Secondary Logo

Journal Logo

Malignant Tumors of the Pelvis: An Outcome Study

Mankin, Henry, J; Hornicek, Francis, J; Temple, H, Thomas; Gebhardt, Mark, C

Clinical Orthopaedics and Related Research: August 2004 - Volume 425 - Issue - p 212-217

This is a report of outcomes after a review of the demographic, diagnostic, therapeutic, and survival data for patients with pelvic primary and secondary tumors treated during the past 28 years. Using a computerized system it has been possible to assess the results for 206 patients with bone and soft tissue sarcomas and metastatic carcinomas to define the variation in outcome and the factors which statistically show an effect on survival. The data were compared with data for other anatomic sites. Based on our study, it is apparent that the outcome for all the tumors was approximately 50% survival with only soft tissue sarcomas having a poorer result. There were only minimal to moderate differences in outcome on the basis of gender, age, type of surgery, or adjunctive therapy. Patients who had intralesional surgery did less well as did patients with higher Musculoskeletal Tumor Society stages. Comparing the results for these patients with results for patients with the same stage and diagnosis but with tumors located in other sites showed significant differences. Results for patients with pelvic allograft compared with results of patients who had femoral allografts for the same diagnoses showed a poorer outcome for the patients who had pelvic allografts. Several possible explanations are provided for these variations in results.

From the Orthopaedic Oncology Service, Massachusetts General Hospital and Children’s Hospital, Harvard Medical School, Boston, MA.

Received: June 30, 2003

Revised: September 22, 2003; November 21, 2003

Accepted: January 23, 2004

Dr. Temple is currently at the University of Miami.

Each author certifies that the institution has approved the human protocol for this investigation and that all experimentation was conducted in conformity with ethical principles of research.

Correspondence to: Henry J. Mankin, MD, Gray 6 Orthopaedics, Massachusetts General Hospital, Boston, MA 02114. Phone: 617-726-2735; Fax: 617-726-6823;

The treatment of malignant tumors, which involve the bones of the pelvis or the immediately adjacent soft tissues has been considered difficult. Since the earliest descriptions of resective and amputative surgery,4,12,14,15,21,22,26,33,37 many authors have reported a lower rate of patient survival than for patients with the same diagnosis in other sites, along with multiple complications of the surgical procedure or reconstruction.3,6–9,11,13,14,16,20,27–31,,36,38,39,41

During the previous 28 years we have treated and maintained followup data on 206 patients with malignant bone and soft tissue primary or metastatic malignant tumors of the pelvic bones or the immediately adjacent soft tissue structures. Surgical treatments have included internal partial resective surgical excision and reconstruction (144 patients), total resection and allograft replacement (41 patients), and formal external hemipelvectomy (21 patients). All but 31 of the patients received chemotherapy or radiation treatment or both. The outcome for the entire series shows that 50% of the patients have died of disease. The remaining 50% have survived but in some cases, with physical limitations and disabilities. The current study documents the poor survival rate for these patients and explains why these tumors are difficult to treat.

Back to Top | Article Outline


During the past 31 years the Orthopaedic Oncology Service has maintained a database in which demographic, diagnostic, therapeutic, complication, and outcome information has been recorded for more than 15,000 patients with benign or malignant tumors affecting the skeleton or adjacent soft parts.25 The system records patient information from the time of initial contact to the most recent visit, and when the final outcome for the patient is not known, it is possible through several sources to gain information as to survival and the patient’s life status and adjustments.34

Two-hundred six patients were identified as having malignant tumors involving the pelvis and adjacent soft parts and the information obtained from them is the basis of this presentation. An approximately equal number of patients with benign or low-grade tumors were excluded from the study along with a small number of patients (approximately 30) with high-grade tumors for whom it was impossible to obtain survival information.

Examination of the demographic data for the 206 patients shows that there were 107 males and 99 females with an average age of 47 ± 20 years (range, 3–89 years). The average duration of followup was 5.2 ± 5.2 years (range, 0.5–24 years). Forty-one of the patients had total pelvic allografts, which for all but four included the acetabulum and required a bipolar hip replacement as part of the reconstruction. Another 21 patients had external radical hemipelvectomy and the remaining 144 had some form of resective surgery, usually including the ilium, pubis, or ischium but sparing the acetabulum. Thirty-one of the patients had MSTS Stage I disease, 105 patients as had Stage II disease, and 70 patients had Stage III disease.18 The diagnoses for the patients were mostly highly malignant tumors including 45 with osteosarcomas, 46 with chondrosarcomas, 22 with Ewing’s sarcomas, 27 with malignant fibrous histiocytomas, and 47 with metastatic carcinoma. There were four patients with chordomas, which had recurred locally after sacral resection and involved the pelvic bones.

