A fifty-one-year-old man (Case 1)
and a twenty-six-year-old woman (Case 2) had a limb-sparing
operation with sciatic nerve resection. Both patients had a malignant
peripheral nerve-sheath tumor that developed in the sciatic nerve
in the proximal half of the posterior aspect of the thigh. The surgical stage
according to the Musculoskeletal Tumor Society Classification9 was IIB in each patient. Wide resection
of the tumor involving the sciatic nerve (both the tibial nerve
and common peroneal nerve branches) was performed. The muscles that
were resected included the semimembranosus, semitendinosus, and long
head of the biceps femoris in both patients. There were no postoperative
complications. At the time of gait analysis, at five years for one
patient and at two years for the other, neither patient had distant
metastasis or local recurrence.
Because of the sciatic nerve resection, neither patient had muscle
strength, sensation, or proprioception distal to the ankle joint
of the affected limb. In Case 1, the muscle strengths of the involved
limb measured by manual muscle testing10 were
5 for the knee extensors, 3 for the knee flexors, 0 for the ankle
and foot extensors, and 0 for the ankle and foot flexors. In Case
2, the strength was 5 for the knee extensors, 3 for the knee flexors,
0 for the ankle and foot extensors, and 0 for the ankle and foot
flexors. The functional score according to the functional evaluation
system of the International Society of Limb Salvage (ISOLS)11 was 17 (57%) for Case 1
and 15 (50%) for Case 2 (Table I).
The patients had no pain or phantom sensation in the treated limb
(ISOLS pain score of 5) and were satisfied with the treatment result
(ISOLS emotional acceptance score of 4). However, they had partial
occupational restriction (ISOLS function score of 1) and a major
cosmetic gait abnormality with obvious stance-time asymmetry (ISOLS
gait score of 1). Although they could walk outside, walking on stairs
or slopes was difficult (ISOLS walking ability score of 3). Because
both patients required an ankle-foot orthosis to support
the ankle, the highest score that they could be assigned in the
supports category was 3 of 5 (the ISOLS supports score was 3 for
Case 1 and 1 for Case 2). One patient (Case 2) used a cane for walking
outside. She preferred to use a cane for walking a distance because
she felt that it reduced fatigue. During gait analysis sessions,
both patients walked while wearing a plastic ankle-foot
orthosis but not with a cane or crutches.
Gait Analysis
Six individuals with a hip disarticulation (four men and two women;
mean age, twenty-three years) who used a Canadian hip disarticulation
prosthesis with a satisfactory fit and eight healthy subjects (four
men and four women; mean age, twenty-four years) who did not have
neuromuscular or skeletal disease were studied for comparison.
The temporal-distance parameters and kinematic data on walking
were obtained with a motion analyzer12 (ELITE
Plus; BTS [Bioengineering Technology and Systems],
Milan, Italy). The system included four 100-hertz television
cameras, and it detected the positions of reflective markers glued
on the subjects. The markers were placed bilaterally on the acromion,
the anterior superior iliac crest, the center of the greater trochanter,
the lateral joint line of the knee, the lateral malleolus of the
ankle, the lateral side of the base of the calcaneus, and the lateral
side of the base of the fifth metatarsal. The three-dimensional
coordinates of each marker were computed, and the temporal-distance
and kinematic variables (lateral and vertical displacements and
arc of rotation) were calculated with original software written
by one of us (T.M.).
Gait analysis showed that the walking velocity of our two patients
with sciatic nerve resection was slow, the stride length was short,
and the stance-time asymmetry resembled that of the subjects with
hip disarticulation (Table II). Both patients had excessive lateral
displacement of the single sinusoidal curve for each gait cycle
along the line of progression. In contrast, the vertical displacement
of the pelvis during a stride was small. The patients reduced forward
displacement of the uninvolved limb during the stance-time-period
of the involved limb. The arc of pelvic rotation during a stride
was also reduced (Table III).
Ground reaction forces were measured with a force-plate (Kistler,
Winterthur, Switzerland) built into the walkway (Fig. 1). There were
two distinct peaks (at the loading response and terminal stance
phases) separated by a mid-stance valley in the vertical force of
the unaffected limb of the two patients and of both limbs of the
healthy controls. The curve for the the involved limb of the patients
with sciatic nerve resection showed a plateau. In the horizontal
fore-aft plane, the anteriorly directed loading response and the
posteriorly directed driving force of the involved limb were reduced
in both patients.
