➢ Osteolysis is a process mounted by the host immune system that relies on several variables, including patient-related factors, type of insert material, modes of wear, and implant design.
➢ Imaging techniques such as radiography, computed tomography (CT) scans, magnetic resonance imaging (MRI), and tomosynthesis aid in diagnosing osteolysis.
➢ Surgical options for the treatment of osteolysis include the insertion of bone grafts, bone cement, and prosthetic augmentation.
➢ Although no approved pharmacological therapies for the specific treatment of osteolysis exist, the use of bisphosphonates and statins decreases the risk of osteolysis.
Osteolysis has long been an issue in joint arthroplasties. The early durability problems associated with Teflon (polytetrafluoroethylene) acetabular components as described by Charnley could be considered as the first manifestations of osteolysis, which he described as tissue “caseation and sterile sinus formation.”1,2 Polyethylene was turned to as a bearing material after Teflon exhibited poor wear properties in some of the earliest hip arthroplasties3. Later, the term cement disease4 was introduced when the process of osteolysis was mistakenly thought to be a reaction to the polymethylmethacrylate (PMMA) cement fixing the implant. However, the development of cementless implants did not resolve the osteolytic lesion problem, and, as such, the disease was then classified as a particle disease5. The most common reason for the failure of primary total knee arthroplasty (TKA) frequently has been cited to be aseptic loosening6,7. Osteolysis is the most pronounced manifestation of the biological process responsible for aseptic loosening, and, as such, osteolysis may be perceived as the most common cause of failure following primary TKA. The present review examines various mechanisms responsible for osteolysis as well as the diagnostic and treatment modalities that are currently available to orthopaedic surgeons and their patients.
Mechanisms of Osteolysis
The debris generated from prosthetic wear triggers a cascade of macrophage cytokines, such as interleukin-1-beta (IL-1β) and tumor necrosis factor-alpha (TNF-α), among others, resulting in osteoclastic bone resorption and eventually leading to osteolysis8. Howie et al.9 demonstrated that wear particles phagocytized by macrophages elicited different responses depending on particle size. Particles measuring <5 μm in size generated a strong mononuclear macrophage response, whereas larger particles resulted in more multinucleated giant cells. Another potential osteolytic mechanism is the formation of a membrane that produces collagenase and prostaglandin E2. Goldring et al.10,11 reported that, after total hip arthroplasty, a synovial-like membrane capable of producing both collagenase and prostaglandin E2 forms at the cement-bone interface. The loss of osseous support around the implant results in abnormal stress and force distributions at the bone-prosthesis interface during locomotion, potentially leading to periprosthetic fractures and premature implant failure12. Although the term osteolysis is frequently used interchangeably with aseptic loosening, these processes are in fact different. Aseptic loosening is an umbrella term that is used to describe total joint arthroplasty failure associated with inadequate initial fixation, mechanical loss of fixation over time, or biological loss of fixation caused by immunological reactions to wear particles, whereas osteolysis refers to the host immunological response that results in implant loosening.
Depending on the location and type of wear patterns, the debris particles within the prosthetic knee joint may consist of polyethylene, PMMA cement, or metal. Commonly described locations for the generation of wear debris include the tibial-femoral articulation surfaces, the posterior face of the tibial insert (backside wear), and the implant-bone interface13-15. Although wear can occur at all of these locations, polyethylene forms the great majority of wear particles16. As such, this review will focus on polyethylene wear rather than metal or cement wear.
Osteolysis also may be examined by assessing mechanical wear patterns. Several potential modes of polyethylene wear may result from the combination of rolling, sliding, and rotational motions that may occur in an artificial knee joint. The tribological interactions of solid surfaces within the artificial knee lead to the development of surface abrasions and erosions, particularly on the softer polyethylene surface17. In another mode of wear, adhesion, the polyethylene material smears onto the metal component of the joint17. Furthermore, fatigue failure is the formation of subsurface cracks in the polyethylene caused by cycles of loading and unloading of the joint, which then propagate and create particles that are shed into the joint space17-19. Of these wear patterns, adhesion has been shown to produce the particles with the greatest biological activity (diameter, <1 μm)20.
