Sequential kidney biopsies from deceased donors often show substantial differences in histologic categorization, which may cause some organs to be unnecessarily discarded, according to a retrospective study. "Although we think of procurement biopsies as providing objective/reliable data on organ quality, as currently practiced they instead often provide information that is not reproducible and not associated with allograft outcomes," Dr. S. Ali Husain of Columbia University, in New York City, told Reuters Health by email. In a paper in the Clinical Journal of the American Society of Nephrology, Dr. Husain and colleagues note that as many as 20% of deceased donor kidneys are discarded. Moreover, 38% of all discards follow frozen section biopsies performed around the time of organ recovery. In addition, the authors say, procurement biopsies are usually frozen sections that oversample scarred subcapsular cortex "and are interpreted by on-call pathologists with limited expertise in kidney pathology." The team studied data on more than 1,000 deceased-donor kidneys transplanted at their institution between 2006 and 2016. Of these, 60% had multiple procurement biopsies. Most (98%) had their first biopsy performed elsewhere and their second biopsy performed locally. Ten kidneys had three procurement biopsies. The first and the last were used for comparison. Overall, say the researchers, "histologic categorization was discordant in 35% of repeat biopsies, and kappa agreement statistics for each histologic compartment ranged from just 0.12 - 0.17." Histologic concordance was 85% for kidneys whose first biopsy was categorized as optimal but just 27% for those whose first biopsy was categorized as suboptimal. Optimal as opposed to suboptimal histology on second biopsy was tied to significantly better allograft survival, but this was not the case for first-biopsy histology. Thus, concluded Dr. Husain, "unfavorable procurement biopsy histology should not be used alone to make organ utilization decisions. Rather, such findings may be the result of sampling error, and repeating a biopsy in a standardized fashion may help identify kidneys whose quality was underestimated by their first biopsies." Dr. Minnie Sarwal, director of the Precision Transplant Medicine Initiative at the University of California, San Francisco, told Reuters Health by email, "The results of the study highlight the problem with the biopsy, its variability in sampling and possibly readability, as correlations are weak/absent." "A take-home message from this," concluded Dr. Sarwal, who was not involved in the research, "would be to consider more automated AI-based histological analysis to eliminate human bias, and to consider more biologically robust platforms such as RNA or protein measurements from the tissue, to more accurately relate to functional outcomes." —David Douglas Source
