Over a 25-year period we have treated 36 patients

Osteosarcoma of the pelvis R. J. Grimer, S. R. Carter, R. M. Tillman, D. Spooner, D. C. Mangham, Y. Kabukcuoglu From the Royal Orthopaedic Hospital Oncology Service, Birmingham, England Over a 25-year period we have treated 36 patients with osteosarcoma of the pelvis. Of the tumours, 24 (67%) were primary osteosarcomas and 12 (33%) arose either after irradiation or in association with Paget s disease. Six patients had a hindquarter amputation and 12 were treated by a limb-salvage procedure with intrapelvic excision. The five-year survival rate of all the patients with pelvic osteosarcoma was 18%, while for 17 treated by chemotherapy and surgery it was 41%. The prognosis for patients presenting with metastases or with secondary osteosarcoma was appalling and none survived after 29 months. No patient over the age of 50 years when seen initially survived for a year. Youth and a good response to chemotherapy along with complete surgical excision offer the best chance of cure. J Bone Joint Surg [Br] 1999;81-B:796-802. Received 25 June 1998; Accepted after revision 12 February 1999 Osteosarcoma of the pelvis remains one of the greatest challenges in orthopaedic oncology. 1 The tumours are often large and present late. Many arise secondary to Paget s disease or radiotherapy and are difficult, or impossible, to resect without serious morbidity. These tumours should behave like any other osteosarcoma, but in practice many are chondroblastic and respond poorly to chemotherapy. 2 There is likely to be a similar response to radiotherapy. As a result, we rely on surgical excision for any chance of cure. This is technically R. J. Grimer, FRCS, Consultant Orthopaedic Surgeon S. R. Carter, FRCS, Consultant Orthopaedic Surgeon R. M. Tillman, FRCS, Consultant Orthopaedic Surgeon D. Spooner, MRCP, Consultant Clinical Oncologist D. C. Mangham, MRCPath, Consultant Pathologist The Royal Orthopaedic Hospital NHS Trust, Bristol Road South, Northfield, Birmingham B31 2AP, UK. Y. Kabukcuoglu, Visiting Fellow Department of Orthopaedics, Sisli Etfal Hospital, Istanbul, Turkey. Correspondence should be sent to Mr R. J. Grimer. 1999 British Editorial Society of Bone and Joint Surgery 0301-620X/99/59241 $2.00 challenging, not least because of the difficulties involved in imaging and in the accuracy of definition of the true extent of the tumour before operation. Hemipelvectomy has been the accepted surgical treatment of primary malignant tumours of the pelvic bones for many years. This has cosmetic and functional disadvantages because of the loss of the limb associated with a high rate of morbidity. 3,4 Improvements in preoperative imaging and adjuvant chemotherapy have allowed the development of several methods for reconstruction of the hip and hemipelvis which give a functional and cosmetically acceptable limb in patients who have had pelvic resection for malignancies. These techniques include iliofemoral and ischiofemoral arthrodesis, surgical pseudarthrosis, reconstruction with pelvic allografts, the use of custom-made endoprostheses, the use of the saddle prosthesis and reimplantation of the excised hemipelvis after sterilisation by autoclaving or radiation. 5-12 None of these procedures is straightforward and in patients with a poor prognosis extensive surgery is not justified. Nevertheless, failure to resect the tumour leads to inevitable death. We have reviewed our results of the management of patients with osteosarcoma of the pelvis to try to define groups with good or bad prognostic indicators which may help to decide on treatment. Patients and Methods Between 1971 and 1996, 36 patients (5%) were treated for osteosarcoma of the pelvis of a total of 724 patients with osteosarcoma (Table I). There were 18 women and 18 men with a mean age of 35.3 years (11 to 85) and median of 23 years (Fig. 1). The tumours were found in the right hemipelvis in 14, and in the left in 22, most commonly in the ilium (18). Of the tumours, 24 (67%) were primary osteosarcomas and 12 (33%) arose secondary to another condition. The histological types of osteosarcoma were high-grade intramedullary osteosarcoma in 22 (61%), low-grade osteosarcoma in two (6%) (one being parosteal) and secondary osteosarcoma in 12 (33%) of which two occurred after irradiation, two arose in diaphyseal aclasia and eight were in association with Paget s disease. The two post-irradiation 796 THE JOURNAL OF BONE AND JOINT SURGERY

