Validity and reliability of nerve and muscle ultrasound

Abstract

INTRODUCTION: Nerve and muscle ultrasound has been studied in several conditions, but validity and reliability have not been assessed systematically. METHODS: Nerve cross-sectional area and muscle thickness were measured ultrasonographically at several sites in 4 cadavers, which were then dissected, and actual measurements were obtained. To assess intrarater and interrater reliability, between 3 and 5 ultrasonographers, with varying experience levels, made repeated measurements on healthy volunteers. RESULTS: Correlation coefficients for nerve and muscle validity were >0.968 (P < 0.001), and for intrarater reliability were >0.901 (P < 0.001) for still and real-time images. Correlation coefficients for interrater reliability were more varied, but for still images they were all significant at the P < 0.001 (0.542-0.998) level, and for real-time images they were significant at the P < 0.05 level for half the sites (0.243-0.981). CONCLUSION: Overall, nerve and muscle ultrasound is a valid and reliable diagnostic imaging technique.