Large-head metal-on-metal (MoM) hip implants gained popularity in the early 2000s as a result of their theoretical advantages of low wear and decreased risk of dislocation . However, by 2010 it became clear that some large-head MoM implants were associated with higher-than-expected revision rates, and their use rapidly declined . One MoM implant system was the Articular Surface Replacement (ASR) Hip System (DePuy Orthopaedics, Warsaw, IN, USA), which was implanted in an estimated 93,000 patients worldwide before a manufacturer recall in 2011 . Many patients therefore remain with the ASR Hip System in situ and require updated followup guidelines.
Numerous MoM implants underwent revision because of adverse local tissue reactions (ALTRs) secondary to implant wear . ALTRs can range from mild aseptic lymphocyte-dominated vasculitis-associated lesions to inflammatory lesions coupled with periprosthetic tissue masses (pseudotumors) and necrosis . These reactions have been described interchangeably as adverse reactions to metal debris, aseptic lymphocytic vasculitis-associated lesions, metallosis, and pseudotumors . ALTRs occur in as many as 69% (107 of 156) of patients who received the ASR XL THA  and in as many as 36% of patients (27 of 75) who underwent total hip resurfacing . It is important to identify ALTR in its early stages because MoM revision resulting from ALTR has been shown to lead to worse survival, inferior patient-reported outcomes, and higher rates of complications including dislocation and infection .
Although there is general agreement concerning the association between elevated blood metal ion levels and ALTR incidence in MoM THA [2, 4, 5, 27] and in MoM hip resurfacing arthroplasty (HRA) [18, 20, 29], one recent study found that blood ion levels may not be able to distinguish cases of ALTR at 5 to 7 years from surgery . The consensus regarding the relationship between patient symptoms and ALTR presence, however, is weaker. Several studies have shown similar pseudotumor incidence between symptomatic and asymptomatic patient cohorts [13, 27], and some have even demonstrated a higher incidence of ALTR in asymptomatic patients [3, 15, 19].
In our study, we therefore asked: (1) Was the presence of symptoms as measured by patient-recorded outcome measures (PROMs) associated with ALTR presence and severity as noted on metal artifact reduction sequence (MARS)-MRI in patients treated with one design of MoM THA and HRA? (2) Could reliable thresholds for blood metal ion levels be determined that were associated with ALTR presence on MARS-MRI?
Patients and Methods
Study Cohort and Followup Data
This retrospective study presents a secondary analysis on data drawn from a prospective, international, multicenter study of the recalled ASR Hip System. In the larger prospective study, patients are followed annually for 5 years for the purpose of identifying risk factors for revision and determining followup guidelines. Patients were consented between 2012 and 2015, after the recall of the ASR Hip System (average time from index surgery, 7.5 years; range, 3.8-11.7 years) from 16 centers in six different countries. Institutional review board approval was obtained at each participating center. In the present analysis, only data from the enrollment visit were considered. In addition, to minimize selection bias, only patients from two centers performing MARS-MRI on all patients regardless of clinical presentation were included. Finally, only patients with unilateral implants were considered to avoid confounding effects of an additional ion source in cases of bilateral MoM implants. The final study cohort consisted of 327 patients with a unilateral ASR hip replacement, 145 (44.3%) with a MoM THA, and 182 (55.7%) with a MoM HRA (Fig. 1).
In addition to receiving a MARS-MRI, patients also completed PROMs and received whole blood metal ion testing. All data were obtained within a 3-month window. Two PROMs were obtained for all patients: (1) the Harris hip score (HHS) measure of hip function (0-100, 0 being worst); and (2) a visual analog scale (VAS) for pain (0-10, 0 being least pain). In addition, all patients included in the study underwent testing for whole blood metal ion levels (cobalt [Co] and chromium [Cr]) at a local certified laboratory.
At both centers, MRI scanners were adjusted to produce MARS. Scanning was performed with 1.5-T clinical scanners (GE Healthcare, Waukesha, WI, USA; and Siemens Healthcare, Erlangen, Germany). Sequences used in imaging were coronal, axial, and sagittal proton density or T1-weighted fast spin echo as well as coronal and axial short tau inversion recovery sequences.
MARS-MRIs were assessed for the presence ALTR lesions using the Anderson classification system, which has been shown to have the highest inter- and intrareader reliabilities of currently available schemes [1, 30]. In this classification, ALTRs are classified as mild (C1), moderate (C2), or severe (C3). The classifications were devised to indicate the urgency to treat such that C1 means no intervention but followup recommended, C2 lesions should be considered for revision, and C3 lesions are in urgent need of revision. We therefore grouped C2 and C3 lesions together to indicate ALTR presence, our primary outcome.
