may still have to have hy-
poglycemic medicines increased or stop niacin because of major changes in
glucose control [52]. Uric acid levels may increase 10% to 15% [53] and one
must warn patients with high baseline serum uric acid levels or a history of
gout of the possibility of an acute gouty attack.
The Coronary Drug Project (Table 5), a secondary prevention study in
those with CHD, demonstrated a 5-year 27% decrease in nonfatal MI
1202
Table 5
Clinical trials of niacin therapy for dyslipidemia in patients with diabetes
CDP [55,56] Combineda Niacin, 3 g, versus placebo 3908 5 MI 27 3.3 30
SMITH
Baseline FPG R126 mg/dL Niacin, 3 g, versus placebo d 5 MI 57 8.4 12b
Combineda Niacin, 3 g, versus placebo 3908 15 Total mortality 11 6.2 16
FPG R126 mg/dL Niacin, 3 g, versus placebo d 15 Total mortality 17 6.5 15
a
b
Not significantly different from combined results.
Abbreviations: AAR, absolute risk reduction over length of study; CDP, Coronary Drug Project; FPG, fasting plasma glucose; MI, myocardial infarction;
NNT, number needed to treat; RRR, relative risk reduction, all statistically significant unless otherwise indicated.
and a 15-year 11% reduction in total mortality in those using 3 g of imme-

diate-release, or crystalline, niacin per day [54,55]. A recent substudy [56]
showed that although the niacin group at 1 year had fasting plasma glucose
5 mg/dL greater than the placebo group and a 1-hour post–glucose-chal-
lenge plasma glucose 15 mg/dL greater, the decrease in relative risk, or
the preventive effect, was the same. This same benefit was seen even in those
with fasting glucoses greater than 126 mg/dL, which defines diabetes today
(see Table 5). In addition, the preventive effects were the same no matter
how great the glucose increase from baseline to year 1 in the fasting or
1-hour, postchallenge glucose levels.
The HDL-Atherosclerosis Treatment Study (HATS) [57] was a 3-year,
double-blind quantitative coronary angiography imaging study with one
group randomized to simvastatin in increasing doses to achieve an LDL
cholesterol less than 90 mg/dL, and increasing doses of immediate-release
niacin, to increase HDL cholesterol by 10 mg/dL at year 1. Comparative
values of on-trial lipids between simvastatin-niacin and placebo groups
were as follows: LDL cholesterol 77 versus 116 mg/dL, HDL cholesterol
38 versus 34 mg/dL, and triglycerides 146 versus 196 mg/dL. The group
on the simvastatin-niacin combination therapy progressed only 0.7% com-
pared with a 3.9% progression with placebo (P !.001). Although not pow-
ered as a clinical event reduction trial, there was a 60% reduction in
coronary clinical events, including coronary revascularization, in the sim-
vastatin-niacin group.
Two newer placebo-controlled, double-blind imaging studies (ARBITER
2 and ARBITER 3) [58,59] using carotid intimal medial thickness have
explored the addition of intermediate-release niacin, 1000 mg at night, to
ongoing statin therapy versus continuing the statin therapy alone in 160 sub-
jects with CHD (90% men, 28% with diabetes, 75% with hypertension).
Simvastatin greater than or equal to 20 mg per night was baseline statin
therapy with LDL cholesterol levels in the two groups of 87 and 91 mg/dL
and at goal. In ARBITER 2 at 1 year, the 1-g, intermediate-release niacin
had raised HDL cholesterol from 39 mg/dL to 47 mg/dL in the niacin-statin
arm, producing no progression in that group (0.014 mm/y, P ¼ .23,) com-
pared with progression in those on statin alone (0.044 mm/y, P !.001).
The study showed no difference in clinical events.
In ARBITER 3 [59], the subjects who had been on statin and niacin pla-
cebo could switch to 1 g of intermediate-release niacin, whereas those on in-
termediate-release niacin could continue it another 12 months. Over the 24
months average baseline triglycerides decreased from 165 to 130 mg/dL
(P !.001); LDL cholesterol decreased from 88 to 79 mg/dL (P !.001);
and HDL cholesterol increased from 40 to 49 mg/dL (P !.001) with no sig-
nificant increase in fasting glucose. With these lipid changes, at 1 year ca-
rotid intimal medial thickness in those taking statin therapy alone
progressed 0.52 mm, whereas those on 1 g of intermediate-release niacin
plus continuing statin regressed 0.027 mm (P !.001). At 24 months there
1204 SMITH
was further regression of 0.041 mm (P !.001) with the added niacin com-
pared with placebo. The only lipid change independently associated with re-
gression was an increase in HDL cholesterol.
This is one of the rare clinical studies comparing statin therapy alone ver-
sus statin therapy in combination with another agent looking at CHD imag-
ing or events. Statin therapy and fibric acids together produce better lipid
results than statins alone. In the 18-week SAFARI study, simvastatin, 20
mg, plus fenofibrate, 160 mg, versus simvastatin, 20 mg, alone doubled
the 20% reduction in triglycerides to 43%, doubled the 10% increase in
HDL cholesterol to 19%, and increased the 26% reduction in LDL choles-
terol to 31% with no adverse side effects [60]. Intermediate niacin, 2000 mg
per day, added to lovastatin, 40 mg per day, produced a 47% decrease in
LDL cholesterol, 41% decrease in triglycerides, and 30% increase in
HDL cholesterol rising to 41% by the end of 1 year [61]. One would predict
that these advantageous changes in the lipid profile with combination ther-
apy would predict better imaging and clinical outcomes in trials now
underway.
The Action to Control Cardiovascular Risk in Diabetes (ACCORD)
trial, which will compare simvastatin, 40 mg, alone versus simvastatin,
40 mg, plus fenofibrate is ongoing and should document the magnitude
of such preventive benefits but will only be published in 2009 to 2010.
AIM-HIGH is a 4-year, multicenter, secondary prevention trial of simvasta-
tin plus intermediate-release niacin versus simvastatin alone in persons with
triglycerides greater than 150 mg/dL and HDL cholesterol lower than
40 mg/dL; this study has just finished recruitment.
For persons with insulin resistance and CHD, or with several other car-
diovascular risk factors, whose LDL cholesterol reaches a level significantly
below 100 mg/dL, but whose triglycerides remain greater than 200 mg/dL,
the NCEP ATP III suggests the option of combination therapy. Certainly,
such combinations could be justified in those with dyslipidemia of insulin
resistance who have evidence of progression of coronary disease on one
lipid-altering agent.
Monitoring results of lipid therapy

In the presence of higher triglycerides and lower HDL cholesterol levels,
there may be other ways of targeting atherogenic lipoproteins rather than
just through LDL cholesterol, calculated by the Friedewald equation in
the standard lipid profile. LDL cholesterol values have less accuracy using
this methodology and equation as triglycerides exceed 200 mg/dL and the
method should not be used in those with triglycerides exceeding 400 mg/dL.
Direct methods of measuring LDL cholesterol in such situations are
available but do not provide information on LDL size or particle number,
or on other atherogenic lipoproteins present in the insulin-resistant state,
such as intermediate-density lipoproteins and more importantly small rem-

nant VLDL cholesterol. For those interested, there are several methods of
evaluating risk beyond LDL cholesterol concentration:
1. Non-HDL cholesterol (standard lipid profile)
2. Apolipoprotein B
3. Vertical density gradient ultracentrifugation
4. Segmented gradient gel electrophoresis
5. Nuclear magnetic resonance
Vertical ultracentrifugation, gradient gel electrophoresis, and NMR all
indicate size of the predominant LDL particle peak. In most persons with-
out insulin resistance, the LDL particle is larger and typed as large, or type
A. With insulin resistance, the LDL particle is smaller and termed small,
small-dense, or type B or A-B. This does not, however, suggest a goal for
where these particles need to be for maximal preventive purposes, nor are
there many clinical or imaging trials suggestive of what are such levels.
One possible method for determining such goal levels is to use popula-
tion-percentile levels equivalent to standard-lipid-profile LDL cholesterol
goal levels as outlined in Table 2. LDL cholesterol less than 100 mg/dL in
the standard lipid profile is below the twentieth percentile in the Framing-
ham Offspring Population [15], with less than 70 mg/dL being below the
fifth percentile. Equivalent non-HDL cholesterol values are less than 130
or less than 100 mg/dL. By looking at the twentieth and fifth percentiles
for apolipoprotein B [15] and LDL particle number by NMR in the same
population [13], one can get approximate equivalent values of less than 85
and less than 65 mg/dL for apolipoprotein B and less than 1100 and less
than 800 nmol/L for NMR particle number (!1000 and !700 in the
MESA population) [3].
The NCEP ATP III guidelines recommend the non-HDL cholesterol
goals as a substitute for apolipoprotein B, and have suggested the use of
such advanced lipoprotein testing as possibly useful. Occasionally, one
has a diabetic patient with CHD progressively occluding vessels despite
being at an LDL cholesterol goal level less than 70 mg/dL and triglycerides
less than 150 mg/dL. Advanced lipoprotein testing can determine whether in
such a case the patient has small, dense LDL particles and can quantify
them. In such cases the apolipoprotein B value or NMR particle number
rather than being at or below the fifth population percentile (!70 mg/dL,
the patient’s standard LDL cholesterol value) may be at a much higher fif-
tieth percentile level such as an apolipoprotein B of 110 mg/dL or an LDL
particle number of 1300 nmol/L. One can then aim to drive these LDL par-
ticle numbers or equivalents to lower levels by increasing LDL-lowering
therapy with higher-dose statins, or by using combination therapy, which
can increase LDL particle size and further reduce LDL particle number.
Often, such attempts can lead to standard lipid profile LDL cholesterol of
50 10 mg/dL.
1206 SMITH
Targeting such lower numbers as being more preventive in diabetic

patients with CHD is at this stage hypothetical and without proof of supe-
rior efficacy to lowering LDL cholesterol to less than 70 mg/dL. Certainly,
a more detailed look at targets for LDL cholesterol should never obfuscate
lowering triglycerides and raising HDL cholesterol in difficult or progressive
patients with diabetes and CHD. Although some statins may increase LDL
particle size as they simultaneously lower LDL particle number, fibric acids
and niacin added to a statin may accomplish the same thing while simulta-
neously improving triglyceride and HDL cholesterol levels.
Summary
The increased risk of CHD and its sequelae in the insulin-resistant patient
demands attention to all risk factors contributing to that risk. This article
has focused on treating the dyslipidemia associated with insulin resistance.
Statins are the first choice of a lipid-altering medication because of the
well-demonstrated preventive benefits, which are equivalent in relative risk
reduction in those with and without insulin resistance. In an occasional sub-
analysis, those with the dyslipidemic features of insulin resistance versus
those with LDL cholesterol elevations alone have shown greater
statin-induced relative and absolute CHD risk reductions.
National guidelines and clinical trial data support the following LDL
cholesterol goals in all persons, and in those with insulin resistance:
LDL cholesterol
Primary prevention, use NCEP guidelines plus one additional risk fac-
tor for metabolic syndrome when present
Lower risk !130 mg/dL
Higher risk !100 mg/dL
Secondary prevention: !70 mg/dL
Non-HDL cholesterol
Primary prevention
Lower risk !160 mg/dL
Higher risk !130 mg/dL
Secondary prevention: !100 mg/dL
To achieve these goals, high-dose statins or combination lipid-lowering
therapy often has to be used. The safety of statins at the highest doses in per-
sons with diabetes and at ages greater than 65 years has been clearly dem-
onstrated over a 5-year period.
No data exist to define an LDL cholesterol level below which further low-
ering is not efficacious or is unsafe. In the TNT trial in the quintile of 2000
patients with on-trial LDL cholesterol less than 64 mg/dL, there were fur-
ther reductions in events and no increase in adverse effects compared with
all of the quintiles who had higher on-trial LDL cholesterol levels.
Fibric acids have a proved CHD preventive effect in those with insulin
resistance manifested by triglycerides above 200 mg/dL and especially in
those with HDL cholesterol less than 35 to 40 mg/dL or an LDL/HDL cho-
lesterol ratio greater than 4.5. In comparing studies in such patients fibric
acids have shown greater relative risk reductions than statins, in others
less relative risk reductions.
Of all lipid-altering agents, niacin raises HDL cholesterol the most while
simultaneously lowering triglycerides and LDL cholesterol and has shown
decreases in MI after 5 years and 15-year total mortality even in those
with diabetes or baseline fasting blood glucoses greater than 126 mg/dL.
As a final alternative in those with hypertriglyceridemia, taking fish oils
containing 3000 mg of eicosapentanoic/docosahexanoic acid per day can
lower triglycerides by increasing intracellular catabolism of nascent VLDL
particles. Such doses have yet to be studied to see if this translates into re-
duced CHD events.
Trials of combination therapy of statins with fibric acids, and statins with
niacin, are currently underway, which will demonstrate the efficacy and
safety of these combinations versus statins alone, but are several years
away from completion. One imaging trial of adding 1 g of niacin to a statin
versus the statin alone has shown efficacy in not only stopping progressive
increases in carotid intimal medial thickening but in producing regression.
Such studies set the stage for the next lipid-altering trials measuring the ben-
efits of raising HDL cholesterol while simultaneously lowering triglycerides
and lowering LDL cholesterol.
References
[1] Otvos JD, Jeyarajah EJ, Bennett DW, et al. Development of a proton nuclear magnetic
resonance spectroscopic method for determining plasma lipoprotein concentrations and
subspecies distributions from a single, rapid measurement. Clin Chem 1992;38(9):
1632–8.
[2] Kathiresan S, Otvos JD, Sullivan LM, et al. Increased small low-density lipoprotein particle
number: a prominent feature of the metabolic syndrome in the Framingham Heart Study.
Circulation 2006;113(1):20–9.
[3] Mora S, Szklo M, Otvos JD, et al. LDL particle subclasses, LDL particle size, and carotid
atherosclerosis in the Multi-Ethnic Study of Atherosclerosis (MESA). Atherosclerosis
2007;192(1):211–7.
[4] Garvey WT, Kwon S, Zheng D, et al. Effects of insulin resistance and type 2 diabetes on
lipoprotein subclass particle size and concentration determined by nuclear magnetic reso-
nance. Diabetes 2003;52(2):453–62.
[5] UK Prospective Diabetes Study 27. Plasma lipids and lipoproteins at diagnosis of NIDDM
by age and sex. Diabetes Care 1997;20(11):1683–7.
[6] Feingold KR, Grunfeld C, Pang M, et al. LDL subclass phenotypes and triglyceride metab-
olism in non-insulin-dependent diabetes. Arterioscler Thromb 1992;12(12):1496–502.
[7] Selby JV, Austin MA, Newman B, et al. LDL subclass phenotypes and the insulin resistance
syndrome in women. Circulation 1993;88(2):381–7.
[8] Ginsberg HN, Zhang YL, Hernandez-Ono A. Regulation of plasma triglycerides in insulin
resistance and diabetes. Arch Med Res 2005;36(3):232–40.
1208 SMITH
[9] Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL heterogeneity.
J Lipid Res 2002;43(9):1363–79.
[10] Kissebah AH. Low density lipoprotein metabolism in non-insulin-dependent diabetes
mellitus. Diabetes Metab Rev 1987;3(3):619–51.
[11] Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density
lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem
1972;18(6):499–502.
[12] Cromwell WC, Otvos JD. Heterogeneity of low-density lipoprotein particle number in
patients with type 2 diabetes mellitus and low-density lipoprotein cholesterol ! 100 mg/
dl. Am J Cardiol 2006;98(12):1599–602.
[13] Otvos JD, Jeyarajah EJ, Cromwell WC. Measurement issues related to lipoprotein heteroge-
neity. Am J Cardiol 2002;90(8A):22i–9i.
[14] Expert panel on detection EaToHBCiA. Executive summary of the third report of the Na-
tional Cholesterol Education Program (NCEP) Expert panel on detection, evaluation, and
treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;
285(19):2486–97.
[15] Contois JH, McNamara JR, Lammi-Keefe CJ, et al. Reference intervals for plasma apolipo-
protein B determined with a standardized commercial immunoturbidimetric assay: results
from the Framingham Offspring Study. Clin Chem 1996;42(4):515–23.
[16] Bachorik PS, Lovejoy KL, Carroll MD, et al. Apolipoprotein B and AI distributions in the
United States, 1988-1991: results of the National Health and Nutrition Examination Survey
III (NHANES III). Clin Chem 1997;43(12):2364–78.
[17] Grundy SM. Low-density lipoprotein, non-high-density lipoprotein, and apolipoprotein B
as targets of lipid-lowering therapy. Circulation 2002;106(20):2526–9.
[20] Gerstein HC, Yusuf S, Bosch J, et al. Effect of rosiglitazone on the frequency of diabetes in
[21] Lawrence JM, Reid J, Taylor GJ, et al. Favorable effects of pioglitazone and metformin com-
pared with gliclazide on lipoprotein subfractions in overweight patients with early type 2
diabetes. Diabetes Care 2004;27(1):41–6.
[22] Goldberg RB, Kendall DM, Deeg MA, et al. A comparison of lipid and glycemic effects of
pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes
Care 2005;28(7):1547–54.
[23] 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(9131):854–65.
[24] Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular
events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone
Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 2005;
366(9493):1279–89.
[25] Jones PH, Davidson MH, Stein EA, et al. Comparison of the efficacy and safety of rosuvas-
tatin versus atorvastatin, simvastatin, and pravastatin across doses (STELLAR* Trial). Am
J Cardiol 2003;92(2):152–60.
[26] Vijan S, Hayward RA. Pharmacologic lipid-lowering therapy in type 2 diabetes mellitus:
background paper for the American College of Physicians. Ann Intern Med 2004;140(8):
650–8.
[27] Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart disease with pravastatin
in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N
Engl J Med 1995;333(20):1301–7.
protein as a predictor of coronary heart disease and diabetes in the West of Scotland Coro-
nary Prevention Study. Circulation 2003;108(4):414–9.
[29] Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary prevention of cardiovascular
disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes
Study (CARDS): multicentre randomised placebo-controlled trial. Lancet 2004;364(9435):
685–96.
[30] Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterol-lowering treat-
ment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials
of statins. Lancet 2005;366(9493):1267–78.
[31] Ballantyne CM, Olsson AG, Cook TJ, et al. Influence of low high-density lipoprotein cho-
lesterol and elevated triglyceride on coronary heart disease events and response to simvasta-
tin therapy in 4S. Circulation 2001;104(25):3046–51.
[32] Cannon CP, Steinberg BA, Murphy SA, et al. Meta-analysis of cardiovascular outcomes
trials comparing intensive versus moderate statin therapy. J Am Coll Cardiol 2006;48(3):
438–45.
[33] Deedwania P, Barter P, Carmena R, et al. Reduction of low-density lipoprotein cholesterol
in patients with coronary heart disease and metabolic syndrome: analysis of the Treating to
New Targets study. Lancet 2006;368(9539):919–28.
[34] Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the Na-
tional Cholesterol Education Program Adult Treatment Panel III Guidelines. J Am Coll
Cardiol 2004;44(3):720–32.
[35] Smith SC Jr, Allen J, Blair SN, et al. AHA/ACC guidelines for secondary prevention for pa-
tients with coronary and other atherosclerotic vascular disease: 2006 update: endorsed by the
National Heart, Lung, and Blood Institute. Circulation 2006;113(19):2363–72.
drome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific
Statement. Circulation 2005;112(17):2735–52.
[37] Dayspring T, Pokrywka G. Fibrate therapy in patients with metabolic syndrome and diabe-
tes mellitus. Curr Atheroscler Rep 2006;8(5):356–64.
[38] Rubins HB, Robins SJ, Collins D, et al. Gemfibrozil for the secondary prevention of coro-
nary heart disease in men with low levels of high-density lipoprotein cholesterol. Veterans
Affairs High-Density Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J
Med 1999;341(6):410–8.
[39] Rubins HB, Robins SJ, Collins D, et al. Diabetes, plasma insulin, and cardiovascular disease:
subgroup analysis from the Department of Veterans Affairs high-density lipoprotein inter-
vention trial (VA-HIT). Arch Intern Med 2002;162(22):2597–604.
[40] Frick MH, Elo O, Haapa K, et al. Helsinki Heart Study: primary-prevention trial
with gemfibrozil in middle-aged men with dyslipidemia. Safety of treatment, changes
in risk factors, and incidence of coronary heart disease. N Engl J Med 1987;317(20):
1237–45.
[41] Manninen V, Tenkanen L, Koskinen P, et al. Joint effects of serum triglyceride and LDL
cholesterol and HDL cholesterol concentrations on coronary heart disease risk in the Hel-
sinki Heart Study: implications for treatment. Circulation 1992;85(1):37–45.
[42] Keech A, Simes RJ, Barter P, et al. Effects of long-term fenofibrate therapy on cardiovascu-
lar events in 9795 people with type 2 diabetes mellitus (the FIELD study): randomised con-
[43] Keech A, Simes J, Barter P, et al. Correction to the FIELD study report. Lancet 2006;
368(9545):1415.
[44] Secondary prevention by raising HDL cholesterol and reducing triglycerides in patients with
coronary artery disease: the Bezafibrate Infarction Prevention (BIP) study. Circulation 2000;
102(1):21–7.
[45] Knopp RH. Drug treatment of lipid disorders. N Engl J Med 1999;341(7):498–511.
1210 SMITH
[46] Goldberg A, Alagona P Jr, Capuzzi DM, et al. Multiple-dose efficacy and safety of an
extended-release form of niacin in the management of hyperlipidemia. Am J Cardiol 2000;
85(9):1100–5.
[47] Meyers CD, Carr MC, Park S, et al. Varying cost and free nicotinic acid content in over-the-
counter niacin preparations for dyslipidemia. Ann Intern Med 2003;139(12):996–1002.
[48] McKenney JM, Proctor JD, Harris S, et al. A comparison of the efficacy and toxic effects of
sustained- vs immediate-release niacin in hypercholesterolemic patients. JAMA 1994;271(9):
672–7.
[49] Knopp RH, Alagona P, Davidson M, et al. Equivalent efficacy of a time-release form of
niacin (Niaspan) given once-a-night versus plain niacin in the management of hyperlipid-
emia. Metabolism 1998;47(9):1097–104.
[50] Knopp RH. Evaluating niacin in its various forms. Am J Cardiol 2000;86(12A):51L–6L.
[51] Garg A, Grundy SM. Nicotinic acid as therapy for dyslipidemia in non-insulin-dependent
diabetes mellitus. JAMA 1990;264(6):723–6.
[52] Grundy SM, Vega GL, McGovern ME, et al. Efficacy, safety, and tolerability of once-daily
niacin for the treatment of dyslipidemia associated with type 2 diabetes: results of the assess-
ment of diabetes control and evaluation of the efficacy of niaspan trial. Arch Intern Med
2002;162(14):1568–76.
[53] Elam MB, Hunninghake DB, Davis KB, et al. Effect of niacin on lipid and lipoprotein levels
and glycemic control in patients with diabetes and peripheral arterial disease: the ADMIT
study: a randomized trial. Arterial Disease Multiple Intervention Trial. JAMA 2000;
284(10):1263–70.
[54] Clofibrate and niacin in coronary heart disease. JAMA 1975;231(4):360–81.
[55] Canner PL, Berge KG, Wenger NK, et al. Fifteen year mortality in Coronary Drug Project
patients: long-term benefit with niacin. J Am Coll Cardiol 1986;8(6):1245–55.
[56] Canner PL, Furberg CD, Terrin ML, et al. Benefits of niacin by glycemic status in patients
with healed myocardial infarction (from the Coronary Drug Project). Am J Cardiol 2005;
95(2):254–7.
[57] Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the
combination for the prevention of coronary disease. N Engl J Med 2001;345(22):1583–92.
[58] Taylor AJ, Sullenberger LE, Lee HJ, et al. Arterial biology for the investigation of the treat-
ment effects of reducing cholesterol (ARBITER) 2: a double-blind, placebo-controlled study
of extended-release niacin on atherosclerosis progression in secondary prevention patients
treated with statins. Circulation 2004;110(23):3512–7.
[59] Taylor AJ, Lee HJ, Sullenberger LE. The effect of 24 months of combination statin and
extended-release niacin on carotid intima-media thickness: ARBITER 3. Curr Med Res
Opin 2006;22:2243–50.
[60] Grundy SM, Vega GL, Yuan Z, et al. Effectiveness and tolerability of simvastatin plus feno-
fibrate for combined hyperlipidemia (the SAFARI trial). Am J Cardiol 2005;95(4):462–8.
[61] Kashyap ML, McGovern ME, Berra K, et al. Long-term safety and efficacy of a once-daily
niacin/lovastatin formulation for patients with dyslipidemia. Am J Cardiol 2002;89(6):
672–8.
[62] LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in
patients with stable coronary disease. N Engl J Med 2005;352(14):1425–35.
Med Clin N Am 91 (2007) 1211–1223
Treatment of Hypertension and Other

