Planes, Sequences, and Parameters
MRI is the preferred modality for the investigation of intra-articular hip pathology because it provides exquisite contrast resolution that allows for distinction and characterization of the labrum, cartilage, joint space, capsule, compact and cancellous bone, and regional soft tissues14-16. MRI does not involve ionizing radiation, which is important because many patients being investigated for labral pathology are young2.
Unique to MRI is its multiplanar image acquisition capability. True axial, sagittal, and coronal-plane imaging allows for assessment of anatomy and joint alignment, although, due to variations in anatomy and positioning of the patient, those planes may be suboptimal for the evaluation of smaller, variably located and oriented, intra-articular structures such as the labrum. Thus oblique-plane imaging, relative to specific landmarks, is advised17. The most commonly used approach is to obtain images in three oblique planes relative to the acetabular rim (hence, relative to the labrum), prescribing them from true axial-plane images. The oblique axial and oblique coronal planes will be perpendicular to the anterior-posterior and superior-inferior regions of the labrum, respectively. Three-dimensional-volume isotropic voxel acquisitions with multiplanar and radial reformations have recently become feasible in terms of time and imaging quality and can be applied to MRI of the labrum16.
MRI pulse-sequence selection depends on the technique used, with multiple, variable combinations of the sequence classes able to provide equally diagnostic studies. Spin-echo, fast-spin-echo, and gradient-recalled-echo sequences can all be used in the MRI investigation of labral tears. The sequences may be of variable weighting, including T1, proton-density, and T2, with or without fat suppression applied. A minimum of one fluid (edema)-sensitive sequence (a proton-density sequence with fat suppression, a T2-weighted sequence with fat suppression, or a short tau inversion recovery sequence), should always be part of any protocol17,18.
The labrum is best imaged with use of a surface coil placed around the hip joint, with a small, targeted field of view. The patient is placed supine, and the feet are fastened together to limit motion. There is a wide range of imaging parameters that are commonly used to image the hip labrum, depending on MRI scanner and coil technical specifications, field strength (1.5 or 3.0 T), sequence selection, and radiologist preference. In general, however, the basic parameters are a field of view of 14 to 18 cm, a slice thickness of 3 to 4 mm, and a resolution matrix from 256 × 256 to 512 × 512. Newer, three-dimensional gradient-recalled-echo sequences allow for even thinner slice thicknesses, often as low as 1 mm. The goal is to maximize in-plane and out-of-plane resolution while having a sufficient signal-to-noise ratio and reasonable total scan times.
MRI Appearance of the Acetabular Labrum
The normal labrum is typically well defined, with a pointed triangular shape with sharp margins and with very low signal intensity across all MRI sequences19. There is a firm, continuous attachment of the labrum to the osseous acetabular rim and the acetabular cartilage. This interface between the hyaline cartilage and labrum is referred to as the chondrolabral junction20.
There is a spectrum of morphologic labral changes that may be demonstrated by MRI, including degeneration, substance tearing, and detachment from the acetabular rim. The degenerated labrum may appear on MRI with increased size, globularization, increased intra-substance signal, and surface irregularity21.
Labral substance tears will be demonstrated by increased intra-substance signal extending to the capsular or articular surface of the labrum, with the signal often as high as that of synovial fluid22. With labral detachments, synovial-fluid-intensity signal will undermine the base of the labrum, interposing between the labrum and acetabular rim23. This principle is used to diagnose labral tears regardless of whether or not gadolinium contrast has been added (directly or indirectly) to the synovial fluid23-26. However, it is known by clinical experience and by what has been written in the literature that many tears will not appear with signal intensity as high as fluid because the tear cleft may fill in with granulation tissue, which will prevent synovial fluid or gadolinium solution from entering it. In addition, a volume-averaging effect may occur between the high signal of the tear cleft and the adjacent low signal of the labral tissue, rendering the signal of the tear hypointense relative to that of joint fluid27.
Pitfalls that may lead to a false diagnosis of a labral tear include a sublabral sulcus or recess, a cleft at the junction between the anteroinferior labrum and the transverse ligament, cartilage undercutting of the labrum, and increased signal intensity at the chondrolabral junction25,26,28-30.
Paralabral cysts are small, fluid-filled cysts that can develop secondary to labral degeneration, tears, and detachments. They demonstrate variable degrees of high signal intensity on fluid-sensitive sequences and may fill with synovial fluid and/or gadolinium contrast31.
