Comparison of the NIH Stroke Scale and the Modified NIH Stroke Scale by Classification and Location of Cerebral Infarction

Bu, Seon Hee; Yoon, Sung Sang; Chang, Hye Jung; Kwon, Young Dae; Park, Key Chung; Ahn, Tae Beom; Chang, Dae Il; Chung, Kyung Cheon

Abstract

BACKGROUND: The National Institutes of Health Stroke Scale (NIHSS) is regarded as a valid and reliable tool to measure the severity of neurological deficits in acute stroke, but has been criticized for its complexity and variability. Therefore, the modified NIHSS (mNIHSS) was developed, eliminating redundant and less reliable items from the full version of the NIHSS. The aim of the present study was to evaluate the validity of the mNIHSS according to the subtypes of stroke and the location of affected arterial territories.
METHODS
The severity of stroke in 155 patients with acute cerebral infarction was measured. Each patient was evaluated by two neurologists using both the NIHSS and mNIHSS, and the stroke subtype was determined according to the TOAST classification. The vascular territory of lesion was classified into an anterior and posterior circulation. The criterion-related validity was evaluated by the Pearson Correlation Coefficient between the NIHSS and mNIHSS scores.
RESULTS
When considering the NIHSS scores as the gold criteria, the Pearson correlation coefficients of the mNIHSS were 0.96 in the subtype of large artery atherosclerosis, 0.91 in small vessel occlusion, 0.98 in cardioembolism, and 0.99 in undetermined etiology. On the other hand, the correlation coefficient was 0.98 in patients with an anterior circulation infarction, and was 0.94 in patients with a posterior circulation infarction.
CONCLUSIONS
The criterion-related validity of the mNIHSS scoring system was very high in general. However, the correlations were relatively low in patients with the TOAST subtype of small vessel occlusion and also a posterior circulation infarction.