Peritoneal membrane function was assessed in patients undergoing continuous ambulatory peritoneal dialysis (CAPD) using parameters derived from urea kinetic modeling and the peritoneal equilibration test (PET). Their relationships with other nutritional markers and overall morbidity were determined. Data regarding the patients' nutritional status as determined by total body nitrogen (TBN) measurements, hospital admissions, and infectious complications within the last 12 months were reviewed. Total dialysate clearance (Kt/V) delivered was highly dependent on residual renal function (P < 0.0001). Kt/V derived from peritoneal clearance diminished with increasing age (P < 0.05). A higher delivered total Kt/V was associated with higher normalized protein catabolic rates (P < 0.002), which in turn were associated with improved TBN (P < 0.05). Hospital admissions decreased with improved normalized protein catabolic rates (P < 0.05), and higher serum albumin and total protein levels (P < 0.01 and P < 0.002, respectively). Infectious complications correlated positively with time on dialysis (P < 0.01), and correlated negatively with TBN measurements (P = 0.05). No correlations were found between infectious complications and serum albumin level or peritoneal protein loss. However, the total duration of hospitalization was shortened with higher serum albumin and total protein levels (P < 0.0001 and P < 0.002, respectively). Although Kt/V determinations did not correlate with clearances determined by the PET, the PET-determined creatinine transport rate correlated with TBN (P < 0.05) but not with infectious complications. In conclusion, nutritional parameters correlate with outcome on continuous ambulatory peritoneal dialysis. An integral relationship exists between nutritional status and dialysis delivery, which is best assessed by urea kinetic modeling.