The study reviewed all the possible features available in the computerized system, which potentially could affect survival. The statistical analyses were done using chi square calculations as assessed by the Mantel-Haenzel and Fisher exact tests, Cox regression studies, and Kaplan-Meier parametric estimation and plots.5,10,23

Back to Top | Article Outline


One hundred four of the 206 patients (approximately 50%) have survived for a mean of 8 ± 6 years (range, 2–24 years). The remaining 102 patients died of disease at a mean of 3 ± 3 years (range, 0.5–15 years) (Fig 1).

Fig 1.

Fig 1.

The survival data varied depending on the patient’s diagnosis. Regardless of diagnosis, most of the patients have an approximately equivalent rate to the 50% survival rate for the entire series, with the exception of the 27 patients with malignant fibrous histiocytomas, who have a significantly reduced survival rate of 30% (Table 1).

Table 1

Table 1

Gender had no effect (p < 0.83) on patient outcome, but patient age was a significant factor. The 101 patients older than 50 years had a 44% survival rate compared with the 105 patients younger than 50 years, who had a 57% survival rate (p < 0.02). As a partial explanation of the age-related survival statistics, the patients with malignant fibrous histiocytoma and metastatic carcinoma (mean age, 59 years with survival rates of 30% and 49%, respectively) were older than patients with osteosarcoma or Ewing’s sarcoma (mean age, 29 years with survival rates of 58% and 50% respectively).

Musculoskeletal Tumor Society stage had a significant effect on outcome. The 31 patients classified as having Stages IA and IB disease had a very high survivorship rate (81%) whereas 105 patients classified as having Stages IIA or IIB disease and 70 classified as having Stage III disease had a 45% survivorship rate (Table 2). There was no difference between the outcome for patients with Stage II and Stage III disease (chi square = 0.02, p < 0.91).

Table 2

Table 2

The 21 patients who had a total hemipelvectomy and the 144 patients who had some type of resection surgery had similar survival rates (39% and 47%, respectively); patients with resection and allograft replacements had a slightly better prognosis (54% survival rate) but the difference was not significant (Table 2).

Local recurrence had a limited effect on survival. Fourteen patients had a local recurrence and of that group 10 died of disease (29% survival rate, p < 0.03 when compared with the value of 50% for the total series). Metastasis had a considerably greater effect on survival. Of 122 patients who had metastases at the time of first treatment or patients who had metastases develop, 99 died of disease (19% survival rate, p < 0.0003). If the 70 patients with Stage III disease are not included, the number of recurrences for this group of 136 patients is 10 (7%) and seven of these patients died of disease (30% survival rate). The number of patients who had metastases develop was 76 (56%). Sixty-four of these patients died (16% survival rate). Both of these rates are significantly different from the 53% overall survival rate for the 136 patients (p < 0.00003).

The effect of the use of adjuvant chemotherapy or radiation on survival show better survival results for patients who received no chemotherapy or radiation than patients who received chemotherapy alone (Table 2). The statistical difference by the Mantel-Haenzel test is p < 0.02. However, these data are presumed to be inaccurate based on the types of patients who comprised these groups. More than ½ of the 31 patients who received no adjuvant treatment had MSTS Stages IA or IB disease, whereas only three of the 26 patients who received chemotherapy only had Stage III disease and four had Stage IA or Stage IB disease.

The margins obtained during surgery are another concern. Information regarding the margins was available for only 149 of the 206 patients. A significant difference in patient outcome occurred in relation to the three possible margins for the surgical procedures. The values show a marked decrease in survival for patients who had intralesional surgery compared with patients who had marginal or wide resection margins. The validity of this observation might be challenged on the basis that 18 of the 29 patients who had intralesional surgery had Stage III metastatic carcinoma, which presumably was palliative.

