NIHSS Score and Arteriographic Findings in Acute Ischemic Stroke

BACKGROUND AND PURPOSE: To test the hypothesis that the National Institutes of Health Stroke Scale (NIHSS) score is associated with the findings of arteriography performed within the first hours after ischemic stroke. METHODS: We analyzed NIHSS scores on hospital admission and clinical and arteriographic findings of 226 consecutive patients (94 women, 132 men; mean age 62+/-12 years) who underwent arteriography within 6 hours of symptom onset in carotid stroke and within 12 hours in vertebrobasilar stroke. RESULTS: From stroke onset to hospital admission, 155+/-97 minutes elapsed, and from stroke onset to arteriography 245+/-100 minutes elapsed. Median NIHSS was 14 (range 3 to 38), and scores differed depending on the arteriographic findings (P<0.001). NIHSS scores in basilar, internal carotid, and middle cerebral artery M1 and M2 segment occlusions (central occlusions) were higher than in more peripherally located, nonvisible, or absent occlusions. Patients with NIHSS scores > or =10 had positive predictive values (PPVs) to show arterial occlusions in 97% of carotid and 96% of vertebrobasilar strokes. With an NIHSS score of > or =12, PPV to find a central occlusion was 91%. In a multivariate analysis, NIHSS subitems such as "level of consciousness questions," "gaze," "motor leg," and "neglect" were predictors of central occlusions. CONCLUSIONS: There is a significant association of NIHSS scores and the presence and location of a vessel occlusion. With an NIHSS score > or =10, a vessel occlusion will likely be seen on arteriography, and with a score > or =12, its location will probably be central.