 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
The Journal of Bone and Joint Surgery (American). 2008;90:637-654. doi:10.2106/JBJS.G.01012 © 2008 The Journal of Bone and Joint Surgery, Inc.
Metal-on-Metal Hip Resurfacing Arthroplasty
1 Melbourne Orthopaedic Group, 33 The Avenue, Windsor, Victoria 3181, Australia. E-mail address: ashimmin{at}optusnet.com.au
Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. One or more of the authors or a member of his or her immediate family received, in any one year, payments or other benefits in excess of $10,000 or a commitment or agreement to provide such benefits from a commercial entity (Wright Medical Technology). Also, commercial entities (Wright Medical Technology, DePuy, and Zimmer) paid or directed in any one year, or agreed to pay or direct, benefits in excess of $10,000 to a research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which one or more of the authors, or a member of his or her immediate family, is affiliated or associated.
The main advantage of hip resurfacing is bone conservation for patients likely to outlive a primary conventional hip replacement. Previous attempts at hip resurfacing failed predominantly because of the consequences of a high amount of wear of thin polyethylene acetabular components and poor femoral component fixation. With correct patient selection, surgeon education, and operative technique, survivorship at five years is comparable with that of traditional hip replacements. Hip resurfacing has its own unique set of complications, including a fractured neck of the femur. It is necessary to understand the risk factors prior to performing the procedure.
Hip resurfacing arthroplasty, which is viewed by many as representing an evolution from the mold arthroplasty procedure of Smith-Petersen, has been performed with a variety of materials, designs, operative approaches, techniques, and fixation methods1-6. In the 1970s, despite failures related to technical errors and concerns about osteonecrosis, the procedure was perceived as a major advance in the treatment of young and active patients1,7-10. Unfortunately, the materials available at the time had insufficient wear resistance, and high rates of failure due to wear-debris-induced osteolysis and loosening led to the abandonment of the first-generation hip resurfacing prostheses5,11,12. Problems were also associated with poor fixation methods and the lack of standardization of operative approach, technique, and patient selection. In contrast, prosthetic longevity in small series of hemiresurfacing procedures (in which the femoral head is resurfaced and articulates against the natural acetabulum)13-15 and analysis of the results of metal-on-metal total hip replacements16-20 suggested that, with improved bearing materials, the operation could still be useful as an option for primary arthroplasty in young patients who are likely to require a future revision hip procedure.
There are many suggested advantages of hip resurfacing arthroplasty. These include bone conservation21-23, improved function as a consequence of retention of the femoral head and neck and more precise biomechanical restoration24,25, decreased morbidity at the time of revision arthroplasty26,27, reduced dislocation rates28,29, normal femoral loading and reduced stress-shielding30, simpler management of a degenerated hip with a deformity in the proximal femoral metaphysis (after trauma or osteotomy)31,32, an improved outcome in the event of infection, and a reduced prevalence of thromboembolic phenomena as a consequence of not using instruments in the femur33. There are limited or inconsistent data to support some of these claims, but we will review the available evidence.
Metal-on-Metal Bearings Prosthetic bearing surfaces for hip resurfacing operations are currently manufactured from high-carbon (0.20% to 0.25%) cobalt-chromium-molybdenum alloy. The implants are either cast or forged, and casting may be followed by heat treatments that do not change the chemical compositions but do alter the microstructure (for example, porosity). The relationship of these different microstructures to long-term wear resistance is controversial6,34-36, although recent hip-simulator studies examining the effect of heat treatments have not demonstrated an important difference37,38. Lubrication theory predicts that, in contrast to the extensive wear resulting from large-diameter polyethylene resurfacing components39, the large-diameter metal-on-metal components could potentially result in very low wear if other important factors such as surface smoothness and, in particular, diametral clearance (the difference between the diameters of the femoral head and acetabular cup) are optimized40-43. While the optimal clearance to achieve elastohydrodynamic lubrication and avoid equatorial seizing is still being studied and debated, tribologists recommend that the diametral clearance be as small as possible in large-diameter bearings41,42. This requirement must be balanced against practical limitations of manufacturing tolerances and also must take into account the possibility that deformation of the acetabular component may occur when it is implanted into the acetabulum with a press-fit of 1 to 2 mm. Further deformation of the acetabular component, and subsequent reduction of the effective clearance, may also occur with physiological loading. The effect of deformation of the acetabular component on clearance has been studied experimentally in cadaver pelves or foam models44 and with the use of finite-element modeling45. Both studies showed that the most important factor influencing deformation of the acetabular component was the wall thickness of the component, although diametral clearance and component diameter also were important. Deformation was greater in acetabular components manufactured with thin (2 to 4-mm) walls to conserve pelvic bone. Manufacturers of these thin metal-on-metal acetabular resurfacing components must ensure that deformation of the component does not adversely affect clearance, as this would lead to increased bearing wear.
