Total knee arthroplasty is one of the most successful orthopaedic procedures. It improves quality of life and has high patient satisfaction and excellent longevity, with survivorship of >90% at fifteen to twenty years. However, mechanical failure remains a complication. Debris particles, especially from polyethylene, affect the long-term durability of the implants1-3. Polyethylene debris incites a chemical and cellular inflammatory reaction, resulting in bone resorption and osteolysis. The extent of the osteolysis is dependent on the volume, size, and shape of the polyethylene debris. The main causative factor leading to osteolysis is small particulate debris, which stimulates an inflammatory foreign-body cellular response, resulting in bone resorption4. In contrast, the large polyethylene particles associated with delamination of the polyethylene and fatigue wear do not elicit the same cellular response2,3.
Wear is a motion-driven abrasive or adhesive process, while delamination is a stress-driven mechanical process. In total knee arthroplasty, polyethylene wear occurs from a combination of rolling, sliding, and rotational motions on the bearing surfaces. Polyethylene damage in total knee arthroplasty occurs on the articular surface as abrasive wear due to the sliding motion of the femoral component on the tibia. Pitting, delamination, and flaking are due to the cyclic loading and oxidation of the polyethylene. Tibial post wear in posterior stabilized designs can be seen as delamination, deformation, or fracture. Tibial backside wear occurs as abrasive wear, pitting, flaking, or delamination. The magnitude and degree of polyethylene wear is multifactorial and depends on the material properties of the implant along with the alignment and stability of the prosthesis.
There are numerous causes of osteolysis, including the component design, quality of the polyethylene, manufacturing process, and sterilization techniques2,3,5-9. The impact of the component design depends on the conformity of the femoral component on the tibial polyethylene articular surface and the resultant contact stress. Articular surface designs with small contact areas between the femoral and tibial components distribute forces over a limited area and increase the stress on the tibial polyethylene. Some posterior cruciate-retaining prostheses have this design feature, resulting in increased contact stress on the polyethylene and predisposing it to wear and osteolysis, especially if there is condylar lift-off and edge-loading of the femoral component on the flat tibial polyethylene surface. Furthermore, abnormal and erratic kinematics, such as paradoxical anterior femoral translation with flexion in some cruciate-retaining designs, potentially increases polyethylene wear10. The round-on-round designs are more conforming and lead to lower contact stresses and less polyethylene wear. These partially conforming designs improve anteroposterior stability and kinematics with a lower rate of polyethylene wear11. Retrieval studies have demonstrated that these more conforming designs have lower wear rates and fewer eccentric wear patterns than the less conforming designs5,12-14.
The introduction of modular tibial components resulted in new modes of failure of total knee replacement. Modular tibial baseplates with screw holes have been implicated as sources of polyethylene backside damage, and they create a portal for polyethylene debris to enter the tibia. Some of the earliest reports on osteolysis implicated cementless designs with holes in the baseplate, tibial fixation screws, or discontinuous porous surfaces that served as conduits for debris, ultimately leading to osteolysis2,3. Later studies found that osteolysis also occurred in cemented designs with the invasion of the polyethylene debris through voids in the cement mantle7. Another design feature, introduced with the advent of tibial modularity, is the tibial polyethylene locking mechanism. Depending on the security of the locking mechanism, micromotion occurs between the tibial polyethylene and the baseplate, leading to backside wear of the tibial polyethylene component. Numerous studies have shown that micromotion between the tibial baseplate and polyethylene component occurs to a varying degree with most implant designs, and this motion progresses over time. The surface finish of the tibial baseplate also impacts the extent of backside wear. Manufacturers have attempted to reduce backside wear by improving the locking mechanism, improving mobile-bearing designs, and reintroducing monoblock tibial components.
Regarding the quality and manufacturing of the polyethylene, these factors vary among manufacturers15. The tibial polyethylene component is usually manufactured by either a net-shape molded technique or is machined from bar stock16. Machining of the polyethylene may predispose the polyethylene to surface and subsurface irregularities, which can lead to delamination and failure17. Compression-molded polyethylene has exhibited less wear than components machined from extruded bar stock. Finally, the method of sterilization impacts the mechanical properties and the durability of the tibial polyethylene component. Polyethylene sterilized by gamma irradiation in air produces free radicals with oxidation of the polyethylene, which leads to premature mechanical failure, generation of polyethylene debris, and ultimately osteolysis. Current implants are sterilized in either ethylene oxide gas or gamma irradiation in an inert oxygen-free environment. The recent introduction of highly cross-linked polyethylene created a stronger three-dimensional structure of the polyethylene, resulting in improved wear-resistant polyethylene with less oxidation and a diminution of residual free radicals. Additional developments include the introduction of oxygen-quenching additives such as vitamin E (alpha-tocopherol) to further stabilize the polyethylene, improve the mechanical properties, and improve wear resistance18.
