Afourteen-year-old boy presented to his primary-care physician because of
pain in the right hip of several weeks' duration. He initially described the
pain as diffuse about the hip but later stated that it had become localized to
the groin area, along the medial aspect of the proximal part of the thigh. He
had recently completed a six-week session at summer football camp, which
required repetitive high-impact activities, including running 2 to 3 mi (3.2
to 4.8 km) three times a week and participating in sprint workouts twice a
week. After summer football camp had ended, he immediately began playing
high-school football. The patient initially was treated with nonsteroidal
anti-inflammatory medication and modification of activity and was instructed
to remain non-weight-bearing on the right side and to use crutches for one
week. Radiographs were not made. The symptoms improved with this treatment,
and the patient returned to playing competitive football. The pain recurred
intermittently during the next month and began to increase in frequency and
severity. It became localized to the groin as the season progressed. Beyond
his participation in football, the patient denied any episode of trauma
without a known mechanism of injury. He also denied constitutional symptoms or
pain either at night or at rest. The medical history revealed no pertinent
findings, and the patient was taking no medications and had no dietary
deficiencies. He was then referred for orthopaedic consultation.
Physical examination revealed a well-developed, muscular male patient
(weight, 90 kg) who demonstrated an antalgic gait with a normal
foot-progression angle. The right hip had a full range of motion that was not
associated with pain except with extreme internal rotation in extension.
Anteroposterior and frog-leg lateral radiographs of the right hip and
anteroposterior radiographs of the right femur were made. There was a
nondescript, linear disruption of the normal trabecular pattern on the lateral
side of the right femoral neck, which was only visible on the anteroposterior
radiograph of the right hip (Fig.
1). Because of the suspicious radiographic and physical findings,
magnetic resonance imaging was performed. T2-weighted images revealed an area
of high signal intensity that was consistent with acute bone edema originating
from the lateral femoral neck, and T1-weighted spin-echo images revealed an
area of decreased signal intensity that was consistent with a nondisplaced,
tension-sided femoral neck fracture (Figs.
2-A and
2-B).
Since the fracture had not responded to nonoperative treatment for several
months and because of the potential for displacement and subsequent
complications, we recommended operative stabilization. Under fluoroscopic
visualization, we performed internal fixation by placing two 7.3-mm cannulated
screws percutaneously across the femoral neck without permitting them to
extend across the physis. The patient reported a rapid resolution of hip
discomfort after the operation. After three days, he was permitted to walk
with toe-touch weight-bearing with crutches for four weeks, and he was also
permitted to begin a lower-extremity strengthening program under the
supervision of a physical therapist. He then progressed to weight-bearing as
tolerated, followed by running, strengthening, and agility training. Follow-up
radiographs showed satisfactory healing of the fracture. At eight weeks, he
exhibited a full, painless range of motion as well as a normalization of gait.
Radiographs made at one year revealed good alignment with no evidence of a
fracture line (Figs. 3-A and
3-B). At the time of the latest follow-up, 1.5 years after the
operation, the patient was symptom-free and had resumed all of his preinjury
activities.
The differential diagnosis for a child or adolescent who has a limp with
pain in the hip or the proximal portion of the thigh is extensive and includes
infection, lower-extremity limb-length discrepancy, developmental dysplasia of
the hip, slipped capital femoral epiphysis, transient synovitis,
Legg-Calvé-Perthes disease, trauma, leukemia, tumor, and
fracture11-16.
Achieving a correct diagnosis in children can be difficult. In most instances,
the information gained from a thorough medical history of the patient,
appropriate radiographs, and the physical examination will lead to a proper
diagnosis. In particular, the diagnosis of slipped capital femoral epiphysis
must be placed high on the list of differential diagnoses for an adolescent;
adequate radiographs, physical examination, and, on occasion, magnetic
resonance imaging are necessary to rule out this common diagnosis. In the case
of our patient, magnetic resonance imaging did not reveal any periphyseal
edema, which is a common finding in patients who have a slipped capital
femoral epiphysis.
