Abstract W P26: Itemized NIHSS Subsets Predict MRI Positive Strokes In Patients With Mild Deficits

Abstract

Background: While imaging is useful in confirming the diagnosis of ischemic stroke, negative diffusion weighted imaging (DWI) is reported in up to 10% of patients. We aim to (1) identify predictors of MRI-positive stroke from the itemized NIHSS, and (2) to correlate subsets with infarct volume. Methods: Data were derived from the Stroke Warning Information and Faster Treatment study from 2006 to 2009 among patients with mild deficits (NIHSS 0-5) and attending physician final diagnosis of stroke. Using Medical Image Processing, Analysis, and Visualization (MIPAV, NIH) (Version 7.1.1), we calculated lesion volume (cm3) from DWI sequence. Univariate models studied the association between itemized NIHSS subsets, including cortical deficits (visual field cut, aphasia, or neglect), and presence of DWI hyperintensity and lesion volume. Multivariable regression assessed factors predicting DWI-positive strokes; p<0.05 was considered significant. Results: Of 611 patients with a discharge diagnosis of stroke, 498 underwent MRI and 29.5% were DWI negative. On multivariate analysis, predictors of a positive DWI were NIHSS score of 3-5 (OR= 2.5, 95%CI:1.1-5.5), motor deficits (OR= 1.9, 95%CI:1.1-3.4), and ataxia (OR=3.0, CI:1.5-6.1). All patients with neglect and visual deficits were DWI positive (table). The mean lesion volume in cm3 was larger in patients with NIHSS 3-5 vs. NHSS 0-2 (49.0 vs. 17.3, p=0.002), cortical deficits (112.6 vs. 22.9, p<0.001), neglect (236.6 vs. 29.7, p<0.001), and visual deficits (245.7 vs. 26.4, p<0.001). Other subsets showed no differences. Conclusion: NIHSS score subsets predict DWI positivity and lesion volume in mild strokes. The presence of neglect or visual field deficit on the NIHSS subsets is highly likely to have an MRI correlate even in patients with low NIHSS.