Limb salvage: When, where, and how?

From an era where amputation was the only option to the current day function preserving resections and complex reconstructions has been a major advance in the treatment of musculoskeletal sarcomas. The objectives of extremity reconstruction after oncologic resection include providing skeletal stability where necessary, adequate wound coverage to allow early subsequent adjuvant therapy, optimising the aesthetic outcome and preservation of functional capability with early return to function. This article highlights the concepts of surgical margins in oncology, discusses the principles governing safe surgical resection in these tumors and summarises the current modalities and recent developments relevant to reconstruction after limb salvage. The rationale of choice of a particular resection modality and the unique challenges of reconstruction in skeletally immature individuals are also discussed. Striking the right balance between adequate resection, while yet retaining or reconstructing tissue for acceptable function and cosmesis is a difficult task. Complications are not uncommon and patients and their families need to be counseled regarding the potential setbacks that may occur in the course of their eventual road to recovery, Limb salvage entails a well orchestrated effort involving various specialties and better outcomes are likely to be achieved with centralization of expertise at regional centers.

INTRODUCTION

The last few decades have seen rapid strides in the field of musculoskeletal oncology. Amputation no longer remains the only option to achieve local control in malignant bone and soft tissue tumors of the extremity. Function preserving alternatives in these lesions have now become the norm without compromising on overall disease survival and have resulted in a documented improvement in overall quality of life of patients compared with those with an amputation.1 Various studies have confirmed that limb salvage is superior to amputation in preserving function.23 Although certain species such as starfish or salamanders are capable of regenerating whole extremities, in humans surgeons need to rely on advances and innovations in biological and mechanical reconstruction for restoration of function after major resections for tumors.4 The advent of better imaging modalities, more effective chemotherapy, improved radiotherapy techniques, a better understanding of anatomy with continuous refinement in surgical techniques and advances in prosthesis design and materials have all played a part in enabling this goal.

Though the number of limb salvage surgeries undertaken for malignant tumors of the extremity has increased the principles that govern surgical resection of bone and soft tissue tumors have remained unchanged. Limb salvage is recommended only if:

The ability to achieve adequate margins is not compromised. An adequate margin is one that would result in an acceptably low rate of local recurrence of the tumor. If the surgeon is unable to achieve adequate margins in his endeavor to salvage the limb then an amputation is preferred

The salvaged limb will provide function superior to that offered by a prosthetic limb after an amputation. A desensate salvaged limb with inadequate motors defeats the very purpose of limb salvage, which aims at improving the patient's quality of life.

Balancing these two opposing goals can often be a Herculean challenge, especially in patients with large tumors. Kawaguchi's concept of “barrier effects” has helped surgeons better understand evaluation of margins of resection.5 Though conventionally quantitative parameters were used to define resection margins Kawaguchi converted anatomical structures (any tissue that has resistance against tumor invasion such as muscle fascia, joint capsule, tendon, tendon sheath, epineurium, vascular sheath, and joint cartilage) into definitive thickness of normal tissue and classified them as either a thick barrier or a thin barrier. For purposes of margin evaluation a thick barrier was equivalent of 3 cm thickness of normal tissue and a thin barrier was considered to be 2 cm. Intact joint cartilage was equivalent of 5 cm thickness of normal tissue. By considering barrier effects translated into concrete distance equivalents, oncologically safe surgery can be planned at sites where barriers exist by using margins less than those mandated by true physical distance.

Though, it remains our endeavor to offer limb salvage to majority of our patients certain adverse factors often make these surgeries more complex. Poorly placed biopsy incisions, major vascular involvement, encasement of a major motor nerve, preoperative infection and inadequate motors after resection are hurdles that may need to be overcome using the advances in microsurgical techniques that offer the ability to transfer motors, graft nerves and vessels and provide adequate soft tissue even after extensive resections of overlying skin, muscles, and neurovascular structures.6

Though surgical resection remains the mainstay of treatment in musculoskeletal tumors it is uncommon for a patient with a high grade sarcoma to be treated by surgery alone. Adjuvant modalities like chemotherapy and radiotherapy play an essential part in the integrated management of these patients and the surgeon must be aware that continuous interaction and coordination between the various treating disciplines is important in order to provide the different treatment modalities in the most optimum sequence at appropriate times.