Wear of Retrieved Components Wear-depth measurements of retrieved modern hip-resurfacing components have generally been in agreement with hip-simulator predictions of low wear of well-manufactured, well-positioned implants48,49. Two of us (P.C. and P.E.B.) and colleagues49 reported wear measurements of well-functioning implants that ranged from less than detectable (at 2-µm resolution) to several micrometers per year. Wear rates of >100 µm/yr were associated with malpositioned sockets49. Those failures commonly were associated with tissue metallosis and a focal concentrated wear zone on the femoral component as the result of edge loading (Fig. 1); wear-induced osteolysis has been noted in such cases. There was a wide range of diametral clearances (123 to 400 µm), with metallosis occurring with very high clearance.
Reaction to Debris Locally, particulate metal has been shown to cause the release of inflammatory cytokines from macrophages50,51 and wear-induced osteolysis is occasionally reported, usually as a consequence of metallosis from malfunctioning implants52,53. The migration of particulate metal and corrosion products to distant end organs has been reported in association with metal-on-polyethylene hip and knee components retrieved at autopsy54,55, and these products can induce pathological changes such as histiocytosis, fibrosis, or necrosis47. The possibility of long-term consequences of chronic particulate metal release, including carcinogenicity or other metabolic disorders, is often noted as a concern56. However, a meta-analysis showed no increased risk to patients with metal-on-metal conventional total hip replacement implants57. Nevertheless, MacDonald cautioned that studies to detect a rise in such adverse effects would require many thousands of patients to be followed for several decades58. Such data are not available from studies on hip resurfacing procedures. Osteolysis as a result of metal allergy rather than wear debris has been reported in association with metal-on-metal hip replacements in a small number of cases59. Although the prevalence of wear-debris-induced osteolysis and allergic reactions in patients who have had a metal-on-metal hip resurfacing arthroplasty appears to be <1%59,60, the longest current clinical experience has been limited to two or three implant designs and more clinical data are required to determine the true prevalence of these complications.
All metal-on-metal hip resurfacing arthroplasties in current clinical use are done with cementless acetabular fixation with use of ingrowth surfaces of cast cobalt-chromium beads or plasma-sprayed titanium with or without hydroxyapatite. While there is interest in cementless femoral fixation, the majority of femoral resurfacing components are cemented. Fixation philosophy varies with regard to the presence of a mantle of cement and the depth of cement penetration into the femoral head, which in turn leads to differences in cement volume and viscosity. Concerns have been raised regarding the effect of the depth of cement penetration on the viability and integrity of the remaining femoral head bone61,62 and thermal necrosis of the interfacial bone contributing to failure49,63,64. Clinical results to date tend to suggest that, in most cases, current cementing practices provide adequate femoral fixation. However, the authors of at least one study advocated cementation of the peg in the presence of poor-quality bone, which they reported to have encouraging results65. This finding should be balanced against those of recent finite-element-analysis models that have shown that cementation of the peg could favor stress-shielding of the underlying femoral head, leading to loosening in the long term66.
