Examination of survival after transfer from peritoneal dialysis to hemodialysis.

Data on survival after transfer from peritoneal dialysis (PD) to hemodialysis (HD) is conflicting. We reviewed two decades of outcomes in a PD program to examine short-term survival after transfer from PD to HD. Of 379 patients on PD, 33% transferred to HD. The reasons for transfer were PD-related infections (340%), uremia or failure to thrive (26%), PD catheter problems or loss of mechanical skills (15%), dementia or unable to train (7%), noncompliant with PD (7%), other (10%, including gastrointestinal complications, hernia, encapsulating peritoneal sclerosis, preference, loss of ultrafiltration), and cardiac (2%). All of those transferring for "other" reasons survived 6 months, and as did all except 1 who transferred for uremia (p = 0.035). Overall survival was 92% at 3 months and 85% at 6 months. Using multivariate logistic regression analysis, only score on the Charlson comorbidity index at PD start was a risk factor for dying in the first 6 months on HD: for each 1 point increase in CCI score, the hazard ratio for death was 1.4 (95% confidence interval: 1.16 to 1.74; p = 0.005). To summarize, starting a patient on PD and waiting until uremia to transfer to HD does not have a negative impact on survival. In a program with relatively low PD-related infectious complications, such complications accounted for only one third of transfers to HD.