405 Volumetric Quantification of Subarachnoid Hemorrhage Independently Predicts Hydrocephalus and Seizures. (1st March 2022)

• Record Type: Journal Article
• Authors: Daou, Badih J.; Khalsa, Siri S. S.; Anand, Sharath; Williamson, Craig; Rajajee, Venkatakrishna; Sheehan, Kyle; Pandey, Aditya S.
• Is Part Of: Neurosurgery. Volume 68(2022)Supplement 1
• Page Start: 93
• Page End: 93
• Publication Date: 2022-03-01
• Subjects: Nervous system -- Surgery -- Periodicals; 617.48005
• DOI: 10.1227/NEU.0000000000001880_405
• Languages: English
• ISSNs: 0148-396X

Abstract

Abstract : INTRODUCTION: Hydrocephalus and seizures greatly impact outcomes of patients with aneurysmal subarachnoid hemorrhage (aSAH); however, reliable tools to predict these outcomes are lacking. METHODS: Total hemorrhage volume following aneurysm rupture was retrospectively analyzed on presentation CT imaging using a custom semi-automated computer program developed in MATLAB that employs intensity-based k-means clustering to automatically separate blood voxels from other tissues. Volume data was added to a prospectively maintained SAH database. Through logistic regression analysis, we evaluated the association of hemorrhage volume with shunted-hydrocephalus and seizures and evaluated the diagnostic accuracy by evaluation of the area under the receiver operating characteristic curve (AUC). RESULTS: 288 consecutive patients with aSAH comprised the study population. Mean total hemorrhage volume was 74.9 ml. Thirty-eight patients developed seizures (13.2%). Mean hemorrhage volume in patients who developed seizures was significantly higher than patients with no seizures (mean difference 17.3 ml, P=0.01). In multivariate analysis, larger hemorrhage volume on initial CT scan and hemorrhage volume >50 ml (OR 2.81, P=0.047 CI 1.03-7.80) were predictive of seizures. 48 patients developed shunt-dependent hydrocephalus (17%). The mean hemorrhage volume in patients who developed shunt-dependent hydrocephalus was significantly higher than patients who did not (mean difference 17.2 ml, P=0.006). Larger hemorrhage volume and hemorrhage volume >50 ml (OR 2.45, P=0.03, CI 1.08-5.54) were predictive of shunt-dependent hydrocephalus. Hemorrhage volume had adequate discrimination for the development of seizures (AUC = 0.635) and shunted hydrocephalus (AUC = 0.629). CONCLUSION: Hemorrhage volume is an independent predictor of seizures and hydrocephalus in aSAH. Further evaluation of quantitative volumetric analysis may complement existing scales used in clinical practice and assist in prognostication and management.