A 56-year-old man with no significant medical history had a left medial unicondylar arthroplasty done at an outside institution for isolated medial knee osteoarthritis. He presented to our institution with several months of chronic knee pain and feelings of instability. He had no new trauma and no fevers or other constitutional symptoms suggestive of infection. On physical examination, he had a well-healed midline incision with a large effusion but no surrounding erythema. His active range of motion was 15 to 90 degrees with 10 mm varus and valgus laxity. The erythrocyte sedimentation rate (ESR) and C-reactive protein level (CRP) were within normal limits; aspiration of the knee revealed only 137 white blood cells with no bacteria. Standing radiographs ( Fig. 21.1 , A and B ) showed an unstable medial unicondylar arthroplasty with severe subsidence of the tibial tray and a large bony defect in the medial plateau. The overall alignment of the knee was valgus, which suggested a lateral plateau bony deficiency as well. Intraoperatively, the femoral and tibial components were both found to be grossly loose. On the tibial side, there was a large contained defect on the medial plateau and an uncontained defect on the lateral plateau. After all bony cuts were made and components were trialed, the defects were addressed. Because the medial defect was contained, morselized bone graft was packed into the defect. However, a wedge augment was used on the lateral side because the bony defect was uncontained there. A constrained condylar knee implant was used because of the lack of stability. The soft tissue envelope closed well, and the patient was made weight bearing as tolerated, with therapy focusing on increasing the range of motion. Postoperative radiographs are shown in Figure 21.1 , C and D .
This chapter describes an overview and surgical technique of using allograft for the management of bone defects in revision total knee replacement. A classification system is described and may be used to help guide treatment. Surgical complications, outcomes, and postoperative care are also discussed.
Preoperative planning is critical. The more a surgeon knows preoperatively about potential bone defects, the more prepared he or she can be for the revision. It is important to know that there is not one right way of performing a revision, and the use of structural bone graft is just one tool of several available to the arthroplasty surgeon.
Preoperative radiography and computed tomography should be used to help determine the extent of bony defects.
Care should be taken in removing the implant to keep as much native bone stock as possible.
If screws or Steinmann pins are used, they should be secured to native bone so that the hardware is buried within the allograft and will not touch the final implanted prosthesis.
Cement should be used to secure a stemmed prosthesis to a bulk allograft, because the allograft cannot be relied on for bony ingrowth.
Revision knee arthroplasty in the setting of an active or latent infection with the use of allograft will result in failure.
Use of allograft or implants that are too large may result in problems with wound healing.
Use of allografts without addressing incompetent collateral ligaments will result in failure.
Revision total knee arthroplasty (TKA) poses a challenge for arthroplasty surgeons for many reasons. One of the factors that make this operation difficult is the loss of native bone on both the femoral and the tibial side. Bone loss has many causes, including implant macromotion, infection, osteolysis, periprosthetic fracture, stress shielding, and iatrogenic bone loss that occurs during the explantation of components in revision cases ( Fig. 21.2 ).
With all failed TKAs, it is imperative to rule out infection as the cause of the failure, because the presence of infection may necessitate at least a two-staged operation to reconstruct the joint. Once infection has been ruled out through various laboratory studies and serologies—including erythrocyte sedimentation rate, C-reactive protein levels, and complete blood count with differential (and possibly aspiration)—plans for reconstruction can begin.
Preoperatively, the surgeon can get a gross sense of the presence of bone loss with plain radiographs. However, the best way to evaluate bone loss preoperatively is with computed tomography (CT). The CT scan will help show the extent of the bone loss and whether the defects are contained. This is important for preoperative planning in deciding whether to use metal augments, morselized bone graft, structural allograft, or a tumor prosthesis. More bone loss can be anticipated during the actual surgery, because the explantation of the prosthesis inevitably results in some bone loss, which hopefully will be minimal with proper technique.
Although it is beneficial to evaluate bone loss preoperatively to appropriately plan reconstruction, the ultimate evaluation occurs intraoperatively. The staging system used is the Anderson Orthopaedic Research Institute (AORI) classification for bone defects ( Table 21.1 ).
Indication and Contraindications
Once it has been decided that a patient needs a revision TKA and the prosthesis is not infected, it is important to assess the bone stock available. This may be done with the use of plain radiographs as well as a CT scan. These studies will help elucidate any bony defects and help the surgeon order the proper equipment for the revision. An absolute contraindication for the use of bone graft in a revision TKA is an infection, either latent or active.
There are several indications for the use of bone graft in the revision setting. In general, small contained defects (those contained within a rim of cortical bone), usually less than 5 mm deep, can be treated with morselized allograft on either the femoral or the tibial side. They can also be treated with cement and screws should the surgeon choose to do so. Larger bone defects, on both the femoral and the tibial side, can be treated with femoral head allograft secured with Steinmann pins. Another option for large contained defects is metal allografts and wedges, which are not discussed in this chapter. Large uncontained defects can be treated with an allograft–prosthesis composite or a tumor prosthesis.
Proper preoperative planning helps the surgeon determine what equipment will be needed for revision TKA. As with any revision surgery, the first objective is to obtain appropriate exposure. Next, the surgeon must remove components while retaining as much native bone as possible. This may be accomplished with the use of sagittal saws, thin osteotomes, and Gigli saws. Cement may be removed with a high-speed bur, which can also be used to remove sclerotic bone. Cancellous bone chips are needed for small contained defects, whereas cadaveric femoral head will be needed for large defects. Both male and female acetabular reamers are needed to properly shape the native bone and the femoral head to fit the defect. Steinmann pins or cancellous screws are needed to secure the allograft. If a large uncontained defect is encountered, the surgeon may decide to re-create the anatomy with a tibial/femoral prosthetic composite, in which case a properly sized distal femur or proximal tibia bulk allograft will be needed. Another option for a large uncontained defect on the tibial side is to re-create the deficient cortical bone with mesh and then fill the deficient cancellous bone with morselized cancellous allograft. After this is performed, the final implant may be cemented into place.
With any of these procedures, the surgeon should always have a secondary plan should problems with the allograft be encountered. For this reason, metal augments, cones, and wedges should be available in the room. If the bony defect is so large that the collateral ligaments are compromised, it may be best to forego any allograft reconstruction and revise to a hinged prosthesis.