The number of hip resurfacing designs continues to increase, and most major implant manufacturers now offer these devices. While they share common design features, there are variations such as internal geometry, metallurgy, peg design, and acetabular fixation surfaces (Table I). The allowance for a cement mantle in some resurfacing designs influences the amount, type, viscosity, timing, and application of the cement67. The effect of these features on implant wear, clinical performance, and longevity will need to be investigated in large long-term cohort studies.
Patient Selection The ideal candidate for a metal-on-metal hip resurfacing operation is young and active and has isolated hip disease with good proximal femoral bone quality and morphology and normal kidney function68. This type of physiology will allow the patient to resume an active lifestyle once pain relief is achieved by an arthroplasty procedure. Available registry data suggest that hip resurfacing arthroplasty is a reasonable option for men less than sixty-five years old and women less than fifty-five years old69. Relative contraindications for hip resurfacing arthroplasty include inflammatory arthritis, severe acetabular dysplasia, poor proximal femoral bone geometry (such as a short femoral neck with a high-riding greater trochanter), poor femoral bone stock due to large femoral head cysts, erosive arthritis, a woman of child-bearing age, known metal sensitivities, and a limb-length discrepancy of >2 cm68,70. Absolute contraindications include osteoporotic proximal femoral bone (such as in elderly patients or those taking corticosteroids long-term) because of the increased risk of fracture, compromised renal function because of the risk of impaired excretion of metal ions, and a proximal femoral tumor. One of us (P.E.B.) and colleagues71 calculated a surface arthroplasty risk index (SARI) on the basis of data from a group of patients less than forty years old. The patients were assigned a numerical value on the basis of the presence of four risk factors: femoral head cysts of >1 cm (2 points), a weight of <82 kg (2 points), previous proximal femoral surgery (1 point), and a University of California at Los Angeles (UCLA) activity score of >6 (1 point). They reported that a SARI score of >3 represented a twelvefold increase in the risk of early failure or adverse radiographic change. This index remained valid regardless of the underlying diagnosis72 and implant design73.
Hip Resurfacing Arthroplasty in Women of Child-Bearing Age Recently, Ziaee et al.76 confirmed previous reports that both cobalt and chromium ions cross the placenta. It was also established that the placenta has a modulatory (reducing) effect on the concentration of these ions that reach the fetus. It should be noted that, in all ten cases studied, the serum levels of maternal metal ions were within the typical range. The effect on the fetus of a mother who is an outlier from that typical range, with extremely high ion levels, is still unknown. At this time, women of child-bearing age should be informed of the theoretical risks to the fetus associated with metal ion exposure. Because of these theoretical risks, these patients should consider conventional total hip arthroplasty with bearing surfaces other than those consisting of metal on metal. Ziaee et al. recommend that women of child-bearing age who strongly desire a metal-on-metal hip resurfacing arthroplasty delay childbirth for approximately two years after implantation, when the run-in wear phase has finished76.
Reintroduction of Hip Resurfacing Arthroplasty into Current Clinical Practice Prior to commencing a hip resurfacing arthroplasty, surgeons must be aware that there is a learning curve associated with this operation77, and they are strongly urged to undertake additional training, including didactic courses, supervised cadaver dissection, independent cadaver dissection, and visits to watch surgeons who have experience with the performance of hip resurfacing operations.
Approach The optimal operative approach for hip resurfacing arthroplasty is controversial, and each proposed technique has advantages and disadvantages (Table II). In contrast to the initial experience with metal-on-polyethylene hip resurfacing arthroplasties, the majority of which were done through approaches that dislocated the hip anteriorly, most current-generation metal-on-metal hip resurfacing procedures for which results have been reported were performed through the posterior approach. The advantages of the posterior approach include the excellent exposure after circumferential capsulotomy, the preservation of the hip abductor muscles, and its ease of reproducibility by the majority of surgeons. However, with the release of the short external rotator muscles, the main blood supply to the retinacular vessels of the femoral head (the ascending branch of the medial circumflex artery) is sacrificed. This may lead to a compromised blood supply and osteonecrosis.
