Journal of Surgical Oncology 47:17-20 (1991)

Primary Soft Tissue Tumours of the Pelvis

Causing Referred Pain in the Leg

~~~ ~~ ~ ______

Referred pain in the leg is occasionally due to a pelvic soft tissue tumour.
Among 11 patients who presented this way, one had a lymphoma, one had
a benign schwannoma, and nine had soft tissue sarcomas. Most patients
had undergone a variety of procedures, including laminectomy ,before the
correct diagnosis was established. In five cases, an accurate diagnosis was
obtained by needle biopsy. The lymphoma responded to chemotherapy,
and the benign schwannoma was excised. Of the nine patients with soft
tissue sarcoma, six underwent marginalhtracapsular excision, three
receiving supplementary radiotherapy, and two were treated by nonsur-
gical means. Hindquarter amputation was technically impossible or
inappropriate in these cases. All those with high-grade tumours have died
or have metastases. Of four patients with low-grade tumours, three have
exhibited only slow disease progression. Careful judgment and a precise
histopathological diagnosis are required in planning treatment for patients
with pelvic soft tissue tumours causing referred pain in the leg.

KEYWORDS:pelvis, soft tissue sarcoma, treatment planning

INTRODUCTION referred pain in the leg. The records of these patients

While the most common cause of referred pain in the
leg is a prolapsed intervertebral disc, alternative diag-
noses must always be considered. When the underlying
cause is nerve infiltration by advanced pelvic malig-
nancy, the diagnosis may be evident from the history and
clinical examination. By contrast, primary soft tissue
tumours often cause little in the way of symptoms and
frequently remain unrecognized until they are large,
particularly when they are related to the pelvic girdle
[Barret al., 19891. We have reviewed a series of patients
with soft tissue tumours related to the pelvic girdle
causing referred pain in the leg to examine their manner
of presentation, diagnosis, and subsequent management.
PATIENTS AND METHODS
Among more than 800 patients referred to the Soft
Tissue Sarcoma Unit at the Royal Marsden Hospital
between 1970 and 1990, 248 patients had pelvic or
retroperitoneal tumours of whom 11 presented with
0 1991 Wiley-Liss, Inc.
were reviewed for details of their clinical history, diag-
nosis, and management.
At operation, surgical procedures were categorized
according to Enneking’s classification [Enneking et al.,
19811. Zntracapsular excision is equivalent to surgical
debulking, marginal excision refers to enucleation of the
tumour in the plane of its capsule, wide excision involves
excision of the tumour with a margin of normal tissue,
and radical excision implies excision of the entire
anatomical compartment within its fascia.
In patients with tumours lying within the pelvic cavity,
the tumour was removed through a laparotomy incision.
In those in whom the bulk of the tumour occupied the
buttock, the tumour was approached through a curved
incision, as illustrated in Figure 1 .
Accepted for publication January 18, 1991.
Address reprint requests to Mr. J. Meirion Thomas, Royal Marsden
Hospital, Fulham Road, London, England SW3 6JJ.
18 Ball et al.
Fig. 1. Postoperative appearance of the surgical incision used to
approach a soft tissue tumour occupying the buttock. Note the scar
from a previous laminetomy.
RESULTS
There were eight female and three male patients. Mean
age at presentation was 50 years (range 21-81 years).
Pain was referred in the distribution of the femoral nerve
in three patients and in that of the sciatic nerve in eight.
Five patients had objective neurological signs including
absent reflexes and muscle wasting.
In three patients, a mass was detected at the same time
that referred pain was first experienced. One of these
patients had previously undergone intracapsular excision
of a soft tissue tumour on the right side of the pelvis and
a diagnosis of malignant fibrous histiocytoma (MFH) had
already been established. Local recurrence had occurred
despite adjuvant radiotherapy. In the other two patients,
the diagnosis was obtained by incision biopsy in one and
Tru-Cut biopsy in another.
Eight patients presented with referred pain in the
absence of symptoms or signs of an underlying tumour,
although in one patient a diagnosis of von Recklinghaus-
en’s neurofibromatosis had been previously established.
The median period from the onset of symptoms to
definitive diagnosis among these patients was 21 months
(range 4-36 months)-the shortest period being for the
patient with neurofibromatosis, the longest for a patient
with a low-grade liposarcoma (Table I).
In seven of the patients who presented with referred
pain alone, a mass had become palpable by the time of
definitive diagnosis, and in one patient an abnormality on
plain radiography was apparent. In another patient bony
metastases were evident on plain radiography. Four
patients had consulted more than one specialist before the
correct diagnosis was reached. Two had undergone
radiculography , one had undergone laparoscopy , and
another had undergone an inconclusive laparotomy. One
patient had been treated by physiotherapy, two by spinal
epidural and one had undergone a laminectomy (Fig. 1).
In six patients, the tumour was situated in the pelvic
cavity (Fig. 2), while in eight the bulk of the tumour
occupied the buttock; in three of these cases there was
extension through the greater sciatic foramen (Fig. 3).
The diagnosis was eventually established by open biopsy
in one patient, Tru-Cut biopsy in four and laparotomy in
another three after assessment by computed tomography
(CT).
