Studies of the adequacy of peritoneal dialysis and recommendations have assumed that renal and peritoneal clearances are comparable and therefore additive. The CANUSA data were reanalyzed in an effort to address this assumption. Among the 680 patients in the original CANUSA study, 601 had all of the variables of interest for this report. Adequacy of dialysis was estimated from GFR (mean of renal urea and creatinine clearance) and from peritoneal creatinine clearance. The Cox proportional-hazards model was used to evaluate the time-dependent association of these independent variables with patient survival. For each 5 L/wk per 1.73 m(2) increment in GFR, there was a 12% decrease in the relative risk (RR) of death (RR, 0.88; 95% confidence interval [CI], 0.83 to 0.94) but no association with peritoneal creatinine clearance (RR, 1.00; 95% CI, 0.90 to 1.10). Estimates of fluid removal (24-h urine volume, net peritoneal ultrafiltration, and total fluid removal) then were added to the Cox model. For a 250-ml increment in urine volume, there was a 36% decrease in the RR of death (RR, 0.64; 95% CI, 0.51 to 0.80). The association of patient survival with GFR disappeared (RR, 0.99; 95% CI, 0.94 to 1.04). However, neither net peritoneal ultrafiltration nor total fluid removal was associated with patient survival. Although these results may be explained partly, statistically, by less variability in peritoneal clearance than in GFR, the latter seems to be physiologically more important than the former. The assumption of equivalence of peritoneal and renal clearances is not supported by these data. Recommendations for adequate peritoneal dialysis need to be reevaluated in light of these observations.