A second issue is related to the use of allografts in the pelvis. A comparison of the data for 41 patients treated with pelvic allografts with the data of 526 patients with allografts for high-grade tumors located in other anatomic sites shows considerable differences (Table 3). The patients who had pelvic resection and allograft replacement were older but the length of followup was similar. The survival for the patients who had pelvic allografts was considerably less (p < 0.007) than for patients with allograft replacement in other sites. Furthermore the recurrence rate, frequency of metastasis, and infection were much higher in the patients who had pelvic allograft surgery, but the likelihood of amputation was the same in both groups. The current clinical assessment for patients with pelvic allografts was 51% rated as good or excellent, whereas for 526 patients with nonpelvic allografts done for Stage II or Stage III malignant tumors, the clinical success rate was 68% (p < 0.02).

Table 3

Table 3

The survival rates for patients with pelvic tumors compared with survival rates for patients with the same diagnoses and stages of disease but with tumors located in the femur are significantly different. Two hundred fifty-one patients with Stage II osteosarcomas and 53 patients with Stage II chondrosarcomas of the femur have had an almost 80% survival rate and the data for 143 patients with malignant histiocytomas of the thigh are similar. The survival rates for patients with pelvic osteosarcomas and chondrosarcomas are less than 60% and for patients with the pelvic malignant fibrous histiocytoma, it is approximately 30%.

Back to Top | Article Outline


Although some authors have reported reasonably good results for patients with some pelvic lesions1,24,35 (Table 4), there is little doubt on the basis of literature surveys that the treatment of patients with high-grade sarcoma and metastatic carcinoma of the pelvis is considerably more difficult than for other sites.6,11,13,27,29,41 The failure rate is higher, the disability levels are greater, and the patient’s survival rate is considerably poorer.6,19,23,24,30,32,39 There are several possible explanations for this.

Table 4

Table 4

The size of the tumors often is much greater in the pelvis than in other sites. Although the pelvis has a collection of nerves passing through it, it still is possible for tumors at this site to grow much larger with less recognition by the patient compared with other more distal sites, or in the upper extremities. A lesion 10 cm in longest diameter in the calf or thigh is hard to ignore but a similarly sized lesion in the pelvis can exist without patient awareness, providing fracture or neural or vascular compression does not occur.

Surgical resection often is much more complex for pelvic tumors, based on the proximity of blood vessels and nerves. Avoiding viscera and trying to save the femoral or sciatic nerves or the iliac and femoral vessels and even some of the smaller arteries and veins can lead not only to a marginal resection but at times an intralesional one. In the aforementioned reported data, patients who have had intralesional surgical resections for pelvic tumors have very high mortality rates compared with patients who have had marginal or wide resections.6,19,23,24,30,32,39

Reconstructive surgery about the pelvis is complex and at times difficult. Restoring anatomy for the hip, maintaining the length of the limb, and retaining function for the muscles are not simple and at times require more time and ingenuity than procedures in other anatomic sites.1,3,6,7,13,14,16,17–19,22,27,29–31,36,39,41 The duration of surgery increases the likelihood of infection and anesthetic complications for the patient. Most of the patients who had pelvic surgery for high-grade tumors spent several days in the intensive care unit at our institution and frequently remained intubated and received intravenous fluids for a much longer time than patients who were treated for lesions in other anatomic sites.

Insertion of a pelvic allograft has some advantages in comparison with autograft reconstruction or metallic inserts but also has the highest rate of failure of any allograft procedure.2,40,42 At times dealing with the complex complications may exceed the potential value of the procedure for the patient. It also is likely that the pelvic allograft has the highest rate of donor bacterial contamination, and many banks treat pelvic components with radiation to reduce the likelihood of an infection in the host. These devices presumably are less likely to be infective, but perhaps are more fragile mechanically because of the exposure to high-dose radiation.

Radiation and chemotherapy were given to most of the patients in this series and presumably were in part responsible for the continued survival of ½ of the patients. The patients who did not receive such adjuvant care were too few to assess the role of these measures but one must presume on the basis of our understanding of the treatment protocols for osteosarcoma, Ewing’s sarcoma, and malignant fibrous histiocytoma that the outcome would have been far poorer.19,32,35

One of the more speculative explanations for the poor results in this study is related to vascularity and temperature. The vascularity of pelvic tumors is probably considerably greater than in other more distal sites. Another feature of the vascular issue is the internal body temperature at the site of the tumor, which is considerably higher than that of more distal tumors. This may play a role in increasing the malignancy of the tumor and the risk of metastasis.