Histological diagnoses are given in Table I. One
patient had a B-cell lymphoma, one had a benign
schwannoma, and nine had a variety of soft tissue
sarcomas, including one extraskeletal chondrosarcoma.
Of the soft tissue sarcomas, five were high grade and four
low grade.
The patient with lymphoma was treated with chemo-
therapy, achieving complete resolution of the mass
within 3 months. The patient with a benign schwannoma
had the tumour completely excised at laparotomy. Of the
nine patients with soft tissue sarcomas, including the
three who originally presented with a mass, two under-
went marginal excision (one with sacrifice of the femoral
nerve), and four underwent intracapsular excision. Of
these six patients, one received preoperative radiother-
apy, two were given postoperative radiotherapy, and
another received systemic chemotherapy after surgery. A
further patient received radiotherapy alone, another (with
metastases) received chemotherapy alone, and one re-
ceived no treatment apart from analgesia (Table I).
None of the patients with soft tissue sarcoma under-
went hindquarter amputation. This was considered tech-
nically unfeasible in the three patients who were treated
by nonsurgical means and in three of the four patients
who underwent intracapsular excision. One patient, with
a low-grade (lipoma-like) liposarcoma, was technically
suitable for hindquarter amputation but this was regarded
as inappropriate treatment in view of the favourable
prognosis of this type of tumour, and an intracapsular
excision was performed.
None of the three patients who originally presented
with a mass as well as referred pain survived more than
six months: all had high-grade tumours. Of the eight
patients who initially presented with referred pain alone,
all but one remain alive at a median period of 27 months
(range 3-45) following referral. Of those with sarcoma
presenting with referred pain alone, one has died and two
with high-grade tumours have metastases. In contrast,
three patients with low-grade tumours have exhibited
only slow progression of residual disease. Only two
patients have obtained symptomatic relief of referred pain
as a result of surgery and two patients, including the one
in whom the femoral nerve was sacrificed, have a
permanent neurological deficit.
 Pelvic Soft Tissue Tumours 19
TABLE 1. Referred Pain From Pelvic Soft Tissue Tumours: Patient Details, Histology,
and Management*
Time
to diagnosis FO~~OW-UP
Diagnosis (mo) Agelsex Grade Treatment (mo)
Referred pain and mass
MFH - 701M High Analgesia d.4
Liposarcoma - 811F High M, RIT d.5
Leiom yosarcoma - 611F High I, CIT d.4
Referred pain alone
Schwannoma 12 251F Benign M 45 NED
Lymphoma 5 411F High CIT 3 NED
MPNST 35 481F Low RIT 40 LD
Extraskeletal 18 53lF Low I, RIT 15 LD
chondrosarcoma
Liposarcoma 24 591M Low I 21 LD
Haemangioperic ytoma 31 561F Low RIT, I d.11
MPNST 4 211M High M 21 LD, mets
MFH 24 31lF High CIT 12 LD, mets
* MFH, malignant fibrous histiocytoma; MPNST, malignant peripheral nerve sheath tumour; M,
(marginal) excision; I, intracapsular excision (debulking); RIT, radiotherapy; CIT, chemotherapy; LD,
local disease; NED, no evidence of disease; mets, metastases; d, death.

DISCUSSION
The present series consists of patients with a miscel-
laneous group of tumours whose common characteristic
was their presentation with referred pain in the leg. All
the tumours were related to the pelvic girdle, lying either
within the pelvic cavity or in the buttock region. Most
tumours remained undetected for several months after the
onset of pain. There was no obvious relationship between
the duration of symptoms before detection and the
pathological subtype of the primary tumour but the period
before detection was longer in patients with low-grade
tumours than in those with high-grade tumours.
Several patients received inappropriate treatment, in-
cluding surgery, before the definitive diagnosis was
established. This is hardly surprising, given the rarity of
Fig. 2. CT scan showing a malignant peripheral nerve sheath tumour
these tumours. In only one patient with von Reckling- arising from the femoral nerve in a patient with von Recklinghausen’s
hausen’s neurofibromatosis, which is associated with neurofibromatosis.
about a 5% risk of malignancy [Fisher, 19891, was the
diagnosis suspected soon after the onset of symptoms.
Pelvic CT at an earlier stage would perhaps have resulted
in detection of these tumours before they became clini- sarcoma and the subtype and grade can usually be
cally palpable and in some cases this might have led to determined by an experienced pathologist from a Tru-Cut
more successful intervention. specimen [Ball et al., 19901.
The importance of obtaining a precise histological The conventional approach to tumours in the buttock is
diagnosis is illustrated by the patient with lymphoma in through an oblique incision overlying the tumour mass
whom surgery was thereby avoided. In five cases, the [Karakousis, 19851. We prefer the curved incision illus-
diagnosis was obtained by Tru-Cut biopsy, a technique trated above because it enables the sciatic nerve to be
we strongly recommend, as it does not compromise identified beyond the perimeter of the tumour at the
subsequent surgery and can be performed under local outset. In a patient with a tumour confined to the buttock,
anaesthetic. Contrary to popular belief a generous tissue widelradical excision can be undertaken using this inci-
sample is not required for an accurate diagnosis of sion. Moreover, if a hindquarter amputation is subse-
20 Ball et al.
Fig. 3. CT scan showing a dumbbell tumour (haemangiopericy-
toma), T, in the greater sciatic foramen.
quently considered necessary, it is possible to incorporate
the incision in the definitive operation.
