Houdek, Matthew T. MD; Kralovec, Michael E. MD; Andrews, Karen L. MD
Hemipelvectomy-level amputations are typically performed at large tertiary medical centers for the treatment of soft tissue and bone tumors1,2; however, other indications include infection, ischemia, previous myelopathy with recurrent nonhealing wounds, and extensive trauma.3 Of the malignant processes that occur in the pelvis, 10%–15% are primary bone tumors and 5% are soft tissue sarcomas.4–8 For several decades, amputation was the treatment of choice for advanced extremity tumors. Although surgery is the most common treatment of pelvic tumors, radiation therapy and chemotherapy may also be used. An external or standard hemipelvectomy is most often performed for tumors that cannot be adequately resected through limb-sparing techniques and to provide palliation for patients with pain or distress that cannot be managed with lesser means.9,10
Trauma-related hemipelvectomy is a rare, devastating, and often fatal injury that poses a number of challenges to the treating orthopedic trauma surgeon.11 Similarly, combat-related hemipelvectomy is generally indicated if there is insufficient soft tissue coverage after an injury to the proximal limb that is complicated by life-threatening local infection and/or a necrotic and dysvascular hemipelvis after early ligation of critical intrapelvic vasculature.11 Typically, these patients are young and otherwise healthy, requiring long-term medical and rehabilitative care after the amputation.11,12
Hemipelvectomy-level amputation involves amputation of the lower extremity along with the hemipelvis.1 This procedure was first successfully performed in 1895; however, it remains a rare procedure.13 The complexity of the surgical anatomy makes hemipelvectomies one of the most technically demanding and invasive surgical procedures.4,14
Transpelvic amputation surgery frequently has more variability than other amputation levels because of the amount of tissue involved. More recently, with multimodality therapy and advances in radiographic imaging, surgical techniques have been developed that allow limb salvage.9 For cases in which limb salvage can be achieved, an internal hemipelvectomy can be accomplished through an abdominoinguinal incision.15 In this procedure, the pelvis is either reconstructed with one of a variety of techniques or left unreconstructed. Reconstructive options include saddle prostheses, allograft/prosthetic composite, or custom-made implants. The goal of these reconstructions is to provide the patient with a stable functional hip joint for ambulation and restore anatomic leg length.10
There are various techniques for an external hemipelvectomy that are based on the amount of bone removed and the location of the soft tissue flap used for coverage of the wound (Fig. 1).16 The standard hemipelvectomy implies that the innominate bone is removed from the pubic symphysis to the sacroiliac joint.16 Variations in this procedure include the extended and the modified hemipelvectomy. In the extended procedure, portions of the sacrum, the lumbar spine, or the contralateral innominate bone are removed.16 Modified hemipelvectomies occur when less than the entire innominate bone is removed.16
Soft tissue coverage is a large part of the preoperative planning of the hemipelvectomy, with three common flaps used: posterior, anterior, and thigh fillet.16 The posterior approach (Fig. 2A) is used for most hemipelvectomies. In this procedure, a myocutaneous posterior flap is created consisting of skin, subcutaneous fat, fascia lata, and the gluteus musculature.15,16 This flap is used for tumors that are located anteriorly, and the skin and gluteal musculature are not involved.1 In cases in which the tumor invades the upper portion of the buttock, this tissue is commonly removed to achieve an adequate margin. To provide coverage in these cases, an anterior flap (Fig. 2B) can be created.1,15,16 This anterior flap is based on the superficial femoral vessels and contains skin, subcutaneous fat, fascia lata, and a portion of the sartorius muscle.1,16 The thigh flap is made up of the musculature of the thigh and supplied by the profunda femoral vessels.16 One of the unique advantages to the use of the anterior and the thigh fillet flap is that these both include femoral cutaneous nerves, providing a sensate flap.16
Early on, it was reported that half of patients undergoing a hemipelvectomy died of complications related to the surgical procedure.17 Through advances in surgical technique and anesthesia, this number has decreased to 0%–10%.2,12 Although postoperative complications are common, these can be successfully managed in the vast majority of patients. With progress in surgical oncology, chemotherapy, and radiation therapy, life expectancy is prolonged when comprehensive treatment is begun before metastasis occurs.
