The Stages in Diabetic Renal Disease

With Emphasis on the Stage

of Incipient Diabetic Nephropathy
C. E. MOGENSEN, C. K. CHRISTENSEN, AND E. VITTINGHUS

SUMMARY

Alterations in renal function and structure are found

even at the onset of diabetes mellitus. Studies performed over the last decade now allow definition of a
series of stages in the development of renal changes
in diabetes. Such a classification may be useful both
in clinical work and in research activities.
Stage 1 is characterized by early hyperfunction and
hypertrophy. These changes are found at diagnosis,
before insulin treatment. Increased urinary albumin excretion, aggravated during physical exercise, is also a
characteristic finding. Changes are at least partly reversible by insulin treatment.
Stage 2 develops silently over many years and is
characterized by morphologic lesions without signs of
clinical disease. However, kidney function tests and
morphometry on biopsy specimens reveal changes.
The function is characterized by increased GFR. During good diabetes control, abumin excretion is normal;
however, physical exercise unmasks changes in albuminuria not demonstrable in the resting situation. During poor diabetes control albumin excretion goes up
both at rest and during exercise. A number of patients
continue in stage 2 throughout their lives.
Stage 3, incipient diabetic nephropathy, is the forerunner of overt diabetic nephropathy. Its main manifestation is abnormally elevated urinary albumin excretion, as measured by radioimmunoassay. A level
higher than the values found in normal subjects but
lower than in clinical disease is the main characteristic
of this stage, which appeared to be between 15 and
300 |xg/min in the baseline situation. A slow, gradual
increase over the years is a prominent feature in this
very decisive phase of renal disease in diabetes when
blood pressure is rising. The increased rate in albumin
excretion is higher in patients with increased blood
pressure. GFR is still supranormal and antihypertensive treatment in this phase is under investigation, using the physical exercise test.
From the Second University Clinic of Internal Medicine, Kommunehospitalet,
8000 Aarhus C, Denmark.
Address reprint requests to C. E. Mogensen at the above address.

64

Stage 4 is overt diabetic nephropathy, the classic

entity characterized by persistent proteinuria (>0.5 g/
24 h). When the associated high blood pressure is left
untreated, renal function (GFR) declines, the mean fall
rate being around 1 ml/min/mo. Long-term antihypertensive treatment reduces the fall rate by about 60%
and thus postpones uremia considerably.
Stage 5 is end-stage renal failure with uremia due to
diabetic nephropathy. As many as 25% of the population presently entering the end-stage renal failure programs in the United States are diabetic. Diabetic nephropathy and diabetic vasculopathy constitute a
major medical problem in society today. DIABETES 32
(Suppl. 2):64-78, 1983.

n comparison with other renal disorders, the natural history of renal involvement in insulin-dependent diabetes
follows a very characteristic and probably unique pattern. For a long period of 10-20 yr or more, structural
components in the kidney are enlarged with concomitant
hyperfunction. Thereafter, in a large proportion of the patients (around 30-50%), renal function starts to deteriorate,
first evidenced by proteinuria.12 The initial phase of abnormal
increase in protein excretion is not detectable by ordinary
laboratory methods; it requires more sensitive methods, such
as radioimmunoassay34 or sophisticated immunochemical
methods.
Thus, a series of stages in the development of renal changes
in diabetes is recognizable:
(1) Early hypertrophy-hyperfunction. Changes are found
at diagnosis before insulin treatment is started. They may
persist for many years when diabetes control is poor.356
(2) Glomerular lesions without clinical disease. This silent
stage develops over many years with structural lesions on
biopsy but without clinical or laboratory signs of renal disease.78 About 30-50% of patients develop clinical nephropathy, and the remaining 50-70% do not. Further studies of
this phase are necessary to identify factors of importance
for this progression into stages 3 and 4.

DIABETES, VOL 32, SUPPL 2, JUNE 1983

C. E. MOGENSEN, C. K. CHRISTENSEN. AND E VITTINGHUS

Standard
control

Onset of Clinical Insulin

diabetes diagnosis treatment

GFR

+ 40%-

Proteinuria: fall in GFR

( +increased blood pressure)

1 I

Non-proteinuric patients

N(SD) -40%-

days

weeks

20

years

25

30 years

Exercise
Albumin
excretion

Proteinuria

+ 100% -

Non-proteinuria

N(SD) IV

III
FIGURE 1. Development of kidney function
changes in juvenile diabetes mellitus: GFR and
albumin excretion.

days

weeks

20

years

25

30

more general concept of the possible linking of the early

changes, both the hemodynamic15 and metabolic/structural
entities,67 to the late clinical nephropathy is important for the
understanding of diabetic renal disease,615 and possibly more
so for understanding the mechanism that governs the final
common pathway to decline in renal function and nephron
destruction in a variety of nephropathies.16 Figure 1 provides
an outline of the changes described above. A different pattern may be relevant for type II diabetes.1314

(3) Incipient diabetic nephropathy. The primary manifestation of this stage is abnormally elevated baseline urinary
albumin excretion as measured by radioimmunoassay. Based
on recent studies the range between 15 and 300 (j.g/min is
used for diabetics in good metabolic control examined in
the laboratory in the sitting or recumbent position. Slowly
increasing albumin excretion over several years is also a
feature in this very decisive phase of renal disease in diabetes.
(4) Overt diabetic nephropathy. This is the classic stage
with its major characteristic being a persistent proteinuria
(>0.5 g/24 h) in the absence of findings of nondiabetic renal
disease.9 Renal function declines when associated high blood
pressure is left untreated; the mean fall rate in GFR is 1 ml/
min/mo.10"12
(5) End-stage renal failure due to diabetic nephropathy.
Uremia caused by diabetes mellitus, both insulin-dependent
and non-insulin-dependent, is found in as many as 25% of
the population presently entering end-stage renal failure programs in the U.S. and constitutes a major medical problem;
perhaps the most crucial problem in clinical nephrology.11314
Classifying patients into these five categories may be useful, both in clinical and in research activities. Identification
of the determinants for the transition from one stage to the
next is vital if related questions are to be clarified. Also, the

STAGE 1. EARLY HYPERTROPHY-HYPERFUNCTION

Systematic studies over the last decade reveal that multiple

renal changes are already present at the clinical onset of
diabetes mellitus. These changes are predominant with regard to structure, biochemistry, and function, justifying the
definition of a new entity, and the early hypertrophy-hyperfunction stage.36717 Renal size, in human diabetes and in
the experimental animal, is markedly increased early in the
course and concomitantly increased function also is found
both in human and experimental animal diabetes.1719 Standard insulin treatment leads to reduction in both renal size
and GFR over 3 mo, suggesting that the changes stem from
metabolic effects that are only slowly reversible.3 Partial normalization of the functional parameter is found earlier.5 It is
possible that in the early course of diabetes mellitus in-

TABLE 1
Experimental setup in the exercise provocation test
Periods
Start
drinking water
every 20 min

-120

Pre-exercise
1

Exercise
1

40

Postexercise
1

80

140 min

Measurements: Heart rate by counting pulse rate.