Surgery in osteosarcoma and Ewing sarcoma is usually preceded by neoadjuvant chemotherapy for a duration of approximately 3 months (based on the specific protocol used) prior to resection. In these patients who are immuno compromised it is necessary to ensure that the blood counts have regained appropriate levels (Absolute Neutrophil Count >1500) prior to proceeding with surgery. Even though the tumor may have shown a good response to neoadjuvant chemotherapy the margins of surgical resection are based on the initial extent of disease (or the imaging that shows maximum extent of disease) to ensure adequate oncologic clearance. As most of these patients would also need adjuvant therapy following surgery, surgery must be planned and executed in such a manner so as not to unduly disrupt the delivery of these adjuvant modalities. Problems in wound healing or postoperative infection can result in a delay in the postoperative delivery of these modalities, which could compromise both local and distant disease control.

Limb salvage therefore requires a well coordinated multidisciplinary approach involving varied specialties. The objectives of extremity reconstruction after oncologic resection include providing skeletal stability where necessary, adequate wound coverage to allow early subsequent adjuvant therapy, optimizing the aesthetic outcome and preservation of functional capability with early return to function.

This article summarizes the current modalities and recent developments relevant to reconstruction after limb salvage.

Search strategy

A comprehensive literature search of the Medline database using various keyword combinations of “sarcoma, bone tumor, limb salvage, megaprosthesis” was executed.

The search was filtered to include studies published from 2005 onwards. Of the 628 results retrieved suitable abstracts were reviewed. Case reports were excluded. Where appropriate, full articles were read and cross references retrieved. Based on the literature collated, a summary of the current modalities and recent developments relevant to reconstruction after limb salvage is presented.

Reconstruction options for limb salvage in extremity bone tumors

There are a variety of reconstruction options after excision of bone tumors. Metallic prostheses (megaprostheses), which span the resection gap and allow for movement of the joint form the mainstay in limb salvage surgery for reconstruction after tumor resection, providing both mobility and stability.7 They provide an immediate return to function and unlike biologic alternatives (bone) are not affected by ongoing adjuvant chemotherapy and radiotherapy. Ten year prosthesis survival rates have ranged upward of 80% in most series from large volume centres.8 Studies have shown that a cemented, rotating-hinge prosthesis with a hydroxyapatite collar offers the best chance of long term survival for prosthesis around the knee with a remarkably low rate of aseptic loosening.9 It has been suggested that bone ingrowth may provide more durable biologic fixation of megaprostheses and help reduce the incidence of aseptic loosening which is one of the commonest causes of failure of endoprosthetic reconstruction with conventional stems. The uncemented Compress® Compliant PreStress Implant (Biomet, Warsaw, IN) (uses a novel spring system, which achieves immediate, high compression fixation that induces bone hypertrophy and avoids stress shielding. This may help avoid eventual aseptic failure of the prosthesis.10

Low cost, locally manufactured prosthesis are also now available and these remain the workhorse for surgeons in resource challenged settings for prosthetic reconstructions after limb salvage.1112 Though these prostheses did have initial problems with early failure, the advent of better manufacturing techniques and increasing surgeon involvement in design development have helped create a durable prosthesis option at more affordable costs over the past decade. These prostheses are now routinely being used even for total bone resections and total femur and total humerus replacements are not uncommon.1314

A composite of an allograft and prosthesis can also be used for reconstruction in certain situations. An allograft replaces the segment of bone resected, while prosthesis implanted in the allograft and host bone replaces the articular surface.15 The allograft helps restore bone stock (which may be beneficial in subsequent revision surgery) and provides a biological surface for soft tissue attachment while the prosthesis provides a reliable and stable articulation and support for the allograft. Though the functional rates in most series are comparable to those after use of a modular prosthesis alone, the complication rates, including infection are higher.1516 Hence, appropriate case selection is critical. Areas where an allograft-prosthetic composite reconstruction provides a distinct functional advantage are reconstructions of the proximal femur. In these the median hip abductor strength is greater for patients with an allograft-prosthetic composite when compared with those with prosthesis alone.17