Osteolysis also may be facilitated by hydrodynamic forces21. Inflammatory mediators such as histamine and nitrous oxide cause an increase in the permeability of capillary endothelium, leading to an increase in the interstitial fluid volume and hydrostatic pressure levels. This higher pressure can increase the spread of wear particles within the joint space and lead to osteocyte death and subsequent resorption of bone by osteoclasts22.
Potential Risk Factors
There are various patient-specific risk factors for the development of wear debris and osteolysis following primary TKA. Patient age and sex correlate with the risk of osteolysis. One study demonstrated that every one-year decrease in patient age was associated with an approximately 5% increased likelihood of failure due to wear23. Males have been reported to have a 2.8-times greater risk than females for the development of wear-related failure following TKA23. These two risk factors are likely due to higher postoperative activity levels in younger, male patients24. One study demonstrated that patients less than sixty years old are 30% more active than those more than sixty years old, with males being 28% more active than females24. In clinical studies, patient body habitus has not been shown to be a consistent factor associated with polyethylene wear, possibly because higher body mass index scores have been associated with lower activity levels25.
Type of Insert Material
Another factor related to the development of osteolysis following TKA is the type of polyethylene material in the tibial insert. The three types of polyethylene used for TKA are conventional (non-cross-linked) polyethylene, ultra-high molecular weight polyethylene (UHMWPE), and highly cross-linked polyethylene (HXLPE).
Conventional polyethylene can be manufactured by either machining or compression molding, with the latter type having been shown to exhibit less posterior wear26,27. UHMWPE is a subset of polyethylene materials with a molecular weight of 2 to 6 million units26,27. The UHMWPE type used for TKA is a semicrystalline polymer with approximately 50% of the chains in a solid crystalline phase and the remainder in an amorphous phase26. HXLPE is formed from UHMWPE via gamma radiation, which breaks the long carbon-hydrogen chains and allows cross-links to form within the amorphous phase. The cross-links confer increased wear resistance to the polymer while also lowering its tensile strength and resistance to fatigue crack propagation28-30. However, the radiation forms free radicals, which initiates the degradation of the polymer31,32. This oxidation can lead to early mechanical failure, resulting in the production of polyethylene wear particles and potentially osteolysis.
Free radicals are much more prevalent when the polyethylene component is sterilized with gamma radiation in air. This method has since been replaced by sterilization with either gaseous ethylene oxide or gamma radiation in an inert gas, both of which generate fewer free radicals33. Improved sterilization techniques have demonstrated a substantial decrease in radiographic evidence of osteolysis at five to ten years postoperatively34. There are several post-irradiation processes that can be performed on polyethylene to further improve its material properties. Post-irradiation melting of HXLPE further reduces its wear rate and free radicals trapped within the crystalline region35. However, this process also reduces its crystalline structure, resulting in a loss of material strength36. Another option is post-irradiation thermal annealing of the HXLPE, which avoids the reduction in the crystalline structure. However, oxidation remains problematic because of incomplete elimination of free radicals37,38.
Second-generation HXLPE is infused with vitamin E, an antioxidant that reduces residual free radicals39,40. One in vitro study indicated that vitamin E prevents the oxidation of irradiated UHMWPE when subjected to accelerated aging via cyclic loading or squalene absorption41. This prevention of oxidation would eliminate the need to melt and anneal the polyethylene after irradiation, resulting in higher strength and fatigue resistance than first-generation HXLPE42-45. A more recent animal study demonstrated that wear debris from vitamin E-infused cross-linked UHMWPE produced a considerably attenuated osteolytic response compared with standard UHMWPE46.
There have been conflicting results concerning the superiority of HXLPE to conventional polyethylene in the setting of knee arthroplasty. Fisher et al.47 compared the in vitro wear rates and debris production of conventional and cross-linked polyethylene and found that HXLPE had lower wear rates. However, the HXLPE wear particles were found to be more biologically active, producing a higher TNF-α response47. Similar results were reported in a more recent study, in which HXLPE was associated with fewer and smaller, but more biologically active, wear particles48. However, different results were observed in an in vitro study, in which HXLPE was associated with decreased wear rates and the same particle size compared with conventional polyethylene40. In addition, a retrospective study showed no significant difference between HXLPE and conventional polyethylene in terms of the rate of osteolysis following TKA49. In addition to material performance, another factor to consider when comparing polyethylene inserts is cost. “Premium” (cross-linked) polyethylene components cost ∼33% more than conventional components50. The disadvantages of HXLPE, including susceptibility to fatigue failure and higher cost, coupled with conflicting results related to its advantages, call into question whether it is the superior material for tibial inserts. More research in this area is needed to definitively answer this question.