Blood pressure by auscultatory technique.
Urinary albumin excretion by RIA.
Urinary B-2-microglobulin excretion by RIA.

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

65

STAGES IN DIABETIC RENAL DISEASE

100

60

20

0.3

(o-O) before insulin

(-) after 6-15 days
of insulin treatment
in 6 diabetic patients
Urinary
B-2-microglobulin

0.1

1/2.

Pre

Exercise

2. M i
Post

2.
periods

Exercise loads; 450 and 600 kpm/min.

M - mean of 2.exercise and Lpost period.
FIGURE 2. Urinary albumin and p-2-microglobulin excretion at the onset of diabetes, before and after insulin treatment.

creased size of the kidney and associated phenomena are

linked to the development of at least some aspects of the
late diabetic glomerulosclerosis.
It has also been established that the important parameter
used in the evaluation of renal function in diabetes mellitus,
urinary albumin excretion, is elevated in uncontrolled diabetes mellitus. These changes are even more pronounced
during a renal provocative test, i.e., the exercise provocation
test.418
The usual procedure in this test is shown in Table 1. It is
important that a long baseline period, 2 h with water drinking
every 20 min, is used, because a very modest temporary
increase in urinary albumin excretion may be seen when
normal subjects start to drink water. Larger loads of water,
e.g., 1 l_ water during 5 min, lead to a more pronounced
increase in urinary albumin excretion.20 After 1.5 h, excretion
rate is stable and the experimental urine collection may start.
Exercise loads are used that do not normally provoke increase in albumin excretion in nondiabetic subjects of similar
age and sex to the diabetics examined in this study.
As depicted in Figure 2, pronounced increases in albumin
excretion occur during the exercise provocation test in untreated diabetics that is an aggravation of already higher
baseline value.18 After 6-15 days of ordinary insulin treatment, a perfectly normal test is seen, e.g., normal baseline
values and no increase in albumin excretion during exercise.
Exercise-induced changes in blood pressure were normal
in these patients. No significant difference was found either
before or after treatment. However, a significant fall in exercise-induced tachycardia was observed after treatment.18
Beta-2-microglobulin excretion did not increase during the
exercise test, indicating undisturbed tubular function. The
abnormal excretion can be attributed to increased glomerular permeability and not a decreased tubular reabsorption
of proteins. The exact mechanism behind the exercise-induced increase in albumin excretion remains to be clarified.
An abnormally high intraglomerular pressure may operate
to produce increased permeability to protein in the untreated

66

situation, further aggravated by exercise. This abnormality

is readily normalized by standard insulin treatment.
After some years of diabetes, abnormal albumin excretion
emerges during exercise. In this situation, the abnormalities
probably involve an abnormal structure of the glomerular
basement membrane. The changes are even more marked
in diabetics with increased baseline albuminuria (incipient
diabetic nephropathy). In this stage, the mechanism must
be different from abnormalities emerging in the untreated
situation at diagnosis of diabetes.
These data, along with earlier described changes of GFR,
RPF, and renal size, suggest a combination of structural and
functional changes early in diabetes. No data are available
on the intraglomerular pressure in human diabetes, but in
the experimental animal, an increased transglomerular pressure gradient has been demonstrated.15
STAGE 2. GLOMERULAR LESIONS WITHOUT CLINICAL
DISEASE

For many years after the start of insulin treatment, during

standard control, the characteristic feature is that all standard clinical parameters related to renal function are normal.
However, using the classical renal function test outlined by
Homer Smith, numerous studies have shown a very consistent elevation of glomerular filtration rate. This increase is
on the order of 20-30%, whereas the increase in the untreated situation is on the order of 30-40%. 358 The elevated
renal function is most likely linked to the imperfect metabolic
control obtained during our standard insulin regimens. GFR
was shown to be more increased than extracellular volume
in type I diabetes, suggesting what has been termed "a real
renal hyperfunction."21 Changes in growth hormone and
glucagon may also influence the final level of renal function.23~25 In studies primarily in type II diabetes, exchangeable body sodium was increased.26
Data on renal plasma flow are conflicting.2225 Using labeled hippuran for measuring renal plasma flow, a slight

Urinary albumin

500
200

50E
E

XXX

20

- D (11)

xxx

= N (10)
5- T

yr

2-

Urinary B-2-microglobulin
0.2
0.05

Pre 300
periods

450

600

750

900

1200

Exercise loads in kpm/min.

FIGURE 3. Graded exercise in diabetic and normal man.

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

MOGENSEN, C. K. CHRISTENSEN. AND E. VITTINGHUS

(jg/min

|jg/min

200

150
18 years
Diabetes duration

100

50

12

18 years
Diabetes duration

15
years
Diabetes duration

FIGURE 4. Albumin excretion in the development of incipient diabetic nephropathy: example of courses in three patients.

increase has been observed in some studies but normal or

even depressed values have been found in others.3 This
discrepancy may be explained by different degrees of metabolic control during the clearance procedure. Renal extraction of para-amino hippurate (PAH) proved to be normal
in diabetic patients,27 so there is no firm evidence to suggest
that enhanced GFR is simply a consequence of increased
flow to the kidney. Indeed, when calculating RPF per gram
kidney weight, the value is depressed.5 Well-preserved renal
function is found in all patients without proteinuria despite
very long diabetes duration 3 This means that the majority of
long-standing diabetics continues to have elevated renal
function and a much better prognosis than those with proteinuria.12
The well-preserved kidney function found in many patients
is evidenced by the fact that normal albumin excretion persists, despite the fact that pronounced structural alterations
may be found when renal biopsy specimens are examined.
It is characteristic that a gradual increase in the thickness
of the glomerular basement membrane is found with increase
in duration of diabetes followed by changes in the volume
of the glomerular mesangium.7
The apparent paradox, i.e., structural abnormalities associated with normal function, was the impetus for our development of a provocation test. It is based on the concept
that some kind of provocation could disclose changes not
readily demonstrable in the baseline situation when abnormalities are still suspected. Exercise seems a reasonable

