Allograft Prosthetic Composite and Endoprosthetic Replacement of the Proximal Femur: A Review

RECONSTRUCTION 2023, Jan, 26 | 12:00 am

Allograft Prosthetic Composite and Endoprosthetic Replacement of the Proximal Femur: A Review

Allograft prosthetic composite (APC) and endoprosthetic replacement of the proximal femur are surgical procedures used to treat severe bone loss in the hip. This article reviews the indications, contraindications, techniques, and outcomes of these procedures.

INTRODUCTION

Reconstruction of the proximal femur after tumor resection or in instances of failed arthroplasty remains a significant challenge, despite advances in surgical technique and prosthetic design. Depending on the pathology, a number of reconstructive techniques are in use, the most common being allograft prosthetic composite (APC) and endoprosthetic replacement (EPR) (1,2,3,4,5,6,7,8,9,10,11,12). In addition to sarcomatous lesions and failed hip arthroplasty with uncontained, circumferential, segmental bone loss (13), indications for an APC or segmental arthroplasty include metastatic disease and failed internal fixation of hip fractures. Culprits in the development of this magnitude of bone loss can include infection, aseptic loosening, or nonunion of fracture with implant cutout, among others.

INDICATIONS: WHEN TO CHOOSE EPR VERSUS APC?

It is important to consider the goals of surgery when contemplating an APC or endoprosthetic reconstruction. In a patient with bony malignancy and a short anticipated survival, full weight bearing and immediate mobilization are paramount to relieving pain and ensuring quality of life (1). As a result, endoprosthetic reconstruction is most commonly favored in older patients with metastatic disease (Fig. 29-1). Patients at significant risk of problems with bone healing, those requiring radiotherapy or a prolonged course of chemotherapy, as well as individuals in whom the abductor mechanism cannot be reconstructed are also best managed with an EPR (Table 29-1). That segmental

endoprostheses are readily available, are modular, and carry less risk of infection contribute to their widespread use (13). Disadvantages of this mode of reconstruction include a 10% rate of aseptic loosening (1,9,10,11,14) (Fig. 29-2), trochanteric escape (Fig. 29-3), and dislocation, which has been reported in up to 28% in some studies (1,9,10,13,15). Revision surgery requiring complete implant removal may also be more difficult if a long, cemented stem is used.

TABLE 29-1 Ideal Patient Selection for an Allograft Prosthetic Composite or an Endoprosthetic Reconstruction

FIGURE 29-1 A: Magnetic resonance imaging demonstrating a metastatic lesion in the right proximal femur. This 69-year-old female patient had a history of metastatic breast carcinoma and had developed significant functional pain. B: Postoperative plain radiographs demonstrating a cemented proximal femoral EPR with a hemiarthroplasty articulation.

Patient Characteristic

Allograft Prosthetic Composite Reconstruction

Endoprosthetic Reconstruction

Age Diagnosis

Metastatic disease

Younger

Older

Yes

Sarcoma

Possible

Yes

Failed arthroplasty

Yes

Abductor repair

Possible

Not possible

Large capacious

intramedullary canal

Yes

FIGURE 29-2 Aseptic loosening of a proximal femoral endoprosthetic reconstruction.

FIGURE 29-3 Plain radiograph of a cemented proximal femoral endoprosthetic reconstruction showing trochanteric escape with significant proximal migration of the fragment.

Implant durability and restoration of function, although always desirable, becomes much more critical in patients with long anticipated survival (1,13). Consequently, the arthroplasty patients or a young individual with good prognostic factors who is expected to survive a sarcoma represent potential candidates for an APC. This construct has the advantage of restoring bone stock, providing an anchor to which the abductors can be attached, and facilitating further revision surgery due to the shorter and uncemented nature of the prosthesis within the host proximal femur (1,13,15). Disadvantages primarily include the risk of infection, which can be as high as 20%, and nonunion at the host-graft junction in 10% of cases (4,5,6,12,13,15,16). The possibility of disease transmission (HIV, hepatitis, other viral disease), albeit rare, is also a consideration in recommending this surgical modality (13,15). Abductor dysfunction and graft resorption are significant concerns as are fractures, occurring in up to 60% of patients (13,15,16,17).