Cardiovascular Risk Factors in Patients
with Metabolic Syndrome
Takeki Suzuki, MDa,1,
Shunichi Homma, MD, FACCb,*
a
Division of Cardiology, Department of Medicine, University of Vermont College of Medicine,
111 Colchester Avenue, Burlington, VT 05403, USA
b
Division of Cardiology, Department of Medicine, Columbia University College of Physicians
and Surgeons, 630 West 168th Street, New York, NY 10032, USA
Metabolic syndrome (MetS) (also known as syndrome X [1] and insulin

resistance syndrome [2]), is a concurrence of hypertension, abdominal obe-
sity, impaired fasting glucose (IFG), and dyslipidemia. It has been shown to
be a risk factor for cardiovascular disease [3–7]. The clustering of cardiome-
tabolic risk factors has been known for a long time [8].
Insulin resistance has been proposed as the underlying pathophysiologic
cause of MetS [9]. The prevalence of MetS has increased over decades
among United States adults as the prevalence of obesity increased [10,11].
Various definitions of MetS have been proposed [12–15]. In 1998, the World
Health Organization defined the syndrome [13]. In 2001, the National Cho-
lesterol Education Program (NCEP) Adult Treatment Panel III (ATP III)
defined MetS as one potential secondary target of therapy beyond low-
density lipoprotein (LDL)–lowering therapy [16]. The NCEP ATP III defi-
nition was later modified by the American Heart Association and National
Heart Lung and Blood Institute Panel in 2005 to correspond with the new
American Diabetes Association criteria for IFG [17,18]. Detailed descrip-
tion of clinical definitions [19], controversies [20], and underlying patho-
physiology of MetS [21] are discussed elsewhere in this issue.
1
Present address: Fletcher Allen Health Care, McClure One, 111 Colchester Avenue,
Burlington, VT 05401, USA.
E-mail address: sh23@columbia.edu (S. Homma).
1212 SUZUKI & HOMMA
Strategy to treat patients with metabolic syndrome

The primary goal of clinical management of MetS is to reduce the risk of
clinical atherosclerotic cardiovascular disease [18]. In patients who have not
yet developed it, prevention of type 2 diabetes mellitus is another important
goal.
Risk assessment
The first step is to assess an individual’s risk status for atherosclerotic
cardiovascular disease. Major risk factors other than LDL and diabetes mel-
litus are as follows: cigarette smoking; hypertension (blood pressure R140/
90 mm Hg or on antihypertensive medication); low high-density lipoprotein
(HDL) cholesterol (!40 mg/dL); family history of premature coronary
heart disease ([CHD] in male first-degree relative !55 years of age, in
female first-degree relative !65 years of age); and age (men R45 years;
women R55 years) [16]. By using the Framingham risk score, individuals
with MetS should be categorized based on absolute 10-year risk for CHD:
high risk (10-year risk O20%); moderately high risk (2þ risk factors and
10-year risk 10%–20%); moderate risk (2þ risk factors and 10-year
risk !10%); or low risk (0–1 risk factor and 10-year risk !10%) [16].
Individuals with the clinical form of atherosclerotic cardiovascular disease
or with diabetes are in the high-risk category [12,22]. Risk factors in the Fra-
mingham risk score include age, total cholesterol, HDL cholesterol, blood
pressure, and cigarette smoking. Electronic 10-year risk calculators are
available at http://www.nhlbi.nih.gov/guidelines/cholesterol/.
MetS is a target of therapy beyond LDL lowering and it is not an adequate
tool to estimate 10-year risk for CHD [23]. Although cigarette smoking is not
related to MetS, smoking deserves special attention as a cardiovascular risk
factor and efforts should be made to bring about smoking cessation in any cur-
rent smokers.
ATP III recommended two major approaches to manage individuals with
the MetS: to reduce the underlying causes (obesity and physical inactivity);
and to treat hypertension and other cardiometabolic risk factors in MetS
[16]. A summary of clinical management of MetS is shown in Table 1.
The two approaches are discussed next.
Reduction of underlying causes

Excess fat (obesity) and physical inactivity are two main underlying
causes of MetS. These promote development of insulin resistance. Thera-
peutic lifestyle changes are the first steps to be considered for individuals
with MetS. Therapeutic lifestyle changes should stress weight reduction
and physical activity. Weight reduction is achieved with a comprehensive
approach using exercise and healthy diet.
TREATMENT OF HYPERTENSION 1213
Several randomized controlled studies have shown the usefulness of ther-

apeutic lifestyle changes [24–27]. In the Da Qiang IGT and Diabetes Study,
577 participants with impaired glucose tolerance (IGT) were randomly
assigned to the following four groups: (1) diet, (2) exercise, (3) diet plus ex-
ercise, and (4) placebo. The diet, exercise, and diet-plus-exercise interven-
tions decreased the incidence of diabetes mellitus over 6 years by 31%,
46%, and 42%, respectively [26]. The Finnish Diabetes Prevention Study
Group involved 522 middle-aged, overweight subjects with IGT and showed
that lifestyle changes decreased the risk of diabetes by 58% [27]. Follow-up
of the Finnish Diabetes Prevention Study showed sustained lifestyle changes
and a reduction in diabetes incidence after the individual counseling was
stopped [28].
Reduction in the incidence of type 2 diabetes mellitus among high-risk
persons was also observed in the United States Diabetes Prevention Pro-
gram [24]. Diabetes Prevention Program assigned 3234 subjects with IGT
to placebo, metformin, or a life-style modification program. Over 2.8 years
of follow-up, lifestyle intervention reduced the incidence of diabetes by
58%, identical to the result of the Finnish Diabetes Prevention Study. In
the Diabetes Prevention Program, the prevalence of and progression to
MetS were examined [29]. The prevalence of MetS in the Diabetes Preven-
tion Program population with IGT was 53% at baseline and the incidence
of MetS in the lifestyle group was reduced by 41% when compared with
the placebo group. Of interest, the dramatic effect of lifestyle on the preven-
tion of incident MetS and the reduction of its overall prevalence seemed to
be most strongly related to a reduction in waist circumference and in blood
pressure, and not through a correction of the lipid abnormalities of triglyc-
eride and HDL cholesterol [29]. Next, each component of lifestyle that needs
to be modified is discussed.
Abdominal obesity
Weight reduction is the first-line intervention in obese individuals with
MetS. Weight reduction has a synergistic effect on LDL lowering and de-
creases all of the risk factors of MetS. Obesity guidelines put emphasis on
weight reduction, using behavioral change to reduce caloric intake and
increase physical activity [30]. The first goal is to achieve a 7% to 10%
decrease in total body weight from baseline over 6 to 12 months [18]. This
requires a modest caloric intake reduction of 500 to 1000 calories per day.
Physical inactivity
Current guidelines on physical activity recommend practical, regular, and
moderate regimens of physical activity (ie, 30 minutes or more of moderate-
intensity physical activity, such as brisk walking daily) [31,32]. Patients are
advised progressively to increase their physical activity. More exercise
(ie, 1 hour daily) is even more efficacious for weight control [33]. Avoiding
1214
Table 1
Summary of clinical management of metabolic syndrome
Underlying risk factors Clinical management
Abdominal obesity Weight reduction: reduce weight by 7%–10% over 6–12 months; ultimate goal to achieve desirable
weight (BMI !25 kg/m2)
Physical inactivity Practical, regular, and moderate physical activity for 30–60 min
Atherogenic diet Reduced intakes of saturated fats (!7% of total calories), trans fats, and cholesterol (!200 mg/d)
Reduced simple sugars; increased intakes of fruits, vegetables, and whole grains
Nutritional intervention
Hypertension and other cardiometabolic risk factors
Hypertension Follow JNC-7 report
SUZUKI & HOMMA

Goal: 140/90 for individuals with isolated hypertension and 130/80 for individuals with diabetes
mellitus or chronic kidney disease
Elevated fasting glucose For IFG, encourage weight reduction and increased physical activity; metformin, 850 mg twice per
day, for individuals with IFG and IGT and any of the following [53]:
!60 years of age
BMI R35 kg/m2
Family history of diabetes in first-degree relatives
Elevated triglycerides
Reduced HDL cholesterol
Hypertension
Hemoglobin A1c O6%
For type 2 diabetes mellitus, lifestyle therapy and medication therapy to achieve hemoglobin A1c
!7%
Atherogenic dyslipidemia Primary target: elevated LDL
Goal: high-risk patients !100 mg/dL
For very high-risk patients, LDL !70 mg/dL is optional
Moderately high-risk patients
(2þ risk factorsa and 10-year risk %20%) !130 mg/dL
LDL !100 mg/dL is optional
Moderate-risk patients
(2þ risk factors and 10-year risk !10%) !130 mg/dL
Lower-risk patients (0–1 risk factor) !160 mg/dL
Treatment: lifestyle changes plus LDL-lowering medication
Secondary target: elevated non-HDL
Goal: High-risk patients !130 mg/dL
Moderately high-risk patients !160 mg/dL
Non-HDL !130 mg/dL is optional
Moderate-risk patients !160 mg/dL
Lower-risk patients !190 mg/dL
Treatment: intensify LDL-lowering therapy or add fibrate or nicotinic acid
Tertiary target: reduced HDL
TREATMENT OF HYPERTENSION
No specific goal
Prothrombotic and proinflammatory state Low-dose aspirin for high-risk patients
CRP may be useful for monitoring, but currently not recommended
Miscellaneous
Current smoking Smoking cessation
a
Risk factors include cigarette smoking; hypertension (blood pressure R140/90 mm Hg or on antihypertensive medication); low HDL cholesterol
(!40 mg/dL); family history of premature CHD (CHD in male first-degree relative !55 years of age, CHD in female first-degree relative !65 years of
age); and age (men R45 years; women R55 years).
Abbreviations: BMI, body mass index; CHD, coronary heart disease; CRP, C-reactive protein; HDL, high-density lipoprotein; IFG, impaired fasting
glucose; IGT, impaired glucose tolerance; JNC-7, Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure;
LDL, low-density lipoprotein.
Data from Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and
treatment of high blood cholesterol in adults (adult treatment panel III). JAMA 2001;285:2486–97; and Grundy SM, Cleeman JI, Daniels SR, et al.
Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute scientific statement.
Circulation 2005;112:2735–52.
1215
1216 SUZUKI & HOMMA
common sedentary activities in leisure time (television watching and com-

puter games) is advised. American Heart Association guidelines recommend
exercise testing before vigorous exercise in selected patients with cardiovas-
cular disease and other patients with symptoms of or at high risk of cardio-
vascular disease [32].
Atherogenic and diabetogenic diets

As a part of therapeutic lifestyle changes, ATP III recommends reduced
intake of saturated fats (!7% of total calories); trans fats; and cholesterol
(!200 mg/d) [16]. In addition, there should be reduced consumption of sim-
ple sugars, and increased intake of fruits, vegetables, and whole grains.
Physicians are, at every stage of patient care, encouraged to refer patients
to dietitians or other qualified nutritionists for optimal nutritional interven-
tion. American Heart Association dietary guidelines are also available for
further information [34].
Hypertension and other cardiometabolic risk factors

Beyond reducing underlying causes, such as abdominal obesity, physical
inactivity, and atherogenic and diabetogenic diets, treatment should be
directed against treating hypertension and other cardiometabolic, or cardio-
vascular risk factors of MetS. Controversy exists on this subject [35]. There
is no question, however, that the cardiometabolic risk factors are prone to
cluster. Hypertension and other cardiometabolic risk factors for MetS,
such as IFG and dyslipidemia, are discussed. Prothrombotic and proinflam-
matory states, which are closely related to MetS, are also discussed as poten-
tial therapeutic targets of MetS.
Hypertension
Hypertension is one of the cardinal components of MetS. Hypertension
itself is a major and independent cardiovascular risk factor [36]. It has
been shown that patients with hypertension are more likely to be insulin
resistant [37,38], and hypertension tends to cluster with other metabolic
risk factors [39].
Currently, there is no guideline for treating hypertension specifically in
individuals with MetS. The authors use the 7th Report of the Joint National
Committee on Prevention, Detection, Evaluation, and Treatment of High
Blood Pressure (JNC-7) guidelines to treat hypertension in such individuals
[40]. First, lifestyle changes should be emphasized in all individuals with
MetS. Mild elevations of blood pressure often can be effectively controlled
by therapeutic lifestyle changes. The Dietary Approaches to Stop Hyperten-
sion diet, a diet rich in fruits, vegetables, and low-fat dairy foods and with
reduced saturated and total fat, has been shown substantially to lower blood
pressure [41]. In patients with MetS, the Dietary Approaches to Stop Hyper-
tension diet resulted in reduction of the metabolic risk factors both in men
and women [42].
The JNC-7 report recommends that patients with systolic blood pressure
of greater than or equal to 140 mm Hg or diastolic blood pressure of greater
than or equal to 90 mm Hg require antihypertensive treatment. The blood
pressure goals are less than 140/90 mm Hg in patients with isolated
hypertension and less than 130/80 mm Hg in patients with diabetes mellitus
or chronic kidney disease. In diabetic patients, angiotensin-converting
enzyme inhibitors (ACE-Is) or angiotensin receptor blockers (ARBs) are
recommended.
Beyond diabetes mellitus, ACE-Is or ARBs can be a good option for
hypertensive patients with MetS. Recent research has focused on the role
of the renin-angiotensin system in MetS [43]. At multiple levels, there is
a cross-talk between the renin-angiotensin system and insulin signaling,
and the renin-angiotensin system blockage by ACE-Is or ARBs may have
a favorable effect on glucose profiles. This favorable impact of ACE-Is or
ARBs on glucose metabolism has been suggested by previous post hoc anal-
yses from clinical trials and a meta-analysis [44–46]. There is no clinical trial,
however, showing the effect of ACE-Is on decreasing the incidence of diabe-
tes. The DREAM trial evaluated the effects of ramipril and rosiglitazone on
the primary end points of incident diabetes and death in subjects with IGT
or IFG. This study did not show a beneficial effect of ramipril on their pri-
mary outcome, incident diabetes, over a 3-year period, but ramipril signifi-
cantly increased regression to normoglycemia [47]. Given these, the use of
ACE-Is or ARBs in patients with the MetS but without diabetes mellitus
may be considered. If patients take ACE-Is for another indication, such
as myocardial infarction, hypertension in diabetes mellitus, or congestive
heart failure, improvement in glycemia may be another benefit.
Combination antihypertensive therapy is another consideration. Patients
with hypertension (R140/90) may require combination antihypertensive
therapy to control blood pressure. In this regard, most investigators in the
hypertension field believe that the potential benefit of low-dose diuretics
(12.5 mg of hydrochlorothiazide or its equivalent) in combination with
ACE-Is or ARBs outweighs risk. The combination of diuretics and
b-blockers may worsen metabolic profiles [48], however, and should be
avoided.
The goal blood pressure is not known in individuals with MetS, but to
use the same blood pressure goal as diabetes mellitus (!130/80) may be
reasonable.
Elevated fasting glucose

Elevated fasting glucose (R100 mg/dL) is included in the components of
the MetS. Patients with IFG and diabetes mellitus are included in NCEP
1218 SUZUKI & HOMMA
ATP III definition of MetS. Beyond therapeutic lifestyle changes, metformin

[24], thiazolidinedione [49–51], and acarbose [52] have been shown to lower
the risk of diabetes mellitus in patients with IFG or IGT. In the Diabetes
Prevention Program, metformin reduced the incidence of diabetes by 31%
[24]. None of these drugs is recommended solely for preventing diabetes
in individuals with MetS, however, because of insufficient evidence on
long-term safety and cost-effectiveness. Weight reduction and increased
physical activity are recommended for individuals with MetS and elevated
fasting glucose.
Recently, the American Diabetes Association issued a new consensus
statement that recommends that metformin be considered as drug therapy
for individuals with IFG-IGT [53]. If metformin (850 mg twice per day) is
to be used, both abnormalities (IFG and IGT) must be documented and
individuals must have any of the following:
!60 years of age
BMI R35 kg/m2
Family history of diabetes in first-degree relatives
Elevated triglycerides
Reduced HDL cholesterol
Hypertension
Hemoglobin A1c O6%
If individuals with MetS meet these criteria, metformin can be a treatment
of choice.
For patients with type 2 diabetes mellitus, treatment is directed toward
a hemoglobin A1c of less than 7% [54]. Management of diabetes is discussed
elsewhere in this issue.
Atherogenic dyslipidemia
Atherogenic dyslipidemia consists of an aggregation of lipoprotein
abnormalities including elevated serum triglyceride and apolipoprotein B,
increased small LDL particles, and reduced level of HDL cholesterol [18].
As emphasized by NCEP guidelines [12], LDL cholesterol is the primary
target of lipid-lowering therapy even among individuals with MetS. Statins
(3-hydroxy-3-methylglutaryl-coenzyme A [HMG-CoA] reductase inhibi-
tors) reduce LDL cholesterol by competitively inhibiting HMG-CoA reduc-
tase, the rate-limiting step in cholesterol biosynthesis [55]. Statins occupy
a portion of the binding site of HMG-CoA, blocking access of the substrate
HMG-CoA to HMG-CoA reductase [56]. Ezetimibe and bile acid seques-
trants are other LDL-lowering drugs. The target LDL level is less than
100 mg/dL for patients with CHD and with CHD risk equivalents, such
as diabetes and 10-year risk of CHD greater than 20%. In very high-risk
individuals (including patients with established CHD plus other high-risk
conditions including MetS), an LDL goal of 70 mg/dL is a therapeutic
option [57]. For individuals with multiple (two plus) risk factors in whom
10-year risk for CHD is less than or equal to 20%, LDL cholesterol should
be less than 130 mg/dL. For moderately high-risk patients with 10-year risk
between 10% and 20%, LDL less than 100 mg/dL is optional. For lower-
risk patients in whom 10-year risk for CHD is less than 10%, the target
LDL cholesterol level is 160 mg/dL.
Once the primary target is achieved, the secondary target of lipid-lower-
ing therapy is non-HDL cholesterol in patients with high triglyceride levels
(R200 mg/dL). Non-HDL cholesterol is the sum of LDL plus very LDL
cholesterol. Goals for non-HDL cholesterol levels are 30 mg/dL higher
than goals for LDL cholesterol in each risk category (ie, non-HDL goal
for CHD and CHD risk equivalent group is !130 mg/dL, whereas LDL
goal for the group is !100 mg/dL). There are two drug therapy options
to achieve the secondary goal. The first option is to intensify statin therapy,
which has been used for lowering LDL. The second option is to add nico-
tinic acid or fibrate. When fibrate is used in combination with a statin, cau-
tion should be heeded for risk of severe myopathy especially when higher
doses of the statin are used.
Special consideration should be made for patients in whom serum triglyc-
eride levels are greater than or equal to 500 mg/dL. These individuals have
a higher risk of acute pancreatitis secondary to very high triglyceride levels.
Drug treatment (fibrate or nicotinic acid) and low-fat diet, weight reduction,
and increased physical activity should be initiated to lower triglyceride levels
and prevent development of acute pancreatitis.
The tertiary target of atherogenic dyslipidemia is HDL cholesterol. No
specific goal is set for HDL cholesterol levels, but efforts should be made
to raise HDL to the highest extent possible with standard therapies for ath-
erogenic dyslipidemia [18].
Prothrombotic and proinflammatory states