OSTEOSARCOMA OF THE PELVIS 797 Table I. Details of 36 patients with osteosarcoma of the pelvis Case Age Necrosis Cause of Death at Alive at number (yr) Gender Year Site of tumour Type Cause* Stage Type of surgery Margins CT (%) Intention death (mth) (mth) 1 11 M 1992 Ilium/acetabulum Primary 2b Excision and re-implantation I Yes - Curative Metastases 13 2 13 F 1994 Ilium/acetabulum Primary 2b Excision and re-implantation I Yes 0 Curative Metastases 35 3 14 F 1995 Ilium/acetabulum Primary 2b Excision and re-implantation W Yes 65 Curative - 34 4 15 M 1983 Ischium/acetabulum Primary 2b Pelvic endoprosthesis W Yes 95 Curative - 172 5 15 M 1994 Ilium Primary 2b Hindquarter amputation W Yes 50 Curative - 48 6 16 F 1988 Ilium Primary 3 - - Yes - Palliative Metastases 7 7 16 F 1987 Ilium/s-i joint Secondary RT 2b Excision and fibular strut I Yes 20 Curative Blood loss 27 8 16 M 1983 Ilium Primary 2b Excision and fibular strut W Yes - Curative Metastases 19 9 17 F 1990 Ilium Primary 2b - - Yes - Palliative Metastases 37 10 17 F 1991 S-i joint Primary 2b - - Yes - Palliative Metastases 17 11 17 M 1981 Ilium Primary 2b Hindquarter amputation W Yes - Curative - 194 12 19 F 1989 Ilium Primary 3 - - Yes - Palliative Metastases 14 13 19 F 1989 Ilium Primary 2b Excision M Yes - Curative - 103 14 20 F 1995 Ilium/acetabulum Primary 2b Excision and re-implantation I Yes 90 Curative - 33 15 20 M 1996 Pubis Primary 1b Excision W - - Curative - 18 16 20 F 1986 Ilium Primary 2b Hindquarter amputation I Yes 40 Curative Metastases 7 17 22 F 1993 Ilium/sacrum Primary 2b - - Yes - Palliative Metastases 5 18 23 M 1994 Ilium/sacrum/pub Primary 2b - - Yes - Palliative Metastases 24 19 23 F 1995 Ilium/sacrum Primary 2b - - Yes - Palliative Metastases 15 20 24 M 1990 Ilium/sacrum/pub Secondary RT 3 - - - - Palliative Metastases 22 21 25 M 1993 Ilium/sacrum Primary 2b - - Yes - Palliative Septicaemia 12 22 9 F 1991 Ilium Primary 2b Excision and fibular strut M Yes 40 Curative Metastases 14 23 33 M 1994 Ilium Secondary DA 2b Excision and fibular strut M Yes - Curative Metastases 29 24 34 F 1971 Ilium Primary 3 Hindquarter amputation I Yes - Palliative Metastases 3 25 49 M 1995 Pubis Primary 1b Excision M - 95 Curative - 29 26 54 M 1994 Ilium Secondary DA 3 - - Yes - Palliative Metastases 10 27 55 F 1989 Ilium Primary 2b Hindquarter amputation I Yes 50 Curative Metastases 16 28 61 M 1983 Ilium Secondary Paget's 3 - - - - Palliative Metastases 1 29 62 M 1974 Ilium Secondary Paget's 2b - - Yes - Palliative Metastases 5 30 62 M 1991 Ilium/sacrum Secondary Paget's 3 - - Yes - Palliative Metastases 6 31 67 F 1988 Ilium/sacrum Secondary Paget's 2b Hindquarter amputation W - - Palliative Metastases 12 32 72 M 1993 Acetabulum Primary 3 - - - - Palliative Metastases 2 33 74 F 1996 Ilium/sacrum Secondary Paget's 2b - - - - Palliative Metastases 11 34 75 F 1989 Ilium Secondary Paget's 2b - - - - Palliative Metastases 10 35 77 M 1990 Ilium Secondary Paget's 2b - - - - Palliative Metastases 11 36 85 M 1994 Ilium Secondary Paget's 3 - - - - Palliative Metastases 7 * RT, radiation-induced; DA, diaphyseal aclasia I, intralesional; M, marginal; W, wide Chemotherapy Death/Alive = months from diagnosis VOL. 81-B, NO. 5, SEPTEMBER 1999

798 R. J. GRIMER, S. R. CARTER, R. M. TILLMAN, D. SPOONER, D. C. MANGHAM, Y. KABUKCUOGLU Fig. 1 The age distribution of the 36 patients with osteosarcoma of the pelvis at the time of diagnosis and the number of patients with primary and secondary disease. sarcomas occurred in young people, one of whom had received pelvic irradiation 11 years earlier for an ovarian teratoma while the other had had radiotherapy for Ewing s sarcoma of the pelvis nine years previously. The patients were staged according to the system of the Musculoskeletal Tumour Society. 