PREOPERATIVE PLANNING

In the setting of a known or suspected neoplasm, the appropriate oncologic workup should be carried out, including imaging, blood work, biopsy, and tumor staging. Magnetic resonance imaging of the tumor should cover the entire lesion, and ideally, span the length of the entire bone involved to identify any skip metastases. The site of the femoral osteotomy should be planned out with a minimum distance of 2 cm from the most distal aspect of the lesion. Computed tomography (CT) of the chest, abdomen, and pelvis should be obtained, in addition to other nuclear medicine scans where appropriate. If the lesion is not known to be metastatic in nature, the case should be discussed in a multidisciplinary tumor board meeting, and neo- or adjuvant therapy organized with medical or radiation oncologists where needed.

Cases of failed arthroplasty should be worked up for infection. Inflammatory markers, such as ESR and CRP,

should be ordered and joint aspiration carried out under sterile technique with ultrasound or CT guidance. Samples should be sent for a cell count, aerobic and anaerobic cultures, as well as mycobacterium and acid-fast bacilli. In our institution, aspirates are also sent for a synovial fluid CRP and alpha-defensin test (18,19,20), which together show promise in diagnosing periprosthetic joint infection with a sensitivity of 97% and a specificity of 100% (18). If the patient is indeed infected, a two-stage exchange, with placement of an antibiotic-impregnated spacer and a prolonged course of intravenous antibiotics in consultation with colleagues with expertise in infectious disease, should precede reconstruction.

Medical and nutritional optimization should be prioritized in order to reduce the risk of medical complications and afford the best chance of achieving bony union. Smoking is a relative contraindication to an APC. Informed consent should be obtained after an in-depth discussion of the risks and benefits of surgery.

ALLOGRAFT SELECTION

To select the size of implant or allograft, full-length radiographs of the femur and an anteroposterior radiograph of the pelvis should be obtained. Leg lengths should also be assessed clinically and radiographically for discrepancy, and any deformity of the distal portion of the femur should be noted. A CT scan can be helpful in measuring the size of the host bone canal at the level of the planned resection. Plain radiographs of the allograft are mandatory to identify that the correct bone has been ordered and to measure the width of the canal (Fig. 29-4). The intramedullary canal of the allograft should be slightly larger than that of the distal host bone, in order to facilitate cementing the prosthesis into the allograft while maintaining press-fit fixation in the host bone (Table 29-2). Measuring the host canal size is crucial because a fully porous-coated, distal fixation stem (Fig. 29-5) may not have sizes to accommodate a highly capacious canal. The technique described involves cementing the implant into the graft in one step and press-fitting the fully porous-coated cylindrical stem into the host bone. Unless an exceptionally large allograft is secured, one is limited to a maximum of a 16-mm stem. Larger diameter host bone requires a different reconstructive choice. This situation is not common with neoplastic resections but can be an issue in a revision arthroplasty situation. An endoprosthetic reconstruction, an intussusception allograft (13), or a plated allograft with a shorter cemented stem could be considered. Regardless of which method of reconstruction is planned, an EPR should always be available as a backup when attempting an APC in case of unanticipated complications. In most cases, the surgeon will only have one allograft available, so it is prudent to have a plan in place should the graft be the wrong size or become contaminated during the case. We always thaw and inspect the graft just prior to the start of anesthesia induction.

The manner of graft preparation and storage is important to consider when selecting an allograft because of the possibility of immune rejection and allograft antigenicity (15,21,22). These factors are thought to play an important role in bone resorption and subsequent fracture. For this reason,

1 2

fresh allograft should be avoided, despite superior resistance to bending and torsional stress (15,23). Irradiated bone is weaker and should also be avoided. Disease transmission is rare (21,22,23). The American Association of Tissue Banks-accredited tissue bank has stringent guidelines on graft preparation and provides bone that is deep frozen, our preferred method of processing. Furthermore, we order allograft with soft tissue attachments, namely the gluteal tendons and hip capsule (Fig. 29-6). This aids in abductor mechanism repair and needs to be specifically requested to avoid receiving a graft with no soft tissues whatsoever. It is important to meet with the allograft representative prior to ordering the graft. We prefer to review radiographs and specimen photographs prior to the final order.

FIGURE 29-4 Plain radiograph of an allograft proximal femur taken preoperatively. The canal width should be larger when compared with that of the host bone at the level of resection.

TABLE 29-2 Selecting an Allograft: Important Characteristics

Allograft Characteristics

Reason

Deep frozen

Prevent graft antigenicity, immune rejection, and

bone resorption

Intramedullary canal larger than host bone

Allow cementation of prosthesis into graft and press-

fit fixation of distal prosthesis into host bone

Longer than anticipated resection

Allows adjustments in length intraoperatively

Intact abductor mechanism (musculotendinous

portion of gluteus medius and minimus)

Allows reconstruction of abductor mechanism

Intact hip capsule

Aids in stability

FIGURE 29-5 Fully porous-coated, cylindrical stem for use in an APC reconstruction. Picture showing the implant trial, osteotomized allograft proximal femur, and the final implant.

