Introduction: To develop and validate a nomogram to predict remnant kidney function after living-donor kidney donation.

Method: For nomogram construction and validation, all the donors were randomly divided into two cohorts, including training cohorts and validation cohorts. Then we identified independent prognostic factors using univariate analysis and multivariate logistic regression models. A nomogram for predicting 1-year eGFR was constructed based on these identified prognostic factors. The performance of the nomogram was validated both internally in training cohort and externally in validating cohort.

Results: Age and pre-donation eGFR were significantly identified in multivariate analysis. Finally, a nomogram was constructed by incorporating these two independent predictors. The C-indexes for eGFR prediction in the nomogram were 0.761 and 0.782 for the training set and validation set. The calibration plot showed good agreement between the actual observations and the predicted outcomes both in training set and validation set.

Conclusion: This model might be a simple, but useful guide to predict remnant kidney function after donation, which could be an important clinical tool to improve the selection of living donors.