Hemicortical resections after appropriate preoperative planning in select cases of low grade malignant tumors have been shown to be oncologically safe. Such bone preserving resections enable biological reconstructions, which are more durable in the long term.18 The resultant defect can be reconstructed using allografts, recycled tumor bone or autografts.1920

The advent of computer-assisted tumor surgery (CATS) has increased the accuracy of intended bone resection and has also enabled multiplanar osteotomies.21 CATS allows integration of imaging with surgical instrumentation using specially developed software and may be beneficial in resection and reconstruction of pelvic, sacral and difficult joint-preserving tumor surgery. It provides a useful tool in achieving a better balance between disease resection and preservation of function in anatomically challenging locations.22 To further reduce technical errors during navigation-assisted bone tumor resection, surgeons have also experimented with direct magnetic resonance imaging (MRI)-guided navigation surgery without image fusion using absorbable pins as temporary implanted bone markers that prevent artifacts on MRI.23 Computer aided design allows for the manufacture of customized prosthesis to accommodate for individual patient specific complex resections and reconstructions aimed at maximizing residual bone stock.24

Infection remains an inherent danger with the use of large metallic implants in immuno compromised patients.25 Various agents coated on the external surface of the prosthesis have helped reduce the rate of infection. Silver-coated prosthesis, vancomycin coated prosthesis and iodine impregnated implants have all shown promise in early results.262728 In a study comprising 158 patients with tumors in whom iodine-supported implants were used to prevent infection only three cases developed acute infection. In another series, the infection rate in patients with a silver-coated megaprosthesis was compared with the data for patients in whom uncoated titanium megaprosthesis was implanted. The infection rate was substantially reduced from 18% in the titanium group to 6% in the silver group. Whereas, 39% of patients in the titanium group ultimately had to undergo amputation when periprosthetic infection developed, these mutilating surgical procedures were not necessary in the silver-coated group.

Children, because of the dynamic nature of growing bones pose a unique challenge. They have a narrow medullary cavity (which limits the size of intramedullary prosthetic stems), continually remodeling bone and pose a greater functional demand on their reconstructions. The issue of ultimate limb length discrepancy at skeletal maturity also influences the choice of reconstruction especially in the lower limb. The use of free vascularized epiphyseal transfer after resection of extremity tumors has not been commonly employed with only the occasional case reports present in literature.2930 An expandable prosthesis is a commonly used solution to the problem of limb length discrepancy that would result in young children offered salvage with a megaprosthesis.31 The prostheses have special mechanisms to lengthen them at periodic intervals. Though these involve small incisions, the repeated surgical exposures for expansion increase the chances of infection with its subsequent sequelae. The newer generation expandable prostheses which permit expansion with noninvasive techniques can help reduce this risk. These prostheses have special mechanisms to lengthen them at periodic intervals by allowing graduated extension when subjected to a controlled external magnetic field [Figure 1]. However, the high cost of these “expandable” prosthesis and limited availability of a “low cost” alternative precludes the use of this option in a large majority of growing children.

Figure 1

Immediate postoperative radiograph of a noninvasive expandable prosthesis (left). The same prosthesis after serial expansion in vivo (right). The double headed white arrow demonstrates the area of expansion

Where cost constraints preclude the use of expandable prosthesis in a majority of children with residual growth, rotationplasty remains an often used alternative.32 Rotationplasty, first described by Borrgreve in 1930 and van Nes in 1950, and applied for tumors by Salzer et al. in 1981 is also useful occasionally in adults where large tumors necessitate resection of extensive skin and surrounding soft tissue precluding conventional limb salvage.