The role of devascularization in femoral failure has been questioned62,78-80. It has been suggested that the presence of osteoarthritis favors the development of an intraosseous blood supply within the femoral head, hence reducing the role of the extraosseous vessels81. Blood flow and oxygen tension studies of osteoarthritic femoral heads, however, have not confirmed this62,82,83. Some surgeons who perform hip resurfacing arthroplasty advocate a vascular-sparing approach84,85, although compromising the extraosseous blood supply does not appear to be a major clinical issue with current designs of hip resurfacing prostheses implanted through the posterior approach49,82,86. Exposure of the acetabulum without a prior femoral neck resection can present technical challenges. Accurate placement of a guide pin in the femoral neck is necessary to avoid varus positioning of the component and notching of the femoral neck. Computer navigation systems may be useful for the performance of the procedure87,88. Because of the technical challenges, one report has suggested that the so-called learning curve for accurate placement of the femoral component is in excess of twenty cases89. It is recommended that surgeons initially use the approach with which they are most comfortable.
Biomechanical Reconstruction
Although relative valgus orientation is favorable to implant survivorship, it results in a decreased femoral offset92. Silva et al.70 and Loughead et al.95 documented an average decrease in femoral offset ranging from 4.5 to 8 mm. This, combined with a limited capacity to correct a limb-length discrepancy of >2 cm, has put into question the capacity of hip resurfacing arthroplasty to properly restore hip biomechanics. However, both of these studies were retrospective, and, in a more recent prospective randomized clinical trial comparing hip resurfacing arthroplasty with conventional total hip replacement, Girard et al. found that a greater percentage of resurfaced hips had the offset reconstructed to within 4 mm of that on the normal, contralateral side25. The lack of modularity of the femoral component represents a major difference between hip resurfacing prostheses and conventional total hip replacement devices, especially when the surgeon attempts to optimize the femoral head-neck offset in order to minimize the risk of impingement and maximize the range of motion92. The best method for optimizing the head-neck offset during hip resurfacing arthroplasty is still not known. One technique involves the removal of prominent osteophytes on the anterior aspect of the head and neck to restore femoral head sphericity, optimizing component sizing and facilitating accurate guidewire placement96. Although osteophyte removal could weaken the femoral neck if it is done too aggressively, the arthritic femoral head is usually enlarged and thus the surgeon may tend to favor the use of a larger femoral component if the osteophytes are preserved. This will also result in the implantation of an acetabular component that is larger than what might have been used in a conventional total hip replacement92.
Because the reintroduction of the hip resurfacing concept is relatively recent, there is a lack of published long-term results. Most publications are from centers in which the authors have been involved with the design of the prostheses. The short-term (up to two-year) to medium-term (five-year) results of clinical outcome and survivorship studies from independent centers using contemporary hip resurfacing implants have been similar to, and complication rates have been comparable with, those of conventional total hip replacements97. With the exception of designs that involved a cemented acetabular component73,98, current hip resurfacing prostheses (those involving cementless acetabular fixation—i.e., hybrid fixation) have been associated with encouraging early results. Investigators from one center reported a 99.8% survivorship at a mean of 3.3 years in a group of 446 hips in patients less than fifty-five years old with a diagnosis of osteoarthritis99. They also reported on 144 consecutive cases followed for a minimum of five years; the survivorship was 99% for aseptic cases33.
