Concerns have been raised about the possibility of protein restriction resulting in malnutrition and poor subsequent survival on dialysis. However, no studies have examined patients treated with protein restriction to determine their subsequent survival on dialysis. This study prospectively monitored 67 patients with established chronic renal failure (mean initial serum creatinine of 4.3 mg/dL) who were treated with a very low-protein diet (0.3 g/kg per day) supplemented with either essential amino acids or a ketoacid-amino acid mixture and observed closely for clinical complications. Forty-four patients required dialysis. Once dialysis was started, dietary treatment was no longer prescribed. The cumulative mortality rate during the first 2 yr after starting dialysis was 7% (95% confidence interval, 0 to 16%). During this period, only two deaths occurred compared with 11.5 deaths expected on the basis of national mortality rates adjusted for age, sex, race, and cause of renal disease (P = 0.002). However, the protective effect was limited to the first 2 yr on dialysis. Thereafter, mortality rates increased, resulting in a total of 10 deaths during 96.4 person-years of follow-up, which was not significantly lower than the 14.9 deaths expected (P = 0.25). Extrapolation of sequential serum creatinine measurements made before dietary treatment suggests that the improved survival cannot be due to the early initiation of dialysis. Although the lack of an internal control group and data on dialysis lends uncertainty, the large difference in mortality rate between these patients and the nationwide experience indicates that protein restriction and close clinical monitoring predialysis does not worsen and may substantially improve survival during the first 2 yr on dialysis. These findings point out the importance of studying predialysis treatments as a means for lowering mortality on dialysis.