Case 1. A fifty-two-year-old woman was admitted to the hospital following alcohol intoxication. Reportedly, she was unconscious for an unknown period of time. At the time of admission, the patient had no sensation on the plantar aspect of the right foot, no pain with passive stretch of the toes, and no palpable posterior tibial or dorsalis pedis pulses. The leg was very tense to palpation. The compartment pressure, measured with use of a handheld monitor (Stryker Intracompartmental Pressure Monitor System; Stryker Instruments, Kalamazoo, Michigan), was 40 mm Hg in the superficial posterior compartment. Her blood pressure on admission was 100/40 mm Hg, and she was anuric. Initial laboratory tests indicated a metabolic acidosis due to renal insufficiency. Blood urea nitrogen was 28 mg/dL (10 mmol/L) (normal, 6 to 22 mg/dL; 2.1 to 7.9 mmol/L); serum creatinine, 4.9 mg/dL (433 µmol/L) (normal, 0.6 to 1.2 mg/dL; 53 to 106 µmol/L); serum creatine phosphokinase, 22,900 U/L (normal, 15 to 165 U/L); and blood pH, 7.34 (normal, 7.37 to 7.45). She had no fracture of the lower extremity.
She was taken to the operating room and underwent a four-compartment fasciotomy with use of a standard two-incision technique2. Both the medial and lateral incisions were 18 cm in length. Following release of all four compartments, circulation returned to the foot, with palpable pulses. The initial intraoperative examination revealed bleeding from the muscles in all compartments with limited areas of dark discoloration and minimal contraction to mechanical stimulation. Both incisions were left open.
The patient underwent two more débridements at forty-eight hours and ninety-six hours following the first procedure. During each procedure, muscle viability was reassessed and we found that the response to mechanical stimulation was gradually disappearing. Intraoperative cultures obtained at the time showed polymicrobial flora, including methicillin-resistant Staphylococcus aureus and groups A, B, C, and G streptococcus. On the fifth day of hospitalization, she had spiking fevers to 39.5°C despite wide-spectrum antibiotic coverage (1 g of vancomycin and 400 mg of ciprofloxacin administered intravenously every twelve hours). The next day, the patient underwent a fourth operation. We noted that the muscles in all compartments were unresponsive to mechanical stimuli, and therefore we excised them in toto. The peroneal vascular bundle was found to be occluded with a thrombus distal to the trifurcation. The anterior and posterior tibial neurovascular bundles were intact, and there was a good perfusion to the skin covering the tibia. At the end of this procedure, the wound was packed with laparotomy pads and the skin was approximated loosely. The leg was placed in a posterior splint with a bulky dressing.
During the next procedure, two days later, negative suction drains (Hemovac; Zimmer, Dover, Ohio) were placed along the lateral and medial sides of the leg and were maintained in place for sixteen days until the output decreased to 20 mL over two consecutive days in total from both drains. Antibiotic treatment was discontinued forty-eight hours after the last procedure. The leg was placed in a posterior leg splint until the removal of both drains, and weight-bearing was not permitted. Following drain removal, the patient was managed with an ankle-foot orthosis and was allowed to bear weight as tolerated. No additional surgery was required, and the patient was discharged home.
On her last examination at six months, she stated that she was able to feel the lateral and medial aspects of the midpart of the leg. Semmes-Weinstein monofilament testing, however, revealed no response to a 6.5-mm monofilament at those locations. She had no sensation on the dorsal and plantar aspects of the foot (Fig. 1). She had intact posterior tibial and dorsalis pedis pulses. Passive ankle motion was 80°, and she had flexible clawing of the toes. She could only walk with the aid of the ankle-foot orthosis (see Appendix). Her overall general health status was assessed with use of the Short Form-36 (SF-36) generic questionnaire3 (Table I). She did not seek any additional treatment at that point as she was going through rehabilitation and psychotherapy programs. She was taking narcotic analgesic medication along with antidepressants.