Prothrombotic and proinflammatory states are related to MetS. Patients
with MetS often have elevated prothrombotic biomarkers, such as fibrino-
gen and plasminogen activator inhibitor-1. Likewise, cytokines, such as
tumor necrosis factor-a and interleukin-6, and acute-phase reactants, such
as C-reactive protein, are elevated in patients with MetS [58–61]. There is
cumulating evidence that proinflammatory state is an underlying patho-
physiology of MetS, and high-sensitivity C-reactive protein has been
proposed to be incorporated into MetS criteria [62]. Currently, no defini-
tions have included proinflammatory state as a component of the syndrome.
There is no specific therapy directly acting on prothrombotic or proinflam-
matory state. Low-dose aspirin may be considered for patients with CHD,
diabetes, or high-risk profile based on the Framingham risk score (10-year
risk of CHD O20%) to prevent arterial thrombosis. Medications, such as sta-
tins, nicotinic acid, fibrates, ACE inhibitors, and thiazolidinediones, can
1220 SUZUKI & HOMMA
decrease C-reactive protein levels, but use of these medications solely to re-
duce C-reactive protein cannot be recommended. C-reactive protein may be
used to monitor proinflammatory state in individuals. A high-sensitivity C-re-
active protein level greater than 3 mg/dL may be used as a cutpoint for a high-
risk [63]. At present, routine use of C-reactive protein for individuals with
MetS is not recommended.
Summary
Treatment of MetS consists of multiple components and there is no spe-
cific treatment exclusively for the syndrome. Therapeutic lifestyle changes
with weight reduction, increased physical activity, and healthy diet are
first-line therapies for individuals with MetS. Risk stratification using the
Framingham risk score may be applied to detect patients with higher risk
of CHD. In addition to therapeutic lifestyle changes, medical treatment
for each component of the syndrome is recommended. Hypertension is
one of the components of the syndrome and should be treated aggressively
following JNC-7 report. ACE-Is or ARBs may be the treatment of choice
for hypertension in MetS, especially when either type 2 diabetes mellitus
or chronic kidney disease is present. Combination therapy of ACE-Is or
ARBs with low-dose diuretics may be considered when blood pressure is
not well controlled by monotherapy.
References
[1] Reaven GM. Role of insulin resistance in human disease (syndrome X): an expanded defini-
tion. Annu Rev Med 1993;44:121–31.
[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.
with the metabolic syndrome. Diabetes Care 2001;24(4):683–9.
[4] Lakka HM, Laaksonen DE, Lakka TA, et al. The metabolic syndrome and total and cardio-
vascular disease mortality in middle-aged men. JAMA 2002;288(21):2709–16.
[5] McNeill AM, Katz R, Girman CJ, et al. Metabolic syndrome and cardiovascular
disease in older people: the cardiovascular health study. J Am Geriatr Soc 2006;
54(9):1317–24.
[6] Wilson PW, Kannel WB, Silbershatz H, et al. Clustering of metabolic factors and coronary
heart disease. Arch Intern Med 1999;159(10):1104–9.
[7] Wilson PW, D’Agostino RB, Parise H, et al. Metabolic syndrome as a precursor of cardio-
vascular disease and type 2 diabetes mellitus. Circulation 2005;112(20):3066–72.
[8] Avogaro P, Crepaldi G, Enzi G, et al. Associazione di iperlipidemia, diabete mellito e obesita
di medio grado. Acta Diabetol Lat 1967;4:36–41.
[9] Reaven GM. Banting lecture 1988. Role of insulin resistance in human disease. Diabetes
1988;37(12):1595–607.
U.S. Adults. Diabetes Care 2004;27(10):2444–9.
[11] Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United
States, 1999–2004. JAMA 2006;295(13):1549–55.
[12] Third report of the National Cholesterol Education Program (NCEP) expert panel on detec-
tion, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel
III): final report. Circulation 2002;106(25):3143–421.
a WHO consultation. Diabet Med 1998;15(7):539–53.
[14] Einhorn D, Reaven GM, Cobin RH, et al. American college of endocrinology position state-
ment on the insulin resistance syndrome. Endocr Pract 2003;9(3):237–52.
[15] International Diabetes Federation. The IDF consensus worldwide definition of the meta-
bolic syndrome. Available at: http://www.idf.org/webdata/docs/mets_def_update2006.pdf.
[16] Executive summary of the third report of the National Cholesterol Education Program
(NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in
adults (adult treatment panel III). JAMA 2001;285(19):2486–97.
[17] Genuth S, Alberti KG, Bennett P, et al. Follow-up report on the diagnosis of diabetes mel-
litus. Diabetes Care 2003;26(11):3160–7.
drome: an American Heart Association/National Heart, Lung, and Blood Institute scientific
statement. Circulation 2005;112(17):2735–52.
[19] Pi-Sunyer FX. What is the metabolic syndrome? A clinical definition of the characteristics
and epidemiology. Med Clin North Am 2007, in press.
[20] Brietzke S. Controversies in the diagnostic criteria for the metabolic syndrome. Med Clin
North Am 2007, in press.
[21] Lann D, LeRoith D. Insulin resistance as the underlying cause for the metabolic syndrome.
Med Clin North Am 2007, in press.
[22] Haffner SM, Lehto S, Ronnemaa T, et al. Mortality from coronary heart disease in subjects
with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarc-
tion. N Engl J Med 1998;339(4):229–34.
[23] Wannamethee SG, Shaper AG, Lennon L, et al. Metabolic syndrome vs Framingham risk
score for prediction of coronary heart disease, stroke, and type 2 diabetes mellitus. Arch
Intern Med 2005;165(22):2644–50.
[25] Kosaka K, Noda M, Kuzuya T. Prevention of type 2 diabetes by lifestyle intervention: a
Japanese trial in IGT males. Diabetes Res Clin Pract 2005;67(2):152–62.
[26] Pan XR, Li GW, Hu YH, et al. Effects of diet and exercise in preventing NIDDM in people
with impaired glucose tolerance. The Da Qing IGT and diabetes study. Diabetes Care 1997;
20(4):537–44.
344(18):1343–50.
[28] Lindstrom J, Ilanne-Parikka P, Peltonen M, et al. Sustained reduction in the incidence of
type 2 diabetes by lifestyle intervention: follow-up of the Finnish diabetes prevention study.
Lancet 2006;368(9548):1673–9.
[29] Orchard TJ, Temprosa M, Goldberg R, et al. The effect of metformin and intensive lifestyle
intervention on the metabolic syndrome: the diabetes prevention program randomized trial.
Ann Intern Med 2005;142(8):611–9.
[30] Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in
adults–the evidence report. National Institutes of Health. Obes Res 1998;6(Suppl 2):51S–209S.
[31] Pate RR, Pratt M, Blair SN, et al. Physical activity and public health. A recommendation
from the Centers for Disease Control and Prevention and the American College of Sports
Medicine. JAMA 1995;273(5):402–7.
1222 SUZUKI & HOMMA
[32] Thompson PD, Buchner D, Pina IL, et al. Exercise and physical activity in the prevention
and treatment of atherosclerotic cardiovascular disease: a statement from the Council on
Clinical Cardiology (subcommittee on exercise, rehabilitation, and prevention) and the
Council on Nutrition, Physical Activity, and Metabolism (subcommittee on physical activ-
ity). Circulation 2003;107(24):3109–16.
[33] Grundy SM, Hansen B, Smith SC Jr, et al. Clinical management of metabolic syndrome:
report of the American Heart Association/National Heart, Lung, and Blood Institute/Amer-
ican Diabetes Association conference on scientific issues related to management. Circulation
2004;109(4):551–6.
[34] Krauss RM, Eckel RH, Howard B, et al. AHA dietary guidelines: revision 2000: a statement
for healthcare professionals from the nutrition committee of the American Heart Associa-
tion. Circulation 2000;102(18):2284–99.
the Study of Diabetes. Diabetes Care 2005;28(9):2289–304.
[36] Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to
vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective
studies. Lancet 2002;360(9349):1903–13.
[37] Ferrannini E, Buzzigoli G, Bonadonna R, et al. Insulin resistance in essential hypertension.
N Engl J Med 1987;317(6):350–7.
[38] Swislocki AL, Hoffman BB, Reaven GM. Insulin resistance, glucose intolerance and hyper-
insulinemia in patients with hypertension. Am J Hypertens 1989;2(6 Pt 1):419–23.
[39] Zavaroni I, Mazza S, Dall’Aglio E, et al. Prevalence of hyperinsulinaemia in patients with
high blood pressure. J Intern Med 1992;231(3):235–40.
[40] Chobanian AV, Bakris GL, Black HR, et al. The seventh report of the Joint National Com-
mittee on Prevention, Detection, Evaluation, And Treatment of High Blood Pressure: the
JNC 7 Report. JAMA 2003;289(19):2560–72.
[41] Appel LJ, Moore TJ, Obarzanek E, et al. A clinical trial of the effects of dietary patterns on
blood pressure. DASH collaborative research group. N Engl J Med 1997;336(16):1117–24.
[42] Azadbakht L, Mirmiran P, Esmaillzadeh A, et al. Beneficial effects of a dietary approaches to
stop hypertension eating plan on features of the metabolic syndrome. Diabetes Care 2005;
28(12):2823–31.
[43] Prasad A, Quyyumi AA. Renin-angiotensin system and angiotensin receptor blockers in the
metabolic syndrome. Circulation 2004;110(11):1507–12.
[44] Abuissa H, Jones PG, Marso SP, et al. Angiotensin-converting enzyme inhibitors or angio-
tensin receptor blockers for prevention of type 2 diabetes: a meta-analysis of randomized
clinical trials. J Am Coll Cardiol 2005;46(5):821–6.
[45] Hansson L, Lindholm LH, Niskanen L, et al. Effect of angiotensin-converting-enzyme inhi-
bition compared with conventional therapy on cardiovascular morbidity and mortality in
hypertension: the Captopril Prevention Project (CAPPP) randomised trial. Lancet 1999;
353(9153):611–6.
[46] Yusuf S, Sleight P, Pogue J, et al. Effects of an angiotensin-converting-enzyme inhibitor,
ramipril, on cardiovascular events in high-risk patients. The Heart Outcomes Prevention
Evaluation Study investigators. N Engl J Med 2000;342(3):145–53.
[47] Bosch J, Yusuf S, Gerstein HC, et al. Effect of ramipril on the incidence of diabetes. N Engl
J Med 2006;355(15):1551–62.
[48] Lindholm LH, Persson M, Alaupovic P, et al. Metabolic outcome during 1 year in newly
detected hypertensives: results of the Antihypertensive Treatment and Lipid Profile in
a North of Sweden Efficacy Evaluation (ALPINE Study). J Hypertens 2003;21(8):1563–74.
[49] Buchanan TA, Xiang AH, Peters RK, et al. Preservation of pancreatic beta-cell function and
prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high-risk
Hispanic women. Diabetes 2002;51(9):2796–803.
[50] Gerstein HC, Yusuf S, Bosch J, et al. Effect of Rosiglitazone on the frequency of diabetes in
[51] Knowler WC, Hamman RF, Edelstein SL, et al. Prevention of type 2 diabetes with troglita-
zone in the diabetes prevention program. Diabetes 2005;54(4):1150–6.
[53] Nathan DM, Davidson MB, DeFronzo RA, et al. Impaired fasting glucose and impaired
glucose tolerance: implications for care. Diabetes Care 2007;30(3):753–9.
[54] Standards of medical care in diabetes. Diabetes Care 2005;28(Suppl 1):S4–36.
[55] Corsini A, Maggi FM, Catapano AL. Pharmacology of competitive inhibitors of
HMG-CoA reductase. Pharmacol Res 1995;31(1):9–27.
[56] Istvan ES, Deisenhofer J. Structural mechanism for statin inhibition of HMG-CoA reduc-
tase. Science 2001;292(5519):1160–4.
[57] Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the
National Cholesterol Education Program Adult Treatment Panel III Guidelines. Circulation
2004;110(2):227–39.
[58] Dandona P, Aljada A, Bandyopadhyay A. Inflammation: the link between insulin resistance,
obesity and diabetes. Trends Immunol 2004;25(1):4–7.
[59] Frohlich M, Imhof A, Berg G, et al. Association between C-reactive protein and features of
the metabolic syndrome: a population-based study. Diabetes Care 2000;23(12):1835–9.
[60] Ridker PM, Buring JE, Cook NR, et al. C-reactive protein, the metabolic syndrome, and risk
of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American
women. Circulation 2003;107(3):391–7.
[61] Wannamethee SG, Lowe GDO, Shaper AG, et al. The metabolic syndrome and insulin
resistance: relationship to haemostatic and inflammatory markers in older non-diabetic
men. Atherosclerosis 2005;181(1):101–8.
[62] Ridker PM, Wilson PW, Grundy SM. Should C-reactive protein be added to metabolic syn-
drome and to assessment of global cardiovascular risk? Circulation 2004;109(23):2818–25.
[63] Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascu-
lar disease: application to clinical and public health practice: a statement for healthcare pro-
fessionals from the centers for disease control and prevention and the American Heart
Association. Circulation 2003;107(3):499–511.
Med Clin N Am 91 (2007) 1225–1253
Medical Therapy for ObesitydCurrent

Status and Future Hopes
George A. Bray, MD, MACP
Pennington Center, 6400 Perkins Road, Baton Rouge, LA 70808, USA
Obesity is often described as an epidemic. In this context, it is essential to

develop ways of preventing more people from becoming obese. But where
prevention fails, treatment may be necessary. A number of different strate-
gies have been used to treat obesity, including diet, exercise, behavior ther-
apy, medications, and surgery. Criteria for selecting among these treatments
involve evaluating the risks to the individual from their obesity and balanc-
ing that against any possible problems with the treatment. Because all
medications inherently have more risks than diet and exercise, deciding to
use medications should only be done for people where the benefit justifies
the risk [1].
This process of evaluation is particularly important when we realize that
drug treatment for obesity has been tarnished by a number of unfortunate
problems over the years. Since the introduction of thyroid hormone to treat
obesity in 1893, almost every drug that has been tried in obese patients has
caused undesirable outcomes, necessitating its termination. Thus, caution
must be used in accepting any new drug for the treatment of obesity unless
the safety profile would make it acceptable for almost everyone.
Another issue surrounding drug treatment of obesity is the perception
that because patients regain weight when drugs are stopped, the drugs are
ineffective. Quite the contrary is true. Obesity is a chronic disease that has
many causes. However, cure is rare, and treatment is thus aimed at pallia-
tion, that is, producing and maintaining weight loss. Physicians do not
expect to cure diseases such as hypertension or hypercholesterolemia with
medications. Rather, they expect to palliate them. When the medications
for any of these chronic diseases are discontinued, the disease is expected
The author has received research support from Merck, and has served as a consultant to
sanofi-aventis, Merck, Schering-Plough, Eli Lilly, Amgen, and Amylin.
E-mail address: brayga@pbrc.edu
1226 BRAY
to recur. This means that medications only work when they are used. The
same argument applies for medications used to treat obesity.
If an individual is to lose weight, he or she must go into negative energy
balance, where the energy taken in as food is less on average than the energy
needed for daily activities. Thus, the current group of medications can be
divided into two broad categories: those that act primarily on the central
nervous system to reduce food intake and those that act primarily outside
the brain. Wherever the primary site of action may be, however, the net
effect must be a reduction in food intake and an increase in energy expen-
diture. There currently are several drugs available in the United States to
treat obesity. Table 1 [2–6] summarizes these drugs.
Drugs that reduce food intake primarily by acting

in the central nervous system
The drugs considered in this category are rimonabant, sibutramine,
phentermine, and the other sympathomimetic drugs.
Rimonabant
Rimonabant is approved in Europe but not yet (as of June 2007) in the
United States. The stimulation of food intake by tetrahydrocannabinol
found in the marijuana plant occurs by stimulation of cannabinoid recep-
tors. There are two cannabinoid receptors, CB-1 (470 amino acids in length)
and CB-2 (360 amino acids in length), which respond to endogenous endo-
cannabinoids, of which at least two have been found, anandamide and
2-arachidonoyl-glycerol. The CB-1 receptors are distributed through the
brain in the areas related to feeding, as well as on fat cells and in the gastro-
intestinal (GI) tract. Rimonabant is a specific antagonist of the CB-1
Table 1
Drugs approved by the US Food and Drug Administration (FDA) for treatment of obesity
Generic name Status Usual dose Comments
Drugs approved by the US FDA for long-term treatment of overweight patients
Orlistat 120 mg tid May have gastrointestinal
side effects
Sibutramine 5 mg/d–15 mg/d Raises blood pressure
Rimonabant NDA 20 mg/d Reduces weight gain after
withdrawn cessation of smoking
Drugs approved by the US FDA for short-term treatment of overweight patients
Benzphetamine DEA-III 15 mg/d–30 mg/d in AM Short-term use only
Diethylpropion DEA-IV 25 mg tid; 75 mg in AM Short-term use only
Phendimetrazine DEA-III 35 mg tid before meals Short-term use only
Phentermine DEA-IV 18.75 mg–37.5 mg tid; Short-term use only
15 mg/d–30 mg/d
in AM of slow-release
MEDICAL THERAPY FOR OBESITY 1227
receptor, and inhibits intake of sweet or palatable food in marmosets as well

as high-fat food intake in rats, but not in rats fed standard chow. In addition
to inhibiting intake of highly palatable food, rimonabant may increase
energy expenditure in rodents. Genetically engineered mice that lack the
CB-1 receptor are lean and resistant to diet-induced obesity.
The results of three phase III trials of rimonabant for the treatment of obe-
sity have been published [7–9] and results from a fourth have been presented
at a scientific meeting. In a one-year trial, 1507 subjects with body mass index
(BMI) greater than 30 kg/m2 (or greater than 27 kg/m2 with treated or
untreated dyslipidemia, hypertension, or both) were randomly assigned to
receive rimonabant (5 mg per day or 20 mg per day) or placebo, in addition
to a diet calculated to produce a 600 kcal per day deficit [7]. The mean weight
loss (plus or minus the standard deviation) at 1 year was 3.4 kg plus or mi-
nus 5.7 kg, 6.6 kg plus or minus 7.2 kg, and 1.8 kg plus or minus 6.4 kg in
the rimonabant 5-mg, 20-mg, and placebo groups, respectively. More pa-
tients in the group receiving 20 mg per day of rimonabant as compared
with placebo achieved a weight loss of greater than 5% (51% versus 19%)
or 10% (27% versus 7%). The 20-mg per day dose of rimonabant produced
significantly greater improvements in waist circumference, high-density lipo-
protein (HDL), triglycerides, insulin resistance, and prevalence of the meta-
bolic syndrome than placebo. Side effects, including mood changes, nausea
and vomiting, diarrhea, headache, dizziness, and anxiety were more frequent
in the rimonabant 20-mg group than the 5-mg or placebo groups. However,
dropout rates were similar in all three groups. Thus, rimonabant 20 mg per
day resulted in clinically meaningful weight loss, reduction in waist circum-
ference, and improvements in several metabolic risk factors.
In a second 1-year study, 1018 obese subjects with dyslipidemia and
a BMI between 27 kg/m2 and 40 kg/m2 were randomized equally to placebo,
rimonabant 5 mg per day, or rimonabant 20 mg per day [8]. Weight loss was
2% in the placebo group and 8.5% in the 20-mg rimonabant group. In the
group receiving 20 mg per day of rimonabant, waist circumference was
reduced 9 cm, triglycerides were reduced by 15%, and HDL cholesterol
was increased by 23% as compared with 3.5 cm, 3% and 12%, respectively,
in the placebo group. In the 20-mg per day groups, the low-density lipopro-
tein (LDL) particle size increased, adiponectin increased, glucose decreased,
insulin decreased, c-reactive protein decreased, and the metabolic syndrome
prevalence was cut in half. There was no increase in depression or anxiety,
and neither pulse nor blood pressure increased.
The third study conducted in North America was a 2-year double-blind
trial that randomized 3,040 obese subjects without diabetes to placebo, 5-mg
rimonabant or 20-mg rimonabant (Fig. 1) [9]. At 1 year, half of the rimona-
bant groups were rerandomized to placebo. At 1-year, weight loss was
2.8 kg in the placebo group and 8.6 kg in the 20-mg per day rimonabant
group. In the fourth study, 1,047 subjects with type 2 diabetes were random-
ized to 1 year of treatment. The weight losses were not quite as great as in
1228 BRAY
Effect of Rimonabant on Body Weight

over 2 Years
0
Placebo
-2
Weight Loss (kg)
-4
Rimonabant 5 mg
-6
Placebo
-8
Rimonabant 20 mg
-10
0 12 24 36 48 60 72 84 96 108
Weeks of Treatment
Fig. 1. Effect on body weight of rimonabant in a 2-year, randomized, placebo-controlled clin-

ical trial (Adapted from Pi-Sunyer FX, Aronne LJ, Heshmati HM, et al. Effect of rimonabant,
a cannabinoid-1 receptor blocker, on weight and cardiometabolic risk factors in overweight
or obese patients: RIO-North America: a randomized controlled trial. JAMA 2006;295(7):
761–75; with permission).
the other three groups, but there was consistent improvement in the comor-
bid risk factors.
Safety
There were significantly more psychiatric side effects with the higher dose
of rimonabant in the first year of treatment. Because patients with depression
were excluded from the initial phase III studies, there is no information on
how this drug works in depressed patients or those taking antidepressants.
Sibutramine
Sibutramine is approved by the Food and Drug Administraton (FDA) for
long-term use. Sibutramine has been evaluated extensively in several placebo-
controlled, double-blind multicenter clinical trials lasting 6 to 24 months and
including men and women of all ethnic groups, with ages ranging from 18
years to 65 years and with a BMI between 27 kg/m2 and 40 kg/m2 [2,4–6,
10,11]. In a clinical trial lasting 8 weeks, sibutramine produced a dose-de-
pendent weight loss with doses of 5 mg and 20 mg per day [6]. In a 6-month
dose-ranging study of 1,047 subjects, 67% treated with sibutramine achieved
a 5% weight loss from baseline, and 35% lost 10% or more [6]. There was
a clear dose-response effect in this 24-week trial, and subjects regained weight
when the drug was stopped, indicating that the drug remained effective when
used. Data from this multicenter trial are shown in Fig. 2 [6].
In a 1-year trial of 456 subjects who received sibutramine (10 mg or 15 mg
per day) or placebo, 56% of those who stayed in the trial for 12 months lost at
End of
Treatment
0
Mean % Change in Weight
-2
-4
Placebo
-6 1 mg
5 mg
10 mg
-8 15 mg
20 mg
30 mg
-10
0 4 8 12 16 20 24 2 2 3
Treatment
Fig. 2. Six-month randomized placebo-controlled dose-ranging trial with sibutramine at six

doses and placebo. (Reproduced from Bray et al. Sibutramine produces dose-related weight
loss. Obes Res 1999;7:189–98; with permission.)
least 5% of their initial body weight, and 30% of the subjects lost 10% of their
initial body weight while taking the 10-mg dose [12]. In a third trial of subjects
who initially lost weight eating a very low-calorie diet before being random-
ized to sibutramine or placebo, sibutramine (10 mg per day) produced addi-
tional weight loss, whereas the placebo-treated subjects regained weight [13].
The sibutramine trial of obesity reduction and maintenance lasted 2 years
and provided evidence for weight maintenance [14]. Seven centers partici-
pated in this trial, in which subjects were initially enrolled in a 6-month
open-label phase and treated with 10 mg per day of sibutramine. Of the sub-
jects who lost more than 8 kg, two thirds were then randomized to sibutr-
amine, and one third to placebo. During the 18-month double-blind phase
of this trial, the placebo-treated subjects steadily regained weight, maintain-
ing only 20% of their weight loss at the end of the trial. In contrast, the
subjects treated with sibutramine maintained their weight for 12 months
and then regained an average of only 2 kg, thus maintaining 80% of their
initial weight loss after 2 years [14]. Despite the higher weight loss with si-
butramine at the end of the 18 months of controlled observation, the blood
pressure levels of the sibutramine-treated subjects were still higher than in
the subjects treated with placebo.
The possibility of using sibutramine as intermittent therapy has been
tested in a randomized, placebo-controlled trial lasting 52 weeks [15]. The
subjects randomized to sibutramine received one of two regimens. One group
received continuous treatment with 15 mg per day for 1 year, and the other
had two 6-week periods when sibutramine was withdrawn. During these
periods when the drug was replaced by placebo, there was a small regain
in weight that was lost when the drug was again resumed. At the end of
the trial, the continuous-therapy and intermittent-therapy groups had lost
the same amount of weight.
1230 BRAY
Some trials have reported the use of sibutramine to treat patients with
hypertension. In a 52-week trial involving subjects with hypertension whose
blood pressure levels were controlled with calcium channel blockers with or
without beta-blockers or thiazides [16], sibutramine doses were increased
from 5 mg to 20 mg per day during the first 6 weeks. Weight loss was
significantly greater in the sibutramine-treated subjects, averaging 4.4 kg
(4.7%), as compared with 0.5 kg (0.7%) in the placebo-treated group. Dia-
stolic blood pressure levels decreased 1.3 mm Hg in the placebo-treated
group, and increased 2 mm Hg in the sibutramine-treated group. The sys-
tolic blood pressure levels increased 1.5 mm Hg in the placebo-treated group
and 2.7 mm Hg in the sibutramine-treated group. Heart rate was unchanged
in the placebo-treated subjects, but increased by an average of 4.9 beats per
minute in the sibutramine-treated subjects.
In two studies, subjects with diabetes were treated for 12 weeks or 24 weeks
with sibutramine. In the 12-week trial, subjects with diabetes treated with si-
butramine at 15 mg per day lost 2.4 kg (2.8%), compared with 0.1 kg (0.12%)
in the placebo group. In this study, hemoglobin A1C levels decreased 0.3% in
the drug-treated group and remained stable in the placebo group. Fasting glu-
cose values decreased 0.3 mg/dL in the drug-treated subjects and increased
1.4 mg/dL in the placebo-treated group. In the 24-week trial, the dose of si-
butramine was increased from 5 mg to 20 mg per day over 6 weeks [17].
Among those who completed the treatment, weight loss was 4.3 kg (4.3%)
in the sibutramine-treated subjects, as compared with 0.3 kg (0.3%) in pla-
cebo-treated subjects. Hemoglobin A1C levels decreased 1.67% in the
drug-treated group, compared with 0.53% in the placebo-treated group.
These changes in glucose and hemoglobin A1C levels were expected from
the amount of weight loss associated with drug treatment.
Sibutramine has also been used in children [18–20]. In a large 12-month-
long multi-center trial, 498 adolescents aged 12 to 16 were randomized to
treatment with placebo or sibutramine, 10 mg per day, which could be
increased to 15 mg per day in those who had not lost more than 10% of their
body weight by 6 months [20]. After 12 months, the mean absolute change
in BMI was 2.9 kg/m2 (8.2%) in the sibutramine group as compared
with 0.3 kg/m2 (0.8%) in the placebo group (P! 0.001). Triglycerides,
HDL-cholesterol, and insulin sensitivity improved, and there was no signif-
icant difference in the changes in either systolic or diastolic blood pressure.
Sibutramine has also been studied as part of a behavioral weight-loss
program. With sibutramine alone and minimal behavioral intervention,
the weight loss over 12 months was approximately 5.0 kg plus or minus
7.4 kg over 12 months. Behavior modification alone produced a weight
loss of 6.7 kg plus or minus 7.9 kg. Adding a brief behavioral therapy
session to a group that also received sibutramine produced a slightly larger
weight loss of 7.5 kg plus or minus 8.0 kg. When the intensive lifestyle
intervention was combined with sibutramine, the weight loss increased
to 12.1 kg plus or minus 9.8 kg [21].
Sibutramine is available in 5-, 10-, and 15-mg doses; 10 mg per day as

a single dose is the recommended starting level, with titration up or down
depending on response. Doses higher than 15 mg per day are not recommen-
ded. Of the patients who lost 2 kg (4 lb) in the first 4 weeks of treatment,
60% achieved a weight loss of more than 5%, as compared with less than
10% in those who did not lose 2 kg (4 lb) in 4 weeks. Combining data
from the 11 studies on sibutramine showed a reduction in triglyceride, total
cholesterol, and LDL cholesterol levels, and an increase in HDL cholesterol
levels, that were related to the magnitude of the weight loss.
Safety
Sibutramine increases blood pressure levels in normotensive patients or
prevents the decrease that might have occurred with weight loss. The
magnitude of the change may be dose related, so lower doses are preferred.
Systolic and diastolic blood pressure levels increase an average of plus
0.8 mm Hg and plus 0.6 mm Hg, and pulse increases approximately four
to five beats per minute. Caution should be used when combining sibutr-
amine with other drugs that may increase blood pressure levels. Sibutramine
is contraindicated in patients with a history of coronary artery disease, con-
gestive heart failure, cardiac arrhythmias, or stroke. Sibutramine should not
be used with selective serotonin reuptake inhibitors or monoamine oxidase
inhibitors, and there should be a 2-week interval between terminating mono-
amine oxidase inhibitors and beginning sibutramine. Because sibutramine is
metabolized by the cytochrome P-450 enzyme system (isozyme CYP3A4), it
may interfere with the metabolism of erythromycin and ketoconazole.
Sympathomimetic drugs: pharmacology and efficacy

The sympathomimetic drugs, benzphetamine, diethylpropion, phendime-
trazine and phentermine, are grouped together because they act like norepi-
nephrine. Drugs in this group work by a variety of mechanisms, including
the blockade of norepinephrine reuptake from synaptic granules [6].
All of these drugs are absorbed orally and reach peak blood concen-
trations within a short period. In addition, the half-life in blood is short
for all except the metabolites of sibutramine, which have a long half-life.
The two metabolites of sibutramine are active, but this is not true for the
metabolites of other drugs in this group. Liver metabolism inactivates a large
fraction of these drugs before excretion. Side effects include dry mouth,
constipation, and insomnia. Food intake is suppressed either by delaying
the onset of a meal or by producing early satiety.
The efficacy of an appetite-suppressing drug can be established through
randomized, double-blind clinical trials that show a significantly greater
weight loss than in the placebo group, and a weight loss that is more than
5% below that with placebo [2,4,10]. Clinical trials of sympathomimetic
drugs conducted before 1975 were generally short because it was widely
1232 BRAY
believed that short-term treatment would ‘‘cure’’ obesity. This was un-
founded optimism, and because the trials had a short duration and often
used a crossover design, they provided few long-term data. The focus here
is on longer-term trials lasting 24 weeks or more that include an adequate
control group.
One of the longest of these clinical trials of drugs in this group lasted 36
weeks and compared placebo treatment with continuous phentermine or
intermittent phentermine [6]. Both continuous and intermittent phentermine
therapy produced more weight loss than placebo. In the drug-free periods,
the subjects treated intermittently slowed their weight loss, only to lose
weight more rapidly when the drug was reinstituted. Phentermine and dieth-
ylpropion are classified by the US Drug Enforcement Agency as schedule IV
drugs; benzphetamine and phendimetrazine are schedule III drugs. This reg-
ulatory classification indicates the United States government’s belief that
they have the potential for abuse, although this potential appears to be
very low. Phentermine and diethylpropion are approved for only a ‘‘few
weeks,’’ which is usually interpreted as up to 12 weeks. Weight loss with
phentermine and diethylpropion persists for the duration of treatment,
suggesting that tolerance does not develop to these drugs. If tolerance were
to develop, the drugs would be expected to lose their effectiveness and
patients would require increased amounts of the drug to maintain weight
loss. This does not occur.
Safety of sympathomimetic drugs

The side-effect profiles for sympathomimetic drugs are similar [5,6].
These agents produce insomnia, dry mouth, asthenia, and constipation.
The safety of older sympathomimetic appetite suppressant drugs has been
the subject of considerable controversy because dextroamphetamine is
addictive. The sympathomimetic drugs phentermine, diethylpropion, benz-
phetamine, and phendimetrazine have very little abuse potential, as assessed
by the low rate of reinforcement when the drugs are self-injected intrave-
nously by test animals [6]. Sympathomimetic drugs can also increase blood
pressure levels.
Drugs that act on peripheral metabolism

Lipase inhibitor orlistat: pharmacology and efficacy
Orlistat is a potent and selective inhibitor of pancreatic lipase that
reduces the intestinal digestion of fat. The drug has a dose-dependent effect
on fecal fat loss, increasing it to approximately 30% on a diet that has 30%
of its energy as fat. Orlistat has little effect in subjects eating a low-fat diet,
as might be anticipated from its mechanism of action [6].
A number of long-term clinical trials (1–2 years) with orlistat have been
published [2,4,10,11]. The results of a 2-year trial are shown in Fig. 3. The
Orlistat and Body Weight

Percent change in body weight -0
Placebo tid
-1
Orlistat 120 mg
-2 tid
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
- 0 1 2 3 4 5 6 7 8 9 10 11
Week
S D D
Mildly hypocaloric Weight maintenance

(eucaloric diet)
Fig. 3. Double-blind randomized clinical trial of orlistat versus placebo with a rerandomization
of participants after the first year (Adapted from Sjostrom L, Rissanen A, Andersen T, et al.
Randomised placebo-controlled trial of orlistat for weight loss and prevention of weight regain
in obese patients. European Multicentre Orlistat Study Group. Lancet 1998;352(9123):167–72;
with permission).
trial consisted of two parts: In the first year, subjects received a hypocaloric
diet calculated to be 500 kcal per day less than the subject’s requirements
[22]. During the second year, the diet was calculated to maintain weight.
By the end of year one, the placebo-treated subjects lost 6.1% of their initial
body weight and the drug-treated subjects lost 10.2%. The subjects were
randomized at the end of year one. Those switched from orlistat to placebo
gained weight from 10% to 6% below baseline. Those switched from
placebo to orlistat lost weight from 6% to 8.1% below baseline, which
was essentially identical to the 7.9% loss in the subjects treated with
orlistat for the full 2 years.
In a second 2-year study, 892 subjects were randomized. One group re-
mained on placebo throughout the 2 years (97 subjects), and a second group
remained on orlistat (120 mg three times per day) for 2 years (109 subjects). At
the end of 1 year, two-thirds of the group treated with orlistat for 1 year were
changed to orlistat (60 mg three times per day) (102 subjects), and the others
were switched to placebo (95 subjects). After 1 year, the weight loss was 8.7
kg in the orlistat-treated group and 5.8 kg in the placebo group (P! 0.001).
During the second year, those switched to placebo after 1 year reached the
same weight as those treated with placebo for 2 years (4.5% in those with
placebo for 2 years and 4.2% in those switched to placebo during year two).
In a third 2-year study, 783 subjects remained in the placebo or orlistat-
treated groups at 60 mg or 120 mg three times per day for the entire 2 years.
1234 BRAY
After 1 year with a weight-loss diet, the placebo group lost 7 kg, which was
significantly less than the 9.6 kg lost by the group treated with orlistat,
60 mg three times per day, or the 9.8 kg lost by the group treated with orli-
stat, 120 mg three times per day. During the second year, when the diet was
liberalized to a ‘‘weight maintenance’’ diet, all three groups regained some
weight. At the end of 2 years, the subjects in the placebo group were 4.3 kg
below baseline, the subjects treated with orlistat, 60 mg three times per day,
were 6.8 kg below baseline, and the subjects who took orlistat 120 mg three
times per day were 7.6 kg below baseline.
The final 2-year trial that has been published evaluated 796 subjects in
a general-practice setting. After 1 year of treatment with orlistat 120 mg
three times per day, the orlistat-treated patients (n ¼ 117) had lost 8.8 kg,
as compared with 4.3 kg in the placebo group (n ¼ 91). During the second
year, when the diet was liberalized to ‘‘maintain body weight,’’ both groups
regained some weight. At the end of 2 years, the orlistat group was 5.2 kg
below their baseline weight, compared with 1.5 kg below baseline for the
group treated with placebo.
A 4-year double-blind, randomized, placebo-controlled trial with
orlistat treated a total of 3,304 overweight subjects, 21% of whom had
impaired glucose tolerance [23]. The lowest body weight was achieved
during the first year, and was more than 11% below baseline in the orli-
stat-treated group and 6% below baseline in the placebo-treated group.
Over the remaining 3 years of the trial, there was a small regain in weight,
such that by the end of 4 years, the orlistat-treated subjects were 6.9%
below baseline, compared with 4.1% for those receiving placebo. The
trial also showed a 37% reduction in the conversion of subjects from
impaired glucose tolerance to diabetes; essentially all of this benefit oc-
curred in the subjects with impaired glucose tolerance at enrollment into
the trial.
Orlistat has also been used to treat obese children. A multicenter trial
tested the effect of orlistat in 539 obese adolescents [24]. Subjects were
randomized to placebo or orlistat 120 mg three times a day and a mildly
hypocaloric diet containing 30% fat. By the end of the study, BMI had
decreased 0.55 kg/m2 in the drug-treated group, but had increased more
than 0.31 kg/m2 in the placebo group. By the end of the study, weight had
increased by 0.51 kg in the orlistat-treated group, compared with more than
3.14 kg in the placebo-treated group. This difference was due to differences
in body fat. The side effects were gastrointestinal in origin, as expected from
the mode of action of orlistat.
Weight maintenance with orlistat was evaluated in a 1-year study. Sub-
jects were enrolled if they had lost more than 8% of their body weight
over 6 months while eating a 1,000-kcal per day (4,180-kJ/day) diet. The
729 subjects were randomized to receive placebo or orlistat at 30 mg, 60
mg, or 120 mg three times per day for 12 months. At the end of this time,
the placebo-treated patients had regained 56% of their body weight, as
compared with 32.4% regain in the group treated with orlistat, 120 mg three
times per day. The other two doses of orlistat were not different from
placebo in preventing the regain of weight.
Patients with diabetes treated with orlistat, 120 mg three times per day for
1 year, lost 6.5% of their body weight, as compared with a 4.2% loss in
the placebo-treated group [25]. The subjects with diabetes also showed a sig-
nificantly greater decrease in hemoglobin A1C levels. In another study of
orlistat and weight loss, investigators pooled data on 675 subjects from three
of the 2-year studies, described previously, in which glucose tolerance tests
were available [26]. During treatment, 6.6% of the subjects taking orlistat
converted from a normal to an impaired glucose tolerance test, compared
with 10.8% in the placebo-treated group. None of the orlistat-treated sub-
jects who originally had normal glucose tolerance developed diabetes,
compared with 1.2% in the placebo-treated group. Of those who initially
had normal glucose tolerance, 7.6% in the placebo group but only 3% in
the orlistat-treated group developed diabetes.
Safety of orlistat
Orlistat is not absorbed to any significant degree, and its side effects are
thus related to the blockade of triglyceride digestion in the intestine [27].
Fecal fat loss and related GI symptoms are common initially, but they sub-
side as patients learn to use the drug. The quality of life in patients treated
with orlistat may improve despite concerns about GI symptoms. Orlistat
can cause small but significant decreases in fat-soluble vitamins. Levels
usually remain within the normal range, but a few patients may need vita-
min supplementation. Because it is impossible to tell which patients need
vitamins, it is wise to provide a multivitamin routinely, with instructions
to take it before bedtime. Orlistat does not seem to affect the absorption
of other drugs, with the exception of acyclovir.
Combining orlistat and sibutramine

Because orlistat works peripherally to reduce triglyceride digestion in the
GI tract, and sibutramine works on noradrenergic and serotonergic reup-
take mechanisms in the brain, their mechanisms of action do not overlap
and combining them might provide additive weight loss. To test this possi-
bility, investigators randomly assigned patients to orlistat or placebo after
1 year of treatment with sibutramine [28]. During the additional 4 months
of treatment, there was no further weight loss. Thus, we have no data that
adding orlistat and sibutramine is beneficial.
Drugs that increase energy expenditure

There are no effective candidate drugs in this class. Trials of b-3 adrenergic
agonists have not yielded a successful molecule [29].
1236 BRAY
Drugs that have been used for obesity, but are not approved for this purpose
Fluoxetine and sertraline
Fluoxetine and sertraline are both selective serotonin reuptake inhibitors
that block serotonin transporters, thus prolonging the action of serotonin.
These drugs both reduce food intake. In a 2-week placebo-controlled trial, flu-
oxetine at a dose of 60 mg per day produced a 27% decrease in food intake
[30]. Both fluoxetine and sertraline are approved by the FDA for treatment
of depression. In 8- to 16-week long clinical trials with depressed subjects,
sertraline gave an average weight loss of 0.45 kg to 0.91 kg. Fluoxetine,
at a dose of 60 mg per day (three times the usual dose for treatment of depres-
sion), was effective in reducing body weight in overweight subjects. A meta-
analysis of six studies using fluoxetine showed a wide range of results, with
a mean weight loss in one study of 14.5 kg and a weight gain of more than
0.40 kg in another study [4]. In the meta-analysis by Avenell and colleagues
[11], the weight loss at 12 months was 0.33 kg (95%CI 1.49 kg to 0.82
kg). Goldstein and colleagues [31] reviewed the trials with fluoxetine that in-
cluded one 36-week trial in type 2 diabetic subjects, a 52-week trial in subjects
with uncomplicated overweight, and two 60-week trials in subjects with dys-
lipidemia, diabetes, or both. A total of 719 subjects were randomized to fluox-
etine and 722 to placebo. Five hundred twenty-two subjects on fluoxetine and
504 subjects on placebo completed 6 months of treatment. Weight losses in the
placebo and fluoxetine groups at 6 months and 1 year were 2.2 kg, 4.8 kg,
and 1.8 kg, 2.4 kg, respectively. The regain of 50% of the lost weight
during the second 6 months of treatment with fluoxetine makes this drug
inappropriate for the long-term treatment of obesity. Fluoxetine and sertra-
line, although not good drugs for long-term treatment of obesity, may be pre-
ferred for the treatment of depressed obese patients over some of the tricyclic
antidepressants that are associated with significant weight gain.
Bupropion
Bupropion is a norepinephrine and dopamine reuptake inhibitor that is
approved for the treatment of depression and for help in smoking cessation.
In one clinical trial, 50 overweight subjects were randomized to bupropion
or placebo for 8 weeks, with a blinded extension for responders to 24 weeks.
The dose of bupropion was increased to a maximum of 200 mg twice daily in
conjunction with a calorie-restricted diet. At 8 weeks, 18 subjects in the bup-
ropion group lost 6.2% plus or minus 3.1% of body weight, as compared
with 1.6% plus or minus 2.9% for the 13 subjects in the placebo group
(P! 0.0001). After 24 weeks, the 14 responders to bupropion lost 12.9%
plus or minus 5.6% of initial body weight, of which 75% was fat as determined
by dual energy X-ray absorptometry [32].
Two multicenter clinical trials, one in obese subjects with depressive
symptoms and one in uncomplicated overweight subjects, followed this
study. In the study of overweight subjects with depressive symptom ratings

of 10 to 30 on a Beck Depression Inventory, 213 subjects were randomized
to 400 mg per day of bupropion and 209 subjects were assigned to placebo
for 24 weeks. The 121 subjects in the bupropion group who completed the
trial lost 6.0% plus or minus 0.5% of initial body weight, as compared
with 2.8% plus or minus 0.5% in the 108 subjects in the placebo group
(P! 0.0001) [33]. The study in uncomplicated overweight subjects random-
ized 327 subjects to bupropion 300 mg per day, bupropion 400 mg per day,
or placebo in equal proportions. At 24 weeks, 69% of those randomized re-
mained in the study and the percent losses of initial body weight were 5%
plus or minus 1%, 7.2% plus or minus 1%, and 10.1% plus or minus 1%
for the placebo, bupropion 300 mg, and bupropion 400 mg groups, respec-
tively (P! 0.0001). The placebo group was randomized to the 300-mg or
400-mg group at 24 weeks and the trial was extended to week 48. By the
end of the trial the dropout rate was 41%, and the weight losses in the bu-
propion 300-mg and bupropion 400-mg groups were 6.2% plus or minus
1.25% and 7.2% plus or minus 1.5% of initial body weight, respectively
[34]. Thus, it appears that nondepressed subjects may respond to bupropion
with weight loss to a greater extent than those with depressive symptoms.
Topiramate
Topiramate is approved for treatment of selected seizure disorders. It is
a weak carbonic anhydrase inhibitor. Topiramate also modulates the effects
at receptors for the gamma-aminobutyric acid (GABAA) receptor and
the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/
kainate subtype of the glutamate receptor. This drug also exhibits state-
dependent blockade of voltage-dependent sodium (Naþ) or calcium (Ca2þ)
channels. These mechanisms are believed to contribute to its antiepileptic
properties. The modulation of GABAA receptors may provide one potential
mechanism to reduce food intake, although other mechanisms, yet to be
described, may be more important in defining its effects on body weight [35].
Topiramate is an antiepileptic drug that was discovered to give weight loss
during the clinical trials for epilepsy. Weight losses of 3.9% of initial weight
were seen at 3 months, and losses of 7.3% of initial weight were seen at 1
year [36]. Bray and colleagues [37] reported a 6-month, placebo-controlled,
dose-ranging study of topiramate. Three hundred eighty-five obese subjects
were randomized to placebo or topiramate at 64 mg per day, 96 mg per
day, 192 mg per day, or 384 mg per day. These doses were gradually reached
by a tapering increase and were reduced in a similar manner at the end of the
trial. Weight loss from baseline to 24 weeks was 2.6%, 5%, 4.8%,
6.3%, and 6.3% in the placebo, 64 mg, 96 mg, 192 mg, and 384 mg groups,
respectively. The most frequent adverse events were paresthesias, somnolence,
and difficulty with concentration, memory, and attention.
This trial was followed by two other multicenter trials. The first trial
randomized 1289 obese subjects to placebo or topiramate 89 mg per day,
1238 BRAY
192 mg per day, or 256 mg per day. This trial was terminated early because
of the sponsor’s decision to pursue a time-release form of the drug. The 854
subjects who completed 1 year of the trial before it was terminated lost
1.7%, 7%, 9.1%, and 9.7% of their initial body weight in the
placebo, 89 mg, 192 mg, and 256 mg groups, respectively. Subjects in the
topiramate groups had significant improvement in blood pressure and glu-
cose tolerance [38]. The second trial enrolled 701 subjects who were treated
with a very low-calorie diet to induce an 8% loss of initial body weight. The
560 subjects who achieved an 8% weight loss were randomized to topira-
mate 96 mg per day, 192 mg per day, or placebo. This study was also termi-
nated. At the time of early termination, 293 subjects had completed 44
weeks. The topiramate groups lost 15.4% and 16.5% of their baseline
weight, while the placebo group lost 8.9% [39]. Although topiramate is still
available as an antiepileptic drug, the development program to obtain an
indication for overweight was terminated by the sponsor because of the
associated adverse events.
Zonisamide
Zonisamide is an antiepileptic drug that has serotonergic and dopaminer-
gic activity in addition to inhibiting sodium and calcium channels. Weight
loss was noted in the clinical trials for the treatment of epilepsy, again sug-
gesting a potential agent for weight loss. Gadde and colleagues [40] tested
this possibility by performing a 16-week randomized controlled trial in 60
obese subjects. Subjects were placed on a calorie-restricted diet and random-
ized to zonisamide or placebo. The zonisamide was started at 100 mg per
day and increased to 400 mg per day. At 12 weeks, those subjects who
had not lost 5% of initial body weight were increased to 600 mg per day.
The zonisamide group lost 6.6% of initial body weight at 16 weeks, as
compared with 1% in the placebo group. Thirty-seven subjects completing
the 16-week trial elected to continue for 32 weeks: 20 in the zonisamide
group and 17 in the placebo group. At the end of 32 weeks, the 19 subjects
in the zonisamide group lost 9.6% of their initial body weight, as com-
pared with 1.6% for the 17 subjects in the placebo group.
Lamotrigine
Lamotrigine is a third antiepileptic drug that has been evaluated for its
effects on body weight [41]. In a double-blind, randomized, placebo-
controlled trial, the dose of lamotrigine was escalated from 25 mg per day
to 200 mg per day over 6 weeks. The effect on weight loss was compared
with placebo treatment over 26 weeks in 40 healthy overweight (BMI
30 kg/m2 –40 kg/m2) adults over 18 years of age. At the end of the trial
body weight was marginally lower (P ¼ 0.062) in the lamotrigine-treated
group (6.4 kg) than in the placebo-treated group (1.2 kg) [42].
Metformin
Metformin is a biguanide that is approved for the treatment of diabetes
mellitus. This drug reduces hepatic glucose production, decreases intestinal
absorption from the GI tract, and enhances insulin sensitivity. In clinical tri-
als where metformin was compared with sulfonylureas, it produced weight
loss [6]. In one French trial, biguanides and the prevention of the risk of
obesity (BIGPRO), metformin was compared with placebo in a 1-year
multi-center study in 324 middle-aged subjects with upper body adiposity
and insulin resistance syndrome (metabolic syndrome). The subjects on met-
formin lost significantly more weight (1 kg–2 kg) than the placebo group,
and the study concluded that metformin may have a role in the primary pre-
vention of type 2 diabetes [43]. In a meta-analysis of three of these studies,
Avenell and colleagues [11] reported a weight loss at 12 months of 1.09 kg
(95%CI 2.29 kg to 0.11 kg).
The best trial of metformin, however, is the Diabetes Prevention Program
(DPP) study of individuals with impaired glucose tolerance. This study
included a double-blind comparison of metformin 850 mg twice a day versus
placebo. During the 2.8 years of this trial, the 1073 subjects treated with
metformin lost 2.5% of their body weight (P! 0.001), compared with
the 1082 subjects treated with placebo, and the conversion from impaired
glucose tolerance to diabetes was reduced by 31%, compared with placebo.
In the DPP trial, metformin was more effective in reducing the development
of diabetes in the subgroup who were most overweight, and in the younger
members of the cohort [44]. Although metformin does not produce
enough weight loss (5%) to qualify as a ‘‘weight-loss drug’’ (FDA criteria
require greater than or equal to 5% weight loss), it would appear to be
a very useful choice for overweight individuals who have diabetes or are
at high risk for diabetes. One area where metformin has found use is in
treating overweight women with the polycystic ovary syndrome, where the
modest weight loss may contribute to increased fertility and reduced insulin
resistance [45].
Pramlintide
Amylin is a peptide found in the beta cell of the pancreas, that is cose-
creted along with insulin to circulate in the blood. Both amylin and insulin
are deficient in type 1 diabetics where beta cells are immunologically
destroyed. Pramlintide, a synthetic amylin analog, has a prolonged biolog-
ical half-life [46]. Pramlintide is approved by the FDA for the treatment of
diabetes. Unlike insulin and many other diabetic medications, pramlintide is
associated with weight loss. In a study where 651 subjects with type 1 diabe-
tes were randomized to placebo or subcutaneous pramlintide 60 mcg three
or four times a day along with an insulin injection, the hemoglobin A1c de-
creased 0.29% to 0.34%, and weight decreased 1.2 kg relative to placebo
[47]. Maggs and colleagues [48] analyzed the data from two 1-year studies in
1240 BRAY
insulin-treated type 2 diabetic subjects randomized to pramlintide 120 mcg

twice a day or 150 mcg three times a day. Weight decreased by 2.6 kg and
hemoglobin A1c decreased 0.5%. When weight loss was then analyzed by
ethnic group, African Americans lost 4 kg, Caucasians lost 2.4 kg,
and Hispanics lost 2.3 kg, and the improvement in diabetes correlated
with the weight loss, suggesting that pramlintide is effective in ethnic groups
with the greatest burden from overweight. The most common adverse
event was nausea, which was usually mild and confined to the first 4 weeks of
therapy.
Exenatide
Glucagon-like peptide-1 (GLP-1) is derived from the processing of the
proglucagon peptide, which is secreted by L-cells in the terminal ileum in re-
sponse to a meal. Increased GLP-1 inhibits glucagon secretion, stimulates
insulin secretion, stimulates gluconeogenesis, and delays gastric emptying
[49]. It has been postulated to be responsible for the superior weight loss
and superior improvement in diabetes seen after gastric bypass surgery for
overweight [50,51]. GLP-1 is rapidly degraded by dipeptidyl peptidase-4
(DPP-4), an enzyme that is elevated in the obese. Bypass operations
for overweight increase GLP-1, but do not change the levels of DPP-4
[46,52].
Exenatide (Exendin-4) is a 39-amino acid peptide that is produced in the
salivary gland of the Gila monster lizard. It has 53% homology with GLP-1,
but it has a much longer half-life. Exenatide decreases food intake and body
weight gain in Zucker rats while lowering HgbA1c [53]. It also increases
beta-cell mass to a greater extent than would be expected for the degree
of insulin resistance [54]. Exendin-4 induces satiety and weight loss in
Zucker rats with peripheral administration, and crosses the blood-brain bar-
rier to act in the central nervous system [55,56]. Exenatide is approved by
the FDA for treatment of type 2 diabetics who are inadequately controlled
while being treated with either metformin or sulfonylureas.
In human beings, exenatide reduces fasting and postprandial glucose
levels, slows gastric emptying, and decreases food intake by 19% [57]. The
side effects of exenatide in human beings are headache, nausea, and vomit-
ing that are lessened by gradual dose escalation [58]. Several clinical trials of
30 weeks duration have been reported using exenatide at 10 mcg subcutane-
ously per day or a placebo [59–61]. In one trial with 377 type 2 diabetic sub-
jects who were failing maximal sulfonylurea therapy, exenatide produced
a fall of 0.74% more in HgbA1c than placebo. Fasting glucose also de-
creased and there was a progressive weight loss of 1.6 kg [61]. The interest-
ing feature of this weight loss is that it occurred without lifestyle change,
diet, or exercise. In a 26-week randomized control trial, exenatide produced
a 2.3 kg weight loss, compared with a gain of more than 1.8 kg in the
group receiving insulin glargine [62].
Somatostatin
Somatostatin is a small peptide that is released in the GI tract and in the
brain. It inhibits the release of most peptides, including insulin, glucagon,
and growth hormone, among others. Overweight caused by hypothalamic
injury has been associated with hypersecretion of insulin [63]. Lustig and
colleagues [64] treated eight children with overweight caused by hypotha-
lamic damage with octreotide injections to decrease insulin hypersecretion.
These children gained 6 kg in the 6 months before octreotide treatment
and lost 4.8 kg in the 6 months on octreotide, an analog of somatostatin.
The weight loss was correlated with the reduction of insulin secretion on
a glucose tolerance test. This open-label trial was followed by a randomized
controlled trial of octreotide treatment in children with hypothalamic
overweight. The subjects received octreotide 5 mg per kg to 15 mg per kg
per day or placebo for 6 months. The children on octreotide gained 1.6 kg,
compared with 9.1 kg for those in the placebo group [65].
This same group of investigators postulated that there might be a subset
of obese subjects who were insulin hypersecretors and that these subjects
would respond with weight loss to treatment with octreotide. Following
an oral glucose tolerance test in which glucose and insulin were measured,
44 subjects were treated with a long-acting form of octreotide 40 mg per
month for 6 months. These subjects lost weight, reduced food intake, and
had a reduced carbohydrate intake. Weight loss was greatest in those with
insulin hypersecretion and the amount of weight lost was correlated with
the reduction in insulin hypersecretion [66]. In a multicenter randomized
controlled trial, 172 obese subjects (144 women and 28 men) who had insulin
hypersecretion during a glucose tolerance test at screening received long-
acting octreotide in doses of 20 mg per month, 40 mg per month, 60 mg
per month, or placebo for 6 months. The greatest weight loss was 3.5%
to 3.8% of initial body weight in the two higher dose groups, an amount
that was statistically significant, but not enough to meet the criteria for
approval by the FDA [67,68].
Octreotide has been shown to decrease gastric emptying [69]. Treatment
of patients with the Prader-Willi syndrome who have elevated ghrelin levels
does not cause weight loss, but ghrelin levels are normalized. The reason for
the lack of weight loss was postulated to be the reduction of PYY, a satiating
gastrointestinal hormone that also decreased [70].
Drugs in clinical trial

Growth hormone and growth hormone fragment
Growth hormone is a pituitary peptide that is essential for the adolescent
growth spurt. Bioengineered growth hormone is widely used to treat short
stature, as well as growth hormone deficiency in adults [71], and has been
1242 BRAY
used by athletes to build muscle, as one of its effects is to enhance protein

accretion. Growth hormone has been consistently shown to increase body
protein and to reduce total body fat, particularly visceral fat, making it a po-
tential agent for treatment of the overweight patient.
In a small clinical trial, 18 newly diagnosed overweight diabetic subjects
were randomly assigned to placebo or growth hormone injection, along with
dietary restriction, in a double-blind study [72]. Investigators found a greater
decrease in visceral fat, an increase in lean body mass, and improved insulin
sensitivity during this 12-week trial. In a 12-month randomized, double-
blind, placebo-controlled clinical trial, 40 postmenopausal women were as-
signed to daily placebo injections or injections of growth hormone (0.67 mg
per day). After 1 year, Franco and colleagues [73] reported that the women
had significantly reduced their abdominal and visceral adiposity, and had
improved their insulin sensitivity and total and low-density lipoprotein
cholesterol concentrations.
A group working in Australia has identified a fragment of growth hor-
mone that is lipolytic. This compound, called AOD9604, is a modified frag-
ment of the amino acids in growth hormone from 177 to 191, and is orally
active. It is said to bind to the fat cell, stimulating lipolysis and inhibiting
re-esterification without stimulating growth. A 12-week multi-center trial
randomized 300 obese subjects to one of five daily doses (1 mg, 5 mg, 10 mg,
20 mg, and 30 mg) of AOD9604 or placebo. The 1-mg dose was the most
effective for weight loss. Subjects on the 1-mg dose lost 2.6 kg, as compared
with 0.8 kg in the placebo group, and the rate of weight loss was constant
throughout the trial [74]. Phase III trials are evidently in the planning stages.
Leptin
The lack of leptin, a hormone derived from the fat cell, causes massive
overweight in animals and man. Its replacement reverses the overweight asso-
ciated with this deficiency state. The discovery of leptin generated hope that
leptin would be an effective treatment for many overweight patients. Leptin
at subcutaneous doses of 0.0 mg/kg, 0.01 mg/kg, 0.05 mg/kg, 0.1 mg/kg,
and 0.3 mg/kg daily was tested in lean [75] and obese [76] human beings of
both sexes. Lean subjects were treated for 4 weeks and lost 0.4 kg to 1.9
kg. Obese subjects were treated for 24 weeks and a dose-response relationship
for weight loss was seen, with the 0.3 mg/kg group losing 7.1 kg [77].
Pegylated leptin allows for weekly, rather than daily, injections. Although
pegylated leptin at 20 mg per week and 60 mg per week in obese subjects
over 8 to 12 weeks did not give any weight loss above placebo, pegylated leptin
at 80 mg per week combined with a very low-calorie diet for 46 days gave
2.8 kg more weight loss in 12 subjects randomized to leptin, as compared
with the 10 subjects randomized to placebo (P! 0.03) [75].
In contrast with the disappointing results in overweight patients, leptin
has been found to ameliorate many of the symptoms of lipodystrophy,
a disease with reduced or absent fat stores. Nine female patients with lipo-
dystrophy and a serum leptin level of less than 4 ng/mL were treated with
recombinant methionyl human leptin for 4 months. Eight of the women
had diabetes. During treatment with leptin, glycosylated hemoglobin de-
creased an average of 1.9% during the 4 months of therapy, and the triglyc-
eride levels decreased by 60%. Liver volume was also reduced by an average
of 28% and resting metabolic rate decreased significantly with therapy [78].
A reduction in body weight produced by eating a low-calorie diet is associ-
ated with decreased 24-hour energy expenditure, and decreased leptin and
thyroid hormone levels. When body weight was reduced by 10%, circulating
triiodothyronine, thyroxin, and leptin concentrations were decreased. All of
these endocrine changes were reversed by administration of replacement
doses of recombinant human methionyl-leptin. Total energy expenditure in-
creased in all subjects during treatment with leptin, indicating that decreased
leptin may account for some aspects of the endocrine adaptations to weight
loss [79].
Neuropeptide-Y receptor antagonists

Neuropeptide Y (NPY) is a widely distributed neuropeptide that has five
receptors, Y-1, Y-2, Y-4, Y-5, and Y-6. Neuropeptide Y stimulates food
intake, inhibits energy expenditure, and increases body weight by activating
Y-1 and Y-5 receptors in the hypothalamus [80]. Levels of NPY in the
hypothalamus are temporally related to food intake and are elevated with
energy depletion. Surprisingly, NPY-knockout mice have no phenotype.
NPY-5 receptor antagonists fall into two categories, those that reduce
food intake and those that do not, but those that do reduce food intake
seem to do so through a mechanism separate from Y-5. Thus, Y-5 receptor
antagonists do not appear promising as antioverweight agents [81]. Y-1
receptor antagonists appear to have greater potential as antioverweight
agents. A dihydropyridine neuropeptide Y-1 antagonist inhibited NPY-
induced feeding in satiated rats [82]. Another Y-1 receptor antagonist, J-104,
870, suppressed food intake when given orally to Zucker rats [83].
A study measuring NPY in obese human beings casts doubt on the
importance of the NPY antagonists in the treatment of overweight in man.
Obese women had lower NPY levels than lean women, and weight loss with
a 400 kcal per day diet and adrenergic agonists (caffeine and ephedrine or
caffeine, ephedrine, and yohimbine) did not change NPY levels at rest or
after exercise [84].
Several clinical trials with a selective Y-5 receptor antagonist have been
completed. The first was a 2-year randomized, placebo-controlled trial
that included two doses. The dose was selected based on displacement of
receptor ligand in human positron emission tomography studies [85]. There
was a significantly greater weight loss with the antagonist, indicating that
NPY is involved in regulation of human body weight, but the magnitude
1244 BRAY
of the effect was not deemed to be clinically significant. The second trial was
designed to test the effect of the antagonist on the prevention of weight gain
induced by providing subjects with a very low-calorie diet before randomi-
zation. Again, there was a significant effect, but it was not large enough to
warrant continued pursuit of this drug [85]. Other drugs are under evalua-
tion, but no data are available.
Serotonin 2C receptor agonists

Mice lacking the, 5-hydroxytryptophan (5HT) -2C receptor have increased
food intake because they take longer to be satiated. These mice also are resis-
tant to fenfluramine, a serotonin agonist that causes weight loss. A human
mutation of the 5HT-2C receptor is associated with early-onset increases in
human body weight [86–87]. The precursor of serotonin, 5-HT, reduces
food intake and body weight in clinical studies [88,89]. Fenfluramine
[90,91] and dexfenfluramine [92], two drugs that act on the serotonin system
but were withdrawn from the market in 1997 because of cardiovascular side
effects, also reduce food intake in human studies. Meta-chlorophenylpipera-
zine, a direct serotonin agonist, reduces food intake by 28% in women and
20% in men [93]. Another serotoninergic drug, sumatriptan, which acts on
the 5-HT1B/1D receptor, also reduced food intake in human subjects [94].
The robust effects of agonists toward the HT-2C receptors in suppressing
food intake have stimulated the development of several new compounds.
Only one of these has advanced to formal clinical trials. The results of
a phase II dose-ranging study for lorcaserin (APD356) have been presented.
A total of 459 male and female subjects, with a BMI between 29 kg/m2 and
46 kg/m2 and an average weight of 100 kg, were enrolled in a randomized,
double-blind controlled trial comparing placebo against 10 mg and 15 mg
given once daily, and 10 mg given twice daily (20 mg per day). During the
12 weeks of the trial the placebo group lost 0.32 kg (N ¼ 88 completers),
as compared with 1.8 kg in the 10-mg per day dose given once daily
(N ¼ 86), 2.6 kg in the 15-mg per day dose (N ¼ 82 completers), and
3.6 kg in the 10 mg twice per day (20 mg total) (N ¼ 77 completers). Side
effects that were higher in the active treatment groups than in the placebo
group were headache, nausea, dizziness, vomiting, and dry mouth. No cardiac
valvular changes were noted [95]. Additional clinical trials are underway.
PYY 3-36
PYY 3-36 is a hormone produced by the L-cells in the gastrointestinal
tract that is secreted in proportion to the caloric content of a meal. PYY
3-36 levels are lower during fasting and after a meal in overweight subjects,
as compared with lean subjects. Caloric intake at a lunch buffet was reduced
by 30% in 12 obese subjects and by 29% in 12 lean subjects 2 hours after an
intravenous infusion of PYY 3-36 [96]. Thrice daily nasal administration
over 6 days was well tolerated and reduced caloric intake by about 30%
while giving 0.6 kg weight loss [97]. Development of a nasal spray formula-
tion for PYY 3-36 has undergone Phase I clinical trials. Based on the re-
views, Merck and Company severed its commercial relationship with
Nastech on March 1, 2006. Nastech, the developer of the nasal formulation,
plans to continue developing this product.
Oxyntomodulin
Oxyntomodulin is a gastrointestinal peptide produced in the L-cells of
the intestine that is released in response to food. Animals injected with oxy-
ntomodulin have a reduction in body fat and food intake. In a short-term
clinical study, oxyntomodulin reduced food intake by 19.3%, compared
with a placebo infusion. In a 4-week randomized, double-blind, placebo-
controlled trial, overweight volunteers injected oxyntomodulin subcutane-
ously 3 times a day 30 minutes before meals. Body weight was reduced
2.3 kg plus or minus 0.4 kg in the group receiving oxyntomodulin, as com-
pared with 0.5 kg plus or minus 0.5 kg in the placebo group. Serum levels
of leptin decreased and adiponectin increased in the group receiving oxynto-
modulin. Energy intake in the treated group decreased by 170 kcal plus or
minus 37 kcal (25% plus or minus 5%) at the beginning study meal, and
by 250 kcal plus or minus 63 kcal (35% plus or minus 9%) at the final
meal [98]. Further studies on this intriguing peptide are awaited.
Cholecystokinin
Cholecystokinin decreases food intake by causing subjects to stop eating
sooner [99]. Although the relationship between cholecystokinin and satiety
has been known for many years, development as a weight-loss agent has
been slow because of concerns regarding pancreatitis. Because the human
pancreas has no cholecystokinin-A receptors, an orally active compound
that is a selective agonist of the cholecystokinin-A receptor is being evalu-
ated in clinical trials, but no reports of those trials have yet appeared.
Oleoylestrone
Oleoylestrone is a weakly estrogenic compound that is produced in fat
cells, carried in the blood on HDL particles, and feeds back to the central
nervous system to reduce food intake while maintaining energy expenditure.
Oleoylestrone is orally active and has been used to treat one morbidly obese
male without an accompanying weight-loss program. Oleoylestrone was
given in doses of 150 mmol to 300 mmol per day in ten consecutive 10-day
courses of treatment separated by at least 2 months. Weight dropped 38.5
kg and BMI dropped from 51.9 kg/m2 to 40.5 kg/m2 over 27 months, and
weight was still declining at the time of the report [100]. Oleoylestrone
was well tolerated and there were no estrogenic side effects observed. Phar-
maceutical company-sponsored phase I trials are presently in progress.
1246 BRAY
Melanin concentrating hormone receptor-1 antagonist

Melanin concentrating hormone and alpha melanocyte stimulating
hormone (a-MSH) have opposite effects on skin coloration in fish, and excess
melanin concentrating hormone blocks the reduction in food intake by
a-MSH when both are injected into the cerebral ventricles of rats [101]. Mel-
anin concentrating hormone has two receptors, MCH-1R and MCH-2R.
Mice without the MCH-1 receptor have increased activity, increased temper-
ature, and increased sympathetic tone [102]. Over-expression of the MCH-1
receptor and chronic infusion of an MCH-1 agonist cause enhanced feeding,
caloric efficiency, and weight gain, while an MCH-1 antagonist reduces food
intake and body weight gain without an effect on lean tissue [103]. MCH-1
antagonists reduce food intake by decreasing meal size, and also act as anti-
depressants and anxiolytics [104,105]. An orally active MCH-1 receptor
antagonist that has good plasma levels and central nervous system exposure
induced weight loss in obese mice with chronic treatment [106]. A number of
other MCH-1 antagonists reduce food intake and body weight in experimen-
tal animals [107]. No human studies have been reported.
Cetilistat-a pancreatic lipase inhibitor

Although orlistat, a lipase inhibitor, is already approved for the treatment
of overweight, cetilistat (ATL-962), another GI lipase inhibitor, is also in de-
velopment. A 5-day trial of cetilistat in 90 normal volunteers was conducted
on an inpatient unit. There was a three- to seven-fold increase in fecal fat that
was dose-dependent, but only 11% of subjects had more than one oily stool.
It was suggested that this lipase inhibitor may have fewer gastrointestinal
adverse events when compared with orlistat [108]. A 12-week randomized,
double-blind, clinical trial of cetilistat showed a significantly greater decrease
in body weight with cetilistat than placebo (4.1 kg versus 2.4 kg, 120 mg three
times per day). Weight loss with cetilistat was similar to orlistat, but there
were fewer GI side effects with cetilistat.
Ghrelin antagonists
Ghrelin is a small 28-amino acid peptide synthesized in the stomach. Its
active form contains an octanoate group on the third amino acid. The level
of ghrelin rises with fasting and declines after eating, suggesting it may be
a signal to begin meals. Chronic administration produces hyperphagia
and weight gain in animals. Moreover, overweight subjects have lower levels
than normal-weight individuals. Ghrelin acts at the growth hormone secre-
tagogue receptor (GHSR) to produce its effects. A group of growth hor-
mone stimulating peptides that also act on this GHSR are known to
increase food intake in human subjects [109]. Antagonists to this receptor
might thus be useful drugs for treating overweight patients; this is supported
by suppression of food intake and attenuated weight regain in diet-induced

obese mice treated with such drugs.
Combination of drugs that produce weight loss

The first important clinical trial combining drugs that acted by separate
mechanisms used phentermine and fenfluramine [110]. This trial showed
a highly significant weight loss of nearly 15% below baseline, with fewer
side effects by using combination therapy. This combination became very
popular [111], but because of reports of aortic valvular regurgitation associ-
ated with its use, fenfluramine was withdrawn from the market worldwide
on September 15, 1997 [112].
Several other combinations of existing drugs are now under development.
One of these is the combination of phentermine with topiramate, where
weight losses of over 10 kg have been reported. A second is a combination
of phentermine with zonisamide. A third is the combination of naltrexone
with bupropion, where additive weight loss has been noted. Initial data
have been published on all of these combinations, but longer-term studies
are needed to evaluate the potential drug-drug interactions and side effects
produced.
Summary
There are presently comparatively few drugs available for the treatment
of overweight patients, and their effectiveness is limited to palliation of
the chronic disease of obesity. However, drug development that is now
underway is more rapid than in the past, and investigators anticipate the
discovery of safe and effective pharmacologic strategies for the management
of obesity and its very serious complications.
References
[1] Bray G. Contemporary diagnosis and management of obesity. Newtown (PA): Handbooks
in Health Care; 2003.
[2] Padwal R, Li SK, Lau DC. Long-term pharmacotherapy for overweight and obesity: a sys-
tematic review and meta-analysis of randomized controlled trials. Int J Obes Relat Metab
Disord 2003;27(12):1437–46.
[3] Shekelle PG, Hardy ML, Morton SC, et al. Efficacy and safety of ephedra and ephedrine for
weight loss and athletic performance: a meta-analysis. JAMA 2003;289(12):1537–45.
[4] Li Z, Maglione M, Tu W, et al. Meta-analysis: pharmacologic treatment of obesity. Ann
Intern Med 2005;142(7):532–46.
[5] Bray G, Greenway F. Pharmacological treatment of the overweight patient. Pharmacol
Rev 2007;59(2):151–84.
[6] Bray GA, Greenway FL. Current and potential drugs for treatment of obesity. Endocr Rev
1999;20(6):805–75.
1248 BRAY
[7] Van Gaal LF, Rissanen AM, Scheen AJ, et al. Effects of the cannabinoid-1 receptor
blocker rimonabant on weight reduction and cardiovascular risk factors in overweight
patients: 1-year experience from the RIO-Europe study. Lancet 2005;365(9468):1389–97.
[8] Despres JP, Golay A, Sjostrom L. Effects of rimonabant on metabolic risk factors in
overweight patients with dyslipidemia. N Engl J Med 2005;353(20):2121–34.
[9] Pi-Sunyer FX, Aronne LJ, Heshmati HM, et al. Effect of rimonabant, a cannabinoid-1
receptor blocker, on weight and cardiometabolic risk factors in overweight or obese
patients: RIO-North America: a randomized controlled trial. JAMA 2006;295(7):761–75.
[10] Shekelle PG, Morton SC, Maglione M, et al. Pharmacological and surgical treatment of
obesity. Evid Rep Technol Assess (Summ) 2004;103:1–6.
[11] Avenell A, Brown TJ, McGee MA, et al. What interventions should we add to weight
reducing diets in adults with obesity? A systematic review of randomized controlled trials
of adding drug therapy, exercise, behaviour therapy or combinations of these interventions.
J Hum Nutr Diet 2004;17(4):293–316.
[12] Smith IG, Goulder MA. Randomized placebo-controlled trial of long-term treatment with
sibutramine in mild to moderate obesity. J Fam Pract 2001;50(6):505–12.
[13] Apfelbaum M, Vague P, Ziegler O, et al. Long-term maintenance of weight loss after a very-
low-calorie diet: a randomized blinded trial of the efficacy and tolerability of sibutramine.
Am J Med 1999;106(2):179–84.
[14] James WP, Astrup A, Finer N, et al. Effect of sibutramine on weight maintenance after
weight loss: a randomised trial. STORM Study Group. Sibutramine Trial of Obesity
Reduction and Maintenance. Lancet 2000;356(9248):2119–25.
[15] Wirth A, Krause J. Long-term weight loss with sibutramine: a randomized controlled trial.
JAMA 2001;286(11):1331–9.
[16] McMahon FG, Fujioka K, Singh BN, et al. Efficacy and safety of sibutramine in obese
white and African American patients with hypertension: a 1-year, double-blind, placebo-
controlled, multicenter trial. Arch Intern Med 2000;160(14):2185–91.
[17] Fujioka K, Seaton TB, Rowe E, et al. Weight loss with sibutramine improves glycaemic
control and other metabolic parameters in obese patients with type 2 diabetes mellitus.
Diabetes Obes Metab 2000;2(3):175–87.
[18] Berkowitz RI, Wadden TA, Tershakovec AM, et al. Behavior therapy and sibutramine for
the treatment of adolescent obesity: a randomized controlled trial. JAMA 2003;289(14):
1805–12.
[19] Godoy-Matos A, Carraro L, Vieira A, et al. Treatment of obese adolescents with sibutr-
amine: a randomized, double-blind, controlled study. J Clin Endocrinol Metab 2005;
90(3):1460–5.
[20] Berkowitz C, Fujioka K, Daniels S, et al. Weight loss in adolescents treated with sibutr-
amine: a randomized, placebo-controlled trial. Ann Intern Med 2006;145(2):81–90.
[21] Wadden TA, Berkowitz RI, Womble LG, et al. Randomized trial of lifestyle modification
and pharmacotherapy for obesity. N Engl J Med 2005;353(20):2111–20.
[22] Sjostrom L, Rissanen A, Andersen T, et al. Randomised placebo-controlled trial of orlistat
for weight loss and prevention of weight regain in obese patients. European Multicentre
Orlistat Study Group. Lancet 1998;352(9123):167–72.
[23] Torgerson JS, Hauptman J, Boldrin MN, et al. XENical in the prevention of diabetes in
obese subjects (XENDOS) study: a randomized study of orlistat as an adjunct to lifestyle
changes for the prevention of type 2 diabetes in obese patients. Diabetes Care 2004;27(1):
155–61.
[24] Chanoine JP, Hampl S, Jensen C, et al. Effect of orlistat on weight and body composition in
obese adolescents: a randomized controlled trial. JAMA 2005;293(23):2873–83.
[25] Hollander PA, Elbein SC, Hirsch IB, et al. Role of orlistat in the treatment of obese patients
with type 2 diabetes. A 1-year randomized double-blind study. Diabetes Care 1998;21(8):
1288–94.
[26] Heymsfield SB, Segal KR, Hauptman J, et al. Effects of weight loss with orlistat on glucose
tolerance and progression to type 2 diabetes in obese adults. Arch Intern Med 2000;160(9):
1321–6.
[27] Zhi J, Mulligan TE, Hauptman JB. Long-term systemic exposure of orlistat, a lipase
inhibitor, and its metabolites in obese patients. J Clin Pharmacol 1999;39(1):41–6.
[28] Wadden TA, Berkowitz RI, Womble LG, et al. Effects of sibutramine plus orlistat in obese
women following 1 year of treatment by sibutramine alone: a placebo-controlled trial. Obes
Res 2000;8(6):431–7.
[29] Redman LM, de Jonge L, Fang X, et al. Lack of an effect of a novel {beta}3-adrenoceptor
agonist, TAK-677 on energy metabolism in obese individuals: a double-blind, placebo-
controlled randomized study. J Clin Endocrinol Metab 2007;92(2):527–31.
[30] Lawton CL, Wales JK, Hill AJ, et al. Serotoninergic manipulation, meal-induced
satiety and eating pattern: effect of fluoxetine in obese female subjects. Obes Res 1995;
3(4):345–56.
[31] Goldstein DJ, Rampey AH Jr, Roback PJ, et al. Efficacy and safety of long-term fluoxetine
treatment of obesitydmaximizing success. Obes Res 1995;3(Suppl 4):481S–90S.
[32] Gadde KM, Parker CB, Maner LG, et al. Bupropion for weight loss: an investigation of
efficacy and tolerability in overweight and obese women. Obes Res 2001;9(9):544–51.
[33] Jain AK, Kaplan RA, Gadde KM, et al. Bupropion SR vs. placebo for weight loss in obese
patients with depressive symptoms. Obes Res 2002;10(10):1049–56.
[34] Anderson JW, Greenway FL, Fujioka K, et al. Bupropion SR enhances weight loss:
a 48-week double-blind, placebo-controlled trial. Obes Res 2002;10(7):633–41.
[35] Astrup A, Toubro S. Topiramate: a new potential pharmacological treatment for obesity.
Obes Res 2004;12(Suppl):167S–73S.
[36] Ben-Menachem E, Axelsen M, Johanson EH, et al. Predictors of weight loss in adults with
topiramate-treated epilepsy. Obes Res 2003;11(4):556–62.
[37] Bray GA, Hollander P, Klein S, et al. A 6-month randomized, placebo-controlled,
dose-ranging trial of topiramate for weight loss in obesity. Obes Res 2003;11(6):722–33.
[38] Wilding J, Van Gaal L, Rissanen A, et al. A randomized double-blind placebo-controlled
study of the long-term efficacy and safety of topiramate in the treatment of obese subjects.
Int J Obes Relat Metab Disord 2004;28(11):1399–410.
[39] Astrup A, Caterson I, Zelissen P, et al. Topiramate: long-term maintenance of weight loss
induced by a low-calorie diet in obese subjects. Obes Res 2004;12(10):1658–69.
[40] Gadde KM, Franciscy DM, Wagner HR 2nd, et al. Zonisamide for weight loss in obese
adults: a randomized controlled trial. JAMA 2003;289(14):1820–5.
[41] Devinsky O, Vuong A, Hammer A, et al. Stable weight during lamotrigine therapy: a review
of 32 studies. Neurology 2000;54(4):973–5.
[42] Merideth CH. A single-center, double-blind, placebo-controlled evaluation of lamotrigine
in the treatment of obesity in adults. J Clin Psychiatry 2006;67(2):258–62.
[43] Fontbonne A, Charles MA, Juhan-Vague I, et al. The effect of metformin on the metabolic
abnormalities associated with upper-body fat distribution. BIGPRO Study Group.
Diabetes Care 1996;19(9):920–6.
[45] Ortega-Gonzalez C, Luna S, Hernandez L, et al. Responses of serum androgen and insulin
resistance to metformin and pioglitazone in obese, insulin-resistant women with polycystic
ovary syndrome. J Clin Endocrinol Metab 2005;90(3):1360–5.
[46] Riddle MC, Drucker DJ. Emerging therapies mimicking the effects of amylin and glucagon-
like peptide 1. Diabetes Care 2006;29(2):435–49.
[47] Ratner RE, Dickey R, Fineman M, et al. Amylin replacement with pramlintide as an
adjunct to insulin therapy improves long-term glycaemic and weight control in type 1 dia-
betes mellitus: a 1-year, randomized controlled trial. Diabet Med 2004;21(11):1204–12.
1250 BRAY
[48] Maggs D, Shen L, Strobel S, et al. Effect of pramlintide on A1C and body weight in insulin-
treated African Americans and Hispanics with type 2 diabetes: a pooled post hoc analysis.
Metabolism 2003;52(12):1638–42.
[49] Patriti A, Facchiano E, Sanna A, et al. The enteroinsular axis and the recovery from type 2
diabetes after bariatric surgery. Obes Surg 2004;14(6):840–8.
[50] Small CJ, Bloom SR. Gut hormones as peripheral anti obesity targets. Curr Drug Targets
CNS Neurol Disord 2004;3(5):379–88.
[51] Greenway SE, Greenway FL 3rd, Klein S. Effects of obesity surgery on non-insulin-depen-
dent diabetes mellitus. Arch Surg 2002;137(10):1109–17.
[52] Lugari R, Dei Cas A, Ugolotti D, et al. Glucagon-like peptide 1 (GLP-1) secretion and
plasma dipeptidyl peptidase IV (DPP-IV) activity in morbidly obese patients undergoing
biliopancreatic diversion. Horm Metab Res 2004;36(2):111–5.
[53] Szayna M, Doyle ME, Betkey JA, et al. Exendin-4 decelerates food intake, weight gain, and
fat deposition in Zucker rats. Endocrinology 2000;141(6):1936–41.
[54] Gedulin BR, Nikoulina SE, Smith PA, et al. Exenatide (exendin-4) improves insulin sensi-
tivity and {beta}-cell mass in insulin-resistant obese fa/fa Zucker rats independent of glyce-
mia and body weight. Endocrinology 2005;146(4):2069–76.
[55] Rodriquez de Fonseca F, Navarro M, Alvarez E, et al. Peripheral versus central effects of
glucagon-like peptide-1 receptor agonists on satiety and body weight loss in Zucker obese
rats. Metabolism 2000;49(6):709–17.
[56] Kastin AJ, Akerstrom V. Entry of exendin-4 into brain is rapid but may be limited at high
doses. Int J Obes Relat Metab Disord 2003;27(3):313–8.
[57] Edwards CM, Stanley SA, Davis R, et al. Exendin-4 reduces fasting and postprandial glu-
cose and decreases energy intake in healthy volunteers. Am J Physiol Endocrinol Metab
2001;281(1):E155–61.
[58] Fineman MS, Shen LZ, Taylor K, et al. Effectiveness of progressive dose-escalation of
exenatide (exendin-4) in reducing dose-limiting side effects in subjects with type 2 diabetes.
Diabetes Metab Res Rev 2004;20(5):411–7.
[59] DeFronzo RA, Ratner RE, Han J, et al. Effects of exenatide (exendin-4) on glycemic con-
trol and weight over 30 weeks in metformin-treated patients with type 2 diabetes. Diabetes
Care 2005;28(5):1092–100.
[60] Kendall DM, Riddle MC, Rosenstock J, et al. Effects of exenatide (exendin-4) on glycemic
control over 30 weeks in patients with type 2 diabetes treated with metformin and a sulfo-
nylurea. Diabetes Care 2005;28(5):1083–91.
[61] Buse JB, Henry RR, Han J, et al. Effects of exenatide (exendin-4) on glycemic control over
30 weeks in sulfonylurea-treated patients with type 2 diabetes. Diabetes Care 2004;27(11):
2628–35.
[62] Heine RJ, Van Gaal LF, Johns D, et al. Exenatide versus insulin glargine in patients with sub-
optimally controlled type 2 diabetes: a randomized trial. Ann Intern Med 2005;143(8):559–69.
[63] Bray GA, Gallagher TF Jr. Manifestations of hypothalamic obesity in man: a comprehen-
sive investigation of eight patients and a review of the literature. Medicine (Baltimore) 1975;
54(4):301–30.
[64] Lustig RH, Rose SR, Burghen GA, et al. Hypothalamic obesity caused by cranial insult
in children: altered glucose and insulin dynamics and reversal by a somatostatin agonist.
J Pediatr 1999;135(2 Pt 1):162–8.
[65] Lustig RH, Hinds PS, Ringwald-Smith K, et al. Octreotide therapy of pediatric hypotha-
lamic obesity: a double-blind, placebo-controlled trial. J Clin Endocrinol Metab 2003;
88(6):2586–92.
[66] Velasquez-Mieyer PA, Cowan PA, Arheart KL, et al. Suppression of insulin secretion is
associated with weight loss and altered macronutrient intake and preference in a subset
of obese adults. Int J Obes Relat Metab Disord 2003;27(2):219–26.
[67] Lustig R, Greenway F, Velasquez D, et al. Weight loss in obese adults with insulin hyper-
secretion treated with Sandostatin LAR Depot. Obes Res 2003;11(Suppl):A25.
[68] Lustig RH, Greenway F, Velasquez-Mieyer P, et al. A multicenter, randomized, double-