In addition, maximal mass or cystic lesion ALTR diameter and synovial thickness of the ALTR were determined for hips in which ALTR (C1, C2, or C3) was found. Synovial thickness was defined as the maximal thickness of solid or cystic lesions either contained by the pseudocapsule or communicating with the disrupted pseudocapsule as measured on axial images . Such measurements have been shown to correlate with aseptic lymphocyte-dominated vasculitis-associated lesions scores and with tissue damage observed intraoperatively . Synovial thickness therefore served to quantify ALTR severity and was considered as our secondary outcome.
One author (RM) who had 3 years of experience reviewing MARS-MRIs and who was blinded to the clinical result assessed all MARS-MRIs. A validation set of 35 MARS-MRI was read three times by our reader to measure reliability and by a musculoskeletal radiologist (HP) with more than 10 years of experience reviewing MARS-MRI to gauge accuracy. Reliability proved excellent (intraclass correlation = 0.82), and there was substantial agreement between the readers (κ coefficient = 0.64).
A total of 49 of 145 (34%) patients who had undergone ASR MoM THA developed a moderate or severe (Anderson Grade C2/C3) ALTR by the time of their followup (mean, 7.2 years; range, 3.8-11.4 years). For patients in the ASR MoM HRA group, we found a 16% (29 of 182 patients) prevalence of moderate or severe ALTR at a mean of 7.8 years (range, 4.0-11.7 years) from surgery (Table 1).
Evaluation of Symptoms
We stratified the patients into two groups (symptomatic and asymptomatic) based on the HHS and VAS pain score. Each outcome measure was dichotomized based on the defined patient acceptable symptom state (PASS), the value above which the patient deems their current symptom state as acceptable. For the HHS, scores > 74 points  were considered as acceptable; all lower scores were classified as poor. For VAS pain, all scores ≤ 3.5  were deemed acceptable; all others were deemed poor. Patients who reported poor scores on either the HHS or VAS pain were labeled as clinically symptomatic. All others were categorized as asymptomatic.
Based on these classifications, symptoms were observed in 60 of 145 (41%) patients treated with ASR MoM THA and in 45 of 182 (25%) patients treated with ASR MoM HRA (Table 1).
All patients in both implant groups were considered separately for all analyses. We used a multivariable binary logistic regression to identify variables independently associated with ALTR. The following patient and implant factors that have previously been shown or theorized to be associated with ALTR were entered in the model as independent variables: age, sex, head size, followup time, Co and Cr ion levels, and symptom state. Backward elimination was used to iteratively remove variables from the model until the optimal fit was achieved.
In addition, we used the Mann-Whitney U test to analyze whether symptom state was associated with ALTR diameter or synovial thickness.
Finally, for blood metal ion levels that proved to be independently associated with ALTR, the most sensitive and specific threshold values were identified with receiver operator characteristic (ROC) analysis and the Youden method .
Statistical analyses were performed using SPSS, Version 19.0 (IBM, Armonk, NY, USA).
Association of Symptoms With ALTR
ALTR in Patients With MoM THAs
After controlling for the potential confounding variables of sex, femoral head size, and followup time, we found that the presence of symptoms (odds ratio [OR], 2.9; 95% confidence interval [CI], 1.5-6.0; p = 0.007) was associated with an increased likelihood of a patient having moderate or severe ALTR (Table 2). In a subanalysis of the patients who underwent ASR MoM THA and developed ALTR, the lesion or mass size did not vary between symptomatic and asymptomatic cohorts (mean difference = 3.1 mm; 95% CI, 2.0-5.1; p = 0.098), but synovial thickness did (mean difference = 2.5 mm; 95% CI, 1.1-6.8; p = 0.030; Fig. 2).
ALTR in Patients With MoM HRAs
After controlling for the aforementioned potentially confounding variables, we found no association between the level of symptoms and ALTR in patients who received a MoM HRA; symptom state was excluded from the multivariable model because it did not add any predictive value (Table 2). When considering patients with an ALTR only, we found that neither ALTR lesion nor mass size (mean difference = 2.1 mm; 95% CI, 1.1-6.2; p = 0.378) nor synovial thickness (mean difference = 0.9 mm; 95% CI, 0.1-1.8; p = 0.442; Fig. 2) was associated with symptoms (Table 3).