To clarify the role of the quadriceps muscles during walking in
these patients, dynamic electromyograms were made with an electromyographic
amplifier (Omniace RT3200N; NEC, Tokyo, Japan) and the use of telemetric
surface electrodes. Cathodes were placed on the motor point of selected
muscles, near the center of the muscle, and anodes were placed 3
cm distal to the cathode. As expected, there were no major activities
in the tibialis anterior and gastrocnemius muscles of the involved
limb of the patients. The rectus femoris muscle of the involved
limb had prolonged excessive waves in amplitude and density during
the stance phase.
Limb-sparing is now widely accepted as a treatment option for
the majority of soft-tissue sarcomas of the extremities3,13. There are numerous reports about
the low recurrence rates after resection of the tumors with adequately
wide margins14-17. However,
studies about the functional loss following the procedure are scarce4,18. In the present study, the walking
function and gait characteristics of two patients who had had limb-sparing
surgery with sciatic nerve resection were analyzed with use of a
nonparametric objective evaluation system and a subjective method
of gait analysis.
According to the evaluation system of the ISOLS11, the functional level of the patients
with sciatic nerve resection was as low as 50% of normal.
The functional level of these patients was inferior to that achieved
by prosthetic knee replacement or rotationplasty, which has been
reported to be about 80% of normal6,7,19. However, their
results were better than those of the patients with a hip disarticulation
and a well-fitting prosthesis in our comparison group.
Among parameters of gait analysis, free-walking velocity
is one of the most useful indicators of the overall walking efficiency20.
Decreased velocity associated with a reduced stride length is consistent
with a more unstable and inefficient gait pattern. Patients with
arthritis at either the hip or the knee joint decrease the walking
velocity in an effort to relieve pain. Although our patients had
no pain during walking, they walked much more slowly than controls
did, and they even walked more slowly than subjects with a hip disarticulation and
those with an above-the-knee amputation (as reported in the literature21,22).
Proprioceptive impairment obstructs walking because it prevents
the patient from knowing the position of the leg and the type of
contact with the floor. As a result, the patient is unsure when
it is safe to transfer body weight onto the limb. Muscle weakness
may also contribute to the slow walking velocity. We thought that
the slow walking velocity of our patients was due mainly to the
combined loss of the sensory and motor function of the foot and
ankle.
The magnitude of the vertical ground reaction force changes with
variations in gait speed23. The decreased walking velocity of our
patients may have been responsible for the changes of the vertical
ground reaction force. The reduced fore-aft force on the involved
limb was considered to be the result of both the decreased walking
velocity and the loss of motor function of the foot and ankle. The
anteriorly directed force at the loading response phase was reduced
secondary to the slower walking velocity. A loss of motor function
of the flexors in the foot decreases the driving force during toe-off
and was probably the primary reason for the reduction of the posteriorly
directed force at terminal stance.
The electromyograms of the patients demonstrated excessive activity
in the quadriceps muscles of the involved limb during stance phase;
this may be the compensatory action required to stabilize the knee
and ankle, which were unstable because of the sensory and motor
impairments following the resection of the sciatic nerve and the
hamstring muscles. Additional inefficient muscle work is likely
to increase energy consumption6,24-26,
so patients with sciatic nerve resection may walk with unnecessarily
high energy consumption. Studies of the energy cost of walking for
subjects with sciatic nerve resection would be useful.
As has been demonstrated in patients with proximal femoral replacement27, an assistive device such as a cane
might alter the patterns and efficiency of the gait of patients
with sciatic nerve resection. Kinematic and kinetic studies of the
effects of assistive devices on the walking ability of such patients
are warranted.
Surgery for malignant tumors can be radical, involving the sciatic
nerve, without necessarily depriving the patient of the lower limb.
The present study showed, however, that such limb-sparing
surgery results in a major walking disability. This observation
is in accordance with the findings of Suzuki et al., who reported
the importance of normal foot function for successful limb salvage28. Both patients in the current study
walked more slowly than did subjects with a hip disarticulation.
Amputation has a major cosmetic disadvantage, but it offers greater
opportunities for sports participation29,30.
Nevertheless, both of our patients stated that, despite the functional
limitation, the limb-sparing was preferable to an amputation.
The type of operation (limb-sparing or amputation) must
be decided with each patient. The results of this study may help
surgeons and patients to make a decision regarding limb-sparing
surgery with sciatic nerve resection for the treatment of a soft-tissue
sarcoma.
Note: The authors thank Cherie McCown and Aki Yoshida for assistance
with the preparation of this manuscript.