Case 2. A thirty-four-year-old woman was found unconscious in a crouched position following alcohol intoxication. On admission, she had elevated compartment pressures of 40 mm Hg, measured with a handheld monitor (Stryker Intracompartmental Pressure Monitor System), in the right leg with no palpable pulses. Sensation was absent on the plantar and dorsal aspects of the foot. The duration of the crush injury was not known at the time of the initial evaluation. Her blood pressure on admission was 110/55 mm Hg. She was making 45 mL of urine per hour. Her temperature was 38°C. Initial laboratory tests showed metabolic acidosis with renal insufficiency. Blood urea nitrogen was 45 mg/dL (16.1 mmol/L) (normal, 6 to 22 mg/dL; 2.1 to 7.9 mmol/L); serum creatinine, 3.9 mg/dL (345 µmol/L) (normal, 0.6 to 1.2 mg/dL; 53 to 106 µmol/L); serum creatine phosphokinase, 11,916 U/L (normal, 15 to 165 U/L); and blood pH, 7.33 (normal, 7.37 to 7.45). She had no fracture of the lower extremity. She was taken to the operating room immediately for a four-compartment fasciotomy.
Two 15-cm incisions were used to release all four compartments, and the pedal pulses returned to normal after the fasciotomy. Parts of the anterior compartment muscle groups were not viable as evidenced by changes in color and turgor and the lack of contractility to mechanical stimulation. Following débridement of these muscles, the patient was taken to the operating room three more times at intervals of forty-eight hours. During these procedures, the remaining muscle groups were not responsive to mechanical stimuli and exhibited marginal blood flow and signs of patchy ecchymosis. On day 8, all remaining muscle groups were removed. The leg wounds were packed for forty-eight hours, and a bulky dressing and a posterior splint were applied. Forty-eight hours later, the packing material was removed, the wounds were closed, and two negative suction drains were placed, one medial and one lateral. Antibiotic treatment (1 g of vancomycin and 400 mg of ciprofloxacin administered intravenously every twelve hours) was discontinued on day 10, forty-eight hours after the last procedure. No additional surgery was required. The drains were removed on day 28, when their combined output was <20 mL over two consecutive days. The leg was maintained in a posterior splint, and no weight-bearing was permitted until drain removal. Then, she was managed with an ankle-foot orthosis and was allowed to bear weight as tolerated. The patient remained in the medical intensive care unit for the treatment of renal insufficiency, and eventually she was discharged forty-five days after admission. Shortly after discharge, she attended an alcohol rehabilitation program.
On her last visit at three years after the injury, the patient had 15° of passive range of ankle motion with clawing of the toes. Semmes-Weinstein monofilament testing showed a positive response to a 4.2-mm monofilament over the plantar aspect of the first metatarsal head and to a 3.61-mm monofilament over the dorsal aspect of the third metatarsal. When walking without the ankle-foot orthosis, she had a stable stance phase, but she had to lift the injured leg higher during swing phase compared with the contralateral leg. She did not wish to undergo an ankle fusion (see Appendix). Her overall health status was assessed with the SF-36 questionnaire (Table I). She graduated from the alcohol rehabilitation program and returned to full-time employment as a cashier.
Case 3. A thirty-three-year-old woman was trapped for six hours under a collapsed building following an earthquake. On admission at another hospital, she reportedly had signs of compartment syndrome involving both lower extremities with intact pulses but no sensation on the plantar aspects of the feet. She had no fractures. She underwent four-compartment fasciotomies of both lower extremities with use of a two-incision technique. Following seven serial débridements at that facility, the patient was referred to our tertiary-care center for a second opinion fifteen days after the original injury.