Wound complications after hemipelvectomy have been reported to range from 20% to 80%.14,16,18,19 In a recent study, age, smoking, American Society of Anesthesiologists (ASA) class tumor characteristics, intent of operation, preoperative radiation or chemotherapy, diabetes, obesity, or volume of blood transfusion was not statistically significant in predicting hemipelvectomy wound morbidity.14 Length and extent of the operation and surgical wound class were statistically significant variables that were associated with development of wound infection and flap necrosis.14,20 Although spinal or visceral resection increased wound infection rate, it did not increase flap necrosis rates.14 All types of soft tissue flaps have been shown to provide adequate immediate soft tissue coverage. The level of vascular ligation strongly influences posterior hemipelvectomy flap necrosis.16
Phantom pain is frequently seen with hemipelvectomy-level amputation, with studies showing up to 90% of patients reporting phantom pain.12,21,22 One factor that has been shown to increase the risk for phantom pain and one of the most common chief complaints of patients undergoing hemipelvectomy is poorly controlled preoperative pain.22,23 Unlike phantom pain, residual limb spasm is more typically seen after traumatic amputation.12 Residual limb spasm is a movement disorder (dyskinesia) that occurs weeks to months after amputation.20 Although the exact pathogenesis of this process is not fully understood, it is thought to be the result of injury to peripheral nerves either at the time of trauma or during surgery. The pain can be debilitating.12,20 The incidence rate for both of these complications highlights the need for a multimodal pain and palliative care team for the treatment of both preoperative and postoperative pain.22
Rehabilitation after hemipelvectomy is optimally managed by a multidisciplinary integrated team. Understanding the functional outcomes for this population assists the rehabilitation team to counsel patients, plan goals, and determine discharge needs. The most important rehabilitation goal is the optimal restoration of the patient’s functional independence.24
A retrospective study found no difference in function at three different time frames between internal hemipelvectomy (limb sparing) and external hemipelvectomy (amputation).25 Refaat et al.26 noted similarities in the quality-of-life of patients who had amputation and those who had limb salvage for treatment of high-grade sarcomas of the pelvis.25 Guo et al.27 found that 43% of patients with internal hemipelvectomy and 53% of patients with external hemipelvectomy underwent acute inpatient rehabilitation. The median rehabilitation length of stay was 22 days for the patients after hemipelvectomy and 20 days after internal hemipelvectomy.
As soon as the patients’ standing tolerance allows and the drains are removed, they are cast for a pelvic leveler. Given the morbidity associated with hemipelvectomy surgery, some patients are unable to undergo successful prosthetic rehabilitation because of medical and wound healing complications. Depending on the associated neurologic deficits after hemipelvectomy, patients can (1) function from a wheelchair base (and require comprehensive rehabilitation to address seating and equipment needs and optimize neurogenic bowel and bladder management); (2) use a combination of a wheelchair, a prosthesis, and assistive devices; or (3) primarily use a prosthesis.
Ten to twelve weeks after hospital dismissal, after the flap/incision is completely healed, patients are ready to proceed with casting for a hemipelvectomy prosthesis if indicated. They are instructed to continue to use compression (compression wrap or a garment) until that time to stabilize the volume of the residual soft tissue.
Factors such as age, sex, etiology, level of amputation, and general health play important roles in determining prosthetic use.28 Some authors have found that patients who accept high-level prostheses were younger than those who rejected them.28,29 The etiology of amputation has been shown to be a factor in predicting prosthetic use. Tumor patients are more likely to be successful prosthetic users.30,31
The literature on this level of amputation is mostly limited to case reports, case series, or descriptive articles discussing prosthetic management. If prosthetic rehabilitation is indicated, a person should be fit with prosthetic components that enable him/her to function as his/her individual medical conditions and possible technology allow.32 A study from 1975 evaluating energy expenditure in patients with transfemoral-level amputation had difficulty finding people with this level of amputation who could walk at a constant, comfortable rate for 4 mins and subjects who walked with an unlocked knee.33 Certainly, prosthetic rehabilitation has made significant advances since that time.
