Kidney International, Vol. 62 (2002), pp.

220228

Effect of dietary protein restriction on prognosis in patients

with diabetic nephropathy
HENRIK P. HANSEN, ELLIS TAUBER-LASSEN, BERIT R. JENSEN, and HANS-HENRIK PARVING
Steno Diabetes Center, Gentofte, Denmark

Effect of dietary protein restriction on prognosis in patients

with diabetic nephropathy.
Background. Recent data suggest that dietary protein restriction improves survival and delays the progression to endstage renal disease (ESRD) in non-diabetic nephropathies. The
purpose of our study was to determine the effect of dietary
protein restriction on survival and progression to ESRD in
diabetic nephropathy.
Methods. A four-year prospective, controlled trial with concealed randomization was performed comparing the effects of
a low-protein diet (0.6 g/kg/day) with a usual-protein diet. The
study included 82 type 1 diabetic patients with progressive
diabetic nephropathy [pre-study mean decline in glomerular
filtration rate (GFR) 7.1 mL/min/year (95% CI, 5.8 to 8.5)].
The main outcome measures were decline in GFR and development of ESRD or death.
Results. During the follow-up period the usual-protein diet
group consumed 1.02 g/kg/day (95% CI; 0.95 to 1.10) as compared with 0.89 (0.83 to 0.95) in the low-protein diet group
(P 0.005). The mean declines in GFR were 3.9 mL/min/year
(2.7 to 5.2) in the usual-protein diet group and 3.8 (2.8 to 4.8)
in the low-protein diet group. ESRD or death occurred in 27%
of patients on a usual-protein diet as compared with 10% on
a low-protein diet (log-rank test; P 0.042). The relative risk
of ESRD or death was 0.23 (0.07 to 0.72) for patients assigned
to a low-protein diet, after an adjustment at baseline for the
presence of cardiovascular disease (P 0.01). Blood pressure
and glycemic control were comparable in the two diet groups
during the follow-up period.
Conclusion. Moderate dietary protein restriction improves
prognosis in type 1 diabetic patients with progressive diabetic
nephropathy in addition to the beneficial effect of antihypertensive treatment.

dence of end-stage renal disease (ESRD) or death in patients with non-diabetic nephropathies and slows the progression of diabetic nephropathy [3]. However, the latter conclusion was based on 108 type 1 diabetic patients in
five studies (mean length of follow-up, 4.5 to 35 months)
[48] applying changes in urinary albumin excretion rate
or the decline in GFR or creatinine clearance as endpoints. Flaws in design, randomization procedure, patient selection, methods and end-points in addition to
the confounding impact of antihypertensive treatment
suggest that the above-mentioned conclusion in relation
to diabetes should be interpreted with caution [912].
We report the results of a four year, prospective controlled trial with concealed randomization, comparing
the effect of a low-protein diet with a usual-protein diet
in 82 type 1 diabetic patients with progressive diabetic
nephropathy. The study tested the hypothesis that a reduction in dietary protein intake retards the progression
to ESRD and improves survival in type 1 diabetic patients with diabetic nephropathy.
METHODS
The study was a prospective, randomized, unmasked,
controlled trial carried out at the Steno Diabetes Center.
With concealed randomization the patients were in blocks
of two according to the level of GFR (50, 50x75,
75x100, 100 mL/min/1.73 m2), assigned to receive
either a usual-protein diet or a low-protein diet. All
patients were recruited at the Steno Diabetes Center
and seen by the same physician (HPH) and dietitian
(ETL) at each visit. The local ethics committee approved
the study, and all patients gave written informed consent.

Dietary protein restriction slows the progression of

renal disease and improves survival in animals with varies glomerulopathies [1, 2]. Recently, a meta-analysis
suggested that dietary protein restriction lowers the inci-

Patients and randomization

The following inclusion criteria were used: Patients
between 18 and 60 years of age with type 1 diabetes
mellitus for at least 10 years, with an onset before the
age of 35 years, presence of diabetic retinopathy, albuminuria 300 mg/24 h in at least two out of three sterile
urine samples, and no clinical or laboratory evidence of

Key words: type 1 diabetes, survival, ESRD, GFR, progressive renal

disease, protein restriction, diabetic nephropathy.
Received for publication September 4, 2001
and in revised form January 31, 2002
Accepted for publication February 21, 2002

