*Purpose: Among pediatric kidney donors the discard rate is highest among donors under 3-years-old. This study aims to identify factors predicting discard and construct a discard risk index (pkDSRI).

*Methods: The United Network of Organ Sharing (UNOS) database was used to retrospectively analyze 3,972 deceased donors (<=3-years-old) through univariate and multivariate logistic regressions. Donors were divided into a training set (n = 2,648 donors, 66.67%) and a validation set (n = 1,324 donors, 33.33%). The training set was used to derive an index for kidney discard. Factors significant in the univariate analysis (p< 0.05) and multivariate analysis were incorporated into a pkDSRI score. The pkDSRI was internally validated with the validation set.

*Results: Eighty-nine potential risk factors for kidney discard were examined and identified, and 34 were significant predictors (p <0.05). The most significant factor was: Age<1 (odds ratio [OR], 6.39; confidence interval [CI], 3.33-12.25). Equation 1 presents the risk equation used to calculate the DSRI. The DSRI has a C-statistic of 0.79 for the training set and 0.77 for the validation set.

*Conclusions: The pkDSRI uses 5 significant risk factors, including elevated lab values (creatinine) donor age, and region (UNOS regions 5, 10), to accurately predict risk of kidney discard with a C-stat of 0.79 in a population experiencing a high discard rate (28.93%). The pkDSRI can be used to predict discard and identify high probability discard donors for aggressive allocation.

« Back to 2020 American Transplant Congress