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

choice since extreme exercise is known to provoke proteinuria in normal man. Our exercise test employed an exercise load insufficient to cause albumin excretion in normal
18,28-30

The exact relationship between intensity of exercise and

increase in aibuminuria was recently established.18 This study
confirmed that diabetic patients, after some years' duration
of the disease, exhibit increases in albumin excretion at a
work load of 600 kpm/min during a 20-min period. The increase in albumin excretion is much more pronounced at
moderately high work loads, as shown in Figure 3. Comparable controls show significant increase only at a much
higher work load, namely 1200 kpm/min for 20 min, although
there was a clear tendency for an increment in aibuminuria
at 900 kpm/min. The increase in albumin excretion is of glo-

TABLE 2
Yearly increase in albumin excretion in incipient and overt
diabetic nephropathy
Incipient diabetic nephropathy
(N = 7)
(range -1.2-76)
Overt diabetic nephropathy
(N = 7)

2328 ^ ^ 3206 (SD)

yr
(range -99-9087)

In both cases, progression in nephropathy was blood-pressure

dependent.

67

STAGES IN DIABETIC RENAL DISEASE

Studies over 5.8 years

ug/min
year

R = 0.79
2p = 3.4%
n -7

50

new manner of treatment is promising for the amelioration

of renal complications, as well as other microvascular lesions. It is indeed possible that vigorous control of metabolism must be instituted early in the course, before organ
damage both in the kidney and the retina is established.
A number of patients with still-normal baseline albuminuria
are expected to develop increased protein excretion and
later to develop overt nephropathy.
We are currently examining the possibility that the exercise
provocation test may serve as an identifier of patients prone
to nephropathy development. Preliminary data show that a

ug/min

GFR
154

GFR
142

GFR
126

150

100

85

95

105
mm Hg
Diastolic pressure
50

FIGURE 5. Yearly increase in albumin excretion plotted against diastolic pressure in seven patients with incipient diabetic nephropathy.

merular origin both in normal and diabetic subjects. This

conclusion is drawn from the observation that beta-2-microglobulin excretion was not affected by exercise, except
from a small increase in diabetic patients at 1200 kpm/min.
In percent of the maximal increase in heart rate, the elevation in albumin excretion in the diabetics was significant
at 55% (-600 kpm/min) and the increase in albumin excretion was more pronounced at higher work loads (Figure 3).
The albumin excretion in the controls showed a significant
increase only at 88% (-1200 kpm/min) of their maximal increase in heart rate, but there was a tendency to elevation
already at 65%. The diabetics showed a significant difference to the normals in albumin excretion when they were
working at 55-82% of the maximal heart rate. Then the best
test area seems to be from 55% to 65% of the maximal heart
rate, where none of the controls showed elevated albumin
excretion.
Systolic blood pressure tended to increase markedly in
diabetic patients although no significant difference could be
established and there was no correlation between increases
in albuminuria and blood pressure increase, either in normals
or diabetics.1830 However, it is difficult to completely rule out
that an abnormal exercise-induced blood pressure increase
contributes to albuminuria in diabetic patients in this early
phase of renal involvement. Exercise-induced increases in
blood pressure and heart rate in diabetic patients are readily
depressed by acute administration of the cardioselective
beta-blocking agent metropolol in doses of 5-10 mg i.v.30
The recent studies by Koivisto et al.31 and by Viberti et al.32
documented normalization of exercise-induced albuminuria
after treatment by continuous subcutaneous administration
of insulin using portable pumps. This normalization of exercise-induced albuminuria probably reflects functional
changes rather than a regression of structural lesions. This

68

O
18

years

Diabetes duration

GFR
137

GFR
135

GF R
13

300
/

250

I
200

150

100

50

0
years
Diabetes duration

FIGURE 6. Albumin excretion before and during antihypertensive

treatment using cardioselective beta-blocking agents.

DIABETES, VOL 32, SUPPL. 2, JUNE 1983

C. E. MOGENSEN, C. K. CHRISTENSEN, AND E. VITTINGHUS

600

500

400

300

200

100

-2

Baseline

450

600kpm 1

EXERCISE

Post exercise

FIGURE 7. Urinary albumin excretion during exercise in young male

diabetics with incipient diabetic nephropathy.

rather long observation period is necessary. A group of patients was reexamined 5 yr after the original exercise test
without any definitive data on the predictive value of the
exercise test. The predominant finding was that very few
patients developed nephropathy in this period of time.
STAGE 3. INCIPIENT DIABETIC NEPHROPATHY

There can be no doubt that proteinuria is an important prognostic indicator in the course of diabetes mellitus. It is not
very harmful in itself, but it signals an increase in blood

pressure and future decline in renal function. This is a wellestablished fact, confirmed in numerous studies.
What is not clear, however, is the role of the incipient rise
in albuminuria or microalbuminuria found in diabetic patients
after a duration of diabetes of about 10-15 yr. We call this
phase incipient diabetic nephropathy to distinguish it from
the overt diabetic nephropathy. Little data are available on
this phase of renal disorder in diabetes despite the fact that
this period is of particular interest from a clinical point of
view, since it is the stage where effective intervention stands
a better chance. We are dealing with patients predicted to
develop pronounced renal damage, i.e., the interphase between well-preserved even supranormal kidney function, and
the phase where deterioration in renal function is manifest
and where effective intervention becomes difficult.
We propose to define incipient diabetic nephropathy as
baseline urinary albumin excretion between 15 and 300 |xg/
min; this is above the normal range but below the values
generally found in overt diabetic nephropathy. Importantly,
GFR is well preserved, even supranormal in these patients.
It may prove useful to use a more narrow range, thus excluding patients with low excretion, who in long-term studies
would appear to show a nonprogressive pattern. Reducing
the upper limit may also prove useful, e.g., omitting patients
with more pronounced alterations in whom the deleterious
course with regard to renal damage may be more difficult
to modify.
We have recently studied seven male patients mean age
28 yr at entry of the study, with a 14 yr mean duration of
diabetes. These patients have until now been observed in a
mean period of 6 yr. The above-mentioned definition was
used (15-300 fjug/min) as a well-defined baseline value.
The main purpose of the study is to evaluate progression
of incipient phase of nephropathy, manifested by increase
in albuminuria (or microalbuminuria, see below) and to relate
occurrence of abnormal albumin excretion to changes in
glomerular filtration rate and renal plasma flow. It would be
of interest to examine blood pressure changes in relation to
increase in albuminuria. Intervention with antihypertensive
treatment may prove more effective in this early phase than

