A multicenter study comparing the accuracy of MRI to arthroscopy for the diagnosis of glenohumeral joint pathologies

Abstract

Background: The number of patients with shoulder pathology who seek medical treatment is on the increase. It affects approximately18-26% of adult population. The symptoms can sometimes be debilitating affecting not only the person’s occupation but also activities of daily living. Among the diagnostic modalities for glenohumeral pathologies, clinical examination remains the key but MRI and arthroscopy are more accurate and can play a complimentary role. The MRI is highly sensitive, specific, non-invasive with no radiation. On the contrary, shoulder arthroscopy is considered to be the “Gold Standard” for diagnosis and treatment of glenohumeral joint pathologies. It is accurate and less invasiveness compared to open shoulder surgery. However, it is expensive. Objective: The aim was to determine the degree of accuracy of MRI compared to arthroscopy in the diagnosis of glenohumeral joint pathologies. Design: A multicenter prospective consecutive cross-sectional study. The sites included: The Nairobi, Aga Khan University, Mater, Kikuyu, Kijabe, Coptic and MP Shah hospitals. Methodology: The patients with traumatic soft tissue injury or degenerative syndromes of the glenohumeral joint were recruited over a period of 8 months. Clinical examinations were done followed by MRI and then arthroscopy. The SPSS version 25 computer software was used to code the collected data. The final results were presented in charts, tables and graphical forms. The Sensitivity, Specificity, PPV, and NPV were calculated to determine the accuracy of the MRI and clinical examination. This were compared to the findings of arthroscopy. For the categorical variables, chisquare test was used. The P value < 0.05 was considered significant. Results: The sample size was 74 participants. The age range was 16 to 73 years with an average of 48 years. The male to female ratio was 1:1. Majority 46 (62%) of the patients had joint pathology on the right side while 28(37.8%) was on the left. This might be related to hand dominance. The three modalities of investigations found the frequencies of the glenohumeral joint pathologies are as follows. The Rotator Cuff Tears ranged from 35 to 42 (47.3% to 56.8%), Subacromial Impingement Syndrome ranged from 21-24 (28.4 % to 32.4%), and Bankart lesions ranged from 9-10 (12.2 % to 13.5%). The result revealed a strong positive relationship between MRI and arthroscopic finding for Rotator Cuff Tear (r = 0.663, p<0.05), Subacromial Impingement Syndrome (r = 0.652, p<0.05) and Bankart lesion (r = 0.699). However, the clinical examination showed a moderate positive relationship for Rotator Cuff Tear (r =0.46, p<0.05) and Subacromial Impingement Syndrome (r = 0.445, p<0.05). The sensitivity for MRI ranges from 0.7 for Bankart lesions to 0.914 for Rotator Cuff Tears. Furthermore, the positive predictive value was 0.762 (76%) for Rotator cuff tear and 0.8 (80%) for Subacromial Impingement Syndrome. Conclusion: The study revealed a significant correlation between clinical examination, MRI and arthroscopy for the diagnosis of glenohumeral pathologies. Both MRI and clinical examination are complimentary to each other. Consequently, in low income countries, arthroscopy can be done after thorough clinical examination without preliminary MRI in resources limited situations based on the resolution of the surgeon