External Validation of the PLASE Score Prediction Model for PDA Surgery

The PLASE score-a predictive model for subsequent patent ductus arteriosus (PDA) surgery using simple echocardiographic indices on day 3-has demonstrated strong predictive ability. To our knowledge, no predictive model for PDA surgery has undergone external validation. This study aimed to externally validate and update the PLASE score using an independent cohort. The PLASE score was applied to a validation dataset comprising infants admitted to Kanagawa Children's Medical Center between 2019 and 2024 with a gestational age of 23 to 29 weeks. Discriminative performance was assessed using the area under the receiver operating characteristic curve (ROC-AUC) and the Brier score. Calibration analysis was also conducted. The PLASE score was subsequently updated using the entire dataset. The validation cohort included 127 patients-23 who underwent PDA surgery and 104 who did not. The PLASE score exhibited good discriminative ability, with a ROC-AUC of 0.833 (95% confidence interval [CI], 0.755-0.911) and a Brier score of 0.125. After model updating, the ROC-AUC improved to 0.846 (95% CI, 0.810-0.882). The PLASE score demonstrated high predictive performance in this external validation cohort. Based on the integrated dataset, a revised model (PLASE score version 2) was developed, offering a more robust and clinically applicable tool for current neonatal care.