Between January 1975 and December 2000, eighty-one patients with a pelvic
sarcoma underwent acetabular resection. Eighteen patients died of the disease,
five were treated for subsequently diagnosed metastases or for progression of
the disease, and one had a secondary malignant lesion develop. Nine patients
were lost to follow-up. Two patients were excluded because no quality-of-life
questionnaire was available in the patient's native language. The
questionnaire was available only in German and English. In addition, one
patient was less than sixteen years old at the time of follow-up. Therefore,
functional evaluations and quality-of-life examinations were available for
forty-five patients who had undergone resection of the acetabulum, and these
patients formed the study group.
Quality of life was assessed with use of the European Organization for
Research and Treatment of Cancer core quality-of-life questionnaire (EORTC
QLQ-C30)4. The
questionnaire incorporates nine multi-item scales: five functional scales
(physical, role, emotional, social, and cognitive function), three symptom
scales (pain, fatigue, and financial difficulties), and a global health and
quality-of-life scale. Each health profile is scored on a 100-point scale: a
higher score in the functional scales means a lesser disability, whereas a
higher mean score for symptoms reflects more problems. The European
Organization for Research and Treatment of Cancer quality-of-life
questionnaire is widely used for cancer patients. An approach to aid
interpretation is the use of normative or population-based reference values.
The values for the general population are based on a large randomized
trial5.
Function was assessed with use of the Musculoskeletal Tumor Society system
developed by Enneking et
al.6. The score is a
clinically based and standardized evaluation system for the comparison of
function after different surgical treatments in patients with a tumor in an
extremity. The Musculoskeletal Tumor Society evaluation system (1993 version)
is based on the analysis of factors pertinent to the patient as a whole (pain,
restriction in activities and/or occupation, and emotional acceptance) and
factors specific to the affected limb (the use of walking supports, walking
ability, and gait). Each parameter is given a value ranging from 0 to 5,
according to specific criteria. A parameter can be excluded if it is not
applicable to the patient. The overall result is expressed as a summation of
each parameter, which then is converted into a percentage of the maximum
possible score.
The type of surgical treatment, the number of surgical revisions, and the
occupational and marital status of each patient were recorded.
The median age of the patients at the time of acetabular resection was 30.4
years (range, 4.2 to 73.5 years). The median age at the time of follow-up was
35.7 years (range, 16.1 to 83.2 years). The median time since the index
surgery to the time of the latest follow-up was sixty-nine months (range,
sixteen to 239 months). There were twenty-nine men and sixteen women in the
study group. The diagnosis was chondrosarcoma in twenty-one patients, Ewing
sarcoma or peripheral neuroectodermal tumor in seventeen patients, and
osteosarcoma in seven patients.
Ten patients received chemotherapy and surgical treatment without radiation
therapy. Fourteen patients received radiation therapy in addition to surgical
treatment and chemotherapy. Twenty-one patients received no chemotherapy and
no radiation. The acetabular lesions were excised with a variety of resections
according to the classification system of Enneking and
Dunham7,8.
Seven patients had a type-PI-II resection; one, a type-PII resection; six, a
type-PII-III resection; eleven, a type-PI-II-III resection; eight, a
type-PI-II-III-H1 resection; four, a type-PII-III-H1 resection; seven, a
type-PI-II-III-H resection (amputation); and one, a type-PII-III-H resection
(amputation) (Table I,
Fig. 1).
Surgical Procedures
Twenty patients had a hip transposition, which involved moving the femoral
head proximally to the lateral side of the sacrum or the underside of the
resected ilium after resection of the acetabulum. The joint capsule was
reconstructed with use of a polyethylene terephthalate mesh tube (Trevira;
Implantcast, Buxtehude, Germany), which was fixed to the pelvis with
transosseous sutures or bone anchors (Mitek anchors; Ethicon, Westwood,
Massachusetts) and formed a pouch for the femoral head. Soft tissues were
reattached to the
tube9 (Figs.
2 and
3). Ten of these patients
underwent preoperative and postoperative chemotherapy and radiation therapy
(five patients had radiation before surgery, and five had it after surgery),
and seven patients underwent preoperative and postoperative chemotherapy.
In ten patients, a prosthetic reconstruction was performed, following
resection of the acetabulum and femoral head, with use of a Vitallium pelvic
prosthesis (Howmedica, Kiel, Germany) (Fig.
4). One of these patients underwent preoperative and postoperative
chemotherapy, and two patients had preoperative and postoperative chemotherapy
and radiation therapy after surgery.
