Osteolysis of the bone surrounding total hip replacements is a relatively common and potentially serious complication. This chapter reviews the etiology of osteolysis around total hip arthroplasty and treatment options for osteolysis around well-fixed acetabular and femoral components.
Plain radiographs are unable to detect subtle findings.
Patients with asymptomatic acetabular osteolysis must be educated and closely monitored with serial radiographs.
No pharmacologic agent has been identified that can either prevent the progression of osteolysis or reduce the size of an osteolytic lesion.
Osteolysis can occur without affecting implant stability in uncemented acetabular components.
Strategies for treating osteolysis around a well-fixed acetabular component include either a polyethylene liner exchange with debridement of accessible lesions or revision of the acetabular component and bone grafting of the osteolytic lesions.
Contraindications for liner exchange include a loose acetabular component, a malpositioned cup, an acetabular component with an ongrowth surface in the presence of a large osteolytic defect, a badly damaged outer shell, and lack of availability of a highly cross-linked polyethylene liner in a young patient.
The stability and positioning of the cup must be ascertained at the time of the procedure.
The largest size femoral head with the appropriate thickness of polyethylene should be selected.
Osteolytic lesions must be thoroughly debrided and then may be bone grafted.
Whenever possible, obtain the stickers that identify the components used in the prior total hip arthroplasty.
The optimal bone graft material has not been identified.
The equipment, components, and graft material must be available to perform a complete revision even when planning a polyethylene liner exchange.
Stability of the construct must be carefully assessed because dislocation can occur postoperatively.
Postoperative bracing may be useful because patients tend to have less pain than with primary total hip arthroplasty and often ignore dislocation precautions.
Osteolysis of the bone surrounding total hip replacements is a relatively common and potentially serious complication. Osteolysis is diagnosed radiographically and is characterized by areas of radiolucency in the bone adjacent to the implant. Osteolysis occurs as a result of the generation of wear debris and may be associated with pain and implant loosening, necessitating the need for revision surgery. The management of osteolysis is complicated because the disorder often develops silently and the components are often well fixed. This chapter reviews the etiology of osteolysis around total hip arthroplasty and treatment options for osteolysis around well-fixed acetabular and femoral components.
Osteolysis results from a cellular response to particulate wear and corrosion debris generated within a total hip arthroplasty. The most common reason for revision total hip arthroplasty is aseptic loosening caused by osteolysis. Wear debris and corrosion products from prosthetic implants initiate osteolysis by activating macrophages, a critical event in this process. Macrophage activation ultimately leads to osteoclastogenesis and bone resorption. A number of other cytokines also are activated in this process that affect bone turnover, including prostaglandin E2, tumor necrosis factor-α (TNF-α), interleukin-1, and interleukin-6. Fibroblasts regulate expression of the bone-resorbing metalloproteinases, which are responsible for catabolism of the organic components of bone. This collection of chemokines, cytokines, and mediators can stimulate cells to activate osteoclastic bone resorption around both acetabular and femoral components ( Fig. 39-1 ). Although these cytokines are involved in varying degrees in the osteolysis pathway, attempts to correlate the level of various mediators with the degree of osteolysis have failed. RANK ligand (RANKL) and osteoprotegerin (OPG) are key factors in the development of osteolytic lesions. RANKL promotes the stimulation of nuclear factor κB, which results in osteoclast differentiation and maturation. Osteoprotegerin is a natural RANKL antagonist secreted by osteoblasts that inhibits osteoclast activation. RANKL production is increased in response to resorption signals. Osteolytic bone resorption has been hypothesized to develop because of an imbalance in RANKL and OPG production. Periprosthetic tissues produce an increased amount of RANKL in response to wear debris, and OPG is decreased in the local bone microenvironment.
