Diabetic Nephropathy

Julia Breyer Lewis

Malluche et al. - Clinical Nephrolo gy, Dialysis and Transpla ntation - I-7

been estimated that patients with Type I DM

Key Points and nephropathy (manifested by proteinuria)
have a 100-fold greater risk of death relative
to a nondiabetic population. Patients with DM
Both alterations in systemic and glomerular without kidney disease have only a 2-fold
hemodynamics and changes in metabolism increase in relative mortality [2].
due to hyperglycemia are crucial factors in the
pathogenesis of diabetic nephropathy.
Microalbuminuria and proteinuria are key
markers which identify patients with diabetes
mellitus (DM) who have developed diabetic
Epidemiology
nephropathy. Intensive glycemic control dra-
matically reduces the risk of developing dia- The cumulative incidence of nephropathy
betic nephropathy. in patients with Type I DM of 40 years dura-

I.7
Blood pressure control slows the progres- tion is 40%. Several reports suggest that the
sion of diabetic nephropathy even after dete- cumulative incidence of diabetic nephropathy
rioration of renal function starts. Angiotensin- is declining, which may reflect better blood
converting enzyme (ACE) inhibition slows sugar control in the diabetic population as a
the progression of diabetic nephropathy and whole. The annual incidence peaks just before
is indicated for all normotensive or hyperten- 20 years duration of DM and declines there-
sive diabetic patients who have microalbu- after (Figure 1) [2]. The decline in the annual
minuria or proteinuria. incidence of diabetic nephropathy over time
Diabetic nephropathy is the most common
cause of end-stage renal disease (ESRD) in
the United States leading to enrollment in the
Medicare ESRD program. Patients with DM
currently account for 35% of the population
requiring renal replacement therapy. Patients
with Type II diabetes mellitus (DM) constitute
60% of the diabetic population in the ESRD
program. The cost to Medicare for these pa-
tients’ renal replacement therapy alone ex-
ceeds 2 billion dollars per year [1]. In addition
to its significant impact on health care costs,
Figure 1. Cumulative incidence of diabetic neph-
diabetic kidney disease significantly shortens ropathy in relation to duration of diabetes in 907
the lifespan of the patient with diabetes. It has Type I patients.

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 1

Chapter I - Clinical Nephrology and Hypertension

suggests that the pool of susceptible patients II DM already have diabetic nephropathy at
becomes exhausted. Diabetic nephropathy the time of diagnosis of DM. However, in the
with proteinuria rarely develops before 10 Pima Indian study, the age of onset of DM is
years duration of DM or after 30 years of DM. well documented and precedes proteinuria
Thus, the development of diabetic neph- from diabetic nephropathy by approximately
ropathy is not simply a function of duration of 20 years. Assuming that the natural history of
DM nor an inevitable complication in all pa- diabetic nephropathy in Pima Indians is rep-
tients. Rather, the data suggest that only a resentative of all patients with Type II DM, the
subset of 40% of patients with Type I DM duration of DM required to develop neph-
develop diabetic nephropathy, perhaps be- ropathy may be similar for Type I and Type II
cause of a genetic susceptibility or an environ- DM.
mental exposure. Finally, there are important racial differ-
Reports of familial clustering of diabetic ences that contribute to the risk of developing
nephropathy are consistent with a genetic sus- diabetic nephropathy. African Americans,
ceptibility to the development of this compli- Mexican Americans, American Indians,
cation. In families with multiple siblings who Maori, and Polynesians with DM have a
have Type I DM, those of a proband with greater risk of developing diabetic neph-
diabetic nephropathy are more likely to have ropathy and a more rapid progression to
proteinuria and renal insufficiency secondary ESRD than Caucasians [3]. It is not known
to diabetic nephropathy than the siblings of whether these racial differences are the result
duration-matched pro-bands without neph- of genetic, environmental, or other factors.
ropathy. Similar data supporting genetic sus-
ceptibility to diabetic nephropathy have also
been reported in families with Type II DM.
Early studies suggested that the cumulative
incidence of diabetic nephropathy in patients Pathogenesis and Risk
with Type II DM was less than in patients with Factors
Type I DM. However, a recent careful natural
history study done in Pima Indians with Type
II DM reported a cumulative incidence of Many factors have been postulated to be
diabetic nephropathy comparable to that important in the pathogenesis of diabetic
found in Type I DM patients [3]. Furthermore, nephropathy. It appears that exposure to both
in Western Europe it has been reported that up the systemic hemodynamic effects of DM and
to 40% of patients with Type II DM developed to the diabetic milieu are critical factors for
nephropathy, a cumulative incidence quite the development of diabetic kidney disease. In
similar to that found in Type I DM [4]. In part, studies with diabetic rats and humans, unilat-
some of the discrepancies between these re- eral renal artery stenosis has been reported to
ports may reflect the very high mortality rate protect the kidney distal to the blocked renal
from atherosclerotic disease in patients with artery from developing the morphologic
Type II DM and renal disease, resulting in a changes of DM. Thus, exposure to the sys-
bias against the progression to ESRD. temic blood pressure appears to contribute to
For most Type II patients, the age of onset the development of diabetic kidney disease.
of DM is unknown and may precede the diag- Normal kidneys transplanted into patients
nosis by many years. Some patients with Type with DM develop diabetic lesions over time,

