Abstract
To determine whether three-dimensional quantitative assessment of aneurysm wall enhancement (AWE) improves the identification of high-risk intracranial aneurysms (IAs) compared to existing risk scoring systems. This cross-sectional, multicenter observational study included patients with saccular IAs who prospectively underwent 3T high-resolution MRI between 2018 and 2023 at the University of Iowa, Changhai Hospital, and Ren Ji Hospital. IAs were classified as symptomatic if associated with rupture, sentinel headache, cranial nerve palsy, or neurological symptoms due to mass effect. AWE was quantified using three-dimensional aneurysm wall mapping. The diagnostic performance of the PHASES score was compared to a multivariate logistic regression model developed using all-subsets regression, incorporating PHASES components along with sex, smoking status, aneurysm morphology, and AWE data. A total of 334 patients with 387 saccular IAs were analyzed; 73 aneurysms (19%) were symptomatic. In multivariate analysis, three-dimensional circumferential AWE (3D-CAWE) was independently associated with symptomatic presentation (OR 6.1, 95% CI 2.1–19.5; p < 0.001). The PHASES score achieved an area under the curve (AUC) of 0.64 (95% CI 0.56–0.71) for discriminating symptomatic IAs. A comprehensive model including age, smoking status, size ratio, and 3D-CAWE significantly outperformed PHASES, achieving an AUC of 0.80 (95% CI 0.74–0.85; ΔAUC = 0.16, 95% CI 0.09–0.23; p < 0.001). A comprehensive risk model that integrates clinical variables, aneurysm morphological features, and 3D-CAWE significantly improves the classification of symptomatic IAs compared to the PHASES score alone. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10143-025-04120-w.