Factors Affecting Long-term First Renal Graft Survival Identified in a 10-year, 5-centre Retrospective National UK Study

Medcalf JF, James J, Bankhart J. on Behalf of the STEPP Research Group
British Transplantation Society Annual Meeting, Manchester March 2007

Introduction
Long-term renal graft survival has progressively improved, but more slowly than improvements in acute rejection. This long term transplant outcome study will collect detailed information on 3000 first renal transplant recipients. This is a presentation of graft survival in those transplanted between 1992 and 2001.

Methods
All patients who had a first renal transplant were identified in five UK transplant centres. Data pertaining to: transplant centre, patient age, sex, ethnic group, time on dialysis pre-transplant, transplant donor type (live/cadaver), year of transplant (1992-1996/1997-2001) and diabetes as the attributed cause of renal failure were collected. Patients were considered to be at risk of graft failure until they died, or until the end of December 2001 if death had not occurred by this point. 1659 patients were included in the final analysis, 403 (24%) of whom had experienced graft failure by this time. Univariable analysis was initially performed using Log-Rank tests. Factors significantly affecting graft survival were then included in a Cox multivariable analysis.

Results
2373 patients have been identified to date. Mean age at transplant 44.3yrs (SD14.4), 1491 Male, 882 Female. 1158 were transplanted 1992-1997, 1215 were transplanted 1998-2002. Ethnic gp was known in 1830(1583 white, 247 non-white).

Univariable analysis showed differences in graft survival between transplant centres (p=0.001), year of transplant (1997-2001 > 1992-1996, p=0.003), donor type (live survival > cadaver, p=0.002) and ethnic group (white > non-white p<0.001). Significant differences also existed between recipient ages; patients aged 40-49 had significantly better graft survival than either younger (<40yrs) or older recipients (60yrs). Multivariable analysis showed that recipient middle age, live donor type, recipient white ethnic group, 1997-2001 time cohort, and certain sites predict longer graft survival. There was a significant interaction between time-on-dialysis and donor age (p=0.0008), such that at age <30yrs a longer period of time on dialysis predicted better graft survival, whilst at older ages the opposite is true.

Conclusion
There were significant graft survival differences in the study cohort associated with donor and recipient characteristics, and between centres. Detailed data collected over the next two years will allow investigation of quality of life, and other patient reported outcomes, other important donor and recipient characteristics, treatment, and models of care affecting patient and graft survival.