Predictive Value of Non-Contrast Brain CT Patterns in Differentiating Aneurysmal and Non-Aneurysmal Subarachnoid Hemorrhage

Abstract

Introduction This study aims to evaluate the prognostic value of CT hemorrhage patterns in identifying aneurysmal versus non-aneurysmal subarachnoid hemorrhage (SAH) and in predicting aneurysm localization, to improve early diagnosis and treatment. Methods A retrospective study included 246 patients with spontaneous SAH who underwent CT and digital subtraction angiography (DSA) from 2020 to 2024. CT hemorrhage patterns were classified into eight types, and their correlation with aneurysm presence and location was analyzed. Statistical analysis was performed using SPSS, with odds ratios (OR) calculated to assess the association between CT patterns and aneurysm localization. Results Aneurysms were identified in 67.1% of patients, with the most common locations being the anterior communicating artery (ACom) and right middle cerebral artery (RMCA). A significant association was found between aneurysms and certain CT hemorrhage patterns. Type 1 (basal cistern) and Type 4 (ventricular extension) were most strongly linked to aneurysms, with Type 4B and Type 4R patterns predominantly seen in anterior-circulation aneurysms. In contrast, Type 8 patterns, associated with perimesencephalic hemorrhage, were more frequent in non-aneurysmal cases, often linked to vascular malformations such as arteriovenous malformations (AVMs) and fistulas. Conclusion This CT-based pattern classification serves as a complementary triage tool; these CT patterns may provide insight into vascular lesion location but do not replace DSA, the gold standard for definitive vascular imaging. Further research is needed to confirm these findings and assess their broader clinical applicability. © 2026 The Association of University Radiologists.