Patients returning to haemodialysis or peritoneal dialysis after a failed kidney transplantation sometimes have a renal allograft left in situ for some urine production. Low-dose immunosuppressive medication is often continued in such patients. To evaluate the morbidity and mortality between patients in time periods with (group A) or without (group B) low-dose maintenance immunosuppression, the present study was initiated. In a multi-centre cohort study we analysed data from patient files, which showed failure after at least 3 months graft function between 10 August 1972 and 4 April 1996, including 197 kidney transplantations. A total of 1.7 versus 0.51 infections per patient year was found in groups A and B, respectively (odds ratio [OR]: 3.4, 95% confidence interval [CI]: 2.5-4.5). There was an increased mortality in group A compared to group B (OR 3.4, 95% CI: 1.8-6.3), both from infectious disease (OR 2.8, 95% CI: 1.1-7.0), and cardiovascular disease (OR 4.9, 95% CI: 1.8-13.5). Continuation of immunosuppressive medication did not lead to fewer rejections (defined as a painful, tender graft and/or haematuria and/or low-grade non-infectious fever). Transplantectomy-related morbidity and mortality were acceptable. The increase in morbidity and mortality associated with low-dose maintenance immunosuppression argues in favour of stopping these medicaments when failed renal allograft patients return to dialysis.