Supplementary Material for: Validation of FLAIR Hyperintense Lesions as Imaging Biomarkers to Predict the Outcome of Acute Stroke after Intra-Arterial Thrombolysis following Intravenous Tissue Plasminogen Activator

<b><i>Background:</i></b> Intravenous tissue plasminogen activator (tPA) given within 4.5 h of symptom onset is accepted as the standard treatment of ischemic stroke. Persistent occlusion of cerebral arteries despite intravenous thrombolysis and unremitting neurologic deficits lead us to consider additional intra-arterial approaches. The aim of this study was to elucidate the potential of fluid-attenuated inversion recovery (FLAIR) MRI performed during or immediately after intravenous thrombolysis for predicting clinical outcomes of subsequent intra-arterial thrombolysis. <b><i>Methods:</i></b> With a prospective stroke registry database of patients hospitalized in our institution from January 2004 to February 2010, we identified ischemic stroke patients with the following conditions: (1) presentation within 2.5 h of onset, (2) treated with intravenous tPA based on brain CT, (3) persistent occlusion on subsequent MRI/MR angiography, including a FLAIR sequence, and (4) eventually treated with intra-arterial thrombolysis. Demographic, clinical and laboratory findings including initial National Institutes of Health Stroke Scale (NIHSS), follow-up NIHSS at the 7th day or discharge, modified Rankin scale (mRS) score at 3 months, and symptomatic hemorrhagic transformation were captured. FLAIR images were reviewed by 2 investigators blinded to clinical information independently and dichotomized into the absence and presence of FLAIR change within the diffusion-restriction lesions. <b><i>Results:</i></b> Of the 57 patients who met these conditions, FLAIR-hyperintense lesions (FHL) were observed in 32 (56.1%). The FHL-negative group was 69.1 ± 12.1 years old on average and the FHL-positive group 67.3 ± 11.0 years old. In both groups, hypertension was the most common vascular risk factor, cardioembolic stroke was the most common subtype, and distal middle cerebral artery was the most common site of occlusion. The incidence of symptomatic hemorrhagic transformation was 4.0% in the FHL-negative group and 9.4% in the FHL-positive group (p = 0.62). NIHSS scores of 0-1 on the 7th day of hospitalization or at discharge were observed in 36% of the FHL-negative group and in 9.4% of the FHL-positive group; mRS scores of 0-1 at 3 months was 32% in the FHL-negative group and 21% in the FHL-positive group. An ordinal logistic regression analysis showed that the presence of FHL was associated with higher 7-day NIHSS scores (adjusted for relevant covariates) but not with higher 3-month mRS scores. <b><i>Conclusions:</i></b> This study suggests that the FHL might be used as imaging biomarker to predict outcomes for additional intra-arterial thrombolysis in patients treated with intravenous tPA.