As skin, quadriceps and the vessel can be sacrificed to provide a wider margin this procedure is also applicable in cases with extensive involvement of the quadriceps or where skin has undergone prior radiation. It is also used to salvage cases with uncontrolled infection following a prosthetic replacement.33 Though the most common site where rotationplasty is used is for tumors of the distal femur and proximal tibia, it is also suitable for tumors involving the entire femur. For young children it is possible to simply resect the femur and insert the lateral tibial plateau into the hip joint, producing a rotated limb with a false hip34 [Figure 2]. This will remodel to produce a remarkably good hip joint. In older adults too, the same technique is applicable. Here, instead of articulating the tibial plateau with the acetabulum, a cemented bipolar prosthesis is inserted into the proximal tibia which in turn articulates with the acetabulum to form a “new hip joint.”

Figure 2

Rotationplasty (a) preoperative magnetic resonance imaging (osteosarcoma) showing involvement of the entire femur, (b) schematic figure showing insertion of the lateral tibial plateau into the hip joint after resecting the femur, producing a rotated limb with a false hip (c) rotationplasty radiograph on followup

In some cultures, the cosmetic deformity produced by this operation makes it difficult to accept, but the functional benefits of this procedure have made it popular, especially in areas where cost constraints involved in use of an expandable prosthesis for children can often be a dominating factor in decision making.32 A 10-year followup study of patients with rotationplasty found no reduction in psychosocial adaptation, and similar life contentment as in healthy persons. Based on these findings, the authors recommended rotationplasty instead of amputation whenever conventional limb salvage was not possible.35

A similar concept utilizing resection of all surrounding skin, soft tissue and bone to ensure adequate oncologic clearance is applicable in large or contaminated lesions of the upper limb around the elbow.36 The entire intercalary segment is resected leaving only neurovascular continuity between the distal and proximal fragments. Internal fixation between the residual humerus and ulna in appropriate position facilitates a bony arthrodesis. Care must be taken to resect an adequate segment of the proximal radius to ensure that subsequent pronosupination is retained. The resultant “fused elbow” retains excellent hand function. The upper limb though cosmetically shorter than the opposite side offers a residual function far better than after an above elbow amputation.

Biological means of reconstruction using autografts, allografts and reimplantation of sterilized tumor bone (after autoclaving/pasteurisation/irradiation) offer an attractive alternative option in certain scenarios. Conventional strut allografts and fibula autografts have the disadvantage of being unable to provide a mobile articulating surface. Thus, spanning a defect involving the articular surface with these would result in an arthrodesis and inability to move that particular joint. Osteochondral allografts were used in an attempt to recreate a mobile joint, but long term data has not been very encouraging with one series reporting that 70% of patients experienced an event (fracture/progressive arthritis/nonunion/infection) during the followup period. 60% of the allografts needed removal at a mean of 5-year.37

Though arthrodesis of the knee may not be a favored option after resection because of the functional limitations, it is still a popular alternative in the developing countries. It is difficult to argue against the durability of this robust and inexpensive reconstruction, especially after resection of large benign lesions where the patient is expected to have a normal life expectancy. The physical demands placed on a reconstruction in patients whose livelihood depends on hard manual labor can be a deterrent to the use of prosthesis. The potential costs of possible future revision surgeries often makes patients opt for arthrodesis as the primary reconstruction modality. Another situation where an arthrodesis may be a possible option is in large tumors when a considerable amount of quadriceps muscle is sacrificed in order to obtain adequate oncologic clearance. A mobile joint may not be the best reconstruction modality in this scenario.

Traditionally autografts and allografts have been used to bridge the defect in order to achieve an arthrodesis. Some authors have also successfully demonstrated the use of a two-ring Ilizarov construct with bifocal bone transport over a thin-diameter long intramedullary nail as a cost effective means of achieving effective union with least complications.38 Besides conventional means of arthrodesis, it is possible to use cement spacers coupled with inexpensive internal fixation devices as a primary reconstruction option for limb salvage in these cases39 [Figure 3]. Besides being cost effective, cement spacers provide other advantages as well. The operating time is shorter compared with methods using biological reconstructs, which require either harvesting an auto graft or shaping an allograft to ensure optimal fit. The necessity of a specialized tissue bank for procuring allografts is not a constraint. Unlike biological reconstructions, cement spacers are unaffected by adjuvant treatment modalities. The rehabilitation schedule does not depend on evidence of graft incorporation or “hypertrophy” of the graft and patients are ambulant with immediate weight-bearing on a stable limb. In cases with fungating, infected tumors at presentation there is an obvious reluctance to use either prosthesis or grafts. Here, antibiotic impregnated cement spacers provide an alternative to an amputation. A potentially exciting addition could be the incorporation of antineoplastic drugs in the bone cement to enhance local drug delivery in large tumors. Healey has shown that polymethylmethacrylate cement mixed with doxorubicin retained adequate strength and sufficient amounts of the drug eluted to have potential biologic activity.40