r 0.72
2p- 0 395%

FIGURE 8. Increase in albumin excretion rate

during exercise plotted against increase in systolic blood pressure during exercise (600 kpm/
min).

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

Increase in systolic blood

_ _ j _ pressure (600 kpm/min)
150

69

STAGES IN DIABETIC RENAL DISEASE

ug/min.

GFR

600

137

400

200
155

122/90

135

-2

-1

450 kpm 600 kpm

Lpostex

Baseline

EXERCISE

Postexercise

ug/min,

300

200

100
115/85

126

-2

Lpostex
Baseline

EXERCISE

Postexercise

FIGURE 9. Albumin excretion before and after antihypertensive treatment. Data from exercise provocation test.

in overt diabetic nephropathy where decline in renal function

can be reduced but not arrested.1011 Figure 4 shows examples of the progression in three patients examined several
times over the years with respect to baseline albuminuria. It
is evident that a rather variable course is found (Table 2).
Blood pressure was monitored in the observation period
and a correlation between blood pressure level and increase
in albuminuria was established as documented in Figure 5.
Most likely the increase in albuminuria is caused by the high
blood pressure level observed in some patients, although it
cannot be excluded that albuminuria may produce increase
in systemic blood pressure. Also, both changes may be
correlated with the general severity of the disease, although
no relevant parameter was identified. During the observation
period, there were no changes in the metabolic control as
evaluated by fasting plasma glucose. Beta-2-microglobulin

70

excretion did not increase, indicating that tubular function is

undisturbed and that the changes observed in albumin excretion are due to progressive glomerular changes.
Renal function was well preserved in these patients. Thus
glomerular filtration rate was still clearly elevated both at
entry in the study and after an observation period of 6 yr
[142 ml/min 10 (SD) -> 139 ml/min 18 (SD)]. In contrast,
a significant fall in renal plasma flow was noted in this period.
Initial renal plasma flow values compare well with the level
typically found in diabetic patients without proteinuria (530550 ml/min).322-26 There was a significant increase in diastolic
blood pressure during the observation period [86 mm Hg
9 (SD) -> 92 12]. It is tempting to speculate that the
changes in RPF reflect microcirculatory changes in glomeruli
and may be related to the incipient blood pressure rise observed.
Based on these observations, we are conducting a trial of
the effect of antihypertensive treatment on renal function
using cardioselective beta-blocking agents in incipient diabetic nephropathy. Preliminary data are shown in Figure 6
in patients followed for 6 and 1.5 yr during hypertensive
treatment. A dramatic decrease in albuminuria is observed
after the initiation of antihypertensive treatment. Intervention
in this phase would seem logical since renal function is still
well preserved but appears bound to decline. In the subsequent phase, overt diabetic nephropathy, GFR is already
declining and its rate of fall can be ameliorated by antihypertensive treatment. However, more effective treatment resulting in total elimination of decline in renal function appears
feasible, and further trials should establish whether antihypertensive treatment in this early phase will prove effective
before the downhill course of renal function, characteristic
for diabetic nephropathy, is established.
Exercise-induced albuminuria. We have recently studied
14 patients in the stage of incipient diabetic nephropathy by
the exercise provocation test, using procedures described
earlier33 (Table 1). These patients exhibit well-preserved kidney function. Systolic blood pressure is not above the normal
range, although the mean values are higher than those found
in normal subjects. However, diastolic pressure is slightly
but significantly increased in these patients, namely 92.5 mm
Hg 8.1 versus 79.5 12.4 (SD) in normals (2P < 0.01).
Systolic pressures were 130.2 11.8 versus 120 14.1 in
normals. There was no correlation between baseline albuminuria and blood pressure. All subjects were males with a
mean age of 26.3 yr and a mean duration of diabetes of 16
yr.
Baseline albumin excretion values were increased as a
criterion for subject selection. At an exercise load of 450
kpm/min for 20 min further increase in albumin excretion is
observed (Figure 7). Patients with normal baseline excretion
values only respond with a rise in albumin to 600 kpm/20
min and above. The increase in incipient nephropathy is
further aggravated at 600 kpm/min and in the first postexercise period, and thereafter values return to the preexercise
level. These studies demonstrate that even light exercise
further amplifies the already abnormal baseline values. The
variability of diabetic albuminuria commonly seen in clinical
practice may be explained by changes in physical activity
during 24-h collection periods.
Blood pressure involvement was unmasked during the ex-

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

C. E. MOGENSEN. C. K. CHRISTENSEN. AND E. VITTINGHUS

+Albuminuria (418 ug/min)

-Albuminuria (7.2 ug/mm)

MSEM

FIGURE 10. Percentage change in creatinine

clearance after 8 IU insulin i.v. in five patients
with and without slight albuminuria. Difference in
percentage fall between groups: 3.9% versus
17.1% (2P = 0.05).

