Impact of dialysis modality on morbimortality of kidney transplant recipients after allograft failure. Analysis in the presence of competing events.

Background And Objective: The number of patients who start dialysis due to graft failure increases every day. The best dialysis modality for this type of patient is not well defined and most patients are referred to HD. The objective of our study is to evaluate the impact of the dialysis modality on morbidity and mortality in transplant patients who start dialysis after graft failure.

Material And Methods: A multicentre retrospective observation and cohort study was performed to compare the evolution of patients who started dialysis after graft failure from January 2000 to December 2013. One group started on PD and the other on HD. The patients were followed until the change of dialysis technique, retransplantation or death. Anthropometric data, comorbidity, estimated glomerular filtration rate (eGFR) at start of dialysis, the presence of an optimal access for dialysis, the appearance of graft intolerance and retransplantation were analyzed. We studied the causes for the first 10 hospital admissions after starting dialysis. For the statistical analysis, the presence of competitive events that hindered the observation of the event of interest, death or hospital admission was analyzed.

Results: 175 patients were included, 86 in DP and 89 in HD. The patients who started PD were younger, had less comorbidity and started dialysis with lower eGFR than those on HD. The mean follow-up was 34 ± 33 months, with a median of 24 months (IQR 7-50 months), Patients on HD had longer follow-up than patients on PD (35 vs. 18 months, p = < 0.001). The mortality risk factors were age sHR 1.06 (95% CI: 1.03-1.106, p = 0.000), non-optimal use of access for dialysis sHR 3.00 (95% CI: 1.507-5.982, p = 0.028) and the dialysis modality sHR (PD/HD) 0.36 (95% CI: 0.148-0.890, p = 0.028). Patients on PD had a lower risk of hospital admission sHR [DP/HD] 0.52 (95% CI: 0.369-0.743, p = < 0.001) and less probability of developing graft intolerance HR 0.307 (95% CI 0.142-0.758, p = 0.009).

Conclusions: With the limitations of a retrospective and non-randomized study, it is the first time nationwide that PD shows in terms of survival to be better than HD during the first year and a half after the kidney graft failure. The presence of a non-optimal access for dialysis was an independent and modifiable risk factor for mortality. Early referral of patients to advanced chronic kidney disease units is essential for the patient to choose the technique that best suits their circumstances and to prepare an optimal access for the start of dialysis.