Purpose: To compare non-enhanced magnetic resonance imaging (MRI), MR arthrography, and computed tomography (CT) arthography in the evaluation of labral, chondral, and ligamental lesions of the hip using arthroscopy as the gold standard.

Methods and Materials: Prospective evaluation of 36 patients (19 male, 17 female; ranging between 16 and 69 years of age; average, 37 years) with non-enhanced MRI, MR arthrography, and CT arthrography between January 2002 and January 2003. A 1.5-T MR unit and an 8-row multislice CT scanner were used. Arthroscopy was performed by one experienced orthopedic surgeon after a maximum of 8 weeks from the imaging studies. A solution of 15 mL of nonionic iodinated contrast medium, 5 mL of bupivacaine, and 0.1 mL of paramagnetic contrast medium was injected into the hips under fluoroscopy. Two experienced musculoskeletal radiologists reviewed by consensus all the available images for each patient.

Results: Non-enhanced MRI and MR arthrography showed 8 cases with degenerative changes of the labrum (evidenced by increased intrasubstance signal intensity and/or contour irregularity), 28 labral tears, 3 perilabral cysts associated with labral tears, 5 chondral defects, 3 cases of hyperthrophy of the ligamentum teres, and 1 case of hyperthropy of the pulvinar. CT arthrography showed the same 28 labral tears and 5 chondral defects. Degenerative changes of the labrum, perilabral cysts, and ligamental lesions could not be identified on CT arthrography. The chondral defects were more conspicuous on CT arthrography than on MRI. Using arthroscopy as the gold standard, non-enhanced MRI and MR arthrography had 74% and 94% of sensitivity in the detection of any type of hip lesion, respectively.

Conclusion: MR arthrography is more sensitive for labral and ligamental lesions of the hip than either non-enhanced MRI or CT arthrography. Chondral defects are more conspicuous on CT arthrography than on non-enhanced MRI or MR arthrography.