2
 7 Breyer Lewis - Diabetic Nephropathy

whereas kidneys inadvertently harvested disease. Together, these studies imply that
from patients with DM transplanted into non- alterations in the renin-angiotensin system
diabetic ESRD patients showed resolution of play an important role in the pathogenesis of
their diabetic lesions on microscopy. This diabetic nephropathy. Additional important
suggests that exposure to the diabetic milieu, factors in this process may be alternations in
rather than an intrinsic defect in the kidneys the production of prostaglandins or kinins.
predisposes these patients to the development Identifying the risk factors for development
of diabetic nephropathy. of diabetic nephropathy in humans may help
Hyperglycemia, insulin insufficiency, aug- identify important pathogenic mechanisms in
mented glucagon and growth hormone levels, the development of this disease [8]. Signifi-
and increased ketogenesis have all been im- cant risk factors include a family history of
plicated in the pathogenesis of diabetic neph- diabetic nephropathy, the presence of hyper-
ropathy. More recently, it has been demon- tension, poor glycemic control, smoking, and
strated that hyperglycemia results in the gly- increased plasma pro-renin activity. In large
cosylation of long-lived proteins that population studies, patients with DM have
crosslink to form advanced glycosylation end significantly elevated blood pressures com-
products (AGEs) [5]. In animal models, the pared with non-diabetic controls. The pres-
accumulation of AGEs is associated with the ence of proteinuria in these patients is predic-
end-organ complications of DM. Similarly, in tive of the presence of hypertension. Some,
humans the accumulation of AGEs is in- but not all, studies suggest that at the time of
creased in patients with renal insufficiency. diagnosis of DM, patients destined to develop
Inhibitors of AGE formation are beneficial in diabetic nephropathy have significantly

I.7
retarding the progression of diabetic neph- higher mean arterial blood pressures than
ropathy in experimental animals and are cur- those patients with DM who never develop
rently being tested in human clinical trials. nephropathy. Furthermore, patients with both
Hyperglycemia also leads to the shunting of Type I and Type II DM and nephropathy have
glucose down the polyol pathway with the a predisposition to hypertension, as indicated
accumulation of sorbitol in the lens, nerves, by elevated sodium-lithium counter transport
and kidney. In animal models, this pathway activity in erythrocytes and a strong parental
has been implicated in the development of history of hypertension as compared to dia-
diabetic nephropathy, but it is unclear what betic patients without nephropathy. Com-
role it may play in human diabetic kidney bined, these data make a compelling argue-
disease [6]. ment that a genetic predisposition to hyper-
Alterations in the renin-angiotensin system tension is a risk factor for developing diabetic
also contribute to the development of diabetic nephropathy. However, since it has been re-
nephropathy in animal and human models, as ported that up to 25% of patients with Type I
do abnormal intraglomerular pressures in ani- DM and established diabetic nephropathy are
mal models. In the diabetic rat, decreasing normotensive, a genetic susceptibility to hy-
intraglomerular pressures by decreasing pertension cannot be the only risk factor in the
angiotensin (Ang) II-mediated efferent arteri- development of diabetic nephropathy.
ole resistance preserves glomerular structure Hyperglycemia is clearly necessary for the
and function [7]. Similarly, in humans, inter- development of diabetic nephropathy in ani-
rupting the renin-angiotensin system (see be- mals and humans. Many retrospective clinical
low) slows the progression of diabetic kidney studies have documented a relationship be-

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 3

Chapter I - Clinical Nephrology and Hypertension

tween poor blood sugar control and the sub-

sequent development of diabetic nephropathy.
Patients with poor blood sugar control de-
velop diabetic nephropathy earlier. Poor gly-
cemic control also synergistically increases
the risk for diabetic nephropathy in patients
genetically predisposed to hypertension.
Smoking is a risk factor identified in multi-
ple epidemiological studies for the develop-
ment and more rapid progression of diabetic
Figure 2. Five distinctive lesions in glomeruli of
nephropathy. The mechanism for the effects
patients with insulin-dependent diabetes mellitus.
of smoking on kidney disease is unclear. In- A = Kimmelstiel-Wilson nodule; B = intercapillary
creased plasma pro-renin activity has also glomerulosclerosis;C = thickening of glomerular
basement membrane; D = mesangial expansion; E
been associated with an increased risk of pa-
= fibrin cap lesion.
tients developing diabetic retinopathy and
nephropathy, although there is considerable
overlap in the plasma pro-renin activity be-
tween patients with and without complica- sion. Although DM is primarily a vascular
tions. (including glomerular) disease, alterations in
the structure and function of the renal tubu-
lointerstitium are also present. The degree of
interstitial fibrosis also correlates with the
reduction in GFR. Many patients will have
morphologic changes in the kidney consistent
Pathology with the effects of DM but never develop
proteinuria or other clinical manifestations of
diabetic nephropathy.
Renal biopsies are normal in patients at the
time of diagnosis of Type I DM. Within as The morphologic changes found in patients
little as 1 – 2 years, however, morphologic with Type II DM and diabetic nephropathy
changes appear, including glomerular base- were long thought to be indistinguishable
ment membrane thickening, mesangial ex- from those in patients with Type I DM. Care-
pansion, nodular and diffuse forms of intra- ful prospective biopsy studies have demon-
capillary glomerulosclerosis, the capsular strated that patients with Type I and Type II
drop lesion, and the fibrin cap and arteriolar DM share the nodular form of intracapillary
hyalinosis (Figure 2). Glomerular basement glomerulosclerosis (Kimmelstiel-Wilson-le-
membrane thickening is a sensitive indicator sions). However, patients with Type II DM
for the presence of DM, but does not predict have more prominent non-nodular glomeru-
which patients will develop clinically signifi- losclerosis, arteriolar hyalinosis, and other
cant nephropathy. In contrast, mesangial ex- vascular changes.
pansion has been demonstrated to correlate
with clinically significant diabetic neph-
ropathy [9]. Glomerular filtration rate (GFR),
as measured by creatinine clearance, declines
linearly with the degree of mesangial expan-

