The clinical symptoms associated with an interdigital neuroma
were initially reported by Civinni in 18351 and
later by Durlacher in 18452. Although
Morton3 mistakenly attributed
this condition to an "affection of the fourth metatarsophalangeal
joint," his name, nonetheless, is most commonly associated
with the lesion. Perineural fibrosis has been documented on histological
evaluation of surgical specimens4-24.
This finding has led many investigators to propose a symptomatic
entrapment neuropathy of the interdigital nerve at the distal extent
of the transverse intermetatarsal ligament as the cause of an interdigital
neuroma4,8,10,13,14,17,19,25-29.
The demographics of interdigital neuromas have been well documented;
they include a higher prevalence in women11,15,23,24,26,28,30-34,
a mean age at onset in the fifth decade15,16,23,24,31,32,35,36,
the most common location in the third intermetatarsal space4,5,8,20,23,26,31,37, and exacerbation
of symptoms with constricting shoe-wear4,5,9,15,16,34,38-41.
Although several authors have reported a high rate of satisfactory
postoperative results4,8,11,15,24,26,34,36,41,
many reports are characterized by short-term follow-up18,26,33,34,42 and a lack of postoperative
clinical evaluation11,12,31,33,36,43,44.
Other confounding factors in published reports on the operative
treatment of neuromas include the involvement of multiple surgeons35, simultaneous surgery in an adjacent
web space11,23,31,35,43, and the
inclusion of patients with both primary and recurrent neuromas15,31,35, making the evaluation of
success rates following primary neuroma excision difficult to assess.
Confounding diagnoses such as inflammatory and degenerative arthritis
as well as instability of a lesser metatarsophalangeal joint require
preoperative radiographic assessment, which has previously been documented
in only one study35.
The purpose of the present study was to document the long-term
clinical results of surgical excisions of primary interdigital neuromas
performed by a single surgeon.
Demographics
From January 1991 to January 1997, 121 consecutive patients (131
feet, 136 neuromas) underwent surgical excision of a persistently
symptomatic interdigital neuroma by the senior author (M.J.C.).
Their records were retrospectively reviewed, and thirty-nine
patients (forty-two feet, forty-four neuromas) were excluded.
The reasons for exclusion were death (two patients), a diagnosis
of inflammatory arthritis (two patients), a diagnosis of recurrent
neuroma following failed prior surgery (eleven patients), and a
diagnosis of concomitant instability of a lesser metatarsophalangeal
joint (twenty-four patients). A diagnosis of instability of a lesser
metatarsophalangeal joint required radiographic demonstration of
medial deviation of the second or third metatarsophalangeal joint
of >5°44 with concurrent
pain on palpation of the plantar aspect of the second or third metatarsophalangeal
joint44-46, a positive drawer
sign47, and relief of pain following
an intra-articular injection of lidocaine hydrochloride48. Of the remaining eighty-two
patients who had undergone excision of an interdigital neuroma,
sixty-six (81%) (seventy-one feet, seventy-four
neuromas) were successfully contacted and returned for examination
by another orthopaedic surgeon (T.P.). All of the patients’ records,
including preoperative and postoperative chart notes and radiographs,
operative reports, and pathology reports, were reviewed.
During the study period, fifty-eight patients had unilateral neuroma
excision, five patients had bilateral neuroma excision, and three
patients had neuromas in adjacent intermetatarsal spaces in the
same foot excised at different times (with nine to twenty-six
months between the initial and second resections) by the senior
author; thus, there was a total of seventy-four neuroma
excisions in seventy-one feet. In addition to the five patients
who had a neuroma on each foot excised within the time-frame of
this study, nine patients had a neuroma excised from the contralateral
foot not within the time-frame of this study (one to sixteen years [average,
nine years] before or after the study). Also, in addition
to the three patients who had adjacent excisions performed by the
senior author during the study period, one patient had an adjacent neuroma
excised by another surgeon twelve months prior to the index procedure,
one had an adjacent neuroma excised by another surgeon twelve months
after the index neuroma excision, and one patient had an adjacent
neuroma excised by the senior author two years following the conclusion
of this study.
