Between 1987 and 1997, we surgically treated tarsal tunnel syndrome
in thirty-seven patients with a total of forty-four involved feet
(twenty-three right and twenty-one left feet). Of the thirty-seven
patients, seventeen were male and twenty were female, and the average
age was thirty-one years (range, eleven to sixty-five years). Six
patients had a history of trauma to the foot and/or the
ankle. Three of them had been treated nonoperatively for a fracture
of the calcaneus; two had had a sprained foot or ankle, which was
not treated; and the other one had had a contusion in the region
of the tarsal tunnel. Flatfoot deformity was found in four patients (seven
feet); two of these patients (four feet) were found to have an accessory
navicular bone bilaterally. Diabetes mellitus was present in two
patients.
The chief symptom of these patients was numbness or pain in the
foot or a combination of the two symptoms (Table I). The severity
of the pain and the numbness varied, both among the patients and
within an individual patient, and the intensity of the symptoms
varied according to the amount of daily activity or sports participation.
Five patients had pain or numbness after vigorous exercise. Seven
patients (seven feet) reported nighttime pain in the foot. The mean
interval between the onset of symptoms and the final diagnosis in
our clinic was four months (range, one to twelve months).
The physical examination revealed sensory disturbances, ranging
from slight to moderate hypoesthesias, in all of the patients. Thirty-five
feet had hypoesthesias on the medial side of the sole in the distribution
of the medial plantar nerve, seven had hypoesthesias across the
whole sole in the distribution of both the medial and the lateral
plantar nerve, and two had hypoesthesias in the distribution of
both the medial plantar nerve and the medial calcaneal branch of
the tibial nerve. Local tenderness was induced by gentle compression
behind the medial malleolus, with the ankle in a neutral position,
in forty-three feet. The degree of tenderness varied dramatically,
however. In the same position, a Tinel sign, ranging from vague
to definite, was noted in forty-one feet and was absent in three.
Electrodiagnostic studies were carried out on thirty-eight feet.
The terminal motor latency of the medial plantar nerve was prolonged
in thirteen feet. The sensory nerve-conduction velocity between
the medial malleolus and the great toe was reduced in seven feet. Action
potentials of the tibial sensory nerve could not be elicited in
sixteen feet.
In twenty-two patients (twenty-five feet), a space-occupying
lesion or a talocalcaneal coalition was confirmed by imaging examinations.
In nineteen feet without a well-defined lesion, a surgical procedure
was performed because of the failure of nonoperative treatment (medication,
physical therapy, or the use of orthotic devices) after three to
four months. The tibial nerve was completely decompressed in all
of the procedures. The decompression was combined with the Kidner
procedure9 in two patients (four
feet) with a flatfoot deformity, and it was combined with a medial
displacement osteotomy of the calcaneus as well as the Kidner procedure
in one patient (two feet).
The physical test of dorsiflexion of the ankle, eversion of the
foot, and dorsiflexion of all of the toes that was described above
was performed before, during, and after the operation in all of
the patients. Before the operation, we held the foot in the test
position for five to ten seconds and questioned the patient about
any changes in symptoms. With the foot held in the same position,
we checked for the Tinel sign and for local tenderness of the nerve.
The surgical procedure was performed under tourniquet control,
with a curved medial skin incision extending from above to below
the medial malleolus. The flexor retinaculum was cut, but the laciniate
ligament was left intact. Then the tibial nerve, the posterior tibial
vessels, and both the flexor hallucis longus and the flexor digitorum
longus tendons were identified. When the physical test was performed
during the operation, we observed the change in the anatomical relationship
of the structures within the tarsal tunnel. Then we cut the laciniate ligament,
completely releasing the tibial nerve. Again, we observed the anatomical
relationship of the structures within the tunnel as the test was
repeated.
Postoperatively, the patient wore a short leg cast, with the
ankle and foot in a neutral position, for two weeks. Periodically,
we assessed whether the symptoms and the signs disappeared, and
when they had, we determined whether the physical test could reproduce
them. The average duration of follow-up was three years and eleven
months (range, two years to thirteen years and three months).
The test was also performed on fifty normal volunteers (100 feet)
who did not have a history of trauma to the feet or ankles. Twenty-five
of the volunteers were male, with an average age of thirty-five
years (range, ten to eighty-two years), and twenty-five were female,
with an average age of forty-one years (range, eleven to eighty-six
years). We judged whether each subject’s arch was normal,
low, or high by observing the standing heel profile from behind
and by measuring the shape of the sole with a pedoscope10. Eighty-eight
feet were found to have a normal longitudinal arch, six had a low
arch, and the remaining six had a high arch. We observed whether
any symptoms or signs could be induced by performing the test on
the feet of the volunteers.
Performance of the test induced no signs or symptoms in the feet
of the normal volunteers.
When the test was performed before the surgical procedure, nine
of the twenty feet with numbness had an intensification of the numbness,
five of the twenty had no change, and six (four with a ganglion
cyst and two with flatfoot deformity) had pain. Of the seventeen
feet with pain, fifteen had an intensification of the pain and two had
no change. Of the seven feet with both symptoms, four had intensification
of the pain, two had intensification of the numbness, and one had
no change. The tenderness intensified in forty-two of the forty-three
feet that had had tenderness previously, and it was induced in the
one that had not. The pain usually was induced immediately after
the foot was placed in the test position; however, it took several
seconds for numbness to develop. A Tinel sign became more pronounced
in forty-one feet, and the sign was induced in three feet in which it
had been absent.