13 Preoperative studies included plain radiography of the pelvis with MRI or CT of the region around the tumour. A total body technetium bone scan and CT of the chest were carried out before biopsy. All patients had an open biopsy in line with the planned definitive incision. Two patients (6%) were stage 1b, 25 (69%) were stage 2b and nine (25%) were stage 3, with metastatic disease present at the time of diagnosis. Five of 12 patients with secondary osteosarcoma had metastases at diagnosis, compared with four of the 24 with a primary tumour. There were no patients over the age of 50 years who did not have either metastases at presentation or a tumour secondary to some other condition. Most of the lesions were principally in the ilium with nine also invading the sacrum or sacroiliac joint, while four invaded the acetabular area. One tumour just invaded the acetabulum, two were in the pubis/ischium and one in the ischium and acetabulum. We began to use modern techniques of chemotherapy in 1983. All patients seen since then were offered neoadjuvant chemotherapy consisting of doxorubicin and cisplatin or a high-dose methotrexate regime based on the protocols of the European Osteosarcoma Intergroup in use at the time of treatment. 14,15 Twenty-six patients had chemotherapy before and after operation. Chemotherapy was not given because of advanced age (>65 years) in six patients, low- Fig. 2a Case 4. Radiographs showing a) the extent of the tumour and b) after resection and insertion of a prosthesis. Fig. 2b THE JOURNAL OF BONE AND JOINT SURGERY

OSTEOSARCOMA OF THE PELVIS 799 grade tumour in two patients and advanced disease in one. It was refused by one patient who had a radiation-induced sarcoma. Five of these patients had radiotherapy for palliative treatment. Three patients had neither chemotherapy, radiotherapy nor surgery because of advanced disease. Tumours which were resectable surgically were found in 18 patients. Limb-salvage surgery was carried out if technically possible to obtain both adequate margins of resection and a functional limb. In general, involvement of the sciatic notch or of the femoral vessels by the tumour prevented limb salvage. Intrapelvic excision was used in 12 patients with resection and autograft (fibular strut grafting of the ilium) in four, resection and a prosthesis in one (Fig. 2) and resection alone in three. Four patients were treated by resection of the affected part of the bone, removal of the soft tissue and bulk of the tumour followed by extracorporeal irradiation of the resected area with 90 Gy and reimplantation of the bone. According to the classification system of Enneking and Dunham 4 four of the resections were type I, four type I-II, one type II-III, and three type III (Fig. 3). Six patients had a hindquarter amputation because of the extent of the tumour which usually meant involvement of the sciatic nerve or iliac vessels. The margins of excision were verified by the pathologist and defined according to the system of the Musculoskeletal Tumour Society. 16 The surgical margins were wide in seven patients, marginal in four and intralesional in seven. The resection margins obtained by hindquarter amputation with three wide and three intralesional were no better than those achieved by limb-salvage surgery in which there were four wide, four marginal and four intralesional. Attempts were made to obtain a wide margin in every case, but this was often impossible without sacrifice of major nerves, vessels or intra-abdominal structures. The intralesional excisions were not always appreciated at the time of surgery and were most common in the margins of bone resection. In these cases further excision was sometimes carried out or radiotherapy given. Surgery was not undertaken because of the presence of metastatic disease in eight patients, age and/or infirmity in four, progression of disease in two and inoperability in two. Overall, 16 patients were considered to have had potentially curative treatment with chemotherapy and an attempt at surgical excision. The remaining 20 only had palliative treatment which usually involved either chemotherapy, radiotherapy or surgery. Results The 36 patients have been followed up for between 18 and 194 months. At the latest follow-up, six patients (17%) were free from local or metastatic disease and two (5.5%) were alive with disease, both after resection of pulmonary metastases. These developed in 28 patients (78%), 26 of whom had died at the latest follow-up. Two patients without metastases died, one from septicaemia secondary to chemotherapy-induced neutropenia and the other at operation from massive haemorrhage during an attempt to resect a recurrent radiation-induced sarcoma of the sacroiliac joint. The overall survival was estimated by the Kaplan-Meier method. 17 The five-year survival rate of patients with pelvic osteosarcoma was 18% (Fig. 4), for the 16 patients with curative treatment by chemotherapy and surgical excision 41% (Fig. 5) and for the 20 patients with palliative treatment alone 0% (Wilcoxon signed-rank test, p < 0.001). The mean survival for patients receiving palliative treatment was ten months. The patients with secondary osteosarcoma had a Fig. 3 The types of resection according to the classification system of Enneking and Dunham 4. Fig. 4 Cumulative survival of all 36 patients with osteosarcoma of the pelvis. VOL. 81-B, NO. 5, SEPTEMBER 1999

