Reliability and Limits of Agreement of the Supraspinatus Muscle Anatomical Cross-Sectional Area Assessment by Ultrasonography

The purpose of this study was to evaluate the reliability and limits of agreement for assessment of the anatomical cross-sectional area (ACSA) of the supraspinatus muscle using B-mode ultrasonography. Sixteen participants were examined with two different protocols, on two different days. There were no statistically significant differences (p > 0.05) in ACSA values between days 1 and 2 or between protocols 1 and 2; the average intra-class correlation coefficient ranged from 0.93 to 0.96. The limits of agreement for supraspinatus ACSA were, in both protocols, about ± 1 cm2. Our findings revealed that both protocols had high reliability in distinguishing differences of about 1 cm2 between groups or after interventions and that ultrasonography can be used for experimental designs in which the expected changes in ACSA would be higher than 14%.