Background and Purpose:
It is well established that earlier treatment times are associated with better outcomes in acute stroke patients receiving thrombolysis. There is also an association between time from stroke onset and lesion visibility on FLAIR MRI. We hypothesized that lesion visibility on FLAIR, independent of time, may be a predictor of outcome in stroke patients with known onset.
Methods:
We analyzed data from acute ischemic stroke patients presenting over the last 10 years who were screened with MRI and treated with IV thrombolysis within 4.5 hours from known onset. Three independent readers assessed whether acute ischemic lesions seen on Diffusion Weighted Imaging were also FLAIR-positive based on visual inspection. Multivariable regression analysis was used to obtain an adjusted odds ratio of favorable clinical and radiological outcomes based on FLAIR-positivity.
Results:
Of 310 stroke patients, 24% had lesion visibility on initial FLAIR MRI. The interrater agreement for the FLAIR-positive assessment was 84% (κ=0.604, 95% CI 0.557-0.652). Patients with FLAIR-positive lesions were younger (67 vs 73 years, p=0.028), had more right hemispheric strokes (57% vs 42%, p=0.018), were imaged later (127 vs 104 minutes, p=0.010), had more frequent blood-brain barrier disruption (44% vs 26%, p=0.004), less frequent early neurologic improvement (30% vs 58%, p<0.001), and less frequent favorable 90-day functional outcome (49% vs 63%, p=0.039). Following multivariable logistic regression, older age, greater NIH Stroke Scale, lesion visibility on FLAIR, but not time-from-onset, were independently associated with less favorable outcome.
Conclusions:
FLAIR-positive acute ischemic stroke within 4.5 hours of known onset was associated with less favorable 90-day outcome after IV thrombolysis. When compared with time, lesion visibility on FLAIR was more strongly associated with outcome.
Automated quantitative MRI volumetry reports support diagnostic interpretation in dementia: a multi-rater, clinical accuracy study