At another center, a 94.4% survivorship was reported at four years after 400 hip resurfacing prostheses performed with hybrid fixation in a group of patients (average age, forty-eight years) in which only 66% had a primary diagnosis of osteoarthritis100. The authors of that paper acknowledged that the limits of the hip resurfacing concept were probably being extended at their center. When their data were separated according to whether the patients had a surface arthroplasty risk index (SARI) of >3 or More recently, one of us (A.S.) and colleagues reported a 99.13% cumulative survivorship at a mean of five years in a prospective study of 230 resurfaced hips101. At another center, the clinical and radiographic results of metal-on-metal hip resurfacing arthroplasty were compared with those of conventional total hip replacement in two groups of fifty-four patients matched for age, gender, body mass index, and activity level102. At five to seven years, the total hip replacement group had a revision or intent-to-revise rate of 8% (four of fifty-one hips) due to polyethylene wear and osteolysis. The hip resurfacing group had a revision or intent-to-revise rate of 6% (four of sixty-three hips) due to femoral component migration. Both of these revised or intent-to-revise rates are higher than what one would expect from other published reports69. There is debate about the range of motion achieved with hip resurfacing arthroplasty. In one report, the mean flexion range was 111.2°, which is lower than that reported after conventional total hip replacements103. In a prospective, randomized, controlled trial comparing seventy-six patients with a conventional hip replacement and eighty with a hip resurfacing arthroplasty, Lavigne et al. demonstrated no difference between groups in terms of the total arc of motion in all planes104. The Australian Joint Replacement Registry has been in operation since 1999, the year in which hip resurfacing arthroplasty was introduced into clinical practice in that country69. The registry is efficient in capturing data on >95% of all inserted and revised implants and, at the time of writing, contained information on 113,327 primary total hip replacements, including 8945 hip resurfacing operations. The registry data suggested that men with a hip resurfacing implant who are less than sixty-five years old have the same revision rate at four years as do men of the same age who have a conventional total hip replacement. The revision rate for women with a hip resurfacing implant is twice that for men with a hip resurfacing implant. The primary pathological condition can affect the revision rate, with inflammatory arthropathy, osteonecrosis, and developmental dysplasia associated with higher rates of early failure69. Patients with advanced osteonecrosis are a particularly challenging group to treat with hip resurfacing arthroplasty. Large femoral head defects as well as associated risk factors such as continued use of corticosteroids substantially compromise bone quality and the surface area available for implant fixation in the femur72. Revell et al. recently reported a survivorship of 93% in a consecutive series of seventy-three osteonecrotic hips followed for an average of 6.1 years after resurfacing105. Mont et al. reported similar results (a survivorship of 94%), at four years, for forty-two hips with osteonecrosis106. Results from these series suggest that hip resurfacing arthroplasty should be approached with caution for this patient population. The question of how much of the remaining viable head and neck bone is required for successful hip resurfacing arthroplasty has not been answered in the literature. Patients with a diagnosis of osteonecrosis presenting with a SARI score of >3 are at a higher risk for early failure72. The decision to proceed with hip resurfacing must be made after an assessment of the full risk profile of the particular patient. The question of whether to treat defects with cancellous bone graft or to fill them with cement has also not been adequately answered in the literature.
Functional outcomes and quality-of-life scores following hip resurfacing arthroplasty have been reported to be similar to those after conventional total hip replacement103. A prospective randomized trial suggested a slight increase in quality-of-life scores and activity levels in association with hip resurfacing arthroplasty107. Initial gait analyses of patients who had undergone hip resurfacing arthroplasty demonstrated confounding results, consisting of an increased peak abduction moment and decreased peak adduction moment108. These findings were thought to lead to increased stresses in the femoral neck and predispose the patient to component loosening and femoral neck fracture. In a more recent paper, Mont et al.109 suggested, on the basis of gait analysis, that the results of hip resurfacing arthroplasty were superior to those of conventional hip replacement (for example, walking was faster and hip kinematics were more normal). In contrast, a recently completed but as yet unpublished study performed by one of us (A.S.) and colleagues demonstrated a different outcome. In this study, gait and radiographic biomechanical analyses were used to compare hip resurfacing arthroplasty with conventional total hip replacement. It was concluded that a total hip replacement with accurate biomechanical reconstruction of the hip performed the same as a well-functioning hip resurfacing implant110. Some orthopaedic surgeon innovators of hip resurfacing implants as well as industry marketers have suggested that it is realistic for patients to return to high-impact sports. However, while the use of low-wearing surfaces such as metal-on-metal bearings, the bone-conserving nature of the operation, and the promising medium-term results do allow some optimism for young patients, caution should be used when advising patients about sustained high activity until such time that long-term results after high-activity levels become available.