Preoperative Findings
The indication for surgery was intractable pain isolated to either
the second or the third intermetatarsal space that was refractory
to conservative treatment such as the use of orthotic devices, metatarsal
pads, nonsteroidal anti-inflammatory medications, or shoe-wear
modification; this was present in all seventy-four cases. The magnitude
of the preoperative pain, however, was not quantitated. A vague
description of ill-defined pain in the forefoot was typical at the
initial evaluation, following which the patients were requested
to exacerbate the symptoms with the use of constricting shoe-wear
and increased physical activity in an attempt to help them to isolate
the pain to a discrete area. If at repeat evaluation the exact location
of the pain remained unclear, 1% lidocaine hydrochloride
was injected into the most symptomatic interspace in an attempt
to temporarily alleviate the symptoms. The temporary relief of symptoms
following an injection was correlated with the history and the findings
on physical examination to localize the painful interspace. Diagnostic
injections were used in forty-four (59%) of the seventy-four
cases to assist in determining the location of the neuroma. The
number of visits prior to the surgery averaged 2.1 (range, one to
four). Metatarsal pads or orthotic devices were used preoperatively
by twenty-six (39%) of the sixty-six patients. Radiation
of neuritic pain (twenty-six [35%] of
the seventy-four cases), numbness of an adjacent digit (twenty-two [30%]),
and a positive Mulder sign49 (Fig. 1) (thirty [(41%])
were found on preoperative physical examination.
Follow-up Evaluation
The average duration of postoperative follow-up was
sixty-nine months (range, thirty-eight to 121 months). At the time of
final follow-up, patients were asked to characterize the pain as
none, mild, moderate, or severe. They were also asked to characterize
their footwear restrictions as none (can wear any style of footwear
at any time), mild (unable to wear narrow shoes or high heels but
able to wear all other types of shoes), or major (difficulty in
finding comfortable shoes). In addition, they were asked to characterize
activity restrictions as none (no restrictions, with the ability
to participate in any sports or recreational activities), mild (mild
restrictions, but not enough to interfere with everyday activities),
or major (severe restrictions, with an inability to perform daily
activities or work). Patients were asked to rate their overall satisfaction
with the result of the surgery as excellent, good, fair, or poor
according to our previously published scale50.
With this scale, patients rate the result as excellent if they have
no problems related to the foot, are very satisfied, have mild or
no pain, and walk without difficulty; as good if they have few problems,
are satisfied, have mild pain, walk without difficulty or with mild
difficulty, and would have the surgery again under similar circumstances;
as fair if they have moderate pain, some difficulty walking, and
reservations about the success of the surgery; and as poor if they
have continued pain, have little improvement in walking, and regret
having had the surgery.
At the time of final follow-up, the physical examination included
inspection and palpation of the foot with attention to any sensory
deficits (on light touch with pinprick), the presence of intractable
plantar keratoses, fat-pad atrophy, incisional pain, and interspace
or metatarsophalangeal joint tenderness. A positive percussion test
of the involved nerve, the presence of a Mulder sign49 (Fig. 1), and the presence of instability
of a lesser metatarsophalangeal joint (a drawer sign46-48) (Fig. 2) were evaluated as well.
Radiographic Evaluation
Preoperative weight-bearing radiographs were reviewed and compared
with standardized postoperative weight-bearing radiographs at the
time of final follow-up. These radiographs were inspected for evidence
of degenerative or inflammatory arthritis as well as for evidence
of malalignment associated with instability of a lesser metatarsophalangeal
joint as defined above.
Surgical Technique
All operations were performed, with use of a standard technique,
by the senior author (M.J.C.). Under tourniquet control, a 3-cm
dorsal longitudinal incision is centered over the involved interspace.
The transverse intermetatarsal ligament is placed under tension
with a self-retaining retractor and is divided. A small periosteal
elevator is then used to delineate the involved nerve both proximally
and distally. The digital nerves distal to the bifurcation of the
common digital nerve are transected. The dissection is then carried
proximally into the interspace, isolating the common digital nerve
approximately 3 cm proximal to the bifurcation, where it is transected.
(As described by Mann and Reynolds15 and
by others24,51, it is important
to sever any adjacent capsular nerve branches at this time as they
may prevent proximal migration of the nerve stump, increasing the
possibility of a recurrent symptomatic neuroma.)
The patient is allowed to walk wearing a postoperative shoe with
full weight-bearing, as pain permits, immediately following the
surgery.
All statistical analyses were performed with use of Microsoft Excel
(Redmond, Washington). A chi-square test was used to compare levels
of postoperative subjective satisfaction. The comparisons of postoperative
satisfaction following bilateral surgery and that following surgery
in adjacent interspaces were performed with use of a Fisher exact
test (a variation of the chi-square test) because one of the cell
sizes was quite small.