2002 by the International Society of Nephrology

220

Hansen et al: Protein restriction and diabetic nephropathy

other kidney or urinary tract disease [13], GFR above

20 mL/min/1.73 m2 and a pre-study decline in GFR 2
mL/min/year (progressive diabetic nephropathy). Since
patients continued their usual antihypertensive treatment, reduction in albuminuria even into the normal
range was observed in some of the patients at entry,
as previously demonstrated by us [14, 15]. Presence of
pregnancy, a history of congestive heart failure or myocardial infarction or coronary bypass surgery within the
last three months resulted in exclusion. Eighty-two type 1
diabetic patients who fulfilled the inclusion criteria consecutively entered the study between May 1995 and April
1996 (Fig. 1).
Study diet
After randomization an iso-caloric low-protein diet of
0.6 g/kg/day was prescribed to patients in the low-protein
diet group. The same dietitian (ETL) gave nutritional
advises, based on estimated protein intake, at least every
three months during the whole study period (see below).
All patients in the low-protein diet group received supplementation of calcium of 500 mg/day. The usual-protein diet consisted of the patients pre-study diet. In the
low-protein diet group (N 4), urine albumin losses 2
g/day were replaced by increasing dietary protein on a
gram-for-gram basis, only if the patient was compliant
with the prescribed low-protein diet.
Procedures and measurements
The planned duration of follow-up was four years in
each patient. At randomization and scheduled visits every three-month all patients gave a complete history of
medication, underwent examination of weight, urinary
albumin-, sodium- and urea excretion, serum albumin,
serum urea, hemoglobin, hemoglobin A1c (normal range;
4.1 to 6.4%), blood pressure, serum total-cholesterol and
high-density lipoprotein (HDL) cholesterol. GFR, serum
triglycerides, calcium and phosphorous, anthropometric
measurements, nutritional status and smoking habits
were evaluated every six months. Apart from the urine
collections, all the measurements were carried out with
the patients in the supine position. Chemical and biochemical analyses of serum and urine were performed
by standard laboratory techniques.
Dietary protein intake was estimated on the basis of
three consecutive 24-hour urine samples completed
before each visit, using the urinary excretion of urea
nitrogen (UUN) as follows [16]: estimated protein intake (g/day) 6.25 [UUN (g/day) standard body
weight (kg) 0.031 (g/kg/day)]. All consumed protein
is reported.
Before randomization, GFR was measured either by
the plasma clearance of 51Cr-EDTA (N 58) or the
formula of Cockroft and Gault (usual-protein diet group;
N 11, low-protein diet group; N 13) [17]. At random-

221

ization and during follow-up, GFR was measured on the

basis of the plasma clearance of an intravenous injection
of 51Cr-EDTA over a four-hour period starting at 8 am
[18]. The results were standardized for 1.73 m2 body surface area, using the same surface for each patient during
the study.
Blood pressure was measured on the right arm with
a Hawksley Random Zero Sphygmomanometer. The individual blood pressure level was determined as the
mean of at least two measurements. The mean blood
pressure was calculated {[(systolic blood pressure (2
diastolic blood pressure)]/3}.
Body mass index (BMI) was calculated as weight (kg)
per height (m)2. Retinopathy was assessed by fundus
photography and graded simplex or proliferative retinopathy. Present smokers were defined as persons smoking more than one cigarette/cigar/pipe per day.
A 12-lead electrocardiogram was recorded and subsequently coded using the Minnesota Rating Scale [19]
independently by two trained observers, who were masked
to the clinical status of the patients. Coronary heart disease was diagnosed if the electrocardiogram showed
signs of myocardial infarction or ischemia, or if patients
reported a history of myocardial infarction according to
the World Health Organization criteria [20] and validated by hospital records.
Outcome measures
The main outcome measures were the rate of decline
in GFR and the cumulative incidence of ESRD requiring
dialysis or transplantation, and death. Slopes were calculated on the basis of the baseline GFR and all follow-up
values. Before the development of uremic symptoms,
patients (N 6) were referred to the department of
nephrology at the University Hospital of Herlev or Rigshospitalet, when GFR deteriorated below 10 to 20 mL/
min/1.73 m2, where accepted criteria for initiation of dialysis and transplantation were applied on patients from
both diet groups. These patients continued their scheduled visits and treatment in the study until the end of the
four-year follow-up. Cause of death was obtained from the
death certificate. An independent observer without knowledge of randomization reviewed all death certificates and
the primary cause of death was recorded. Conditions
requiring withdrawal from the study included intercurrent illness (cancer, N 3; stroke, N 1) that precluded
the patients continued participation in the study.
Statistical analysis
Data for blood pressure, hemoglobin A1c and albuminuria during the period before randomization were averaged for each year for each patient and used to determine
a mean value for the entire period. Linear regression
analysis was used to estimate the rate of decline in GFR
for each patient before and during the present study

222

Hansen et al: Protein restriction and diabetic nephropathy

Table 1. Clinical characteristics of type 1 diabetic patients with diabetic nephropathy at the time of randomization