Figure 3

(a) Preoperative magnetic resonance imaging (osteosarcoma) of the distal femur with additional lesions in the ipsilateral tibia (b) followup radiograph demonstrating the cement spacer with a long interlocking nail for fixation after en bloc extra articular resection

It is also possible to subsequently successfully carry out a secondary revision of the cement spacer to a prosthesis or biological arthrodesis if the need arises or the patient desires it.41

In children and adolescents with open physis when metaphyseal tumors are located close to the growth plate without transgression of the physis, Canadell's technique of physeal distraction allows for preservation of the epiphysis while ensuring a safe margin of excision.42 Physeal separation by gradual external fixator distraction is the first part of tumor resection. With the rupture of the growth plate, the metaphyseal osteotomy is already performed preoperatively and tumor resection can be completed by an intraoperative diaphyseal osteotomy followed by appropriate reconstruction.43

For tumors that involve the diaphyseal portion of a bone, an intercalary resection and reconstruction can be performed that saves the joints at either end. In these cases, the excised segment of bone can be replaced with either a metallic diaphyseal prosthesis or bone in the form of a strut allograft or fibular autograft. The pedicled or free vascularized fibula graft is among the most commonly used grafts in musculoskeletal oncology across various sites.4445 While it was initially hoped that massive allografts would become fully incorporated into the host, retrieval data shows that only a small percentage of the allograft actually becomes revascularized while the rest remains necrotic.46 Rather than a biologic replacement for the excised bone segment the allograft functions as a biologic spacer. Allografts have their share of complications too which include infection, nonunion and late fractures.4748 A retrospective review of patients who underwent massive proximal humeral allograft replacement for primary or secondary bone tumors documented an overall revision rate of the allografts of 75%.49 In another study, in spite of stringent patient selection in the use of large bulk structural allografts for limb preservation, 11 of 23 patients experienced at least one complication requiring a second procedure.50 Recommendations to reduce the number of failures include the use of bridging osteosynthesis with plate fixation and reconsidering the use of allografts for reconstructions of defects that are ≥15 cm, especially in older patients.51

A single vascularized fibula is often not strong enough to withstand the loading after reconstruction of large defects, especially in the lower extremity and fractures are not uncommon. In an attempt to improve the incorporation of allografts, while providing additional structural stability to the vascularized fibula, a combination using a strut allograft with a vascularized fibula autograft has been advocated.5253

Sequential controlled bone transportation as proposed by Ilizarov has also been used to fill defects created after skeletal resection.54 Defects are usually large and the process of bone regeneration can be a long drawn affair. The presence of multiple external pins and wires for extensive periods required by this method can be a problem in patients who are immuno compromised because of chemotherapy and therefore susceptible to pin tract infections. The quality of the bone regenerate may also be altered because of adjuvant therapies.

Recently, there has been a lot of interest in using the patient's own tumor bone and replacing it after it has been sterilized.55 Methods of sterilization described have included the use of autoclaving, microwave, pasteurizing, liquid nitrogen, and radiotherapy (extra corporeal radiotherapy). The principle is the same; the tumor-bearing bone is excised with adequate margins, all soft tissue and macroscopic tumor removed and the remaining bone sterilized by any of the above methods before being reimplanted [Figure 4]. Although the bone is dead the advantage is that it functions as a “size matched” allograft. An essential prerequisite is that the bone should initially not be damaged significantly by the tumor otherwise it would become too weak to use once sterilized. The problems inherent with allograft usage remain and hence it too can be combined with a vascularized graft. The technique is relatively time consuming, but inexpensive to use. As the patient's own bone is used it avoids the logistic issues involved in allograft procurement and the fear of disease transmission.