Another material choice is the type of metal used in the implant. One study51 showed that titanium-aluminum-vanadium particles had a greater immunological effect than cobalt-chromium (CoCr) particles in terms of the release of IL-1β, TNF-α, and IL-6. This difference was due to the titanium particles being minimally toxic to macrophages, which led to higher levels of IL-1β release and a more robust inflammatory response. The CoCr particles were much more toxic to macrophages and thus had little effect on the release of pro-inflammatory mediators.
Backside Wear Related to Implant Design
Backside wear is due to the micromotion between the tibial baseplate and the polyethylene insert during physiological loading52,53. Backside wear also occurs in association with elastic and plastic deformation, which are seen in thin polyethylene inserts54. As a result, the minimum recommended thickness for polyethylene inserts is 8 mm52,55-57. Although previous studies19,58-60 have indicated that the articulation surface is the predominant source of polyethylene debris, more recent studies have indicated that backside wear volume is comparable with articulation surface debris volume61,62. Backside wear increases when motion at the articular surface is reduced63. Schwarzkopf et al.64 reported that while highly conforming articulation surfaces have lower contact stresses and wear at the articulation interface, they also generate higher levels of backside wear. Wear debris particles generated from the backside of the tibial insert have been shown to be smaller than particles from the articular surface and may have increased biological activity65.
As the modular tibial implant design (i.e., a polyethylene insert in a metal baseplate) introduces a new potential source of wear debris, its benefits should be assessed in comparison with those associated with all-polyethylene tibial components. The modular design allows the bearing surface to be matched with the femoral implant component, independent of the tibial baseplate66. Finite element analysis also has demonstrated improved fixation via reduced stresses at the bone-prosthesis interface when the modular tibial component has been compared with the all-polyethylene tibial component67-70. Several recent retrospective studies71,72 have shown identical ten-year survival rates for modular and all-polyethylene tibial implants. All-polyethylene tibial components are approximately 30% less expensive than metal-backed tibial components and effectively eliminate biologically active backside wear71. As such, the all-polyethylene tibial design may be an economically and biomechanically superior alternative to modularity. Nevertheless, the majority of primary TKAs are performed with use of modular tibial inserts72; thus, additional studies comparing the biomechanical outcomes associated with the two designs are warranted.
There are other design variables that may affect wear rates and osteolysis following TKA. One additional decision related to procedures involving modular tibial components is whether the polyethylene insert should be secured with use of a fixed-bearing or rotating-platform design. With fixed-bearing designs the polyethylene insert is press-fitted into the metal tibial tray, whereas with rotating-platform designs the insert can be rotated relative to the tray. One study demonstrated that volumetric wear rates under intermediate to high kinematic loading were up to four times lower following TKAs performed with rotating-platform as compared with fixed-bearing designs73. Another design choice is whether to use a posterior-stabilized or posterior-cruciate-retaining prosthesis. Posterior-stabilized designs tend to have highly conforming articulation surfaces that transfer stresses to the stem-bone and baseplate-polyethylene insert interfaces74,75, leading to increased motion and wear at the backside interface.
Diagnosing and Monitoring the Progression of Osteolysis Following TKA
The ability to accurately diagnose and monitor the progression of osteolysis following TKA is critical. An earlier diagnosis allows for a less-invasive and more-efficient revision TKA. The difficulty of achieving this target is mainly due to the absence of symptoms in the early stages of the disease and the low sensitivity of current imaging modalities. By the time polyethylene wear is detectable, osteolysis likely has progressed and has resulted in implant loosening76. Four imaging techniques are used to detect polyethylene wear and osteolysis: radiography, computed tomography (CT) scans, magnetic resonance imaging (MRI), and tomosynthesis. Tomosynthesis allows for high-resolution limited-angle tomography and is less limited by metal artifacts than CT scanning.