FIGURE 29-6 Allograft proximal femur is ordered with soft tissue attachments to aid in the reconstruction of the abductor mechanism. A: Full-length allograft prior to osteotomy. B: Abductor tendons and (C) hip capsule are left intact and repaired to the host soft tissues.

PATIENT POSITIONING

After induction of anesthesia, the patient is intubated, and a urinary catheter is placed in the bladder. Preferably, general anesthesia is employed in combination with a periarticular injection. This injection is composed of ketorolac, epinephrine, morphine, and ropivacaine and can be infused into the joint and surrounding tissues in layers as the wound is being closed. Compression stockings and mechanical compression devices are placed on the contralateral lower leg to aid in preventing the development of venous thromboembolism. Leg length is evaluated in the supine position clinically, and the hip and knee are gently put through a range of motion. This latter maneuver is not performed if there is any concern for impending fracture. The patient is placed in the lateral decubitus position with the affected side facing up, and all bony prominences are well padded. A Montreal positioner or bean bag is used to position the patient, and care is taken to

ensure that the pelvis is in a neutral position. An axillary roll is placed to protect critical structures in the axilla. The leg is prepared with an iodine scrub in two stages. Sterile drapes are applied to the leg, followed by an iodine-impregnated adhesive drape to all-exposed skin. The knee should not be covered by impermeable drapes, lest extension of the incision is required. Great care is taken to minimize any contamination by the groin, as the incidence of infection following these cases is not insignificant. The leg lengths in this position are again checked with the affected side in slight abduction, neutral, and adduction.

SURGICAL APPROACH AND DEALING WITH THE ABDUCTOR MECHANISM

A direct lateral approach is carried out, taking the incision through subcutaneous tissue a few centimeters distal to the planned osteotomy site. Biopsy tracts are excised en bloc with tumor or previous arthroplasty incisions if there is adequate skin remaining for wound closure. Care should be taken throughout the approach to cauterize any bleeding vessels, as there is potential for significant blood loss in these cases (24). The tensor fascia lata and iliotibial band are incised, exposing the vastus lateralis. The patient's diagnosis will then dictate which structures in the thigh can be preserved. In the case of a sarcoma, a wide excision should be undertaken, leaving a normal cuff of muscle on the tumor. If the proximal femur and greater trochanter are affected by tumor, the gluteus maximus, medius, and minimus should all be tenotomized as long as possible and reflected proximally. The vastus lateralis tendon should be tenotomized and reflected distally, although a portion of this muscle often must remain to cover tumor. In the setting of revision arthroplasty, a direct lateral approach to the hip can be undertaken.

If preoperative MRI scans confirm that the greater trochanter is free of sarcoma, metastatic deposits, or significant bone loss from osteolysis, a wafer of bone approximately 1 cm in thickness can be preserved. A trochanteric slide osteotomy is the preferred scenario, leaving the abductors as well as the vastus lateralis insertion attached to the trochanteric fragment (Fig. 29-7). These structures can then be reflected anteriorly during the remainder of the dissection. If the vastus lateralis muscle has significant tumoral involvement proximally but the trochanter can be preserved, then a classic trochanteric osteotomy is the safest option.

Options for dealing with the abductor mechanism are found in Table 29-3. Likewise, on the allograft side, if the host trochanter can be preserved, then the graft trochanter can be removed with an oscillating saw on the back table in preparation for reconstruction (Fig. 29-7).

The vastus lateralis can then be carefully reflected off the femur to expose the osteotomy site or the junction of normal and osteolytic bone, as in the case of complex revision arthroplasty. In revision arthroplasty, as much host bone as possible is preserved to maximize surface area for healing to the allograft. Circumferential dissection is carried out, taking care to tag muscles for later repair. The surgeon should identify and ligate all intramuscular perforating vessels as they exit the posterior compartment over the length of the resection, as these have the potential for significant bleeding. Careful dissection of the femoral neurovascular bundle should be undertaken if it is close to the tumor on the medial side. The deep femoral artery and vein commonly require ligation in the setting of tumor.

FIGURE 29-7 Intraoperative picture depicting inset proximal femoral allograft. Trochanteric slide has been performed and is being retracted anteromedially using a gynecological tenaculum. A trochanteric osteotomy has been performed on the allograft side in preparation for reconstruction.

