External validation of the VALE scoring system for hemorrhage risk in pediatric AVM patients

Brain arteriovenous malformations (AVMs) are the leading cause of spontaneous intracranial hemorrhage in children. The recently developed VALE scoring system has demonstrated predictive value for hemorrhage risk in adult AVM patients; however, its applicability to pediatric populations remains unknown. We performed a retrospective analysis of a prospectively maintained vascular database at University Hospital Frankfurt. Pediatric patients (≤ 18 years) diagnosed with a single brain AVM between 2005 and 2023 were included. VALE scores were calculated according to the original model using ventricular system involvement, associated venous aneurysms, deep location, and exclusively deep venous drainage. Logistic regression analyses were performed to evaluate associations with hemorrhagic presentation. Discriminatory performance was assessed using receiver operating characteristic (ROC) analysis. A total of 52 pediatric AVM patients were included, of whom 31 (60%) presented with hemorrhage. Associated venous aneurysms were observed in 29% of ruptured AVMs compared with 5% of unruptured lesions and represented the only VALE component that remained significant in multivariable analysis. The complete VALE score demonstrated limited discriminatory performance, yielding an area under the ROC curve (AUC) of 0.608 (95% CI 0.454-0.762). Classification into VALE-defined risk categories showed similarly limited predictive ability (AUC 0.545, 95% CI 0.390-0.700), with no significant increase in hemorrhage risk across risk groups. This study represents the first external validation of the VALE scoring system in a pediatric AVM cohort. In our population, the VALE score demonstrated limited discriminatory performance and could not be conclusively validated. While associated venous aneurysms remained associated with hemorrhagic presentation, larger multicenter studies are required to further evaluate the applicability of the VALE score in pediatric patients and to determine whether pediatric-specific hemorrhage risk prediction models may be beneficial.