▪ Objective: To compare patient and graft survival of recipients of kidneys from living, unrelated donors (LUDs); cadaveric donors; and living, related donors (LRDs) matched for zero (mismatched), one, or two (identical) haplotypes.
▪ Design: Cohort study.
▪ Setting: Sixty-three renal transplantation centers affiliatedfiliated with the Brazilian Transplantation Registry (accounting for more than 95% of the transplantation activity in Brazil).
▪ Patients: Patients having renal transplantation between January 1987 and March 1991. Of 2892 patients, 165 (6%) received transplants from LUDs; 964 (33%), from cadaveric donors; 183 (6%), from zero haplotype, HLA-matched LRDs; 1259 (44%), from one haplotypematched LRDs; and 321 (11%), from two haplotype-matched LRDs.
▪ Measurements: Patient and graft survival. Patients were followed for an average of 15.8 months.
▪ Results: After adjustment for age, race, diagnosis of primary disease, history of previous transplantation, cyclosporine use, and number of transplants from LUDs per center, patient survival did not differ statistically for recipients of kidneys from LUDs and recipients of cadaveric kidneys (risk ratio [RR], 1.16; 95% CI, 0.68 to 1.98). Little difference was seen between the adjusted death rate for recipients of zero haplotype-matched LRDs and recipients of cadaveric kidneys (RR, 1.13; CI, 0.69 to 1.87). Similarly, in a multivariate analysis, recipients of kidneys taken from LUDs and zero haplotype-matched LRDs had a risk for graft failure that did not differ statistically from that of cadaveric kidney recipients (RR, 0.74; CI, 0.45 to 1.22 and RR, 0.82; CI, 0.53 to 1.25, respectively).
▪ Conclusions: Graft survival for recipients of kidneys from LUDs is similar to that from zero haplotype-matched LRDs and is at least as good as that achieved with cadaveric transplants.