Variable
Sex female/male
Age years
Duration of diabetes years
Duration of diabetic nephropathy years
Insulin u/kg/24 h
Dietary protein intake g/kg/24 h
Standard body weight kg
Body mass index kg/m2
Systolic blood pressure mm Hg
Diastolic blood pressure mm Hg
Retinopathy simplex/proliferative
GFR mL/min/1.73 m2
Albuminuriaa mg/24 h
Urinary sodium excretion rate mmol/24 h
Cardiovascular events N (%)
Coronary heart disease
Stroke
I/D polymorphism in the ACE gene (%)
No. of antihypertensive drugs 0/1/2/3/3
Smoking N (%)
Hemoglobin A1c (%)
Blood hemoglobin mmol/L
Serum total-cholesterol mmol/L
Serum HDL-cholesterol mmol/L
Serum triglicerids mmol/L
Serum creatinine lmol/L
Serum urea mmol/L
Serum albumin g/L
Serum calcium mmol/L
Serum phosphorous mmol/L

Usual-protein diet
N 41

Low-protein diet
N 41

18/23
41 (9)
28 (8)
10 (5)
0.60 (0.16)
1.04 (0.25)
68 (9)
25 (3)
138 (18)
79 (10)
12/29
67 (32)
737 (1.2)
166 (58)

11/30
40 (8)
27 (7)
9 (5)
0.67 (0.18)
0.97 (0.26)
72 (8)
25 (4)
142 (17)
81 (9)
9/32
69 (30)
681 (1.2)
151 (52)

8 (20)
2 (5)
20/53/27
7/7/15/9/3
19 (46)
9.7 (1.4)
8.2 (1.1)
5.4 (1.0)
1.43 (0.45)
1.17 (0.62)
139 (62)
9.1 (4.6)
36 (4)
1.24 (0.05)
1.30 (0.18)

8 (20)
3 (7)
21/54/25
8/6/13/9/5
18 (44)
9.8 (1.6)
8.2 (1.3)
5.5 (1.4)
1.32 (0.49)
1.70 (2.03)
133 (48)
9.1 (4.2)
37 (3)
1.23 (0.05)
1.25 (0.23)

None of the variables differed significantly between diet groups. Abbreviations are: I/D, insertion/deletion; ACE, angiotensin-converting enzyme; GFR, glomerular
filtration rate. Data are mean (SD) unless specified.
a
Geometric mean (antilog SE)

using all the determinations of GFR. A minimum of one

year of follow-up with at least three measurements of
GFR was required for a patient to be included in the
slope analysis. Variables from the first visit after randomization and during the whole study were used to determine the mean value for each patient. Using an intentionto-treat approach, we related the rates of decline in
glomerular filtration, and cumulative incidence of ESRD
or death to the prescribed diet group. Kaplan-Meier
estimates of time to ESRD or death curves were compared by log-rank test. In the analysis of predictors of
ESRD or death, the Cox regression model was used. A
multiple regression analysis was performed with rate of
decline in GFR as the dependent variable and backward
selection including variables (dietary protein intake,
mean arterial blood pressure, hemoglobin A1c, albuminuria and total cholesterol) with P 0.10 in an univariate
analysis. Before the present study, we analyzed the rate
of decline in GFR in 106 type 1 diabetic patients with
progressive diabetic nephropathy. Applying the obtained standard deviation (3.2) for the rate of decline in
GFR (6.8 mL/min/year), in a sample size calculation we
found that 34 patients were necessary in each diet group
to detect a difference in the rate of decline in GFR of

2.5 mL/min/year between the two groups ( 0.05,

0.10). Data are presented as means (SD or 95% CI)
except when indicated. Owing to the skewed distribution
values for albuminuria was logarithmically transformed
before statistical analysis and given as geometric mean
(range or 95% CI). Paired and unpaired t tests were used
to compare the results within or between the two diet
groups. All calculations were made using SPSS for Windows (SPSS Inc., Chicago, IL, USA). P values less than
0.05 (two-sided) were considered to indicate significance.
RESULTS
Clinical characteristics of patients in the two diet
groups are shown in Table 1. No patients were lost to
follow-up (Fig. 1).
Dietary protein intake and nutritional status
After randomization and within the first three months
of the study, there was an initial decline in the dietary
protein intake of 0.06 g/kg/day (P 0.24) in the usualprotein diet group and 0.15 g/kg/day (P 0.01) in the
low-protein diet group (P 0.22 between groups; Fig. 2).
After three months and during follow-up, the patients

Hansen et al: Protein restriction and diabetic nephropathy

223

Fig. 1. The study profile.

Fig. 2. Estimated dietary protein intake in type 1 diabetic patients with progressive nephropathy. Symbols are: () usual-protein diet; () lowprotein diet. Values for the estimated dietary protein intake are mean and the 95% CI is indicated by the whiskers. The numbers of patients with
an estimated dietary protein intake at each visit are shown below the panel.

