Magnetic resonance imaging accuracy for the diagnosis of superior labrum anterior-posterior lesions in the community setting: eighty-three arthroscopically confirmed cases.

Magnetic resonance imaging (MRI) has been shown in the literature to have high accuracy for the diagnosis of superior labrum anterior-to-posterior (SLAP) lesions. However, these studies have been performed at specialized institutions and have not been evaluated for community settings. Our hypothesis was that MRI diagnosis of SLAP lesions in the community setting is not as accurate as suggested by previous literature. The MRI reports for 83 consecutive surgically confirmed type II SLAP lesions were evaluated. The findings regarding the presence of SLAP lesions were documented from the community radiologists' MRI readings. Two blinded, fellowship-trained musculoskeletal radiologists evaluated these same 83 MRIs, as well as 17 MRIs with surgically confirmed normal superior labra. The musculoskeletal radiologists also used a visual analog scale to subjectively grade the quality of each MRI. MRIs were performed at 14 different centers and were read by 28 different community radiologists. There were 37 noncontrast MRIs and 46 MRI arthrograms. The community radiologists accurately identified 51% of the SLAP lesions. The sensitivities for the community radiologists were significantly better for the MRI arthrograms than for the noncontrast MRIs (P = .0002). For the two musculoskeletal radiologists, the sensitivities were 60% and 67.5%, and the specificities were 71% and 76.5%. The accuracies for the musculoskeletal radiologists were 62% and 69%. One of the musculoskeletal radiologists had significant correlation between the visual analog score and MRI diagnostic accuracy (P = .0006). The sensitivity for one of the musculoskeletal radiologists was significantly greater than the sensitivity for the community radiologists (P = .0063). The accuracies between the other musculoskeletal radiologist and the community radiologists were not significantly different (P = .0577). In this community setting, MRI was not accurate for the diagnosis of SLAP lesions. Musculoskeletal radiologists were more accurate than the community radiologists; however, the musculoskeletal radiologists were not as accurate as previous literature might predict.