Biologic Treatment Options for the Hip: A Narrative Review

H. Thomas Temple, MD


Curr Orthop Pract. 2019;30(6):501-509. 

In This Article

Proximal Femoral Allografts

The goal of reconstruction of the proximal femur is to restore bone stock and reproduce abductor function. A classification system describing complications for metal implants was developed by Henderson et al.[20] The principle modes of failure of proximal femoral endoprostheses are: soft tissue and aseptic loosening. These failures are mitigated in part with biologic reconstructions due to the presence of soft tissues on the grafts and the biologic potential allowing allograft host-junction healing. Roque et al.[21] retrospectively reviewed 150 patients from a single institution who underwent resection and reconstruction of the proximal femur. The types of reconstruction were: osteoarticular allograft, allograft-prosthesis, intercalary, and allograft arthrodesis. The overall success rate was 77%, with the best results in the allograft-prosthesis (82%) and intercalary (87%) procedures. In patients with graft failures, the most common causes were infection (15 patients) and fracture (26 patients).[21]

Osteoarticular allografts have a limited role in hip reconstruction after tumor resection. Jofe et al.[22] reviewed 44 patients who had an allograft prosthesis or allograft alone to replace the proximal femur. Major complications were principally infection (five) and fracture (six). They concluded that although the groups were not quite comparable, the patients in the allograft prosthesis group had better outcomes. The principal modes of failure of osteoarticular grafts are fracture and arthrosis over time. The later complication is not insignificant. Verbeek et al.[23] reviewed complications in patients with total knee or hip arthroplasty in retained allografts. They reported complications in 61% of patients, the most common being structural (fracture). Of those patients who underwent removal of the allograft and endoprosthetic reconstruction, complications were observed in 81% of patients, most commonly aseptic loosening and infection. Patients with a primary allograft prosthetic composite also had a very high complication rate at revision (85%).[23]

Allograft prosthetic composites (APC) theoretically have the advantage of preserving bone stock, providing soft-tissue attachments for abductor tendons and avoiding degenerative changes in the hip that invariably occur in patients with osteoarticular allografts. The biomechanical advantages of abductor insertion results in improved gait compared to patients with proximal femoral endoprostheses.[24] Although restoring bone stock is a significant advantage of APC reconstruction of the proximal femur, preserving bone stock in the face of revision of this construct is often not possible. In a study of 203 proximal femoral allograft prosthetic composites studied by Wilke et al.,[25] preservation of bone stock using this mode of reconstruction was called into question. They reported 27 revisions due to aseptic loosening or fracture of the allograft. At the time of revision, only seven allografts were either wholly (three) or partially (four) retained. Most patients required removal of the graft and replacement with an endoprosthesis.[25] Despite the cited shortfalls, APC reconstruction is very useful in replacement of large segments of the proximal femur and maintaining reasonable abductor function and hip stability. This is illustrated in this 58-year-old woman who underwent total hip arthroplasty 5 yr prior to presenting with right hip and groin pain. She was found to have permeative bone destruction in the medial proximal femoral cortex and extraosseous new bone formation seen on anteroposterior and lateral radiographs (Figure 4A and B). MRI (sagittal STIR) revealed a soft-tissue mass involving the proximal femur and acetabulum. The mass was hyperintense on STIR images and had a lobular appearance (Figure 4C and D). A biopsy was done and interpreted as an intermediate-grade chondrosarcoma. This patient had an extraarticular resection of the pelvis (type II) and the proximal femur. The defect was reconstructed using a pelvic-femoral allograft and a cemented long-stem bipolar prosthesis (Figure 4E and F). The allograft healed well, and the patient became an independent ambulator using a cane.

Figure 4.

A and B, Anteroposterior and lateral pelvic radiographs of the right hip demonstrating a total hip arthroplasty that is stable and in good position with an area of permeative bone destruction over the lesser trochanter with soft-tissue mineralization (arrows). The matrix has an arcs and rings morphology. C and D, Coronal and axial STIR images reveal a hyperintense signal abnormality with a lobular growth pattern in the soft tissue around the hip with involvement of the acetabulum. A subsequent biopsy was interpreted as an intermediate grade chondrosarcoma. E and F, An anteroposterior radiograph shows a pelvic allograft that is well fixed with a biopolar hemiarthroplasty and a long-stem cemented implant allograft femur composite. The allograft host junction is indicated by the dark arrow.

Proximal graft resorption is also a prominent sequelae of structural allograft replacement of the proximal femur. This is most likely due to revascularization through the attachments of the abductor muscles. A similar phenomenon and high incidence of fracture are observed in the proximal humerus through the soft-tissue attachments, principally, the rotator cuff tendons. Bone resorption can lead to aseptic loosening of the implant or fracture of the allograft. Graft resorption is seen in a 23-yr-old woman with multicentric giant cell tumor who underwent resection of the proximal femur and reconstruction using an allograft prosthetic composite. She developed significant graft resorption in 18 mo (Figure 5A) and hip pain requiring revision to a total hip arthroplasty and allograft strut grafting (Figure 5B). Her pain abated, and she currently ambulates independently without assistive devices.

