- Email: [email protected]
Are Metal Ion Levels a Useful Trigger for Surgical Intervention?
The Journal of Arthroplasty Vol. 27 No. 8 Suppl. 1 2012
Are Metal Ion Levels a Useful Trigger for Surgical Intervention? William L. Griffin, MD,* Thomas K. Fehring, MD,* James C. Kudrna, MD,y Robert H. Schmidt, MD,z Michael J. Christie, MD,§ Susan M. Odum, MEd,‖ and Anne C. Dennos, BS‖
Abstract: The purpose of this study was to determine if cobalt and chromium ion levels can predict soft tissue damage at total hip revision. This study included 90 metal-on-metal total hip patients with preoperative cobalt and chromium ion levels. Tissue damage noted at revision surgery was graded on a 4-point scale. Sensitivity, specificity, and predictive values were calculated for various threshold values. Receiver operating characteristic analysis was conducted. Using 7 ppb as a threshold, cobalt and chromium ion levels had poor sensitivity and specificity (Co, 65% and 56%; Cr, 29% and 75%). Positive predictive values for cobalt and chromium were only 48% and 26% respectively. The area under the curve was 0.37 for cobalt and 0.44 for chromium. The length of time to revision significantly correlated with tissue damage (P = .001). Ion levels are unreliable predictors of periarticular soft tissue damage and should not be used in isolation as surgical intervention triggers. Keywords: metal ion levels, total hip revision, cobalt, chromium. © 2012 Elsevier Inc. All rights reserved.
surgical intervention [5]. The medical device alert from the British Orthopedic Association concerning MOM implants chose cobalt and chromium ion levels of 7 ppb as a threshold for concern [6]. Because cobalt (Co) and chromium (Cr) are essential elements, there is an effective renal clearance mechanism for these metals. Wear at the MOM articulation produces nonsoluble metal debris as well as soluble metal ions that can be measured in the blood [7]. These ion levels present in the blood can come from several sources: wear at the articulation, corrosion at any taper junction, abrasion of loose components, and normal body stores [1]. The blood levels of these ions represent a balance between ion production from the device and renal excretion, and these levels can be affected by changes in activity levels as well as renal function. Based on current recommendations, surgeons may be compelled to revise well-functioning MOM implants based solely on a metal ion level of greater than 7 ppb [6]. In other reports, patients with metal ion levels well below 7 ppb were found to have significant tissue damage at the time of surgery [8]. Currently, there are no evidence-based data to support the use of a specific metal ion threshold as an indicator of periarticular tissue damage and thus a trigger for surgical intervention. The purpose of this study is to determine if cobalt and chromium ion levels can predict soft tissue damage at the time of MOM revision.
Metal-on-metal (MOM) hip articulations were thought to be a potential solution for wear-related failures in total hip arthroplasties while also improving stability through the use of larger diameter femoral heads. Although MOM bearings produce considerably less wear than metal-on-poly bearings, the debris generated can lead to adverse local tissue reactions and systemic effects from prolonged exposure to metal ions [1-3]. The clinical evaluation and treatment of patients presenting with symptomatic MOM bearing implants are difficult. It is important to have reliable tests to help determine when to intervene surgically to prevent tissue damage. Ultrasound and magnetic resonance imaging with metal artifact reduction sequences have been used to assess periarticular reactions secondary to metal wear debris [4]. In addition, it has been suggested that metal ion levels may have prognostic value in determining the presence of articular wear and the timing or need for
From the *OrthoCarolina Hip and Knee Center, Charlotte, North Carolina; yIllinois Bone and Joint Institute, Glenview, Illinois; zTexas Hip and Knee Center, Fort Worth, Texas; §Southern Joint Replacement Institute, Nashville, Tennessee; and ‖OrthoCarolina Research Institute, Charlotte, North Carolina. Submitted January 13, 2012; accepted March 13, 2012. The Conflict of Interest statement associated with this article can be found at doi:10.1016/j.arth.2012.03.020. Reprint requests: William L. Griffin, MD, OrthoCarolina Hip and Knee Center, 2001 Vail Ave, Suite 200-A, Charlotte, NC 28207. © 2012 Elsevier Inc. All rights reserved. 0883-5403/2708-0008$36.00/0 doi:10.1016/j.arth.2012.03.020
32
Metal Ion Levels for Surgical Intervention Griffin et al
Materials and Methods We conducted a multicenter retrospective review of a series of MOM total hips that were revised. All 4 participating centers obtained institutional review board approval. Cases were identified by queries of institutional prospective total joint registries and practice management systems. Electronic health records and registry data were reviewed to document dates of primary total hip surgery and revision total hip surgery to determine time in situ. In addition, reason for revision and implant details from the primary surgery were recorded. Whole blood cobalt and chromium ion levels were obtained from LabCorp and Nation Medical Services using an Inductively Coupled Mass Spectrometry (ICPMS) technique on whole blood and reported in micrograms per liter (1 μg/L = 1 ppb). In the event that there was more than 1 ion level workup, the result closest to the revision date was selected. Operative notes were reviewed by the senior author (WLG) to evaluate the severity of metallosis and/or tissue damage at the time of the revision surgery. The severity of tissue damage was categorized using the following tissue grade scale, developed by the authors: grade 0, no metallosis; grade 1, metallic staining, intracapsular fluid; grade 2, extracapsular fluid; grade 3, extracapsular tissue necrosis. One hundred sixteen symptomatic MOM total hip revision cases with complete cobalt and chromium ion levels were included. Four patients who underwent bilateral MOM total hip revision (8 hips) and 15 patients with an unrevised MOM hip on the contralateral side (15 hips) were excluded. In addition, 4 patients (4 hips) were excluded because of insufficient data to evaluate tissue damage. This left a total of 89 symptomatic MOM revision cases available for this study. Of these 89 cases, 84 were monoblock implants (82 DePuy ASR TM [Warsaw, Ind], 1 Conserve R [Wright Medical Group Inc., Memphis, Tenn], and 1 Magnum TM [Biomet, Warsaw, Ind]), and 5 were modular implants (Pinnacle R, DePuy Orthopedics, Warsaw, Ind). Reasons for revision are documented in Table 1. Standard descriptive statistics were calculated. The association between ion level and tissue grade was determined by a Spearman correlation coefficient. Oneway analysis of variance was used to determine any differences between ion levels, tissue grade, and time in situ. Using 7 ppb as the cutoff, sensitivity, specificity,
Table 1. Frequency of Revision Diagnoses Reason for Revision
Frequency
Aseptic loosening Suspected metallosis Pain Noise
17 30 42 10
There was overlap with the reasons for revision.
