Revision of Metal-on-Metal Bearing Surfaces in Hip Arthroplasty
A 60-year-old woman, who had left primary metal-on-metal hip resurfacing 3.5 years earlier for severe osteoarthritis, presented with increasing left hip pain, clicking, and swelling over the preceding several months. Her postoperative course had been uncomplicated, and she had excellent pain relief for more than 3 years. She denied fevers, shakes, or chills. On physical examination, she was found to have a normal gait and equal leg lengths. She was neurovascularly intact but had a painful range of motion of her hip and audible squeaking.
Chromium and cobalt metal ion levels were elevated to 65 and 53 ppb, respectively. Results for the erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, and white blood cell (WBC) count were within normal limits. Plain radiographs showed a left total hip resurfacing in an unchanged position with no evidence of loosening. Computed tomography revealed a 1.6 × 1.4 cm lytic lesion in the supraacetabular region of the left ilium, which was consistent with a pseudotumor ( Figs. 73A.1 and 73A.2 ).
The patient subsequently underwent resurfacing revision with conversion to a ceramic-on-polyethylene total hip arthroplasty ( Figs. 73A.3 to 73A.11 ) using a posterior Kocher-Langenbeck approach. On opening the hip capsule, 50 to 60 mL of grayish fluid was released. An extensive pseudocapsule with the grayish staining of metallosis was observed in the posterior and anteroinferior aspects of the hip joint. This was carefully debulked to maintain as much viable tissue as possible while completely excising the damaged tissue. A femoral neck osteotomy was made around the stem of the size 42 femoral resurfacing component. Next, the well-fixed, well-positioned, 48-mm resurfacing cup was removed, and a 52-mm, press-fit trabecular metal acetabular shell with a polyethylene liner was placed. The femoral resurfacing component was converted to a standard single-taper, proximally fixed femoral component with a modular ceramic head.
Histologic analysis of the tissue sampled intraoperatively revealed dense fibrous tissue with reactive changes, chronic inflammation, and particulate-laden macrophages consistent with an adverse reaction to metal debris. The patient was allowed to bear weight as tolerated and had an uncomplicated postoperative course. At the 3-month follow-up evaluation, she had a slight Trendelenburg gait, no instability, and a good range of motion.
In this chapter, we focus on the complications associated with the use of metal-on-metal (MoM) bearing surfaces in hip arthroplasty, the workup of symptomatic and asymptomatic individuals with these prostheses, and techniques used in revision or conversion that are specific to MoM bearings.
Although popular media have focused on the adverse effects of metal ions, most revisions performed on MoM prostheses are not for adverse reactions to metal debris (ARMD) but rather for mechanical problems such as aseptic loosening and fractures.
Common causes of failure should be excluded before testing for ARMD, but physicians should remain alert to the possibility of these tissue reactions due to the high degree of associated morbidity.
ARMD should be considered in any symptomatic patient presenting with unexplained pain or mechanical symptoms.
The use of a clinical evaluation algorithm, early revision, larger revision femoral heads, and patient education can reduce the rate of complications and repeat revisions.
Advanced imaging to look for fluid collection seems a more useful tool than obtaining ion levels at this time.
In revision for ARMD, the bearing surface should be revised to a non-MoM bearing to prevent recurrence of ARMD or continued symptoms.
If infection cannot be ruled out preoperatively, a two-staged revision should be planned with intraoperative tissue sampling to help distinguish between infection and ARMD. Only after infection is ruled out should a single-stage revision for ARMD be considered.
In revision for pseudotumors, especially those infiltrating periprosthetic vessels and nerves, a multidisciplinary team, including a vascular or plastic surgeon, may need to be assembled to remove the entire lesion.
In cases of ARMD, especially pseudotumors, the surgical site should be thoroughly cleared of all metal debris and nonviable tissue to prevent recurrence of pseudotumors or other ARMD. Débridement of the surgical site of metal debris should be done with care to preserve as much muscle and soft tissue as possible to maintain hip stability and avoid dislocation.