4
 7 Breyer Lewis - Diabetic Nephropathy

Figure 3. The natural history

of diabetic nephropathy.

above, studies of the Pima Indians indicate

Natural History analogous progression in Type II DM com-
pared to Type I [3, 4].
The natural history of diabetic nephropathy
is best understood in patients with Type I DM
and is illustrated in Figure 3. Functional Changes
At the time of diagnosis of Type I DM,
functional changes in the kidney are mani- At the onset of DM, kidney size is increased
fested in virtually all patients [7]. Within a few and albuminuria is present, but these changes

I.7
years, morphologic changes occur in the kid- usually reverse with blood sugar control. The
neys of most patients. Glomerular hyperfiltra- iniation of insulin therapy also lowers the
tion, elevations in the systemic mean arterial dramatically increased GFR, although it re-
blood pressure, and poor blood sugar control mains supranormal compared to a nondiabetic
all appear to be important features in this early population. In both retrospective and prospec-
period. Nephropathy is manifested early by tive studies, higher GFRs at the time of diag-
microalbuminuria, with the majority of pa- nosis of DM predicted the later development
tients progressing to overt proteinuria. On of diabetic nephropathy [10]. A GFR of 125
average, proteinuria is present in the 40% of mL/min or less had a negative predictive value
patients who were destined to develop dia- of 95% in Type I patients, while a GFR > 125
betic nephropathy 10 – 25 years after the onset mL/min had a 53% positive predictive value
of Type I DM. Once proteinuria is established, for the development of nephropathy. How-
renal function declines inexorably, with 50% ever, because of overlap in these patient popu-
of patients reaching ESRD within 7 – 10 lations, an early measurement of GFR alone
years. cannot solely predict diabetic renal disease. In
Patients with Type II DM frequently have patients with Type II DM, reports vary widely
an unknown age of onset of DM and many about the presence or absence of glomerular
co-existing illnesses complicating their hyperfiltration. This may reflect the difficulty
course. For example, Type II diabetics have a in establishing the length of DM in these
high cardiovascular mortality and often do not patients. In the Pima Indians, early glomerular
survive long enough to develop ESRD from hyperfiltration similar to that reported in pa-
diabetic nephropathy. However, as noted tients with Type I DM has been documented.

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 5

Chapter I - Clinical Nephrology and Hypertension

In animal models of DM, glomerular hyper- tients, many of whom do not survive to de-
filtration is associated with increments in re- velop proteinuria or ESRD.
nal flow because of renal dilatation and in-
creases in the glomerular transcapillary hy-
draulic pressure gradient. These changes lead
to accelerated glomerulosclerosis. It is likely Proteinuria and Declining Renal
that similar glomerular damage associated Function
with glomerular hyperfiltration occurs in hu-
mans. Glomerular hyperfiltration has also For the vast majority of patients with dia-
been associated with increased glomerular betic nephropathy, proteinuria inevitably pro-
surface area and renal hypertrophy. gresses to renal decline. In patients with either
Type I or Type II DM, the proteinuria typically
ranges from 0.5 gm to more than 20 gm/24
hour. In early nephropathy, albuminuria is
Microalbuminuria secondary to a loss of the anionic charge bar-
rier, whereas in established nephropathy,
The presence of microalbuminuria (albu- proteinuria results from an increased number
min excretion of 30 – 300 mg/24 hour) in of nonselective enlarged pores. Proteinuria
patients with DM is highly predictive of the typically occurs 17 years (± 6 years) after the
progression to proteinuria and renal insuffi- onset of Type I DM.
ciency in the next 10 – 15 years. The reported Once proteinuria is established, GFR be-
prevalence of microalbuminuria in patients gins to decline, with an average reported rate
with Type I DM varies widely, depending, at of loss between 3 – 12 mL/min/yr. The decline
least in part, on how the population is selected in renal function can be hastened by other
in terms of DM duration. Overall, approxi- diabetic complications such as neurogenic
mately 20% of patients with Type I DM will bladder, urinary tract infections, papillary ne-
have microalbuminuria in cross-sectional crosis, and exposure to renal toxins such as IV
studies. Poor glycemic control and elevations contrast. Other factors that predict a faster
in mean arterial blood pressures also correlate progression include uncontrolled systemic
with the development of microalbuminuria hypertension, uncontrolled blood sugars,
[11]. Once present, microalbuminuria is higher protein excretion rates, hyper-
highly predictive of progression to overt pro- cholesterolemia, smoking, and a parental his-
teinuria and renal insufficiency. Between 75% tory of DM. Certain racial groups, such as
and 100% of patients with documented mi- African Americans, Mexican Americans, Na-
croalbuminuria go on to proteinuria and de- tive Americans, Polynesians, and Maori, all
clining renal function. The GFR in patients have a faster rate of progression to ESRD than
with microalbuminuria is well preserved or Caucasians with diabetic nephropathy [4].
increased, but may begin to decrease once the
urinary albumin excretion rate exceeds 100
µg/min. Type II patients also develop microal-
buminuria, which also predicts progression to
proteinuria and ESRD. However, microalbu-
minuria is associated with a high mortality
rate due to atherosclerotic events in these pa-