Femoral Neck Fracture Fracture of the femoral neck is the most common complication of hip resurfacing arthroplasty, and it is unique to this type of hip arthroplasty. According to Steffen et al.82, the fracture rates reported by ten different surgeons who performed hip resurfacing arthroplasty ranged from 0% (in a study by Daniel et al.99, who analyzed 446 hips resurfaced with metal-on-metal implants) to 12% (in a study by Capello et al.111, who reported on sixty-eight hips resurfaced with metal-on-polyethylene implants). A multisurgeon national audit of the first 3429 metal-on-metal hip resurfacing arthroplasties performed in Australia over a four-year period demonstrated a fracture rate of 1.46% at a mean of 15.4 weeks (range, zero to fifty-six weeks) postoperatively112. Risk factors for femoral neck fracture include a combination of patient-associated, technique-associated, and postoperative factors. Patient-associated factors include gender and proximal femoral bone quality. The rate of fracture in women is twice that reported in men86,112. This high fracture rate may be a consequence of the reduced bone density in postmenopausal women or the increased risk of overpenetration of cement in osteoporotic bone. Technique-associated factors include notching of the superior part of the femoral neck (Figs. 3-A and 3-B) and varus femoral placement relative to the anatomical neck, which is consistent with the biomechanical principles suggested by Freeman81 and by one of us (P.E.B.) and colleagues90. With the femoral component in the varus position, there are increased tensile forces in the superior-lateral aspect of the femoral neck, increased shear stresses at the head-neck junction of the prosthesis, and increased compressive forces on bone that is likely to be weak under compression (Fig. 2-A)81. Intraoperative factors such as poor exposure, incomplete seating of the femoral component, and inaccurate direction of impaction on the implant may also contribute to fracture risk49,86.
In view of the prevalence of femoral neck fractures, a period of protected weight-bearing should be considered in some or all cases to allow the operatively machined femoral neck time to heal and remodel86. If any of the known risk factors (e.g., a large notch or incomplete seating) are identified intraoperatively, conversion to a conventional hip replacement should be considered intraoperatively. This requires appropriate preoperative patient counseling and informed consent. If a nondisplaced fracture does occur postoperatively, it may be treated successfully with nonoperative means—i.e., with a period of protected weight-bearing113. Displaced femoral neck fractures require conversion to a conventional total hip replacement. Two of us (P.C. and P.E.B.) and colleagues carried out extensive implant retrieval studies on ninety-eight metal-on-metal hip resurfacing prostheses, including twenty-eight that were associated with a fracture of the femoral neck49. The majority of these fractures had been noted within two months after the operation and had occurred through an area of active bone repair at the femoral neck-component junction. In contrast, in seven cases with an average time to fracture of fifteen months, a substantial proximal segment of the head was fully devascularized and necrotic and the fracture had occurred between the interface of the dead and viable segments of bone, within the area covered by the femoral component. Similar findings were reported by Morlock et al., in an analysis of fracture patterns and histological characteristics in association with fifty-five failures of femoral resurfacing48. Many of those failures were considered to have a dual-phase mode. The original trauma to the bone occurred at the time of the operation; healing was initiated, but the actual failure occurred several weeks or months later. This suggests that fractures that occur in the short-term may be related to patient selection or biomechanical or technical factors, whereas those occurring later may be associated with other factors such as impaired healing.