blind, placebo-controlled, dose-finding trial of a long-acting formulation of octreotide in
promoting weight loss in obese adults with insulin hypersecretion. Int J Obes (Lond)
2006;30(2):331–41.
[69] Foxx-Orenstein A, Camilleri M, Stephens D, et al. Effect of a somatostatin analogue on
gastric motor and sensory functions in healthy humans. Gut 2003;52(11):1555–61.
[70] Tan TM, Vanderpump M, Khoo B, et al. Somatostatin infusion lowers plasma ghrelin
without reducing appetite in adults with Prader-Willi syndrome. J Clin Endocrinol Metab
2004;89(8):4162–5.
[71] Hoffman AR, Kuntze JE, Baptista J, et al. Growth hormone (GH) replacement therapy
in adult-onset GH deficiency: effects on body composition in men and women in a dou-
ble-blind, randomized, placebo-controlled trial. J Clin Endocrinol Metab 2004;89(5):
2048–56.
[72] Nam SY, Kim KR, Cha BS, et al. Low-dose growth hormone treatment combined with diet
restriction decreases insulin resistance by reducing visceral fat and increasing muscle mass
in obese type 2 diabetic patients. Int J Obes Relat Metab Disord 2001;25(8):1101–7.
[73] Franco C, Brandberg J, Lonn L, et al. Growth hormone treatment reduces abdominal
visceral fat in postmenopausal women with abdominal obesity: a 12-month placebo-
controlled trial. J Clin Endocrinol Metab 2005;90(3):1466–74.
[74] Metabolic.com. AOD 9604 phase 2b clinical trial successful. Available at: http://www.
metabolic.com.au/files/T5SH4035T6/ASX_%20AOD9604_result%20announcement.pdf.
Accessed August 10, 2007.
[75] Hukshorn CJ, Westerterp-Plantenga MS, Saris WH. Pegylated human recombinant leptin
(PEG-OB) causes additional weight loss in severely energy-restricted, overweight men. Am
J Clin Nutr 2003;77(4):771–6.
[76] Hukshorn CJ, Saris WH, Westerterp-Plantenga MS, et al. Weekly subcutaneous pegylated
recombinant native human leptin (PEG-OB) administration in obese men. J Clin Endocri-
nol Metab 2000;85(11):4003–9.
[77] Heymsfield SB, Greenberg AS, Fujioka K, et al. Recombinant leptin for weight loss in
obese and lean adults: a randomized, controlled, dose-escalation trial. JAMA 1999;
282(16):1568–75.
[78] Oral EA, Simha V, Ruiz E, et al. Leptin-replacement therapy for lipodystrophy. N Engl
J Med 2002;346(8):570–8.
[79] Rosenbaum M, Murphy EM, Heymsfield SB, et al. Low dose leptin administration reverses
effects of sustained weight-reduction on energy expenditure and circulating concentrations
of thyroid hormones. J Clin Endocrinol Metab 2002;87(5):2391–4.
[80] Parker E, Van Heek M, Stamford A. Neuropeptide Y receptors as targets for anti-obesity
drug development: perspective and current status. Eur J Pharmacol 2002;440(2–3):173–87.
[81] Levens NR, Della-Zuana O. Neuropeptide Y Y5 receptor antagonists as anti-obesity drugs.
Curr Opin Investig Drugs 2003;4(10):1198–204.
[82] Poindexter GS, Bruce MA, LeBoulluec KL, et al. Dihydropyridine neuropeptide Y Y(1)
receptor antagonists. Bioorg Med Chem Lett 2002;12(3):379–82.
[83] Kanatani A, Hata M, Mashiko S, et al. A typical Y1 receptor regulates feeding behaviors:
effects of a potent and selective Y1 antagonist, J-115814. Mol Pharmacol 2001;59(3):501–5.
[84] Zahorska-Markiewicz B, Obuchowicz E, Waluga M, et al. Neuropeptide Y in obese women
during treatment with adrenergic modulation drugs. Med Sci Monit 2001;7(3):403–8.
[85] Erondu N, Gantz I, Musser B, et al. Neuropeptide Y5 receptor antagonism does not induce
clinically meaningful weight loss in overweight and obese adults. Cell Metab 2006;4(4):
275–82.
[86] Gibson WT, Ebersole BJ, Bhattacharyya S, et al. Mutational analysis of the serotonin re-
ceptor 5HT2c in severe early-onset human obesity. Can J Physiol Pharmacol 2004;82(6):
426–9.
[87] Nilsson B, et al. 5-HT2C receptor agonists. J Med Chem 2006;49(14):4023–34.
1252 BRAY
[88] Cangiano C, Ceci F, Cascino A, et al. Eating behavior and adherence to dietary prescrip-
tions in obese adult subjects treated with 5-hydroxytryptophan. Am J Clin Nutr 1992;
56(5):863–7.
[89] Cangiano C, Laviano A, Del Ben M, et al. Effects of oral 5-hydroxy-tryptophan on energy
intake and macronutrient selection in non-insulin dependent diabetic patients. Int J Obes
Relat Metab Disord 1998;22(7):648–54.
[90] Rogers PJ, Blundell JE. Effect of anorexic drugs on food intake and the micro-structure of
eating in human subjects. Psychopharmacology (Berl) 1979;66(2):159–65.
[91] Foltin RW, Haney M, Comer SD, et al. Effect of fenfluramine on food intake, mood, and
performance of humans living in a residential laboratory. Physiol Behav 1996;59(2):
295–305.
[92] Drent ML, Zelissen PM, Koppeschaar HP, et al. The effect of dexfenfluramine on eating
habits in a Dutch ambulatory android overweight population with an overconsumption
of snacks. Int J Obes Relat Metab Disord 1995;19(5):299–304.
[93] Walsh AE, Smith KA, Oldman AD, et al. m-chlorophenylpiperazine decreases food intake
in a test meal. Psychopharmacology (Berl) 1994;116(1):120–2.
[94] Boeles S, Williams C, Campling GM, et al. Sumatriptan decreases food intake and increases
plasma growth hormone in healthy women. Psychopharmacology (Berl) 1997;129(2):
179–82.
[95] Smith SR, Prosser W, Donahue D, et al. APD356, an orally-active selective 5HT2c agonist
reduces body weight in obese men and women. Diabetes and Metabolism 2006;55(Suppl1):
A80.
[96] Batterham RL, Cohen MA, Ellis SM, et al. Inhibition of food intake in obese subjects by
peptide YY3-36. N Engl J Med 2003;349(10):941–8.
[97] Brandt G, Quay S. Intranasal peptide YY 3-36: phase 1 dose ranging and dose sequencing
studies. Obes Res 2004;12(Suppl):A28.
[98] Wynne K, Park AJ, Small CJ, et al. Subcutaneous oxyntomodulin reduces body weight in
overweight and obese subjects: a double-blind, randomized, controlled trial. Diabetes 2005;
54(8):2390–5.
[99] Pi-Sunyer X, Kissileff HR, Thornton J, et al. C-terminal octapeptide of cholecystokinin
decreases food intake in obese men. Physiol Behav 1982;29(4):627–30.
[100] Alemany M, Fernandez-Lopez JA, Petrobelli A, et al. [Weight loss in a patient with
morbid obesity under treatment with oleoyl-estrone]. Med Clin (Barc) 2003;121(13):
496–9 [in Spanish].
[101] Ludwig DS, Mountjoy KG, Tatro JB, et al. Melanin-concentrating hormone: a functional
melanocortin antagonist in the hypothalamus. Am J Physiol 1998;274(4 Pt 1):E627–33.
[102] Astrand A, Bohlooly YM, Larsdotter S, et al. Mice lacking melanin-concentrating hor-
mone receptor 1 demonstrate increased heart rate associated with altered autonomic activ-
ity. Am J Physiol Regul Integr Comp Physiol 2004;287(4):R749–58.
[103] Shearman LP, Camacho RE, Sloan Stribling D, et al. Chronic MCH-1 receptor modulation
alters appetite, body weight and adiposity in rats. Eur J Pharmacol 2003;475(1–3):37–47.
[104] Kowalski TJ, Farley C, Cohen-Williams ME, et al. Melanin-concentrating hormone-1
receptor antagonism decreases feeding by reducing meal size. Eur J Pharmacol 2004;
497(1):41–7.
[105] Borowsky B, Durkin MM, Ogozalek K, et al. Antidepressant, anxiolytic and anorectic
effects of a melanin-concentrating hormone-1 receptor antagonist. Nat Med 2002;8(8):
825–30.
[106] Souers AJ, Gao J, Brune M, et al. Identification of 2-(4-benzyloxyphenyl)-N- [1-(2-pyrroli-
din-1-yl-ethyl)-1H-indazol-6-yl]acetamide, an orally efficacious melanin-concentrating hor-
mone receptor 1 antagonist for the treatment of obesity. J Med Chem 2005;48(5):1318–21.
[107] Handlon AL, Zhou H. Melanin-concentrating hormone-1 receptor antagonists for the
treatment of obesity. J Med Chem 2006;49(14):4017–22.
[108] Dunk C, Enunwa M, De La Monte S, et al. Increased fecal fat excretion in normal volun-
teers treated with lipase inhibitor ATL-962. Int J Obes Relat Metab Disord 2002;26(Suppl):
S135.
[109] Laferrere B, Abraham C, Russell CD, et al. Growth hormone releasing peptide-2
(GHRP-2), like ghrelin, increases food intake in healthy men. J Clin Endocrinol Metab
2005;90(2):611–4.
[110] Weintraub M. Long-term weight control: the National Heart, Lung, and Blood Institute
funded multimodal intervention study. Clin Pharmacol Ther 1992;51(5):581–5.
[111] Stafford RS, Radley DC. National trends in antiobesity medication use. Arch Intern Med
2003;163(9):1046–50.
[112] Connolly HM, Crary JL, McGoon MD, et al. Valvular heart disease associated with
fenfluramine-phentermine. N Engl J Med 1997;337(9):581–8.
Med Clin N Am 91 (2007) 1255–1271
Bariatric Surgery for Morbid ObesitydA

Cure for Metabolic Syndrome?
Subhash Kini, MD, FRCS, FRCS (Ed), FRCS (Glas)a,*,
Daniel M. Herron, MD, FACSa,
Robert T. Yanagisawa, MDb
a
Department of Surgery, Mount Sinai School of Medicine, 5 East 98th Street,
15th floor, New York, NY 10029, USA
b
Division of Endocrinology, Diabetes, and Bone Diseases, Mount Sinai School of Medicine,
Box 1055, 1 Gustave L. Levy Place, New York, NY 10029, USA
Bariatric operations produce durable weight loss of up to 75% or more of

excess body weight. Bariatric surgery is an extremely effective tool for
improving the metabolic syndrome. Weight loss operations correct the met-
abolic syndrome because of a reduction of visceral adipose tissue mass. Gas-
tric bypass and biliopancreatic diversion-duodenal switch (BPD-DS) also
stimulate incretins through earlier presentation of food to the ileum.
Weight-loss surgery significantly decreases overall mortality. A large two
cohort observational study from Canada [1] shows that there is a reduction
in the relative risk of death by 89% in morbidly obese patients who undergo
bariatric surgery, when compared with those who do not have surgery.
With the majority of weight loss occurring within 6 to 12 months of sur-
gery, and acceptably low rates of weight regain, bariatric surgery exceeds the
results of any other treatment modality available today. When used appro-
priately, it improves many aspects of the metabolic syndrome, including
type 2 diabetes, hypertension, and dyslipidemia. This article reviews indica-
tions for bariatric surgery, the different operations, and patient selection for
each type. The availability of various surgical options means that an individ-
ualized approach can be used for each patient. Careful preoperative prepa-
ration and postoperative management are keys to successful outcome.
E-mail address: subhash.kini@mountsinai.org (S. Kini).
1256 KINI et al
Indications for bariatric surgery on metabolic syndrome

Body mass index (BMI) is defined as weight (in kilograms) divided by
height (in meters) squared. While BMI does have some limitations as a mea-
sure of total body fat, it is the most simple and practical way to evaluate the
degree of excess weight in a routine clinical setting. According to current
National Institutes of Health (NIH) guidelines, bariatric surgery is indicated
for patients with a BMI greater or equal to 40 kg/m2 or those with BMI
greater or equal to 35 kg/m2, with one or more obesity-associated comorbid-
ities such as diabetes, hypertension, coronary artery disease, or sleep apnea
[2].
Morbidly obese patients with metabolic syndrome should strongly con-
sider bariatric surgery before they develop irreversible sequelae of their dis-
ease. They comprise a subset of obese individuals who have shown much
higher risk for obesity-associated complications and should therefore be
treated more aggressively [3,4]. Unlike obese but metabolically healthy indi-
viduals, those with metabolic syndrome should begin medical management
early, with an aggressive combination of medical and behavioral interven-
tion including both diet and exercise. If the best efforts at medical manage-
ment fail, surgical intervention should be strongly considered; this is
preferable to waiting until irreversible complications occur. Current NIH
guidelines for bariatric surgery do not specifically address metabolic syn-
drome as an indication.
Patients must realize that bariatric surgery is strictly a tool for weight loss
and, like any other tool, can be misused or used ineffectively. Even with the
best surgical techniques, surgical failure can occur. Inadequate weight loss
has been defined differently by different authors, but typically refers to
weight loss of less than 30% to 50% of excess body weight or significant
weight regain. In order for bariatric surgery to be most successful, patients
have to learn how to use it effectively. Well-informed patients with a care-
fully planned bariatric operation to target their metabolic syndrome should
do very well.
Surgical options
As of 2007 there are a number of different surgical options available for
the treatment of severe obesity. These operations result in weight loss
through a combination of gastric volume restriction, malabsorption, and
hormonal changes. Gastric bypass, performed either open or laparoscopi-
cally, is currently the most popular operation in the United States [5]. Lap-
aroscopic adjustable gastric banding is the next most popular option and is
rapidly gaining in popularity. Biliopancreatic diversion with duodenal
switch, the most technically complex and metabolically invasive weight
loss procedure, represents a fairly small percentage of weight loss
BARIATRIC SURGERY FOR MORBID OBESITY 1257
operations. Although vertical banded gastroplasty (VBG) used to be one of

the most common bariatric operations, its popularity has waned in recent
years, owing to an increasing database suggesting poor long-term outcomes
[6]. Laparoscopic sleeve gastrectomy, also referred to as vertical gastrec-
tomy, is a newer operation that is gaining in popularity [7]. A number of
investigational procedures, such as gastric electrical stimulation and endo-
scopic restrictive procedures, are currently being evaluated and may enter
the clinical mainstream in the near future [8].
Gastric bypass
The gastric bypass, first performed in the United States in 1967, is cur-
rently the most frequently performed weight-loss operation in the United
States [9]. The operation, which was initially performed through an open ap-
proach exclusively, is now commonly performed laparoscopically, resulting
in shorter hospital stays and fewer wound complications. Many bariatric
surgeons consider the operation the ‘‘gold standard’’ against which all other
weight-loss operations are measured [10].
Since its inception, the gastric bypass has undergone substantial evolu-
tion. Most surgeons use a linear surgical stapler to create a small gastric
pouch based on the lesser curvature of the stomach. The volume of this
pouch is typically 15 mL to 20 mL. The pouch is anastomosed to a segment
of small intestine, referred to as the Roux limb (Fig. 1). Food passes from
the gastric pouch into the Roux limb, which is typically 50 cm to 150 cm
in length.
Although the mechanism of action of the gastric bypass is still not com-
pletely understood, most surgeons feel that the operation is primarily a re-
strictive one. Because the gastric pouch is so small, patients experience
a feeling of satiety after eating a fairly small meal. The gastric bypass also
provides a small degree of malabsorption because of the separation of the
food path (alimentary limb) from the biliopancreatic secretions (biliopancre-
atic limb). The amount of malabsorption the operation causes is directly
linked to the length of the alimentary and biliopancreatic limbs.
Gastric bypass also discourages patients from eating sweets, because of
the dumping syndrome, which is caused by the rapid passage of simple
sugars and high osmolarity foods from the pouch into the Roux limb.
Patients with dumping syndrome typically feel cold and clammy or warm
and flushed and frequently feel the need to lie down until the sensation
passes. In general, dumping syndrome (sometimes referred to as the ‘‘post-
operative police officer’’) is perceived by patients as unpleasant and results
in avoidance of sweets.
Recently, it has become increasingly appreciated that gastric bypass may
cause weight loss through hormonal mechanisms. Although controversial,
Cummings and colleagues’ [11] study demonstrated that serum ghrelin levels
are remarkably decreased after gastric bypass. Additionally, diurnal
1258 KINI et al
Fig. 1. Roux-en-Y gastric bypass. (Courtesy of D.M. Herron, MD, New York, NY. Ó 2007
Daniel M. Herron.)
variation in ghrelin levels appeared to be eliminated after surgery. While

some subsequent investigations failed to support these findings, most bariat-
ric specialists feel that the endocrine mechanism of gastric bypass-induced
weight loss is very important and warrants further study [12].
The amount of weight loss after gastric bypass is variable, and depends
quite substantially on patient behavior after surgery. Most bariatric sur-
geons expend a considerable amount of effort educating patients on the im-
portance of exercise and dietary moderation in conjunction with gastric
bypass. The weight-loss typically occurs during the first 6 to 12 months after
operation, after which weight tends to stabilize. While appetite is severely
diminished during the first few months after gastric bypass, it tends to return
after the first year. Some patients, especially if noncompliant with postoper-
ative dietary and exercise recommendations, will subsequently regain some
of their lost weight.
Overall, patients typically lose one half to three quarters of their initial
excess body weight. In a long-term follow-up study performed by MacLean
and colleagues [13], weight-loss success was defined as a reduction in BMI to
less than 35 kg/m2. With this definition, patients with an initial BMI less
than 50 kg/m2 achieved a 93% success rate of postoperative weight loss.
Super-obese patients, with a body mass index above 50 kg/m2, frequently
lose more weight in absolute terms. However, only 57% of super-obese
patients in MacLean’s study ultimately achieved a BMI below 35 kg/m2.
Gastric bypass was performed laparoscopically for the first time in 1994
and reported in 1996 [14]. A perspective, randomized trial by Nguyen and
colleagues [15], comparing laparoscopic and open technique, demonstrated
many benefits of a laparoscopic approach, including shorter postoperative
recovery, decreased impairment of pulmonary function, less postoperative
pain and fewer wound related complications. Additionally, the laparoscopic
approach is perceived by many patients to be vastly preferable to open sur-
gery. The largest laparoscopic series show results that are as good or better
than open series [10]. Typically, patients are able to return to normal activ-
ities within 2 to 3 weeks after their operation.
The most common complication after laparoscopic gastric bypass surgery
is stenosis of the gastrojejunal anastomosis. This occurs in 2% to 15% of
patients and is typically treated with endoscopic balloon dilatation [16].
Marginal ulcers may occur just distal to the anastomosis in 1% or less of
patients because of exposure of the small intestinal mucosa to the gastric
acid; this is typically managed medically with proton pump inhibitors [17].
The incidence of postoperative bowel obstruction varies. When the mesen-
teric defects are closed the incidence is approximately 1% [18]. Despite
the fact that laparoscopic gastric bypass is a major operation performed
on severely obese patients with substantial cardiac, pulmonary, and endo-
crine comorbidities, perioperative mortality remains low, ranging from
0.0% to 0.4% [19,20].
Laparoscopic adjustable gastric band

In June of 2001, the Food and Drug Administration approved the use of
a laparoscopically placed adjustable gastric band. This device consists of
a silicone rubber balloon that is wrapped around the upper portion of the
stomach, immediately distal to the gastroesophageal junction. The device
is connected, through a piece of flexible silicone tubing, to a subcutaneous
access port typically mounted to the abdominal wall muscle (Fig. 2).
Unlike the gastric bypass, the adjustable gastric band allows for an ad-
justable amount of gastric restriction. By injecting saline into the subcuta-
neous port, the balloon expands and narrows the gastric lumen within
the band. Ideally, the band is adjusted to provide approximately 1 to 2
pounds of weight loss per week. It is important to avoid excessive tightening
of the band, which may result in frequent vomiting and esophageal
dilatation.
In a recent series of 1014 consecutive cases, Ponce and colleagues [21] re-
ported postoperative excess weight loss of 40%, 53%, and 62% at 1, 2, and
3 years respectively. At 36 and 48 months, three quarters of patients had
achieved greater than 50% excess weight loss. Complications in this series
were relatively low. Less than 1% of bands required surgical explantation
and there was no reported mortality. Results in the super morbidly obese
population may not be as good; in a retrospective 2006 study comparing
1260 KINI et al
Fig. 2. Laparoscopic adjustable gastric band and access port. (Courtesy of D.M. Herron, MD,
New York, NY. Ó 2007 Daniel M. Herron.)
laparoscopic adjustable gastric banding with gastric bypass, adjustable gas-

tric band patients experienced a greater incidence of late complications, re-
operations, lower weight loss and decreased overall satisfaction [22].
Biliopancreatic diversion with duodenal switch

The biliopancreatic diversion with duodenal switch, commonly referred
to as the BPD-DS or merely the duodenal switch, is both the most techni-
cally complex and least commonly performed bariatric operation in the
United States [23,24]. The operation is a modification of a highly malab-
sorptive operation first reported by Scopinaro and colleagues [25] in Italy
in the 1970s. The operation includes a ‘‘sleeve gastrectomy’’ where the
greater curvature of the stomach is resected and the lesser curvature and py-
lorus left intact. The duodenum is surgically divided in its first portion. The
small section of duodenum just distal to the pylorus is connected to the last
250 cm of small bowel. The biliopancreatic limb, containing all the bile and
pancreatic secretions, merges with the alimentary path approximately 100
cm from the terminal ileum. It is only in this final common channel that nor-
mal protein and fat absorption occurs (Fig. 3).
Weight loss after the duodenal switch procedure is excellent. In Marceau
and colleagues’ [23] series of 457 patients, subjects lost 73% of their initial
excess body weight. However, as would be expected from a highly malab-
sorptive operation, patients suffer from the side effects of fat and protein
malabsorption. Most patients experience between three to five soft bowel
movements per day and pass a large amount of noxious flatus. Such side ef-
fects are noted to be a ‘‘major problem’’ by 30% to 40% the patients [23].
Additionally, abdominal cramping and bloating are frequent side effects,
occurring more than once a week in one third of duodenal switch patients.
Fig. 3. Biliopancreatic diversion with duodenal switch. (Courtesy of D.M. Herron, MD,
New York, NY. Ó 2007 Daniel M. Herron.)
As might be expected for such a technically complex operation, BPD-DS

suffers from a higher incidence of major morbidity (15%) and mortality
(0.5%–2.5%) than other bariatric operations [24,26]. These issues are re-
sponsible for the failure of this operation to gain the widespread acceptance
that gastric bypass and gastric banding enjoy.
Laparoscopic sleeve gastrectomy

Recently, there has been increasing interest in the laparoscopic sleeve gas-
trectomy, also referred to as a vertical gastrectomy [27]. This operation rep-
resents the restrictive portion of the BPD-DS without the intestinal bypass
(Fig. 4). In a series of 216 patients undergoing vertical gastrectomy, Lee and
colleagues [7] reported loss of 59% initial excess body weight at one year,
compared with 47% for gastric band patients and 75% for gastric bypass
patients. Some investigators are optimistic that this procedure may provide
weight loss as good or better than that obtained with the adjustable gastric
band, without the need to implant a medical device and frequent adjusting.
At present, we still await long-term outcome data.
Preoperative workup
The initial patient encounter is a long and involved one. The encounter
starts with a thorough history and a physical examination, followed by a
detailed discussion of the various weight loss operations, including alterna-
tives, risks and benefits. The ‘‘tool concept’’ is explained to the patient: the
1262 KINI et al
Fig. 4. Sleeve gastrectomy or vertical gastrectomy. (Courtesy of D.M. Herron, MD, New York,
NY. Ó 2007 Daniel M. Herron.)
fact that the operation is a ‘‘tool,’’ one of the three components of this
weight-loss process. Exercise and compliance with a new dietary regimen af-
ter weight-loss surgery are the other two equally important components of
this weight-loss process.
It is stressed that not all excess weight will be lost: 50% to 75% excess
body weight loss can be expected with the gastric bypass, sleeve gastrec-
tomy, and BPD-DS procedures, and 40% to 50% excess body weight loss
can be expected with the adjustable gastric band procedure. The patient
needs to understand the necessity of lifelong follow-up and vitamins and
mineral supplementation after weight-loss surgery. Failure to comply with
the postoperative regimen may result in long-term severe complications up
to and including death. The preoperative work-up typically includes baseline
blood work, psychologic assessment and clearance, nutritional assessment,
education and clearance, and an upper gastrointestinal endoscopy.
The baseline blood work includes tests to screen for nutritional defi-
ciencies (serum calcium, phosphorus, alkaline phosphatase, 25-hydroxy
vitamin D, parathormone, iron, iron binding capacity, ferritin, folic acid, vi-
tamin B12, total protein, albumin), baseline blood counts, serum chemistry,
lipid profile, glycosylated hemoglobin, and liver function.
The psychologic assessment determines if the patient is seeking to have
this surgery for health reasons, rather than strictly cosmetic ones, and if
the patient’s expectations are realistic. A history of a severe eating disorder,
substance abuse, or mental instability, such as suicide attempts, are relative
contraindications. If the patient currently has one of these problems they
must get therapy and clearance from the psychiatrist or psychologist before
proceeding to surgery. A patient’s noncompliance with psychotropic medi-
cation prescriptions and therapy appointments may predict noncompliance
with the postoperative regimen. The patient is encouraged to develop a per-
sonal support system (spouse, significant other, family, or friend) and join
a support group.
All patients have a formal consultation with a bariatric nutritionist who
screens for eating disorders, and educates about the change in diet after sur-
gery. Additionally, the nutritionist can assess whether patients are moti-
vated enough to comply with the required dietary modification.
All patients undergo an upper endoscopy to rule out other upper gas-
trointestinal pathology, to check for any significant hiatal hernia, and to
check for evidence of reflux esophagitis. Helicobacter pylori, if present,
is treated.
Further medical workup depends on the patient’s medical condition. A
cardiology stress test and/or echocardiogram is ordered as necessary. A pos-
itive stress test is followed up with an angiogram. A polysomnography is or-
dered based on the patient’s symptoms, such as excessive daytime
drowsiness. Obstructive sleep apnea, if present, is treated with continuous
positive airway pressure (CPAP) or bilevel positive airway pressure for at
least 3 weeks before surgery.
If a patient has biliary colic symptoms, an abdominal ultrasound is per-
formed to confirm the presence of gallstones.
At the second preoperative visit, the results from the preoperative consul-
tations are reviewed. All medical conditions, such as hypertension and dia-
betes, must be well controlled at the time of the surgery. The patient is given
a quiz to assess comprehension and retention of information given at the
first preoperative visit. This quiz is graded and reviewed with the patient.
Gaps in their knowledge are reinforced with further education. The poten-
tial complications of surgery are reviewed in depth, the expected weight loss
is discussed, and compliance expectations are reinforced. All of the patient’s
questions are answered to their satisfaction. After confirmation of sufficient
comprehension, consent forms are reviewed in depth. Preoperative and post-
operative instructions for surgery are reviewed with the patient with partic-
ular emphasis on importance of protein intake, hydration and follow-up
with their primary physician after surgery.
Perioperative management
When patients come in for their routine preoperative tests, they are seen
by an anesthesiologist for a careful evaluation of their upper airway.
Patients who have a short, thick neck, crowded airway, receding chin, or
are super-super obese (BMI more than 60) are likely to have a difficult
airway.
1264 KINI et al
For all patients with a potentially difficult airway, an awake, fiberoptic