33
positive predictive value, and negative predictive value were calculated to assess the value of both cobalt and chromium ion levels as diagnostic, predictive tools for metallosis. To evaluate optimal threshold values of metal ion levels, a receiver operating characteristic analysis was conducted. All statistical analyses were performed using SAS 9.2.1 (Cary, NC).
Results There was a profound variation in the whole blood metal ion levels. The mean cobalt ion level in the study population was 23 (0-236), and the mean chromium ion level was 9 (0-86). Of the 89 patients, 23 were classified as grade 0, 38 as grade 1, 19 as grade 2, and 9 as grade 3. There was little to no correlation between ion levels and tissue damage. The correlation coefficient for cobalt and tissue grade was 0.192 (P = .073) and 0.162 (P = .132) for chromium and tissue grade. There were no statistical differences in the mean cobalt (P = .64) or chromium (P = .41) ion levels between the 4 tissue grades. There was also no incremental increase, as the severity of tissue damage increased. In fact, the mean cobalt levels for grades 0 (17.41 ppb) and 2 (15.69 ppb) were similar, whereas mean cobalt levels for grades 1 (27.92 ppb) and 3 (26.87 ppb) were similar. For chromium, the mean ion level was highest for grade 1. See Table 2 for metal ion levels by each tissue grade. There were significant (P = .0013) differences in the time in situ of the MOM hip between the 4 tissue grades. Furthermore, the severity of tissue damage increased as the time in situ increased. These results indicate that there may be a time-dose response relationship. When ion levels and time in situ are combined, this correlation is lost. This suggests that the tissue damage seen at revision surgery may be related to the length of time of metal ion exposure but not the exposure dose. See Table 2 for time in situ for each tissue grade. Using 7 ppb as the cutoff value, the diagnostic value of both cobalt and chromium ion levels indicate poor accuracy of these tests as a diagnostic or predictive tool. The sensitivity of 7 ppb as a predictor of patients with moderate to severe tissue damage was 61% for cobalt and 31% for chromium. The specificity or the proportion of patients without moderate to severe tissue damage was 58% for cobalt and 76% for chromium. The positive predictive value or the proportion of patients with tissue damage and cobalt ion levels greater than 7 ppb was only 39.5% and 37.5% for chromium. The negative predictive values were 60.5% and 62.5% for cobalt and chromium, respectively. These values did not improve when using alternate cutoff values ranging from 7 to 50 ppb (Table 3). The receiver operating characteristic curve indicates that the area under curve for cobalt was 0.37 (Fig.). In other words, there is only a 37% probability that cobalt ion levels will be higher in patients with moderate to severe
34 The Journal of Arthroplasty Vol. 27 No. 8 Suppl. 1 September 2012 Table 2. Sensitivity and Specificity of Cobalt and Chromium Metal Ion Levels Cobalt Threshold
Sensitivity
7 ppb 10 ppb 13 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.61 0.39 0.36 0.32 0.32 0.25 0.21 0.14
95% CI 0.41-0.78 0.22-0.59 0.19-0.56 0.17-0.52 0.17-0.52 0.11-0.45 0.09-0.41 0.05-0.33
Specificity 0.58 0.63 0.68 0.7 0.73 0.75 0.81 0.87
95% CI 0.45-0.70 0.49-0.75 0.55-0.79 0.57-0.81 0.60-0.83 0.63-0.85 0.68-0.89 0.76-0.94
Chromium Threshold
Sensitivity
7 ppb 10 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.31 0.24 0.14 0.32 0.07 0.03 0
95% CI 0.16-0.51 0.11-0.44 0.05-0.33 0.17-0.52 0.01-0.24 0.001-0.20 0.0-0.15
Specificity 0.76 0.81 0.82 0.73 0.85 0.85 0.95
95% CI 0.63-0.85 0.68-0.89 0.70-0.90 0.60-0.83 0.74-0.93 0.74-0.93 0.86-0.99
Cobalt Threshold
PPV
95% CI
NPV
95% CI
7 ppb 10 ppb 13 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.4 0.32 0.33 0.33 0.34 0.32 0.33 0.33
0.25-0.56 0.18-0.51 0.18-0.53 0.17-0.54 0.18-0.56 0.15-0.55 0.14-0.59 0.11-0.65
0.6 0.68 0.67 0.67 0.65 0.68 0.67 0.67
0.44-0.75 0.49-0.82 0.47-0.82 0.46-0.83 0.44-0.82 0.45-0.85 0.41-0.86 0.35-0.89
Chromium Threshold
PPV
95% CI
NPV
95% CI
7 ppb 10 ppb 13 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.38 0.37 0.27 0.35 0.18 0.1 0.0
0.20-0.59 0.17-0.61 0.09-0.55 0.18-0.56 0.03-0.52 0.005-0.46 0.0-0.69
0.63 0.63 0.73 0.65 0.82 0.9 1.0
0.41-0.80 0.38-0.83 0.45-0.91 0.44-0.82 0.48-0.97 0.54-0.99 0.31-1.0
Abbreviations: CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.
tissue damage as compared with those who have none to minimal tissue damage. The area under the curve for chromium was slightly higher at 0.44.