After iliac resection through the acetabulum, the defect was reconstructed
with an allograft in three patients and an autograft in one patient. Two of
these patients received preoperative and postoperative chemotherapy and
radiation therapy before surgery, and one patient received preoperative and
postoperative chemotherapy. After resection of the acetabular tumors in the
ischiopubic region, no reconstruction was necessary in three patients. One of
these patients underwent preoperative and postoperative chemotherapy. Four
patients underwent hindquarter amputation as the initial surgical treatment,
and four had a secondary amputation.
Wide margins were achieved in thirty patients, marginal margins in five
patients, and intralesional margins in two patients.
Complications of Surgery
Six of the twenty patients who underwent hip transposition had
complications. Three patients had postoperative wound problems, and three had
a palsy of the peroneal nerve. No patient who underwent hip transposition had
an additional procedure because of an infection or local relapse. In five
patients with a hip transposition, a limb-lengthening was done at an average
of 5.6 years after the tumor surgery. Of the ten patients who underwent
prosthetic reconstruction, eight required a surgical revision because of
complications: four patients had a deep infection, one patient had loosening
of the implant, two patients had a local relapse, and one patient had a palsy
of the femoral nerve. In these ten patients, external hemipelvectomy was
necessary because of persistent infection and local relapse in two patients
each.
The patient who had implantation of an autograft had development of a deep
infection, requiring removal of the graft resulting in a flail hip. Also, an
infection developed in one patient who had had reconstruction with
allograft.
None of the three patients who did not undergo reconstruction had
complications. Of the four patients initially treated with a hindquarter
amputation, two had complications; one of them had a deep infection and one
had a hematoa. Four additional patients underwent subsequent hindquarter
amputation.
The analysis-of-variance procedure, the Likert summated score method, and
the Student t test were used for statistical
analysis10-12.
The different treatment groups (endoprosthetic replacement, hip transposition,
and hindquarter amputation) were compared with regard to function and quality
of life. A p value of 0.05 was considered the threshold for significance.
Functional Results for Entire Group
The mean functional score, according to the Musculoskeletal Tumor Society
system, at the latest follow-up evaluation was 14.5 points (range, 2 to 27
points) of a maximum of 30 points. Nineteen of the forty-five patients
required pain medication for constant pain; two of the nineteen patients had
severe pain requiring continuous use of narcotics, seven had moderate pain
requiring periodic use of narcotics, and ten had mild pain requiring the use
of non-narcotic analgesics. Thirteen patients periodically used non-narcotic
analgesics, and nine patients required no pain medication. Twenty-nine
patients were unable to return to their previous occupation, mainly because
the work required long hours of standing. With regard to the use of walking
supports, twenty-four patients needed a walker or crutches, nine patients used
a single cane, eleven patients used orthotics, and one patient walked without
an assistive device. Ten patients were unable to walk, one patient could walk
at home, two patients could walk at home and very short distances outside, and
thirty-two patients walked in the community; however, for twenty-three of
these patients, the walking distance was limited.
Functional Results According to Surgical Treatment
The functional results in relation to the surgical procedure performed are
shown in Table II. The four
patients with tumor resection and allograft and/or autograft reconstruction
had a mean functional score of 62.5% (range, 53.3% to 80%) of a possible score
of 100%. All of these patients had alterations in their gait. The scores for
the use of walking aids were generally poor; no patient had a score of >3
points (of a possible 5 points). The mean emotional acceptance score was 70%
(range, 20% to 100%) of a possible score of 100%.
The eight patients who underwent hindquarter amputation had a mean
functional score of 26.3% (range, 6.7% to 40%). All of these patients used a
wheelchair or two crutches in the house. The mean emotional acceptance score
was 62.5% (range, 40% to 80%). Four of these eight patients required constant
pain medication (one with severe pain required continuous use of narcotics,
two with moderate pain required periodic use of narcotics, and one with mild
pain used non-narcotic analgesics).
The twenty patients with a hip transposition had a mean functional score of
60.7% (range, 16.6% to 83.3%) and a mean emotional acceptance score of 74%
(range, 40% to 100%). Seven of the twenty patients required constant pain
medication (one patient required continuous use of narcotics, one required
periodic use of narcotics, and five used non-narcotic analgesics). Three of
the twenty patients required a cane, and four needed two crutches. All
patients in this group had a limb-length discrepancy and required a shoe lift
(median lift, 7 cm; range, 5 to 12 cm). A persistent deficit in muscle power
at the hip was reflected in an alteration of gait in seventeen of the twenty
patients.