224

Hansen et al: Protein restriction and diabetic nephropathy

Fig. 3. Cumulative incidence of end-stage renal disease (ESRD) or death in type 1 diabetic patients with progressive diabetic nephropathy in the
usual-protein group (dashed lines) and the low-protein diet group (solid line). Log rank text, P 0.042. The numbers at the bottom denote the
number of patients in each group at risk for the event at baseline and after each six month period.

in the usual-protein diet group consumed a mean of 1.02

(0.95 to 1.10) g/kg/day, as compared with 0.89 (0.83 to
0.95) g/kg/day in the low-protein diet group (P 0.005,
Fig. 2). From baseline during follow-up the mid-arm
circumference, serum albumin and body weight were
comparable in the two diet groups (data not shown).
Dialysis, transplantation and death
At the end of the study six patients had required dialysis or renal transplantation and nine patients had died.
Causes of death were heart failure (N 4) and myocardial infarction (N 5). Dialysis, transplantation or death
occurred in 27% of patients (dialysis, N 3; transplantation, N 1; death, N 7) on the usual-protein diet, as
compared with 10% in patients (dialysis, N 2; death,
N 2) on the low-protein diet (log-rank test, P 0.042;
Fig. 3). The curves reflecting the cumulative incidence
of ESRD or death continued the separation during the
whole study period. A Cox regression analysis revealed
a relative risk of ESRD or death of 0.23 (0.07 to 0.72;
P 0.01) for the patients assigned to the low-protein
diet group, compared with those assigned to the usualprotein diet group, after adjustment for the presence at
baseline of cardiovascular disease.
Renal function
Before randomization the rate of decline in GFR in
the whole group of patients was 7.1 (CI, 5.8 to 8.5) mL/

min/year during a median follow-up of 4.9 (range, 1.0 to

11.2) years. Within the first six months of follow-up there
was a comparable significant decline in GFR of 4.1 mL/
min (P 0.01) in the usual-protein diet group and 4.4
mL/min (P 0.01) in the low-protein diet group (P
0.87 between groups). From randomization the rate of
decline in GFR slowed significantly in both diet groups
during the four years of follow-up (3.9 mL/min/year in
the usual-protein diet group vs. 3.8 in the low-protein diet
group), while urinary albumin excretion rate remained
unchanged (Table 2). The average improvement of the
rate of decline in GFR, comparing slopes before and
after randomization, were comparable in the two diet
groups; 2.7 (1.1 to 4.3) mL/min/year in the usual-protein
diet group and 3.7 (1.6 to 5.9) mL/min/year in the lowprotein diet group (P 0.44). Geometric mean of albuminuria was comparable between diet groups during the
follow-up (Table 2).
Glycemic control, blood pressure, serum lipids
and urea
Hemoglobin A1c and blood pressure values were comparable during follow-up in the two diet groups (Table 2).
Blood pressure was equally significantly reduced during
the study compared to the pre-study period in both diet
groups, while hemoglobin A1c was significantly reduced
in the low-protein diet group (Table 2). During follow-

Hansen et al: Protein restriction and diabetic nephropathy

Table 2. Rate of decline in GFR, albuminuria, blood pressure and
hemoglobin A1c before and during the study in type 1 diabetic
patients with diabetic nephropathy

Follow-up years
Before the study
During the study
Rate of decline in GFR
mL/min/year
Before the study
During the study
Albuminuria mg/24 h
Before the study
During the study
Blood pressure mm Hg
Before the study
Systolic
Diastolic
Mean
During the study
Systolic
Diastolic
Mean
Hemoglobin A1c %
Before the study
During the study

Usual-protein diet
N 34

Low-protein diet
N 38

5.1 (1.0 to 10.1)

4.0 (1.5 to 4.2)

5.0 (1.0 to 11.2)

4.0 (2.1 to 4.3)

6.6 (5.2 to 8.1)

3.9 (2.7 to 5.2)a

7.6 (4.9 to 10.2)

3.8 (2.8 to 4.8)a

721 (502 to 1036)

614 (389 to 969)

690 (547 to 871)

542 (382 to 769)

138 (133 to 144)

85 (82 to 87)
102 (100 to 105)

140 (136 to 144)

85 (83 to 88)
104 (101 to 106)

140 (135 to 146)

79 (76 to 81)b
99 (97 to 102)c

142 (138 to 146)

80 (78 to 83)b
101 (99 to 103)c

9.6 (9.2 to 9.9)

9.6 (9.3 to 10.0)

9.8 (9.4 to 10.1)

