Object: Predicting which patients with aneurysmal subarachnoid hemorrhage (SAH) will develop delayed ischemic neurological deficit (DIND) due to vasospasm remains subjective and unreliable. The authors analyzed the utility of a novel software-based technique to quantify hemorrhage volume in patients with Fisher Grade 3 aneurysmal SAH.
Methods: Patients with aneurysmal SAH in whom a computerized tomography (CT) scan was performed within 72 hours of ictus and demonstrated Fisher Grade 3 SAH were analyzed. Severe DIND was defined as new onset complete focal deficit or coma. Moderate DIND was defined as new onset partial focal deficit or impaired consciousness without coma. Fifteen consecutive patients with severe DIND, 13 consecutive patients with moderate DIND, and 12 consecutive patients without DIND were analyzed. Software-based volumetric quantification was performed on digitized admission CT scans by a single examiner blinded to clinical information. There was no significant difference in age, sex, admission Hunt and Hess grade, or time to admission CT scan among the three groups (none, moderate, or severe DIND). Patients with severe DIND had a significantly higher cisternal volume of hemorrhage (median 30.5 cm3) than patients with moderate DIND (median 12.4 cm3) and patients without DIND (median 10.3 cm3; p < 0.001). Intraparenchymal hemorrhage and intraventricular hemorrhage were not associated with DIND. All 13 patients with cisternal volumes greater than 20 cm3 developed DIND, compared with 15 of 27 patients with volumes less than 20 cm3 (p = 0.004).
Conclusions: The authors developed a simple and potentially widely applicable method to quantify SAH on CT scans. A greater volume of cisternal hemorrhage on an admission CT scan in patients with Fisher Grade 3 aneurysmal SAH is highly associated with DIND. A threshold of cisternal hemorrhage volume (> 20 cm3) may exist above which patients are very likely to develop DIND. Prospective application of software-based volumetric quantification of cisternal SAH may predict which patients will develop DIND.