Volumetric Assessment of Perimesencephalic Subarachnoid Hemorrhage

Abstract

Introduction: Perimesencephalic subarachnoid hemorrhage (pmSAH) is a rare, typically benign subtype of non-aneurysmal subarachnoid hemorrhage (SAH). While the majority of patients demonstrate a positive recovery trajectory, a subset of patients experiences complications, including vasospasm, hydrocephalus, or delayed cerebral ischemia (DCI). Reliable imaging markers for risk stratification are lacking. This study evaluates whether volumetric CT-based biomarkers-validated in aneurysmal SAH (aSAH)-are also predictive for pmSAH.

Methods: In this retrospective single-center study, 72 patients with confirmed pmSAH between 2011 and 2024 were analyzed. The automated volumetric segmentation was performed using 3D Slicer and TotalSegmentator to quantify intracranial volume (ICV), brain volume (BV), cerebrospinal fluid (CSF), and selective sulcal volume (SSV). The associations between volumetric parameters and clinical presentation, complications, and functional outcome (Glasgow Outcome Scale, GOS) were assessed using non-parametric statistics and Spearman correlation.

Results: The median intracranial volume was 1352.7 mL, brain volume 1247.3 mL, cerebrospinal fluid volume 95.9 mL, and selective sulcal volume 19.4 mL. Vomiting at presentation was associated with higher CSF and SSV values (p = 0.04 and p = 0.005, respectively), but no significant volumetric differences were found regarding other symptoms or complications (vasospasm, hydrocephalus, DCI). GOS scores were uniformly high (median = 5), and none of the volumetric markers significantly correlated with outcome or complication rate (all p > 0.05).

Conclusion: In contrast to aSAH, volumetric CT biomarkers such as ICV, BV, CSF, and SSV do not offer predictive value in patients with pmSAH. Risk stratification should continue to rely on initial hemorrhage pattern and volume, clinical monitoring, and individualized assessment rather than other volumetric parameters.

Keywords: Delayed cerebral ischemia; Hydrocephalus; Perimesencephalic subarachnoid hemorrhage; Subarachnoid hemorrhage; Vasospasm; Volumetric analysis.

Fig. 1

Representative axial non-contrast CT demonstrating automated volumetric segmentation: of cerebrospinal fluid (CSF, blue overlay), hemorrhage (red overlay), and brain volume (BV, yellow overlay) in a patient with pmSAH

Fig. 2

Box-and-whisker plots illustrating the distribution of intracranial volume (ICV), brain volume (BV), cerebrospinal fluid (CSF), and selective sulcal volume (SSV), stratified by the presence or absence of in-hospital complications. The first column shows patients with and without hydrocephalus, the second column shows patients with and without vasospasm, and the third column shows patients with and without delayed cerebral ischemia (DCI). The boxes represent the interquartile range (IQR), the horizontal lines denote the median, and the whiskers span the minimum to maximum values. No statistically significant differences were observed between complication groups for any volumetric parameter