Discussion There are a number of factors that can contribute to the increased wear of MOM bearings and the generation of Co and Cr ions. These include manufacturing and design variables, surgical technique, and patient variables. In terms of the manufacturing techniques, evaluation of the original MOM designs and bench top studies has identified design parameters that can improve the wear of MOM bearings. The hardness of the CrCo material has been improved with the use of a
Fig. Receiver operating characteristic curve for cobalt and chromium metal ion levels.
higher carbon content and forging of the material, both of which have decreased wear rates. Improving the roundness and smoothness and keeping the tolerances between 80 and 120 μm have also improved the wear characteristics [9,10]. The design of the implant can also lead to higher failure rates. Wear characteristics vary across MOM bearings, which has been particularly evident with 1piece designs. Because the 1-piece shells were originally intended for resurfacing arthroplasties, there are a number of design compromises. To accommodate the native femoral neck and avoid impingement, 1-piece shells are less than a full 180° hemisphere. By reducing valuable articular surface at the rim, the shell is more prone to edge loading and runaway wear. In an attempt to maximize head size and minimize acetabular bone resection, the resurfacing shells were made relatively thin. Thinner shells can deform during insertion with a press-fit technique and interfere with the tight tolerances of an MOM articulation [11]. In an attempt to increase the stiffness of these shells, the manufacturers thickened the dome region of the cup. The amount of this thickening varies by manufacturer and cup size and has the unintended consequence of moving the center of rotation further outside the shell, again decreasing the articular surface available to the head. A few of the 1Table 3. Difference in Metal Ion Levels Mean Time In Situ (Months) by Tissue Grade (Analysis of Variance) Tissue Grade
Cobalt
P
0 1 2 3
17.41 27.92 15.69 26.87
.644
Chromium 6.06 12.7 6.13 7.26
P .41
Time in Situ 29.58 33.55 44.01 46.54
P .0013
Metal Ion Levels for Surgical Intervention Griffin et al
piece designs also added a cut-out at the rim to allow for an insertional device, removing even more articular surface. When combined, the net effect of these design features makes the 1-piece shells very susceptible to edge loading and increased wear. With this loss of articular surface, a cup placed at 50° abduction may behave like a cup inserted at 63° of abduction [12]. Numerous studies have identified that increased cup abduction and cup anteversion angles can lead to higher wear rates. A less appreciated factor contributing to failure is excessive anteversion. Women in particular may have more combined anteversion. The femur may have significant native anteversion. This in combination with acetabular version may lead to edge loading and aberrant wear in extension during ambulation. Fisher [13] noted a 10-fold increase in wear with a 55° abduction angle, and Morlock et al [14] reported a 27fold increase in wear with 59° of abduction. De Haan et al [15] postulated that when the weight-bearing vector is within 10 mm of the cup's rim, the MOM bearing loses its fluid film lubrication and boundary wear ensues with higher wear rates. This may explain why women with smaller cups and smaller articular surfaces have been more prone to failure secondary to wear. The 10-mm arc comprises a larger portion of the articular surface in smaller cups than in men with larger cups [16]. The wear rate of metal bearings is also subject to patient variables as with any bearing surface. Younger, more active patients place more demands on the bearing surface and are also the patients most likely to have received an MOM bearing for those very reasons [17]. When metal debris is generated through wear, it is accepted that soluble Co and Cr ions would be present in the blood and provide a reliable measure of MOM bearing performance. Leopold et al [18] initially reported the use of metal ion levels as a trigger for revision surgery. They documented postoperative serial titanium levels in a total knee study and noted a dramatic increase in titanium levels in 1 patient with no obvious clinical or radiographic abnormalities. Revision was performed, and the patient had worn through the polyethylene of the metal backed patella leading to titanium on titanium articulation in the patellofemoral joint. Jacobsson et al [19] reported elevated ion levels in the McKee-Farrar design at the 20-year follow-up point with no apparent ill effects. In a prospective randomized study comparing MOM to metal-poly bearings, MacDonald et al [20] measured ion levels and noted increasing levels in 41% of the MOM patients. The study was discontinued because of this finding, but the elevated ion levels did not seem to correlate with any symptoms. The cytotoxicity of metal debris is poorly understood. In this study, tissue damage at the time of revision did show some correlation to the time in situ and may indicate a time- and dose-related phenomenon. Patients
35
have variable hypersensitivity responses to metallic debris and may also have variable metal ion exposure threshold levels that lead to adverse reactions. One theory has suggested that the acidic environment found with crevice corrosion at the head-neck taper may be a factor. Recent evidence suggests that larger heads may contribute to an increase in taper wear and corrosion [21]. Abrasive wear of a CoCr bearing produces trivalent Cr as metallic debris. In an acidic environment, this trivalent Cr can be converted into a hexavalent Cr, which is able to more easily cross the cell membrane barrier and is more cytotoxic. So, the combination of increased wear along with a bad taper may be significant [7,22]. There are several limitations of our study inherent with multicenter observational studies. Ion levels were measured by several different laboratories and different technicians, which may have affected our ion level measurements. Although most of our revisions involved ASR cups, other designs were included, and extrapolation to all MOM total hip arthroplasty designs may not be valid. The method used for grading soft tissue damage was a subjective interpretation of the operative note by the senior author, and there was no attempt to determine intraobserver and interobserver variability. Nonetheless, the study is representative of real situations encountered in clinical practice. It would be very useful to surgeons and patients alike if there were a simple blood test that could be used to predict MOM bearing function and subsequent tissue damage before major problems occur. This study highlights the difficulties of such a test. Although this is the largest series reporting on the relationship between ion levels and tissue damage, there are currently thousands of patients with MOM bearings being evaluated and undergoing revision surgery. Pooling of several large data sets might give a clearer clinical understanding of increased ion levels and define the levels that warrant closer observation and potentially revision. Currently with an arbitrary 7-ppb threshold, patients with a lack of symptoms are being operated on prematurely, and others with significant damage may have a false sense of security from low ion levels. Ion levels should not be used alone to monitor MOM bearings and should not be used in isolation for determining when revision surgery is warranted. Patient history and physical examination with metal ion levels, cross-sectional imaging with ultrasound, or magnetic resonance imaging with metal artifact reduction sequences should be reviewed in combination before considering revision. The possibility of infection must also be considered from the outset in these cases. In this study, we found that tissue damage did positively correlate with time in situ. This would suggest that, in those cases where metallosis is suspected, delaying
36 The Journal of Arthroplasty Vol. 27 No. 8 Suppl. 1 September 2012 surgical intervention is not a good option. In addition, a good understanding of the specific implant design involved, its position on x-ray, and the track record of the implant will help guide surgical decision making.