Component Loosening Early in the evolution of hip resurfacing arthroplasty, there was a general perception that the operative exposure and technique caused femoral head osteonecrosis and subsequent failure of the implant, although the histologically verified reported prevalence of this complication was generally low114,115. The majority of failures occurred as a result of the poor wear characteristics of the metal-polyethylene bearing combination5,11. Modern hip resurfacing arthroplasty has mostly been done through the posterior approach, and, while thirteen fractures in one series of 377 hip resurfacing arthroplasties were attributed to devascularization-induced weakening of the bone79, larger histological studies of hip resurfacing failures demonstrated mostly interfacial bone necrosis, appositional bone-healing, and generally good viability of the bone48,49. Insufficient or improper cement fixation was also an issue with the first generation of metal-on-polyethylene hip resurfacing arthroplasties. In one study of 170 resurfaced hips, at a median of ten years, larger femoral components (implying a larger fixation area) had significantly better survivorship than the smaller sizes (59% compared with 39%)116. Similarly, current-generation metal-on-metal hip resurfacing arthroplasties have had better short-term results in patients with a higher body mass index117 and in those without large femoral head defects118. In addition, the presence of femoral head defects can lead to a higher volume of cement being used, which may cause thermal necrosis of the surrounding bone and further undermine fixation of the component during the reparative phase49. Implant retrieval studies of resurfaced femoral heads have shown an extensive variation from the desired amount and distribution of cement48,49. For example, the total area of cement measured in sections from forty-five resurfaced femoral heads that had failed because of femoral neck fracture, femoral loosening, or a nonfemoral cause (for example, acetabular loosening or malposition) was found to range from 11% to 89%. There was significantly more cement in association with femoral components that failed because of loosening (50% of the head area) than there was in association with other modes of failure (for example, 36% of the head area in association with a femoral neck fracture and 39.5% in association with nonfemoral causes)49. Loosening because of insufficient cement penetration has also been reported100. Concerns about thermal necrosis with extensive cementation have been noted63, and histological studies have demonstrated that this does occur in hip resurfacing arthroplasties48,49, but healing of the bone without an intervening membrane has also been found, even with extensive cement usage49.
Metal Allergy and Hypersensitivity The periprosthetic tissues of patients who have had revision surgery because of suspected metal sensitivity are typically characterized by the presence of extensive perivascular or diffuse infiltrates of both B and T lymphocytes59,125,126. These may be found in conjunction with large areas of necrosis but typically without notable wear debris124. To distinguish these lesions from T-cell-dominated delayed-type hypersensitivity, the term ALVAL (aseptic lymphocytic vasculitis associated lesions) was introduced to describe these histological features59.
Femoral Impingement
Serum and Whole-Blood Metal Ions Typically, blood testing has been done for research, and use of metal ion testing to monitor the wear of metal-on-metal bearings is controversial. Jacobs et al.130 proposed that accurate monitoring of the parts per billion concentrations of metal ions in patients could possibly provide some insights into implant performance but that the methodology is still technically challenging and the interpretation of the results requires an extensive database with correlative clinical information. Such databases are still in the process of being developed, although more centers are gathering and reporting such data. In contrast, Clarke et al.36 suggested that, since rates of wear of metal-on-metal implants are undetectable with radiographic techniques, monitoring of metal ion levels will help to identify patients at risk for wear-related problems. Issues that present some confusion in the interpretation of reports of metal ion levels include the variability in the units of measurement and the specimens used (serum, whole blood, red blood cells, or urine). The majority of reported results are based on serum levels. Recently, Daniel et al.131 questioned the appropriateness of using serum levels and recommended using whole-blood levels as a surrogate measure of systemic exposure to metal ions.
Radiographic Monitoring
Radiographic Classification
Roentgen Stereophotogrammetric Analysis and Ein Bild Roentgen Analyse Until the radiographic changes indicative of component loosening that were noted above become apparent, studies of component migration may be useful to monitor the early clinical performance of implants. Two methods are available and have been validated for the evaluation of hip resurfacing prostheses. One is Ein Bild Roentgen Analyse-femoral component analysis, in which the center of the femoral head, the tip of the peg, and femoral landmarks are used to define reference points, which are then compared on a series of routine radiographic pairs in order to detect component migration135. The method relies on reproducible radiographs. In a group of twenty-eight hips followed for a minimum of 3.5 years, Ein Bild Roentgen Analyse detected migration of >2 mm earlier than the onset of clinical symptoms or radiographic evidence of failure135. With the other method, roentgen stereophotogrammetric analysis, implanted markers are used as reference points with which to assess migration of an implant relative to the femur; this method is reported to have an accuracy of 0.1 mm136. It was applied in two studies to assess early migration in patients who had hip resurfacing implants136,137. No femoral component migration was detected in either study. One limitation of roentgen stereophotogrammetric analysis is that, as a result of the high variability of bone quality in patients as well as potential selection bias, the number of hips studied has been quite small and may not be representative of all patients undergoing hip resurfacing arthroplasty. Conversely, Ein Bild Roentgen Analyse permits the analysis of a greater number of hips but is not as accurate as roentgen stereophotogrammetric analysis.