9.5 (9.1 to 9.9)c

Data are mean (95% CI). Except follow-up (median and range) and albuminuria (geometric mean and 95% CI), no significant differences were found between
diet groups. A minimum of 1 year of follow-up during the study with at least 3
measurements of GFR was required for a patient to be included in this analysis
(N 72).
a
P 0.005, bP 0.001, cP 0.05 compared to the period before the study

up, serum total-cholesterol [5.8 (5.5 to 6.1) mmol/L and

5.5 (5.2 to 5.8), respectively], high density lipoprotein
(HDL)-cholesterol [1.44 (1.31 to 1.58) mmol/L and 1.42
(1.26 to 1.58), respectively], triglycerides [1.47 (1.29 to
1.65) mmol/L and 1.59 (1.25 to 1.92), respectively] and
serum urea [11.8 (9.9 to 13.7) mmol/L and 10.9 (8.9 to
12.8), respectively] were comparable in the usual-protein
diet group and the low-protein diet group.
Medication and smoking
At baseline and during follow-up, a comparable number of patients in both diet groups received antihypertensive treatment, angiotensin-converting-enzyme inhibitors,
diuretics, and blockers, calcium channel blockers,
low-dose acetylsalicylic acid and lipid lowering agents,
respectively (Table 3). At baseline and at follow-up numbers of smokers were comparable in the two diet groups.
Numbers of smokers was reduced to 13 in the usualprotein diet group, and to 14 in the low-protein diet
group during follow-up.
Regression analyses for pooled data
For all patients with cardiovascular disease at baseline
(regardless of randomization) the relative risk of ESRD
or death were 14.2 (4.4 to 45.6; P 0.0001). Age and
sex were excluded from the model. Univariate regression
analyses revealed significant or borderline significant as-

225

sociations between the rate of decline in GFR and mean

values during follow-up of blood pressure (P 0.001),
albuminuria (P 0.001), hemoglobin A1c (P 0.03),
serum cholesterol (P 0.04) and dietary protein intake
(P 0.07). A multiple linear regression analysis using
these parameters as independent variables and the rate
of decline in GFR as the dependent variable, showed
that an increase in mean blood pressure (of 10 mm Hg),
hemoglobin A1c (of 1%) and albuminuria (10-fold) were
associated with a significant worsening in the rate of
decline in GFR of 1.2 (0.2 to 2.3), 0.71 (0.08 to 1.35) and
1.94 (0.40 to 3.48) mL/min/year, respectively.
DISCUSSION
The main finding from our prospective, randomized
controlled trial is that patients with type 1 diabetes suffering from progressive diabetic nephropathy experience a
beneficial effect of moderate restriction in dietary protein on the development of ESRD or death. The beneficial effect of protein restriction appeared within the first
year, and persisted with continued treatment, as also has
been demonstrated in non-diabetic nephropathies [21],
suggesting that type 1 diabetic patients with progressive
diabetic nephropathy are highly sensitive to dietary protein restriction. Despite the differences in event rates,
the decline in GFR in the two treatment groups did not
differ significantly, while both diet groups demonstrated
a progressive time-dependent reduction in the rate of
decline in GFR as compared to the period before randomization. Blood pressure, albuminuria, and glycemic
control were independent risk factors for the deterioration in GFR [15]. Statistical analyses were by intentionto-treat, that is, patients were included in their assigned
diet group, regardless of the achieved dietary protein
intake. A lack of adherence to the prescribed low-protein
diet by some patients would result in an underestimation
of the true beneficial effect of the diet. The achieved
level of long-term dietary protein restriction reflects real
life in an outpatient clinic set-up. Since diabetic patients
have other restrictions to the diet, this may reduce compliance to an additional low-protein diet. Although better compliance can be obtained by applying much more
intensive dietary counseling [4, 5, 22], it is not easy to
lower dietary protein intake to less than 0.8 g/kg/day
over extended periods of time [23], and patients with
advanced non-diabetic renal disease are only able to
lower their protein intake by 0.1 to 0.2 g/kg/day despite
intensive nutritional counseling [24]. Our study did not
reveal the mechanisms of the beneficial effect of moderate protein restriction, since neither the measured cardiovascular risk factors, nor the pharmacological treatment, or progression promoters in relation to kidney
disease showed any significant differences between the
two diet groups.

226

Hansen et al: Protein restriction and diabetic nephropathy

Table 3. Medication at baseline and follow-up in 82 type 1 diabetic patients with diabetic nephropathy randomized to
either a usual-protein diet or a low-protein diet
At baseline N (%)

Variable
Antihypertensive treatment
Angiotensin-converting enzyme inhibitors
Diuretics
- and -blockers
Calcium channel blockers
Low-dose acetylsalicylic acid
Lipid lowering agents (statins)

Usual-protein
34
29
26
7
8
7
4

(83)
(85)
(76)
(21)
(24)
(17)
(10)

At follow-up N (%)

Low-protein
33
29
28
9
10
5
3

(80)
(88)
(85)
(27)
(30)
(12)
(7)

Usual-protein
38
33
33
9
10
13
12

(93)
(87)
(87)
(24)
(26)
(32)
(29)

Low-protein
37
33
33
9
9
9
14

(90)
(89)
(89)
(24)
(24)
(22)
(34)

There were no significant differences in medication between diet groups at baseline or at follow-up.