800 R. J. GRIMER, S. R. CARTER, R. M. TILLMAN, D. SPOONER, D. C. MANGHAM, Y. KABUKCUOGLU Fig. 5 Cumulative survival of patients with osteosarcoma of the pelvis according to whether they had palliative or curative treatment. Fig. 6 Cumulative survival of patients with osteosarcoma of the pelvis according to whether they had primary or secondary disease. worse prognosis than those with a primary tumour (Fig. 6). The five-year survival rate of patients with primary highgrade intramedullary osteosarcoma was 24.2% for 22 patients. Of the 12 patients with secondary osteosarcoma none survived for more than 29 months. The survival rate of the patients with high-grade non-metastatic osteosarcoma was 30% at five years. The patients with osteosarcoma secondary to Paget s disease survived for a mean of eight months, as did those who presented with metastases. The survival rate of the patients was examined by age groups. It was 33.3% at five years for the 16 patients who were less than 20 years old, but there were only two survivors in the older age group and both had low-grade tumours. Nine patients had a histological assessment of the effectiveness of chemotherapy. Only two had necrosis of more than 90% and they remain alive, although one has undergone a resection for pulmonary metastases. In the other seven the percentage of necrosis was very disappointing with a mean of 40%, and a survival of 21% at four years. There was no difference in survival between those patients having hindquarter amputation and those with limb-salvage surgery, being 33% and 37% at five years, respectively. The relationship of the surgical margins to overall survival was significant (p < 0.05). The five-year survival was 61.4% for patients with wide and marginal surgical excision, while it was nil for patients with intralesional excision. There was a local recurrence in only two of the 18 patients who had surgical excision (11%). One (case 7), with a radiation-induced sarcoma of the ilium, had resection of a sacroiliac joint and a fibular strut graft. After two years she developed local recurrence and at attempted hindquarter amputation she died because of excessive intraoperative haemorrhage. The other local recurrence arose in a patient (case 2) who had re-implantation of the ilium after sterilisation with extracorporeal radiotherapy. She developed pulmonary metastases after 23 months and refused any further treatment. Some months later an occult intrapelvic soft-tissue recurrence was detected. It did not appear to have arisen from the re-implanted bone. Both these local recurrences arose in patients with microscopically positive margins after the first operation and a poor response to chemotherapy. There were no local recurrences in any of the other patients, despite the high incidence of inadequate margins; 11 of 18 patients had marginal or intralesional excisions. We did not see any pelvic instability in patients who had resection and reconstruction. Four were treated by nonvascularised fibular iliosacral grafts; there was one surgical failure which required re-exploration and repositioning of the graft. Mild sacroiliac pain was present in three patients, but it was not sufficiently severe to necessitate a sacroiliac fusion. Deep infection developed in one patient with a type I-II resection treated by irradiation and re-implantation of the pelvic bone. The infected graft was resected which left the patient with a flail hip. Superficial infection developed in three patients (16%) and was controlled with antibiotics. None of the patients with limb salvage had a major nerve palsy. Vascular or urological complications developed in three (16%), but these were minor and were treated conservatively. Discussion Pelvic osteosarcoma constitutes 4% to 10% of osteosarcomas of all sites 18,19 and accounted for 5% of the osteosarcomas treated at our centre. The peak incidence in this series was in the second decade and there was a second peak in the eighth because of Paget s disease. The high incidence of osteosarcoma arising secondary to other conditions (33%) and the high proportion of cases with metastases at presentation (25%) have not previously been noted. There are only a few published accounts of the outcome of osteosarcoma of the pelvis. 16,20-24 The overall survival THE JOURNAL OF BONE AND JOINT SURGERY

