tlooto

A Predictive Model for Distinguishing Non-Aneurysmal Subarachnoid Hemorrhage from Aneurysmal Subarachnoid Hemorrhage Using Clinical and Radiographic Data.

Pichaporn Boonliang, A. Kitkhuandee, W. Munkong, Phumtham Limwattananon, Pichayen Duangthongphon

2026.4.1World Neurosurgery

Paper details

Journal
World Neurosurgery
Published
2026.4.1
Citations
0
Fields
Medicine

Abstract

BACKGROUND This study identified risk factors that distinguish non-aneurysmal subarachnoid hemorrhage (naSAH) from aneurysmal subarachnoid hemorrhage (aSAH). It also assessed a clinical-radiographic predictive model for risk stratification, especially when initial computed tomography angiography (CTA) is negative.

METHODS A retrospective study of 275 patients with spontaneous SAH was conducted. Multivariate logistic regression identified independent predictors of naSAH. The model's performance was evaluated using the area under the receiver operating characteristic curve (AUROC).

RESULTS Independent predictors of naSAH included perimesencephalic SAH (PMSAH) (OR: 9.46, P < 0.001), good World Federation of Neurosurgical Societies (WFNS) grades 1-3 (OR: 2.72, P = 0.008), and diabetes mellitus (DM) (OR: 2.78, P = 0.046). Furthermore, a model combining CTA negativity, PMSAH, good WFNS grade, and DM demonstrated an AUROC of 0.9587. Notably, when all 3 clinical features were present and CTA was negative, the predicted probability of naSAH was 98%.

CONCLUSIONS PMSAH, good WFNS grade, and DM are strongly associated with naSAH. While these factors increase the pre-test probability of non-aneurysmal etiology, digital subtraction angiography (DSA) remains the gold standard for definitive diagnosis. This model serves as a supplementary tool for clinical counseling and prioritizing diagnostic urgency.

Citation format

BOONLIANG, Pichaporn, et al. A predictive model for distinguishing non-aneurysmal subarachnoid hemorrhage from aneurysmal subarachnoid hemorrhage using clinical and radiographic data. World Neurosurgery, 2026, 210: 124978.