Effects of Dietary Protein Restriction on the Rate of Decline in Renal Function


Synopsis: A prudent approach in patients with chronic renal failure is the judicious implementation of the USRDA for protein, which is 0.8 g/kg/d-enough protein to maintain adequate nutrition.

Source: Kasiske BL, et al. Am J Kidney Dis 1998;31:954-961.

Dietary protein restriction (dpr) may delay the need for renal replacement therapy in patients with chronic renal failure. What remains to be established is if DPR actually reduces the rate of decline in renal function. Protein restriction ameliorates uremic symptoms and signs, and, thus, the need for and initiation of dialysis and transplantation may be delayed because of symptomatic improvement, rather than due to preservation of renal function. This meta-analysis addresses the focused question of the effect of DPR on the rate of decline in renal function in both randomized and non-randomized trials. The effect of demographic differences and underlying renal disease is also assessed.

There were a total of 13 randomized controlled trials in the literature, and the data were pooled. There were 1919 patients included in 13 such studies, with a mean follow-up of 21.8 months (range, 6-36 months). The total number of diabetics in the trials was only 102. The average dietary protein intake was 0.7 g/kg/d in the treatment group compared with 1-1.3 g/kg/d in the control group. The difference in protein intake between treatment and control groups was 0.3 g/kg/d in the non-diabetics and 0.5 g/kg/d in the diabetic patients. The pooled data demonstrated that DPR reduced the rate of decline in renal function, as measured by glomerular filtration rate (GFR), by only 0.5-0.7 mL/min/yr. Smaller studies (< 100 patients) more often reported positive results, suggesting a possible publication bias in favor of low protein diets.

The randomized and non-randomized trials, when analyzed together, comprised 24 studies and included 2248 patients (range, 12-840 patients in each study). Of these, 13 trials were randomized (as described above), and 11 studies used control data but were non-randomized. There were 129 diabetic patients altogether in six studies. The mean follow-up was 21.1 months. The mean dietary protein intake was 0.6 g/kg/d in the treatment group and 1 g/kg/d in the controls.

Overall, the benefit of DPR on decline in renal function was greater in non-randomized vs. randomized trials (regression coefficient, -5.2 mL/min/yr; P < 0.05), and in diabetics compared to non-diabetes (5.4 mL/min/yr; P < 0.05). There was a trend toward a greater effect of DPR with longer study duration and each additional year of follow-up (2.1 mL/min/yr), but this was not statistically significant. The other parameters-age, gender, body weight, other causes of renal disease (except diabetes), baseline renal function, mean arterial pressure, and baseline proteinuria-did not influence the effect of reduced protein intake on decline in GFR.


There are several cogent and pragmatic questions with regard to the effects of dietary protein restriction and renal failure.

    1. Does protein restriction delay the need for dialysis and transplantation? The answer is yes.

    2. Does protein restriction significantly reduce the rate of decline in renal function? The answer is no.

    3. Should protein intake be restricted in advanced chronic renal failure?

      a) What is the benefit?

      b) What is the risk?

      c) What should the protein intake be?

Protein restriction with GFR < 25 mL/min will result in improvement in symptomatology and a delay in the appearance of uremic symptoms. The risk of severe protein restriction is malnutrition, which is an independent risk factor for increased mortality in renal failure. A prudent approach is the judicious implementation of the USRDA for protein, which is 0.8 g/kg/d-enough protein to maintain adequate nutrition. This is a far cry from the over-indulgence which makes the protein intake about 1.4-1.5 g/kg/d in the average American diet, and is wedged between the protein-restricted and control groups in this meta-analysis.