Physical examination and magnetic resonance imaging in the diagnosis of superior labrum anterior-posterior lesions of the shoulder: a sensitivity analysis.

PURPOSE: The overall purpose of our study was to examine the sensitivity of physical examination, magnetic resonance imaging (MRI), and magnetic resonance (MR) arthrogram for the identification of arthroscopically confirmed SLAP lesions of the shoulder.

METHODS

An analysis of 51 consecutive patients with arthroscopically confirmed SLAP lesions and no history of shoulder dislocation was performed. Before undergoing surgery, all patients underwent a standardized physical examination and had either an MRI and/or MR arthrogram performed. Sensitivity analysis was then performed on the results of both the physical examination maneuvers and the radiologic imaging compared to the arthroscopic findings at surgery.

RESULTS

The sensitivity of O'Brien's (active compression) test was 90%, whereas the Mayo (dynamic) shear was 80% and Jobe's relocation test was 76%. The sensitivity of a physical examination with any 1 of these 3 SLAP provocative tests being positive was 100%. Neer's sign (41%) and Hawkin's impingement tests (31%) each had low sensitivity for SLAP lesions. The sensitivity of MRI for SLAP lesions was 67% when interpreted by the performing surgeon, 53% when read by a radiologist. When the MR arthrograms were analyzed alone, the sensitivity was 72% (surgeon) and 50% (radiologist), respectively.

CONCLUSIONS

All 3 physical examination maneuvers traditionally considered provocative for SLAP pathology (O'Brien's, Mayo shear, and Jobe's relocation) were sensitive for the diagnosis of SLAP lesions. MRI and MR arthrogram imaging had lower sensitivity than these physical examination tests in diagnosing SLAP lesions. Patient history, demographics, and the surgeon's physical examination should remain central to the diagnosis of SLAP lesions.

LEVEL OF EVIDENCE: Level II, development of diagnostic criteria on basis of consecutive patients with universally applied gold standard.