The diagnostic value of direct CT arthrography using MDCT in the evaluation of acetabular labral tear: with arthroscopic correlation

The purpose of this study was first, to determine the sensitivity, specificity, and accuracy of MDCT arthrography (CTA) for the diagnosis of acetabular labral tear and sulcus; second, to correlate tear types using the Lage classification system on CTA compared with the arthroscopic classification; and third, to correlate CTA localization with arthroscopic localization. Direct CTA was performed using 16- or 64-slice MDCT in 126 hips (124 patients) who had chronic groin pain and positive impingement test. Images were reviewed and evaluated by two experienced musculoskeletal radiologists preoperatively. CTA findings were compared with arthroscopic findings in 58 hips (56 patients) under consensus by two orthopedic surgeons. Forty-one of the 58 hips were diagnosed as labral tears on CT arthrography. Forty-three of the 58 hips were shown to have a labral tear on arthroscopy. Sensitivity, specificity, and accuracy for detecting labral tear and sulcus by CTA were 90.7%, 86.7%, and 89.7%, and 93.8%, 97.6% and 96.6% respectively for observer 1, and 90.7% and 80.0%, 87.9% and 87.5%, 95.2%, and 93.1 % respectively for observer 2. Thirty-five out of 41 hips (85%) that were diagnosed with labral tear on CTA correlated substantially with arthroscopic Lage classification (kappa coefficient = 0.65). CTA and arthroscopic findings showed similar distribution patterns of the tears with most lesions located in antero- and postero-superior areas (p = 0.013). Direct CT arthrography using MDCT may be a useful diagnostic technique in the detection of acetabular labral tear.