Radiographs are primarily used as the initial screening tool for the detection of both osteolytic lesions and polyethylene wear77. Osteolytic lesions appear as radiolucent areas in the bone adjacent to the prosthesis, whereas polyethylene wear manifests as narrowing of the joint space. In order for radiographs to detect polyethylene wear, they must be tangential to the articular surface77. Although tibial and patellar osteolysis are often identified on radiographs of the knee, femoral osteolysis is usually underestimated because of the concealing nature of the femoral component77. In order to circumvent this issue, Nadaud et al.78 recommended using oblique radiographs to analyze femoral bone for osteolysis. Goldvasser et al.76 reported that osteolysis has a higher likelihood of being identified correctly through radiographic analysis at the time of revision when a systematic approach and scoring algorithm are used. Specifically, the mean total delamination score was shown to be indicative of osteolysis (p = 0.05) and the radiographic score was significantly higher when osteolysis was mentioned in the surgical notes (p = 0.033)76. The researchers also noted that anteroposterior radiographs of the tibia were most sensitive for the detection of osteolysis76.
Solomon et al.79, in a cadaveric study, assessed the capacity of CT scans with metal artifact-suppression software, MRI, and fluoroscopically guided radiography to detect larger (0.7 to 14-cm3) osteolytic knee defects. MRI and CT scanning were superior to fluoroscopically guided radiography, with no significant difference in overall sensitivity or specificity between the MRI and CT. However, MRI displayed superior sensitivity compared with CT for detecting femoral defects.
Another study compared the ability of MRI, CT scans with metal artifact-suppression software, radiographs, and tomosynthesis to detect smaller (0.7-cm3) femoral bone defects80. Tomosynthesis was demonstrated to be superior for the early detection of osteolytic lesions. These lesions were completely undetected on both radiographs and MRI scans because they were obscured by the metal femoral implant. CT scanning detected smaller lesions, but with less sensitivity (61.5% compared with 85.4%) and specificity (64.1% compared with 87.2%) than tomosynthesis. Tomosynthesis also was associated with a 94.4% lower dosage of radiation and an approximately 70% lower cost in comparison with CT scanning.
The above studies indicate that both CT and MRI are viable options for assessing the extent of larger osteolytic lesions to aid in preoperative revision TKA planning. Concurrently, tomosynthesis shows potential as a screening tool to detect early small osteolytic lesions. Radiographs are less effective than MRI and CT scanning even for the detection of large lesions but are commonly used as a screening tool because of their low cost.
There has been a shift in many developed countries to use digital imaging in place of radiographs. Zotti et al.81, in a cadaveric study, examined the ability to detect periprosthetic femoral and tibial osteolytic lesions with use of CT scans and fluoroscopically guided radiographs. In that study, plain radiographs were found to be superior to digital imaging for detecting lesions.
Another potential tool for aiding in the diagnosis of osteolytic lesions is biomarkers, which help to assess metabolic processes82. A reliable biomarker for osteolysis will improve patient outcomes as patients may remain asymptomatic until substantial bone loss has occurred. A known marker of osteoclast activity is the level of tartrate-resistant acid phosphatase (TRAP) in synovial fluid83. Kim et al.84 also reported a significant correlation (p < 0.0001) between synovial fluid TRAP levels and osteolytic pit formation on dentin slices. More recent studies85-89 have yielded conflicting results, failing to produce a reliable biomarker for osteolysis. These studies have examined candidate biomarkers such as TRAP 5b, CTX-1 (C-terminal telopeptide of type-I collagen), OPG (osteoprotegerin), cathepsin K, and RANKL (receptor activator of nuclear factor kappa-B ligand). More research is needed to elucidate biological markers that indicate osteolysis.
Treatment of Osteolysis
The surgical treatment of osteolysis around an artificial knee joint depends on the size of the lesion. The available treatment modalities for repairing bone lesions following primary TKA include the insertion of prosthetic metal wedges, bone grafts, and bone cement. Bone cement may be used to fill small (≤10-mm) isolated defects at the joint surface90-94. A limitation of bone cement is that the polymerization process is exothermic, releasing heat that can cause thermal necrosis of surrounding bone and vascular tissue, leading to aseptic loosening95,96.
For defects measuring >10 mm in diameter, metal augmentation or bone grafts are usually indicated94. Bone graft is often used in younger patients in an attempt to preserve bone94,97, whereas prosthetic augmentations are frequently used in elderly and low-activity patients94.
