Center effects in renal transplantation.

1. One-year graft survival rates for first cadaver transplants to adult recipients ranged from 60-93% at transplant centers reporting more than 50 transplants to the UNOS Renal Transplant Registry between October 1987 and December 1991. 2. There was no apparent correlation between center size and success rates for primary or repeat cadaveric transplants when centers were grouped according to high, average, and low 1-year graft survival rates. 3. Fifteen "high" centers had 88%, 15 "average" centers had 80%, and 15 "low" centers had 69% 1-year graft survival rates, respectively. 4. Projected half-lives for transplants surviving the first year were not significantly different among the 3 center groups, suggesting that long-term survival did not correlate strictly with 1-year survival. 5. The 20% difference in 1-year graft survival rates between the high and low center groups was reduced to 10% and to 5% when transplants functioning at discharge or at 6 months, respectively, were considered. Thus, approximately half of the center effect was associated with events that occurred during the transplant hospitalization. 6. At high and average centers, less than 10% of kidneys did not function on the first day compared with 17% at low centers (p less than 0.01). Twenty percent of patients at high and average centers required dialysis during the first week compared with 37% at low centers (p less than 0.01). Less than 5% of kidneys never functioned during the transplant hospitalization at high and average centers compared with 10% at low centers (p less than 0.01). 7. Graft survival rates for patients with early graft dysfunction were significantly higher at high centers than at average or low centers, suggesting that successful management of patients with poor early function differed among the center groups. 8. Low centers transplanted more Blacks and fewer healthy patients than high centers, but when stratified for these variables, the center differences in graft outcome were undiminished. 9. Other patient mix variables, including age, sensitization, and original disease, did not account for variation in survival rates for the center groups. High centers transplanted more diabetics (33%) than average (25%) or low (18%) centers. 10. There were no significant differences among the center groups in the incidence of early rejection episodes. However, graft survival following rejection was 10-30% lower at average and low centers than at high centers.(ABSTRACT TRUNCATED AT 400 WORDS)