OSTEOSARCOMA OF THE PELVIS 801 has been variously reported to be between 20% and 47% in non-metastatic high-grade pelvic osteosarcoma. Successful treatment always involves effective chemotherapy and surgical resection of the tumour. There are early reports of successful local control using intra-arterial chemotherapy and radiation without surgical intervention, but the longterm prognosis is uncertain. 23 The ability to obtain a complete surgical resection is critical in the planning of surgery for any patient with a pelvic osteosarcoma. Effective chemotherapy can make this easier to achieve, but the response of pelvic osteosarcoma to chemotherapy is poor, and even with modern techniques a high percentage of necrosis cannot be relied upon. Our rate of 11% of local recurrence in patients after resection is consistent with other reports. 1,6-8,24 Kawai et al 23,24 noted that it was more difficult to obtain a wide margin in tumours involving the sacrum and our observations confirm this. We found that most tumours with frank involvement of the sacrum extended at least to the spinal canal or across to the opposite side and were usually considered to be inoperable because of the high risk of local recurrence. Kawai et al 23 had a local recurrence rate of 32% which tends to confirm this. Hindquarter amputation would be expected to offer a better prognosis if adequacy of surgical resection was the only deciding factor, but this has not been shown to be the case. Not only are the margins of excision with a hindquarter amputation no better but survival is not improved, although these patients may have worse tumours in the first place. Fahey et al 1 noted a high incidence of vascular invasion, particularly in the chondroblastic lesions. We did not identify this as a particular problem, although we are very much aware of this potential complication and agree that in all chondroid tumours of the pelvis careful imaging of the association of the tumour with the great veins is mandatory. There is a tendency to try to avoid hindquarter amputation, particularly when the prognosis is so poor. The devastating functional and psychological effects of this mutilation should not be underestimated and we reserve this procedure now for patients with a realistic prospect of cure for whom no other reconstructive option is available. It is never acceptable, however, to compromise the margins of resection in order to obtain limb salvage in a potentially curable patient. Involvement of the great vessels or a major nerve by the tumour means that any lesser procedure than hindquarter amputation is unacceptable. The methods which we have used to reconstruct the pelvis after resection of the tumour are all well known, as are their complications. 25,26 There is no single solution for every pelvic tumour and various authors have previously described the options for the different sites within the pelvis. Our experience of pelvic endoprostheses has been gratifying, but only one patient in this series proved suitable for its use. 10 The incidence of tumours involving both ilium and acetabulum has resulted in our use of irradiated and re-implanted bone in four patients. This practice has previously been described by Harrington, 9 who had reasonable results in the short term. In our four patients, two died, one became infected requiring removal of the graft and the fourth is well after 2.5 years with a functional score of 85% and has returned to school. It is certainly a technique worthy of further investigation. Osteosarcoma of the pelvis remains a daunting prospect for orthopaedic oncologists. The prognosis is appalling for patients with secondary osteosarcomas and all those with metastases at presentation. Effective chemotherapy combined with complete surgical excision is mandatory to achieve cure. If this can be combined with limb salvage and reconstruction the functional results are likely to be better than after amputation. 