TABLE 29-3 Methods of Reconstructing the Abductor Mechanism

Method of Reconstruction	Material Used
Trochanteric slide	16-gauge wire
Classic trochanteric osteotomy	16-gauge wire
Soft tissue repair (graft to host abductor tendons)	Fiberwire suture
Soft tissue repair (host abductor tendons to prosthesis)	Fiberwire suture
Muscle repair directly to tensor fascia	Fiberwire
No repair	N/a

FEMORAL OSTEOTOMY

The osteotomy site is then identified by measuring the distance from a predetermined landmark on imaging, such as the most proximal tip of the greater trochanter. The rotational orientation is marked so that it will be easily recognizable once the proximal femur is removed. If the diagnosis is not malignant, the hip can be dislocated and the implants removed. Removal of implants is greatly facilitated by resection of the proximal femur; residual cement down the canal is removed by conventional techniques. Although a number of osteotomy configurations are possible for an APC (13,15), our preferred method is a transverse cut, which allows delicate rotational adjustments (Fig. 29-8). One critique of this type of osteotomy is that it is less stable; however, we have not had frequent problems with healing of the host-graft junction, and a transverse osteotomy avoids the technical

difficulty of matching rotation and contour on a step cut. We also have not needed to use cables or strut allograft or to clam shell any remaining proximal host bone to augment the construct.

Once the osteotomy has been completed, a bone clamp is placed on the end of the proximal femoral segment and used to lift up while detaching the remaining musculature from the linea aspera and posteromedial thigh. As much as possible, the entire hip capsule is carefully dissected off the proximal femur and carefully sutured around the neck of the prosthesis during wound closure. Retention of this structure is an important supplement for hip stability. Capsular retention and repair is all the more critical with an endoprosthesis, and its closure may be supplemented with alloderm, gortex graft, or mesh to aid in hip stability.

With the proximal femur now completely removed, the acetabulum is prepared and reamed. Most commonly, an uncemented press-fit acetabulum is preferred, with the exception of widely metastatic lesions with significant destruction of the supra-acetabular region. If much of the hip capsule is resected, consideration of a dual mobility, or constrained acetabular component, may be entertained due to the risk of dislocation (Fig. 29-9).

FIGURE 29-8 The allograft and fully porous-coated stem being inserted into host bone. A transverse cut is shown and is our preferred method of femoral osteotomy. This allows minor rotational adjustments at the time of stem insertion and avoids the technical difficulties of matching step-cuts on host and allograft bone.

FIGURE 29-9 Proximal femoral endoprosthesis with a metal-on-polyethylene articulation and a large femoral head (A) or a constrained liner (B) to mitigate the risk of dislocation.

ALLOGRAFT PROSTHETIC COMPOSITE RECONSTRUCTION

The allograft bone should be soaked in warm saline solution from the start of the case and can be prepared on the back table. It is extremely important to secure the graft to either an allograft vise/holder or the patient with a suture to prevent graft contamination if inadvertently dropped intraoperatively. Using the resected specimen as a guide and verifying with the template length, the allograft can be transversely osteotomized to slightly longer than its desired length to leave room for adjustment at the time of final implant placement. The femoral neck is osteotomized at the appropriate height approximately 1.5 cm from the lesser trochanter. The graft canal is prepared using straight reamers, with the intent of leaving a 2-mm cement mantle. The allograft should not be excessively reamed, so as to prevent significant weakening of the bone that may predispose it to fracture. A cylindrical, long stemmed, fully porous-coated implant with distal fixation is the optimal choice to achieve press-fit, uncemented fixation on the host side while cementing on the allograft side. The host bone should be underreamed using straight, rigid reamers by 0.5 mm to achieve good endosteal contact after removal of any bone cement from prior procedures. Four to six centimeters of scratch fit should be present distal to the osteotomy; if this is not possible, another method of reconstruction should be considered. If the resected proximal femur is not accurate in determining length, we insert the trial prosthesis in the acetabulum with the second from the shortest neck length, apply traction to the limb to equalize leg length or estimate appropriate soft tissue tension and then mark the length at the femoral osteotomy on the prosthesis. This is the distance the stem will protrude out of the graft after it is cemented into position. The trial prosthesis is then placed into the femur and inserted to the marked length. A bone clamp is firmly placed on the trial stem at the level of the mark, and a trial reduction is then performed, assessing limb length and overall stability. Fine-tuning of the osteotomy length can be made at this point in time.

