MR arthrography of the hip is performed as a two-stage procedure. The first step is performed under fluoroscopic guidance. The preferred approach for hip arthrography is a direct anterior approach, targeting the lateral femoral neck or the lateral femoral head (Fig. 15.2). The intra-articular needle position is confirmed by injection of iodinated contrast material. Joint distention is achieved by injection of a dilute solution of gadolinium (12). The standard dilution of gadolinium for joint distention is 2.5 mm of gadopentetate dimeglumine (9). For a single hip this dilution may be achieved by mixing 0.1 mL of gadopentetate dimeglumine in 20 mL of normal saline, however, different gadolinium products may contain different gadolinium concentrations. The joint capacity is 8 to 20 mL. Intra-articular lidocaine injection also may be useful. Pain relief with intra-articular lidocaine further supports an intra-articular source. Failure to produce pain relief, however, does not exclude an intra-articular source. With the use of intra-articular gadolinium, axial, coronal, and sagittal T1-weighted fat suppressed images are acquired through the hip joint using a surface coil and field of view of 14 to 16 cm. It is important to use at least three imaging planes to ensure that all portions of the labrum are adequately assessed. An assessment of the entire pelvis is performed with T1 and short tau inversion recovery (STIR) coronal images that include the symphysis pubis and the sacrum, increasing the field of view to 36 to 40 cm.

Standard imaging planes sufficiently demonstrate all acetabular labral tears. Imaging in the axial oblique plane has the highest rate of detection of acetabular labral tears (10). Addition of radial imaging has not been found to reveal additional labral tears (11).

One pitfall in the interpretation of MR arthrography of the hip is the question about the presence of a normal sulcus or recess at the anterosuperior aspect of the joint. To date, no conclusive data are available about whether such a sulcus may exist normally. The perilabral recesses are located at the labrocapsular junction on the capsular surface of the labrum. A sulcus created by the junction of the transverse ligament and the labrum is also a normal finding (Fig. 15.3). This sulcus should be known as the labroligamentous sulcus.
Another sulcus is located at the posteroinferior aspect of the joint, and it represents a physiologic cleft at the junction of the articular cartilage. The cleft may be partial or complete. Histologic studies have concluded that the articular cartilage and the labrum blend together seamlessly, which suggests that a separation between the articular cartilage and labrum would be abnormal (2,13,14). Cashin et al. studied the prenatal development of the human acetabular labral–chondral complex in 11 fetal hips, aged from 8 weeks of gestation to term. Morphologically, the anterior acetabular labral–chondral complex had a somewhat marginal attachment to the acetabular cartilage and seemed to cap its anterior part. There was an intra-articular labral projection which formed a recess between the articular surface of the acetabulum and the labrum. The posterior labrum differed in that it was directly attached to and continuous with the acetabular cartilage. It lacked a labral projection into the articular space. Under higher magnification, the attachment site of the labrum to the acetabulum anteriorly demonstrated a sharp, abrupt transition at the junction of the cartilaginous acetabulum and the labrum. Anteriorly, the collagen fibers at the junction of the acetabular cartilage ran parallel to the labral–cartilage junction, but posteriorly they were perpendicular to this junction when examined under polarized light (15).

Hodler et al. (2) examined 12 cadaver hips from elderly individuals to make their observations. The conclusions of Seldes et al. (13) and Tan et al. (14) are drawn from studies of 67 cadaveric specimens. Seldes and Tan, however, report tears at the anterosuperior margin in 74% of hips, and 89% of these tears were detachments. In their study, few normal anterosuperior labral margins were examined. The average age of their specimens was 78 years. The finding is of significance in the younger population but not the older population. In their study of six cadaveric hips (age range: 72 to 84 years), Czerny et al. (16) did not identify any sublabral sulci. Dinauer et al. (17) and Petersilge et al. (3) in their studies of 23 and 24 hips, respectively, were unable to confirm the presence of a sulcus at the anterosuperior portion of joint. In an arthroscopic study of 56 hips, Fitzgerald (5) found 41 tears with separation of the articular cartilage and labrum. Most tears were in the anterior aspect of the hip. Based on visual inspection, they reported that evidence of attempted healing of the defect was usually seen but a residual sulcus could be identified. The sulcus measured 2 to 5 mm in width and 8 to 20 mm in length, which would suggest that the sulcus is abnormal. The literature suggests that a normal cleft is located between the margin of the articular cartilage and the labrum and that this normal variant has an incidence of 5% to 6%. The margins of the cartilage edge and the border of the labrum are sharp. In contrast, any extension of contrast into the acetabular (osseous) labrum junction should be considered abnormal. Any irregularity of the margins of the cartilage edge or the adjacent labrum would indicate that the separation between labrum and cartilage is abnormal.

A defect in the capsule between the pubofemoral and iliofemoral ligaments may lead to communication between the joint and the iliopsoas bursa, and contrast may occasionally fill this structure (3). The bursa should not be confused with an extra-articular cyst or ganglion. The bursa typically lies lateral to the iliopsoas tendon, whereas extra-articular cysts are medial and usually adjacent to the acetabulum (18).