The three main criteria for successful prosthetic rehabilitation for patients with high-level amputation are comfort, function, and cosmesis. Recent advances in prosthetic technology have resulted in prosthetic limbs that weigh substantially less than those used in the past (Fig. 3A). The hemipelvectomy/hip disarticulation prosthesis has a custom total contact socket fabricated of soft plastic (polypropylene) that wraps around the pelvis. The current classic “Canadian hip disarticulation”–style prosthesis primarily describes the positioning of the hip joint anteriorly and proximally in relationship to the patient’s contralateral anatomic hip joint. This “new” hip joint positioning has carried on to the present designs since reported in 1957.34 The prosthetic knee is set posterior to the weight line in a position of stability (Fig. 3B).
To optimize the fit of the hemipelvectomy socket, dynamic casting with the patient standing is helpful (Fig. 4). The residual soft tissues, any residual bony prominences, and whether there is intact sensation or muscle flap innervation on the residual limb play an important role in prosthetic fitting and function. Unique impression acquisition techniques have evolved over time to include a method that has its goal of producing a model that not only precompresses and defines the volume of the soft tissue envelope (suspending the standing patient with a sling) but also identifies concentrated pressure–tolerant areas within the slung tissue that can be enhanced to further distribute the weight-bearing forces. This also limits the pistoning caused by inadequate precompression.
Hip disarticulation model acquisition differs from the hemipelvectomy technique in that it is usually done standing with the amputated-side ischial tuberosity bearing weight on a deformable material (such as memory foam) to define the primary weight-bearing skeletal structure. This is combined along with careful definition of both the affected- and contralateral-side iliac crest.
Given the differences of the residual tissues (hemipelvectomy patients do not have any remaining skeletal structure to accept weight-bearing pressure, whereas patients with hip disarticulation–level amputations do), the sockets, although similar in appearance, transfer the patients’ weight to the prosthetic componentry during stance phase in totally different ways. Hemipelvectomy weight bearing is accomplished through precompression of the tissue/musculature of the amputated side. Hip disarticulation weight bearing is directed through the remaining ischial tuberosity and selected soft tissue compression.
Custom soft tissue supplementation may be required to optimize pressure distribution and comfort of the socket (Fig. 5).
Suspension of hemipelvectomy prostheses frequently requires the addition of a shoulder strap (Fig. 6) because of the mobility of the soft tissue envelope between swing and stance phase. Suspending a hip disarticulation prosthesis is usually accomplished by enclosing the iliac crests, therefore limiting pistoning.
Traditional hip joints for this level of amputation are single axis and allow movement in just one plane. Newer single-axis hip joints allow postfabrication adjustments for alignment corrections (hip flexion angle, abduction/adduction, and internal/external rotation), and some have provision for hip extension assist. A recent advancement is a polycentric hip joint,35 with the following clinical benefits:
- It allows leg length reduction during swing phase. This helps reduce the risk for falling and increases security.
- It controls three-dimensional movement of the hip joint during walking (closely resembles the normal hip movement in the sound leg).
- It provides a large flexion angle that makes it easier to accomplish activities of daily living (putting on shoes or getting into a car).
- It is noticeably easier to initiate swing phase with this joint’s integrated expansion springs (energy stored during the stance phase is used during the swing phase initiation to compensate for the missing hip muscles and to reduce the amount of energy needed for walking).
- It allows a dampened, controlled heel strike in the stance phase with significantly reduced hyperlordosis and a natural hip joint extension (the user can accomplish a more controlled, smooth roll-over on the prosthesis under full load).
- It allows an individual stride length setting, which controls the pendulum motion in the swing phase.
- It improves sitting posture (reduces pelvic obliquity to a minimum).
Motion analysis results show that the polycentric hip joint can reduce gait abnormalities when compared with the more commonly used uniplanar hip joint.36 The polycentric hip joint can be used only with an appropriately functional microprocessor knee. This knee will provide the following benefits when used for hemipelvectomy and hip disarticulation prostheses:
- Physiologic gait pattern (hip and knee flexion are initiated concurrently)
- Stability when walking (microprocessor stance control)
- Alternative modes for additional movement patterns not covered by the normal walking mode
- Increased patient safety
Microprocessor knees have sensors that measure biomechanic parameters such as pylon bending moment and knee angular velocity. Data are fed into a microcontroller that modifies mechanical factors to optimize gait control. Typically, this is accomplished via Servo-controlled adjustments to a hydraulic valve that alters the resistance to outflow and thereby alters the dampening of the knee unit. Ultimately, these automatic adjustments allow for ongoing adaptation to the instantaneous functional needs of the amputee.37
Prosthetic training includes education in donning and doffing the prosthesis and progression of mobility from weight shifting to ambulating and functional transfers. Prosthetic training is “a process.” Dynamic training and fitting/alignment adjustments are done concurrently to optimize results.
