Accuracy of abduction-external rotation MRA versus standard MRA in the diagnosis of intra-articular shoulder pathology.

The goal of this study was to compare the accuracy of abduction-external rotation magnetic resonance arthrography (ABERMRA) with standard MRA in the diagnosis of intra-articular shoulder pathology.One hundred three consecutive patients undergoing preoperative direct MRA and subsequent arthroscopic examination were included in the study. Seventy-eight patients underwent standard MRA and 25 underwent ABERMRA. Specialist-trained musculoskeletal radiologists reported all scans, and attending shoulder surgeons performed all arthroscopies. Arthroscopic assessment revealed 11 partial-thickness rotator cuff tears, 3 full-thickness tears, 64 labral lesions (48 soft tissue and 16 significant bony), and 17 superior labrum anterior-posterior (SLAP) tears. The sensitivity/specificity for standard MRA was 0.56/0.99 for partial-thickness rotator cuff tears, 1.00/1.00 for full-thickness rotator cuff tears, 0.75/0.91 for soft tissue labral tears, 0.58/1.00 for significant bony glenoid lesions, and 0.50/0.91 for SLAP tears. Abduction-external rotation magnetic resonance arthrography increased the sensitivity/specificity to 1.00/0.85 for soft tissue labral tears, 0.75/1.00 for significant bony glenoid lesions, and 1.00/1.00 for SLAP tears, although it missed 2 of 2 partial-thickness rotator cuff tears.This study suggests that standard MRA is a valuable investigation tool for instability, SLAP tears, and rotator cuff tears, although limitations exist. Additional ABERMRA sequences appear to improve the diagnostic accuracy of soft tissue anterior and posterior labral tears, SLAP tears, and significant bony glenoid lesions and should be routinely requested by shoulder surgeons when ordering MRAs to obtain the maximum benefit from this invasive investigation.