ercise test. Systolic pressure increased to abnormally high

levels in the diabetics. A highly significant correlation between exercise-induced albumin excretion and systolic pressure was also documented (Figure 8). No correlation was
found between exercise-induced albumin excretion and the
exercise-mediated increase in heart rate.
The exercise test may prove useful in monitoring the progress of nephropathy in the incipient phase, and also during
therapeutical trials such as antihypertensive treatment or
during supercontrol of blood glucose. Figure 9 shows the
effect of antihypertensive treatment for 15 mo and 8 yr on
the exercise-induced albuminuria in patients with incipient
diabetic nephropathy. Marked depression of both baseline
and exercise-induced albuminuria is found and trials are
being undertaken to further establish these findings. The
exact mechanism responsible for the abnormal albuminuria
during exercise in incipient nephropathy remains to be clarified. Other mechanisms may operate in incipient nephropathy compared with newly diagnosed diabetes. Blood pressure may play an important role for the enhancement of
albuminuria in incipient nephropathy, but microcirculatory
changes in glomeruli may also be involved and, importantly,
these patients would exhibit marked morphologic changes,
although biopsies were not performed. Changes in renal
autoregulation in connection with arteriolar hyalinosis may
be an important contributory factor.34 Evidently structural lesions in the glomerulus proper are very likely to be involved
as the mechanism. Tubular alterations do not appear to play
a role in the exercise-induced albuminuria, since beta-2microglobulin excretion is undisturbed in these patients during physical exercise.
The concept of microalbuminuria. Recent work has established an important role for microalbuminuria in monitoring the development of renal changes in diabetic patients.
Our studies show that patients with baseline values greater
than 15 |ig/min but less than 300 have a clear tendency to
progression, agreeing with the report of Viberti and co-workers on long-term follow-up studies in patients with baseline
albuminuria between 30 and 140 |xg/min.35
More work is needed to screen the diabetic population at
risk, e.g., insulin-dependent patients with diabetes duration
of 10 yr or more but no clinically detectible proteinuria. In

DIABETES, VOL. 32, SUPPL 2, JUNE 1983

100

+3

periods

the phase of incipient diabetic nephropathy, a large proportion of the patients have microalbuminuria (defined as
abnormally increased values, but below the clinically detectible level).
There may well exist a stage where effective intervention
is possible both with antihypertensive treatment, but possibly
also with supercontrol of blood glucose by continuous subcutaneous insulin infusion.

^g/min
7O0Or

years
Diabetes duration

FIGURE 11. Albumin excretion in the development of incipient and

overt diabetic nephropathy.

71

STAGES IN DIABETIC RENAL DISEASE

10

20

50

40

50

60

70

loo

no
Months

so

IOO

no

120

GFR
ml /min
100eatment
Start ol treatment

75-

50-

25-

10

20

30

50
Months

I Start of treatment

FIGURE 12. GFR before and during antihypertensive treatment in five diabetic patients with overt diabetic nephropathy.

Possible manifestations of loss of autoregulation in the

kidney in early diabetic nephropathy. Recent studies have
documented that administration of large doses of intravenous insulin provokes increased albumin excretion in patients with normal baseline albumin excretion.36 However,
patients with enhanced baseline values show a paradoxic
decline in albuminuria.37 Marked reduction in creatinine excretion, probably reflecting a fall in GFR, was found after
administration of insulin to patients with early nephropathy
(Figure 10). It is possible that this is a manifestation of impaired renovascular autoregulation in this phase of the disease. Such a loss of autoregulation may also be responsible
for generation of high systemic pressures, e.g., during exercise to the glomerular capillary bed. Blood pressure elevation may be of importance as an aggravator of nephropathy responsible for the rapid progression seen several years
after a slowly progressive increase in microalbuminuria. The
course of an individual patient is depicted in Figure 11.

72

STAGE 4. OVERT DIABETIC NEPHROPATHY

Patients with persistent and pronounced proteinuria all evidence decrease in glomerular filtration rate. In our studies,
the fall rate of GFR has been correlated with diastolic blood
pressure at the end of the control period. The mean decrease
was found to be of the order of 1 ml/min/mo.1011 A similar
decline in renal function has been observed by other
investigators1238 and also during pregnancy in patients with
diabetic nephropathy.39
Antihypertensive trials, conducted for a number of years,
have shown that effective treatment reduces the rate of decline of GFR.101140 In Figure 12 is shown the decline rates of
GFR in individual patients before and during antihypertensive treatment. Progression of nephropathy, measured by
increase in albuminuria, expressed as albumin clearance as
percent of GFR, was markedly ameliorated by antihypertensive treatment (Figure 13). As summarized in Figure 14, progression of nephropathy as rate of decline of GFR is reduced

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

C. E. MOGENSEN, C. K. CHRISTENSEN, AND E. VITTINGHUS

per cent
of GFR

SO
0

10

30

40

00

ro

80

90

100

HO

120

-1

per cent
ol GFR

Albumin clearance
per cent
of GFR

Start ol treatment

Start of treatment
1.0O
o o

0.1Start ol treatment

10

20

30

40

50
Months

Albumin clearance
as % of GRF
1.0 T

Start of treatment

FIGURE 13. Albumin clearance as percent of GFR before and during antihypertensive treatment. Patients with overt diabetic nephropathy.

by 60% during antihypertensive treatment. Progression in

albuminuria was completely halted during long-term treatment, and blood pressure level was significantly reduced
from 162 14 (SD)/103 9 to 144 11 /95 8 (2P < 0.02).
It is important to note that filtration rate still declines significantly during antihypertensive treatment,40 and indeed in
some patients effective blood pressure control was difficult
to achieve, not close to the goal of 140/90 or lower (Figure
15). To examine the effect of intervention at an earlier stage
would therefore be the next logical step in a series of subsequent investigation.
Benefit from antihypertensive treatment in diabetic nephropathy was further supported by the finding of Parving
and co-workers. Their studies agree with our findings.12
The mechanism of proteinuria in the late phase of diabetic

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

nephropathy was studied using very high-molecular-weight

dextran and long renal clearance collection period. Diabetic
patients with heavy proteinuria exhibit an abnormal enhancement in very high-molecular-weight dextran clearance (Figure 16). The changes detectible with the dextran clearance
technique are very small; at highest a few-fold increase in
the very high-molecular-weight dextrans, compared with a
1000-fold increase or more in albumin excretion. These data
agree with the recent studies of Myers and co-workers.4142
The higher-molecular-weight dextran and longer clearance
periods used in our study appear to increase the test's sensitivity; abnormalities may be detectible in less advanced
diabetic nephropathy. In early diabetic nephropathy, dextran
clearance is normal when expressed as percent of GFR.34142
It is likely that the late changes found in the advanced di-

73

STAGES IN DIABETIC RENAL DISEASE

Fall rate of GFR

2.0-

1.24'

