Abstract T P46: Final Infarct Volume Discriminates Outcome In Mild Strokes

Abstract

Introduction: Final infarct volume (FIV) is a well-established predictor of outcomes in moderate and severe stroke. However, our knowledge of how FIV predicts disability in the mild stroke population is very limited. Our objective was to determine if FIV could differentiate good versus poor outcome after mild stroke. We hypothesized that smaller FIV will be associated with favorable clinical outcome. Methods: We used a retrospective, multicenter registry for consecutive patients who presented with mild stroke (NIHSS ≤ 5) within 24 hours of stroke onset. The imaging data included baseline head CT, noninvasive vascular imaging and follow up imaging (24-72 hour MR/3-5 day CT). Baseline ASPECTS, proximal arterial occlusion (PAO), collateral scores and FIV (using MIPAV) were assessed by core lab. Clinical data included age, sex, baseline NIHSS and 90 day modified Rankin Scale (mRS). Logistic regression was used to evaluate associations between favorable outcome (mRS 0-1) and imaging and clinical data, and to determine a data-driven cut-point for FIV based on the maximal sensitivity and specificity of the receiver operating characteristic curve. Results: Among 90 patients with mild stroke, 3 who received intravenous thrombolysis and 22 who did not have follow-up imaging were excluded, leaving 65 patients (mean age 86 years, 38% female, median NIHSS score 4) in whom FIV calculations were performed. An optimal FIV cut-point at 20 cc was identified, for differentiating between favorable and poor outcomes (area under curve 0.73, 95% CI: 0.58-0.88). Of the 45 patients with FIV <20 cc, 37 (82%) had a favorable outcome compared to 5 out of 14 (36%) with FIV ≥ 20 cc (P<.01). In the multivariable model, FIV ≥ 20cc remained strongly associated with poor outcome (adjusted OR, 0.11; 95% CI, 0.02- 0.50, P<.01), irrespective of age, gender, stroke severity, ASPECTS and PAO. A higher collateral score was also found to be associated with favorable outcome (adjusted OR, 2.43; 95% CI, 1.12- 5.27, P=0.02). Conclusions: A final infarct volume cut point of 20 cc was found to best differentiate between the likelihood of good versus poor outcome in patients with mild stroke. Further validation of FIV as a surrogate marker is essential to improve risk prediction in mild stroke patients.