Abstract WP77: Final Infarct Volume May Predict Functional Outcome after Mild, Nondisabling Ischemic Stroke: Results from the PRISMS Trial

Introduction: In patients with acute cerebral ischemia (ACI) within 3h of onset with mild deficit judged as nondisabling at presentation, up to 20% experience poor functional outcome at 90 days. We tested the hypothesis that the presence of infarction at 24h would independently predict worse outcome at 3 months, and also explored the role of final infarct volume (FIV) on outcomes. Methods: The phase 3b, randomized PRISMS trial compared intravenous alteplase to aspirin for mild (NIHSS 0-5), nondisabling stroke at <3 h from onset. In a prespecified analysis, we included patients with 24h MRI performed and excluded acute cerebral ischemia mimics. Central readers assessed Day 24h MRIs for acute infarct, chronic infarcts, and white matter hyperintensity (WMH) burden; FIV and WMH volumes were measured quantitatively using semi-automated software. The primary outcome was 3-month EQ5D scores, a measure of QOL, and the secondary outcome was modified Rankin Scale (mRS) scores, a measure of disability. Multivariable linear regression and proportional odds cumulative logit models were used to evaluate the association between (1) acute infarction presence and (2) FIV on the primary and secondary outcomes, respectively, adjusting for relevant covariates (Table 1). Results: Of 313 patients enrolled, 273 had a final diagnosis of ACI and 212/273 (77%) had 24h MRI scans. Acute infarcts were present In 109 (51%), and median FIV was 1.20 mL (IQR 0.57-2.50). No association between presence of acute infarcts and 3-month EQ5D (p=0.84) or mRS (p=0.17) scores was observed (Table 1A). No association between FIV and EQ5D scores (p=0.31) was observed as well, but FIV was strongly associated with mRS scores (OR 1.11, 95% CI 1.04-1.18, p<0.001), in adjusted models (Table 1B). Conclusions: Among patients presenting with mild, nondisabling ACI, FIV appears to be associated with worse functional outcome. Since this was a post-hoc analysis, this hypothesis requires further prospective study.