Fifty-two (79%) of the sixty-six patients
were female, and fourteen (21%) were male. The average
age at surgery was fifty years (range, fifteen to seventy-seven
years). Thirty-five neuromas occurred in the right foot
and thirty-nine, in the left. Sixteen (22%) of
the seventy-four neuromas occurred in the second intermetatarsal
space, and fifty-eight (78%) occurred in the third
intermetatarsal space (ratio, 1:3.6).
Overall satisfaction was rated as excellent by forty patients (61%),
good by sixteen (24%), fair by five (8%), and
poor by five. At the time of final follow-up, thirteen feet were
mildly painful, nine were moderately painful, and three were severely
painful; the remaining forty-six feet (65%) were pain-free.
A peripheral neuropathy later developed in one of the ten patients
who had an overall satisfaction rating of fair or poor, a peripheral
nerve injury developed following regional anesthesia in another,
a reflex sympathetic dystrophy developed in one, and severe instability
of the second metatarsophalangeal joint developed in one. Of the
five feet that were subjectively rated as poor, four were noted
to have moderate or severe pain localized to the area of the neuroma
resection. The fifth patient who rated the result as poor was dissatisfied because
of persistent numbness of the lesser toes.
Of the fourteen patients who underwent bilateral neuroma excision
(either simultaneously or on a delayed basis), eleven had a satisfaction
rating of good or excellent. Of the six patients who had adjacent
neuroma excisions, with one procedure performed either prior to
or as a delayed procedure after the other, four reported an excellent
or good result and two noted a fair or poor result. With the numbers
available, the proportion of patients with an excellent or good
result was statistically the same whether a unilateral, bilateral,
or adjacent neurectomy had been performed (bilateral compared with
unilateral procedures, p = 0.69; adjacent compared with unilateral
procedures, p = 0.40).
One deep wound infection was treated with incision, drainage,
and open packing, and it went on to heal successfully; at the time
of final follow-up, the patient had an excellent result. Minor wound
complications characterized by persistent erythema or serous drainage
lasting for longer than ten days developed in six patients. These
patients were treated with oral antibiotics, and all wounds healed
successfully. Of these six patients, three rated the result as excellent;
two, as good; and one, as poor. Four of the seventy-one
feet underwent subsequent surgery; a neurectomy in the adjacent
interspace was done in two, and other, unrelated surgery was done
in two. The site where the interdigital neuroma had been excised
was not reexplored in any foot.
Tenderness on palpation in the surgically treated interspace was
noted at the time of final follow-up in forty-five (61%)
of the seventy-four cases, although patients infrequently
complained or were aware of pain in this region. There was tenderness
on direct palpation of the plantar aspect of seven lesser metatarsophalangeal
joints. Three of these feet were rated as poor; one, as fair; two,
as good; and one, as excellent. A positive drawer sign that had
not been present preoperatively was observed in eleven lesser metatarsophalangeal joints
postoperatively. Of these joints, five appeared normal radiographically
and only two had a combination of tenderness on palpation, a positive
drawer sign, and radiographic evidence of instability. These two
lesser metatarsophalangeal joints were in two patients who complained
of pain and rated the result as poor. A positive Mulder sign, present
in thirty of the seventy-four cases preoperatively, was
absent in all but two of the thirty cases postoperatively. The result
in both cases was rated as excellent at the time of final follow-up.
There was a discrepancy between the subjective sensation of numbness
and the result of the sensory examination at the time of final follow-up.
Upon questioning, patients reported numbness in thirty-six
(51%) of the seventy-one feet, but the numbness
was considered bothersome in only four feet. In contrast, the objective
findings included normal sensation in two (3%) of the seventy-one
feet, numbness between the digits in fifty-one (72%),
numbness in the plantar web space in forty-six (65%), and
numbness in the terminal aspect of the digits in twenty-nine
(41%). Several patients had more than one area of numbness
(Fig. 3).
A positive nerve percussion test (Tinel sign) was noted in one dorsal
incision, in a patient who rated the result of the surgery as excellent.
All other incisions were well healed and asymptomatic. A positive
nerve percussion test was observed in twenty-six (35%)
of the seventy-four involved interspaces at the final examination,
although prior to that time none of the patients were aware of this
finding and it was not mentioned as a subjective complaint.
A discrete intractable plantar keratosis developed beneath a lesser
metatarsal head in eleven of the seventy-one feet, although
there was clinical and radiographic evidence of instability of the
metatarsophalangeal joint in only one of them. The result was rated
as good or excellent in nine of these eleven feet and as poor in
two at the time of final follow-up. Diffuse fat-pad atrophy developed
following the surgery in two other patients, yet both rated the
final result as good.