References 1. MacDonald SJ. Metal-on-metal total hip arthroplasty. Clin Orthop Relat Res 2004;429:86. 2. Mabilleau G, Kwon YM, Pandit H, et al. Metal-on-metal hip resurfacing arthroplasty: a review of periprosthetic biological reactions. Sabokbar Acta Orthop 2008;79:734. 3. Tower SS. Arthroprosthetic cobaltism: neurological and cardiac manifestations in two patients with metal-onmetal arthroplasty: a case report. J Bone Joint Surg Am 2010;92:2847. 4. Hayter CL, Potter HG, Su EP. Imaging of metal-on-metal hip resurfacing. Orthop Clin North Am 2011;42:195. 5. Langton DJ, Sprowson AP, Joyce TJ, et al. Blood metal ion concentrations after hip resurfacing arthroplasty: a comparative study of articular surface replacement and Birmingham hip resurfacing arthroplasties. J Bone Joint Surg Br 2009;91:1287. 6. MHRA. Medical Device Alert. Ref: MDA/2010/033. Issues: 22 April 2010. 7. Cobb AG, Schmalzreid TP. The clinical significance of metal ion release from cobalt-chromium metal-on-metal hip joint arthroplasty. Proc Inst Mech Eng H 2006;220:385. 8. De Smet K, De Haan R, Calistri A, et al. Metal ion measurement as a diagnostic tool to identify problems with metal-on-metal hip resurfacing. J Bone Joint Surg Am 2008;90(Suppl 4):202. 9. Medley JB, Chan FW, Krygier JJ, et al. Comparison of alloys and designs in a hip simulator study of metal on metal implants. Clin Orthop Relat Res 1996;329(Suppl):S148. 10. Chan FW, Bobyn JD, Medley JB, et al. The Otto Aufranc Award. Wear and lubrication of metal-on-metal hip implants. Clin Orthop Relat Res 1999;10. 11. Springer BD, Habet NA, Griffin WL, et al. Deformation of 1-piece metal acetabular components. J Arthroplasty 2012;27:48.
12. Griffin WL, Nanson CJ, Springer BD, et al. Reduced articular surface of one-piece cups: a cause of runaway wear and early failure. Clin Orthop Relat Res 2010;468: 2328. 13. Fisher J. Bioengineering reasons for the failure of metalon-metal hip prostheses: an engineer's perspective. J Bone Joint Surg Br 2011;93:1001. 14. Morlock MM, Bishop N, Zustin J, et al. Modes of implant failure after hip resurfacing: morphological and wear analysis of 267 retrieval specimens. J Bone Joint Surg Am 2008;90(Suppl 3):89. 15. DeHaan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg Br 2008;90:1291. 16. Kwon YM, Ostlere SJ, McLardy-Smith P, et al. “Asymptomatic” pseudotumors after metal-on-metal hip resurfacing arthroplasty: prevalence and metal ion study. J Arthroplasty 2011;26:511. 17. Eskelinen A, Remes V, Helenius I, et al. Total hip arthroplasty in the Finnish arthroplasty register. 4,661 primary replacements followed for 0-22 years. Acta Orthop 2005;76:28. 18. Leopold SS, Berger RA, Patterson L, et al. Serum titanium level for diagnosis of a failed, metal-backed patellar component. J Arthroplasty 2000;15:938. 19. Jacobsson S, Djerf K, Wahlstrom O. Twenty-year results of McKee-Farrar versus Charnley prosthesis. Clin Orthop Relat Res 1996;329:S60. 20. MacDonald SJ, McCalden RW, Chess DG, et al. Metalon-metal versus polyethylene in hip arthroplasty: a randomized clinical trial. Clin Orthop Relat Res 2003; 406:282. 21. Takamura KM, Langton D, Gandhi JN, et al. The main issue of large diameter MoM total hip arthroplasty: the taper junction. Presented at the Annual Meeting. February; American Academy of Orthopaedic Surgeons; 2012. 22. Lindgren JU, Brismar BH, Wikstrom AC. Adverse reaction to metal release from a modular metal-on-polyethylene hip prosthesis. J Bone Joint Surg Br 2011;93:1427.