Osteolysis is a progressive condition that should be diagnosed early and treatment initiated prior to a catastrophic failure, such as periprosthetic fracture or substantial bone loss that renders the revision arthroplasty more complicated.
Once an osteolytic lesion is identified, the goal is to intervene prior to component failure or periprosthetic fracture due to the bone loss. Management should address the reason for the osteolysis, restore the region of bone loss, and restore the appropriate component position and limb alignment with a stable implant.
Small areas of limited osteolysis are usually of little clinical importance, but these patients should be followed on an annual basis to determine whether the lesion is progressive or impacting the fixation of the prosthesis. Lesions that are asymptomatic, small, and nonprogressive usually require no treatment. Pharmacologic treatments, such as bisphosphonates and nonsteroidal anti-inflammatory medications, have been used to minimize the osteolytic process, but there are insufficient clinical data to support this course of treatment23.
Isolated osteolytic tibial lesions with well-fixed modular components, either cemented or cementless, with no malalignment, no malrotation, and no instability may be treated with isolated tibial polyethylene exchange and impaction bone-grafting of the lesion (Figs. 3-A and 3-B)24,25. These lesions are usually incidental radiographic findings in asymptomatic patients, but are concerning because of the size of the osteolytic lesion and the potential for later mechanical failure. It is important to determine the extent of the lesion, the integrity of the cortex, and the stability of the component fixation. Even when the above criteria are met, several other factors that must be considered include ensuring that the tibial locking mechanism is intact and functional, the new tibial polyethylene is of good quality and has not been on the shelf too long (an issue with older designs), and that there is no damage to the femoral surface or the tibial baseplate. If the femoral component surface is damaged, it should be exchanged because the irregular femoral surface will damage the new polyethylene tibial component.
Symptomatic patients with expanding osteolytic lesions or cortical disruption, loose components, instability, or periprosthetic fracture should have revision of the total knee replacement. This may involve either single-component revision or total revision. The amount of bone loss should be carefully assessed in planning the revision total knee arthroplasty.
Bone loss is common in revision total knee arthroplasty for osteolysis and can be compounded by subsidence of loose components, pathologic fracture, or removal of well-fixed components. The options available to treat the bone defects include the use of polymethylmethacrylate, autogenous bone graft, morselized allogenic bone graft, structural bone allograft, modular augments, and megaprostheses26-37. The choice of augmentation is dependent on the degree of bone loss, patient age, and surgeon experience. Additional considerations include the need for implant stem extensions, use of constrained prostheses, and correction of anatomic deformity38.
Preoperative assessment with radiographs is inadequate, but a CT scan will demonstrate the amount of bone loss. In a retrospective analysis of thirty-one patients with symptomatic total knee replacements who had osteolytic lesions, radiographs detected only 17% of the osteolytic lesions, whereas CT scans revealed lesions in all cases21. However, the intraoperative findings are often different from those predicted by imaging39. A useful classification system for bone loss is the Anderson Orthopedic Research Institute (AORI) classification system, which grades the bone loss from the femur and tibia independently as 1, 2, or 38,10. The severity of bone loss and the proximity to the femoral epicondyles and the tibial tubercle determine the grade. Bone loss from the femoral condyles or tibial plateau is further subdivided into A or B, depending on whether one (A) or two sides (B) are involved. This classification system provides a useful guide for dealing with the degree of augmentation needed during revision arthroplasty32,40.
If the bone loss is minimal (AORI type 1) and the cortical rim is intact, the small cavitary defects can be augmented with bone cement or cancellous bone chips. In young patients, for whom bone preservation and restoration is important, impaction bone-grafting of these bone defects has been valuable. For patients with greater bone loss (AORI type 2), structural allografts can provide a viable biologic option for restoration of the proximal part of the tibia or the distal part of the femur (Figs. 4-A and 4-B). Another easy option for AORI type-2 defects that is readily available in most revision systems is modular augmentation. These modular augments along with stem extensions—either press-fit or cemented—can be added to the femoral and tibial components and are useful in restoring the femoral and tibial components to their appropriate position. Modular augmentation helps to restore moderate-sized bone defects with biomechanically stable components to allow restoration of the proximal and distal femoral anatomy, assist in the creation of equal flexion and extension gaps during the revision procedure by restoring the joint line, and allow full weight-bearing with functional motion. Severe bone loss (AORI type 3) results in a difficult revision arthroplasty and usually requires bulk structural allografts, tibial or femoral metaphyseal cones, or megaprostheses.
Modular revision implants with augmentation blocks or wedges facilitate the treatment of bone defects. These modular augments attach to the femoral and tibial components, are available in various shapes and sizes, and are fixed to the components with screws or cement. These augments are adaptable, allow intraoperative customization of the components, provide excellent biomechanical properties, and require minimal bone resection as the augments build off the residual bone. Femoral defects can be reconstructed, in most revision arthroplasty systems, with metal augments in increments of 5 mm. Since most bone loss is either from the distal end of the femur, posterior condyles, or both, augments are fixed to the femoral component in an effort to restore the femoral anatomy, posterior condylar offset, and distal femoral joint line. Augmentation of the posterolateral part of the femur also assists with the rotational position of the femoral component. The tibial modular augments are either wedges or blocks and assist in dealing with the bone defects and positioning the tibial component perpendicular to the mechanical axis. Modular wedges are useful when a loose tibial component subsides into a varus position with a resulting angular deformity, while tibial blocks are useful for revising a failed unicondylar knee replacement or a tibial defect that is more symmetrical. With the use of modular augments in revision total knee arthroplasty, good or excellent results have been reported to range from 84% to 98%36,37.
The more challenging revision arthroplasties are those with substantial bone loss (AORI type 2B or 3). Reconstruction choices in these cases are modular augmentation, structural allograft, or megaprostheses. Trabecular metal cones and metaphyseal sleeves for the proximal part of the tibia and distal end of the femur provide a modular alternative for metaphyseal augmentation in patients with severe bone loss and provide a platform for the final components. Trabecular metal cones used in the proximal part of the tibia, implanted in a press-fit cementless manner, recreate the cortical rim and provide a stable platform for the final component. The variety of available shapes addresses both cancellous bone loss and cortical defects (Figs. 5-A, 5-B, and 5-C). The distal femoral cones help to reestablish the metaphyseal-diaphyseal junction and create a stable base for the femoral component. These modular constructs absorb compressive loads and provide both structural and mechanical support. The unique material properties of porous trabecular metal allow it to achieve rapid bone ingrowth and osseointegration with the potential for long-term biologic fixation and restoration of bone stock. Clinical studies with trabecular metal cones have demonstrated excellent short-term outcomes31,33. Further long-term studies are necessary to determine the durability of these new modular augments.
With massive bone loss from osteolysis, it may be preferable, especially in younger patients, to attempt to restore the bone stock with structural allografts28,40. The advantage of this technique is that it is a biologic alternative with restoration of the distal end of the femur or proximal part of the tibia. The disadvantages of structural allograft include nonunion, late collapse, and disease transmission or infection. Most clinical studies have noted clinical success when the allograft-prosthetic composite is rigidly fixed securely to the host bone41. Diaphyseal engaging implant stems are necessary to ensure stability and fixation. In the rare case of severe osteolysis in an older patient with a pathologic fracture, a megaprosthesis with either a distal femoral or proximal tibial replacement may be an alternative.
In the presence of substantial osteolysis, fixation of the final components is important and stem extensions are used to enhance the immediate stability of the implant in the residual bone. Methods of stem fixation are a controversial aspect of revision total knee arthroplasty. Some advocate the use of cemented stem fixation, whereas others advocate the use of cementless press-fit stem extensions. Each has its own advantages and disadvantages35,42. The length and diameter of the extensions should be determined on the basis of the integrity of the residual bone and the dimensions of the intramedullary canal36,43. Cemented stems allow for intraoperative adjustment with unusual anatomy and achieve fixation in large canals and osteopenic bone. It is recommended that short, narrow, non-canal-filling stem extensions be cemented. Longer diaphyseal stems should not be cemented. Cemented stems are difficult to remove if revision is necessary, and since they are not canal-filling, they do not guarantee alignment44. Cementless press-fit stem extensions are easy to use and facilitate component alignment, and diaphyseal engaging stems ensure fixation. In revision total knee arthroplasty, the mechanical stability of the femoral and tibial components is increased by the addition of press-fit canal-filling stems, especially in the presence of poor metaphyseal bone quality or large bone defects43. Long modular stem extensions, which can be canal-filling and diaphyseal engaging, can be used in a press-fit manner by cementing only the condylar and metaphyseal portions of the femoral prosthesis as well as the tibial surface and metaphysis of the tibial component. While canal-filling stems are more predictable in guaranteeing alignment, anatomic variation may create the need for offset stems, especially in the tibia. In an anatomic study, Hicks et al. found that there was substantial variation in alignment between the center of the tibial plateau and the center of the tibial diaphyseal canal. Actual measurements of this difference ranged from 15 mm anterior to the center to 1.5 mm posterior to the center and 8 mm medial to the center to 4.5 mm lateral45. Because of this variability, an offset tibial stem may be necessary in revision arthroplasty to ensure accurate alignment, secure fixation, and correct orientation of the tibial tray without overhang (Figs. 6-A and 6-B).
In conclusion, revision total knee arthroplasty is a complex procedure that needs to address alignment, stability, fixation, and knee motion. For the complex reconstruction in the presence of osteolysis, it is important to determine the reason for failure and then to address the bone defects with appropriate augmentation of the bone loss and a stable well-fixed implant.