Abstract DP214: Use of MRI in Emergency Department Patients with Dizziness and Association with Subsequent Stroke

Background: There are ~4 million annual ED visits for dizziness in the US. While cerebrovascular disease is detected in only a small proportion of these visits, specialist consultation and neuroimaging are often sought. We therefore aimed to describe the use of brain MRI in ED patients presenting with dizziness and to explore associations between its use and subsequent stroke. Methods: We conducted a retrospective study using data from the Healthcare Cost and Utilization Project from 11 states (2016-2021). Only ED visits for dizziness resulting in discharge to home were included. ICD-10-CM codes R42 and H81 in primary discharge position were used to identify dizzy patients and Current Procedural Terminology codes (70551-70552, 70540, and 70543) were used to detect ED brain MRI use. Study variables included patient demographics, comorbidities classified via Charlson Comorbidity Index (CCI), hospital factors (bed size, presence of EM residents and/or medical students, location, and availability of neurological services), and on-hours vs off-hours arrival/discharge. Our primary study outcome was stroke hospitalization within 30 days of index ED visit defined using validated ICD-10-CM codes. We used simple descriptive statistics to report brain MRI use and logistic regression to evaluate associations between its use and subsequent stroke. Results: Among 966,510 dizzy patients with an ED visit, 95,998 (9.9%) had a brain MRI. The proportion of dizzy patients who received brain MRI increased from 7.8% in 2016 to 12.0% in 2021. Mean ED length of stay among those with brain MRI was 621 minutes compared to 273 minutes for those who did not receive brain MRI. There were 1,486 (0.2%) stroke hospitalizations at 30 days. Risk of stroke was lower at 30 days (OR, 0.66; 95% CI, 0.54 - 0.81) in unadjusted analysis when an MRI brain was obtained. After adjusting for patient demographics, CCI, study year, hospital characteristics (teaching status, location, size, presence of neurological services) and presentation timing (on-hours vs off-hours, weekend vs weekday), risk of stroke remained lower at 30 days (aOR, 0.52; 95% CI, 0.36-0.72) among patients receiving brain MRI. Conclusion: In a large, population-based study, short-term stroke risk was halved among ED patients with dizziness discharged to home who received a MRI brain compared to those who did not, highlighting the potential role of MRI for improving the acute assessment of ED patients presenting with dizziness.