Biomechanical evaluation of physical examination tests for rotator cuff tears: a computer simulation study

Background: Numerous physical diagnostic shoulder tests have been established to determine the presence of rotator cuff tears and to identify the affected muscles. However, reported sensitivities and specificities of these tests vary strongly. The aim of this study was to identify diagnostic postures that are biomechanically most sensitive in identifying rotator cuff lesions and compensation mechanisms. Methods: A musculoskeletal modeling study investigating muscle activity in healthy shoulders as well as in shoulders with anterior, superior, and posterosuperior rotator cuff tear patterns, was conducted. Muscle moment arms and muscle synergism for the Lift-off and Bear Hug tests, Jobe and Full-can tests, and Infraspinatus and Hornblower tests were compared for healthy and pathological models. Results: In a healthy model the Lift-off test showed significantly higher subscapularis activity compared to the Bear Hug test (P < .001). Teres minor and infraspinatus activity were threefold and twofold higher, in the Hornblower than the Infraspinatus test, respectively. In superior tests, supraspinatus activity was more than twofold lower than lateral deltoid activity and synergistic activity increase was smallest (Δ 1%-3% in deltoid). Activity increase was highest in posterosuperior tests for the teres minor with 66.4% activity increase in the Infraspinatus test (P < .001) and 81.3% increase in the Hornblower test (P < .001). Conclusions: The Lift-off test was significantly more sensitive in detecting subscapularis tears and the Hornblower test was more effective in assessing teres minor integrity in posterosuperior tears. Both, Jobe and Full-can tests demonstrated low biomechanical sensitivity in the detection of superior rotator cuff tears.