Written by Alexander Turner and Ketan Patel
Spoon Feed
In this substudy of the Canadian TIA Score cohort, using the score plus MRI better predicted subsequent 90-day stroke, particularly in the medium-risk group.
CT misses it, MRI gets it – finding the stroke before it strikes back
The Canadian TIA Score does a good job of risk stratifying TIA patients into low, medium, and high-risk groups. While management in low and high-risk groups is somewhat intuitive and clear, the majority fall in the medium-risk group, without clarity and consensus on next steps of management.
This study evaluated the predictive value of MRI in TIA/minor stroke patients with negative CT scans for subsequent stroke risk. Conducted across 13 Canadian emergency departments, the cohort study (N=1,048) found that 31.5% had acute infarction on MRI. Stroke rates at 90 days were significantly higher in MRI-positive patients (low-risk: 10.0%, medium-risk: 22.3%, high-risk: 24.7%) compared to MRI-negative patients (0.9%, 1.3%, 4.9%). MRI enhances stroke risk stratification, particularly for medium-risk patients.
This study supports rapid MRI for the medium-risk population, while potentially deferring low-risk patients for 7 days. High-risk patients should be medically optimized regardless of MRI findings.
How will this change my practice?
TIAs are a huge gray area when it comes to admission to the hospital and subsequent management once stabilized and seen in the ED. I am willing to apply a Canadian TIA score more consistently and advocate for the patients in the medium-risk group to have a more rapid MRI brain ordered to prevent some of the 90 day poor outcomes in this group.
Editor’s note: Seems like TIA patients needs MRI. I could envision shared decision making for patients with a low-risk score, and arrangement of rapid outpatient MRI. ~Clay Smith
Source
The value of MRI in transient ischemic attack/minor stroke following a negative CT for predicting subsequent stroke. CJEM. 2025 Mar;27(3):215-220. doi: 10.1007/s43678-024-00853-7. Epub 2025 Feb 4. PMID: 39903424