The predominant particle within periprosthetic tissue in total hip arthroplasty with a polyethylene acetabular cup articulating against a femoral head made of either metal or ceramic is polyethylene, with 90% of these particles less than 1 μm in size. Polyethylene comprises 70% to 90% of the debris volume. With respect to cytokine production, particles in the 0.1 to 1.0 μm range are the most biologically active. Other particles that may lead to osteoclast activation include polymethylmethacrylate, titanium alloy, and cobalt-chrome alloy corrosion products. In fact, osteolysis was originally thought to be the result of “cement disease.” The major determinants of particle bioreactivity are size, composition, and concentration. The concentration of wear debris particles from periprosthetic tissues is directly related to the duration of implantation, and billions of particles are generated per gram of tissue. Quantitation of the relative bioreactivity of metal, polymer, and ceramic particles is not possible because all can elicit a cellular response that varies with cell type and the specific response measured.
Histologic analysis of cemented components demonstrates variable-sized cement fragments in the tissue associated with foreign body giant cells. Periprosthetic bone resorption occurs most commonly in a linear manner, resulting in degradation of bone at the cement-bone interface and eventual loosening of the implant. In contrast, osteolysis of the pelvis associated with well-fixed porous coated acetabular components inserted without cement is more commonly localized and expansile. In many cases the acetabular component remains well fixed even when extensive osteolysis is present.
Several reports have established a relation between wear of the polyethylene acetabular cup and osteolysis, and an osteolysis threshold appears to exist regarding annual wear rates. In a study of 48 well-functioning acetabular components at 10-year follow-up evaluation, Dowd et al noted an increased prevalence of osteolysis with higher wear rates. Osteolysis did not develop in any of the nine hips with a true wear rate of less than 0.1 mm/yr. In contrast, osteolysis was noted in all eight hips with a wear rate of greater than 0.3 mm/yr. Orishimo et al reported a fourfold increase in the likelihood of osteolysis for every 0.1-mm increase in the annual linear wear rate. The results of this analysis quantitatively support prior observations that a wear rate of 0.2 mm/yr seems to represent a critical threshold for the development of osteolysis.
Alternative bearing materials have been developed in an attempt to decrease wear rates and osteolysis. Ceramic-on-ceramic prostheses have been shown to have extremely low wear rates (up to 4 mm 3 /yr). Ceramic wear particles generated in vivo and in hip joint simulations under microseparation conditions have a bimodal size distribution with nanometer-sized (5 to 20 nm) and larger particles (0.2 to more than 10 μm). Alumina ceramic wear particles have been shown to be capable of inducing osteolytic cytokine production; however, the volumetric concentration of the particles needed to generate this response is not generally present in vivo. Despite these findings, Yoon et al reported a high prevalence of osteolysis in the pelvis and, to a lesser extent, in the femur in 103 total hip arthroplasties performed with ceramic bearing surfaces. Metal-on-metal articulations produce particles that are in the nanometer-size range and have a very limited capacity to activate macrophages that induce osteolytic cytokines at the volumes generated in vivo. Osteolysis can still occur, as noticed in a recent study by Carr and Desteiger, who reported the presence of femoral osteolysis in 2.6% of 125 patients who underwent metal-on-metal total hip arthroplasties followed up for an average of 3 to 9 years.
A genetically based host variability exists in the reaction to particulate degradation products. These host factors could include the effectiveness of lymphatic drainage and the aggressiveness of the response by histiocytes. Identifying preoperatively who is high risk for the development of osteolysis could change bearing selection for specific patients, but this information is currently not available.
Currently, plain radiographs are the standard for detecting and monitoring osteolysis. Because radiographic evidence of osteolysis often precedes a patient’s symptoms, radiographs are routinely performed on postoperative visits after total hip arthroplasty to detect both osteolysis and wear. Conventional radiographs with a supine, internally rotated hip view are able to identify larger osteolytic lesions. The ability to assess osteolysis is improved by increasing the kV to overpenetrate the x-ray. Judet films also help increase sensitivity of detection. Another method to determine liner thickness and wear more accurately is by using templates of the acetabular implant ( Fig. 39-2 ).