All ten patients who underwent prosthetic replacement were unable to walk
without external supports. The mean functional score was 39.6% (range, 20% to
60%). Seven of the ten patients needed two crutches. Four of them had mild
pain that required the use of non-narcotic analgesics, and one patient had
moderate pain that required the periodic use of narcotics.
In a comparison of the results after hip transposition, prosthetic
replacement, and hindquarter amputation, the patients who had hip
transposition demonstrated the best functional outcome. When function was
analyzed with respect to a restriction in activities, eight of the twenty
patients with a hip transposition were unable to return to their previous
occupation mainly because the work required long hours of standing, whereas
seven of the ten patients with endoprosthetic replacement (p = 0.067) and all
eight patients with hindquarter amputation (p < 0.0001) were unable to
return to their previous occupation. A major functional deficiency with gait
alteration was found in thirteen of the twenty patients with a hip
transposition compared with nine of the ten patients with a pelvic prosthesis
(p = 0.044) and all patients who underwent hindquarter amputation (p =
0.008).
Walking in the community was possible for nineteen of the twenty patients
with a hip transposition compared with six of the ten patients with a
prosthetic replacement (p = 0.004) and two of the eight patients who underwent
hindquarter amputation (p < 0.0001). There was also a significant
difference between prosthetic replacement and hindquarter amputation (p =
0.015) with respect to the ability to walk in the community. The results of
analysis of variance are described in Table
III.
Quality-of-Life Results for Entire Group
The quality-of-life results, based on the quality-of-life questionnaire
(QLQ-C30) of the European Organization for Research and Treatment of Cancer,
for the entire group were compared with those for a general population (see
Appendix). Patients with a pelvic tumor involving the acetabulum demonstrated
deficiencies, especially in role, social, and physical functioning. With
regard to global health status, only a slight difference was found. No
substantial emotional deficiency was found in the patients following
acetabular resection. Patients with a pelvic tumor had more pronounced
pain-related restrictions in the activities of daily living and more financial
difficulties than those in a general population.
Quality-of-Life Results According to Surgical Treatment
Restrictions in the ability to perform hobbies or other daily activities
(role function) were reported by patients with acetabular tumor resection and
prosthetic replacement (mean score, 25 points) and by patients who had a
hindquarter amputation (mean score, 33.3 points).
With regard to physical function, more restrictions were reported by
patients with a hindquarter amputation (mean score, 48.3 points) than by
patients with a hip transposition (mean score, 64 points) or those with an
endoprosthetic replacement (mean score, 60 points).
No pain or mild pain-related restrictions in daily activities were reported
by patients with a hip transposition (mean score, 32.5 points), whereas more
pain-related restrictions were reported by those with an endoprosthesis (mean
score, 51.7 points) and those with a hindquarter amputation (mean score, 41.7
points). The mean score and standard deviation for the functional and symptom
scales of the quality-of-life questionnaire are shown in the Appendix.
With regard to the general evaluation of health, patients who had a hip
transposition, hindquarter amputation, or replacement with a pelvic prosthesis
generally reported that they were in good health; no significant differences
among the groups were found, with the numbers available. Patients who had an
amputation had better results than did patients with endoprosthetic
replacement with regard to global quality of life. With regard to the ability
to perform hobbies, work, or other daily activities, the results were
significantly better for patients with a hip transposition (mean score, 51.6
points) than for those who had replacement with a prosthesis (mean score, 25
points) or who had an amputation (mean score, 33.3 points) (p = 0.043). With
regard to emotional functioning, the mean score for the patients with a hip
transposition (70.4 points), those with prosthetic replacement (67.2 points),
and those with an amputation (61.4 points) were not found to be significantly
different, with the numbers available (p = 0.774). With regard to social
functioning, the best results were reported by patients with a hip
transposition (mean score, 60 points). Fewer restrictions in social
functioning were reported by patients with an amputation (mean score, 52.1
points) than by patients with prosthetic replacement (mean score, 36.7
points). The groups were different with respect to pain-related restrictions,
with a mean score of 32.5 points for patients who had a hip transposition,
51.7 points for those who had prosthetic replacement, and 41.7 points for
those who had an amputation. Financial difficulties were more pronounced in
the group with prosthetic replacement and in patients with hindquarter
amputation. The results of the analysis of variance are shown in the
Appendix.
Limb salvage is possible in patients with a pelvic sarcoma; however, limb
salvage surgery for such patients is an extensive procedure associated with a
considerable rate of local complications and functional and cosmetic
disadvantages1-3,13-15.