Are Metal Ion Levels a Useful Trigger for Surgical Intervention? William L. Griffin, MD,* Thomas K. Fehring, MD,* James C. Kudrna, MD,y Robert H. Schmidt, MD,z Michael J. Christie, MD,§ Susan M. Odum, MEd,‖ and Anne C. Dennos, BS‖
Abstract: The purpose of this study was to determine if cobalt and chromium ion levels can predict soft tissue damage at total hip revision. This study included 90 metal-on-metal total hip patients with preoperative cobalt and chromium ion levels. Tissue damage noted at revision surgery was graded on a 4-point scale. Sensitivity, specificity, and predictive values were calculated for various threshold values. Receiver operating characteristic analysis was conducted. Using 7 ppb as a threshold, cobalt and chromium ion levels had poor sensitivity and specificity (Co, 65% and 56%; Cr, 29% and 75%). Positive predictive values for cobalt and chromium were only 48% and 26% respectively. The area under the curve was 0.37 for cobalt and 0.44 for chromium. The length of time to revision significantly correlated with tissue damage (P = .001). Ion levels are unreliable predictors of periarticular soft tissue damage and should not be used in isolation as surgical intervention triggers. Keywords: metal ion levels, total hip revision, cobalt, chromium. © 2012 Elsevier Inc. All rights reserved.
surgical intervention [5]. The medical device alert from the British Orthopedic Association concerning MOM implants chose cobalt and chromium ion levels of 7 ppb as a threshold for concern [6]. Because cobalt (Co) and chromium (Cr) are essential elements, there is an effective renal clearance mechanism for these metals. Wear at the MOM articulation produces nonsoluble metal debris as well as soluble metal ions that can be measured in the blood [7]. These ion levels present in the blood can come from several sources: wear at the articulation, corrosion at any taper junction, abrasion of loose components, and normal body stores [1]. The blood levels of these ions represent a balance between ion production from the device and renal excretion, and these levels can be affected by changes in activity levels as well as renal function. Based on current recommendations, surgeons may be compelled to revise well-functioning MOM implants based solely on a metal ion level of greater than 7 ppb [6]. In other reports, patients with metal ion levels well below 7 ppb were found to have significant tissue damage at the time of surgery [8]. Currently, there are no evidence-based data to support the use of a specific metal ion threshold as an indicator of periarticular tissue damage and thus a trigger for surgical intervention. The purpose of this study is to determine if cobalt and chromium ion levels can predict soft tissue damage at the time of MOM revision.
Metal-on-metal (MOM) hip articulations were thought to be a potential solution for wear-related failures in total hip arthroplasties while also improving stability through the use of larger diameter femoral heads. Although MOM bearings produce considerably less wear than metal-on-poly bearings, the debris generated can lead to adverse local tissue reactions and systemic effects from prolonged exposure to metal ions [1-3]. The clinical evaluation and treatment of patients presenting with symptomatic MOM bearing implants are difficult. It is important to have reliable tests to help determine when to intervene surgically to prevent tissue damage. Ultrasound and magnetic resonance imaging with metal artifact reduction sequences have been used to assess periarticular reactions secondary to metal wear debris [4]. In addition, it has been suggested that metal ion levels may have prognostic value in determining the presence of articular wear and the timing or need for
From the *OrthoCarolina Hip and Knee Center, Charlotte, North Carolina; yIllinois Bone and Joint Institute, Glenview, Illinois; zTexas Hip and Knee Center, Fort Worth, Texas; §Southern Joint Replacement Institute, Nashville, Tennessee; and ‖OrthoCarolina Research Institute, Charlotte, North Carolina. Submitted January 13, 2012; accepted March 13, 2012. The Conflict of Interest statement associated with this article can be found at doi:10.1016/j.arth.2012.03.020. Reprint requests: William L. Griffin, MD, OrthoCarolina Hip and Knee Center, 2001 Vail Ave, Suite 200-A, Charlotte, NC 28207. © 2012 Elsevier Inc. All rights reserved. 0883-5403/2708-0008$36.00/0 doi:10.1016/j.arth.2012.03.020
32
Metal Ion Levels for Surgical Intervention Griffin et al
Materials and Methods We conducted a multicenter retrospective review of a series of MOM total hips that were revised. All 4 participating centers obtained institutional review board approval. Cases were identified by queries of institutional prospective total joint registries and practice management systems. Electronic health records and registry data were reviewed to document dates of primary total hip surgery and revision total hip surgery to determine time in situ. In addition, reason for revision and implant details from the primary surgery were recorded. Whole blood cobalt and chromium ion levels were obtained from LabCorp and Nation Medical Services using an Inductively Coupled Mass Spectrometry (ICPMS) technique on whole blood and reported in micrograms per liter (1 μg/L = 1 ppb). In the event that there was more than 1 ion level workup, the result closest to the revision date was selected. Operative notes were reviewed by the senior author (WLG) to evaluate the severity of metallosis and/or tissue damage at the time of the revision surgery. The severity of tissue damage was categorized using the following tissue grade scale, developed by the authors: grade 0, no metallosis; grade 1, metallic staining, intracapsular fluid; grade 2, extracapsular fluid; grade 3, extracapsular tissue necrosis. One hundred sixteen symptomatic MOM total hip revision cases with complete cobalt and chromium ion levels were included. Four patients who underwent bilateral MOM total hip revision (8 hips) and 15 patients with an unrevised MOM hip on the contralateral side (15 hips) were excluded. In addition, 4 patients (4 hips) were excluded because of insufficient data to evaluate tissue damage. This left a total of 89 symptomatic MOM revision cases available for this study. Of these 89 cases, 84 were monoblock implants (82 DePuy ASR TM [Warsaw, Ind], 1 Conserve R [Wright Medical Group Inc., Memphis, Tenn], and 1 Magnum TM [Biomet, Warsaw, Ind]), and 5 were modular implants (Pinnacle R, DePuy Orthopedics, Warsaw, Ind). Reasons for revision are documented in Table 1. Standard descriptive statistics were calculated. The association between ion level and tissue grade was determined by a Spearman correlation coefficient. Oneway analysis of variance was used to determine any differences between ion levels, tissue grade, and time in situ. Using 7 ppb as the cutoff, sensitivity, specificity,
Table 1. Frequency of Revision Diagnoses Reason for Revision
Frequency
Aseptic loosening Suspected metallosis Pain Noise
17 30 42 10
There was overlap with the reasons for revision.
