Patients on dialysis and recipients of renal transplants have higher mortality than individuals without kidney disease. In this study we evaluated the possible impact of dialysis therapy before transplantation on patient survival after the transplant. This analysis includes all of the patients who received a cadaveric renal transplant at The Ohio State University from 1984 to 1991 and who remained alive with functioning grafts for at least six months after the transplant (N = 523). After a follow-up of 84 +/- 14 months, 28% of the patients died and 23% lost their grafts. By multivariate analysis, reduced patient survival (censored at the time of graft loss) correlated with these pre-transplant variables: Older age (P < 0.0001), the presence of diabetes (P = 0.0002), smoking (P = 0.009), and the length of time on dialysis (P = 0.0002). Thus, 7% of patients who were never dialyzed, 23% of those dialyzed for less than three years, and 44% of patients dialyzed for > or = three years died post-transplant. By Cox regression, patient survival months correlated with time on dialysis pre-transplant (P = 0.0003). The type of dialysis (CAPD vs. hemodialysis) did not correlate with patient survival. Graft survival, censored for patient death, did not correlate with any of these pre-transplant variables. The relationship between time on dialysis and patient mortality is due to at least two factors: (1) transplant recipients who had dialysis for > or = 3 years had higher mortality due to infections (22%) than those who had dialysis for < 3 years (3%, P = 0.01 by X2); and (2) increasing time on dialysis increases the prevalence of both left ventricular hypertrophy (P = 0.008) and cardiomegaly (P = 0.004), and these relationships are statistically independent of other factors that also correlate with the prevalence of cardiovascular disease. In conclusion, increased time on dialysis prior to renal transplantation is associated with decreased survival of transplant recipients.