Association of early National Institutes of Health Stroke Scale improvement with vessel recanalization and functional outcome after intravenous thrombolysis in ischemic stroke.

BACKGROUND AND
PURPOSE: Early neurological improvement (ENI) after thrombolytic therapy of acute stroke has been linked with recanalization and favorable outcome, although its definition shows considerable variation. We tested the ability of ENI, as defined in previous publications, to predict vessel recanalization and 3-month functional outcome after intravenous thrombolysis recorded in an extensive patient cohort in the Safe Implementation of Thrombolysis in Stroke-International Stroke Thrombolysis Register (SITS-ISTR).
METHODS: Of 21,534 patients registered between December 2002 and December 2008, 798 patients (3.7%) had CT- or MR angiography-documented baseline vessel occlusion and also angiography data at 22 to 36 hours post-treatment. ENI definitions assessed at 2 hours and 24 hours post-treatment were (1) National Institutes of Health Stroke Scale (NIHSS) score improvement ≥4 points from baseline; (2) NIHSS 0, 1, or improvement ≥8; (3) NIHSS ≤3 or improvement ≥10; (4) improvement by 20%; (5) 40% from baseline; or (6) NIHSS score 0 to 1. Receiver operating curve analysis and multiple logistic regression were performed to evaluate the association of ENI with vessel recanalization and favorable functional outcome (modified Rankin Scale score 0 to 2 at 3 months).
RESULTS: ENI at 2 hours had fair accuracy to diagnose recanalization as derived from receiver operating curve analysis. Definitions of improvement based on percent of NIHSS score change from baseline demonstrate better accuracy to diagnose recanalization at 2 hours and 24 hours than the definitions based on NIHSS cutoffs (the best performance at 2 hours was area under the curve 0.633, sensitivity 58%, specificity 69%, positive predictive value 68%, and negative predictive value 59% for 20% improvement; and area under the curve 0.692, sensitivity 69%, specificity 70%, positive predictive value 70%, and negative predictive value 62% for 40% improvement at 24 hours). ENI-predicted functional outcome with OR 2.8 to 6.0 independently from recanalization in the angiography cohort (n=695) and with OR of 6.9 to 9.7 in the whole cohort (n=18 181).
CONCLUSIONS: Early 20% neurological improvement at 2 hours was the best predictor of 3-month functional outcome and recanalization after thrombolysis, although fairly accurate, and may serve as a surrogate marker of recanalization if only imaging evaluation of vessel status is not available. If recanalization status is required after intravenous thrombolysis, vascular imaging is recommended despite ENI.