The Scorecard: Comparing different scoring systems to predict the surgical outcome of pyeloplasty in children
TL;DR
PPS demonstrated superior predictive value for both hydronephrosis improvement and cortical gain, correlating consistently with risk stratification, in children with PUJO.
Abstract
ABSTRACT
Background
Pelvi-ureteric junction obstruction (PUJO) is a functional or anatomical obstruction at the renal pelvis–ureter junction, leading to impaired urine drainage and is the most common cause of antenatally detected hydronephrosis. Several pyeloplasty prediction scoring systems have been developed to quantify disease severity and estimate the need for surgical intervention. This study aimed to apply these scoring systems not only to stratify severity but also to predict surgical outcomes, allowing more informed parental counselling.
Objective
To evaluate the predictive performance of three scoring systems; Pyeloplasty Prediction Score (PPS), Onen Hydronephrosis Grading, and the Hydronephrosis Severity Score (HSS), in foretelling postoperative outcomes in children with PUJO, specifically cortical gain and improvement of hydronephrosis. MATERIALS &
Methods
This descriptive prospective cross-sectional study included 70 children aged 1–13 years who underwent Anderson–Hynes pyeloplasty for PUJO. Patients with secondary causes or elevated eGFR were excluded. At presentation, all children were categorized into risk groups according to PPS, Onen grade, and HSS based on ultrasonography and MAG-3 renogram findings. Postoperative follow-up at 2 months assessed cortical gain and hydronephrosis resolution. RESULTS:In 70 children undergoing pyeloplasty, PPS showed superior prediction of hydronephrosis resolution, outperforming HSS (AUC difference –0.229, p=0.002) and Onen (–0.044, p=0.169), while HSS was inferior to Onen (p=0.004). For cortex gain, PPS exceeded HSS (AUC difference 0.214, p=0.040) but was similar to Onen (p=0.730).
Conclusion
PPS demonstrated superior predictive value for both hydronephrosis improvement and cortical gain, correlating consistently with risk stratification. Onen and HSS did not show proportional predictive performance for these outcomes at 3-month follow-up.