Validity and Reliability of Neurological Scores in Mice Exposed to Middle Cerebral Artery Occlusion

Background and Purpose-The selection of appropriate neurological scores and tests is crucial for the evaluation of stroke consequences. The validity and reliability of neurological deficit scores and tests has repeatedly been questioned in ischemic stroke models in the past. Methods-In 198 male mice exposed to transient intraluminal middle cerebral artery occlusion, we examined the validity and reliability of 11 neurological scores (Bederson score 0-3, Bederson score 0-4, Bederson score 0-5, modified neurological severity [0-14], subjective overall impression [0-10], or simple neurological tests: grip test, latency to move body length test, pole test, wire hanging test, negative geotaxis test, and elevated body swing test) in the acute stroke phase, that is, after 24 hours. Combinations of neurological scores or tests for predicting infarct volume were statistically analyzed. Results-Infarct volume was left skewed (median [Q1-Q3], 78.4 [54.8-101.3] mm(3)). Among all tests, the Bederson (0-5; r=0.63, P<0.001), modified neurological severity (r=0.80, P<0.001), and subjective overall impression (r=-0.63, P<0.001 scores had the highest test validities, using infarct volume as external reference. Subjective overall impression had the best agreement between 5 raters (Kendall W=0.11, P<0.001). The Bederson (0-5) score discriminated infarct volume in mice with small (<= 50 mm(3); r=0.33, P=0.027) and large (>50 mm(3); r=0.48, P<0.001) brain infarcts, all other tests only in mice with large infarcts. Combining subjective overall impression with Bederson (0-5) score explained 47.6% of the variance of infarct volume. Conclusions-Despite their simplicity, the Bederson (0-5) score, modified neurological severity score, and subjective overall impression have reasonable validity and reliability in the acute stroke phase. The Bederson (0-5) score equally distinguishes infarct volume in small and large infarcts.