Prior to cementing, if wires are to be used to fix the trochanteric fragment from the host, then these are now placed into the allograft. The allograft is then cleaned and dried and pressurized (with difficulty) with antibiotic-impregnated cement (Fig. 29-10). To avoid coating the porous-coated distal portion of the stem with cement, the

marked length that will go into the host bone is protected with a covering of Tegaderm prior to cementation into the allograft (Fig. 29-11). The implant is advanced into the graft bone. The cement is contained as best as possible, and additional cement is injected around the proximal body of the implant as it is being inserted (Fig. 29-12). Accurate anteversion is not important during this step. No cement should be left at the host-graft junction as this may interfere with bony union (13,15). After the cement has set, the construct is then inserted into the host bone. The fit should be tight, and the construct should advance 2 mm with each hammer blow. Correct version of the construct is now determined during the insertion. Final tuning of

the butt junction can be made at the start of stem insertion. As the stem is just starting to be inserted, carefully inspect the two soon-to-be-opposing surfaces to ensure that they are parallel; if an adjustment needs to be made at this point, it should be on the host side after the stem is extracted. The remainder of the implant is then impacted carefully into the host bone to achieve press-fit fixation (Fig. 29-13). Morselized autograft can be placed at the host-graft junction. The femoral head is assembled and the hip reduced into the acetabulum and ranged to ensure appropriate stability and length. The hip capsule is then tightly repaired over the femoral neck using fiberwire in a purse string technique (Fig. 29-14).

FIGURE 29-10 Preparation for allograft cementation. A glove is placed over the osteotomized end of the allograft and secured with a boneholding clamp while inserting and pressurizing cement. This aids in pressurization and prevents cement from interdigitating at the host-allograft interface.

FIGURE 29-11 Fully porous-coated, cylindrical stem for use in an APC reconstruction. Host portion covered with Tegaderm to prevent cement interfering with the porous surface.

FIGURE 29-12 Cementing technique. Bone cement has been pressurized into the graft prior to inserting the implant (A). A small slit is made in the surgical glove clamped over the femoral canal to allow the stem to pass

(B). Additional cement is added proximally as the stem draws out the cement from the allograft (C).

FIGURE 29-13 Intraoperative photographs showing a fully porous-coated stem cemented into the allograft. The composite is shown being advanced into the host bone to achieve press-fit fixation. The canal should be underreamed by 0.5 mm to achieve a good scratch fit, and there should be 4 to 6 cm of tightly-fitting prosthesis proximal to the metaphyseal-diaphyseal junction. Version of the composite is determined during the insertion and is based off the position of the tibial shaft. A and B: Abductor mechanism reconstruction can involve wire fixation of the host trochanter to allograft (A and B) or repair of the allograft soft tissues to the host abductor tendons (C).

FIGURE 29-14 Intraoperative photograph of the capsular repair. The capsule is shown repaired using a purse string technique over the femoral neck. This repair is key in preventing hip instability postoperatively.

Abductor Mechanism Repair

Where a trochanteric slide has been performed, a series of holes are drilled in the allograft around the trochanteric osteotomy site. Two drill holes are placed 1 cm medial to the plane of the osteotomy oriented

posterior to anterior and around 2 cm apart. Two drill holes are also placed running cephalad to caudal. Sixteen-gauge wire is advanced through these holes, the fragment is reduced, and the wires are sequentially tightened over the top of the trochanter (Fig. 29-15). A similar type of wiring fixation is performed for a classic trochanteric osteotomy.

If the host trochanter cannot be preserved, but the host abductor mechanism is left intact, the host abductor tendons are carefully repaired in layers to the allograft tendons. If the graft is properly prepared, the allograft tendons will need to be trimmed. If the abductor mechanism cannot be reconstructed, there is no advantage in using an allograft and an endoprosthetic reconstruction would be the better choice in many cases. See Figure 29-16 for postoperative radiographs of an APC.

9 P.390

We have had good success with this technique, with documented findings of soft tissue incorporation when revision has been performed for other reasons; for instance, conversion of a bipolar to a total hemiarthroplasty (Fig. 29-17). Nevertheless, soft tissue failures have occurred (Fig. 29-18) and can pose a real challenge.

FIGURE 29-15 Allograft prosthetic composite shown with the trochanter osteotomized (A). Reconstruction of the abductor mechanism will involve wire fixation of the host greater trochanter to the allograft. Wires are place prior to cementation of the prosthesis into the allograft (B). Postoperative range of motion following the use of this technique is shown (C).