Metal ion testing should not be used alone when deciding to proceed to surgery or to exclude the diagnosis of ARMD. There is no metal ion threshold level above which revision is indicated, and ARMD may occur in the absence of elevated metal ion levels.
Revision of both components leads to better outcomes and fewer complications than a single-sided revision. If the viability of either component is in question, both components should be revised.
The first metal-on-metal total hip arthroplasty (MoMTHA) was performed in 1937 by Philip Wiles. He used stainless steel components affixed to bone with bolts and screws. Use of modern MoM hip prostheses began in the late 1950s with the McKee-Farrar prostheses. After loosening led to early failure of stainless steel and press-fit models, McKee eventually used a cobalt–chromium–molybdenum alloy for the acetabular component combined with a modified Thompson femoral stem; both were fixed with acrylic cement. Ring and others followed this concept, but poor early and midterm survival due to loosening led to the abandonment of MoM in favor of Charnley’s metal-on-polyethylene, low-friction arthroplasty.
The first MoM prostheses had dichotomous survivorship, with implants failing early or having excellent survival, often lasting more than 20 years. When Weber revisited the MoM implant in 1996, he found no adverse effects from wear in a series of 110 MoM hip implants. Another study by Jacobsson and colleagues found that the McKee-Farrar prosthesis had long-term survival rates comparable to those of Charnley THAs.
These findings and the appeal of prostheses with better wear characteristics than polyethylene generated renewed interest in MoM bearings. Improved manufacturing and tribologic characteristics in the new generation of MoM implants have helped their resurgence. However, concerns about unexplained pain and early failure as a result of adverse reactions to metal debris (ARMD) again led to declining use, with exceptions for selected populations such as young men with appropriate anatomy in hip resurfacing. Registry data from Australia and the United Kingdom have shown that MoM hip replacements have higher revision rates than conventional bearing surfaces. With increasing numbers of MoM hip bearings being implanted and revised, issues related to the proper workup, diagnosis, and surgical technique have become all the more salient.
Indications for Revision
Malpositioning, Fractures, and Loosening
Controversy about the use of MoM bearings has focused on ARMD leading to early failure, unexplained pain, and pseudotumor formation. Although the popular media have focused on these adverse reactions, most revisions performed in metal-on-metal hip resurfacing (MoMHR) cases are for mechanical issues such as malpositioning, aseptic loosening, and fracture.
The indications for revision in MoMTHAs largely coincide with those for THAs of other bearing surfaces, with the exception of a portion of MoMTHAs that fail due to the effects of ARMD. For MoMHR, the distribution of revisions due to various causes is slightly different from those of traditional THAs or MoMTHAs ( Table 73A.1 ). The primary mode of failure, which is unique to MoMHR, is periprosthetic fracture of the femoral neck. Fractures usually occur early, often in the early postoperative years. Risk factors for femoral neck fracture include smaller femoral sizes, head cysts, excessive or inadequate cement penetration, notching of the femoral neck, osteopenia, and varus placement of the femoral component ( Box 73A.1 ).
|Causes of Revision||Combined MoMHR and MoMTHA ∗||MoMTHA||MoMHR|
|Ebramzadeh et al, 2011 |
|Browne et al,2010 |
|Milosev et al, 2006 |
|Porat et al,2012 |
|Australian National Joint Registry, 2011||DeSmet et al, 2011 |
|Carrothers et al, 2010 |
|Acetabular loosening||26.9%||25% (unsp)||32.3%||49.2% (unsp)||33.4% (unsp)||27.4% (unsp)||17.6%|
|Both components loosened||11.7%||Both 2.7%|
|Sepsis or infection||6.1%||19%||17.6%||15%||8.2%||3.5%||9.3%|
|Suspected metal allergy or ARMD||5.3%||27%||26.1%||7.1%||5.3%||8.2% (pain)|
|Dislocation or instability||5%||2.9%||1.5%||2.7%||.9%||2.7%|
|Other||2.6%||1% (THA fx)||5.8% (1 cup fx after trauma, 1 metal inlay dissociation from polyethylene liner)||7.7%||1.1%||4.4% (3.5% due to elevated metal ion levels alone)|
Smaller femoral sizes
Excessive or inadequate cement penetration
Notching of the femoral neck
Varus alignment of the femoral component
Other complications of MoMHR include loosening of the femoral component, impingement, and avascular necrosis of the femoral head. Loosening of the MoMHR femoral component is often a result of improper sizing during the primary procedure. This can be particularly challenging in smaller femurs. After the acetabular component is placed, it dictates what femoral component sizes can be used. For smaller femoral heads, small deviations from the center of the head while drilling can lead to gaps between the femoral bone and resurfacing component, leading to increased rates of loosening.