The natural history of diabetic nephropathy is characterized by an early progressive rise in systemic blood
pressure associated with a relentless decline in GFR of
approximately 10 to 15 mL/min/year [2527]. Blood pressure elevation has proved to accelerate the progression
of diabetic nephropathy [28]. Conversely, effective longterm antihypertensive treatment reduces the rate of decline in GFR to approximately 5 mL/min/year [29], in
agreement with our present finding. The inverse correlation between duration of antihypertensive treatment and
rate of decline in GFR demonstrated in the present study
is in agreement with results previously obtained in diabetic and non-diabetic glomerulopathies, where a progressive time-dependent reduction in the rate of decline
was obtained during long-term aggressive antihypertensive treatment [14, 30, 31]. The mechanisms of this time
dependent reduction in the rate of decline in GFR in
diabetic and non-diabetic nephropathies are unknown.
However, animal models of different kidney diseases
suggest a shift in the balance between synthesis and degradation of extracellular mesangial matrix, preservation
of functioning (normal or only slightly damage) glomeruli, and even the possibility of new growth of glomerular
capillaries, as recently reviewed by Fogo [32]. It should
be recalled that the beneficial effect of improved glycemic control in The Diabetes Control and Complications Trial on the initiation and progression diabetic micro-angiopathy [33], or reversal of structural lesions in
diabetic glomerulopathy during normalization of glycemic control with pancreas transplantation [34], are delayed for several years. Importantly, the improvement
in rate of decline in the present study will reduce the
power to detect a difference between the two diet groups.
In the present study, control of blood pressure rather
than dietary protein restriction was of major importance
for the preservation of GFR during follow-up. Previous
experimental data have suggested that a low protein
diet, similar to treatment with an angiotensin-converting
enzyme inhibitor or an angiotensin II receptor blocker,
acts through a blockade of the renal renin-angiotensin
system [35]. However, the combination of a low protein

diet and maximal effective doses of either enalapril or

losartan have proved to reduce renal fibrosis slightly
more effectively than treatment with a low protein diet
or angiotensin II blockade alone [36], suggesting that
dietary protein restriction also acts on pathways independent of the renin-angiotensin system. Nevertheless, since
the majority of the renal protective effect of dietary
protein restriction seems to be mediated through the
renin-angiotensin system, simultaneous treatment with
an angiotensin-converting enzyme inhibitor, widely used
in the MDRD (Modification of Diet in Renal Disease)
study [22] and the present study, may have blunted the
additive beneficial effect of dietary protein restriction
on the decline in GFR in these two studies and albuminuria in the present study. When rate of decline in GFR
is slow, a time-to-event analysis has greater statistical
power to detect a beneficial effect of an intervention
than an analysis based on the slope of GFR [37]. The
REIN (Ramipril Efficacy in Nephropathy) and MDRD
studies support the validity of this analysis in non-diabetic nephropathies [21, 38].
It should be emphasized, however, that a valid determination of the rate of decline in GFR in patients with
chronic renal disease requires a reliable method for the
determination of GFR, repeated measurements of the
GFR, and a follow-up of at least two years [39]. These
requirements were fulfilled in our prospective study, except for one patient with a follow-up period of only 18
months. We have previously documented the validity of
the formula of Cockroft and Gault for the determination
of the rate of decline in GFR [40].
We performed a prospective, randomized controlled
trial as compared to the previous trials examining the
effect of low-protein diet in type 1 diabetic patients, which
were carried out as non-randomized, or partly randomized, and/or self-controlled studies [48]. Type 1 diabetic
patients with normo- and microalbuminuria were enrolled in one study [8], while only type 1 diabetic patients
with macroalbuminuria were enrolled in the remaining
studies. None of the previous studies required that patients also should have progressive diabetic nephropathy.