6
 7 Breyer Lewis - Diabetic Nephropathy

disease in diabetic patients with renal disease

End-stage Renal Disease far exceeds that in duration-matched patients
(ESRD) with DM without nephropathy. Other morbid-
ity stems from an excessive rate of cerebral
and peripheral vascular complications. Low
Once patients with DM reach ESRD, their serum albumin and infection also complicate
options include hemodialysis, peritoneal di- renal replacement therapy for patients with
alysis, or transplantation. ESRD patients with DM. Thus, despite advances malnutrition, in-
DM have higher mortality rates than nondia- fection, and atherosclerotic diseases shorten
betic ESRD patients. Peritoneal dialysis the life of the patient with DM and ESRD.
avoids the problems associated with hemodia-
lysis, such as vascular access, hemodynamic
instability, and systemic heparin, but is com-
plicated by increased glucose loads, worsen-
ing triglyceride levels, and peritonitis. Clinical Manifestations
Hemodialysis provides higher small molecu-
lar weight substance clearances. Some studies
and Diagnosis
suggest that higher clearances decrease mor-
tality, particularly in patients with DM and Microalbuminuria appears to be an impor-
ESRD, although no prospective studies have tant marker for the development of proteinuria
yet directly compared these two modalities. and declining renal function. Current recom-
Survival is longer for diabetc patients re- mendations are that all patients with Type I

I.7
ceiving a living related donor transplant com- DM of greater than 5 years duration should be
pared to patients remaining on dialysis. Two- screened yearly for microalbuminuria. It is
year renal graft survival and patient survival important to specifically order testing for uri-
in Type I DM with living related donor trans- nary albumin excretion or microalbuminuria,
plants are comparable to survival in nondia- because of the low sensivity of routine urine
betic ESRD patients. However, Type I DM protein determinations for this condition [10].
patients who receive a cadaveric renal trans- Urinary albumin excretion rates exceed
plant have a higher morbidity and mortality 300 mg/day in patients with proteinuria meas-
than nondiabetic ESRD patients. Patients re- ured either by dipstick or by the traditional
ceiving combined renal and pancreas trans- 24-hour urine protein assay. With sensitive
plants have a higher morbidity rate than those assays for microalbuminuria, urinary albumin
undergoing renal transplant alone, but have excretions < 30 mg/24 hour can be detected.
improved hyperglycemia, and both motor and Patients with normal kidneys have urinary
sensory nerve function. No benefits are de- albumin excretion rates < 30 mg/24 hour or
monstrable for pancreatic polypeptide secre- 20 µg/min, while microalbuminuria is
tion, preservation of kidney function, or reti- defined as albumin excretion rates of
nopathy. 30 – 300 mg/24 hour or 20 – 200 µg/min.
The increased morbidity and mortality rates The presence of microalbuminuria can be
observed in ESRD patients with DM are evaluated in a 24-hour urine collection, an
mainly secondary to complications of overnight urine collection that can be extrapo-
atherosclerotic disease. Many investigators lated to 24 hours, or a spot urine measurement
have shown that the risk for cardiovascular of the albumin-creatinine ratio (Table 1).

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 7

Chapter I - Clinical Nephrology and Hypertension

betic retinopathy. Therefore, the absence of

Table 1. Measurement of Microalbuminuria
diabetic retinopathy as determined by either 7
field fundus photos or fluorescein angiogra-
– Test Type I DM patients of greater than 5 phy should again suggest the presence of an
years duration every year.
– Test Type II DM patients at the time of
alternative diagnosis. Only 60 – 65% of pa-
diagnosis and every year. tients with Type II DM and diabetic neph-
– Rule out causes of transient microalbu- ropathy have diabetic retinopathy, making it a
minuria: hyperglycemia, UTI, PE, essential
HTN, CHF, water loading.
less helpful factor in their evaluation. All pa-
– If the albumin excretion rate is elevated, tients with DM and proteinuria should be
repeat 3 times over 3 to 6 months to define evaluated carefully for the presence of other
persistent microalbuminuria.
– Normal albumin excretion: <30 mg/24 hour; systemic diseases that can cause proteinuria,
microalbuminuria: 30 – 300 mg/24 hour; such as a gammopathy, hepatitis B, systemic
proteinuria: >300 mg/24 hour. lupus erythematosus (SLE) or amyloidosis.
Abnormalities in the urinalysis, e.g. hema-
UTI: urinary tract infection; PE: physical exercise; turia or red blood cell casts, should also alert
HTN: hypertension; CHF: congestive heart failure. one to the possibility of another kidney dis-
ease. A renal ultrasound will rule out anatomic
abnormalities and evaluate the patient for kid-
ney size. Most patients with diabetic neph-
ropathy have large kidneys, especially in rela-
Transient causes of microalbuminuria such as tion to their decreased GFR.
hyperglycemia, urinary tract infection (UTI), In a prospective biopsy series in patients
physical exercise, uncontrolled hypertension, with Type I DM and more recently Type II
congestive heart failure, or water loading DM, ≤ 10% of the patients who had the typi-
should be considered. Thus, an elevated albu- cal clinical features noted above were found
min excretion rate should be repeated on 3 to have a diagnosis by biopsy other than dia-
occasions over 3 – 6 months to confirm per- betic nephropathy. Of those patients with an
sistent microalbuminuria. Patients with Type alternate diagnosis, even fewer had a diagno-
II DM should be screened at the time of diag- sis identified for which there was a specific
nosis and yearly thereafter [12]. Identifying therapy indicated. Thus, in the vast majority
patients with microalbuminuria allows meas- of patients with DM and proteinuria, a renal
ures to be implemented to slow or halt the biopsy is not necessary. If, however, the pa-
progression of their kidney disease much ear- tient has any atypical clinical features as out-
lier, before significant renal insufficiency de- lined above, a renal biopsy should be per-
velops. formed.
Proteinuria in a diabetic patient warrants There are unique aspects to the manage-
evaluation. In patients with Type I DM, the ment of the patient with DM and progressive
onset of proteinuria occurs between 10 and 25 renal insufficiency. Insulin requirements de-
years after diagnosis. Thus, patients who first crease in uremia, both because the kidney is
present with proteinuria before 10 years or responsible for 30 – 40% of the metabolism
after 25 years duration of DM should be sus- of insulin and the loss of appetite as renal
pected of having another cause of kidney dis- function declines. Thus, the majority of pa-
ease. In patients with Type I DM, 90 – 95% of tients on insulin will have decreased insulin
patients with diabetic nephropathy have dia- requirements, and many Type II patients will