FIGURE 29-16 Postoperative radiographs of APC reconstruction showing early graft healing (A), full graft incorporation (B and C), and a united trochanter/abductor mechanism reconstruction (D).

FIGURE 29-17 Young patient treated for Ewing sarcoma at age 11 with a proximal femoral resection and reconstruction using an APC (A). This patient required conversion of a bipolar hemiarthroplasty, seen at the time of retrieval (B) to a total hip arthroplasty. Intraoperative photograph demonstrates full healing of a soft tissue repair of the abductor mechanism (C).

FIGURE 29-18 Soft tissue failure of an APC abductor reconstruction. The trochanter was found bare of any soft tissue attachments at the time of revision.

ENDOPROSTHETIC RECONSTRUCTION

P.391

After the bone and any contiguous sarcomatous mass is removed and sent to pathology, a sample from the osteotomy site should be removed with a curette and sent for frozen section to document that no tumor cells remain. The instrumentation of the distal femur should not proceed until this is confirmed. In the setting of a confirmed metastatic lesion or nontumor diagnosis, this step can be omitted. Our preferred technique is either a cemented, stemmed endoprosthesis (Fig. 29-19) or compressive osseointegration technology (Fig. 29-20). The latter has been used increasingly over the last

P.392

5 to 7 years, especially in tumor resections leaving a short distal femoral segment and concerns with fixation beyond the metaphyseal diaphyseal junction. A press-fit option is also available, in most cases a fully porous-coated cylindrical cobalt-chromium stem (Fig. 29-21).

FIGURE 29-19 Cemented proximal femoral endoprosthesis with a short stem implanted in the setting of revision arthroplasty (A) and a long stemmed, cemented implant used for reconstruction after the resection of a proximal femoral sarcoma with considerable diaphyseal involvement (B).

FIGURE 29-20 Compressive osseointegration prostheses.

For a stemmed implant, a tenaculum is placed just distal to the osteotomy site. Reaming should progress until there is significant cortical chatter and endosteal contact. Flexible reamers should

be available, lest there is a tight isthmus or the proximal bony resection is quite long. Based on the resection length of the specimen, the modular intercalary segments and proximal body should be assembled along with the stem on the back table. A trial reduction of the hip should result in a stable construct with equal leg lengths and appropriate soft tissue tension. After appropriate trialing, antibiotic-impregnated cement is mixed while the femur is being prepared. The canal is irrigated and brushed. A cement restrictor is inserted to the appropriate length.

Once the cement has had adequate time to begin curing, the implant is inserted and pressurized into the canal. The correct version of the implant is determined by the position of the tibial shaft. Whenever possible, we coat the body of the endoprosthesis with antibiotic cement prior to wound closure to reduce the incidence of infection. The capsule is closed in a similar fashion to the APC technique (Fig. 29-14).

FIGURE 29-21 Plain radiograph and photograph of cementless proximal femoral endoprostheses (A and B) and a fully porous-coated stem (C).

If a compressive osseointegration implant is selected, a double reamer is inserted into the distal femoral bone. The size will depend on if a short or long traction bar has been selected. The bone is then reamed up to the appropriate size, and the anchor plug is trialed by hand to verify that it fits snugly. The anchor plug is removed and the external guide is assembled, taking care to ensure that all drills for the distal fixation pins pass through the anchor plug. The final anchor plug is then inserted into the host bone with the external guide attached. The distal pinholes are drilled, and the pins tapped into place sequentially. The desired amount of compressive force is then selected; typically, 800 pounds in the femur, which occurs through the compression of Belleville washers. The spindle and adaptor are inserted over the traction bar, and a nut driver is used to tighten the nut until a proximal washer spins freely, indicating that the construct has been tightened to the appropriate level. This portion of the implant comprises the compression mechanism, which aims to achieve biologic fixation of the implant to host bone according to Wolff's law. Bony hypertrophy is often seen at the bone-implant interface (Fig. 29-22).

Correct version is determined at this point. The modular segments are assembled on the back table, inserted over the taper junction, and trialed. If the hip is stable through a range of motion and

the leg lengths are appropriate, the final modular implants are assembled, joined to the compress fixation device, and the hip is reduced. Where endoprosthetic reconstruction has been selected as the appropriate treatment, we favor compressive osseointegration in all young active patients, patients with insufficient remaining bone to adequately cement an implant (Figs. 29-23 and 29-24), and patients undergoing second-stage endoprosthetic reconstruction for infection. In the latter case, it bears mentioning that removal of the implant is greatly simplified if the infection recurs.