Intuitively, the conversion of a hip resurfacing arthroplasty to a conventional total hip replacement by replacing the failed femoral component should be relatively straightforward. The availability of large-diameter cobalt-chromium femoral total hip bearings allows the possibility of retaining the primary resurfacing acetabular implant—for example, in a patient with a femoral neck fracture but no major acetabular defects or malposition of the acetabular component (Fig. 5). Follow-up reports support the conclusion that such conversions can provide a good clinical outcome. For example, one of us (P.E.B.) and colleagues reported a survivorship of 94% at ten years after the primary socket was retained in a group of ninety converted hip resurfacing arthroplasties138. Ball et al. recently reported the results of a match-paired analysis of twenty-one conversions of failed hip resurfacing arthroplasties, in eighteen of which the socket was retained, and fifty-eight primary cementless total hip arthroplasties done during the same period27. At a mean of forty-six months, the functional outcome hip scores were comparable and no other differences were found between the two groups.
When both the acetabular and the femoral component must be revised, a conventional total hip replacement with any of the current bearing combinations could be implanted, except in patients with suspected metal sensitivity, in whom bearing surfaces other than metal-on-metal should be used59.
Considering the poor clinical history of metal-on-polyethylene resurfacing arthroplasties and the concerns over the unknown risks of long-term exposure to metal-on-metal bearings, some surgeons are skeptical about the widespread reintroduction of metal-on-metal hip resurfacing arthroplasty. A recent paper by Lachiewicz139 summarizes these concerns in light of the success of many modern conventional total hip implants and the learning curve involved for surgeons who perform only a few hip resurfacing arthroplasties each year. It also raises the issue of osteoporotic periprosthetic fractures many years after implantation, a problem not addressed in the literature but known to occur with conventional hip replacements140. The lack of long-term follow-up of the new generation of hip resurfacing prostheses leaves many questions unanswered, and surgeons need to be aware that the proponents of hip resurfacing arthroplasty emphasize that it is a technically more demanding operation.
Despite these concerns, we believe that it is reasonable to consider hip resurfacing arthroplasty for younger people. Available information suggests that the twenty-year survival of conventional total hip prostheses is poor. When one considers that the average age of patients who undergo hip arthroplasty is decreasing, and if the current survival rates of hip resurfacing implants continue to be encouraging, younger patients may benefit from simpler revision options with less morbidity due to the bone conservation at the time of the hip resurfacing arthroplasty. Although cement fixation is widely used, there is interest in cementless femoral fixation, particularly for patients returning to high levels of impact activities. Retrieval analysis revealing variable degrees of cement penetration implies difficulty in achieving a consistent cement technique49. Previous experience with the use of cementless femoral components in metal-on-polyethylene resurfacing arthroplasties demonstrated good bone ingrowth fixation in patients and in animal models despite extensive wear debris-induced osteolysis11,141,142. More recently, cementless femoral components have been used in metal-on-metal hip resurfacing arthroplasties. We are aware of only one report describing the results of this procedure, in seventy hips (treated through a posterolateral approach) followed for a mean of 28.5 months133. There were no femoral component-related failures.
While metal ion release from metal-on-metal bearings remains a concern, in terms of its possible effects both locally and systemically, the thin (2 to 4-mm-thick) acetabular components currently in use can only be made from metal, although future material developments may allow hip resurfacing implants to be manufactured from ceramic or other composite materials. Finally, just as modularity has evolved in conventional total hip replacements, it may be possible to achieve similar benefits in hip resurfacing arthroplasty as new materials and designs evolve.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
History of Hip Resurfacing...
3, the four-year survivorship rates were 89% and 97%, respectively. This suggests that patients with a higher risk index were 4.2 times more likely to require revision at four years. 
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |