A twenty-year-old active-duty male United States Marine sustained
multiple injuries as a passenger in an armored vehicle rollover accident. He
was initially resuscitated and stabilized in the theater of war and then was
medically evacuated within forty-eight hours after the injury and evaluated at
the United States National Naval Medical Center in Bethesda, Maryland.
The initial examination at Bethesda showed pain at the thoracolumbar
junction with a palpable step-off. A closed fluid-filled mass was palpable and
tender in the lower back overlying the sacrum and extending to the lower
lumbar area but was not contiguous with the area of thoracolumbar pain. The
initial neurologic examination demonstrated symmetric bilateral dysesthesias
in the saddle area, brisk bilateral knee and ankle reflexes, normal rectal
tone, and an intact bulbocavernosus reflex. He arrived with a bladder catheter
in place. Spine radiographs (Fig.
1) and computed tomographic scans
(Fig. 2) were notable for an L1
burst fracture with retropulsed fragments into the spinal canal, a 25%
subluxation of the right T12-L1 facet joint, and 23° of ky phosis across
the affected vertebral segment. Pelvic radiographs did not demonstrate
fracture or displacement. A T2-weighted magnetic resonance imaging scan was
notable for a signal-enhancing mass between the subcutaneous and deep fascial
layers of the lower lumbosacral spine (Fig.
3). The magnetic resonance imaging scan was not done through the
pelvis and did not include the sacral origin of the gluteus maximus
muscle.
A decompressive L1 laminectomy with reduction of the retropulsed fragments
and a posterior spinal fusion with segmental instrumentation were performed
from T10 to L2 on the third hospital day
(Fig. 4). Decompression was
accomplished through the posterior approach by direct reduction of the bone
fragments around the thecal sac and ligamentotaxis through an intact posterior
longitudinal ligament during distraction. Pedicle screws were used at the
level of injury, as preoperative computed tomography scans showed the pedicles
to be intact. After wound closure, the fluid-filled mass in the lower back was
incised and approximately 1.5 L of liquefied hematoma-seroma was released. A
complete avulsion of the sacral origin and a partial avulsion of the iliac
origin of the gluteus maximus were identified bilaterally. Initial treatment
included copious irrigation, wound débridement, and placement of a
Vacuum-Assisted Closure device (the V.A.C.; Kinetic Concepts, San Antonio,
Texas). Although the wound bed was subjectively clean following the initial
débridement, the decision was made to perform multiple successive
débridements because of the potential for bacterial colonization in
these types of wounds. The gluteal injury was not initially repaired.
After successive trips to the operating room for irrigation and
débridement at forty-eight-hour intervals, the gluteus maximus was
repaired at ninety-six hours. Figure
5 shows the wound bed just prior to repair.
The muscle and its remaining tendinous attachments were mobilized from
adhesions superficially until they could be approximated to the sacral and
iliac origins with minimal tension. Vicryl mesh (polyglactin; Ethicon, Johnson
and Johnson, Somerville, New Jersey) was sutured across the superficial
surface of the muscle with absorbable monofilament suture. Then 3.5-mm anchors
with FiberWire suture (Arthrex, Naples, Florida) were placed on the
dorsolateral portion of the sacrum from lateral to medial to avoid penetration
into the sacroiliac joint. Three suture anchors, which were spaced to evenly
distribute tension to the repair, were used on each side.
The suture was then passed with a free needle from deep to superficial
through the muscle and was tied in a horizontal mattress fashion
(Fig. 6).
Figure 7 shows the final repair
on the right side of the sacrum. Deep and superficial drains were placed
bilaterally for dead space management. The superficial wound was then closed
in layers.
The patient was maintained in a non-weight-bearing status for seventy-two
hours postoperatively. He was then able to sit, to stand, and to walk
independently with use of a cane. He reported minimal pain with all maneuvers
including sitting. At six weeks, the patient reported no pain while sitting or
walking without assistance. On physical examination at the time of the
three-month follow-up, he was able to extend both hips without pain and had a
normal gluteal contour. He had no tenderness at the repair site in the sacrum
or in the lumbar spine. He had grade-5 of 5 strength in all lower-extremity
muscle groups, including the hip extensors, and he had normal bilateral
perianal sensation. He walked without a limp and returned to activities as
tolerated.