Background: To compensate for the shortage of donor kidneys, use of expanded donor criteria (ECD) has been adopted by many transplant centres. Multiple criteria on which to score such kidneys have been proposed but the evidence base for the definitions is derived from retrospective and registry data only. We aimed to see if analysis of ECD in our population would indicate the need to change our donor selection process.
Methods: Data on primary kidney transplants (minimum follow-up two yr) from 1989 to 2004 were reviewed (n = 635). The primary study endpoint was overall graft survival. Published ECD, including the United Network for Organ Sharing (UNOS) ECD criteria were assessed as potential prognostic variables, in a multivariable Cox proportional hazards model.
Results: Patients transplanted after 1996 had improved graft survival compared to those transplanted pre-1996 HR = 0.51 (0.35-0.76), p = 0.0001. Pre-1996 UNOS defined ECD kidneys had a markedly increased risk of graft failure compared to live donor kidneys HR = 3.52 (1.9-6.35), p < 0.001. Post-1996 ECD kidneys had similar prognosis compared to live donor kidneys HR 0.38 (0.1-1.59), p = 0.184. The observed improvement in graft survival was not explained by changes in donor source, cause of end stage renal failure (ESRF), human leukocyte antigen mismatch, recipient age or any histological parameter on implantation biopsy.
Conclusions: The explanation for improved overall graft survival and marked improved survival of ECD kidneys is unclear, but introduction of mycophenolate and subsequent falls in calcineurin inhibitor doses over the study period could be potential factors. These results provide some justification for our current selection and management of ECD kidneys.