Comparison of patient survival in non-diabetic transplant-listed patients initially treated with haemodialysis or peritoneal dialysis

Background. It is still not known whether patients survive longer on one modality of dialysis compared to the other. We have tried to answer this question using data from the Scottish Renal Registry.

Methods. To avoid the confounding effects of co-morbidity, we limited our survival analysis to those patients listed for a renal transplant and excluded patients with a primary renal diagnosis (PRD) of diabetic nephropathy. We studied patients starting dialysis between 01 January 1982 and 31 December 2006.

Results. Three thousand one hundred and ninety-seven patients fulfilled our criteria. A Kaplan-Meier plot showed no difference in survival between initial dialysis modality log-rank P = 0.996). In the Cox regression model, initial dialysis modality was not a significant predictor of survival; hazard ratio = 0.97 (95% CI 0.80 to 1.18) after adjusting for age, sex and PRD. Age at the start of dialysis, hazard ratio = 1.05 (95% CI 1.04 to 1.06) and a PRD group of 'multi-system disease' or 'unknown' were found to significantly influence survival. When survival was also censored for change in modality, there was no difference in survival over the whole study period with the hazard of death for patients on haemodialysis compared to those on peritoneal dialysis being 1.04 (95% CI 0.78 to 1.38; P = 0.803). Age at the start of dialysis remained a significant predictor of death.

Conclusions. This study shows that there was no survival advantage between initial dialysis modalities in nondiabetic patients who are deemed healthy enough for listing for a renal transplant.