Twenty (30%) of the sixty-six patients reported
no shoe-wear restrictions, and thirty-five reported mild
restrictions. Of eleven patients with major shoe-wear restrictions,
five reported continued pain related to the neuroma surgery and the
other six had problems unrelated to the neuroma surgery. Postoperatively,
twenty-three (35%) of the sixty-six patients used
metatarsal pads or orthotic devices although fourteen had not used
them before surgery. Fifteen of those who had used orthotic devices
or metatarsal pads prior to surgery discontinued their use after
surgery. Forty-one (62%) of the sixty-six patients
stated that they had no activity restrictions at the time of long-term
follow-up, whereas twenty-three had mild restrictions
and two had major restrictions.
At the time of surgery, seventy-two of the seventy-four
specimens were submitted for histologic examination. All seventy-two
specimens were reported by the pathologist to be consistent with
an interdigital neuroma.
Comparison of the preoperative and postoperative weight-bearing
anteroposterior radiographs demonstrated that moderate or severe
malalignment of a lesser metatarsophalangeal joint had developed
in six feet. An asymptomatic positive drawer sign had also developed
in three of these six feet. All of these patients had had a negative
response to an intra-articular injection of lidocaine and a negative
drawer test preoperatively. Pain did not develop in the second metatarsophalangeal
joint in any of the six feet. Of these six feet, four were subjectively
rated as excellent and two, as good at the time of final follow-up.
To the best of our knowledge, this investigation represents the longest
reported follow-up in a series of primary resections of interdigital
neuromas. The term metatarsalgia is commonly used to describe ill-defined
forefoot pain, for which a neuroma is only one of several possible
causes. It can be a diagnostic challenge to differentiate other
conditions such as inflammatory or degenerative arthritis, a metatarsal
stress fracture, and instability of a lesser metatarsophalangeal
joint52.
Location of Pain Preoperatively
In this series, fifty-six (85%) of sixty-six
patients reported a good or excellent result. We believe that the
key to this high level of satisfactory results was the careful preoperative
clinical evaluation in which other, confounding diagnoses were eliminated.
Exact localization of pain can be assisted with the use of sequential
injections of a small volume (about 1 ml) of 1% lidocaine
hydrochloride in the intermetatarsal space and later in the adjacent
metatarsophalangeal joints. This test was useful in forty-four
(59%) of seventy-four cases. A patient can compare
the relief obtained with each separate injection to better define
the area of pain. With a larger volume of lidocaine, a larger area
may become numb, leading to an incorrect diagnosis. The use of sequential
injections is only one part of the diagnostic process, which requires
correlation with the history as well as clinical and radiographic
evaluations. In 41% of the cases, there was no need for
an injection because the patient was able to discern the specific
area of discomfort. The low rate of later reexploration of adjacent intermetatarsal
spaces (two of the seventy-one cases) and the fact that no surgically
treated interspace was reexplored following the index surgery support
our premise that careful repetitive evaluation combined with diagnostic
injections is useful in assisting patients who have difficulty in
finding the exact area of pain.
Preoperative Radiographic Examination
The preoperative radiographic examination also helped us to identify
patients with forefoot pain that was not due to a neuroma. Twenty-four
patients with radiographic evidence of instability of a lesser metatarsophalangeal
joint as well as physical evidence of the instability (a positive
drawer sign and metatarsophalangeal joint pain) were eliminated
from this series preoperatively in spite of the fact that they also
had symptoms of an interdigital neuroma. These patients underwent
both reconstruction of the second metatarsophalangeal joint and
neuroma excision; a future report will document the surgical technique
and clinical course of these patients53.
Preoperative Physical Examination
Mulder49 and others8,16 have reported a palpable click
with compression of the involved interspace between the examiner’s
index finger and thumb as the transverse arch is compressed. While
it may be hypothesized that the thickened interdigital nerve subluxates beneath
the adjacent metatarsal heads and the intermetatarsal ligament with
this maneuver, the anatomical basis of this sign has never been
substantiated. Nonetheless, surgery successfully eliminated this
finding in twenty-eight of thirty cases.
Postoperative Findings
Scar Sensitivity
At the time of final follow-up, sensitivity at the site of the nerve
transection was still present in more than one-third of the
cases, although no patient was aware of this finding until he or
she was examined. Patients should be counseled that the nerve transection
leaves a discrete area of tenderness that frequently remains but
that retraction of the cut nerve should result in minimal if any
symptoms.
