Use of multiparametric magnetic resonance imaging as a screening tool for the determination of acute ischemic stroke duration

OBJECTIVE: This study aimed to assess the usability of magnetic resonance imaging (MRI) parameters in the treatment of stroke patients whose symptom onset time is unknown. PATIENTS AND METHODS: We evaluated MRI of the patients whose stroke symptoms began within 12-hours. For quantitative analysis, fluid-attenuated-inversion-recovery (FLAIR) and diffusion-weighted-imaging (DWI) signal -intensity-ratios (SIR) of the lesions were computed. For qualitative analysis, 'mismatch' between visibility of lesion on DWI-FLAIR was evaluated. Patients were analyzed according to the first 4.5/6 hours of stroke onset time. RESULTS: There was a moderate (r=0.569, p<0.001) correlation between symptom MRI time and FLAIR SIR and a weak correlation with DWI SIR (r=0.355, p=0.001). A FLAIR SIR threshold of <= 1.18 for predicted symptom onset 4.5 hours increased specificity (0.77 vs. 0.74) and sensitivity (0.77 vs. 0.69) as compared with visual analysis. A FLAIR SIR threshold of <= 1.19 for predicted symptom onset 6 hours increased sensitivity (0.76 vs. 0.67) and equal specificity (0.75 vs. 0.75) as compared with visual analysis. CONCLUSIONS: In hyperacute ischemic stroke, lesion age can be determined more accurately by the FLAIR SIR analysis than visual analysis. In patients whose stroke onset time is unknown, the FLAIR SIR can be used as a biomarker in the management of stroke patients.