FIGURE 29-22 Left panel: Photograph depicting the compression device, with a porous-coating spindle, traction bar, and anchor plug with holes for the distal fixation pins. Plain radiograph of a proximal femoral compression device (right panel) is also shown at the bone-implant interface, demonstrating ingrowth and hypertrophy of the bone in response to compressive forces. This hypertrophy is also termed the “elephant's foot.”

FIGURE 29-23 A 48-year-old male patient treated elsewhere for a chondrosarcoma of the proximal femur with extraarticular resection and allograft arthrodesis. He presented to our institution following multiple failed attempts at plate osteosynthesis for nonunion. He was treated with resection of the proximal femur and reconstruction using a proximal femoral compression device.

P.395

FIGURE 29-24 A 66-year-old female with a history of Gaucher disease and a multiply operated right hip. Extensive bone loss required a compressive osseointegration prosthesis and placement of the hip in a false acetabulum.

Abductor Mechanism Repair

If a portion of greater trochanter is preserved, it can be repaired by using existing holes in the implant. The intention is not to achieve biologic fixation at the host trochanter-implant junction, rather that the abductor mechanism is held down to the implant while scar tissue forms and ultimately secures the abductors. In our experience, implants whose purpose is to tack down the abductors, such as claw plates, cerclage wires, and tabs, have not proved to be especially useful in preventing migration of the trochanter. Small bone tunnels are created in the fragment, and fiberwire is passed in three separate sections through the implant holes and tied over the trochanteric fragment. If bone was not preserved, then the tendons of the gluteus medius and minimus can be directly sutured to the implant in a similar fashion. Unfortunately, failure of the repair is common and places the patient at risk of instability and gait abnormalities (Fig. 29-18). In either case, the reconstruction should be augmented by suturing the fascia lata to the repair in this region.

POSTOPERATIVE MANAGEMENT

After both an APC and a segmental arthroplasty, the patient is placed in a hip abduction brace immediately postoperatively. Strict hip precautions are maintained to minimize the risk of early postoperative instability. Full weight bearing is allowed in individuals who have been managed with an EPR. In those treated with an APC, toe touch weight bearing is recommended for 6 weeks, then full weight bearing is allowed provided a crutch or cane is used in the opposite hand. Graft host healing is variable but usually is evident in 3 to 6 months. The required rehabilitation and expected outcomes should be discussed in detail with the patient and his or her family prior to the surgery. Surgical drains are left in place postoperatively and generally removed within 24 to 72 hours following surgery. Antibiotics are continued prophylactically for 5 days and include cephazolin and vancomycin. Clindamycin is administered if the patient has a documented allergy to cephazolin.

CONCLUSION

Allograft composite and endoprosthetic reconstructions are complex procedures that place the patient at a significant risk of morbidity. However, they also have the capacity to produce excellent results when procedures are carried out under optimal conditions with appropriately selected

P.396

patients. APCs have become less prevalent over the years but remain an excellent option for young individuals in whom implant durability and function are the main objectives. The procedure can be

technically demanding; however, the advantages are preservation of bone stock, restoration of function through a more robust reconstruction of the abductor mechanism, and a relative ease of revision due to the lack of distal canal violation (13). These patients should be monitored over time due to the risk of bone resorption. Endoprosthetic reconstruction can involve cemented and uncemented stemmed implants, as well as compression osseointegration technology. Implant survival has improved with modern design technology, and as a result, this reconstructive mode has become a mainstream technique for managing massive bone loss in the proximal femur. A major advantage is that the construct allows for immediate postoperative weight bearing, which is essential for patients with a bony malignancy and those on chemotherapy. Hip instability resulting from limitations in reconstructing the abductor mechanism represents an ongoing challenge of this technique.

REFERENCES

Farid Y, Lin PP, Lewis VO, et al.: Endoprosthetic and allograft-prosthetic composite reconstruction of the proximal femur for bone neoplasms. Clin Orthop Relat Res 442: 223-229, 2006.
Bickels J, Meller I, Henshaw RM, et al.: Reconstruction of hip stability after proximal and total femur resections. Clin Orthop Relat Res (375): 218-230, 2000.
Dobbs HS, Scales JT, Wilson JN, et al.: Endoprosthetic replacement of the proximal femur and acetabulum. A survival analysis. J Bone Joint Surg Br 63-B(2): 219-224, 1981
Donati D, Giacomini S, Gozzi E, et al.: Proximal femur reconstruction by an allograft prosthesis composite.