Successful osseointegration of bone graft relies on several graft properties, including (1) osteogenesis, (2) osteoclastic reabsorption, (3) osteoinduction, and (4) osteoconduction. Autograft bone demonstrates all of these properties, whereas allograft bone only demonstrates osteoclastic reabsorption and osteoconduction, allowing bone autografts to heal faster than allografts. Autograft bone is predominately used for smaller isolated lesions98. Bone allografting may be performed with either structural bone or morselized bone that is either impacted99 or loosely packed100,101 during revision surgery. Furthermore, impacted-bone allograft may be used in combination with fine-wire mesh to fill larger uncontained bone defects102.
The tantalum cone prosthesis has shown promise for correcting large bone defects in the metaphyseal region103,104. The modulus of elasticity of the implant is similar to that of bone, making it a reasonable substitute105-108. The prosthesis incorporates advanced osseointegration properties, including a high coefficient of friction, allowing for initial stability of press-fit prostheses. Furthermore, the implant’s high porosity permits osseous ingrowth, ensuring long-term prosthetic stability109-111. The tantalum cone prosthesis is superior to bone graft in that it is not associated with the risk of disease transmission from donor to host, it does not degrade with time, and it has an established structure that provides joint stability105-108. Potential drawbacks to the use of tantalum cones include high cost, the lack of ability to restore bone stock, and the difficulty of removal103.
Three-dimensional printing recently emerged as a potential alternative treatment modality. The technique involves introducing calcium phosphate scaffolds, loaded with growth factors, at the site of osteolysis. Early in vitro studies have demonstrated improvement in the biomechanical and osteogenic properties of the cement; however, additional research should be conducted prior to the implementation of these new technologies in the operating room112.
There is currently no medication specifically approved by the U.S. Food and Drug Administration to prevent periprosthetic osteolysis. One potential new treatment for osteolysis is the anti-inflammatory agent N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer-dexamethasone conjugate (P-Dex). A monthly injection demonstrated the ability to prevent wear-induced osteolysis, without the adverse effects of glucocorticoids113. Although advancements in osteolytic pharmacotherapy are being made, statins and bisphosphonates are currently the only available therapies that have been demonstrated to decrease the risk of osteolysis114,115.
Illustrative Case Report
An eighty-year-old woman presented with a four-month history of pain in the right knee; the pain was exacerbated by activity. The patient had undergone a cruciate-retaining TKA ten years previously for the treatment of osteoarthritis and had not had any complaints prior to the current presentation. The patient reported no swelling, fever, or chills. Physical examination revealed a range of motion of 0° to 100°, with anteroposterior, varus, and valgus stability. A well-healed scar was noted at the incision site, with no signs of inflammation or effusion. Laboratory workup was negative for infection, with the C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), and white blood-cell (WBC) count all falling within the normal range. Preoperative anteroposterior and lateral radiographs of the right knee revealed massive tibial osteolysis as well as some femoral involvement (Figs. 1-A and 1-B). Intraoperative assessment showed that the components were well positioned within normal limits, with signs of polyethylene wear on gross examination. The patient subsequently underwent a revision TKA (Figs. 2-A through 2-D) with bone-grafting and with conversion of the implant to a stemmed posterior-stabilized prosthesis (Fig. 3).
While improved knee implant materials have led to a lower risk of osteolysis, osteolysis remains a common cause of failure after TKA. Although the exact mechanisms remain unclear, the major inflammatory factors and general biological processes have been investigated. Various patient-related risk factors exist, with patient activity level being the most influential. There is controversy with regard to whether HXLPE is superior to conventional polyethylene for the tibial insert. Vitamin E-infused HXLPE shows much promise, however, and future studies may reveal that it is associated with superior results. Commonly used imaging techniques such as CT scanning and MRI do not allow for the early detection of osteolysis, but recent studies have demonstrated that tomosynthesis is a potential early screening tool. The combination of biomarkers with imaging modalities would allow for earlier, less-invasive treatment of osteolytic lesions. Until additional studies assess these variables, revision TKA remains the treatment of choice for patients with osteolysis following primary TKA.
Investigation performed at the Division of Orthopaedics and Rehabilitation, Department of Surgery, Southern Illinois University School of Medicine, Springfield, Illinois
Disclosure: There was no external funding source. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author (or the author’s institution) had a relevant financial relationship in the biomedical arena outside the submitted work.
- Copyright © 2016 by The Journal of Bone and Joint Surgery, Incorporated