Prediction of the anterior shoulder pain source by detecting indirect signs for partial articular subscapularis tendon tears through conventional magnetic resonance imaging.

PURPOSE

To evaluate the diagnostic efficacy of indirect signs for proximal articular-positioned, partial (< 50%), subscapularis tendon tears (facet 1 tears) via conventional magnetic resonance imaging (MRI).

METHODS

A retrospective study was conducted on 67 patients of Yoo's type 1 or 2A tears. Forty-five arthroscopic subacromial decompression and acromioclavicular resection cases served as controls. Indirect signs indicating a facet 1 tear included small defects, superior subscapularis recess (SSR), long head of the biceps (LHBT) configurations, bone edema or cyst formation on lesser tuberosity (LTBEC), and fatty infiltration of subscapularis muscle. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were checked.

RESULTS

SSR was the most sensitive sign (90%). The sensitivities and NPV of LHBT configurations and LTBEC were low (sensitivity: 42.9% and 17.9%, NPV: 56.4% and 44.4%, respectively). The specificities of all indirect signs were relatively high (> 90%). The Chi-squared test and multinomial logistic regression confirmed the significance of small defects, SSRs, and fatty infiltrations for facet 1 tears (p ≤ 0.014). The combined sensitivity and specificity were up to 97.7% and 92.3%, respectively, in the presence of either a small defect or an SSR.

CONCLUSIONS

Conventional MRI alone can detect facet 1 tears through indirect signs (small defects, SSR, and fatty infiltrations of the subscapularis muscle), predicting unspecified anterior shoulder pain due to concealed biceps instability, and facilitating preoperative diagnosis for a facet 1 tear.

LEVEL OF EVIDENCE

III.