intubation is performed with two attending anesthesiologists present (as re-
quired with extubation), at least one of whom is an experienced anesthesiol-
ogist. An emergency airway cart is kept in the room. A nonrebreathing face
mask is used to administer high concentrations of oxygen following extuba-
tion, and the patient must be observed for at least 5 minutes postextubation
before transport out of the operating room.
For all patients with obstructive sleep apnea (OSA), which may be di-
agnosed based on sleep laboratory studies or a strongly suggestive his-
tory, or those with difficult airways, CPAP machines will be used in
the recovery room and after transfer to the floor, at the anesthesiologists’
discretion.
Patient controlled analgesia (PCA), without continuous basal opioid in-
fusion, is a commonly used modality for postoperative pain relief. OSA pa-
tients with PCA are transferred from the recovery room to a step-down
unit where continuous pulse oximetry monitoring is used for the first 24
hours following the operation. A patient is also sent to a step-down bed
if he or she has recently diagnosed sleep apnea or severe sleep apnea, is
more than 65-years-old, has had prolonged surgery, or has severe
comorbidities.
Morbidly obese patients have a higher incidence of thromboembolism.
Deep venous thrombosis prophylaxis consists of early ambulation, calf
pump exercises, subcutaneous heparin, and the routine use of sequential
compression devices.
After gastric bypass the patient is allowed only ice chips. On the first
postoperative day the patient is advanced to a Stage I diet, consisting of
clear, nonsweetened, noncarbonated liquids. On the second postoperative
day the patient is advanced to a Stage II diet, consisting of pureed low cal-
orie foods. The patient is encouraged to use an incentive spirometer and to
expectorate phlegm if present. Oral hypoglycemic agents are not restarted.
Antihypertensive and other medications are started very carefully.
There might be a low index of suspicion of a leak. For example, if the pa-
tient does not ‘‘look right,’’ has a heart rate more than 110 beats per minute,
a tempatrature greater than 101 degrees Fahrenheit, oxygen saturation less
than 90, excessive abdominal pain, or a low urine output, then a water sol-
uble upper gastrointestinal series is obtained to rule out a leak. If there is
a higher index of suspicion of a leak the patient is taken directly for a diag-
nostic laparoscopy. No time is wasted for tests as morbidly obese patients
have a low reserve and tend to deteriorate very rapidly. It must be noted
that tachycardia is sometimes the only sign of a leak.
Dietary guidelines are reinforced by the nutritionist during the inpatient
stay. A patient who has undergone a gastric bypass is usually discharged on
the second postoperative day when he or she is well, afebrile, hemodynam-
ically stable, without abdominal pain, tolerating Stage I or II diet, and
ambulating. The patient is encouraged to continue to use the incentive
spirometer at home for a week and to ambulate frequently for longer

periods of time.
Patients who have other operations also follow a similar postoperative
regimen with some small differences.
Postoperative follow-up
The first postoperative follow-up visit starts at 3 weeks, with reassurance
and encouragement regarding weight loss milestones, such as BMI, percent
of excess body weight loss, comorbidity improvement or resolution, and as-
sessment of side effects or complications. The patient also follows up with
the nutritionist, who assesses calorie, protein, fluid, and supplement intake,
and reinforces the need for compliance with nutritional supplements, pro-
tein intake, and fluid intake. The patient is reminded to start with a
postoperative exercise program and a follow-up schedule. For women, the
importance of birth control is reinforced.
Patients who have had operations other than an adjustable gastric band,
such as gastric bypass, sleeve gastrectomy, or BPD-DS are followed up at
3 weeks, 3 months, 6 months, and then yearly. At each subsequent postoper-
ative visit blood is drawn to assess the status of the above nutrients and pa-
tients get appropriately supplemented with iron, calcium, and multivitamins.
Weight loss occurs with an adjustable gastric band only with regular,
timely adjustments. The band is left empty when first placed. The first
adjustment is performed in the office 6 weeks postoperatively. This allows
for the formation of a capsule around the adjustable gastric band and
thus secures it to the stomach. There is no preset follow-up schedule for gas-
tric band adjustment, which is based on increasing hunger, decreasing
restriction, and slowing of weight loss to less than a pound a week. Fol-
low-up is otherwise similar to that of other operations.
All patients are followed indefinitely on a yearly basis.
Outcome of bariatric surgery on metabolic syndrome

The operative mortality of bariatric surgery has decreased considerably
over the years, making this a safe operation in the hands of experienced
bariatric surgeons. In ‘‘centers of excellence,’’ the operative mortality is
0.35% [28].
Bariatric surgery appears to be the most effective option for the treatment
of severe obesity, resulting in a long-term weight loss, an improved lifestyle,
and an improvement in risk factors associated with obesity [29]. While medical
intervention of obesity has a high likelihood of weight regain, surgical inter-
vention usually results in a long-term, sustained weight loss. The gastric re-
striction created by surgery results in early satiety, which therefore leads to
a reduced volume of intake. Hormonal mechanisms, such as reduction in
1266 KINI et al
ghrelin, an orixogenic hormone secreted by the stomach after Roux-en-Y gas-

tric bypass, may be responsible for the reduction in appetite [10]. Smaller
quantities at each meal decrease the glucose load that the body has to handle.
The prospective controlled Swedish obese subjects study involved obese
subjects who underwent gastric surgery, and contemporaneously matched
conventionally treated obese control subjects. As one of the largest series,
with 4047 subjects, it demonstrates the benefit of bariatric surgery for mor-
bid obesity [30]. The study shows that a significant number of patients with
diabetes can recover with surgical weight loss, based on either the cutoff
values or use of medication to treat diabetes. The mean changes in weight
and risk factors were also more favorable among the subjects treated by gas-
tric bypass than among those treated by gastric banding or other surgical
procedures (Fig. 5).
The duration of diabetes from time of diagnosis seems to predict suc-
cessful improvement of glycemic control postsurgery. Patients with a rela-
tively short duration of diabetes since diagnosis seem to achieve better
glycemic control after bariatric surgery [29]. A shorter duration of diabetes
may better preserve b-cell mass and may contribute to better remission of
diabetes.
The first phase insulin response is typically disrupted early in the course
of diabetes and improves to a near normal level in patients who undergo
gastric bypass, with diabetes of less than 3 to 5 years duration [31]. Recovery
of the first phase insulin response may be the best indicator of diabetes res-
olution in patients who have had gastric bypass (Fig. 6).
The mechanisms underlying the effects of gastric bypass on body weight
and glucose metabolism are still not completely understood, but we can pre-
dict the course of diabetes outcome with surgery. In the immediate
postoperative period, patients are essentially fasting and reacting to the
fast-induced resolution of glucose toxicity and alleviation of insulin resis-
tance. Patients who are on significant doses of insulin preoperatively often
only require minimal basal insulin. Patients gradually tolerate their oral in-
take, but they continue to be in a state of negative energy balance, a condi-
tion that decreases glucose toxicity and improves b-cell function. Eventually,
marked weight reduction with bariatric surgery allows patients to increase
their level of physical activity. Increased physical activity, coupled with
decreased glucose load from smaller quantities of each meal after gastric by-
pass, dramatically improved diabetes control.
Although not conclusively proven, the improvement of insulin secretion
or action may be due to alterations in gut hormones, especially glucagon-
like peptide-1 (GLP-1) release after gastric bypass [32,33], as a result of by-
passing part of the foregut and facilitating delivery of nutrients directly to
the hindgut where it is secreted. GLP-1 is an incretin that stimulates insulin
secretion in response to enteric nutrients [34]. Le Roux and colleagues [35]
found that, compared with lean and obese controls, patients following gas-
tric bypass had increased postprandial plasma peptide YY (PYY) and GLP-1,
Fig. 5. Long-term outcome of bariatric surgery in Swedish obese subject study. Recovery from
diabetes, lipid disturbances, hypertension, and hyperuricemia over 2 and 10 years in surgically
treated subjects and their obese controls. (From Sjostrom L, Lindroos A, Peltonen M, et al.
Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J
Med 2004;351(26):2683; with permission. Copyright Ó 2004, Massachusetts Medical Society.)
1268 KINI et al
Fig. 6. Recovery of the first phase insulin response after gastric bypass. Insulin response after
intravenous glucose infusion in normal control subjects (E) and patients with type 2 diabetes,
before gastric bypass (A), at 3 months (B), 6 months (C), and 12 months (D) after gastric
bypass. (From Polyzogopoulou EV, Kalfarentzos F, Vagenakis AG, et al. Restoration of
euglycemia and normal acute insulin response to glucose in obese subjects with type 2 diabetes
following bariatric surgery. Diabetes 2003;52:101. Reprinted with permission from The
American Diabetes Association. Copyright Ó 2003 American Diabetes Association.)
favoring enhanced satiety. They also noted gastric bypass patients had early
and exaggerated insulin responses, improving glycemic control. None of
these effects were observed in patients losing equivalent weight through gas-
tric banding, which is a purely restrictive operation. Peptide YY is also a gut
hormone that increases satiety, is secreted by L cells in the hindgut, and is
released into the circulation after a meal. GLP-1 and PYY suppress gastro-
intestinal motility, gastric emptying, small intestinal transit, and food
intake.
In the process of selecting the operation, those with diabetes should, per-
haps, consider gastric bypass to maximize incretin effect on insulin secretion
[35]. Purely restrictive bariatric operations (adjustable gastric band and
sleeve gastrectomy) reduce visceral adipose tissue mass. However, since
these operations lack the incretin effect, they have a reduced effect on insulin
resistance (Fig. 7).
Bariatric surgery improves all the components of the metabolic syn-
drome. In a metanalysis of bariatric operations, Buchwald and colleagues
[36] concluded that diabetes improved or resolved in 83%, hypercholesterol-
emia improved in 96%, and hypertension resolved or improved in 87% of
patients who underwent a gastric bypass.
Fig. 7. Increased postprandial plasma insulin, PYY and GLP-1, following gastric bypass: (A)
The insulin response, (C) GLP-1 responses, and (D) PYY response to test a meal in Roux-
En-Y gastric bypass (:), gastric bypass (-), lean (>), and obese (B) subjects. (B) The insulin
increment between baseline and 15 minutes in the four groups. (Adapted from le Roux CW,
Aylwin SJ, Batterham RL, et al. Gut hormone profiles following bariatric surgery favor an
anorectic state, facilitate weight loss, and improve metabolic parameters. Ann Surg 2006;
243:108–114; with permission.)
Summary
Weight loss surgery offers a rational treatment option for morbidly obese
patients with the metabolic syndrome. While there are no standard guidelines
for the optimal timing for patients with metabolic syndrome to have bariatric
surgery, the excellent results presented in this article may call for an earlier sur-
gical treatment of morbidly obese patients with the metabolic syndrome. Re-
search into the incretin modulating mechanisms of bariatric surgery may lead
to the development of pharmacologic treatment options in the future.
References
[1] Christou NV, Sampalis JS, Liberman M, et al. Surgery decreases long-term mortality, mor-
bidity, and health care use in morbidly obese patients. Ann Surg 2004;240:416–23.
[2] Gastrointestinal surgery for severe obesity: National Institutes of Health consensus develop-
ment conference statement. Am J Clin Nutr 1992;55:615S–9S.
[3] Flegal KM, Graubard BI, Williamson DF, et al. Excess deaths associated with underweight,
overweight, and obesity. JAMA 2005;293:1861–7.
1270 KINI et al
[4] Oldridge NB, Stump TE, Nothwehr FK, et al. Prevalence and outcomes of comorbid met-
abolic and cardiovascular conditions in middle- and older-age adults. J Clin Epidemiol
2001;54:928–34.
[5] Buchwald H, Williams SE. Bariatric surgery worldwide 2003. Obes Surg 2004;14:1157–64.
[6] Balsiger BM, Poggio JL, Mai J, et al. Ten and more years after vertical banded gastroplasty
as primary operation for morbid obesity. J Gastrointest Surg 2000;4:598–605.
[7] Lee CM, Cirangle PT, Jossart GH. Vertical gastrectomy for morbid obesity in 216 patients:
report of two-year results. Surg Endosc 2007, epub ahead of print.
[8] Shikora SA, Storch K. Implantable gastric stimulation for the treatment of severe obesity:
the American experience. Surg Obes Relat Dis 2005;1:334–42.
[9] Mason EE, Ito C. Gastric bypass in obesity. Surg Clin North Am 1967;47:1345–51.
[10] Higa KD, Ho T, Boone KB. Laparoscopic Roux-en-Y gastric bypass: technique and 3-year
follow-up. J Laparoendosc Adv Surg Tech A 2001;11:377–82.
[11] Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels after diet-induced weight
loss or gastric bypass surgery. N Engl J Med 2002;346:1623–30.
[12] Stenstrom B, Zhao CM, Tommeras K, et al. Is gastrin partially responsible for body weight
reduction after gastric bypass? Eur Surg Res 2006;38:94–101.
[13] MacLean LD, Rhode BM, Nohr CW. Late outcome of isolated gastric bypass. Ann Surg
2000;231:524–8.
[14] Wittgrove AC, Clark GW, Schubert KR. Laparoscopic gastric bypass, Roux-en-Y: tech-
nique and results in 75 patients with 3–30 months follow-up. Obes Surg 1996;6:500–4.
[15] Nguyen NT, Lee SL, Goldman C, et al. Comparison of pulmonary function and postoper-
ative pain after laparoscopic versus open gastric bypass: a randomized trial. J Am Coll Surg
2001;192:469–76.
[16] Fisher BL, Atkinson JD, Cottam D. Incidence of gastroenterostomy stenosis in laparoscopic
Roux-en-Y gastric bypass using 21- or 25-mm circular stapler: a randomized prospective
blinded study. Surg Obes Relat Dis 2007;3:176–9.
[17] Gumbs AA, Duffy AJ, Bell RL. Incidence and management of marginal ulceration after lap-
aroscopic Roux-Y gastric bypass. Surg Obes Relat Dis 2006;2:460–3.
[18] Comeau E, Gagner M, Inabnet WB, et al. Symptomatic internal hernias after laparoscopic
bariatric surgery. Surg Endosc 2005;19:34–9.
[19] Higa KD, Boone KB, Ho T. Complications of the laparoscopic Roux-en-Y gastric bypass:
1,040 patients–what have we learned? Obes Surg 2000;10:509–13.
[20] Schauer PR, Ikramuddin S, Gourash W, et al. Outcomes after laparoscopic Roux-en-Y gas-
tric bypass for morbid obesity. Ann Surg 2000;232:515–29.
[21] Ponce J, Paynter S, Fromm R. Laparoscopic adjustable gastric banding: 1,014 consecutive
cases. J Am Coll Surg 2005;201:529–35.
[22] Bowne WB, Julliard K, Castro AE, et al. Laparoscopic gastric bypass is superior to adjust-
able gastric band in super morbidly obese patients: a prospective, comparative analysis. Arch
Surg 2006;141:683–9.
[23] Marceau P, Hould FS, Simard S, et al. Biliopancreatic diversion with duodenal switch.
World J Surg 1998;22:947–54.
[24] Hess DS, Hess DW. Biliopancreatic diversion with a duodenal switch. Obes Surg 1998;8:
267–82.
[25] Scopinaro N, Adami GF, Marinari GM, et al. Biliopancreatic diversion. World J Surg 1998;
22:936–46.
[26] Ren CJ, Patterson E, Gagner M. Early results of laparoscopic biliopancreatic diversion with
duodenal switch: a case series of 40 consecutive patients. Obes Surg 2000;10:514–23.
[27] Aggarwal S, Kini SU, Herron DM. Laparoscopic sleeve gastrectomy for morbid obesity:
a review. Surg Obes Relat Dis 2007;3:189–94.
[28] Pratt GM, McLees B, Pories WJ. The ASBS bariatric surgery centers of excellence program:
a blueprint for quality improvement. Surg Obes Relat Dis 2006;2:497–503.
[29] Pories WJ, Swanson MS, MacDonald KG, et al. Who would have thought it? An operation
proves to be the most effective therapy for adult-onset diabetes mellitus. Ann Surg 1995;222:
339–50.
[30] Sjostrom L, Lindroos AK, Peltonen M, et al. Lifestyle, diabetes, and cardiovascular risk fac-
tors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–93.
[31] Polyzogopoulou EV, Kalfarentzos F, Vagenakis AG, et al. Restoration of euglycemia and
normal acute insulin response to glucose in obese subjects with type 2 diabetes following
bariatric surgery. Diabetes 2003;52:1098–103.
[32] Morinigo R, Moize V, Musri M, et al. Glucagon-like peptide-1, peptide YY, hunger, and sa-
tiety after gastric bypass surgery in morbidly obese subjects. J Clin Endocrinol Metab 2006;
91:1735–40.
[33] Laferrere B, Heshka S, Wang K, et al. Incretin levels and effect are markedly enhanced one
month after Roux-en-Y gastric bypass surgery in obese patients with type 2 diabetes. Diabe-
tes Care 2007;30(7):1709–16.
[34] Farilla L, Bulotta A, Hirshberg B, et al. Glucagon-like peptide 1 inhibits cell apoptosis and
improves glucose responsiveness of freshly isolated human islets. Endocrinology 2003;144:
5149–58.
[35] le Roux CW, Aylwin SJ, Batterham RL, et al. Gut hormone profiles following bariatric sur-
gery favor an anorectic state, facilitate weight loss, and improve metabolic parameters. Ann
Surg 2006;243:108–14.
[36] Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-
analysis. JAMA 2004;292:1724–37.

Metabolic Syndrome

Încărcat de

Informații document

Descriere originală:

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Metabolic Syndrome

Încărcat de

Drepturi de autor:

Formate disponibile

Med Clin N Am 91 (2007) xiii–xv

Robert T. Yanagisawa, MD Derek LeRoith, MD, PhD

Why a comprehensive review on metabolic syndrome so many years after

hypertension and to endothelial dysfunction. Primary prevention of meta-

with metabolic syndrome. The eﬀect on the renin–angiotensin system block-

The Metabolic Syndrome:

The metabolic syndrome (MetS) consists of a cluster of risk factors that

* Division of Endocrinology, Diabetes, and Nutrition, St. Luke’s-Roosevelt Hospital,

Box 1. NCEP Adult Treatment Panel III (ATPIII) Definition of the

Box 2. WHO definition of the metabolic syndrome

Box 3. IDF consensus worldwide definition of the metabolic

Predictive value for diabetes

Predictive value for cardiovascular disease

High blood pressure

Combination of the risk factors

Other components of the metabolic syndrome

such as nicotinamide adenine dinucleotide phosphate (NADPH) oxidase,

How clinicians and academicians describe the concept popularly known

Historical perspective: the evolution of the metabolic syndrome

a name given by cardiologists to the phenomenon of angina pectoris associ-

syndrome are ‘‘ambiguous or incomplete’’ and ‘‘the rationale for thresholds

Deﬁning the metabolic syndrome: what is in a name?

Does diagnosing the metabolic syndrome improve the ability to estimate

How clearly deﬁned is the pathophysiology of the metabolic syndrome?

proliferator-activated receptor-a [25]. Adiponectin inhibits tumor necrosis

[33]; essential hypertension [34,35]; impaired glucose tolerance and prediabe-

What is the beneﬁt of metabolic syndrome–focused treatment?

Primary prevention of diabetes mellitus

Nonpharmacologic treatment of hypertension and prehypertension

Primary and secondary coronary heart disease prevention with statins

atorvastatin: after a mean 4 years’ follow-up, primary end points occurred in

Primary and secondary coronary heart disease prevention with ﬁbrates

Secondary prevention of atherosclerotic events with thiazolidinediones

Drug choices in metabolic syndrome

Metabolic syndrome–tailored treatment recommendations

Metabolic syndrome: a look toward the future

adiponectin and elevated adipocyte-derived inﬂammatory peptides responsi-

Insulin Resistance as the Underlying

The metabolic syndrome comprises multiple features, including visceral

Defective insulin secretion and signaling

release of insulin in response to a glucose load [3]. This deﬁciency of acute

action include agents that cause phosphorylation of serine or threonine

Dysregulation of glucose disposal and production

Agents that cause

most important glucose transporters, GLUT4, is regulated by insulin. In re-

endocytosis of the transporter, the result of activation of AMP kinase [23].

Role of free fatty acids

Impaired lipid metabolism

(Apo A-I). Additionally, chylomicron formation and association with its

Adipose tissue, cytokines, and proinﬂammatory states

skeletal muscle, adiponectin aids in glucose transport and FFA oxidation,

Identifying insulin resistance

Other research tools used to measure insulin sensitivity include the

[28] Puigserver P, Rhee J, Donovan J, et al. Insulin-regulated hepatic gluconeogenesis through

[74] Furuhashi M, Ura N, Higashiura K, et al. Blockade of the renin-angiotensin system

America on the Move

The problem we are facing

This work was supported by NIH grant DK 42549.

Individuals maintaining a healthy weight have become the minority. Regret-

Understanding the small change approach to weight management

America on the Move

AOM tools for individuals

AOM tools for groups