8
 7 Breyer Lewis - Diabetic Nephropathy

have decreased oral hypoglycemic require- therapy at higher GFR levels than do patients
ments as ESRD nears. Further, many oral with nondiabetic kidney disease. Choosing
hypoglycemic agents are excreted by the kid- the best renal replacement therapy for diabetic
ney, and their half-life is prolonged in patients patients is complex and requires the consid-
with renal insufficiency. Thus, it is important eration of many factors, including their body
to counsel both Type I and Type II patients habitus, vasculature, and extent of
about hypoglycemia and to monitor their oral atherosclerotic disease and heart function.
hypoglycemic or insulin use. Metformin is Lastly, patients with DM have a high preva-
contraindicated in patients with serum creat- lence of cardiovascular disease. Thus, trans-
inines greater than 1.5 mg/dL and cannot be plant evaluation in these patients requires
used in patients with significant renal insuffi- careful cardiac testing. Many of these patients
ciency. with documented cardiovascular disease also
DM is the most common cause of Type IV have peripheral vascular disease, which can
renal tubular acidosis, which results in hy- complicate their course. For all these reasons,
perkalemia and a hyperchloremic metabolic dialysis teaching and transplant evaluation
acidosis. Thus, patients with this tubular should begin early in patients with diabetic
transport defect can have an electrolyte pat- nephropathy.
tern that is disproportionately abnormal for
the degree of renal insufficiency.
Patients who develop nephropathy fre-
quently have gastroparesis resulting from dia-
betic neuropathy. Symptoms of gastroparesis

I.7
such as nausea and vomiting can mimic those Therapeutic Interventions
of uremia.
It is important to evaluate patients with DM
frequently to document the onset of diabetic The major therapeutic interventions that
nephropathy to allow optimal management have been evaluated for patients with diabetic
and preventative measures. Avoiding exces- nephropathy include antihypertensive ther-
sive use of nonsteroidal anti-inflammatory apy, treatment with ACE inhibitors, improved
agents, the unnecessary use of aminogly- DM control, inhibitors of AGEs formation,
coside antibiotics, and exposure to IV radio- the restriction of dietary protein intake, and
contrast agents is important. Proper interven- treatment of dyslipidemia, as well as a variety
tions for UTI and neurogenic bladder can also of less well-studied interventions (Table 2).
help preserve remaining renal function.
Patients with DM appear to require renal
replacement therapy earlier than patients with Blood Pressure Control
other renal diseases. They have lower GFRs
for any given serum creatinine than nondia- A close correlation exists between the onset
betic patients with renal insufficiency, most and degree of microalbuminuria and the onset
likely related to lower muscle mass. It is not and degree of hypertension. Similarly, the
uncommon for diabetics to be near ESRD presence of hypertension in patients with pro-
with serum creatinines as low as 2 – 3 mg/dL. teinuria is associated with a more rapid de-
Diabetic patients have symptoms of uremia cline in GFR. Multiple studies have shown
and require the initiation of renal replacement that reducing the systemic blood pressure in

