BACKGROUND: A growing number of patients are returning to dialysis after renal transplant failure. The aim of this study is to determine whether peritoneal dialysis (PD) is a safe and good treatment option for these patients. METHODS: All patients returning to PD or haemodialysis (HD) after renal transplant failure before 1 October 2002 at the University Hospital Gasthuisberg, Leuven, Belgium, were evaluated. Data were collected until death, retransplantation (reTx), transfer to HD or PD or until 1 January 2003. RESULTS: Twenty-one patients starting PD (PDpostTx-group) and 39 patients starting HD (HDpostTx-group) after renal transplant failure were included in the study. There were no significant differences in age, sex, serum albumin- and CRP-levels at baseline. The total time on renal replacement therapy at transplant failure and time to transplant failure did not differ between the two groups either. Furthermore, the baseline comorbidity was similar in both groups. During follow-up, the outcome did not differ significantly between the two groups. However, there was a tendency towards higher patient survival and reTx tended to be more frequent in the PDpostTx-group. Moreover, patients in the HDpostTx-group tended to accrue more new comorbidity. The incidence of peritonitis and the evolution of dialysis adequacy (renal and peritoneal Kt/V and creatinine clearances) with time in the PDpostTx-group was similar to that seen in our centre's PD patients who had never undergone transplantation before. CONCLUSIONS: This study suggests that the outcome in patients starting PD after renal transplant failure is at least as good as the outcome in those starting HD. Although these observational findings warrant further confirmation, PD therefore can be regarded as a safe and good treatment option for patients returning to dialysis after renal transplant failure.