Adverse Reactions to Metal Debris
Although reports in the literature state the incidence of ARMD as only 1% to 3% in MoMHRs, these tissue reactions are a focus of this chapter because they are not often seen as a cause of revision with other bearing surfaces and are associated with a high degree of morbidity. The lesions most often referenced in relation to MoM bearings include metallosis, osteolysis, aseptic lymphocyte-dominated vasculitis-associated lesions (ALVALs), and pseudotumors. Metallosis is defined as macroscopic staining, aseptic fibrosis, and necrosis of periprosthetic tissues as a consequence of abnormal wear of the prosthesis. Pseudotumors are usually symptomatic, noninfected cystic or solid soft tissue masses related to the MoM implant that can lead to considerable tissue necrosis and often show ALVALs on histologic examnation. ARMD and adverse local tissue responses are terms frequently used to describe the range of these lesions.
There is inconsistency in the literature about the cause of these lesions, but evidence points to two separate but possibly overlapping causes: a cytotoxic local tissue effect and a hypersensitivity reaction. Cytotoxicity, the result of a foreign body reaction due to increased levels of metal ion particles created through increased wear or corrosion, is suggested by the histologic finding of metal debris–laden macrophages in periarticular tissue, which is often seen in combination with inflammation and surface ulcerations. Wear and increased metal ion levels correlate with unexplained pain, pseudotumors, ALVALs, osteolysis, metallosis, and tissue necrosis, further supporting a foreign body reaction as the cause of ARMD.
Although cytotoxicity related to excess metal debris likely accounts for most ARMD cases, the cause in some patients may instead be hypersensitivity to metal ions. Hypersensitivity as a cause of these reactions was demonstrated in a small group of patients who underwent revision for unexplained groin pain and ARMD but did not have metal ion levels higher than baseline controls, had properly positioned femoral and acetabular components, and did not show signs of excess component wear. These patients likely had a true hypersensitivity reaction to metal debris.
Baseline metal ion levels for patients with well-functioning MoM hips are elevated above those of normal controls for the life of the prostheses, raising concerns about systemic problems. Case reports of neurologic and cardiac problems related to MoM implants confirm that systemic effects can occur; and systemic toxicity should be considered a possible indication for revision. At this time the risk of carcinogenesis has not be considered an indication for revision because no increased risk has been seen in long-term follow-up of first-generation prostheses, even though genotoxicity is seen with exposure to cobalt and chromium found in orthopedic implants.
Teratogenicity is a theoretical concern because cobalt and chromium metal ions can cross the placenta, but it is unlikely in patients with metal ion levels in the normal range for MoM hips. Pregnancy therefore should not be considered an indication for revision, although it is recommended that becoming pregnant be delayed until 2 years after implantation to avoid the running-in phase of metal ion concentrations. Current recommendations are to avoid using MoM bearings in women of childbearing age.
Examination and Imaging
Workup of the painful or problematic MoM hip should initially be conducted in a manner similar to that for other bearing surfaces. Although a high index of suspicion should be maintained for ARMD, the complications commonly causing failure in hip replacements, such as loosening and infection, should be ruled out before evaluation for ARMD.