1.0

During
treatment

Before
treatment
2p = 0.042

60

40

25.8
20

Before
treatment

namic studies.1543 Hemodynamic changes appear to be important modulating factors rather than being directly responsible for the development of late diabetic nephropathy.8
Changes in early diabetes producing late diabetic nephropathy may involve the following mechanisms operating
alone or in combination: (1) increased basement membrane
material, found early as an increase in the filtration surface
area, could be regarded as a forerunner of later increase in
basement membrane thickness, eventually producing clinical glomerulosclerosis; (2) hemodynamic changes, predominantly an increase in filtration pressure, may produce
glomerular structural disruption. Supporting evidence is obtained in the rat model;15 (3) abnormal extravasation of plasma
protein, probably in connection with increased pressure, may
also be of pathogenic importance.812
One should stress that all the changes are strongly dependent upon the metabolic control, supporting the view that
every effort should be undertaken to optimize diabetes control.
We are still faced with the fact that overt diabetic nephropathy (stage 4) develops in only a large minority of the patients
despite the fact that early changes probably are found in all
patients, although to a variable extent. Other factors may
determine the progression and transmission from one phase
to another. In this respect, the phase of incipient diabetic
nephropathy is of particular interest.
Identification of the factor or factors responsible for transition from the rather slow progression in albuminuria in the
incipient phase to the more dramatic increase in overt nephropathy, depicted in Figure 17, is of great importance.
Abnormal blood pressure, linked to glomerular microcirculatory changes, may play a role. Advanced changes in the
arteriolar system may modulate the course by systemic blood
pressure rises that would be transmitted to the glomerular
circulation, resulting in structural disruption in glomerulus.
Table 3 is a recapitulation of the development of renal
lesions in diabetes. It is evident that the phase of incipient
diabetic nephropathy is a very central entity and it would
seem a fruitful working hypothesis that diabetic nephropathy
may be held in this phase with well-preserved kidney function

During
treatment
2p = 0.009

FIGURE 14. Fall rate of GFR and increase in albuminuria before and
during antihypertensive treatment in overt diabetic nephropathy.

abetic nephropathy are quite nonspecific and comparable

to changes found in advanced nephropathy due to other
renal disorders.
Progression from one stage to a more advanced stage.
Early changes in renal function in diabetes may set the stage
for late diabetic nephropathy, as evidenced by hemody-

74

FIGURE 15. Blood pressure levels and decline in kidney function during antihypertensive treatment in diabetic patients with overt nephropathy. Pressure level of -141/92 mm Hg corresponding to a fall rate of
0.2 ml/min/mo.

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

MOGENSEN. C. K. CHRISTENSEN. AND E. VITTINGHUS

ml/min
of GFR

Absolute values

2p = 3.5%

0.5

0.5

40,000

50,000

60,000

ml/min.

70,000

40,000

50,000

60,000

70,000

Per cent
of GFR 5

40

50

60

70

Absolute clearance
values.

40

50

60 70
Mol.-weight*10'
Clearance values expressed
as per cent of GFR.

B
FIGURE 16. (A) Dextran clearance using Macrodex (R) in four proteinuric diabetics and four normals. Long-term dextran clearance period was
used (3 periods of 4 h each). GFR in the normals was in mean 116 ml/min 1 4 (SD), in the diabetics 95.2 35. Urinary albumin excretion was in
the diabetics in mean 2274 (range 523-5450). The normals had albumin excretion values of 8.0 (range 5.1-12.8). With respect to dextran clearance a significant increase in clearance of the 70,000-mol-wt dextrans as percent of GFR was established (2P - 3.5%), whereas there was not
significant difference in lower-molecular-weight dextrans. The individual value for the 70,000-mol-wt was in the normals 0.36, 0.41, 0.28, and
0.44% in contrast to 0.45, 0.47, 0.52, and 0.63% in the diabetics. The diabetic patients with the highest albumin excretion and lowest GFR also
exhibited the most abnormal dextran clearance value of 63% at mol wt 70,000, and throughout the molecular weight ranges, they exhibited the
highest clearances. Dextran was separated according to molecular size as described earlier.3 (B) High-molecular-weight dextran clearance in
normal subjects (hatched area) and six diabetic patients without proteinuria and with normal albumin excretion (dots).

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

75

Incipient, slow
decline

Without treatment: decline 5

ml/min
mo

Still increased by
20-30%

Without treatment: decline ~ 1

ml/min
mo

Not very well studied

Diffuse and nodular

glomerulosclerosis:
Capsular drops. Fibronoid caps. Arteriolar hyalinosis

Glomerular closure

After 10-15 yr (in

30-40% of patients)

After 15-20 yr (in

30-40% of patients)

Final outcome,
after 25-30 yr

Incipient diabetic
nephropathy

Clinical overt diabetic nephropathy

End-stage renal
failure

low

Normal or
slightly increased

Increased by
20-30%

On renal biopsy, increased basement

membrane thickness,
mesangial expansion
stages 3, 4

Detectable after 2
yr of diabetes,
progression
over several yr

Normal or
slightly increased

Renal lesions, without clinical signs

Increased by
20-40%

Increased kidney size.

Increased glomerular
size. Nephron hypertrophy and hyperplasia

RPF

Present at diagnosis of diabetes

(may continue
for many years
when control
poor)

GFR

Early renal hypertrophy-hyperfunction (before

insulin treatment)

Chronology

Stage

Main structural
changes or
lesions

TABLE 3
Stages in a development of renal changes and lesions in diabetes mellitus (type I)

Abnormal to
very high
mol.-weightdextrans
(nonspecific
for diabetes)

Normal

Normal

Dextran
clearance
(% of GFR)

Often some dedine due to nephron closure

High

Abnormal
(~ 160/105)

Not studied
Progressive clinical proteinuria

Not studied

Incipient increase, aggravated

during exercise

Abnormal, aggravation of
baseline abnormalities

15-300 jjig/min.
Increase by
n c ^g/min
yr

Normal

Blood
pressure

Abnormal after
a few years

Increased (before insulin

treatment)

Exerciseinduced

Albumin excretion

Normal in most
patients

Baseline

Functional changes

Neither reversible nor

arrestable

Unknown, to
be investigated

Unknown
(stage I
changes reversible)

Yes

Reversible by
strict insulin
treatment
(e.g., CSII)

Progression reduced, early

treatment advisable (aiming at
140/85-90 mm
Hg)

Probably, studies
are in progress

No hypertension
present. Microcirculatory
glomerular
alterations modifiable?