In final analysis there are some interesting and fascinating issues raised by this research. First, it seems that the mean survival for patients with pelvic tumors is approximately 50%, and this figure is the same for patients with metastatic carcinoma and other high-grade bone sarcomas. The patients with malignant fibrous histiocytomas did less well, presumably because of invasion of viscera, nerves, and blood vessels that made resection more difficult.

Second, gender had no effect and age had only a moderate effect on outcome. Males did as well as females and although older patients fared less well, the difference probably was related to the increased frequency of metastatic disease in the older age group.

Third, the type of surgical treatment seemed to result in little difference in patient outcome. Patients with hemipelvectomies and allograft replacements survived just as well or poorly as patients with partial resections. Fourth, patients who had marginal and wide margins at surgery had similar outcomes at the same 50% level, whereas the 29 patients with intralesional margins did less well.

It seems that regardless of what the tumor is and how it is treated, 50% of the patients will survive, particularly if the margins are marginal or wide and adjunctive treatment is administered. That level of success is not as good as we would like but it is a baseline to see if some efforts will improve the outcome. Based on the data obtained from this study, the approaches we should take include improved surgical techniques, altered adjuvant therapy protocols, and better mechanical restoration systems. These are the directions we must take if we are to improve the status and outlook for these patients.

Back to Top | Article Outline


1. Aboulafia AJ, Malawer MM. Surgical management of pelvic and extremity osteosarcoma. Cancer. 1989;71(Suppl 10):3358–3366.
2. Alo AJ, Ekfors T, Dean PB, et al. Incorporation and clinical results of large allografts of the extremities and pelvis. Clin Orthop. 1991;307:200–213.
3. Apfellstaedt JP, Zhang PJ, Driscoll DL, et al. Various forms of hemipelvectomy for soft tissue sarcomas: Complications, survival and prognostic factors. Surg Oncol. 1995;4:217–222.
4. Ariel IM, Shah JP. The conservative hemipelvectomy. Surg Gynecol Obstet. 1977;144:406–413.
5. Bailar JC, Mosteller F. Guidelines for statistical reporting in articles for medical journals: Amplifications and explanations. Ann Intern Med. 1988;108:258–265.
6. Campanacci M, Capanna R. Functional Results of Reconstruction for Periacetabular Pelvic Resections Requiring Sacrifice of the Hip Joint. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 187–191,1987.
7. Campanacci M, Capanna R. Pelvic resection: The Rizzoli experience. Orthop Clin North Am. 1991;22:65–86.
8. Capanna R, Guernelli N, Ruggieri P, et al. Periacetabular Pelvic Resections. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 141–146,1987.
9. Conrad EU III, Springfield D, Peabody T. Pelvis. In Simon MA, Springfield DS (eds). Surgery for Bone and Soft Tissue Tumors. Philadelphia, Lippincott-Raven 323–341, 1998.
10. Cox DR. Regression models and life tables. J R Stat Soc. 1972;34:187–220.
11. Damron TA, Sim FH. Surgical treatment for metastatic disease of the pelvis and the proximal end of the femur. Inst Course Lect. 2000;49:461–470.
12. Douglass HO, Razack M, Holyoke D. Hemipelvectomy. Arch Surg. 1975;110:82–85.
13. Eilber FR, Eckardt JJ, Grant TG. Resection of Malignant Bone Tumors of the Pelvis: Evaluation of Local Recurrence, Survival and Function. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 136–141, 1987.
14. Eilber FR, Grant TT, Sakai D, et al. Internal hemipelvectomy: Excision of the hemipelvis with limb preservation: An alternative to hemipelvectomy. Cancer. 1979;43:806–809.
15. Enneking WF. Pelvis. In Enneking WF (ed). Musculoskeletal Tumor Surgery. New York, Churchill Livingstone 483–490, 1983.