Hansen et al: Protein restriction and diabetic nephropathy

All previous studies examined small numbers of patients;

the largest study examined 35 subjects [5]. The main
outcome measures in our trial were time-to-event analysis and rate of decline in GFR. Several of the previous
studies evaluated changes in urinary albumin excretion
rate or short-term changes in creatinine clearance [6, 7],
the latter confounded by dietary protein intake [41].
Only Walker et al [4] and Zeller et al [5] performed
valid determinations of the rate of decline in GFR. The
reduction in rate of decline observed by Walker et al [4]
can partly be explained by the previously mentioned
phenomenon of a progressive time-dependent reduction
in the rate of decline in GFR as observed in the present
study and previous studies during antihypertensive treatment [14, 30]. Zeller et al found a rate of decline in GFR
of 3.1 mL/min/year during the low-protein diet (0.72
g/kg/day) [5], similar to what was found in both diet
groups in our study and previously reported during longterm aggressive antihypertensive treatment [13, 30],
while the rate of decline was 12.1 mL/min/year in the
control diet group (1.08 g/kg/day), comparable to what
has been demonstrated during the natural course of diabetic nephropathy (10 to 14 mL/min/year) in patients not
receiving antihypertensive treatment [2527]. However,
baseline proteinuria and mean blood pressure during
follow-up were higher in the control group (4.3 g/day
and 105.5 mm Hg, respectively) compared to the lowprotein diet group (3.1 g/day and 102.3 mm Hg, respectively), which may well have contributed to the accelerated decline in GFR in the control group.
In conclusion, moderate dietary protein restriction improves prognosis in type 1 diabetic patients with progressive diabetic nephropathy in addition to the beneficial
effect of antihypertensive treatment. Our study suggests
that a wider use of dietary protein restriction is indicated
in these patients.

3.
4.
5.
6.
7.
8.

9.
10.
11.
12.
13.

14.

15.
16.

17.
18.

19.

20.

ACKNOWLEDGMENTS
We are indebted to the patients who participated in the study, and
to the Danish Diabetes Association, the Danish Kidney Foundation,
the Poul and Erna Sehested Hansen Foundation, the Enid Ingemann
Foundation, and the Else and Mogens Wedel-Wedelsborg Foundation
for financial support of the study. The study has previously been published in abstract form [Diabetologia 44(Suppl 1), August 2001; J Am
Soc Nephrol 12:September 2001]. We acknowledge the assistance of
Peter Rossing, Bendix Carstensen, Birgitte V. Hansen, Ulla M. Smidt,
Inge-Lise Rossing, Annalise Klausen, and Annika Thierry Larsen.
Reprint requests to Henrik Post Hansen, M.D., Steno Diabetes Center, Niels Steensens Vej 2, 2820 Gentofte, Denmark.
E-mail: [email protected]

REFERENCES
1. Hostetter TH, Meyer TW, Rennke HG, Brenner BM: Chronic
effects of dietary protein in the rat with intact and reduced renal
mass. Kidney Int 30:509517, 1986
2. Nath KA, Kren SM, Hostetter TH: Dietary protein restric-

21.

22.

23.

24.

25.

26.

27.

227

tion in established renal injury in the rat. J Clin Invest 78:1199

1205, 1986
Pedrini MT, Levey AS, Lau J, et al: The effect of dietary protein
restriction on the progression of diabetic and nondiabetic renal
diseases: Meta-analysis. Ann Intern Med 124:627632, 1996
Walker JD, Bending JJ, Dodds RA, et al: Restriction of dietary
protein and progression of renal failure in diabetic nephropathy.
Lancet ii:14111415, 1989
Zeller KR, Whittaker E, Sullivan L, et al: Effect of restricting
dietary protein on the progression of renal failure in patients with
insulin-dependent diabetes mellitus. N Engl J Med 324:7884, 1991
Ciavarella A, Di Mizio G, Stefani S, et al: Reduced albuminuria
after dietary protein restriction in insulin-dependent diabetic patients with clinical nephropathy. Diabetes Care 10:407413, 1987
Barsotti G, Ciardella F, Morelli E, et al: Nutritional treatment
of renal failure in type 1 diabetic nephropathy. Clin Nephrol 29:
280287, 1988
Dullaart RP, Beusekamp BJ, Meijer S, et al: Long-term effects
of protein-restricted diet on albuminuria and renal function in
IDDM patients without clinical nephropathy and hypertension.
Diabetes Care 16:483492, 1993
Parving H-H: Low-protein diet and progression of renal disease
in diabetic nephropathy. (letter) Lancet 335:411, 1990
Parving H-H: Protein restriction and renal failure in diabetes
mellitus. N Engl J Med 324:17431744, 1991
Parving H-H: Effects of dietary protein on renal disease. Ann
Intern Med 126:330331, 1997
Shah N, Horwitz RI, Cancato J: Effects of dietary protein on
renal disease. Ann Intern Med 126:331, 1997
Parving H-H, Andersen AR, Smidt UM, Svendsen PA: Early
aggressive antihypertensive treatment reduces rate of decline in
kidney function in diabetic nephropathy. Lancet i:11751179, 1983
Parving H-H, Andersen AR, Smidt UM, et al: Effect of antihypertensive treatment on kidney function in diabetic nephropathy. BMJ
294:14431447, 1987
Hovind P, Rossing P, Tarnow L, et al: Progression of diabetic
nephropathy. Kidney Int 59:702709, 2001
Maroni BJ, Steinman TI, Mitch WE: A method for estimating
nitrogen intake of patients with chronic renal failure. Kidney Int
27:5865, 1985
Cockcroft DW, Gault MH: Prediction of creatinine clearance
from serum creatinine. Nephron 16:3141, 1976
Brochner-Mortensen J, Rodbro P: Selection of routine method
for determination of glomerular filtration rate in adult patients.
Scand J Clin Lab Invest 36:3545, 1976
Blackburn H, Keys A, Simonson E, et al: The electrocardiogram
in population studies: A classification system. Circulation 21:1160
1175, 1960
Rose GA, Blackburn H: Cardiovascular Survey Methods. Geneva,
WHO, 1968, pp 1188
Levey AS, Greene T, Beck GJ, et al: Dietary protein restriction
and the progression of chronic renal disease: What have all the
results of the MDRD study shown? J Am Soc Nephrol 10:2426
2439, 1999
Klahr S, Levey AS, Beck GJ, et al: For The Modification Of Diet
In Renal Disease Study Group: The effects of dietary protein
restriction and blood-pressure control on the progression of
chronic renal disease. N Engl J Med 330:877884, 1994
Locatelli F: Is the type of protein in the diet more important than
its quantity for slowing progression of chronic renal insufficiency?
Nephrol Dial Transplant 12:391393, 1997
Mehrotra R, Nolph KD: Treatment of advanced renal failure:
Low-protein diets or timely initiation of dialysis. Kidney Int 58:
13811388, 2000
Mogensen CE: Progression of nephropathy in long-term diabetics
with proteinuria and effect of initial antihypertensive treatment.
Scand J Clin Lab Invest 36:383388, 1976
Parving H-H, Smidt UM, Friisberg B, et al: A prospective study
of glomerular filtration rate and arterial blood pressure in insulindependent diabetics with diabetic nephropathy. Diabetologia 20:
457461, 1981
Viberti GC, Bilous RW, Mackintosh D, Keen H: Monitoring