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 9

Chapter I - Clinical Nephrology and Hypertension

the Collaborative Study Group randomized

Table 2. Therapeutic Interventions in Diabetic
Nephropathy
patients with Type I DM and proteinuria to 2
levels of blood pressure control (a mean arte-
rial blood pressure < 92 mmHg or to a mean
Proven Benefit Possible Benefit
arterial blood pressure of 100 – 107 mmHg)
to better define the optimal blood pressure for
Blood sugar control Low protein diet preservation of kidney function. All patients
Blood pressure control Treatment of
ACE inhibition dyslipidemia
in this study received an ACE inhibitor (Rami-
Smoking cessation pril). Preliminary analysis of these results
Prevention of AGEs suggests that the lower blood pressure goal
formation
conferred greater benefit. Indeed, many of the
patients with the lower blood pressure did not
have demonstrable decline in renal function,
but rather a regression in their degree of pro-
teinuria, in some cases to near normal ranges.
diabetic patients with proteinuria and declin-
ing renal function slows the rate of decline of Thus, available data suggest that the blood
renal function and improves survival. These pressure goals for protecting the kidney
studies primarily used conventional antihy- against progressive diabetic nephropathy may
pertensive agents, which at the time did not be lowered to mean arterial blood pressures <
include calcium channel blockers or ACE in- 92 mmHg.
hibitors. Although the studies were performed In addition a reduction in urinary protein or
with relatively few numbers of highly selected albumin excretion might be an appropriate
patients and achieved only modest blood pres- end point of antihypertensive therapy inde-
sure control, a clearly decreased rate of renal pendent of its effect on systemic blood pres-
decline was observed. Similarly, in studies of sure. Several major studies in nondiabetic kid-
patients with microalbuminuria, reducing the ney disease demonstrate that an early reduc-
mean arterial blood pressure resulted in de- tion in urinary protein or albumin excretion
creases in urinary albumin excretion rates. predicts a long-term beneficial effect of the
Even in normotensive patients with microal- intervention. Thus, it has been suggested that
buminuria, the lowering of systemic blood antihypertensive therapy should be advanced
pressure was associated with decreased uri- as tolerated to help reduce urinary protein
nary albumin excretion. excretion.
The definition of adequate blood pressure
control in patients with diabetic nephropathy
has been unclear. The Fifth Joint National ACE Inhibition
Commission defines high normal blood pres-
sure as 140/90 mmHg, and physicians have Data from experiments in diabetic rats first
traditionally used this as a blood pressure goal suggested that specific antihypertensive
for patients with hypertension. A number of agents (ACE inhibitors) could preserve renal
epidemiological studies indicate that the pro- function and structure independent of their
gression of renal disease correlates closely effect on systemic blood pressure. In these
with increased systemic blood pressure. This studies, glomerulosclerosis was only partially
effect was linear, with no lower threshold for mitigated by the use of conventional antihy-
the benefits of lower blood pressure. Recently, pertensive agents, including hydralazine, re-

10
 7 Breyer Lewis - Diabetic Nephropathy

serpine and hydrochlorothiazide. However, Treatment in patients with early diabetic

with ACE inhibitors, albuminuria and nephropathy characterized by microalbu-
glomerulosclerosis were ameliorated. The minuria has also been demonstrated to have a
mechanism of protection was postulated to be beneficial effect [14]. Studies in patients with
the reduction of the tonic constrictor effect of both Type I and Type II DM demonstrated that
Ang II on the efferent arteriole, leading to the use of ACE inhibitors leads to a decrease
lower glomerular intracapillary pressures. in the urinary albumin excretion with far
A number of studies were then conducted fewer patients progressing to overt prote-
in humans with diabetic nephropathy. A meta- inuria. The consistent finding of a beneficial
analysis of antihypertensive therapy in pa- effect of ACE inhibitors in patients with mi-
tients with diabetic nephropathy, showed only croalbuminuria strongly suggests that these
ACE inhibitors decreased proteinuria and pre- patients should be considered for such therapy
served GFR independent of changes in sys- to preserve their remaining renal function.
temic blood pressure. Added benefits of ACE inhibitors are im-
proved insulin sensitivity, and plasma lipid
A recently completed clinical trial of ACE
inhibitors randomized 409 patients with Type profile.
I DM and proteinuria to receive either Capto- Far fewer studies have been performed ex-
pril three times per day or placebo [13]. Pa- amining the efficacy of calcium channel
tients enrolled in this trial had urinary protein blockers in patients with diabetic nephropathy
excretion rates > 500 mg/day and serum creat- [15]. These studies were performed in small
inine concentrations ≤ to 2.5 mg/dL [13]. The numbers of patients, most of whom had Type
primary outcome of the trial was the time to II DM. Recently, it has been suggested that a

I.7
doubling of serum creatinine (to ≥ 2 mg/dL), combination of ACE inhibitors and calcium
representing a halving of the GFR. The use of channel blockers may be of additional benefit.
Captopril led to a risk reduction of 48% for Further investigation is needed to clarify the
doubling of creatinine compared to the pla- role of calcium channel blockers in the treat-
cebo group. Captopril was equally effective in ment of patients with DM and diabetic neph-
reducing the risk of doubling of serum creat- ropathy.
inine in normo- and hypertensive patients and
in African American and Caucasian patients.
Its efficacy was not explained by differences Improved Blood Sugar Control
in baseline urinary protein excretion or fol-
low-up mean arterial blood pressure. Impor- Intensive blood sugar control prevents the
tantly, patients receiving ACE inhibitors had development of diabetic nephropathy and
a risk reduction of 50% in the combined clini- ameliorates established diabetic nephropathy
cal outcome of either death or ESRD. Again, in animal studies. In humans, the Stockholm
this effect of ACE inhibitors was independent Diabetes Intervention Study and the Diabetes
of any effect on systemic blood pressure. Control and Complications Trial have now
There were few adverse effects reported, with conclusively demonstrated that intensive
no ARF and only 6 hyperkalemic events. blood sugar control in Type I diabetics can
Thus, this study convincingly demonstrated delay the development or slow the progres-
that ACE inhibition is renoprotective in both sion of diabetic nephropathy [16]. Patients in
normo- and hypertensive patients with Type I these trials achieved intensive blood sugar
DM and clinically evident proteinuria. control (HgbA1C ∼7.0%) with insulin deliv-