Evaluation should begin with a thorough history and physical examination. As detailed by Browne and colleagues, “The clinical presentation of patients with an adverse metal reaction can be varied and often nonspecific.” The most common presenting symptom of patients needing revision for ARMD is groin pain, followed by mechanical symptoms such as clicking or clunking. An associated local swelling, mass, or rash may be identified during physical examination. The groin pain is often exacerbated by weight-bearing activities and resisted straight leg raises. Patients may also complain of difficulty climbing stairs or rising from a seated position because of weakness in hip flexion and abduction due to pain.
Patients should be asked about rashes (e.g., in response to jewelry), an irritable hip swelling or mass, and problems with wound healing, because these findings may indicate local hypersensitivity or metal cytotoxicity. The general health of the patient should be assessed to elucidate systemic symptoms. In a review by Keegan and co-workers, possible signs of systemic toxicity included tinnitus, vertigo, deafness, blindness, optic nerve atrophy, convulsions, headaches, peripheral neuropathy, cardiomyopathy, hypothyroidism, and polycythemia. The inflammatory element of ARMD may mimic a low-grade infection with systemic symptoms.
Plain radiographs should be obtained to identify fixation of components, confirm lack of infection, and estimate bone stock. An anteroposterior radiograph and a true cross-table lateral radiograph can identify malpositioning of the acetabular component. Extraarticular impingement can lead to pain in an MoM hip, especially in large-diameter MoMTHAs, and radiographs should be examined for extraarticular impingement such as a femur against the anterior inferior iliac spine or iliopectineal eminence. Lucent lines seen in the periprosthetic region can indicate aseptic loosening or infection around the MoM joint. Plain radiographs can also be used to identify femoral neck fractures in MoMHRs.
Plain radiographs are not as useful in identifying soft tissue lesions, and large-diameter femoral heads may obscure visualization, making determination of component orientation difficult. Computed tomography (CT) and magnetic resonance imaging (MRI) are better options for evaluating soft tissue lesions and component positioning. Metal artifact reduction series (MARS) MRI is an excellent way to identify soft tissue changes and mass lesions when radiographs of a symptomatic patient are normal. However, this modality may miss small lesions near the prostheses due to metal artifacts, and ultrasound is an option for their detection. As suggested by Haddad and colleagues, ultrasound allows identification of the masses as solid or cystic and can be used as a guide to aspirate these masses for histologic sampling.
Blood should be drawn for the white blood cell (WBC) count with differential, C-reactive protein (CRP) level, and erythrocyte sedimentation rate (ESR) to rule out infection. No ESR and CRP standardized ranges exist for patients suffering from ARMD. The literature reports inconsistent data in terms of the relation of ESR and CRP values to ARMD, often showing no correlation with severity of local tissue reactions. In assessing the MoM hip, the surgeon should be wary of making a preoperative diagnosis of infection because the sterile soft tissue inflammation of ARMD can lead to elevations in ESR and CRP values. Pandit and colleagues reported three cases of pseudotumors that showed preoperatively elevated ESR and CRP levels but had negative culture results at surgery.
In all cases, elevated WBC, ESR, and CRP values should prompt aspiration of the hip joint for evaluation of infection. The synovial fluid should be cultured along with sensitivity determinations. The WBC count with differential may show a predominance of polymorphonuclear cells (PMNs) in infection (>70%) as opposed to lymphocytosis (up to 40%) in a hypersensitivity response. A predominance of monocytes in the setting of an increased cell count may indicate cytotoxicity as a result of increased wear or hypersensitivity, and it should alert the clinician to the possibility of prosthetic failure due to ARMD. Anecdotal evidence indicates that metal debris can confound automated cell counts, leading to falsely elevated synovial fluid WBC counts, and a manual cell count may be needed to clarify elevated values.