Before ambulation training, patients are taught abdominal strengthening, pelvic tilt/control, and controlling/contracting any muscle/tissue that would influence control or stability of the prosthesis. The patients are also instructed in the function of their prosthetic components. Ambulation training starts in the parallel bars. The patients ambulate using a pelvic tilt to initiate the prosthetic-side step. Weight bearing through the prosthesis increases as the residual tissues tolerate more pressure. Goals for training in the prosthesis include developing equal stance time from side to side and equal step lengths. As patients demonstrate consistency with these parameters, ambulation training is advanced to a walker or pair of crutches. As patients develop skill and confidence walking in the prosthesis on level surfaces, training advances to high-level activities such as climbing stairs, walking on inclines, and using fewer gait aids as possible. In the event of a fall, patients are also taught fall recovery strategies. After patients are able to demonstrate some independence in donning and doffing and safe ambulation, they begin to use their prosthesis outside the therapy environment. At that point, they are provided a schedule to gradually increase the wear time of the prosthesis.
There is limited outcomes literature on patients with hemipelvectomy-level amputation. As a result, most of the outcomes for high-level amputation have been derived from patients with hip disarticulation. Twelve and 30 years ago, walking with a hip disarticulation or hemipelvectomy-level prosthesis was shown to increase energy cost by 80%–250%.38,39 Thirty years ago, the energy cost of walking with crutches (without a prosthesis) at a comfortable walking speed for patients with a hip disarticulation was 45% more than for able-bodied subjects.38 In the past, this was thought to explain why patients with hip disarticulation or hemipelvectomy would prefer to use crutches instead of a prosthesis. Waters et al.40 found that the velocity selected by persons with amputation was significantly lower than that by the control group and that the velocity was lower with higher level of amputation. The rate of oxygen utilization per minute was similar between amputee patients and controls. The patients with amputations modified their walking speed to keep relative energy cost within normal limits. The slower walking speed of amputees is a measure of the loss in efficiency.40 Nowroozi et al.38 reported that patients with hemipelvectomy-level amputations walked approximately 50% slower and spent approximately 125% more energy than able-bodied persons of the same age. They evaluated energy expenditure for eight patients with hip disarticulation and ten patients with hemipelvectomies. Comfortable walking speeds with hip disarticulation and hemipelvectomy-level amputation were 51%–61% of controls, but there was no difference in O2 uptake per minute among the three groups.38 Significant reduction in energy consumption between hip disarticulation and hemipelvectomy has not been demonstrated.38
A cross-sectional descriptive study from the Netherlands showed that patients achieve a relatively high level of activity after hip disarticulation and hemipelvectomy but experience considerable limitations in walking, standing, sitting, and climbing stairs.21
A recent review of prosthetic rehabilitation at the authors’ institution showed that 18 (41.8%) of 43 patients successfully used a prosthesis. Successful use was defined as the patient being discharged from physical therapy with a prosthesis that he/she used at least three times per week. Of the 25 patients who were not successful wearers, 9 patients were not offered prosthetic fitting, 6 patients were unable to be fit with a prosthesis because of wound healing complications, 5 patients never became medically stable enough for fitting, 4 patients were fitted with a prosthesis but chose not to wear it, and 1 patient died before fitting.41 Body mass index, age, marital status, arthritis, depression, cerebrovascular disease, diabetes mellitus, pulmonary disease, previous orthopedic surgery, and dementia were not statistically significant predictors of successful prosthetic fitting. In addition to increased survival, successful prosthetic users reported that their prosthesis made a significant impact on their lives. On average, a successful user wore his/her prosthesis for 5.8 hrs per day. All were able to ambulate with or without the use of gait aids for a mean distance of 158 ft between rest periods.41 The most common reason for patients not wearing a prosthesis was that they were not offered one.41
Prosthetic use after hemipelvectomy improves balance, decreases the need for a gait aid, and optimizes prosthetic restoration. Using a prosthesis helps maintain muscle strength and tone, cardiovascular health, and functional mobility. With new advances in prosthetic components, patients are choosing to use their prostheses for primary mobility. Previously, patients with a hemipelvectomy walked a speed 50% slower than healthy controls.38 Advances in prosthetic management are beginning to optimize functional mobility with the prosthesis. In a recent report, the mean walking speed of 0.84 m/sec observed with a prosthesis was faster than previously reported walking speeds of vascular amputees of both transtibial and transfemoral amputations and similar to patients with traumatic transfemoral amputations.42 There was a marginal statistical difference between a control group and the patient groups (P = 0.06) but no difference (P = 0.44) in the time it took to negotiate the stairs using a prosthesis or crutches.42
Rehabilitation after hemipelvectomy is optimally managed by a multidisciplinary integrated team. Knowledge of rehabilitation options after high-level amputation can facilitate realistic goal setting during presurgical counseling and in the perioperative period. With new advances in prosthetic components, patients are choosing to use their prostheses for primary mobility.
