High performance of critical shoulder angle for diagnosing rotator cuff tears on radiographs.

PURPOSE

Previous studies have reported that a high critical shoulder angle (CSA) is associated with rotator cuff tears (RCTs). However, the available evidence concerning the strength of the association of these parameters with the pathogenesis of RCTs is conflicting. The aim in the present meta-analysis was to assess the diagnostic performance of CSA for detecting RCTs.

METHODS

The PubMed and EMBASE databases were searched for diagnostic accuracy studies of CSA for detecting RCT. Quality of the included studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. We calculated the pooled diagnostic accuracy of CSA, including diagnostic odd ratios (DORs) with 95% confidence intervals (CIs). Bivariate random-effect modeling with forest plots and hierarchical summary receiver operating characteristic curves was performed to evaluate the pooled sensitivity and specificity of CSA. For heterogeneity exploration, we performed meta-regression analyses.

RESULTS

Eight studies including 902 patients (460 patients with RCT and 442 control patients) were included. According to DOR, CSA was an informative feature for RCT (DOR 8; 95% CI 4-16). The overall diagnostic performance of CSA for RCT was good (sensitivity 71% [95% CI 61-80%]; specificity, 77% [95% CI 65-86%]). Meta-regression analyses revealed that the sensitivity of CSA could be higher for differentiating full-thickness RCTs and normal patients. Furthermore, the specificity of CSA could be higher using a cut-off value of 35°.

CONCLUSION

CSA on plain radiographs has good diagnostic performance for RCTs. A cut-off value of 35° is recommended for more accurate diagnosis of full-thickness RCTs. Measuring CSA on plain radiographs may provide clinicians with a readily available and reliable method for detecting RCT in daily practice.

LEVEL OF EVIDENCE

Diagnostic studies, Level III.