Only a few studies have described the functional results or health-related
quality of life following surgical management of patients with bone sarcoma of
the
pelvis16-18.
The ideal method of reconstruction following acetabular excision remains a
subject of controversy. Partial or complete resection of the hemipelvis has
been reported to have an advantage over hindquarter
amputation19.
Options for reconstruction include endoprosthetic replacement and hip
transposition2,3,17.
The complication rate after pelvic reconstruction is
high1,14,15,17.
Complications are related to the large extensive approaches and frequent tumor
involvement of the gluteal muscles. Periacetabular resection and
reconstruction with allograft allows muscle reattachment but is associated
with a high rate of
infection18,20.
If major parts of the ilium are resected and proximal anchorage of the
prosthesis cannot be achieved, long-term stability is difficult to
obtain16,17.
When we compared patients treated with prosthetic replacement and those who
had a hip transposition following resection, the functional results were
significantly better for the patients with hip transposition (p = 0.017),
although the group had major limb-length discrepancies, ranging from 5 to 12
cm. No patient with a hip transposition had an additional operation because of
infection or local relapse. Moreover, the quality-of-life results were
satisfactory.
Patients who underwent a hip transposition had good results with regard to
physical, social, and emotional functioning as well as global health status.
Only mild pain-related restrictions in daily activities were found in this
group. Patients were able to adapt well to their new living conditions,
although strong efforts were necessary for them to deal with problems such as
restricted mobility or changing jobs or job orientation. Good results with
regard to emotional functioning were found in this group. In our opinion, hip
transposition is a good surgical procedure for patients with tumors involving
the acetabular region, especially when a major portion of the ilium has to be
resected. This surgical technique is particularly suitable for physically
active patients who will place a high demand on the reconstruction.
The functional results after implantation of the prosthesis were not
satisfactory, and a high complication rate was found. We believe that one
reason for the worse result was the extensive tumor involvement. Six of the
ten patients had a total hemipelvic resection including the proximal aspect of
the femur. The poor functional results and the high rate of complications were
likely related to the lack of soft-tissue coverage, resection of muscles, and
the creation of a large dead space. In addition, the number of patients with
advanced tumor extension treated with a prosthesis was higher in our series
than that in other
reports3,16-17.
In a study by Gradinger et
al.3, an anatomic
prosthetic reconstruction was performed in nine patients, five of whom had
good functional results after an average duration of followup of twenty-seven
months.
Wirbel et al.16
also reported that good or excellent function was obtained in twenty-nine of
forty-six patients. In the report by Abudu et
al.17, the mean
functional score according to the Musculoskeletal Tumor Society system was 70%
(range, 50% to 90%) in thirteen patients who were free from local or osseous
disease. They reported that limb-salvage surgery and reconstruction with
prosthetic replacement for the treatment of pelvic sarcomas preserves more
function than does hindquarter amputation.
We also found differences in function. None of the eight patients with a
hindquarter amputation was able to walk in the community compared with four of
ten patients with an endoprosthetic replacement (p = 0.015). With regard to
the use of external supports, patients with endoprosthetic replacement had
better results than the patients who had an amputation (p = 0.081). Patients
with a prosthesis had better results with respect to physical and emotional
functioning, whereas patients who had an amputation had better results with
respect to role and social functioning as well as global quality of life.
We believe that hindquarter amputation should be reserved for patients with
a realistic prospect of cure for whom no reconstructive option is available.
We believe that it is not acceptable to compromise the margins of resection in
order to obtain limb salvage. Patients with an acetabular resection were able
to adapt well to their new living conditions, although strong efforts were
necessary for them to deal with problems such as restricted mobility or
changing jobs or job orientation.
Grimer et al.18
stated that involvement of the great vessels or a major nerve by the tumor
means that any lesser procedure than hindquarter amputation is unacceptable.
However, Fuchs et
al.21 showed in
their series that patients with a tumor involving the sciatic nerve can be
treated with limb-sparing surgery and have an acceptable functional outcome.
It has been our experience that resection requiring the excision of the
sciatic nerve was not a contraindication for limb-sparing surgery. Considering
the small number of complications and the good functional and quality-of-life
results, hip transposition after acetabular resection for the treatment of
pelvic sarcomas appears to be the optimal technique for reconstruction.
Tables showing the quality-of-life results and the analyses of those
results are available with the electronic versions of this article, on our web
site at
(go to the article citation and click on "Supplementary Material")
and on our quarterly CD-ROM (call our subscription department, at
781-449-9780, to order the CD-ROM). ?