*Purpose: This study assesses the tradeoffs that physicians make when accepting marginal-quality kidneys, and the extent to which recipient-specific factors influence the acceptability of these kidneys, with the goal of informing policies to reduce kidney discard.

*Methods: We developed and administered a discrete-choice experiment (DCE) to surgeons and nephrologists in the United States who are actively involved in kidney acceptance decisions. The DCE presented a series of kidneys described in terms of kidney donor profile index (KDPI), expected cold ischemia time, donor age, pump parameters, creatinine levels, glomerulosclerosis, and whether the donor had diabetes or donation was made after circulatory death. Respondents were asked to choose between pairs of hypothetical recipients for the kidneys, or to decline the kidney for both recipients. KDPI for the kidneys ranged from 75% to 98%. Recipient characteristics included estimated post-transplant survival (EPTS) score, ejection fraction, human leukocyte antigen mismatch, calculated panel reactive antibody (CPRA), and Karnofsky score. The recipient profiles were designed experimentally to enable estimation of the impact that each characteristic had on the acceptability of kidneys. Regression analysis was used to analyze physicians’ choices and estimate acceptability weights associated with kidney and recipient characteristics.

*Results: A total of 81 physicians answered between 4 and 10 DCE questions. Results showed that respondents considered kidney factors beyond KDPI when making choices. For example, 8 hours of cold ischemia time was acceptable for 85.3% (95% CI: 62.8%–1.07%) of kidneys, but only 64.9% (95% CI: 37.9%–91.9%) and 20.0% (95% CI: -12.2%–52.0%) of kidneys were acceptable after 16 and 30 hours of cold ischemia time, respectively. In terms of recipient characteristics, kidney acceptability generally decreased with the recipient EPTS score and CPRA, but increased again after EPTS was higher than 61.9% (95% CI: 50.4%–73.4%). Respondents were also less likely to give a kidney to a recipient with a Karnofsky score of 70, relative to those with a score of 80 (P<0.001) or 50 (P=0.047).

*Conclusions: While respondents considered a variety of factors, we found systematic effects of kidney and recipient characteristics on the acceptability of marginal-quality kidneys. This information could be critical in developing strategies to better match such kidneys with patients with an appropriate clinical background.

« Back to 2022 American Transplant Congress