The treatment of hip disease in the young adult is a rapidly evolving and growing area of orthopaedic surgery1, driven by improved imaging techniques2,3, safe and reproducible open hip surgery4, and less invasive techniques such as hip arthroscopy5. Currently, the most common indication for hip arthroscopy is the treatment of labral abnormalities6, usually from one of four causes7: trauma, a degenerative condition, dysplasia, and impingement. Labral abnormalities are often associated with acetabular chondral damage8,9, varying in severity from a softening and/or fibrillation to complete cartilage detachment from its osseous bed. Because the treatment of the labral abnormality varies, depending on the underlying etiology and associated pathological condition, it is critical that variant anatomy of the labrum and the labral chondral junction be recognized and treated accordingly7,10. Descriptions of variations of the labral anatomy have focused mainly on the manner in which arthroscopic findings correlate with magnetic resonance imaging (MRI) findings. The primary variation has been termed sublabral sulcus, defined as a cleft between the labrum and adjacent articular cartilage. The reported location of this sublabral sulcus is variable, as is the prevalence, although it is most commonly noted anteriorly. Byrd was the first, as far as we know, to describe as "normal anomalous variations" these partial separations of the labrum from the lateral aspect of the osseous rim of the acetabulum11. However, previous studies have some limitations since the magnetic resonance arthrography (MRA) was reviewed retrospectively to determine the prevalence and location of these sulci, making the exact location difficult to interpret.
We report on a clinical series of five female patients seen with mechanical hip pain secondary to an acetabular labral limbus. The clinical presentation, as well as the intraoperative findings at the time of hip arthroscopy, is described.
The patients were informed that data concerning their cases would be submitted for publication, and they consented. Institutional review board approval was obtained.
Five female patients with a mean age of 19.2 years (range, seventeen to twenty-three years) presented with a history of unilateral groin pain for a mean of 25.2 months (range, twelve to forty-eight months) and with no history of trauma or childhood hip disease. The pain was aggravated by sporting activities in all, and three of the patients also had subjective clicking that occurred with hip flexion and associated rotation. All of the patients had pain with prolonged sitting.
On physical examination, all patients had a normal range of motion of the hip (flexion of >120° and internal rotation of >25°) and a positive impingement sign12,13. The impingement sign, performed with the patient in the supine position with the symptomatic hip flexed >90°, adducted, and internally rotated, reproduces the groin pain.
Our standard radiographic protocol was a well-centered anteroposterior pelvic radiograph and Dunn view of the painful hip14,15. The center-edge angle of Wiberg16, the Tönnis angle17, the presence or absence of a pincer deformity, which was defined as either acetabular retroversion18 (localized anterior overcoverage, where the anterior wall crosses the posterior medial to lateral edge of the acetabulum) or coxa profunda (global overcoverage)9,18, as well as the alpha angle on the Dunn view15 were evaluated. Four of the five patients had evidence of coxa profunda with radiographic findings of the acetabular floor touching or crossing the ilioischial line, but in all patients the center-edge angle was =35° (Table I). None had a cam deformity (Fig. 1).
All patients underwent MRI with gadolinium arthrography (MRA). Each hip was injected with 10 to 15 mL of a diluted (2-mmol) gadolinium-saline solution (Omniscan; GE Healthcare, Princeton, New Jersey) under fluoroscopic guidance by a musculoskeletal fellow or radiologist. The MRI scan was initiated within thirty minutes after the injection. The examinations were carried out on a 1.5-T scanner (Symphony Quantum; Siemens, Erlangen, Germany) with a flexible surface coil. All subjects underwent the routine protocol for MRA of the hip at our institution19, which included an axial three-dimensional isotropic, T1-weighted spoiled-gradient-echo (MPRAGE/Turbo-FLASH) sequence with water excitation (25-cm FOV, 1-mm slice thickness [ST], 256 × 256 matrix, TR/TE/flip angle = 1970 ms/7 ms/15°, and one signal averaged) with radial reformatting; oblique axial fast-spin-echo T1 (180-cm FOV, 4-mm ST, 512 × 256 matrix, TR/TE = 614 ms/17 ms, two signals averaged); coronal fast-spin-echo T1 with fat suppression (3.5-mm ST, 512 × 256 matrix, TR/TE = 553 ms/16 ms, one signal averaged); and sagittal fast-spin-echo T1 with fat suppression (3.5-mm ST, 512 × 256 matrix, TR/TE = 553 ms/16 ms, and two signals averaged). All images were sent to a PACS (Horizon Rad Station 3.3; McKesson, San Francisco, California) and were reviewed by a musculoskeletal radiologist. None of the patients had labral abnormality on MRA, and radial reformats showed normal femoral head and neck contours (Fig. 2). Given the lack of a definite osseous abnormality and a negative MRA, a bupivacaine hip block was performed to confirm that the source of the pain was intra-articular. All five patients had complete relief of the symptoms for at least two hours after this injection20,21.
Arthroscopic Findings
On the basis of the unsuccessful nonsurgical management with activity modification and pain relief with bupivacaine injection, diagnostic hip arthroscopy was done. In all patients, a standard two-portal hip arthroscopy was performed with the patient in the supine position22, with the first portal being introduced anterolaterally under fluoroscopic control and the second directly anterior in line with the anterior superior iliac spine and anterolateral portal. A 70° scope was used to inspect the central compartment. At the time of arthroscopy, there was no evidence of femoral or acetabular articular cartilage damage. The base of the labrum was well fixed to the osseous acetabular rim. However, anteriorly, the intra-articular portion of the labrum extended centrally, overlying the acetabular hyaline cartilage, which we interpreted as a limbus. The extent to which the limbus protruded medially as well as its size varied between patients (Fig. 3). On arthroscopic examination, only the portion of the labrum extending over the acetabular cartilage (the limbus) was mobile. Once this portion was resected (Fig. 4-A), the underlying acetabular articular cartilage was found to be intact. At the time of closure, a local anesthetic solution of bupivacaine was injected into the joint and portal sites. Postoperatively, the patients were managed with crutches and partial weight-bearing (50%) for ten days, followed by physiotherapy with a gradual return to normal activities at the four-week mark. At the one-year follow-up examination, four of the five patients had complete relief of the symptoms. One patient required a repeat arthroscopy eight months after the initial arthroscopy for persistent hip pain. At the time of the repeat arthroscopy, the acetabular labral limbus had been inadequately resected (Fig. 4-B); this was additionally removed and the patient was pain-free at the time of the six-month follow-up. Four of the five patients were able to return to their recreational sporting activities.