Abstract WP358: Impact of an Emergency Department-Based Acute Stroke MRI Protocol on Cost, tPA Administration, and Hospital Length of Stay

Abstract

Background: MRI is more sensitive for the detection of ischemic stroke than CT, but is not routinely used in the Emergency Department (ED) due to limited availability, prolonged scan time, and higher cost. Hypothesis: A 24/7 ED-based rapid MRI (8 minutes scan time) for the diagnosis of acute ischemic stroke will reduce direct costs, tPA administration to stroke mimics, and length of stay. Methods: We included patients who were admitted to the ED under our Brain Attack stroke activation and had either a CT perfusion (CTP) (2013-2015) or, after it became available, a rapid MRI (2015-2018). The primary outcomes are total direct cost (reported as % of the mean cost for CTP due to data reporting restrictions), % of tPA administration given to stroke mimics, hospital length of stay in days. Results: Our cohort includes 215 patients, aged 58.6±19.4 years and 45% male. Acute ischemic stroke was the discharge diagnosis in 79/215 (37%) of patients. The most common non-stroke diagnoses were psychiatric or unmasking (n=57, 27%), complicated migraine (n=43, 20%), seizure (n=13, 6%), and TIA (n=12, 6%). Comparing patients with CTP to those with MRI, including after restricting to patients with stroke mimic (n=136), the rapid MRI was associated with a significant decrease in costs, tPA administration to stroke mimics, and hospital length of stay (Tables 1 and 2). There was also an increase in the number of discharges home from the ED for patients who had the MRI. Conclusion: The introduction of a rapid MRI protocol in our emergency department appears cost-effective and may lead to more accurate decision-making for tPA administration and emergency department disposition in patients suspected of having acute stroke.