Direct Magnetic Resonance Arthrography
The combination of MRI with intra-articular injection of gadolinium-based contrast agents is known as direct magnetic resonance arthrography (d-MRA). This technique is often used to facilitate the evaluation of the small structures in the hip joint, including the labrum, hyaline cartilage, and intra-articular loose bodies12,18. d-MRA is a two-stage procedure in which approximately 8 to 15 mL of dilute gadolinium-saline solution (2 mmol/L) is injected intra-articularly under fluoroscopic guidance, followed by MRI that is initiated, ideally, within thirty minutes of the injection18. The distension effect of the arthrogram may cause separation of the capsular, labral, and osteochondral structures, resulting in increased spatial resolution. The injected contrast solution both outlines normal anatomic structures and identifies pathologic changes such as labral tear clefts, further improving contrast resolution and making the labral pathology more conspicuous17,18,32,33. With d-MRA, the most commonly used sequence is T1, with or without fat suppression, in three planes, in addition to fluid-sensitive sequences18. Figures 1-A and 1-B demonstrate the appearances of both a labral substance tear and a labral detachment.Fig. 1-AFig. 1-A Oblique coronal T1-weighted, fat-suppressed d-MRA image demonstrating the superior (twelve o'clock position) acetabular labrum (white arrow) with a linear focus of high signal intensity consistent with a tear along the deep articular surface close to the base, extending into the substance of the labrum. Note the capsular distension related to the intra-articular injection of gadolinium contrast solution. Fig. 1-B Oblique sagittal T1-weighted, fat-suppressed d-MRA image demonstrating a cleft of high gadolinium signal interposed between the anterosuperior acetabular rim and the labral base anterosuperiorly (white arrow), consistent with a labral detachment.Fig. 2-AFig. 2-A Oblique coronal T1-weighted, fat-suppressed i-MRA image demonstrating a complex tear of the superior acetabular labrum (white arrow). There is high gadolinium signal intensity coursing obliquely through the substance of the labrum, extending through both the deep articular and capsular surfaces. In addition, there is similar high signal interposed between the labral base and the acetabular rim, consistent with labral detachment. Fig. 2-B Oblique axial T1-weighted, fat-suppressed i-MRA image through the midportion of the acetabulum, demonstrating the anterior acetabular labrum (white arrow). The labrum is somewhat globular and frayed and shows intrasubstance gadolinium intensity signal that extend1s to both the deep articular and capsular surfaces, consistent with a complex tear.
Fig. 1-A Oblique coronal T1-weighted, fat-suppressed d-MRA image demonstrating the superior (twelve o'clock position) acetabular labrum (white arrow) with a linear focus of high signal intensity consistent with a tear along the deep articular surface close to the base, extending into the substance of the labrum. Note the capsular distension related to the intra-articular injection of gadolinium contrast solution. Fig. 1-B Oblique sagittal T1-weighted, fat-suppressed d-MRA image demonstrating a cleft of high gadolinium signal interposed between the anterosuperior acetabular rim and the labral base anterosuperiorly (white arrow), consistent with a labral detachment.
Fig. 2-A Oblique coronal T1-weighted, fat-suppressed i-MRA image demonstrating a complex tear of the superior acetabular labrum (white arrow). There is high gadolinium signal intensity coursing obliquely through the substance of the labrum, extending through both the deep articular and capsular surfaces. In addition, there is similar high signal interposed between the labral base and the acetabular rim, consistent with labral detachment. Fig. 2-B Oblique axial T1-weighted, fat-suppressed i-MRA image through the midportion of the acetabulum, demonstrating the anterior acetabular labrum (white arrow). The labrum is somewhat globular and frayed and shows intrasubstance gadolinium intensity signal that extend1s to both the deep articular and capsular surfaces, consistent with a complex tear.
d-MRA is the test of choice for evaluation of the acetabular labrum23-25,34,35. Studies comparing d-MRA to arthroscopy report sensitivities of 63% to 100%, specificities of 44% to 100%, and accuracy values that range from 65% to 96%13,23-25,34-36.
d-MRA is resource-intensive, requiring both fluoroscopy and MRI equipment, temporal coordination of the arthrogram and MRI appointments, and direct involvement by a physician who is skilled in hip arthrography18. In addition, it exposes the patient to radiation and the small risk of complications known to be associated with an arthrogram37.
Indirect Magnetic Resonance Arthrography
Indirect magnetic resonance arthrography (i-MRA) involves the intravenous injection of gadolinium contrast (dose, 0.1 mmol/kg), followed by a variable delay and/or physical activity prior to MRI. Gadolinium contrast will distribute within the joint space, diffusely enhancing the synovial fluid12,18,38, to provide greater contrast resolution between the joint fluid and the labrum, cartilage, and capsule. There will also be enhancement of the background extra-articular soft tissues and vascular structures, both normal and pathologic18. With i-MRA, the same sequence selection as used for d-MRA is adequate—T1-weighted sequences in all three planes in addition to a fluid-sensitive sequence—although with an important caveat. As the concentration of gadolinium within the joint may not be as high with i-MRA as with d-MRA, applying fat suppression to the T1 sequences is strongly advised because it will increase the overall contrast resolution of the study18,38. To understand the vascular physiology of the joint, the performance of multiphasic imaging may be considered18. Figures 2-A and 2-B exemplify the ability of i-MRA to demonstrate complex labral tears.Fig. 3-AFig. 3-A Oblique sagittal proton-density, fat-suppressed 3-T magnetic resonance image, without contrast, demonstrating a labral detachment anterosuperiorly with fluid interposed between the labral base and the acetabular rim, with small loculated paralabral cyst formation (white arrow). Fig. 3-B Oblique axial proton-density, fat-suppressed 3-T magnetic resonance image, without contrast, through the midportion of the acetabulum, demonstrating a complex tear of the anterior portion of the labrum (white arrow) with a linear substance tear and complete detachment.
