An integrated machine learning model enhances delayed graft function prediction in pediatric renal transplantation from deceased donors

Abstract

Background: Kidney transplantation is the optimal renal replacement therapy for children with end-stage renal disease; however, delayed graft function (DGF), a common post-operative complication, may negatively impact the long-term outcomes of both the graft and the pediatric recipient. However, there is limited research on DGF in pediatric kidney transplant recipients. This study aims to develop a predictive model for the risk of DGF occurrence after pediatric kidney transplantation by integrating donor and recipient characteristics and utilizing machine learning algorithms, ultimately providing guidance for clinical decision-making.

Methods: This single-center retrospective cohort study includes all recipients under 18 years of age who underwent single-donor kidney transplantation at our hospital between 2016 and 2023, along with their corresponding donors. Demographic, clinical, and laboratory examination data were collected from both donors and recipients. Univariate logistic regression models and differential analysis were employed to identify features associated with DGF. Subsequently, a risk score for predicting DGF occurrence (DGF-RS) was constructed based on machine learning combinations. Model performance was evaluated using the receiver operating characteristic curves, decision curve analysis (DCA), and other methods.

Results: The study included a total of 140 pediatric kidney transplant recipients, among whom 37 (26.4%) developed DGF. Univariate analysis revealed that high-density lipoprotein cholesterol (HDLC), donor after circulatory death (DCD), warm ischemia time (WIT), cold ischemia time (CIT), gender match, and donor creatinine were significantly associated with DGF (P < 0.05). Based on these six features, the random forest model (mtry = 5, 75%p) exhibited the best predictive performance among 97 machine learning models, with the area under the curve values reaching 0.983, 1, and 0.905 for the entire cohort, training set, and validation set, respectively. This model significantly outperformed single indicators. The DCA curve confirmed the clinical utility of this model.

Conclusions: In this study, we developed a machine learning-based predictive model for DGF following pediatric kidney transplantation, termed DGF-RS, which integrates both donor and recipient characteristics. The model demonstrated excellent predictive accuracy and provides essential guidance for clinical decision-making. These findings contribute to our understanding of the pathogenesis of DGF.

Keywords: DGF; Delayed graft function; Machine learning; Pediatric kidney transplantation; Predict.

Fig. 1

Schematic overview of the development and validation of the DGF-RS model for predicting DGF in pediatric kidney transplant recipients. The process involves data collection, preprocessing, feature selection using univariate LR and comparative analysis, and training and evaluation of various machine learning models. The best-performing model, a RF with an mtry value of 5 and 75% of features retained, was selected as the final DGF-RS model. Abbreviations: DGF, delayed graft function; DGF-RS, DGF-Risk Score; HDLC, high-density lipoprotein cholesterol; DCD, donor after circulatory death; WIT, warm ischemia time; CIT, cold ischemia time

Fig. 2

Identification of key features associated with DGF in pediatric KT. A Forest plot depicting the odds ratios and 95% CIs of variables significantly associated with DGF in univariate LR analysis (P < 0.05). B Comparative analysis of the distribution of these significant features between the DGF and non-DGF groups, revealing notable differences (P < 0.05). Abbreviations: DGF, delayed graft function; KT, kidney transplantation; LR, logistic regression; HDLC, high-density lipoprotein cholesterol; DCD, donor after circulatory death; WIT, warm ischemia time; CIT, cold ischemia time

Fig. 3

Comprehensive evaluation and comparison of machine learning models for the prediction of DGF. A The top 10 best-performing models ranked by their average area under the AUC values on the training and validation sets. B Detailed AUC values of the top 10 models, with the RF model (mtry = 5, 75%p) achieving the highest average AUC of 0.951. C–E Comparison of other commonly used binary classification metrics, including recall, F1-score, precision, and accuracy, among the top models, confirming the superior performance of the RF (mtry = 5, 75%p) model. Abbreviations: DGF, delayed graft function; AUC, area under the curve; RF, random forest

Fig. 4

Comprehensive assessment of the predictive performance of the delayed graft function risk score (DGF-RS) model based on the RF (mtry = 5, 75%p) algorithm. A ROC curves and AUC values of the DGF-RS in the entire cohort, training set, and validation set, demonstrating excellent discriminative ability (AUC: 0.983, 1.000, and 0.905, respectively). B,C DCA illustrating the net benefit of using the DGF-RS for clinical decision-making at different threshold probabilities, providing strong evidence of its potential clinical utility. Abbreviations: RF, random forest; ROC, receiver operating characteristic; AUC, area under the curve; DCA, decision curve analysis