In the present series, none of the patients with sarcoma
underwent hindquarter amputation. In most cases this
was not technically feasible according to criteria we have
reported elsewhere [Watkins and Thomas, 19871 and
only palliative debulking was possible. In three patients,
conservative excision was undertaken even though a
greater margin of clearance might have been achieved by
hindquarter amputation. Two of these patients had high-
grade tumours, and distant metastases subsequently de-
veloped, justifying the decision to proceed conserva-
tively. The third patient had a low-grade liposarcoma,
regarded by some as an atypical lipoma [Azumi et al.,
19871, and the presence of residual disease was accepted
as the price of limb conservation in the knowledge that
long-term survival was unlikely to be prejudiced.
Radiotherapy was used as the sole means of treatment
in one patient with sarcoma and as supplementary
postoperative treatment in two others in whom a satis-
factory margin of clearance was not achieved. In a further
case, radiotherapy was given preoperatively in an attempt
to reduce tumour bulk. Although a partial response was
achieved, the margin of clearance was not improved.
Radiotherapy was only used in patients in whom the
primary component of the tumour occupied the buttock;
it was not used in those with intrapelvic disease because
of the risk of damage to adjacent structures.
A recent report from this unit has drawn attention to the
problems posed by soft tissue sarcoma of the limb girdles
[Barr et a]., 19891. Because of the size and anatomical
position of these tumours, adequate tumour clearance is
often difficult or impossible to achieve by either conser-
vative surgery or amputation and supplementary treat-
ment is usually required. Patients with high-grade tu-
mours have a poor prognosis, and a conservative
approach is justifiable. Patients with low-grade tumours,
on the other hand, have better survival prospects but are
at risk of local recurrence after conservative surgery,
which may not be salvageable by subsequent amputation.
Although it is tempting to advise amputation when
possible as a primary measure in such cases to ensure
local control, this may not be appropriate if residual or
recurrent disease is unlikely to cause serious disability or
compromise survival, as in the case of the patient with a
low-grade liposarcoma.
Surgery failed to alleviate the symptoms of referred
pain in the majority of our cases, despite the reduction in
tumour bulk. Pain control might have been more satis-
factory if complete excision had been accomplished in a
greater proportion of patients, but it is likely that sacrifice
of the nerve would have been required to achieve this
objective. Soft tissue sarcomas are slow to invade
epineurium [Westbury, 19891, so it is sometimes possible
to preserve a major nerve by undertaking a marginal
excision and giving postoperative radiotherapy. This
policy is only acceptable, however, when it is necessary
to undertake marginal excision for reasons other than that
of simply preserving the nerve.
Patients with referred pain in the leg attributable to soft
tissue tumours present a difficult management problem.
The nature of their presentation implies that a major nerve
lies in close proximity to the tumour. Adequate surgical
clearance is therefore unlikely to be achieved without
sacrificing some degree of function, while preservation
of function is unlikely to be compatible with adequate
surgical clearance. An accurate histopathological diag-
nosis is mandatory. Careful judgment is required in
formulating a management policy for these patients, and
a thorough understanding of the behaviour of soft tissue
tumours is essential.
REFERENCES
Azumi N, Curtis J, Kempson RL, Hendrickson MR (1987): Atypical
and malignant neoplasms showing lipomatous differentiation: A
study of 111 cases. Am J Surg Pathol 11:161-183.
Ball ABS, Fisher C, Pittam M, Watkins RM, Westbury G (1990):
Diagnosis of soft tissue tumours by Tru-Cut biopsy. Br J Surg
77:756-758.
Barr LC, Robinson MH, Fisher C, Fallowfield ME, Westbury G
(1989): G Limb conservation for soft tissue sarcomas of the shoulder
and pelvic girdles. Br J Surg 29:1198-1201.
Enneking WF, Spanier SS, Malawer M (1981): The effect of anatomic
setting on the results of surgical procedures for soft part sarcomas of
the thigh. Cancer 47: 1005-1022.
Fisher C (1989): Pathology of soft tissue tumours. In Pinedo HM,
Verwij J (eds): “Treatment of Soft Tissue Sarcomas.” Boston:
Kluwer Academic Publishers, pp. 13-14.
Karakousis CP (ed) (1985): “Atlas of Operations for Soft Tissue
Tumours.” New York: McGraw-Hill, pp. 401407.
Watkins RM, Thomas JM (1987): Role of computed tomography in
selecting patients for hindquarter amputation. Br J Surg 74:711-
714.
Westbury G (1989): Surgery of soft tissue sarcomas. In Pinedo HM,
Venveij J (eds): “Treatment of Soft Tissue Sarcomas.” Boston:
Kluwer Academic Publishers, p. 5 1.