Arrestable or reversible by antihypertensive

treatment

- 25% of end-stage
renal failure patients
in USA are diabetics

Close blood pressui

control in all diabetics (along with
ophthalmoscopy)

Great effort should t

taken to arrest ne
phropathy at this
stage

Progress to clinical nephropathy in 3 0 40% of patients

May link to clinical nephropathy or modulate adversely the

course

Remarks

MOGENSEN, C. K. CHRISTENSEN, AND E. VITTINGHUS

pg/min

Ovnrt nephropathy

2500 pg/min
year
(wide range)

Incipient nephropathy
~

25jjg/mm
(wide range)
Antihypertensive
treatment?

12

15

18

21 years.DD

FIGURE 17. Outline of albumin excretion in the development of incipient and overt diabetic nephropathy.

by antihypertensive treatment or other measures including

optimizing diabetes control.
Great advances in the treatment of end-stage renal failure
in diabetic patients have been achieved over the last decade, and diabetic patients now account for approximately
25% of the population in the U. S. now entering these programs in the stage of uremia. Future emphasis should be
on preventative measures. One logical course is careful metabolic control throughout the lifespan of diabetic patients.
The portable subcutaneous insulin infusion devices present
a technologic and conceptual breakthrough, and with more
effective application these devices may greatly improve the
overall metabolic control.44 46 An additional promising avenue that may be comparable in importance is effective blood
pressure control. Both metabolism and hemodynamics, evidenced by blood pressure rise, should be controlled as effectively as possible.

REFERENCES

1
McCrary, R. F., Pitts, T. 0., and Puschett, J. B.: Diabetic nephropathy.
Natural course, survivorship, and therapy. Am. J. Nephrol. 1981; 1:206-18.
2
Deckert, T., and Poulsen, J E.: Diabetic nephropathy. Fault of destiny?
Diabetologia 1981; 21:178-83.
3
Mogensen, C. E.: Pathophysiology of diabetic complications. Abnormal physiological processes in kidney. In Handbook of Diabetes, Vol. 4: Biochemical Pathology. Brownlee, M., Ed. New York, London, Garland STPM
Press, 1981:23-85.
4
Viberti, G. C, Mackintosh, D., Bilous, R. W., Pickup, J. C, and Keen,
H.: Proteinuria in diabetes mellitus. Role of spontaneous and experimental
variation of glycemia. Kidney Int. 1982; 21:714-20.
5
Christiansen, J. S., Gammelgaard, J., Tronier, B., Svendsen, P. Aa.,

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983

and Parving, H.-H: Kidney function and size in diabetics before and during
intitial insulin treatment. Kidney Int. 1982; 21:683-88.
6
Cortes, P., Dumler, F., Venkatachalam, K. K., and Levin, N. W.: The
effect of diabetes on renal metabolism. In press. Mineral Electrolyte Metab.
1983.
7
0sterby, R.: Diabetic glomerulopathy: structural characteristics of the
early and advanced stages of diabetes mellitus. Diabetes (this issue).
8
Mogensen, C. E.: Introduction. Diabetes mellitus and the kidney. Kidney Int. 1982; 21:673-75.
9
Carstens, S. A., Hebert, L. A., Garancis, J. C, Piering, W. F., and
Lemann, J.: Rapidly progressive glomerulonephritis superimposed on diabetic glomerulosclerosis. Recognition and treatment. JAMA 1982; 247:145357.
10
Mogensen, C. E.: Antihypertensive treatment inhibiting the progression of diabetic nephropathy. Acta Endocrinol. (Copenh.) 1980; 94 (Suppl.
238):103-11.
11
Mogensen, C. E.: Preazotemic diabetic nephropathy. Inhibited by
antihypertensive treatment. In Diabetic Renal/Retinal Syndrome. Friedman, E.
A., and L'Esperance, F. A., Eds. New York, Grune and Stratton, 1980:18398.
12
Parving, H.-H., Andersen, A. R., Smidt, U. M., and Svendsen, P. Aa.:
Early and aggressive antihypertensive treatment reduces the rate of decline
in kidney function in diabetic nephropathy. In press. Lancet 1983.
13
Friedman, E. A.: Nephrology Forum. Diabetic nephropathy. Strategies in prevention and management. Kidney Int. 1982; 21:780-91.
14
Friedman, E. A., and L'Esperance, F. A.: Second Diabetic Renal/
Retinal Syndrome. New York, Grune and Stratton, 1982.
15
Hostetter, T. H., Rennke, H. G., and Brenner, B. M.: The case for
intrarenal hypertension in the initiation and progression of diabetic and other
glomerulopathies. Am. J. Med. 1982; 72:375-80.
16
Hostetter, T. H., Rennke, H. G., and Brenner, B. M.: Compensatory
renal hemodynamic injury. A final common pathway of residual nephron destruction. Am. J. Kidney Dis. 1982; 1:310-14.
17
Mogensen, C. E., Steffes, M. W., Deckert, T., and Christiansen, J. S.:
Functional and morphological renal manifestations in diabetes mellitus. Diabetologia 1981; 21:89-93.
18
Vittinghus, E., and Mogensen, C. E.: Graded exercise and protein
excretion in diabetic man and the effect of insulin treatment. Kidney Int. 1982;
21:725-29.
19
Brown, D. M., Andres, G. A., Hostetter, T. H , Mauer, S. M., Price,
R., and Venkatachalam, M. A.: Kidney complications. Diabetes 1982; 31 (Suppl
1)71-81.
20
Viberti, G. C , Mogensen, C. E., Keen, H., Jacobsen, F. K., Jarrett,
R. J., and Christensen, C. K.: Urinary excretion of albumin in normal man.
The effect of water loading. Scand. J. Clin. Lab. Invest. 1982; 42:147-51.
21
Brochner-Mortensen, J., and Ditzel, J.: Glomerular filtration rate and
extracellular fluid volume in insulin-dependent patients with diabetes mellitus.
Kidney Int. 1982; 21:696-98.
22
Mogensen, C. E.: Renal plasma flow in juvenile diabetes. Acta Endocrinol. (Copenh.) 1981; 94 (Suppl. 242):33-35.
23
Parving, H.-H., Christiansen, J. S., Noer, I., Tronier, B., and Mogensen, C. E.: The effect of glucagon infusion on kidney function in short-term
insulin-dependent juvenile diabetics. Diabetologia 1980; 19:350-54.
24
Christiansen, J. S., Gammelgaard, J., Frandsen, M., Orskov, H., and
Parving, H.-H.: Kidney function and size in type I (insulin-dependent) diabetic
patients before and during growth hormone adminstration for one week. Diabetologia 1982; 22:333-37.
25
Christiansen, J. S., Gammelgaard, J., Orskov, H., Andersen, A. R.,
Telmer, S., and Parving, H.-H.: Kidney function and size in normal subjects
before and during growth hormone administration for one week. Eur. J. Clin.
Invest. 1981; 11:487-90.
26
Weidman, P.: Recent pathogenic aspects in essential hypertension
and hypertension associated with diabetes mellitus. Klin. Wochenschr. 1980;
58:1071-89.
27
Nyberg, G., Granerus, G., and Aurell, M.: Renal extraction ratios for
51
Cr-EDTA, PAH, and glucose in early insulin-dependent diabetic patients.
Kidney Int. 1982; 21:706-708.
28
Mogensen, C. E., Soiling, K., and Vittinghus, E.: Studies on mechanisms of proteinuria using amino acid-induced inhibition of tubular reabsorption in normal and diabetic man. Contr. Nephrol. 1981; 26:50-65.
29
Mogensen, C. E., Christensen, C. K., Christensen, N. J., Gundersen,
H. J. G., Jacobsen, F. K., Pedersen, E. B., and Vittinghus, E.: Renal protein
handling in normal, hypertensive, and diabetic man. Contr. Nephrol. 1981;
24:139-52.
30
Mogensen, C. E., Christensen, C. K., and Vittinghus, E.: Changes in
renal function a n d blood pressure control in diabetes mellitus. With special
reference to exercise-induced changes in albumin excretion and blood pressure. In Second Diabetic Renal/Retinal Syndrome. Friedman, E. A., and L'Esperance, F. A., Eds. New York, Grune and Stratton, 1982:41-64.
31
Koivisto, V. A., Huttunen, N. P., and Vierikko, P.: Continuous subcutaneous insulin infusion corrects exercise-induced albuminuria in juvenile diabetes. Br. M e d . J . 1981; 2 8 2 : 7 7 8 - 7 9 .
32
Viberti, G., Pickup, J . C , Bilous, R. W., Keen, H., and Macintosh, D.:
Correction of exercise-induced microalbuminuria in insulin-dependent dia-

