Comparison of Outcomes of Ischemic Stroke Initially Imaged With Cranial Computed Tomography Alone vs Computed Tomography Plus Magnetic Resonance Imaging

Importance: Patients with acute ischemic stroke often undergo magnetic resonance imaging (MRI) in addition to computed tomography (CT), but its association with clinical outcomes is uncertain. Objective: To assess whether clinical outcomes of patients with acute ischemic stroke with initial CT alone were noninferior to those with additional MRI. Design, Setting, and Participants: A retrospective observational propensity score-matched cohort study of clinical outcomes at discharge and 1 year for patients hospitalized with acute ischemic stroke was conducted at an academic medical center between January 2015 and December 2017. Data collection from an electronic medical record system performed from May 2020 through January 2022 was not completely blinded. Noninferiority margins were based on the designs of previous randomized clinical trials of ischemic stroke treatments. Statistical analysis was performed in January 2022. Participants were adults hospitalized with acute ischemic stroke with admission diagnosis based on CT. Exclusion criteria were primarily missing data. From 508 eligible patients, all 123 cases with additional MRI were propensity-score matched to 123 controls without. Exposure: MRI after initial diagnosis. Main Outcomes and Measures: Death or dependence at hospital discharge (modified Rankin Scale score of 3-6) and stroke or death occurring in survivors within 1 year after discharge. Results: Among 246 participants, the median age was 68 years (IQR, 58-78.8 years) and 131 (53.0%) were men. Death or dependence at discharge occurred more often in patients with additional MRI (59 of 123 48.0%) than in those with CT alone (52 of 123 42.3%; absolute difference, 5.7%; 95% CI, -6.7% to 18.1%), meeting the -7.50% criterion for noninferiority. Stroke or death within 1 year after discharge determined for 225 of 235 (96%) survivors occurred more often in patients with additional MRI (22 of 113 19.5%) than in those with CT alone (14 of 112 12.5%; relative risk, 1.14; 95% CI, 0.86-1.50), meeting the 0.725 relative risk criterion for noninferiority. Conclusions and Relevance: This propensity score-matched cohort study of patients hospitalized with acute ischemic stroke found that a diagnostic imaging strategy of initial CT alone was noninferior to initial CT plus additional MRI with regard to clinical outcomes at discharge and at 1 year. Further research is needed to determine which patients hospitalized with acute ischemic stroke benefit from MRI.