Figure 5.

A and B, Anteroposterior radiographs represent a preoperative radiograph on the left with a composite allograft-bipolar implant. There is extensive resorption of the proximal lateral femur and greater trochanter. On the right is a postoperative radiograph following conversion of the bipolar composite implant to a total hip arthroplasty with augmentation of the proximal femur with strut allografts.

Intercalary allograft reconstruction of the proximal femur is preferred to proximal femoral replacement if adequate proximal femur can be preserved to achieve stable fixation. Alternative biologic options to satisfy large intercalary defects are: vascularized fibulae, autogenous fibulae, combined allograft, and vascularized fibula and segmental transport using osteogenesis. The focus in this section is principally on allograft intercalary reconstruction.

Ortiz-Cruz et al.[26] retrospectively evaluated 104 intercalary allografts in 100 patients; limb salvage was 92%. Nonunion was observed in 31 grafts at one junction or both within 1-year after surgery. Infection, fracture, tumor stage of the lesion and the use of adjuvant chemotherapy and or radiotherapy had adverse effects on graft survival.[26] The following example demonstrates the benefits of intercalary allograft reconstruction in the proximal femur. This case illustrates a 34-year-old active male who developed pain in the left proximal thigh and hip region. A radiograph demonstrated a radiolucent abnormality with thickened cortices and subtle internal matrix that had the appearance of arcs and rings (Figure 6A). There was a hyperintense signal abnormality in the medullary canal with endosteal scalloping but no cortical destruction or soft tissue mass (Figure 6B). Biopsy revealed an intermediate chondrosarcoma. This patient underwent wide local resection and intercalary allograft reconstruction with screw and plate fixation. Both osteosynthesis sites healed without incident and the patient resumed normal function that included high impact activity and running. Over time was evident that the graft underwent revascularization and remodeling (Figure 6C).

Figure 6.

A and B, A radiolucent abnormality was found in the proximal femur with thickening of the cortices and endosteal scalloping. Subtle matrix formation has the appearance of arcs and rings. The corresponding T2-weighted MRI on the right shows a hyperintense signal with heterogeneity, with no obvious soft-tissue mass. A biopsy showed an intermediate chondrosarcoma. C and D, Anteroposterior and lateral radiographs of the right proximal femur revealed a healed intercalary allograft fixed with a screw and side-plate implant. There is obvious remodeling of the graft that can be observed on both anteroposterior and lateral radiographs.

More often, allograft-host junction healing is slow, especially for patients on chemotherapy after reconstruction; as a consequence, nonunions are common. When nonunions supervene, bone grafting at the nonunion site is necessary. Historically, iliac crest bone grafting was used with moderate success. To avoid donor site morbidity and in observance of equal efficacy using 100–300 μm cortical bone, a cellular allograft (micronized bone combined with bone marrow-derived osteoprogenitor cells) has been applied to host-allograft junction nonunions and has been observed to result in rapid and complete healing. The anteroposterior radiograph and MRI in figure (Figure 7A and B) demonstrates a low-grade chondrosarcoma in a 36-year-old woman. She underwent resection of a 13-cm segment of bone (Figure 7C) and reconstruction using a freeze-dried intercalary femoral allograft (Figure 7D). She developed a nonunion at the proximal allograft-host junction (Figure 7E) that underwent bone grafting using a cellular allograft (ViaGraft™, Vivex Biologics, Inc., Atlanta, GA) (Figure 7F) This resulted in rapid healing in short time. At 6mo after surgery (Figure 7G), there was complete healing at the host allograft junction and at 2 yr, remodeling was evident (Figure 7H).

Figure 7.

Anteroposterior radiograph (A), coronal STIR MRI (B), and resected specimen (C) demonstrate a 13-cm chondrosarcoma in the mid femoral diaphysis. D, Postoperative anteroposterior radiographs reveal the proximal and distal allograft host junctions. E and F, Nonunion at the proximal allograft host junction demonstrated radiographically and clinically at the time of grafting with a cellular allograft. G and H, Anteroposterior radiographs at 6 mo (left) and at 2 yr (right) show early healing and consolidation of the proximal allograft host-junction, respectively.

Other considerations for persistent nonunion at the allograft-host junction(s) is (are) the combined use of a structural allograft with a vascularized fibula.[27,28] To mitigate the inconsistent healing at the allograft-host junctions and to reduce fracture risk, controlled temporal administration of embedded bioactive molecules within an allograft has been proposed. Sharmin et al.[29] demonstrated significantly increased bone formation in a rat model coating and loading with VEGF and BMP-2 in an allograft that was placed into a critical defect. Although the regulatory hurdles are considerable, and the leap from animal models, especially rodents, to clinical use in humans is formidable, physiologic augmentation of allografts or synthetic materials with growth factors is compelling.