33
positive predictive value, and negative predictive value were calculated to assess the value of both cobalt and chromium ion levels as diagnostic, predictive tools for metallosis. To evaluate optimal threshold values of metal ion levels, a receiver operating characteristic analysis was conducted. All statistical analyses were performed using SAS 9.2.1 (Cary, NC).
Results There was a profound variation in the whole blood metal ion levels. The mean cobalt ion level in the study population was 23 (0-236), and the mean chromium ion level was 9 (0-86). Of the 89 patients, 23 were classified as grade 0, 38 as grade 1, 19 as grade 2, and 9 as grade 3. There was little to no correlation between ion levels and tissue damage. The correlation coefficient for cobalt and tissue grade was 0.192 (P = .073) and 0.162 (P = .132) for chromium and tissue grade. There were no statistical differences in the mean cobalt (P = .64) or chromium (P = .41) ion levels between the 4 tissue grades. There was also no incremental increase, as the severity of tissue damage increased. In fact, the mean cobalt levels for grades 0 (17.41 ppb) and 2 (15.69 ppb) were similar, whereas mean cobalt levels for grades 1 (27.92 ppb) and 3 (26.87 ppb) were similar. For chromium, the mean ion level was highest for grade 1. See Table 2 for metal ion levels by each tissue grade. There were significant (P = .0013) differences in the time in situ of the MOM hip between the 4 tissue grades. Furthermore, the severity of tissue damage increased as the time in situ increased. These results indicate that there may be a time-dose response relationship. When ion levels and time in situ are combined, this correlation is lost. This suggests that the tissue damage seen at revision surgery may be related to the length of time of metal ion exposure but not the exposure dose. See Table 2 for time in situ for each tissue grade. Using 7 ppb as the cutoff value, the diagnostic value of both cobalt and chromium ion levels indicate poor accuracy of these tests as a diagnostic or predictive tool. The sensitivity of 7 ppb as a predictor of patients with moderate to severe tissue damage was 61% for cobalt and 31% for chromium. The specificity or the proportion of patients without moderate to severe tissue damage was 58% for cobalt and 76% for chromium. The positive predictive value or the proportion of patients with tissue damage and cobalt ion levels greater than 7 ppb was only 39.5% and 37.5% for chromium. The negative predictive values were 60.5% and 62.5% for cobalt and chromium, respectively. These values did not improve when using alternate cutoff values ranging from 7 to 50 ppb (Table 3). The receiver operating characteristic curve indicates that the area under curve for cobalt was 0.37 (Fig.). In other words, there is only a 37% probability that cobalt ion levels will be higher in patients with moderate to severe
34 The Journal of Arthroplasty Vol. 27 No. 8 Suppl. 1 September 2012 Table 2. Sensitivity and Specificity of Cobalt and Chromium Metal Ion Levels Cobalt Threshold
Sensitivity
7 ppb 10 ppb 13 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.61 0.39 0.36 0.32 0.32 0.25 0.21 0.14
95% CI 0.41-0.78 0.22-0.59 0.19-0.56 0.17-0.52 0.17-0.52 0.11-0.45 0.09-0.41 0.05-0.33
Specificity 0.58 0.63 0.68 0.7 0.73 0.75 0.81 0.87
95% CI 0.45-0.70 0.49-0.75 0.55-0.79 0.57-0.81 0.60-0.83 0.63-0.85 0.68-0.89 0.76-0.94
Chromium Threshold
Sensitivity
7 ppb 10 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.31 0.24 0.14 0.32 0.07 0.03 0
95% CI 0.16-0.51 0.11-0.44 0.05-0.33 0.17-0.52 0.01-0.24 0.001-0.20 0.0-0.15
Specificity 0.76 0.81 0.82 0.73 0.85 0.85 0.95
95% CI 0.63-0.85 0.68-0.89 0.70-0.90 0.60-0.83 0.74-0.93 0.74-0.93 0.86-0.99
Cobalt Threshold
PPV
95% CI
NPV
95% CI
7 ppb 10 ppb 13 ppb 15 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.4 0.32 0.33 0.33 0.34 0.32 0.33 0.33
0.25-0.56 0.18-0.51 0.18-0.53 0.17-0.54 0.18-0.56 0.15-0.55 0.14-0.59 0.11-0.65
0.6 0.68 0.67 0.67 0.65 0.68 0.67 0.67
0.44-0.75 0.49-0.82 0.47-0.82 0.46-0.83 0.44-0.82 0.45-0.85 0.41-0.86 0.35-0.89
Chromium Threshold
PPV
95% CI
NPV
95% CI
7 ppb 10 ppb 13 ppb 18 ppb 20 ppb 25 ppb 50 ppb
0.38 0.37 0.27 0.35 0.18 0.1 0.0
0.20-0.59 0.17-0.61 0.09-0.55 0.18-0.56 0.03-0.52 0.005-0.46 0.0-0.69
0.63 0.63 0.73 0.65 0.82 0.9 1.0
0.41-0.80 0.38-0.83 0.45-0.91 0.44-0.82 0.48-0.97 0.54-0.99 0.31-1.0
Abbreviations: CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.
tissue damage as compared with those who have none to minimal tissue damage. The area under the curve for chromium was slightly higher at 0.44.