Clin Orthop Relat Res (394): 192-200, 2002.
Fox EJ, Hau MA, Gebhardt MC, et al.: Long-term followup of proximal femoral allografts. Clin Orthop Relat Res (397): 106-113, 2002.
Hejna MJ, Gitelis S: Allograft prosthetic composite replacement for bone tumors. Semin Surg Oncol 13(1): 18-24, 1997.
Jofe MH, Gebhardt MC, Tomford WW, et al.: Reconstruction for defects of the proximal part of the femur using allograft arthroplasty. J Bone Joint Surg Am 70(4): 507-516, 1988.
Kabukcuoglu Y, Grimer RJ, Tillman RM, et al.: Endoprosthetic replacement for primary malignant tumors of the proximal femur. Clin Orthop Relat Res (358): 8-14, 1999.
Mittermayer F, Krepler P, Dominkus M, et al.: Long-term followup of uncemented tumor endoprostheses for the lower extremity. Clin Orthop Relat Res (388): 167-177, 2001.
Unwin PS, Cannon SR, Grimer RJ, et al.: Aseptic loosening in cemented custom-made prosthetic replacements for bone tumours of the lower limb. J Bone Joint Surg Br 78(1): 5-13, 1996.
Ward WG, Johnston KS, Dorey FJ, et al.: Loosening of massive proximal femoral cemented endoprostheses. Radiographic evidence of loosening mechanism. J Arthroplasty 12(7): 741-750, 1997.
Zehr RJ, Enneking WF, Scarborough MT: Allograft-prosthesis composite versus megaprosthesis in proximal femoral reconstruction. Clin Orthop Relat Res (322): 207-223, 1996.
Gross A: Revision stems: allograft prosthetic composite. In: Barrack R, Rosenberg A, eds. Masters techniques in orthopaedic surgery: the hip. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2005: 385.
Johnson ME, Mankin HJ: Reconstructions after resections of tumors involving the proximal femur. Orthop Clin North Am 22(1): 87-103, 1991.
Babis G, Sakellariou V, Sim F: Allograft prosthetic composite replacement for bone tumours of the proximal femur. In: Sim F, Choong P, Weber KL, eds. Masters techniques in orthopaedic surgery: orthopaedic oncology & complex reconstruction, Philadelphia, PA: Lippincott Williams & Wilkins, 2011: 137.
Farid Y, Lin PP, Lewis VO, et al.: Endoprosthetic and allograft-prosthetic composite reconstruction of the proximal femur for bone neoplasms. Clin Orthop Relat Res 442: 223-229, 2006.
Poitout DG, Lempidakis M, Loncle X, et al.: Massive reconstruction of the acetabulum and proximal femur. Chirurgie 120(5): 254-263, 1994-1995.
Deirmengian C, Kardos K, Kilmartin P, et al.: Combined measurement of synovial fluid alpha-defensin and C-reactive protein levels: highly accurate for diagnosing periprosthetic joint infection. J Bone Joint Surg Am 96(17): 1439-1445, 2014.
Deirmengian C, Kardos K, Kilmartin P, et al.: The alpha-defensin test for periprosthetic joint infection outperforms the leukocyte esterase test strip. Clin Orthop Relat Res 473(1): 198-203, 2015.
Deirmengian C, Kardos K, Kilmartin P, et al.: Diagnosing periprosthetic joint infection: has the era of the biomarker arrived? Clin Orthop Relat Res 472(11): 3254-3262, 2014.
Friedlaender GE, Strong DM, Sell KW: Studies on the antigenicity of bone. II Donor-specific anti-HLA antibodies in human recipients of freeze-dried allografts. J Bone Joint Surg Am 66(1): 107-112, 1984.
Friedlaender GE, Strong DM, Sell KW: Studies on the antigenicity of bone. I Freeze-dried and deep-frozen bone allografts in rabbits. J Bone Joint Surg Am 58(6): 854-858, 1976.
Czitrom A: Biology of bone grafting and principles of bone banking. In: Weinstein S, ed. The pediatric spine: principles and practice. New York, NY: Raven Press, 1994: 1285-1298.
Blackley HR, Davis AM, Hutchison CR, et al.: Proximal femoral allografts for reconstruction of bone stock in revision arthroplasty of the hip. A nine to fifteen-year follow-up J Bone Joint Surg Am 83-A(3): 346-354, 2001.