J Neurosurg 143:490–498, 2025
This retrospective study externally validated four rupture risk scoring systems for unruptured brain arteriovenous malformations (bAVMs) and found their predictive performance ranged from nondiscriminatory to poor. The R2eD AVM scale performed best but still showed limited accuracy, highlighting the need for improved predictive models.
• The study externally validated four rupture risk scoring systems for unruptured brain arteriovenous malformations (bAVMs): Nataf, R2eD AVM, ARI, and VALE.
• Current rupture risk scores showed nondiscriminatory to poor performance in predicting ruptured presentation in a Peruvian single-center population.
• The R2eD AVM scale had the best performance among evaluated scores, but its discrimination was still poor (AUROC 0.664).
• A logistic regression model using size, location, venous tortuosity, and ventricular involvement showed only fair discrimination (AUROC 0.709).
• Key risk factors for rupture included smaller size, periventricular/infratentorial location, absence of venous tortuosity, and ventricular involvement.
• Some variables in published scores, such as venous tortuosity and venous drainage features, were not predictive in this cohort.
• Further research and better predictive factors are needed to improve rupture risk assessment in bAVMs.
• The R2eD AVM score can be used preferentially in clinical practice, but with caution due to its limited performance.
You must be logged in to post a comment.