Plain radiographs are unable to detect subtle findings such as small osteolytic lesions or accurately estimate the size of lesions in the superior ilium and ischium. They also have poor interobserver reliability in determining the rate of osteolysis. Leung et al found only 39% of lesions were detected on anteroposterior pelvis radiographs from specimens taken postmortem from patients who had total hip arthroplasty. Adding an iliac oblique view only increased detection to 52%, whereas CT scans identified 87% of these bony defects. Visualizing lesions developing in the ischium often is difficult, and the surgeon must carefully inspect the plain radiographs for osteolysis when wear or loosening is noted on plain radiographs.
In contrast, computed tomography (CT) provides three-dimensional image analysis and the ability to detect smaller lesions ( Fig. 39-3 ). Claus et al found in a cadaver model that CT scans were able to detect 81% of ischial lesions and 78% of acetabular rim lesions, with a 100% detection rate of lesions 10 cm 3 and larger. These findings were in contrast to an earlier study by Claus et al that found a sensitivity of less than 15% in detecting lesions in the ischium and acetabular rim with four plain radiographic views. Plain radiographs typically underestimate the size of the lesion. Stulberg et al recently advocated the use of CT to identify silent osteolysis. The author assessed the prevalence of CT-identifiable osteolysis in young, active patients after a minimum of 7 years, and silent osteolysis was identified on 48% of CT scans and only 24% of radiographs.
The increased cost and higher irradiation associated with CT scans makes it an unlikely replacement for radiographs in routine monitoring after arthroplasty. CT scans also can be used to delineate the extent of osteolytic lesions noted on screening radiographs, which may facilitate preoperative planning. Further study is necessary to determine a cost-effective way to use CT scans. The senior author (J.R.L.) often obtains a CT scan in a patient with either asymptomatic osteolysis or significant wear but minimal or no osteolysis on plain radiographs to educate the patient and provide guidance regarding the timing of surgical intervention.
No agent has been identified that can either prevent the progression of osteolysis or reduce the size of an osteolytic lesion. The release of TNF-α by macrophages after activation by particulate wear debris initiates the osteolysis process. Very low concentrations of TNF-α have been found to potentiate the effects of RANKL and stimulate osteoclast activation. In a mouse model Childs et al showed a reduction in osteoclast-mediated bone resorption with etanercept, a TNF-α inhibitor, compared with control animals. However, a prospective study by Schwarz et al found no difference in bone resorption after 12 and 24 months in 20 patients with radiographic evidence of acetabular loosening who were randomized to receive etanercept or placebo.
Because cyclooxygenase (COX) synthesis is increased in osteolytic tissue, nonsteroidal antiinflammatory drugs and COX-2 inhibitors have been used in an attempt to decrease osteolysis. Clinical studies have failed to identify a dose of COX-2 inhibitor or other nonsteroidal antiinflammatory molecules capable of preventing or reversing osteolysis. The major reason these medications may prove ineffective in preventing osteolysis is the presence of multiple parallel pathways leading to osteoclast activation independent of prostaglandins.
Bisphosphonates, which have been found to prevent bone loss associated with osteoporosis and metastatic tumors, also have been studied. Although several preclinical animal studies have demonstrated the efficacy of bisphosphonates in limiting osteolysis, this has not been noted in human beings. Lyons et al found no significant effect with oral alendronate therapy for 6 months in patients with aseptic loosening. Rubash et al conducted a multicenter clinical trial that evaluated the effect of alendronate on the radiographic progression of femoral osteolytic lesions after total hip arthroplasty. After 18 months, no significant treatment effect was noted.
A review of pharmacologic therapy in the treatment of osteolysis indicates insufficient evidence exists to recommend the routine use of a specific pharmacologic agent in the management of periprosthetic osteolysis at this time. However, intermittent osteoclast inhibition with a drug with a short half-life could be used in patients who show wear on plain radiographs but have minimal or no lysis. The goal would be to limit the development and progression of osteolysis in these patients. Drugs that inhibit RANK activity should be evaluated to determine their clinical efficacy in this area.