On admission, she had intact pedal pulses bilaterally but had no sensation distal to the knee. Overall, her general status was good except for mild renal insufficiency and metabolic acidosis. On admission, her blood pressure was 120/80 mm Hg and she had the following blood levels: blood urea nitrogen was 15 mg/dL (5.4 mmol/L) (normal, 6 to 22 mg/dL; 2.1 to 7.9 mmol/L); serum creatinine, 1.4 mg/dL (124 µmol/L) (normal, 0.6 to 1.2 mg/dL; 53 to 106 µmol/L); serum creatine phosphokinase, 210 U/L (normal, 15 to 165 U/L); and blood pH, 7.40 (normal, 7.37 to 7.45). She had a fever of 38.5° despite broad-spectrum antibiotic coverage (1 g of vancomycin administered intravenously every twelve hours and 2 g of ceftazidime administered intravenously every eight hours). The prior wound cultures had grown methicillin-resistant Staphylococcus aureus, Enterobacteriaceae, and gram-negative bacilli.
Because of extensive myonecrosis, all of the remaining muscle groups in the left leg and the anterior and lateral compartment muscles in the right leg were excised. The wounds were initially treated open and were irrigated in the operating room on two separate occasions forty-eight hours apart. On day 7, all of the wounds were closed over negative suction drains (two in each leg). Wound drainage continued for thirty-five days, and all drains were removed after there was <20 mL of output over two consecutive days from both drains in each leg. Postoperative dressing and splinting modalities were similar to those used for the previous patients (Cases 1 and 2). The patient remained in the medical intensive care unit for dialysis and was discharged on day 67. She had no sensation in either lower extremity distal to the knees at the time of the discharge.
The patient began to walk with the aid of crutches at three months. At seven months, both feet were responsive to a 3.61-mm Semmes-Weinstein monofilament testing over the plantar aspect of the first metatarsal head and the dorsal aspect of the third metatarsal. At six months, the superficial and deep posterior muscle groups were functioning on the right and she underwent a tibialis posterior to tibialis anterior tendon transfer on that side. On the left, she underwent arthrodesis of the tibiotalar and subtalar joints.
At the time of her last visit, nine years after the injury, the patient was able to walk independently. She was not able to run, but she was able to walk moderately long distances (1 to 2 km). Because the pantalar fusion was in slight equinus, she needed to wear high-heeled shoes. She was responsive to 3.61-mm Semmes-Weinstein monofilament testing on the plantar aspect of the first metatarsal head and the dorsal aspect of the third metatarsal in both lower extremities. The overall health status was assessed with the SF-36 health survey (Table I).
Total muscle excision after a severe crush injury to the lower extremity has been described previously4. Reis and Michaelson reported on seven patients who were initially treated with four-compartment fasciotomies and then had development of extensive myonecrosis requiring secondary débridement. Several of these patients had only bone, skin, and neurovascular bundles, but they were reported to have had a favorable outcome. However, no specific data or long-term follow-up were described.
The role of fasciotomy in the treatment of crush injuries to the lower extremities is somewhat controversial4-6. Reis and Michaelson reported on fifteen patients in two groups with crush injuries treated with or without fasciotomy. They concluded that fasciotomy offered no advantage in patients with a delay in diagnosis, and it often resulted in sepsis and amputation4. In another study of forty-six patients with fifty-eight crushed extremities, Matsuoka et al. provided no evidence that fasciotomy improved the outcome5. In a larger retrospective study describing ninety-five patients with crush injuries, Huang et al. reported a higher rate of infection and amputation in patients with fasciotomies6. The authors of those studies concluded that fasciotomy should be withheld in patients with crush injuries seen after a delay, unless there is an open fracture or a pulseless extremity or the duration of compression is not known in the presence of increased compartment pressure. The fasciotomies applied in the three patients we described are examples of each of these indications. The surgical technique described in the present report may prevent the complications (e.g., amputation and sepsis) seen after fasciotomies. However, making the correct diagnosis in a timely manner is critical, as incomplete excision of necrotic muscle may lead to uncontrolled infection.