Residual Pain and Tenderness
It is not uncommon for a patient to have residual pain in either the
involved interspace or an adjacent lesser metatarsophalangeal joint
following neuroma surgery and yet still have a satisfactory result.
Interspace tenderness was a frequent finding on physical examination
(forty-five [61%] of the seventy-four
cases), but it was rarely recognized by patients prior to the final
follow-up examination and there was persistent interspace
pain related to the neuroma surgery in only six of the forty-five
cases. (Five were rated as having a poor result.)
Problems Related to the Lesser Metatarsophalangeal
Joint
There was no preoperative instability of a lesser metatarsophalangeal
joint in any of the seventy-four cases in this series.
At the time of final follow-up, a single lesser metatarsophalangeal
joint was demonstrated to have a positive drawer sign in eleven
feet but only six were noted to be painful with this maneuver. We
are not sure of the clinical relevance of instability of a lesser
metatarsophalangeal joint in relation to a previous neuroma excision;
however, the instability was not necessarily associated with metatarsophalangeal
joint pain or abnormal radiographic findings. Karges33 reported tenderness of a lesser
metatarsophalangeal joint following neuroma surgery in fifteen of
thirty-five feet seen at the time of follow-up. In the
current series, seven of seventy-four adjacent lesser metatarsophalangeal
joints were tender. Also, an intractable plantar keratosis developed
beneath an adjacent lesser metatarsal head postoperatively in eleven (15%)
of seventy-one feet. Karges noted that, in thirteen (23%)
of fifty-seven cases, an intractable keratotic lesion developed
following surgery. Patients should be alerted to the possibility
of the development of an intractable plantar keratosis following
excision of an interdigital neuroma, but symptoms are seldom severe.
Sensory Deficit
The magnitude of the postoperative sensory deficit has been assessed
in very few studies15,35,43. Mann
and Reynolds15 observed that there
is substantial variability in the cutaneous innervation of both
the web space and the plantar aspect of the foot adjacent to the
web space. Previous investigators have noted normal sensation after
neuroma excision in a relatively high percentage of patients (range,
29%35 to 34%15). In the current series, subjective
numbness was observed in thirty-six (51%) of seventy-one
feet, but it varied considerably from patient to patient. It was
bothersome in only four of the seventy-one feet. All six
patients who had neuromas excised from adjacent web spaces described
wide areas of numbness involving the plantar aspect of the web space,
the tips of the toes, and the area between the toes, confirming
the observation of Benedetti et al.43 that
excision of adjacent common digital nerves leaves a substantial
neurologic deficit. Patients should be counseled preoperatively
that postoperative numbness of a varying degree is commonly associated
with resection of an interdigital neuroma.
Shoe-Wear and Activity Limitations
Shoe-wear8,15,31,33 and activity
limitations8 have previously been
reported following neuroma resection. In the current series, 70% of
the patients had mild or major shoe-wear restrictions and 38% had
mild or major activity restrictions. Patients should be advised
that shoe-wear restrictions are likely following surgery but that
major activity limitations are uncommon.
Surgical Approach
While both a dorsal4,8,9,15,23,26,32,54 and
a plantar approach5,12,13,16, 23,30,33,49,55-57 have
been recommended for the resection of an interdigital neuroma, the
senior author has preferred a dorsal incision because of the low
prevalence of wound complications and the increased ability of patients
to bear weight immediately postoperatively. The minimal rate of
wound complications (two [3%] of seventy-four
incisions) reported in the present series is distinctly lower than
the complication rates following a plantar approach, which have
ranged from 10%33 to
36%24.
The excisions of neuromas from adjacent interspaces were never
performed simultaneously in the current series. Indeed, adjacent
neuromas are uncommon23,28,31,37,43,
and careful preoperative evaluation should enable one to localize
the problem to a single interspace. Friscia et al.31 reported that simultaneous adjacent
web-space exploration, in nine cases, was associated with a much
higher dissatisfaction rate. In our six patients in whom adjacent
neurectomies were done in a staged fashion, a more extensive area
of numbness persisted in the foot and the satisfaction rate was
somewhat lower.
While Greenfield et al.32 concluded
that early excellent results often deteriorate with time after excision
of an interdigital neuroma, we observed a high level of continued
satisfaction after an average duration of follow-up of
sixty-nine months. Residual pain in either an involved
interspace or an adjacent metatarsophalangeal joint, sensitivity
at the level of the nerve transection, and variable areas of numbness
are common findings following resection of an interdigital neuroma,
but they frequently are not recognized by patients and rarely require
treatment.