The history of a patient who has a femoral neck stress fracture will
usually be consistent with that of an active individual who has recently
increased his or her training schedule or intensity. The pain typically is
localized to the groin or the medial aspect of the thigh or is referred to the
distal portion of the femur or the knee, and it is aggravated by activity or
passive internal rotation of the
hip7. The symptoms
can mimic transient synovitis, slipped capital femoral epiphysis, muscle
strains, or benign bone
lesions16. Stress
fractures also can occur in pathologic bone in association with tumors,
osteoporosis, and rheumatoid
arthritis16.
However, when most of the common diagnoses of a painful limp have been
excluded, additional studies will be necessary to determine the exact cause.
Basic hematologic studies usually include a complete blood-cell count,
determination of the erythrocyte sedimentation rate, and measurement of the
level of C-reactive protein to assess the possibility of infection. Additional
imaging studies, such as bone scintigraphy, computed tomography, and magnetic
resonance imaging, may be helpful. Routine radiographs of the pelvis and hip
may not show evidence of a stress fracture for up to four to six weeks after
the onset of symptoms. A technetium bone scan is extremely sensitive and may
show a stress fracture much sooner than radiographs will; however, such scans
are often
nonspecific17,18.
A technetium bone scan probably would not have differentiated between a
compression-sided and a tension-sided stress fracture in our patient, and it
may have been misinterpreted as demonstrating a slipped capital femoral
epiphysis. Magnetic resonance imaging is a very useful technique because of
its increased specificity and its greater ability to identify soft-tissue
lesions about the hip. Short tau inversion recovery magnetic resonance images
are especially helpful in that they can better delineate the chronology of a
stress fracture19.
They are particularly useful for determining the phase of healing and for
ascertaining whether a recurrent or new fracture has occurred at the same
site19. Also, in
comparison with spin-echo images, short tau inversion recovery images have
been shown to provide superior contrast between normal and abnormal
marrow20. Magnetic
resonance imaging should be performed when there is concern regarding the
possibility of an occult fracture or when there is evidence of increased
uptake on bone
scintigraphy18.
Umans et al. showed the value of magnetic resonance imaging in depicting
slipped capital femoral
epiphysis21.
Physeal widening was apparent on the T1-weighted images in every case of
slipped capital femoral epiphysis, including the case of one presumed
so-called pre-slip. T2-weighted images demonstrated synovitis and marrow edema
but did not reveal physeal abnormalities. The authors concluded that magnetic
resonance imaging clearly delineates physeal changes associated with both
pre-slip and slipped capital femoral epiphysis and demonstrates very early
changes at a time when routine radiographs and computed tomographic images may
appear
normal21.
In cases of occult fractures, magnetic resonance imaging tends to show
areas of low signal intensity on T1-weighted spin-echo images and high signal
intensity on T2-weighted images in the area of concern, sometimes exhibiting a
continuous so-called black line within the cortex in the high-signal
area22. Keene and
Lash18 reported the
case of a patient who had a negative bone scan but who was later discovered to
have a femoral neck stress fracture on the basis of magnetic resonance
imaging. There have been other studies, of both children and adults, in which
magnetic resonance imaging was the only modality that depicted the femoral
neck stress
fracture23,24.
Cases in which occult fractures have simulated stress fractures on magnetic
resonance images also have been
reported25-27.
In the case of our patient, the history and the findings of the physical
examination were consistent with a stress fracture. Plain radiographs revealed
a nondescript abnormality on the lateral side of the femoral neck, and
magnetic resonance images were acquired to delineate the nature of the
abnormality. These images revealed a tension-sided femoral neck stress
fracture that we chose to treat in a manner similar to that used for adults.
The patient underwent a successful stabilization procedure with use of two
cannulated screws. At the time of the latest follow-up, 1.5 years after the
operation, the patient was asymptomatic and had returned to full activities,
and there was radiographic evidence of healing of the fracture. Although
compression-sided (medial) femoral neck stress fractures may be treated
nonoperatively with restriction of weight-bearing and modification of
activity, we believe that tension-sided (lateral) femoral neck stress
fractures should be stabilized operatively because of the risk of displacement
and the associated complications such as osteonecrosis. The clinician should
consider a tension-sided stress fracture of the femoral neck in the
differential diagnosis of pain and limping in an adolescent.