Should the allocation of cadaveric kidneys for transplantation be based on HLA matching

In 1996 the European centres participating in the Eurotransplant organization agreed to the implementation of a new allocation system for kidneys of postmortal donors (Eurotransplant Kidney Allocation System, ETKAS). The proposed procedure was based on the following five criteria: HLA mismatch grade, waiting time, mismatch probability (a correction factor increasing the chances for patients with uncommon HLA phenotypes and for homozygous or immunized patients), local transplants (kidneys are shipped only if the HLA match is at least two match grades better than the best compatible local patient), and import‐ export balance per centre. In the course of time several changes have been made in the original procedure, but the basic principles are still used for the exchange of kidneys within Eurotransplant. The brief report of Gillich et al. [1] in the current issue of the journal raises the question of whether the exchange based on HLA matching is still offering an advantage. In a single-centre study the transplant team in Bonn compared the fate of 77 pairs of kidneys of which one was transplanted locally and the other was offered to Eurotransplant for exchange to another centre. For the locally used kidney no attempt to match for HLA antigens with the recipient was made. Consequently, the mean mismatch grade of the local kidney transplants was almost twice that of the exchanged kidneys. Despite this difference in HLA matching, 1-year and 5-year graft survival rates of the kidney pairs did not differ. As expected cold ischaemia time of the locally used kidneys was considerably shorter. The results suggest that HLA matching is not a conditio sine qua non for a successful renal transplantation and that, with current immunosuppressive protocols, ischaemia time has become a more decisive factor. The results are of interest because the study was systematically performed by a single centre and paired kidneys from the same donor were compared. As the authors acknowledge it also has its limitations. The number of patients is relatively small and the waiting times of the recipients in both groups differ. Also, the immunosuppressive protocols used were most likely not similar in the different centres. They advise to perform larger studies to settle this question.