Abstract TP298: Acute Intracerebral Hemorrhage: Prospective Evaluation ofaA Pragmatic Stepwise Diagnostic Algorithm for Underlying Vascular Pathologies

Abstract

BACKGROUND: Intracerebral hemorrhage (ICH) is one of the deadliest form of stroke, but the diagnostic work-up has only been insufficiently defined. We tested the unsefulness of a pragmatic step-wise diagnostic algorithm including CT/CTA, MRI/MRA, and digital subtraction angiography (DSA) on the detection of acutely treatment-relevant underlying pathologies. PATIENTS AND METHODS: We analyzed 114 consecutive patients (42% women, mean age 66 years +/-12) hospitalized for spontaneous ICH. All patients had initial CT with or without CTA. Those without evident underlying cause and no contraindication underwent MRI/MRA. Finally, diagnostic DSA was performed in cases without evident ICH etiology on prior imaging. RESULTS: Of the 114 ICH patients, 2 showed deep ICH with CT evidence of severe small vessel disease. In 2 cases, CTA revealed cerebral venous thrombosis. The remaining 107 (95%) patients without contraindication (n=2 had a pacemaker) underwent subsequent MRI/MRA allowing detection of underlying small vessel disease in 36 (32%) and 14 (12%) with acutely treatment relevant pathology: one cerebral venous thrombosis, six secondary hemorrhagic infarcts, three cavernous malformations, two brain tumors, and two patients had reversible acute vasoconstriction syndrome. Fifty-seven patients (50%) remained without diagnosis of any underlying pathology and underwent DSA revealing 2 dural arteriovenous fistula, 2 arteriovenous malformation, 1 moyamoya syndrome, 1 ruptured aneurysm, and 1 acute cerebral vasoconstriction syndrome. CONCLUSION: Our findings suggest the diagnostic sensitivity of brain CT alone is insufficient for the diagnostic work-up in ICH patients. A pragmatic, stepwise imaging algorithm based on additional CTA, MRI/MRA and DSA lead to the diagnosis of an underlying pathology in 52%, including the detection of acutely treatment-relevant pathologies in 19% (22 of 114 patients).