Response to Letter to Editor: Kidney Machine Perfusion and Improved Long‐Term Graft Survival

We appreciate the letter and feedback provided by Prof. Dr. Peter Schnuelle [1] in response to our retrospective analysis of the UNOS-STAR file, which revealed an association between machine perfusion (MP) use and improved long-term graft outcomes in kidney transplant recipients from deceased donors with cold ischemia time (CIT) > 12 h; particularly mediated by reduction in delayed graft function (DGF) [2]. This letter raises an important point that was challenging to assess in our registry analysis. The main point of discussion concerns selection bias related to donor anatomy. It is possible that, in our mate-kidney analysis, in which one mate-kidney underwent MP and the other did not, kidneys with complex anatomy may have been placed in cold storage more frequently than those with less complex anatomy, which are likely easier to be placed on MP. These donor kidneys with complex anatomy would likely have longer anastomosis times and, hence, a presumed higher risk of DGF, potentially confounding our results. Our limited dataset lacked details on donor anatomy. Therefore, it is a limitation of our study. However, according to the literature, 86% of donor kidneys have a single artery, and only 14% have multiple arteries [3]. Therefore, the likelihood of anatomical complications significantly impacting the results is small. Another study by Miller et al. on the implications of MP on renal anatomy revealed that the number of arteries supplying the right kidney significantly predicted the number of arteries supplying the left kidney [4]. Based on these data, since our study was a mate-kidney analysis, if one kidney had multiple arteries, it is reasonable to infer that the mate-kidney was also likely to have multiple arteries. Thus, one could speculate that the number of organs with complex anatomy was equally distributed between the two groups. There has been an increased utilization of MP in deceased-donor kidneys in the United States. Based on OPTN data, the overall percentage of deceased donor kidneys using MP increased from ~30% in 2010 to 52.8% in 2023 [5]. With the advancements in MP technology and widespread comfort in its use, multiple arteries and complex anatomies are now less of a barrier to its use than in previous years. Even though multiple arteries are likely still contributing to prolonged ischemia and anastomosis times, as described in a case report published by De Carlis et al., it is possible to use MP in kidneys with multiple arteries via arterial graft reconstruction [6]. We agree with the summary statement that prospective, randomized controlled trials (RCTs) are needed to address the knowledge gap regarding the potential long-term benefits of MP for kidney transplant outcomes. We also acknowledge the limitations of prior RCTs, in which the anatomical constraints of the donor kidneys limited the randomization and hence the interpretation of results. Moreover, long-term RCTs are costly and challenging to design. While our registry analysis was robust and nationally representative, its retrospective design has inherent limitations, as we can only establish an association, not causation. While pointing to an overall MP benefit, given the limitations of existing RCTs and observational studies, the evidence remains inconclusive, and the debate over the long-term utility of MP is ongoing. Nonetheless, our study contributes to the body of evidence supporting the use of MP in deceased-donor kidneys with relatively prolonged CIT. Based on the existing data and the increasing MP utilization trend [5], we expect our data to encourage the design of RCTs to assess the impact of MP on kidney transplant outcomes with the goal of minimizing organ discard and improving graft outcomes. Bhavna Chopra: original idea, design, methodology and writing the manuscript. Gabriel Cojuc-Konigsberg: key design, methodology, review and edit manuscript. The authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.