Fig. 3-A Oblique sagittal proton-density, fat-suppressed 3-T magnetic resonance image, without contrast, demonstrating a labral detachment anterosuperiorly with fluid interposed between the labral base and the acetabular rim, with small loculated paralabral cyst formation (white arrow). Fig. 3-B Oblique axial proton-density, fat-suppressed 3-T magnetic resonance image, without contrast, through the midportion of the acetabulum, demonstrating a complex tear of the anterior portion of the labrum (white arrow) with a linear substance tear and complete detachment.
There has been only limited research published recently on the accuracy of i-MRA for the detection of acetabular labral tears. In one small, retrospective study, i-MRA was able to demonstrate 100% of labral tears confirmed at arthroscopy, with only one false-positive result12. In another relatively small study comparing d-MRA and i-MRA, with d-MRA as the reference standard, i-MRA had a sensitivity of 88% and an accuracy of 90% in detecting labral pathology, with substantial agreement between both protocols39. These preliminary studies show promise for the i-MRA technique, although larger population studies are needed before considering i-MRA as the standard test of choice for labral assessment and/or as a replacement for d-MRA.
Compared with d-MRA, i-MRA offers the following benefits to the patient: the intravenous administration of gadolinium is less invasive than the intra-articular injection, and the lack of the fluoroscopically guided arthrogram eliminates radiation exposure. Furthermore, i-MRA has logistical advantages in that only a single radiographic examination has to be scheduled, a radiologist does not have to be directly present for patient preparation or image acquisition, and the MRI can be performed off-site or after hours12,18. Potential disadvantages associated with i-MRA are that the distension effect that d-MRA can give is not realized with i-MRA and that vascular enhancement of paralabral vessels and paralabral or intralabral fibrovascular tissue may lead to false-positive interpretations. There is a small risk associated with systemic administration of gadolinium contrast, the dose of which is much greater than what is injected into the joint with d-MRA12.
Non-Arthrographic Magnetic Resonance Imaging
Non-arthrographic, or routine, MRI of the hip has traditionally been used to assess only large, extra-articular joint structures, including muscle, bone, and the marrow space40. Non-arthrographic MRI, for evaluation of the labrum, is heavily dependent on fluid-sensitive sequences, such as proton-density and T2-weighted sequences (with or without fat suppression) or short tau inversion recovery sequences. On these sequences, joint fluid has high signal intensity as compared with the very low signal intensity of the labrum and capsule and the intermediate signal intensity of hyaline cartilage. This increases the detection of labral substance tears and detachments26. Figures 3-A and 3-B exemplify the ability of non-arthrographic MRI to demonstrate a labral tear. Fluid-sensitive sequences are also optimal for detecting periarticular fluid collections, such as paralabral cysts, that are associated with labral tears41. With non-arthrographic MRI studies, a higher resolution and increased number of sampling averages are used, which may offset the lower contrast and spatial resolution that are associated with routine MRI as compared with d-MRA, although these changes can result in longer scan times15,26.
Only limited studies have been performed to evaluate the accuracy of non-arthrographic, routine MRI in diagnosing acetabular labral tears. The largest study, performed at 1.5 T, found a sensitivity of 96%, a specificity of 33%, and an overall accuracy of 94%26. In a small comparative study, 3-T non-arthrographic MRI was equally effective as 1.5-T d-MRA at detecting labral tears in all subjects42.
With increases in commercially available MRI field strengths in the last decade (from 1.5 to 3.0 T) as well as improvements in MRI hardware and software technologies, non-arthrographic MRI protocols may obviate the need for d-MRA and i-MRA techniques in the future. These technological advances allow for excellent image quality, with higher signal-to-noise ratio, increased spatial and contrast resolution, clinically appropriate scan times, artifact-minimizing strategies, and three-dimensional high-resolution volumetric acquisitions16,26,41. Non-arthrographic MRI of the hip would be the ideal test of choice for investigation of labral tears, as it is noninvasive, free of any radiation or gadolinium exposure, the least resource-intensive and thus least costly, and logistically easier to coordinate when compared with the d-MRA and i-MRA techniques.
Increasingly, the orthopaedic diagnosis and management of musculoskeletal disorders requires adjuvant imaging evidence of disease prior to performing surgical intervention such as arthroscopy. MRI is an effective tool for detecting and characterizing labral tears, with d-MRA being the most commonly used and validated technique for evaluating the labrum. However, i-MRA and non-arthrographic MRI are two less invasive and less resource-intensive techniques that should also be considered. Orthopaedic surgeons and radiologists should strive to develop and implement minimally and noninvasive diagnostic MRI protocols for the investigation of labral pathology.