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 11

Chapter I - Clinical Nephrology and Hypertension

ered either by an insulin pump or by 3 or more gan complications of DM [5]. Aminoguani-

injections daily. These patients had a dramatic dine is a nucleophilic hydrazine compound
decrease in the risk of developing microalbu- that prevents the formation of AGEs and glu-
minuria or progressing from microalbu- cose-derived collagen crosslinks in vitro and
minuria to proteinuria compared to the pa- in vivo. Its use in diabetic rats decreases albu-
tients in the conventional therapy group who minuria and glomerulosclerosis. Clinical tri-
achieved hemoglobin A1Cs of approximately als using aminoguanidine in patients with
9%. Intensive blood sugar control in these Type I and Type II DM and established neph-
studies not only dramatically reduced the risk ropathy are ongoing. This drug is currently
of nephropathy, but also the risk of reti- available only in research settings.
nopathy and neuropathy. The chief adverse
event associated with intensive therapy was a
2 – 3-fold increase in severe hypoglycemic Dietary Protein Restriction
episodes requiring assistance. These hypogly-
cemic episodes occurred in the intensive ther- In many experimental animal models of
apy group despite careful study management renal disease, including DM, high dietary pro-
and support measures beyond what most clin- tein intake accelerates the deterioration of re-
ics could provide. Secondary analysis of these nal function. Low-protein diets improve
studies demonstrated reduced nephropathy glomerular hemodynamics in animal models
risk for any reduction in hemoglobin A1C by vasoconstriction of the afferent arteriole,
levels [17]. This reflects the absence of a resulting in decreased glomerular intracapil-
threshold effect, indicating that any improve- lary pressures. In humans with DM, low pro-
ment in blood sugar control achievable in the tein diets have been demonstrated to reduce
routine clinical setting would reduce the risk glomerular hyperfiltration, decrease urinary
of developing diabetic nephropathy. Thus, all albumin excretion, and slow the rate of de-
patients with Type I DM should achieve the cline of renal function. Although, individual
tightest blood sugar control that can be safely studies in patients with DM have enrolled
achieved in their clinical setting. In Type II small numbers of patients, a recent meta-
diabetics, however, it has not yet been conclu- analysis supported the efficacy of low protein
sively demonstrated that glycemic control diets in patients with diabetic nephropathy
confers a similar advantage. One relatively [18].
small study done in Japan has indicated a
beneficial effect of tighter blood sugar control
in the Type II population. Tight blood sugar
Lipid-lowering Agents
control in this population is very difficult to
achieve and may have more associated risks. Dyslipidemia is common in the diabetic
state and worsened by coexistent diabetic
nephropathy. Patients with poor glycemic
Prevention of the Formation of control usually have hypertriglyceridemia.
AGEs The presence of diabetic nephropathy is asso-
ciated with higher levels of plasma low-den-
It has been demonstrated that AGEs accu- sity lipoprotein (LDL) cholesterol and lipo-
mulate in tissues of diabetic patients and may, protein B (Lp(b)), and lower levels of high-
at least in part, be responsible for the end-or- density lipoprotein (HDL) cholesterol. In

12
 7 Breyer Lewis - Diabetic Nephropathy

many animal models of kidney disease, shown to decrease urinary albumin excretion
dyslipidemia is implicated in causing direct and renal hypertrophy in rats with streptozo-
renal injury and hastening the progression of tocin-induced DM. Although the potential
established renal diseases. Treatment of mechanism of the beneficial effect is un-
dyslipidemia in these animals leads to re- known, it may be mediated by inhibition of
duced glomerular injury. In humans with dia- insulin-like growth factor-1.
betic nephropathy, hyperlipidemia has been Lastly, DM has been characterized by an
identified as a risk factor for a more rapid rate increase in plasma levels of thromboxane B2.
of decline in GFR and increased mortality. Thromboxane synthetase inhibitors signifi-
Recently, several small uncontrolled prelimi- cantly lower urinary protein excretion in dia-
nary studies in diabetic patients with prote- betic rats. In a recent randomized, double-
inuric renal disease showed that treatment masked, placebo-controlled trial in 30 pa-
with β-hydroxy-β-methyl glutaryl-COA tients with DM and microalbuminuria, pico-
(HMG-CoA) reductase inhibitors can lead to tamide (a dual antithromboxane ag-coent that
a stabilization and improvement of renal func- inhibits thromboxane synthetase and blocks
tion [19]. Because dyslipidemia is closely re- the thromboxane receptor) significantly low-
lated to the progression of cardiovascular dis- ered urinary albumin excretion compared to
ease, and highly prevalent in diabetics, lipid- placebo.
lowering agents are recommended, irrespec-
tive of their potential effect on diabetic neph-
ropathy.
References

I.7
Miscellaneous Interventions
[1] Excerpts from United States Renal Data System
1997 Annual Data Report 1997 Incidence and
Smoking is an important risk factor for the prevalence of ESRD. Am J Kidney Dis 30: S40-
development and progression of both diabetic S53
nephropathy and atherosclerotic disease. [2] Andersen AR, Christiansen JS, Andersen NK et al
1983 Diabetic nephropathy in type-I (insulin-de-
Smoking cessation is an important part of the
pendent)-diabetes: An epidemiological study. Dia-
management of all diabetics. betologia 25: 496-501
A variety of other pharmaceutical interven- [3] Nelson RG, Newman JM, Knowler WC, Sievers
tions have been applied to a small number of ML, Kunzelman CL, Pettitt DJ, Moffett CD, Teutsch
SM, Bennett PH 1988 Incidence of end-stage renal
patients in preliminary studies or in animal disease in type-II (non insulin-dependent)-diabetes
models. Pentoxifyline decreased microalbu- mellitus in Pima Indians. Diabetologia 31: 730-
minuria and proteinuria in Type I patients in a 736
placebo-controlled trial. This drug induces [4] Ritz E, Stefanski A 1996 Diabetic nephropathy in
type-II-diabetes: An in-depth review. Am J Kidney
membrane changes that increase erythrocyte
Dis 27 : 167-194
deformability, reducing blood viscosity. Simi-
[5] Brownlee M 1991 Glycosylation products as toxic
larly, dipyramadole has been shown to de- mediators of diabetic complications. Annu Rev
crease urinary albumin excretions in Type I Med 42: 159-166
patients with microalbuminuria. Octreotide, a [6] Heesom AE, Hibberd M, Millward A, Demaine AG
somatostatin analogue, reduced GFR and kid- 1997 Polymorphism in the 5’ -end of the aldose
reductase gene is strongly associated with the de-
ney size in 11 patients with Type I DM and velopment of diabetic nephropathy in Type I dia-
glomerular hyperfiltration. It has also been betes. Diabetes 46: 287-291

