All four patients presented with persistent hip pain radiating to the groin
area that was reproduced with flexion, adduction, and internal rotation (the
so-called impingement test). The mean age of the patients was forty years
(range, thirty-six to forty-five years) at the time of presentation. There
were three men and one woman, and they had a mean duration of symptoms of 3.0
years (range, 0.7 to eight years). Standard hip radiographs that included
anteroposterior pelvic and frog-leg and cross-table lateral views were made of
all patients. None of the patients had radiographic signs of arthritis. All
four hips lacked femoral head-neck offset as seen on the cross-table lateral
radiograph13, which
suggested a cam-type femoroacetabular impingement as described by Ganz et
al.14,15
(Fig. 1). All patients had
evidence of cartilage delamination on magnetic resonance imaging with
gadolinium arthrography, and this finding was confirmed at the time of the
surgery (Figs. 2 and
3). The four patients underwent
surgical dislocation of the hip with a femoral
chondro-osteoplasty16
to correct the osseous abnormality.
Illustrative Case Report
A thirty-eight-year-old man presented with an eight-month history of
progressive pain in the left hip radiating to the groin. There was no history
of trauma. The pain in the hip was worse in the mornings and was aggravated by
long periods of sitting or any forced hip flexion. On physical examination,
the patient was found to have a slightly antalgic gait, no limb-length
discrepancy, and a well-preserved range of motion. Flexion, adduction, and
internal rotation of the hip reproduced a knife-like pain in the groin, which
represented a positive impingement test. The plain radiographs showed no signs
of acetabular dysplasia or arthritis; however, the cross-table lateral
radiograph suggested femoroacetabular impingement. A magnetic resonance scan
with gadolinium arthrography revealed an acetabular labral tear with
associated delamination of the cartilage
(Fig. 4).
Technique of Magnetic Resonance Imaging
Each of the hips was injected from an anterior approach with use of sterile
technique and fluoroscopic guidance to localize the femoral neck. Three
milliliters of nonionic iodinated contrast medium (Omnipaque 300) was injected
to confirm the intra-articular position of the needle, and this was followed
by instillation of 10 to 20 mL of gadolinium saline solution (Omniscan) at a
1:100 dilution. The magnetic resonance imaging was performed with a Siemens
1.5-T Symphony magnetic resonance imaging scanner (Erlangen, Germany). Axial,
coronal, and sagittal fat-saturated images were acquired with use of spine and
body array coils (repetition time, 630 msec; echo time, 12 msec; 256 ×
256 matrix for axial and sagittal images; 512 × 512 matrix for coronal
images; 3-mm slice thickness and two interleaved excitations with use of a 200
mm field of view. The T1-weighted images with fat saturation demonstrated the
delamination of the articular cartilage in the anterolateral quadrant of the
acetabulum with the normal joint signal created by summation of the
T1-weighted signal from the femoral head hyaline cartilage, intra-articular
gadolinium contrast medium, and acetabular hyaline cartilage. The signal of
the delaminated segment was markedly decreased on the T1-weighted images and
was seen best on planes of section that were oriented perpendicular to the
involved cartilage; in our series, these were on the sagittal images. Larger
lesions were apparent on both the sagittal and the coronal images.
Although magnetic resonance imaging is a well-established tool in the
diagnosis of musculoskeletal conditions such as
osteonecrosis17,
acetabular labral
tears18, and
meniscal tears of the
knee19, its role in
assessing articular cartilage of the hip joint has not been completely
defined11,20.
Initial attempts at evaluating articular cartilage of the hip without
gadolinium enhancement were disappointing, and hip arthroscopy was often used
to examine the painful nonarthritic
hip9. However,
arthroscopy is associated with potential complications, which can range from a
simple wound hematoma to septic
arthritis21.
The introduction of different imaging techniques such as radial magnetic
resonance imaging22
and gadolinium
arthrography18 has
resulted in major improvements enabling magnetic resonance visualization of
the hip joint23. In
their review of forty-two hips, Schmid et
al.23 were able to
correlate descriptions of labral tears and cartilage defects based on magnetic
resonance gadolinium arthrography with surgical findings with moderate
accuracy and good interobserver reliability. The cartilage lesions were
described as the presence or absence of chondral degeneration with or without
a labral tear. Schmid et al. did not report on cartilage delamination. In our
series of patients, we found great consistency in the location of the labral
tears and associated cartilage delamination. Lesions were found at the
anterosuperior-lateral margin of the hip joint, in keeping with previously
reported arthroscopic
findings24. The
relationship between labral tears and cartilage delamination—that is,
which comes first—remains to be determined, but both play a role in the
development of hip
arthritis25.
The appearance of the cartilage delamination on magnetic resonance imaging
was quite consistent. On the T1-weighted images, the areas of normal articular
cartilage were observed to have a unilaminar appearance, with the delaminated
cartilage segment identified as a low-signal-intensity curvilinear
intra-articular flap of >1 mm in thickness. In addition, there was a bright
signal deep to the flap that was isointense to the normal articular monolayer
with the intraarticular gadolinium contrast medium. Although the histological
basis for this appearance remains to be determined, the loss of the
T1-weighted signal could be related to several mechanisms, such as fibrous
metaplasia of the hyaline cartilage and/or depletion of the ground substance
proteoglycan from the extracellular matrix. Further investigation, including
histological examination to characterize the delaminated segment, is
necessary.
A physician evaluating a patient presenting with persistent hip pain should
consider the possibility of delamination of the acetabular cartilage,
particularly if pain is elicited with flexion, internal rotation, and
adduction of the hip. We recommend plain radiographs to rule out acetabular
dysplasia and a stress fracture followed by magnetic resonance imaging with
gadolinium arthrography.
Because the natural history of these lesions is not known, nonoperative
management with medication and physical therapy should be the first line of
treatment. However, these chondral lesions are believed to represent an
advanced stage of degeneration of the articular cartilage that, if left
unrecognized, may affect the prognosis following joint-preserving
procedures25,26.
The cartilage delamination seen in this series, and described as chondral
flaps by McCarthy et
al.25, would fit
best in the cartilage-disruption category described by
Buckwalter27. Beck
et al.28 subdivided
cartilage delamination lesions into two categories: debonding (the cartilage
appears macroscopically sound but is mobile, simulating a carpet phenomenon)
and cleavage (there is a frayed edge, which was the case in our series). How
this damage occurs in the acetabulum is unclear. We believe that the cartilage
delaminations that we described were most likely caused by a cam-type
femoroacetabular
impingement14 often
seen in patients who lack femoral head
offset29, have a
history of slipped capital femoral
epiphysis30, and
have osteonecrosis of the
hip31. Surgical
dislocation was performed in all four of our patients since they all presented
with an intracapsular osseous abnormality requiring treatment. The best type
of treatment of these lesions and the long-term outcome of these procedures
are beyond the scope of this paper. However, we believe that treatment of
these abnormalities may help to preserve the hip joint. Nonetheless, hip
arthroscopy may also offer a simple recovery and provide an effective
treatment for these
lesions32.