Abstract TP554: Etiologies of Nontraumatic Convexity Subarachnoid Hemorrhage

Abstract

Introduction: Nontraumatic convexity subarachnoid hemorrhage (cSAH) is a non-aneurysmal variant associated with various etiologies. Methods: We performed retrospective review of consecutive cSAH admitted to a large comprehensive stroke center, from 8/1/06-1/1/16. We abstracted data on demographics, neuroimaging, and clinical presentation; trauma was excluded. Cases were categorized: cryptogenic (negative work-up), susceptibility-weighted imaging-confirmed amyloid angiopathy (AA), posterior reversible encephalopathy syndrome (PRES), imaging-confirmed reversible cerebral vasoconstriction syndrome (RCVS), cerebral venous thrombosis (CVT), large artery occlusion (LA), or other known cause (sepsis, endocarditis, cerebral infarcts, or malignancy). Results: We identified 84 cases of cSAH occurring in 83 patients. Etiology subgroups were: 25% cryptogenic (N=21), 16% AA (N=13), 13% PRES (N=11), 13% RCVS (N=11), 11% CVT (N=9), 8% LA (N=7), and 14% other (N=12). Among cryptogenic patients, 8 had suspected clinical RCVS and 1 had suspected AA. There were significant differences in age, gender, and clinical presentation among etiology subgroups. AA patients were the oldest (mean 75.6 years), while RCVS patients were the youngest (45.6 years, p<0.0001). The majority of AA cohort was male (61%), whereas PRES and RCVS cohorts were exclusively female (100%, p=0.0002). Transient neurologic symptoms were common in AA (69%) but rare in RCVS (0%, p=0.01); headache was common in RCVS (80%) but unusual in AA (15%, p=0.001). Among 11 patients with confirmed RCVS, initial vascular imaging was negative in 6 (55%); repeat vascular imaging was necessary to diagnosis vasoconstriction (mean delay 5 days, range 3 -16 days). Conclusions: To our knowledge, this is the largest case series of cSAH patients. Convexity SAH has a broad range of etiologies, with the most common being cryptogenic, RCVS, PRES, and AA. In our cohort, cSAH due to AA most typically presented as older males with transient neurological symptoms; cSAH due to RCVS presented as younger females with headache. Over half of patients with cSAH due to RCVS required repeat vascular imaging to confirm vasoconstriction. Further research is warranted to clinically characterize this complex variant of SAH.