Malluche et al. - Clinical Nephrology, Dialysis and Transplantation - I-7 13

Chapter I - Clinical Nephrology and Hypertension

[7] Breyer J 1992 Diabetic nephropathy in insulin-de- [17] The Diabetes Control and Complications Trial
pendent patients An in-depth review. Am J Kidney (DCCT) Research Group 1995 The absence of a
Dis 20: 533-547 glycemic threshold for the development of long-
[8] Gall MA, Hougaard P, Borch-Johnsen K, Parving term complications. The perspective of the diabe-
HH 1997 Risk factors for development of incipient tes control and complications trial. Diabetes 45:
and overt diabetic nephropathy in patients with 1289-1298
non-insulin dependent diabetes mellitus: Prospec- [18] Pedrini MT, Levey AS, Lau J, Chalmers TC, Wang
tive, observational study. Brit Med J 314: 783-788 PH 1996 The effect of dietary protein restriction
[9] Steffes MW, Osterby R, Chavers B et al 1989 Per- on the progression of diabetic and nondiabetic
spectives in diabetes-mesangial expansion as a renal diseases: A meta-analysis [see comments].
central mechanism for loss of kidney function in Ann Intern Med 124: 627-632
diabetic patients. Diabetes 38: 1077-1081 [19] Lam KSL, Cheng IKP, Janus ED, Pang RWC 1995
[10] Rudberg S, Persson B, Dahlquist G 1992 Increased Cholesterol-lowering therapy may retard the pro-
glomerular filtration rate as a predictor of diabetic gression of diabetic nephropathy. Diabetologia 38:
nephropathy-an 8-year prospective study. Kidney 604-609
Int 41: 822-838
[11] Bennett PH, Huffner S, Kasicke RI et al 1995
Screening and management of microalbuminuria
in patients with diabetes mellitus: Recommenda-
tions to the Scientific Advisory Board of the Na-
tional Kidney Foundation from an ad hoc commit-
Key References
tee of the Council on Diabetes Mellitus of the
National Kidney Foundation. Am J Kidney Dis 25:
107-112 [**] Ritz E, Stefanski A 1996 Diabetic nephropathy in
[12] Theochari MA, Vyssoulis GP, Toutouzas PK, Bart- type-II-diabetes: An in-depth review. Am J Kidney
socas CS 1996 Arterial blood pressure changes in Dis 27: 167-194
children and adolescents with insulin-dependent Comprehensive review of available information on
diabetes mellitus. J Pediatr 129 (5): 667-670 the epidemiology, natural history and efficacy of
therapeutic interventions in patients with Type II DM
[13]+ Lewis EJ, Hunsicker LG, Bain RP, Rohde RD for
and diabetic nephropathy
the Collaborative Study Group 1993 The effect of
angiotensin-converting-enzyme inhibition on dia-
betic nephropathy. N Engl J Med 329: 1456-1462 [**] The Diabetes Control and Complications Trial
[14] Ravid M, Savin H, Jutrin I, Bental T, Katz B, (DCCT) Research Group 1995 Effect of intensive
Lishner M 1993 Long-term stabilizing effect of therapy on the development and progression of
angiotensin-converting enzyme inhibition on diabetic nephropathy in the diabetes control and
plasma creatinine and on proteinuria in normoten- complications trial. Kidney Int 47: 1703-1721
sive type-II-diabetic patients. Ann Intern Med 118: This clinical trial is a conclusive demonstration that
577-581 intensive glycemic control decreases the risk of
[15] Rossing P, Tarnow L, Boelskifte S, Jensen BR, developing diabetic nephropathy in patients with
Nielsen FS, Parving HH 1997 Differences between Type I DM
nisoldipine and lisinopril on glomerular filtration
rates And albuminuria in hypertensive IDDM pa- [+] Lewis EJ, Hunsicker LG, Bain RP, Rohde RD for
tients with diabetic nephropathy during the first the Collaborative Study Group 1993 The effect of
year of treatment. Diabetes 46: 481-481 angiotensin-converting enzyme inhibition on dia-
[16]**The Diabetes Control and Complications Trial betic nephropathy. N Engl J Med 329: 1456-1462
(DCCT) Research Group 1996 Effect of intensive The clinical trial reported in this paper demonstrates
therapy on the development and progression of the dramatic efficacy of ACE inhibitors in slowing the
diabetic nephropathy in the diabetes control and progression of renal insufficiency in patients with
complications trial. Kidney Int 47: 1703-1721 Type I DM and diabetic nephropathy.

14