None of the patients in our series had development of a cavovarus deformity, which is commonly seen after a Volkmann ischemic contracture of the lower extremity7,8. This may be a result of complete excision of all of the muscles in the leg, eliminating unopposed pull or contracture of muscle-tendon units. Two patients had clawing of the toes develop, likely because of unopposed pull of the reinnervated intrinsic muscles of the foot. Two of the three patients had a return of sensation to the plantar and dorsal aspects of the foot. The other patient (Case 1) was last seen only six months after the injury and may have had further improvement in sensation with the passage of time. The lack of active ankle motion affected the gait as well as daily and recreational activities in all three patients, as reflected in the SF-36 scores. In comparison with normative values, all of the patients scored less in the physical functioning domain9. Activities such as running, bowling, and participating in competitive sports are impossible for them, and lifting heavy objects (>25 kg), bending, kneeling, and stooping seem to be very difficult. Ankle stability can be achieved with one of three methods: fusion of the ankle joint (Case 3), the use of an ankle-foot orthosis (Case 1), or through the development of contractures (Case 2). Two patients (Cases 1 and 2) both had major depressive disorders. Both of these patients scored lower than the average score for the general population in the mental health, emotional role, vitality, and social functioning categories. Whether this was related to the resultant physical impairment or the baseline mental status of the patients is not clear. From a cosmetic standpoint, total four-compartment muscle excision creates a leg that is markedly different compared with the contralateral leg both in appearance and circumference. However, all three patients stated that they still would prefer to have the deformed limb rather than a prosthetic limb.
Several technical points should be emphasized. The success of this technique is dependent on two factors: (1) management of the dead space and (2) survival of the skin covering the leg. For the management of the dead space, we placed two negative suction drains and had them remain in place for a long time (sixteen, twenty-eight, and thirty-five days). We believe this approach prevented hematoma, seroma, or abscess formation. For the survival of the skin covering the leg, two of the four angiosomes of the leg were preserved. In 1987, Taylor and Palmer introduced the angiosome concept and defined the three-dimensional vascular territories supplied by source arteries and veins to each tissue layer between the skin and the bone10. Their studies showed that the leg has four angiosomes11: (1) the anterior tibial artery territory (the skin between the anterior tibial crest and the fibula anterolaterally), (2) the peroneal artery territory (the skin between the fibula and the posterior midline, posterolaterally), (3) the posterior tibial artery territory (the skin between the anterior tibial crest and the posterior midline medially), and (4) the popliteal (sural) artery territory (the skin covering the popliteal fossa and the posterior aspect of the proximal third of the leg).
Skin branches to the anterior surface of the leg come from muscular and septocutaneous perforators from the anterior and posterior tibial arteries. Excision of all muscles in the compartment removes the muscular perforator branches, leaving intact only the septocutaneous branches between the tibia and the skin. These are mostly supplied by the posterior tibial artery. In all of our patients, we paid particular attention to avoid undermining the skin covering the anterior surface of the tibia in order to protect these perforators. Skin branches to the posterior surface of the leg are supplied by the sural, peroneal, and posterior tibial arteries. Among these branches, those that come from the peroneal and the posterior tibial arteries are mostly muscular and are removed during the excision. That basically leaves only the sural artery territory intact among the three sources. Proximally, the sural artery commonly originates from the popliteal artery and usually remains intact, even in severe crush injuries of the leg12. However, distally, it also forms an anastomosis with a septocutaneous perforator from either the peroneal or the posterior tibial artery approximately 5 cm proximal to the tip of the fibula. A circuit is created such that if the proximal arterial supply (the branch from the popliteal artery) is occluded, the flow is reversed through the distal perforator supplying all of the skin of the posterior surface of the leg12. In all three patients, we preserved the distal anastomotic branch of the sural artery in order to ensure the viability of the skin posterior to the leg.
This report shows that total muscle excision following a crush injury to the leg may result in a satisfactory functional outcome in the absence of fracture. Further studies are needed to compare these results with the alternative procedure, amputation.