RESUMO
Shoulder pain is a common musculoskeletal pain in the general population and results in significant disability, quality of life impairment and financial burden to the health care system. This cross-sectional study was carried out among purposively selected 61 adult patients with shoulder pain and or limited range of motion in the Department of Radiology and Imaging, Mymensingh Medical College Hospital, Mymensingh, Bangladesh from September 2018 to August 2020 to determine the usefulness of USG in detection of common shoulder abnormalities, as an initial imaging modality using MRI as reference standard. The majority of the patients 25(40.98%) were in age group of 51-60 years with mean age of 52.98±10.85 years. In the dectection of rotator cuff pathology, the overall sensitivity, specificity and accuracy of USG for any complete tear were 100.0% each, for any partial tear were 79.71%, 96.57% and 91.80%, for any rotator cuff tear were 83.33%, 96.25% and 91.80%, and for any tendinosis were 90.48%, 99.37% and 96.31% respectively. The sensitivity, specificity and accuracy of USG were 88.23%, 92.59% & 90.16% for long head of biceps tendon sheath effusion, 100.0% each for long head of biceps tendon dislocation, 71.11%, 87.50% and 75.41% for glenohumeral joint effusion, and 58.06%, 96.67% and 77.05% for bursal effusion respectively. From this study, it is concluded that high resolution USG showed high sensitivity, specificity and accuracy compared to MRI in detecting common shoulder abnormalities, and could be considered as the first line imaging modality in the evaluation of shoulder pain.