Physical examination, magnetic resonance imaging and the arthroscopy of the shoulder: a correlation of diagnostic tests

Abstract

Background: Shoulder pathology is the third cause of consultation for musculoskeletal disorders; it is a cause of disability and economic losses. The gold standard for its diagnosis and treatment is arthroscopy. However, arthroscopy is an invasive diagnostic technique. Therefore, clinical and imaging methods may be useful as diagnostic techniques before considering arthroscopy. Nevertheless, these non-surgical diagnostic methods may present a poor correlation in diagnostic accuracy when compared to the arthroscopic diagnostic method in the context of shoulder pathology. Methods: A retrospective study including 123 patients who presented shoulder complains, underwent physical examination and a shoulder magnetic resonance imaging (MRI) study prior to a shoulder arthroscopy. A clinical correlation was made using the kappa index, sensitivity and specificity of the clinical and imaging tests were also calculated. Results: The correlation between arthroscopy and the clinical assessment showed a kappa index of 0.25 for rotator cuff injuries, 0.41 biceps pathology, 0.51 sub acromial impingement, 0.51 acromioclavicular arthrosis, 0.28 for SLAP injury. MRI diagnostic method showed a kappa correlation of 0.24 for rotator cuff injuries, 0.59 biceps pathology, 0.39 sub acromial impingement, 0.46 acromioclavicular arthrosis and 0.34 for superior labral anterior posterior (SLAP) injury. Conclusions: After comparing arthroscopy with clinical assessment and the MRI diagnostic method we found that clinical assessment is useful to diagnose biceps pathologies, subacromial impingement, and acromioclavicular arthrosis. On the other hand, MRI studies and diagnosis made by a radiologist showed to be a valuable tool for the diagnosis of biceps pathology and acromioclavicular arthrosis.