Discussion There are a number of factors that can contribute to the increased wear of MOM bearings and the generation of Co and Cr ions. These include manufacturing and design variables, surgical technique, and patient variables. In terms of the manufacturing techniques, evaluation of the original MOM designs and bench top studies has identified design parameters that can improve the wear of MOM bearings. The hardness of the CrCo material has been improved with the use of a
Fig. Receiver operating characteristic curve for cobalt and chromium metal ion levels.
higher carbon content and forging of the material, both of which have decreased wear rates. Improving the roundness and smoothness and keeping the tolerances between 80 and 120 μm have also improved the wear characteristics [9,10]. The design of the implant can also lead to higher failure rates. Wear characteristics vary across MOM bearings, which has been particularly evident with 1piece designs. Because the 1-piece shells were originally intended for resurfacing arthroplasties, there are a number of design compromises. To accommodate the native femoral neck and avoid impingement, 1-piece shells are less than a full 180° hemisphere. By reducing valuable articular surface at the rim, the shell is more prone to edge loading and runaway wear. In an attempt to maximize head size and minimize acetabular bone resection, the resurfacing shells were made relatively thin. Thinner shells can deform during insertion with a press-fit technique and interfere with the tight tolerances of an MOM articulation [11]. In an attempt to increase the stiffness of these shells, the manufacturers thickened the dome region of the cup. The amount of this thickening varies by manufacturer and cup size and has the unintended consequence of moving the center of rotation further outside the shell, again decreasing the articular surface available to the head. A few of the 1Table 3. Difference in Metal Ion Levels Mean Time In Situ (Months) by Tissue Grade (Analysis of Variance) Tissue Grade
Cobalt
P
0 1 2 3
17.41 27.92 15.69 26.87
.644
Chromium 6.06 12.7 6.13 7.26
P .41
Time in Situ 29.58 33.55 44.01 46.54
P .0013
Metal Ion Levels for Surgical Intervention Griffin et al
piece designs also added a cut-out at the rim to allow for an insertional device, removing even more articular surface. When combined, the net effect of these design features makes the 1-piece shells very susceptible to edge loading and increased wear. With this loss of articular surface, a cup placed at 50° abduction may behave like a cup inserted at 63° of abduction [12]. Numerous studies have identified that increased cup abduction and cup anteversion angles can lead to higher wear rates. A less appreciated factor contributing to failure is excessive anteversion. Women in particular may have more combined anteversion. The femur may have significant native anteversion. This in combination with acetabular version may lead to edge loading and aberrant wear in extension during ambulation. Fisher [13] noted a 10-fold increase in wear with a 55° abduction angle, and Morlock et al [14] reported a 27fold increase in wear with 59° of abduction. De Haan et al [15] postulated that when the weight-bearing vector is within 10 mm of the cup's rim, the MOM bearing loses its fluid film lubrication and boundary wear ensues with higher wear rates. This may explain why women with smaller cups and smaller articular surfaces have been more prone to failure secondary to wear. The 10-mm arc comprises a larger portion of the articular surface in smaller cups than in men with larger cups [16]. The wear rate of metal bearings is also subject to patient variables as with any bearing surface. Younger, more active patients place more demands on the bearing surface and are also the patients most likely to have received an MOM bearing for those very reasons [17]. When metal debris is generated through wear, it is accepted that soluble Co and Cr ions would be present in the blood and provide a reliable measure of MOM bearing performance. Leopold et al [18] initially reported the use of metal ion levels as a trigger for revision surgery. They documented postoperative serial titanium levels in a total knee study and noted a dramatic increase in titanium levels in 1 patient with no obvious clinical or radiographic abnormalities. Revision was performed, and the patient had worn through the polyethylene of the metal backed patella leading to titanium on titanium articulation in the patellofemoral joint. Jacobsson et al [19] reported elevated ion levels in the McKee-Farrar design at the 20-year follow-up point with no apparent ill effects. In a prospective randomized study comparing MOM to metal-poly bearings, MacDonald et al [20] measured ion levels and noted increasing levels in 41% of the MOM patients. The study was discontinued because of this finding, but the elevated ion levels did not seem to correlate with any symptoms. The cytotoxicity of metal debris is poorly understood. In this study, tissue damage at the time of revision did show some correlation to the time in situ and may indicate a time- and dose-related phenomenon. Patients
35
have variable hypersensitivity responses to metallic debris and may also have variable metal ion exposure threshold levels that lead to adverse reactions. One theory has suggested that the acidic environment found with crevice corrosion at the head-neck taper may be a factor. Recent evidence suggests that larger heads may contribute to an increase in taper wear and corrosion [21]. Abrasive wear of a CoCr bearing produces trivalent Cr as metallic debris. In an acidic environment, this trivalent Cr can be converted into a hexavalent Cr, which is able to more easily cross the cell membrane barrier and is more cytotoxic. So, the combination of increased wear along with a bad taper may be significant [7,22]. There are several limitations of our study inherent with multicenter observational studies. Ion levels were measured by several different laboratories and different technicians, which may have affected our ion level measurements. Although most of our revisions involved ASR cups, other designs were included, and extrapolation to all MOM total hip arthroplasty designs may not be valid. The method used for grading soft tissue damage was a subjective interpretation of the operative note by the senior author, and there was no attempt to determine intraobserver and interobserver variability. Nonetheless, the study is representative of real situations encountered in clinical practice. It would be very useful to surgeons and patients alike if there were a simple blood test that could be used to predict MOM bearing function and subsequent tissue damage before major problems occur. This study highlights the difficulties of such a test. Although this is the largest series reporting on the relationship between ion levels and tissue damage, there are currently thousands of patients with MOM bearings being evaluated and undergoing revision surgery. Pooling of several large data sets might give a clearer clinical understanding of increased ion levels and define the levels that warrant closer observation and potentially revision. Currently with an arbitrary 7-ppb threshold, patients with a lack of symptoms are being operated on prematurely, and others with significant damage may have a false sense of security from low ion levels. Ion levels should not be used alone to monitor MOM bearings and should not be used in isolation for determining when revision surgery is warranted. Patient history and physical examination with metal ion levels, cross-sectional imaging with ultrasound, or magnetic resonance imaging with metal artifact reduction sequences should be reviewed in combination before considering revision. The possibility of infection must also be considered from the outset in these cases. In this study, we found that tissue damage did positively correlate with time in situ. This would suggest that, in those cases where metallosis is suspected, delaying
36 The Journal of Arthroplasty Vol. 27 No. 8 Suppl. 1 September 2012 surgical intervention is not a good option. In addition, a good understanding of the specific implant design involved, its position on x-ray, and the track record of the implant will help guide surgical decision making.
