Background and purpose: We investigated whether the signal change on fluid-attenuated inversion recovery (FLAIR) can serve as a tissue clock that predicts the clinical outcome after endovascular thrombectomy (EVT), independently of the onset-to-admission time.
Methods: Consecutive patients with acute stroke treated with EVT between September 2014 and December 2018 were enrolled. Based on the parenchymal signal change on FLAIR, patients were classified into FLAIR-negative and FLAIR-positive groups. The clinical characteristics, imaging findings, EVT parameters, and the intracranial hemorrhage defined as Heidelberg Bleeding Classification ≥1c hemorrhage (parenchymal hemorrhage, intraventricular hemorrhage, subarachnoid hemorrhage, and/or subdural hemorrhage) were compared between the 2 groups. A modified Rankin Scale score 0 to 1 at 3 months was considered to represent a good outcome.
Results: Of the 227 patients with EVT during the study period, 140 patients (62%) were classified into the FLAIR-negative group and 87 (38%) were classified into the FLAIR-positive group. In the FLAIR-negative group, the patients were older (P=0.011), the onset-to-image time was shorter (P<0.001), the frequency of cardioembolic stroke was higher (P=0.006), and the rate of intravenous thrombolysis was higher (P<0.001) in comparison to the FLAIR-positive group. Although the rate of complete recanalization after EVT did not differ between the 2 groups (P=0.173), the frequency of both any-intracranial hemorrhage and Heidelberg Bleeding Classification ≥1c hemorrhage were higher in the FLAIR-positive group (P=0.004 and 0.011). At 3 months, the percentage of patients with a good outcome (FLAIR-negative, 41%; FLAIR-positive, 27%) was significantly related to the FLAIR signal change (P=0.047), while the onset-to-image time was not significant (P=0.271). A multivariate regression analysis showed that a FLAIR-negative status was independently associated with a good outcome (odds ratio, 2.10 [95% CI, 1.02–4.31], P=0.044).
Conclusions: A FLAIR-negative status may predict the clinical outcome more accurately than the onset-to-admission time, which may support the role of FLAIR as a tissue clock.
Keywords: hypertension; magnetic resonance imaging; perfusion; probability; thrombectomy.