- Click here for explanation of classification scheme.

The purpose of this meta-analysis was to determine the efficacy of low protein diets in preventing the natural progression of chronic renal failure towards end-stage renal disease and therefore, delaying the need for starting maintenance dialysis.

1. Randomized controlled trials - random assignment to usual protein diet (0.8 g/kg) or moderate protein restriction (0.6 g/kg) or severe protein restriction (0.3 g/kg) regardless of supplementation with amino acids or ketoacids.

2. Cross-over studies in which starting intervention was randomly allocated.

3. Study period of >12 months.

4. Subjects with a diagnosis of moderate to severe CRF as estimated either by serum creatinine, creatinine clearance or GFR.

1. Studies including those with diabetes.

2. Studies including children

Recruitment

Study search was done using:

Medine and Embase between Jan 1966 and June 1999

Hand search of Congress abstracts (Am Soc of Nephrology since 1990, European Dialysis Transplant Assoc since 1985 and International Society of Nephrology since 1987).

Contacting authors of published work.

Design

Meta-analysis of randomized controlled trial of usual protein diet (0.8 g/kg) or moderate protein restriction (0.6 g/kg) or severe protein restriction (0.3 g/kg). Cross-over studies in which starting intervention was randomly allocated were also included.

Blinding used (if applicable)

No blinded follow-up of treatment

Intervention (if applicable)

Standard protein intake (0.8g/kg/day) or greater versus a moderate (0.6g/kg/day) to severe protein restriction (0.3g/kg/day) regardless of supplementation with essential amino acids or ketoacids.

Statistical Analysis

The number of patients originally allocated to each treatment group was noted and an “intention to treat” analysis was used.

Odds ratio, percentage difference, Peto, Mantel-Haensze were conducted.

Data was adjusted for a 1-yr period in order to compare treatment efficiency in trials of different durations.

Timing of Measurements

Trials were conducted from 1975 to 1999. Medline and Embase were searched from January 1966 through June 1999. Congress abstracts; American Society of Nephrology since 1990, European Dialysis Transplant Association since 1985, and International Society of Nephrology since 1987.

Dependent Variables

• Rate of GFR decline measured by plasma clearance of Cr-EDTA
• Rate of GFR decline measured by plasma clearance of iothalamate
• Serum creatinine
• Renal death including death from any cause
• Starting hemo- or peritoneal dialysis
• Receiving a kidney transplant

Independent Variables

• Standard protein intake (0.8g/kg/day) or greater versus a moderate (0.6g/kg/day) to severe protein restriction (0.3g/kg/day) regardless of supplementation with essential amino acids or ketoacids.

 

Initial N: 1,494 patients, 753 in the restricted protein groups and 741 in the unrestricted or higher protein intake groups.

Attrition (final N): See above

Age: Mean age of patients in the studies ranged from 44 to 62 years.

Ethnicity: Not mentioned

Other relevant demographics: Gender (Male/Female) was as follows: .54, .37, .54, .58, .67, .63, and .60 reflecting the higher male prevalence of renal disease.

Anthropometrics:

Location:  Worldwide studies

Other Findings:

The protein intakes for the 7 studies ranged from 0.4 to 0.8 g/kg/d with a mean intake of ~0.6 gm/kg/d.

The level of renal insufficiency was moderate or severe. There was no heterogeneity between studies (P=0.59). There was some heterogeneity between diets.

Based on urinary urea excretion, the actual reduction in dietary protein intake between groups in each study was less than expected and close to 0.2 g/kg/d, 0.25 g/kg/d, 0.3 g/kg/d and 0.35 g/kg/d.

 242 renal deaths were recorded, 101 in the low protein diet group and 141 in the higher protein diet group, giving an odds ratio of 0.61 with a 95% CI of 0.46 to 0.83 (P=0.006).

There was a 39% relative risk reduction in renal death in favor of a restricted protein intake.

Reduction in the Odds of Renal Death in Seven Prospective Randomized Studies of Protein Restriction
Study	Experimental (n/N)	Control (n/N)	Odds Ratio (95% CI fixed)
Jungers 1987	5/10	7/9	.29 [.04,2.11]
Malvy 1999	11/25	17/25	.37 [.12,1.17]
Williams 1991	12/33	11/32	1.09 [.39,3.02]
Ihle 1989	4/34	13/38	.26 [.07,.89]
Rosman 1989	30/130	34/117	.73 [.41,1.30]
Locatelli 1991	21/230	32/226	.61 [.34,1.09]
Klahr 1994	18/291	27/294	.65 [.35,1.21]
Total (95% CI)	101/753	141/741	.61 [.46,.83]

The fact that the actual patient protein intake was greater than prescribed in all studies suggests that regular dietitian services be proposed. Moreover, it has been demonstrated that patients with CRF without intervention by the dietitian will express a progressive decline in protein and energy intakes, potentially contributing to the decline in nutritional markers.

In contrast, feasibility of low protein diets has been shown in large studies with convincing results, highlighting the fact that interested teams can motivate patients to the point of excellent compliance and optimal nutritional benefit, a goal that should be reached for most patients in all renal units. Factors other than diet therapy have demonstrated a renal protective effect and to delay end-stage renal disease. These include angiotensin-converting enzyme inhibitors, blood pressure control, and optimal glucose monitoring in diabetic patients.

It cannot be derived from this meta-analysis that low protein diets reduce the progression of kidney disease since the decision to begin dialysis is often based on BUN and lower intakes of dietary protein would lower serum BUN, postponing dialysis.

University/Hospital:

Hopital Edouard Herriot, University of California

A limitation of some of the studies included in this meta-analysis was the use of serum creatinine rather than GFR as a marker for renal disease. Based on these studies, even a small reduction in dietary protein intake will postpone renal replacement therapy—dialysis or transplant.