Abstract T P42: Diffusion-weighted Imaging-fluid-attenuated Inversion Recovery Mismatch and Fluid-attenuated Inversion Recovery Vascular Hyperintensities in Onset Time Estimation of Acute Ischemic Stroke

Abstract

Background and Purpose: The diffusion-weighted imaging (DWI)-fluid-attenuated inversion recovery (FLAIR) mismatch (DFM) is useful to estimate the onset time of stroke. The presence of FLAIR vascular hyperintensities (FVH) is associated with large diffusion-perfusion mismatch in patients with middle cerebral artery occlusion. We aimed to assess whether the combination of DFM and FVH was useful to predict patients ≤ 4.5 h after stroke onset. Methods: Consecutive patients with acute ischemic stroke who underwent 3.0T or 1.5T MRI including DWI and FLAIR within 12h after onset were registered. Of them, those with middle cerebral artery territory infarction were studied. More than two stroke neurologists judged whether there is DFM. We identified ischemic lesion corresponding to stroke symptom on DWI and then determined whether the FLAIR lesion positivity is negative, subtle (only slightly different from adjacent parenchyma) or evident (a clearly high signal). DFM was defined as the FLAIR negative or subtle corresponding to the DWI lesion. Patients were divided into two groups; the early group, those underwent MRI ≤ 4.5h after onset; and the late group, those underwent > 4.5h. Results: Of 129 patients (56 women, 72±12 years old) studied, 103 patients (45 women, 73±12 years old) and 26 patients (11 women, 70±11 years old) were assigned to the early and the late groups, respectively. Initial NIHSS score (median 7 [IQR 2-15] vs. 3.5 [1-6], p=0.032) was higher, and DFM (67% vs. 35%, p=0.003) and FVH (49% vs. 23%, p=0.019) were more frequently observed in the early group than in the late group. On multivariate analyses adjusted for confounders, DFM (odds ratio 3.35, 95% confidence interval 1.29-9.27; p=0.013) was independently associated with the early group. Patients in the early group were detected with a sensitivity of 0.67, specificity of 0.65, a positive predictive value of 0.88, and a negative predictive value of 0.33 using the presence of DFM and with 0.37, 0.92, 0.95, and 0.27, respectively, using the combination of DFM and FVH. Conclusions: DFM is useful to detect acute ischemic stroke patients ≤ 4.5h of onset with the acceptable sensitivity and specificity. Furthermore, patients with both DFM and FVH are very likely to be ≤ 4.5h of onset, although the sensitivity is low.