27 These rare tumours must be treated in specialist centres with all the resources of a multidisciplinary group working to obtain optimal results. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Fahey M, Spanier SS, Vander Griend RA. Osteosarcoma of the pelvis: a clinical and histopathological study of twenty-five patients. J Bone Joint Surg [Am] 1992;74-A:321-30. 2. Dahlin DC, Unni KK. Bone tumors: general aspects and data on 8542 cases. 4th edition. Springfield, Illinois: Charles C Thomas, 1986:120. 3. Carter SR, Eastwood DM, Grimer RJ, Sneath RS. Hindquarter amputation for tumours of the musculoskeletal system. J Bone Joint Surg [Br] 1990;72-B:490-3. 4. Enneking WF, Dunham WK. Resection and reconstruction for primary neoplasms involving the innominate bone. J Bone Joint Surg [Am] 1978;60-A:731-46. 5. Aboulafia AJ, Buch R, Mathews J, Li W, Malawer MM. Reconstruction using the saddle prosthesis following excision of primary and metastatic periacetabular tumours. Clin Orthop 1995;314:203-13. 6. Gradinger R, Rechl H, Hipp E. Pelvic osteosarcoma: resection, reconstruction, local control and survival statistics. Clin Orthop 1991;270:149-58. 7. O Connor MI, Sim FH. Salvage of the limb in the treatment of malignant pelvic tumors. J Bone Joint Surg [Am] 1989;71-A:481-94. 8. O Connor MI. Malignant pelvic tumours: limb-sparing resection and reconstruction. Semin Surg Oncol 1997;13:49-54. 9. Harrington KD. The use of hemipelvic allografts or autoclaved grafts for reconstruction after wide resections of malignant tumors of the pelvis. J Bone Joint Surg [Am] 1992;74-A:331-41. 10. Abudu A, Grimer RJ, Cannon SR, Carter SR, Sneath RS. Reconstruction of the hemipelvis after the excision of malignant tumors: complications and functional outcome of prostheses. J Bone Joint Surg [Br] 1997;79-B:773-9. 11. Cannon SR, Tillman RM, Grimer RJ, Carter SR, Sneath RS. Treatment of primary bone tumours of the ilium by local resection and fibular strut graft (non-vascularised). In: Campanacci M, Capanna R, eds. Procs 8th international symposium on limb salvage. May 10-12 1995, Florence, Italy. 12. Stell HH. Partial or complete resection of the pelvis. J Bone Joint Surg [Am] 1978;60-A:719-30. 13. Enneking WF. A system of staging musculoskeletal neoplasms. Clin Orthop 1986;204:9-24. 14. Bramwell VH, Burgers M, Sneath R, et al. A comparison of two short intensive adjuvant chemotherapy regimens in operable osteosarcoma of limbs in children and young adults: the first study of the European Osteosarcoma Intergroup. J Clin Orthop 1992;10:1579-91. 15. Souhami RL, Craft AW, Van der Eijken JW, et al. Randomised trial of two regimens of chemotherapy in operable osteosarcoma: a study of the European Osteosarcoma Intergroup. Lancet 1997;350:911-7. VOL. 81-B, NO. 5, SEPTEMBER 1999

802 R. J. GRIMER, S. R. CARTER, R. M. TILLMAN, D. SPOONER, D. C. MANGHAM, Y. KABUKCUOGLU 16. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskeletal sarcoma. Clin Orthop 1980;153:106-20. 17. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81. 18. Huvos AG. Bone tumours: diagnosis, treatment and prognosis. 2nd edition. Philadelphia: WB Saunders Company, 1991. 19. Schajowicz F. Tumors and tumorlike lesions of bone: pathology, radiology and treatment. 2nd edition. Berlin: Springer-Verlag, 1995. 20. Estrada-Aguilar J, Greenberg H, Walling A, et al. Primary treatment of pelvic osteosarcoma: report of five cases. Cancer 1992;69: 1137-45. 21. Marangolo M, Tienghi A, Fiorentini G, et al. Treatment of pelvic osteosarcoma. Ann Oncol 1992;3Suppl:19-21. 22. Shin K-H, Rougraff BT, Simon MA. Oncologic outcome of primary bone sarcoma of the pelvis. Clin Orthop 1994;304:207-17. 23. Kawai A, Huvos AG, Meyers PA, Healey JH. Osteosarcoma of the pelvis: oncologic results of 40 patients. Clin Orth 1998;348:196-207. 24. Kawai A, Healey JH, Boland PJ, et al. Prognostic factors for patients with sarcomas of the pelvic bones. Cancer 1998;82:851-9. 25. Campanacci M, Capanna R. Pelvic resections: the Rizzoli Institute experience. Orthop Clin North Am 1991;22:65-86. 26. Capanna R, van Horn JR, Guernelli N, et al. Complications of pelvic resections. Arch Orthop Trauma Surg 1987;106:71-7. 27. Enneking WF, Menendez LR. Functional evaluation of various reconstructions after periacetabular resection of iliac lesions. In: Enneking WF, ed. Limb salvage in musculoskeletal oncology. New York: Churchill Livingstone, 1987:117-35. THE JOURNAL OF BONE AND JOINT SURGERY