16. Enneking WF, Dunham WK. Resection and reconstruction for primary neoplasms involving the innominate bone. J Bone Joint Surg. 1978;60A:731–746.
17. 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 117–135, 1987.
18. Enneking WF, Spanier SS, Goodman MA. A system for the surgical staging of musculoskelatal sarcoma. Clin Orthop. 1980;153:106–120.
19. Fahey M, Spanier SS. Vander Griend RA: Osteosarcoma of the pelvis. J Bone Joint Surg. 1992;74A:321–330.
20. 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. 1992;74A:331–341.
21. Higinbotham NL, Marcove RC, Casson P. Hemipelvectomy: A clinical study of 100 cases with a five-year-follow-up on 60 patients. Surgery. 1966;59:706–708.
22. Johnson JT. Reconstruction of the pelvic ring following tumor resection. J Bone Joint Surg. 1978;60A:747–751.
23. Kaplan E, Meier P. Non-parametric estimation for incomplete observations. J Am Stat Assoc. 1958;53:457–481.
24. Lane JM, Duane K, Glasser DAB, et al. Periacetabular Resections for Malignant Sarcomas. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 166–169, 1987.
25. Mankin HJ. A computerized system for orthopaedic oncology. Clin Orthop. 2002;398:252–261.
26. Mankin HJ. History of the Treatment of Musculoskeletal Tumours. In Klenerman L (ed). Evolution of Orthopaedic Surgery. London, Royal Society of Medicine Press 191–210, 2002.
27. Masterson EL, Davis AM, Wunder JS, et al. Hindquarter amputation for pelvic tumors: The importance of patient selection. Clin Orthop. 1998;350:187–194.
28. O’Connor MI. Malignant pelvic tumors: Limb-sparing resection and reconstruction. Semin Surg Oncol. 1997;13:49–54.
29. O’Connor MI, Sim FH. Salvage of the limb in the treatment of malignant pelvic tumors. J Bone Joint Surg. 1989;71:481–494.
30. Patterson FR, Peabody TR. Operative management of metastases to the pelvis and acetabulum. Orthop Clin North Am. 2000;31:623–631.
31. Prewitt TW, Alexander HR, Sindelar WF. Hemipelvectomy for soft tissue sarcoma: Clinical results in fifty-three patients. Surg Oncol. 1995;4:261–269.
32. Pring ME, Weber KL, Unni K, et al. Chondrosarcoma of the pelvis: A review of sixty-four cases. J Bone Joint Surg. 2001;83A:1630–1642.
33. Pringle JH. The interpelvic-abdominal amputation with notes on two cases. Br J Surg. 1916;4:283–287.
34. Refaat Y, Gunnoe J, Hornicek FJ, et al. Comparison of quality of life after amputation or limb salvage. Clin Orthop. 2002;397:298–305.
35. Scully SP, Temple HT, O’Keefe RJ, et al. Role of surgical resection in pelvic Ewing’s sarcoma. J Clin Oncol. 1995;13:2336–2341.
36. Shin KH, Rougraff BT, Simon MA. Oncologic outcomes of primary bone sarcomas of the pelvis. Clin Orthop. 1994;304:207–217.
37. Steel HH. Partial or complete resection of the hemipelvis: An alternative to hindquarter amputation for periacetabular chondrosarcoma of the pelvis. J Bone Joint Surg. 1978;60A:719–730.
38. Stephenson RB, Kaufer H, Hankin FM. Partial pelvic resection as an alternative to hindquarter amputation for skeletal neoplasms. Clin Orthop. 1989;242:201–211.
39. Tomeno B, Languepin A, Gerber C. Local Resection with Limb Salvage for the Treatment of Periacetabular Bone Tumors: Functional Results in Nine Cases. In Enneking WF (ed). Limb Salvage in Musculoskeletal Oncology. New York, Churchill Livingstone 147–155, 1987.
40. Vaughan L, Tomford WW, Mankin HJ. Pelvic and Proximal Femoral Resective Surgery and Allograft Replacement. In Evarts CM (ed). Surgery of the Musculoskeletal System. Ed 2. New York, Churchill Livingstone 3185–3210, 1990.
41. Windhager R, Karner J, Kutschera HP, et al. Limb-salvage in periacetabular sarcomas: A review of 21 consecutive cases. Clin Orthop. 1989;331:265–276.
42. Yoshida Y, Osaka S, Mankin HJ. Hemipelvic allograft reconstruction after periacetabular bone tumor resection. J Orthop Sci. 2000;5:198–204.
© 2004 Lippincott Williams & Wilkins, Inc.