During the surgical procedure, some changes were observed in
the patients who did not have a space-occupying lesion. After the
flexor retinaculum was cut, except for the laciniate ligament, the
tibial nerve did not appear to be compressed by the laciniate ligament
and the mobility of the nerve was not restricted when the ankle
joint was plantar flexed and the foot was inverted. When the ankle
joint was dorsiflexed and the foot was everted maximally, the tibial
nerve was stretched and it bulged medially. When all of the toes
were then forcibly dorsiflexed, the flexor hallucis longus muscle
belly entered farther into the tarsal tunnel and pressed upon the
stretched tibial nerve from behind. The tibial nerve became markedly
constricted by the superior edge of the laciniate ligament (Fig. 2-A). This constriction
of the nerve disappeared after the laciniate ligament was released
(Fig. 2-B).
In twenty-five feet, a space-occupying lesion was identified
within the tarsal canal, and, in nineteen feet, no such lesion was
identified. The lesions found at surgery included a talocalcaneal
coalition (in sixteen feet), a ganglion cyst (in five feet), venous
varicosities (in four feet), and an anomalous muscle (in four feet).
In these patients, the anatomical relationships of the tibial nerve and
its surrounding structures were more complicated and were more difficult
to observe without completely cutting the flexor retinaculum, including
the laciniate ligament. The osseous mass of the coalition between
the talus and the calcaneus bulged out between the flexor digitorum
longus tendon and the neurovascular bundle into the tarsal tunnel.
The tibial nerve overrode the osseous bulge and became extremely
stretched in the test position.
Similarly, in the patients whose symptoms were caused by an accessory
flexor digitorum longus muscle, the muscle belly was located in
the tarsal tunnel. During the performance of the physical test,
the tibial nerve became compressed by both the flexor hallucis longus
muscle and the accessory muscle. Complete decompression of the tibial
nerve occurred after excision of the anomalous muscle.
The pain and the uncomfortable feelings in the feet were relieved
immediately after the operation in all patients. Numbness, local
tenderness, and the Tinel sign disappeared in 0.5 to 10.5 months
(average, 2.9 months) postoperatively, except in the three patients
whose symptoms were caused by a fracture of the calcaneus. In these three
patients, a weakly positive Tinel sign and local tenderness were
still noted at the final follow-up examination, and the same signs
were elicited by the physical test. The symptoms could not be reproduced
by a repeat performance of the test in any of the other patients.
The diagnosis of tarsal tunnel syndrome is made primarily on
the basis of the medical history and the findings of the physical
examination. The use of a reliable provocative test may increase
diagnostic accuracy.
Tourniquet tests that reproduce the symptoms by inflating a pneumatic
cuff around the leg have been reported by several authors4,11,12;
however, their accuracy and specificity are not known. The physical
findings of this syndrome can be exacerbated by compression of the
tibial nerve or by applying tension to it. Linscheid et al.12 described a test in which compression
was applied distal to the medial malleolus for sixty seconds; positive findings
were elicited in twenty-seven of thirty-four patients. Other tests
have been performed with the foot and ankle placed in specific positions—that
is, they have been performed with the heel forced into a valgus
position3, with the foot dorsiflexed
during straight-leg raises13,
and with the foot held in inversion4,11,14.
However, details of the reliability of these tests, except for Lam’s
inversion test11, were not described
in the reports. As for Lam’s test, it reproduced the symptoms
in only two of ten patients.
When our test was performed, the symptoms did not change in eight
(18%) of forty-four feet. This percentage is similar to
the results of Linscheid et al. (20.5%)12.
Less time is required for patients to experience increased numbness
or pain with use of our method; when the symptoms changed, they
did so within ten seconds. However, it took sixty seconds to reproduce
symptoms with the method of Linscheid et al.12.
The pain induced by our test is different from that of plantar
fasciitis, although forced extension of the toes does resemble the
method used to stretch and irritate the plantar structures attached
to the heel in that condition. The pain induced by our test is also
different in character from and more severe than that caused by
a talocalcaneal coalition.
To confirm the diagnosis of tarsal tunnel syndrome, both a positive
Tinel sign and local tenderness over the tibial nerve should be
verified by tapping on or compressing the tibial nerve at the tarsal
tunnel in the position described. In addition to inducing a Tinel
sign in patients in whom it had not been evoked previously, testing
in this position resulted in already existing signs becoming more
pronounced.
It should be noted that neither symptoms nor signs were detected
in the feet of the normal volunteers who were tested. Therefore,
our method seems to be more specific than the other methods previously
described for the diagnosis of tarsal tunnel syndrome.
In the tarsal tunnel, the neurovascular bundle is accompanied
by loose adipose tissue and is fixed relatively firmly to its surroundings.
In one cadaveric study, tibial nerve tension was also significantly
increased by eversion (p = 0.0001), dorsiflexion (p = 0.0006),
and combined dorsiflexion-eversion (p = 0.0001)15. In another study,
tarsal tunnel pressure was shown to increase when the foot and ankle
were positioned in full eversion16. Our operative findings indicated
that our physical test is different from the previously reported ones
because it applies both tension and compression to the tibial nerve.
In addition, the flexor hallucis longus muscle belly comes in proximity
to the nerve and compresses it when both the great and the lesser
toes are maximally dorsiflexed.
Clinical symptoms and signs of tibial nerve compression at the
tarsal tunnel were reproduced and/or intensified by this
provocative test in the vast majority of our patients. The maneuver
is safe for the patient and easy for the clinician to perform. This
technique will help to increase the sensitivity of the physical
examination in the diagnosis of tarsal tunnel syndrome, especially
when the clinical signs are vague. References