A pilot study to test the procedure for establishing whether the Orthopaedic Medicine shoulder examination is as accurate as Magnetic Resonance Imaging in diagnosing rotator cuff pathology
Authors: Jonathan Smith
References: Presented at SOM & BESS Conference, 2008 - Poster Prize
Study Design: Prospective, blinded comparison of the Orthopaedic Medicine (OM) examination and Magnetic Resonance Imaging (MRI) to arthroscopic findings of consecutive patients in a hospital setting.
Objective: To design a study and test its ability to determine whether the OM examination is as accurate as MRI at detecting Rotator Cuff (RC) pathology of the shoulder, using arthroscopy as the reference standard.
Background: RC disorders are the most common cause of shoulder pain. Despite this there have been very few prospective studies comparing clinical examination with MRI findings using arthroscopy as the reference standard, and none to date using the OM examination as a standardised protocol.
Methods and Measure: Twenty-two consecutive subjects with shoulder pain who had previously had an MRI scan were prospectively examined and subsequently underwent arthroscopy. The mean age of the subjects was 49 (range 16-75) and 13 were female and 9 male.
Results: When classifying a RC tear as a 'full-thickness tear', MRI had a sensitivity and specificity of 89% and 95% respectively and clinical examination a sensitivity and specificity of 33% and 96% respectively. For detecting the location of RC pathology, MRI had an overall sensitivity and specificity of 64% and 95% respectively and clinical examination an overall sensitivity and specificity of 41% and 94% respectively.
Conclusion: This pilot study was successful in that the design adopted avoided some of the limitations of previous research. The findings suggest that MRI is more accurate than the OM examination at detecting both the degree and location of RC pathology. The OM examination appears to have limited ability at ruling out RC pathology and suggestions based around anatomical studies of the RC have offered plausible explanations for these findings. Further research is needed to validate these claims but it would appear that MRI will continue to be an important diagnostic tool in the diagnosis of RC pathology.