Figure 4

(a) Osteosarcoma in metadiaphyseal area of tibia, (b) radiograph at 18 months after host bone was reimplanted after extra corporeal radiotherapy (medullary canal of reimplanted bone is packed with cement) (arrows show the osteotomy sites)

An interesting concept that merits further investigation is biological reconstruction using pedicled frozen tumor-bearing autografts treated with liquid nitrogen. The operative procedure consists of exposing the tumor, performing one-site osteotomy or joint dislocation, rotating and freezing the tumor lesion in liquid nitrogen and subsequent reconstruction using intramedullary nailing, plates, or composite arthroplasty. This technique of biological reconstruction attempts to save the continuity of anatomical structures thus avoiding the drawbacks of free frozen autografts.56 This method is expected to result in early blood flow recovery, with early union and low complication rates. A study comparing the outcomes of pedicled frozen autografts and free frozen autografts in the lower extremity demonstrated that the union period was shorter and the rate of postoperative complications was lower with the pedicled graft.57

In the leg and the forearm, the use of the “fellow bone” as a vascularized graft (without micro vascular anastamosis) is useful in reconstructing long segmental defects after resection. Fibular centralization or the “fibula-pro-tibia” procedure has been used in combination with strut allografts and resterlized tumor bone. It has also been described as a standalone technique for reconstructing these defects.58 Following adequate resection of the tibia, the fibula is osteotomized proximally and distally at the appropriate level. It is then transposed medially keeping the entire soft tissue attachments intact and fixed proximally to the residual tibia and distally to the residual tibia or talus (in case of an intraarticular resection). This now functions as a vascularized graft resulting in rapid union with subsequent hypertrophy when subject to loading [Figure 5]. Ipsilateral fibula transfer is an easy technique that does not require micro vascular skills and can be accomplished by the primary operating surgeon himself. It also helps reduce operative time compared with transfers requiring vascular anastamosis. Volume reduction of the lower leg due to antero-medial shift of the fibula facilitates skin closure after tumor excision making this an attractive option in large tumors or in cases where excision of biopsy scar entails loss of excessive soft tissue.

Figure 5

(a) Preoperative magnetic resonance imaging of tibia showing an Ewing sarcoma (b) followup radiograph at 36 months showing hypertrophy of the transposed (medial translation into post excision defect) fibula

A similar concept is applicable in the upper limb. The distal ulna can be transposed to fill the defect left after resection of tumors of the distal radius and the wrist is arthrodesed.59 As in the lower limb, volume reduction of the forearm due to radial shift of the ulna facilitates skin closure after tumor excision especially in cases where fungating lesions or extensive soft tissue tumor components entail loss of soft tissue and skin. An ancillary benefit of the resultant one bone forearm is that the complications of ulnar variance, which result when there is inappropriate restoration of length after use of a fibula, iliac crest graft or allograft can be avoided. In spite of creating a one bone forearm, this technique retains prono supination unlike when the ulna is directly fused to the carpus with centralization of the carpus on the ulna.

The “induced membrane” or “Masquelet” technique has been described as an alternative for immediate biological reconstruction after tumor resection in children.60 Inducement of foreign body granulation tissue is a relatively novel therapeutic modality in bone repair. This two stage bone reconstruction method combines inducement of a granulation tissue membrane and subsequent bone autografting as a biphasic technique allowing reconstruction of large bone defects. During the first stage, a cement spacer is inserted after bone resection and stabilization. The cement spacer is removed during a second stage procedure performed after chemotherapy is completed and cortico-cancellous bone autograft is placed in the biological induced chamber created by the cement spacer. This bone grafts consolidate leading to rapid bone union. The advantage of this two stage procedure is that it helps reduce the operating time during the first stage and also reduces early complications despite major bone resection in patients receiving chemotherapy.

