Prediction of Allograft Failure in Pediatric Kidney Transplant Recipients: A Validation Study of the Four-Variable Kidney Failure Risk Equation

Introduction: Recent findings suggest that the four-variable Kidney Failure Risk Equation (KFRE) may be useful in predicting the likelihood of allograft failure in adult kidney transplant (KT) recipients. However, research on its application in pediatric patients is lacking. This study aimed to assess the accuracy of the four-variable KFRE for prediction of the two-year and five-year risk of allograft failure in a cohort of pediatric KT recipients.

Methods: A retrospective observational study of patients undergoing KT in a tertiary pediatric nephrology unit between 2007 and 2017 was conducted. The KFRE risk scores were determined using data collected one-year post-transplantation. Discrimination and calibration properties of the four-variable KFRE were assessed through the area under the receiver operating characteristic curves (AUC) and calibration plots.

Results: Fifty-nine patients with a median age of 12.4 (9.1-15.7) years at KT were included. Eleven (18.6%) were living donor recipients. The median estimated glomerular filtration rate one-year post-transplantation was 62.0 (49.0-75.0) mL/min/1.73 m. One (1.7%) and three (5.1%) patients experienced allograft failure within two and five years following the one-year post-transplantation date, respectively. The four-variable KFRE showed excellent and very good discrimination for the two-year and five-year risks, respectively (AUC 0.966, 95% confidence interval (CI) 0.914-1.000; AUC 0.887, 95% CI 0.732-1.000). Calibration plots demonstrated imprecise calibration.

Conclusion: The four-variable KFRE shows promise in predicting kidney failure progression in pediatric KT recipients with a functioning allograft one-year post-transplantation. Larger-scale studies are essential to confirm its predictive accuracy and establish more definitive conclusions.