Measurement of shoulder abduction angles in dogs: an ex vivo study of accuracy and repeatability

OBJECTIVE: The aim of this study was to determine the accuracy and repeatability of the shoulder abduction test and to assess the effect of transection of the medial shoulder support structures in canine cadavers. MATERIALS AND METHODS: The shoulder abduction angle was measured by three separate observers, both with the shoulder extended and at a neutral angle. Shoulder abduction was then measured, using craniocaudal fluoroscopic images. Arthroscopy was performed in all shoulder joints, with the medial support structures transected in one shoulder of each dog. The three observers again measured shoulder abduction angles in all dogs. Shoulder abduction was measured again using fluoroscopy. Accuracy and repeatability of the abduction test were assessed using linear mixed models. RESULTS: All three observers had different measured abduction angles when compared with fluoroscopy ( < 0.01); however, the experienced surgeon had an error of only 2.9°. Inter-observer repeatability was poor, with all three observers having different abduction measurements ( < 0.001). Intra-observer repeatability, however, indicated no differences on repeated measurements ( = 0.26). Placing the shoulder at a neutral standing angle, and transection of support structures caused an average increase in abduction by 8.2° ( < 0.001) and 4.4° respectively. CONCLUSION: Significant variation exists between observers performing this test, increased accuracy seen in the more experienced observer. Shoulder flexion angle can significantly affect measured abduction angles.