The authors thank Michael Gozola, CP; Linda B. Arneson, PT; and Steven D. Bogard III, PT, for their assistance with the preparation of this manuscript.
1. Karakousis CP, Vezeridis MP.: Variants of hemipelvectomy. Am J Surg. 1983; 145: 273–7
2. Carter SR, Eastwood DM, Grimer RJ, et al.: Hindquarter amputation for tumours of the musculoskeletal system. J Bone Joint Surg Br. 1990; 72: 490–3
3. Mat Saad AZ, Halim AS, Faisham WI, et al.: Soft tissue reconstruction following hemipelvectomy: Eight-year experience and literature review. ScientificWorldJournal. 2012; 2012: 1–9
4. Wirbel RJ, Schulte M, Mutschler WE.: Surgical treatment of pelvic sarcomas: Oncologic and functional outcome. Clin Orthop Relat Res. 2001; 390: 190–205
5. Aboulafia AJ, Malawer MM.: Surgical management of pelvic and extremity osteosarcoma. Cancer. 1993; 71:(suppl): 3358–66
6. Abudu A, Grimer RJ, Cannon SR, et al.: Reconstruction of the hemipelvis after the excision of malignant tumours. Complications and functional outcome of prostheses. J Bone Joint Surg Br. 1997; 79: 773–9
7. Campanacci M, Capanna SR.: Pelvic resections: The Rizzoli Institute experience. Orthop Clin North Am. 1991; 22: 65–86
8. Enneking WF, Dunham WK.: Resection and reconstruction for primary neoplasms involving the innominate bone. J Bone Joint Surg Am. 1978; 60: 731–46
9. Grimer RJ, Chandrasekar CR, Carter SR, et al.: Hindquarter amputation: Is it still needed and what are the outcomes? Bone Joint J. 2013; 95B: 127–31
10. Griesser MJ, Gillette B, Crist M, et al.: Internal and external hemipelvectomy of flail hip in patients with sarcomas: Quality-of-life and functional outcomes. Am J Phys Med Rehabil. 2012; 21: 24–32
11. D’Alleyrand JC, Fleming M, Gordon WT, et al.: Combat-related hemipelvectomy. J Surg Orthop Adv. 2012; 21: 24–32
12. Ebrahimzadeh MH, Kachooei AR, Soroush MR, et al.: Long-term clinical outcomes of war-related hip disarticulation and transpelvic amputation. J Bone Joint Surg Am. 2013; 95: e1141–6
13. Ravtich MM, Wilson TC.: Long-term results of hemipelvectomy. Ann Surg. 1964; 159: 667–82
14. Senchenkov A, Moran SL, Petty PM, et al.: Predictors of complications and outcomes of external hemipelvectomy wounds: Account of 160 consecutive cases. Ann Surg Oncol. 2008; 15: 355–63
15. Karakousis CP, Emrich LJ, Driscoll DL.: Variants of hemiplevectomy and their complications. Am J Surg. 1989; 158: 404–8
16. Senchenkov A, Moran SL, Petty PM, et al.: Soft-tissue reconstruction of external hemipelvectomy defects. Plast Reconst Surg. 2009; 124: 144–55
17. Taylor GW, Rogers WP Jr.: Hindquarter amputation; Experience with eighteen cases. N Engl J Med. 1953; 249: 963–9
18. Higinbotham NL, Marcove RC, Casson P.: Hemipelvectomy: A clinical study of 100 cases with five-year follow-up on 60 patients. Surgery. 1966; 59: 706–8
19. Douglass HO Jr, Razck M, Holyoke ED.: Hemipelvectomy. Arch Surg. 1975; 110: 82–5
20. Marion MH, Gledhill RF, Thompson PD.: Spasms of amputation stumps: A report of 2 cases. Mov Disord. 1989; 4: 354–8