If aspiration is not helpful for either diagnosis, Biant and colleagues have suggested hip arthroscopy to better visualize the hip and take biopsy specimens, allowing a more definitive preoperative diagnosis. Bone scintigraphy can also be used to achieve better insight if preoperative differentiation of infection from ARMD remains elusive. Technetium-99m sulfur colloid imaging techniques have shown a 70% sensitivity and 100% specificity for diagnosing infection. Labeled PMNs in bone show a heterogeneous distribution in cases of infection when they are the main cellular infiltrate. Component loosening correlates with an infiltration of 99m Tc-labeled histiocytes without any colloid-labeled PMNs. In a hypersensitivity-type response in which lymphocytes are the main cellular infiltrate, a homogeneous distribution of 99m Tc and colloid results because lymphocytes are not labeled by either.
Complicating the differential diagnosis of infection versus ARMD are reports of purulent-appearing fluid found intraoperatively during revision for presumed infection for which the final cultures turned out to be negative and tissue pathology was consistent with ARMD. There appears to be no pathognomonic intraoperative findings for ARMD, although Browne and colleagues reported commonly observing metallosis, a shaggy synovium, effusions, and gross damage or wear of components. The only way to differentiate infection from ARMD is intraoperative tissue sampling for histology and cultures, which should be performed whenever infection or ARMD is suspected. ARMD often shows widespread infiltration of histiocytes containing black particles that are consistent with metal particulate debris and areas of necrosis. Viable tissue often exhibits the characteristics of ALVALs. However, the sensitivity and specificity of these histologic changes for ARMD is unknown, and a presumptive diagnosis of ARMD should be made only after infection is ruled out.
Certain populations of patients benefit from metal ion testing to aid in the diagnosis of ARMD. Metal ion levels peak in the first 12 months postoperatively and remain elevated throughout the lifetime of the prosthesis. Well-functioning MoM hips have baseline levels around 2 ppb after unilateral and 3 ppb after bilateral procedures. Poorly functioning MoM hip replacements often have metal ion levels elevated above these levels. For example, Hart and colleagues demonstrated that patients with unexplained painful MoM hip prostheses had double the cobalt and chromium levels of patients with well-functioning hips. As it has become apparent that higher metal ion levels correlate with an increased risk of having or developing ARMD, screening for ARMD using cobalt and chromium levels has become routine.
The gross features of metallosis occur at blood metal ion levels greater than 20 ppb, prompting some to use this as a cut-off value above which revision is indicated. However, testing remains controversial because there is no consensus regarding screening and cut-off values for serum or blood cobalt and chromium concentrations in patients with MoM hips. Moreover, there is no standardized testing protocol, and the means of testing vary between institutions. An extensive institutional database is needed to appropriately interpret serum metal ion levels, and the use of a single laboratory is recommended.
Despite the relative nature of values, metal ion testing is a valid tool when combined with clinical suspicion and other diagnostic tools. Ion level measurements can be obtained from various blood fractions, and the results correlate. However, measurement of whole blood levels of cobalt and chromium is thought to provide a better estimate of systemic metal ion exposure and to be more reliable than measurement of levels in plasma, serum, or red blood cells alone. Inductively coupled plasma-mass spectrometry (ICPMS) has been validated, and the analytic technique is very sensitive for the detection of ions. This method of measurement is not available at all institutions, and in these cases, atomic absorption spectrometry (AAS), which has also been validated but is not as accurate or reliable as ICPMS, can be used.
Although guidelines from the U.S. Food and Drug Administration (FDA) are still pending, the Medicines and Healthcare Products Regulatory Agency (MHRA) of the United Kingdom has set guidelines for testing based on research by Hart and colleagues. Current MHRA recommendations are that all symptomatic patients should receive metal ion testing of cobalt or chromium along with repeat testing of ion levels in 3 months if either is more than 7 ppb. For asymptomatic patients, the value of screening is debatable. However, given reports of asymptomatic soft tissue masses leading to severe soft tissue destruction and extensive soft tissue necrosis seen in otherwise asymptomatic patients, ion testing is recommended for all patients with risk factors for ARMD ( Box 73A.2 ). For asymptomatic patients with no risk factors, screening is unnecessary unless the physician would like to obtain baseline values for future reference. If ion levels are higher than 7 ppb, the MHRA recommends retesting in 3 months to trend levels. Further workup with advanced imaging is also advised.