Log in to MyACC
Ischemic Stroke Initially Imaged With CT Alone vs. CT+MRI
Quick Takes
- Between 1999–2008, use of MRI increased from 28% to 66% of acute ischemic stroke patients, and it remains uncertain if the increased utilization of MRI has resulted in an improvement in patient outcomes.
- The authors used propensity-score matching to compare the outcomes at discharge and at 1 year of patients who received a CT only to patients who received a CT and MRI.
- The authors found that an approach of CT only was noninferior to CT+MRI, though these results should be interpreted with caution.
Study Questions:
Are clinical outcomes of acute ischemic stroke patients evaluated with computed tomography (CT) only noninferior to the outcomes of patients evaluated with CT and magnetic resonance imaging (MRI)?
Methods:
This is a retrospective cohort study of patients admitted to a single tertiary academic medical center between January 2015–December 2017. Included patients were ≥18 years old, had an admission diagnosis of acute ischemic stroke, and had an initial CT scan. Patients who had an MRI scan before receiving an admission diagnosis were not included. Propensity-score matching (per 24 baseline characteristics as well as receipt of intravenous [IV] tissue plasminogen activator [tPA] yes/no and endovascular treatment yes/no) was performed for 123 patients with additional MRI (CT+MRI) to 123 patients who did not have additional MRI (CT-only). The two outcomes of interest were: 1) disability at discharge per a poor modified Rankin scale score of 3-6, and 2) the composite endpoint of stroke or death occurring with 12 months after hospital discharge as determined through electronic medical records and online obituary postings. A noninferiority margin was derived from previous randomized clinical trials of ischemic stroke treatments.
Results:
Of the 246 subjects, median age was 68 years and 53% were men. Only two of the 26 covariates (coronary artery disease and chronic kidney disease) were unbalanced between the CT-only and CT+MRI groups after propensity score matching. Initial CT was positive for acute ischemia in 31.2% of the CT-only group and 19.5% of the CT+MRI group. Disability at discharge was more common in the CT+MRI group (48%) than in the CT-only group (42.3%). Stroke or death within 1 year after discharge occurred more often in patients in the CT+MRI group (19.5%) than in the CT-only group (12.5%).
Conclusions:
The results of this single-center retrospective study of acute ischemic stroke patients suggest that obtaining a CT scan alone results in noninferior patient outcomes when compared to obtaining both a CT and MRI.
Perspective:
Compared to CT+MRI patients, patients who did not undergo MRI in this study were much more likely to have received IV tPA (59% vs. 28.5%)* or endovascular treatment (23.1 vs. 6.5%). This is not surprising, as patients offered IV tPA and endovascular treatment typically have a clear embolic stroke diagnosis and additional imaging (MRI) is often not needed. Propensity-score matching may not have completely compensated for the impact of these interventions. For instance, patients offered acute treatment may have been subjectively more functional at baseline than patients not offered treatment.
Studies that rely on propensity-score matching often include sensitivity analyses as a “spot check” on the results. This study may have benefited from an analysis that excluded patients who received IV tPA and endovascular treatment to ensure that the direction of the results did not change.
These results are unlikely to change management. Stroke neurologists report the benefit of MRI in: 1) determining stroke etiology (Is the ischemia bilateral? Lacunar?); 2) deciding about when to start or restart anticoagulation (How large is the infarct, and is there petechial hemorrhaging in the infarct bed?); and 3) helping patients visualize the size, location, and possible effects of their new central nervous lesion, a question that frequently arises in post-stroke clinics.
*These are very high IV tPA treatment rates. A typical Comprehensive Stroke Center’s treatment rate is <20%.
Clinical Topics: Anticoagulation Management, Dyslipidemia, Noninvasive Imaging, Prevention, Vascular Medicine, Atherosclerotic Disease (CAD/PAD), Lipid Metabolism, Interventions and Imaging, Computed Tomography, Magnetic Resonance Imaging, Nuclear Imaging
Keywords: Anticoagulants, Coronary Artery Disease, Diagnostic Imaging, Electronic Health Records, Embolic Stroke, Endovascular Procedures, Ischemia, Ischemic Stroke, Magnetic Resonance Imaging, Neurologists, Patient Discharge, Renal Insufficiency, Chronic, Secondary Prevention, Stroke, Tissue Plasminogen Activator, Tomography, Tomography, X-Ray Computed
< Back to Listings
