The superior results of living-donor renal transplantation are not completely caused by selection or short cold ischemia time: a single-center, multivariate analysis.

Background. The results of living-donor (LD) renal transplantations are better than those of postmortem-donor (PMD) transplantations. To investigate whether this can be explained by a more favorable patient selection procedure in the LD population, we performed a Cox proportional hazards analysis including variables with a known influence on graft survival. Methods. All patients who underwent transplantations between January 1981 and July 2000 were included in the analysis (n = 1,124, 2.6% missing values). There were 243 LD transplantations (including 30 unrelated) and 881 PMD transplantations. The other variables included were the following: donor and recipient age and gender, recipient original disease, race, current smoking habit, cardiovascular disease, body weight, peak and current panel reactive antibody, number of preceding transplants and type and duration of renal replacement therapy, and time since failure of native kidneys. In addition, the number of human leukocyte antigen identical combinations, first and second warm and cold ischemia periods, left or right kidney and fossa, donor kidney anatomy, donor serum creatinine and proteinuria, and transplantation year were included. Results. In a multivariate model, donor origin (PMD vs. LD) significantly influenced the graft failure risk censored for death independently of any of the other risk factors (P = 0.0303, relative risk = 1.75). There was no time interaction. When the variable cold ischemia time was excluded in the same model, the significance of the influence of donor origin on the graft failure risk increased considerably, whereas the magnitude of the influence was comparable (P = 0.0004, relative risk = 1.92). The influence of all other variables on the graft failure risk was unaffected when the cold ischemia period was excluded. The exclusion of none of the other variables resulted in a comparable effect. Donor origin did not influence the death risk. Conclusion. The superior results of LD versus PAID transplantations can be partly explained by the dichotomy in the cold ischemia period in these populations (selection). However, after adjustment for cold ischemia periods, the influence of donor origin still remained significant, independent of any of the variables introduced. This superiority is possibly caused by factors inherent to the transplanted organ itself, for example, the absence of brain death and cardiovascular instability of the donor before nephrectomy.