Response: DonorNet and the Potential Effects on Organ Utilization

Abstract

To the Editor: We read the article by Gerber and colleagues (1) with particular interest in the placement of locally rejected kidneys, because facilitating kidney placement at this juncture has the greatest potential to reduce the likelihood of discard and extended cold ischemic times (CIT) for transplanted kidneys. The authors suggested that their data ‘demonstrates improvements in the efficiencies of organ placement, including a decrease in the accepted organ offer for select kidney match runs’. They did not point out that for other select kidney match runs, the 90th, 95th and 99th percentiles, acceptor sequence numbers were higher in the post-DonorNet (DN) cohorts compared to pre-DN, indicating that for difficult-to-place kidneys, more offer rejections occurred under DN before finding an acceptor, signifying decreased efficiency of kidney placements. The interpretation of these data is also compromised by the absence of testing for proportional differences in the completion of kidney runs by era. This is a barrier for understanding the impact of DN since any era with proportionately more early closures of the match run would be biased toward lower acceptance sequence numbers. Similarly, another means of addressing efficiency of kidney placement would be to determine how often the match run was closed early because center minimal renal acceptance criteria for CIT had been exceeded. The authors found that the 90th, 95th and 99th percentiles of center number also increased under DN for ECD and DCD kidneys and state that this data suggests ‘an increase in efficiency of placement …’ More likely, this data indicates that more centers were offered and declined difficult-to-place kidneys under DN. Although offering kidneys to more centers may demonstrate improved justice-based allocation, by enforcing allocation rules and reducing offers to aggressive centers lower on the list, it does not appear to represent ‘increased efficiency’. It is stated that ‘for kidney … no significant changes in CIT were observed at any level of sharing’; however, it appears that the asterisk in Figure 2 indicates a significantly longer mean CIT in the later-DN compared to the pre-DN era for national transplants. This provides additional evidence of decreased efficiency in national placement. Unfortunately, CIT is a suboptimal surrogate to evaluate the efficacy of kidney placement because other factors such as the availability of commercial flights and center specific practices, also impact CIT. A potentially better surrogate, if available, would be duration of cold ischemia at the time of offer. Finally, based on the finding of similar kidney discard rates between eras, the authors state that historic arguments suggesting that ‘aggressive’ centers drove organ utilization in the pre-DN period is not supported by their data. However, this conclusion may not be supported by the results since the authors did not investigate which centers were accepting these organs in each era. It may be that the same ‘aggressive’ centers are still driving organ utilization; a concept that is supported by a recent analysis that found the cumulative distribution of imports by center were similar before and after DN (2).