References 1. MacDonald SJ. Metal-on-metal total hip arthroplasty. Clin Orthop Relat Res 2004;429:86. 2. Mabilleau G, Kwon YM, Pandit H, et al. Metal-on-metal hip resurfacing arthroplasty: a review of periprosthetic biological reactions. Sabokbar Acta Orthop 2008;79:734. 3. Tower SS. Arthroprosthetic cobaltism: neurological and cardiac manifestations in two patients with metal-onmetal arthroplasty: a case report. J Bone Joint Surg Am 2010;92:2847. 4. Hayter CL, Potter HG, Su EP. Imaging of metal-on-metal hip resurfacing. Orthop Clin North Am 2011;42:195. 5. Langton DJ, Sprowson AP, Joyce TJ, et al. Blood metal ion concentrations after hip resurfacing arthroplasty: a comparative study of articular surface replacement and Birmingham hip resurfacing arthroplasties. J Bone Joint Surg Br 2009;91:1287. 6. MHRA. Medical Device Alert. Ref: MDA/2010/033. Issues: 22 April 2010. 7. Cobb AG, Schmalzreid TP. The clinical significance of metal ion release from cobalt-chromium metal-on-metal hip joint arthroplasty. Proc Inst Mech Eng H 2006;220:385. 8. De Smet K, De Haan R, Calistri A, et al. Metal ion measurement as a diagnostic tool to identify problems with metal-on-metal hip resurfacing. J Bone Joint Surg Am 2008;90(Suppl 4):202. 9. Medley JB, Chan FW, Krygier JJ, et al. Comparison of alloys and designs in a hip simulator study of metal on metal implants. Clin Orthop Relat Res 1996;329(Suppl):S148. 10. Chan FW, Bobyn JD, Medley JB, et al. The Otto Aufranc Award. Wear and lubrication of metal-on-metal hip implants. Clin Orthop Relat Res 1999;10. 11. Springer BD, Habet NA, Griffin WL, et al. Deformation of 1-piece metal acetabular components. J Arthroplasty 2012;27:48.
12. Griffin WL, Nanson CJ, Springer BD, et al. Reduced articular surface of one-piece cups: a cause of runaway wear and early failure. Clin Orthop Relat Res 2010;468: 2328. 13. Fisher J. Bioengineering reasons for the failure of metalon-metal hip prostheses: an engineer's perspective. J Bone Joint Surg Br 2011;93:1001. 14. Morlock MM, Bishop N, Zustin J, et al. Modes of implant failure after hip resurfacing: morphological and wear analysis of 267 retrieval specimens. J Bone Joint Surg Am 2008;90(Suppl 3):89. 15. DeHaan R, Pattyn C, Gill HS, et al. Correlation between inclination of the acetabular component and metal ion levels in metal-on-metal hip resurfacing replacement. J Bone Joint Surg Br 2008;90:1291. 16. Kwon YM, Ostlere SJ, McLardy-Smith P, et al. “Asymptomatic” pseudotumors after metal-on-metal hip resurfacing arthroplasty: prevalence and metal ion study. J Arthroplasty 2011;26:511. 17. Eskelinen A, Remes V, Helenius I, et al. Total hip arthroplasty in the Finnish arthroplasty register. 4,661 primary replacements followed for 0-22 years. Acta Orthop 2005;76:28. 18. Leopold SS, Berger RA, Patterson L, et al. Serum titanium level for diagnosis of a failed, metal-backed patellar component. J Arthroplasty 2000;15:938. 19. Jacobsson S, Djerf K, Wahlstrom O. Twenty-year results of McKee-Farrar versus Charnley prosthesis. Clin Orthop Relat Res 1996;329:S60. 20. MacDonald SJ, McCalden RW, Chess DG, et al. Metalon-metal versus polyethylene in hip arthroplasty: a randomized clinical trial. Clin Orthop Relat Res 2003; 406:282. 21. Takamura KM, Langton D, Gandhi JN, et al. The main issue of large diameter MoM total hip arthroplasty: the taper junction. Presented at the Annual Meeting. February; American Academy of Orthopaedic Surgeons; 2012. 22. Lindgren JU, Brismar BH, Wikstrom AC. Adverse reaction to metal release from a modular metal-on-polyethylene hip prosthesis. J Bone Joint Surg Br 2011;93:1427.