Pathological fractures in malignant bone tumors are no longer a contra indication for limb salvage. A recent meta-analysis demonstrated that limb salvage in appropriately selected patients of osteosarcoma with pathological fracture did not increase the risk for local recurrence or impact on overall survival rate compared with an amputation.61

Improvement in function after resection need not be restricted to the realm of limb salvage alone. In spite of the surgeon's best efforts there will still remain a group of patients in whom amputation remains the only option. Even in these cases the surgeon should endeavor to improve ultimate function by ensuring that optimal residual limb length is preserved or regained. Various procedures to augment the length of resected bone in order to improve prosthetic fitting and improve function after an amputation have also been described. When amputation just below the knee is necessary the remaining proximal tibia may be too short for a below-knee prosthesis, although the knee may be normal. Including the distal tibia or foot in a long posterior flap by turning it up and increasing the length of a very short proximal tibial stump is one of the methods described.62 The knee is thereby saved, allowing satisfactory use of below-knee prosthesis. This technique is particularly applicable when the distal leg is normal and well vascularised. Even after hip disarticulation functional outcome can be improved by preserving a musculocutaneous flap and placing a modular endoprosthesis in the acetabulum.63 It is worthwhile considering the use of autografts or allografts to augment residual limb length in patients for whom traditional amputation techniques would result in poor function, difficulty in fitting a prosthesis, or greater than necessary anatomic loss.64

Forequarter and hindquarter amputations are not uncommon for large proximal limb tumors. For these large defects that require a free flap, the distal uninvolved portions of these limbs can be harvested as fillet flaps and represent the “spare parts” concept of surgical reconstruction.65 The use of the fillet flap has been shown to be oncologically sound, has no associated donor sites and permits rapid wound healing with an improvement in the quality of life.

Use of nonconventional modalities for local control like microwave induced hyperthermia and high intensity focused ultrasound has also shown promising results.6667 Encouraging local control and overall survival rates have been demonstrated in a few series using these experimental techniques. These techniques may thus eventually have the potential to be utilized as one of the components of limb sparing options in patients with malignant bone tumors.

Though an amputation results in increased disability and decreased functional scores when compared to limb sparing surgery, it must be stressed that limb salvage surgeries are complex endeavors and have their share of complications. Complications were 3 times more common after limb salvage procedures and 4 times more common after endoprosthetic reconstructions compared to after ablative procedures.68 Intraoperative injuries to neurovascular structures, deep infections, delayed unions or fractures when using biological methods of reconstruction, hardware failure and mechanical wear when using implants and prosthesis have all been well documented.69 In the growing skeleton limb length discrepancies and deformities due to partial loss of the physis are additional problems.70 In a study which included 22 patients with osteosarcoma who were treated with a limb sparing surgery and survived more than 20-year from the time of diagnosis, 14% underwent secondary amputation because of long term complications involving the endoprosthesis.71 A retrospective review of 2174 skeletally mature patients who received a large endoprosthesis for tumor resection classified modes of failure as: Soft tissue failures (Type 1), aseptic loosening (Type 2), structural failures (Type 3), infection (Type 4), and tumor progression (Type 5). There were 534 failures with the most common mode of failure in this series being infection (34%). 19% failures were from aseptic loosening at the bone implant interface, and 17% were due to periprosthetic or prosthetic fractures.72

CONCLUSION

The surgeon must decide with the patient what the best surgical procedure is for that individual and he is then responsible for achieving adequate margins and reconstructing the limb if limb salvage is chosen. Properly indicated and executed limb salvage offers the advantage of better function and psychological benefits resulting in an overall improvement in quality of life. It entails a well orchestrated effort involving various specialties and better outcomes are likely to be achieved with centralization of expertise at regional centers with adequate infrastructure so that surgeons and their teams can offer a full range of surgical options to their patients, based upon experience and knowledge. The cost of treatment can be expensive and the postoperative rehabilitation is prolonged often requiring increased inpatient hospital care. Striking the right balance between adequate resection, while yet retaining or reconstructing tissue for acceptable function and cosmesis is a difficult task and complications are not uncommon.6872 Patients and their families need to be counseled regarding the potential setbacks that may occur in the course of their road to recovery, but the eventual satisfaction achieved by both, patient and surgeon after a successful limb salvage is unparalleled and is the elixir that drives oncology surgeons to sail into uncharted waters and to scale new heights.