21. Yari P, Digkstra P, Geertzen J.: Functional outcome of hip disarticulation and hemipelvectomy: A cross-sectional national descriptive study in the Netherlands. Clin Rehabil. 2008; 22: 1127–33
22. Parsons CM, Pimiento JM, Cheong D, et al.: The role of a radical amputations for extremity tumors: A single institution experience and review of the literature. J Surg Oncol. 2012; 105: 149–55
23. Nikolajsen L, Jensen TB.: Phantom limb pain. Br J Anaesth. 2001; 87: 107–16
25. Beck LA, Einertson MJ, Winemiller MH, et al.: Functional outcomes and quality of life after tumor-related hemipelvectomy. Phys Ther. 2008; 88: 916–27
26. Refaat Y, Gunnoe J, Homieck FJ, et al.: Comparison of quality of life after amputation or limb salvage. Clin Orthop Relat Res. 2002; 297: 298–305
27. Guo Y, Fu J, Palmer JL, et al.: Comparison of postoperative rehabilitation in cancer patients undergoing internal and external hemipelvectomy. Arch Phys Med Rehabil. 2011; 92: 620–5
28. Fernandez A, Formigo J.: Are Canadian prosthesis used? A long-term experience. Prosthet Orthot Int. 2005; 29: 177–81
29. McAnelly RD, Refacian M, O’Connell DJ, et al.: Successful prosthetic fitting of a 73-year-old hip disarticulation amputee patient with cardiopulmonary disease. Arch Phys Med Rehabil. 1998; 79: 585–8
30. Ferrapie AL, Brunel L, Besse W, et al.: Lower limb proximal amputation for tumour: A retrospective study of 12 patients. Prosthet Orthot Int. 2003; 27: 179–85
31. Denes Z, Till A.: Rehabilitation of patients after hip disarticulation. Arch Orthop Trauma Surg. 1997; 116: 498–9
32. Kannenberg A, Zacharias B, Mileusnic M, et al.: Activities of daily living: Genium bionic prosthetic knee compared with C-Leg. J Prosthet Orthot. 2013; 25: 110–7
33. Traugh GH, Corcoran PJ, Reyes R.: Energy expenditure of ambulation in patients with above-knee amputations. Arch Phys Med Rehabil. 1975; 56: 67–71
34. McLaurin CA.: The evolution of the Canadian-type hip disarticulation prosthesis. Artif Limbs. 1957; 4: 22–8
35. Blumentritt S, Ludwigs E, Bellmann M, et al.: The new helix 3D hip joint. Orthop Tech. 2008; 5: 1–5
36. Ludwigs E, Bellmann M, Schmalz T, et al.: Biomechanical differences between two exoprosthetic hip joint systems during level walking. Prosthet Orthot Int. 2010; 34: 449–60
37. Marx LJ, Michael JW.: Science, medicine, and the future: Artificial limbs. BMJ. 2001; 323: 732–5
38. Nowroozi F, Salvanelli ML, Gerber LH.: Energy expenditure in hip disarticulation and hemipelvectomy amputees. Arch Phys Med Rehabil. 1983; 64: 300–3
39. Starc G.: Overview of hip disarticulation prostheses. J Bone Joint Surg Am. 2001; 13: 50–3
40. Waters RL, Perry J, Antonelli D, et al.: Energy cost of walking of amputees: The influence of level of amputation. J Bone Joint Surg Am. 1976; 58: 42–6
41. Kralovec ME, Andrews KL, Houdek MT.: Successful prosthetic rehabilitation following high level amputation. Paper presented at: Annual Meeting of the American Academy of Orthopedic Surgery. 2013;
42. Houdek MT, Kralovec ME, Andrews KL, et al.: Functional outcome measures of patients following hemipelvectomy. Paper presented at: Annual Meeting of the Musculoskeletal Tumor Society. 2013;