77

STAGES IN DIABETIC RENAL DISEASE

betics after three weeks of subcutaneous insulin infusion. Diabetes 1981;
30:818-23.
33
Vittinghus, E., and Mogensen, C. E.: Albumin excretion and renal
hemodynamic response to physical exercise in normal and diabetic man.
Scand. J. Clin. Lab. Invest. 1981; 41:627-32.
34
McMillan, D. E.: The functional and degenerative pathology of the
diabetic kidney. In Microcirculation. New York, Academic Press, 1981:22945.
35
Viberti, G. C , Mackintosh, D., and Keen, H.: Determinants of the
penetration of proteins through the glomerular barrier in insulin-dependent
diabetes mellitus. Diabetes (this issue).
36
Mogensen, C. E., Christensen, N. J., and Gundersen, H. J. G.: The
acute effect of insulin in heart rate, plasma noradrenaline, and urinary albumin
excretion. The role of changes in blood glucose. Diabetologia 1980; 18:45357.
37
Christensen, N. J., Gundersen, H. J. G., Mogensen, C. E., and Vittinghus, E.: Intravenous insulin decreases urinary albumin excretion in longterm diabetics with nephropathy. Diabetologia 1980; 18:285-88.
38
Viberti, G. C , Bilous, R. W, Macintosh, D., Bending, J. J , and Keen,
H.: Long-term correction of hyperglycaemia and progression of renal failure
in insulin-dependent diabetes. Br. Med. J. 1983; 286:598-602.
39
Kitzmiller, J. L, Brown, E. R., Phillippe, M., Stark, A. R., Acker, D.,

78

Kaldany, A., Singh, S., and Hare, J. W.: Diabetic nephropathy and perinatal
outcome. Am. J. Obstet. Gynecol. 1981; 141 741-51.
40
M o g e n s e n , C. E.: Long-term antihypertensive treatment inhibiting the
progression of diabetic nephropathy Br Med. J. 1982; 285:685-88.
41
Myers, B. D., Winetz, J. A., Chui, R, and Michaels, A. S.: Mechanisms
of proteinuria in diabetic nephropathy. A study of glomerular barrier function
Kidney Int. 1982; 21:633-41
42
Winetz, J . A., Golbetz, H. V , Spencer, R. J., Lee, J . A., a n d Myers,
B. D.: Glomerular function in advanced human diabetic nephropathy. Kidney
Int. 1982; 21:750-56.
43
Hostetter, T. H., Troy, J. L. and Brenner, B. M.: Glomerular hemodynamics in experimental diabetes mellitus Kidney Int. 1981; 19:410-15.
44
Keen, H., Pickup, J. C , Viberti, G. C, Bilous, R., and Williams, G.:
Aspects of continuous subcutaneous insulin infusion (CSII) in diabetes Diabetes Care 1981; 4:54-57
45
Lauritsen, T., Rrost-Larsen, K , Svendsen, P. Aa., Larsen, H.-W., Deckert, T., Christiansen, J. S., Parving, H.-H., Binder, C, Nerup, J., Deckert, M.,
Larsen, A., Lorup, B., Bojsen, J., and Beck Jensen, L.: Effect of six months
of strict metabolic control on eye and kidney function in insulin-dependent
diabetics with background retinopathy Lancet 1982; 1:121-24.
46
Keen, H., and Viberti, G. C Genesis and evolution of diabetic nephropathy. J. Clin. Pathol. 1981; 34:1261-66.

DIABETES, VOL. 32, SUPPL. 2, JUNE 1983