Reliability and diagnostic accuracy of history and physical examination for diagnosing glenoid labral tears

Abstract

Background: Glenoid labral tears provide a diagnostic challenge.

Hypothesis: Combinations of items in the patient history and physical examination will provide stronger diagnostic accuracy to suggest the presence or absence of glenoid labral tear than will individual items.

Study design: Cohort study (diagnosis); Level of evidence, 1.

Methods: History and examination findings in patients with shoulder pain (N = 55) were compared with arthroscopic findings to determine diagnostic accuracy and intertester reliability.

Results: The intertester reliability of the crank, anterior slide, and active compression tests was 0.20 to 0.24. A combined history of popping or catching and positive crank or anterior slide results yielded specificities of 0.91 and 1.00 and positive likelihood ratios of 3.0 and infinity, respectively. A positive anterior slide result combined with either a positive active compression or crank result yielded specificities of 0.91 and positive likelihood ratio of 2.75 and 3.75, respectively. Requiring only a single positive finding in the combination of popping or catching and the anterior slide or crank yielded sensitivities of 0.82 and 0.89 and negative likelihood ratios of 0.31 and 0.33, respectively.

Conclusion: The diagnostic accuracy of individual tests in previous studies is quite variable, which may be explained in part by the modest reliability of these tests. The combination of popping or catching with a positive crank or anterior slide result or a positive anterior slide result with a positive active compression or crank test result suggests the presence of a labral tear. The combined absence of popping or catching and a negative anterior slide or crank result suggests the absence of a labral tear.