228

28.
29.
30.
31.
32.
33.

34.
35.

Hansen et al: Protein restriction and diabetic nephropathy

glomerular function in diabetic nephropathy. Am J Med 74:256

264, 1983
Rossing P: Promotion, prediction, and prevention of progression
in diabetic nephropathy. Diabetic Med. 15:900919, 1998
Mogensen CE: How to protect the kidney in diabetic patients
With special reference to IDDM. Diabetes 46(Suppl):S104S111, 1997
Bjorck S, Nyberg G, Mulec H, et al: Beneficial effect of angiotensin converting enzyme inhibition on renal function in patients with
diabetic nephropathy. BMJ 293:471474, 1986
Ruggenenti P, Perna A, Gheradi G, et al: Renal function and
requirement for dialysis in chronic nephropathy patients on longterm ramipril: REIN follow-up trial. Lancet 352:12521256, 1998
Fogo AB: Progression and potential regression of glomerulosclerosis. Kidney Int 59:804819, 2001
The Diabetes Control and Complications Trial Research
Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulindependent diabetes mellitus. N Engl J Med 329:977986, 1993
Fioretto P, Steffes MW, Sutherland DER, et al: Reversal of
lesions of diabetic nephropathy after pancreas transplantation.
N Engl J Med 339:6975, 1998
Rosenberg ME, Smith LJ, Correa-Rotter R, Hostetter TH: The

36.
37.
38.
39.

40.
41.

paradox of the renin-angiotensin system in chronic renal disease.

Kidney Int 45:403410, 1994
Peters H, Border WA, Noble NA: Angiotensin II blockade and
low-protein diet produce additive therapeutic effects in experimental glomerulonephritis. Kidney Int 57:14931501, 2000
Greene T, Lau J, Levey AS: Interpretation of clinical studies of
renal disease, in Immunological Renal Diseases, edited by Neilson
EG, Couser WG 1997, pp 887911
Ruggenenti P, Perna A, Gherardi G, et al: Renoprotective properties of ACE-inhibition in non-diabetic nephropathies with nonnephrotic proteinuria. Lancet 354:359364, 1999
Levey AS, Gassman J, Hall PM, Walker WG: Assessing the
progression of renal disease in clinical studies: Effects of duration
of follow-up and regression to the mean. J Am Soc Nephrol 1:1087
1094, 1991
Rossing P, Astrup A-S, Smidt UM, Parving H-H: Monitoring kidney function in diabetic nephropathy. Diabetologia 37:708
712, 1994
Levey AS, Bosch JP, Coggins CH, et al: Effects of diet and antihypertensive therapy on creatinine clearance and serum creatinine
concentration in the modification of diet in renal disease study.
J Am Soc Nephrol 7:556565, 1996