Based on WAKE-UP (2018), THAWS (2020), and smaller single-center trails, the concept of FLAIR signal change as a surrogate timekeeper in hyperacute stroke has shown both safety and efficacy for intravenous thrombolysis. Now, in the age of extended window endovascular thrombectomy (EVT), predictors of good clinical outcome beyond time, age and medical risk factors become prudent for patient selection. In this study, Aoki et al. hypothesize if FLAIR signal change can predict clinical outcome after EVT.
From a prospective registry of 324 consecutive EVT patients presenting with acute ischemic stroke, 227 were retrospectively enrolled from September 2014 to December 2018. Those with premorbid mRS score 0 to 1 with available FLAIR imaging were included; patients with contraindications to MRI were excluded. FLAIR positivity was defined by new hyper-intense signal at DWI-positive lesion site; subtle changes were measured using contralateral signal intensity ratio of 1.2. The median age of patients was 74, NIHSS 15, and symptoms onset to imaging 155 minutes. Ischemic core volume and NIHSS were not significantly different in timing to FLAIR imaging from less than 2 hours to greater than 12 hours.
When compared to 87 FLAIR-positive patients, the 140 FLAIR-negative patients were, on average, older and had shorter onset to imaging, smaller ischemic core, and a higher rate of intravenous thrombolysis. While EVT reperfusion did not differ between groups (P=0.192), the rates of intracerebral hemorrhage were significantly lower with FLAIR-negative patients (39% versus 59%, P=0.011). Using TOAST criteria, FLAIR-negative patients had a higher rate of cardioembolic stroke and lower rate of large-artery atherosclerosis. The authors found FLAIR-negative patients were associated with good outcomes defined as mRS 0-1 at 3 months (41% versus 27%, P=0.047).
While limited by a single-center retrospective design, the authors present a compelling case for FLAIR signal changes as a “tissue clock” for EVT candidates. In addition, they provide insight into the experience of other investigators whose patients, despite EVT, have poor functional outcomes. Access to hyperacute MRI may play a larger role in triage in the future; however, randomized multi-center studies are still needed, particularly when evaluating rates of symptomatic hemorrhage. As FLAIR signal changes morph from a surrogate of time to a measure of viable tissue for reperfusion, when will the question of last known normal (LKN) become last known FLAIR negative (LKFN)?
