The effect of altering eligibility criteria for entry onto a kidney transplant waiting list

Background. This paper concerns the allocation of kidneys from cadaveric donors to patients with end-stage renal disease (ESRD). Currently, the decision as to whether or not a particular patient should go onto the renal transplant waiting list is left to the discretion of the local dialysis centre, and is usually based almost entirely upon consideration of each case on its individual merits. Would this person like to have a renal transplant, is this possible, and would it seem reasonable to give them a chance? It could be argued that such an approach may not make best use of a scarce national resource. In this study we explore the effects of altering the eligibility criteria for transplantation to take explicit and quantitative account of the fact that some patients are more likely to die than others. Methods. We performed a survey of one unit's dialysis patients to ascertain the characteristics used in practice to determine who should go onto the transplant waiting list and who should not. We then created a computer model to simulate a cohort of ESRD patients, initially of the same size and characteristics as that in the unit surveyed, receiving renal replacement therapy over a period of 10 years. Using this model, we compared four strategies for defining eligibility for transplantation: (1) all patients eligible; (2) standard and medium risk patients eligible; (3) only standard risk patients eligible; and (4) no regrafts performed (standard and medium risk according to definitions in the Renal Association Standards Document). Results. Strategies of allowing only standard or standard and medium risk patients onto the waiting list most closely reflected the current decisions made regarding eligibility. The different strategies considered in the models necessarily gave rise to very considerable variation in the size of the waiting list at the end of the 10 year period (range 98-368), which would have important practical implications. The predicted mean time of kidney function varied from 9.8 years for strategy 4 (no regrafts) to 10.8 years for strategy 3 (only standard risk patients eligible). However, the different strategies had very little effect on other parameters, such as numbers of deaths and the size of the dialysis population. Conclusions. Variation in decision making from centre to centre regarding access to renal transplantation could make up to a 10% (1 year) difference in the expected half-life of renal transplants performed. Information about recipient characteristics is therefore required when making comparisons between outcome in one transplant unit with that in another, or when comparing one immunosuppressive regime with another.