Evaluation of the prognostic value of the anatomical characteristics of the bony structures in the shoulder in bursal-sided partial-thickness rotator cuff tears.

Abstract

In recent studies, individual scapular anatomy has been found to be related to degenerative full-thickness rotator cuff tears. However, research on the relationship between the anatomical characteristics of shoulder radiographs and bursal-sided partial-thickness rotator cuff tears (PTRCTs) is limited, and the risk factors for this pathology still need to be determined. The bursal-sided PTRCTs group included 102 patients without a history of shoulder trauma who underwent arthroscopy between January 2021 and October 2022. A total of 102 demographically matched outpatients with intact rotator cuffs were selected as the control group. Radiographs were used to measure the lateral acromial angle (LAA), critical shoulder angle (CSA), greater tuberosity angle (GTA), β-angle, acromion index (AI), acromiohumeral distance (AHD), acromial tilt (AT), acromial slope (AS), acromial type, and acromial spur by two independent observers. Multivariate analyses of these data were used to identify potential risk factors for bursal-sided PTRCTs. Receiver operating characteristic (ROC) analysis was performed to assess the sensitivity and specificity of CSA, GTA, and AI for this type of pathology. The β-angle, AHD, AS and acromion type showed no difference between bursal-sided PTRCTs and controls (p = 0.009, 0.200, 0.747 and 0.078, respectively). CSA, GTA and AI were significantly higher in bursal-sided PTRCTs (p < 0.001). LAA, β-angle and AT were significantly lower in bursal-sided PTRCTs. Multivariate logistic regression analysis demonstrated significant correlations between the acromial spur (p = 0.024), GTA (p = 0.004), CSA (p = 0.003) and AI (p = 0.048) and bursal-sided PTRCTs. The areas under the ROC curves for AI, CSA, and GTA were 0.655 (95% CI 0.580-0.729), 0.714 (95% CI 0.644-0.784), and 0.695 (95% CI 0.622-0.767), respectively. Acromial spur, GTA, CSA, and AI were independent risk factors for bursal-sided PTRCTs. Furthermore, CSA was the most powerful predictor of bursal-sided PTRCTs compared to GTA and AI.