Threshold Cobalt Levels for ALTR
Cobalt was found to be independently associated with ALTR for patients with ASR MoM THA (OR, 1.2; 95% CI, 1.1-1.4; p = 0.001) and ASR HRA (OR, 1.2; 95% CI, 1.2-1.4; p = 0.001). In the patients in the MoM THA group, ROC analysis found that the Co threshold most associated with increased risk of ALTR was 3.2 parts per billion (ppb; area under the curve [AUC] = 0.72; p = 0.008). This cut point demonstrated a sensitivity of 68% and a specificity of 71%. For patients in the MoM HRA group, the most sensitive and specific threshold for Co was found to be 2.9 ppb (AUC = 0.79; p < 0.001). The sensitivity was 79% and the specificity was 69%.
ALTR is a common cause of MoM arthroplasty revision, and therefore it is important to identify it based on its clinical presentation. Numerous national guidelines suggest that patients with MoM arthroplasty undergo blood metal ion testing in addition to being assessed for symptom state [12, 17, 23]. The clinician must use these metrics (blood ion levels and level of symptoms) to gauge ALTR risk. However, recent reports [3, 13, 15, 19, 27] have shown that the ALTR incidence may be high even in asymptomatic patients, and a consensus on the level of blood metal ion levels that might be deemed safe is lacking. We found that symptoms as measured by HHS and pain VAS were associated with an increased risk of having ALTR in patients who received MoM THA but not for those with MoM HRA. Moreover, patients undergoing MoM THA who had more severe ALTR grades were more likely to be symptomatic. Finally, we found that Co was more closely associated with ALTR than Cr, and we determined reliable Co thresholds that are associated with ALTR for patients who underwent MoM THA or MoM HRA.
Our study had some limitations. First, clinical data were collected at a mean time of 7.5 years after the index surgery because the study was started after the recall of the ASR Hip System. This means that our cohort does not include the many patients with ASR implants who were revised in the early followup period; however, this analysis sheds light on the clinical variables associated with ALTR for the patients still implanted and therefore most in need of followup guidelines. Second, we were not able to correlate our ALTR assessment with surgical findings from a revision operation. However, we used MARS-MRI to detect ALTR, which is the gold standard imaging technique for assessing ALTR. We also assessed the images with the most reproducible (Anderson) method  as well as by quantifying synovial thickness, which has previously been shown to correlate with histologic and intraoperative findings . Third, our findings are difficult to compare with others because there is no generally accepted way to quantify symptoms. We dichotomized our PROMs using the PASS concept in an attempt to capture pain and functional status that most closely correlate with what patients perceive as significant. Finally, it is important to note that we included only the recalled ASR Hip System in our study. The ASR Hip System is known to be the worst performing MoM device . Additionally, patients with the ASR Hip System may report symptoms differently knowing that they have a recalled device. Extrapolation of the current results to other MoM implants may not be valid.
In our analysis of the ASR MoM THA cohort, the incidence of ALTR was different between patients with and without symptoms. Furthermore, symptoms were independently associated with ALTR after controlling for potential confounding factors. According to a 2016 meta-analysis, the association between ALTR prevalence and symptom state has been unclear . The association between our symptom measures and ALTR grades demonstrated in this study suggests that in patients with ALTR, the space-occupying lesions may mechanically impede joint motion and reduce function. In addition, such reactions may also stimulate pain receptors through periprosthetic inflammation or tissue necrosis .
In our separate analysis considering patients treated with ASR MoM THA and subsequently diagnosed with ALTR, we found that a thicker synovium was associated with a greater risk of having symptoms. A similar association was found by one other study that evaluated patients with MoM HRA . Those authors found that the synovial volume was greater for symptomatic patients with ALTR. To our knowledge, our study is the first to also demonstrate this association in patients treated with MoM THA. These findings suggest that synovial thickness may be an excellent marker of ALTR severity. Therefore, we suggest that followup MARS-MRIs also be monitored for signs of synovial thickening, because this may indicate potential symptoms. A previous report has shown that among patients undergoing MoM THA with an ALTR, larger pseudotumor size, solid pseudotumors, and anteriorly located pseudotumors were correlated with symptoms . These associations did not exist within our data set.
In contrast with the patients who had MoM THA in our study, the proportion of patients treated with ASR MoM HRA who were diagnosed with an ALTR did not vary between symptomatic and asymptomatic patients. Only one study evaluating MoM HRA found that patients with ALTR reported lower functional scores as measured by the Oxford Hip Score . However, that study found only a 4% prevalence of pseudotumors, which is much lower than other studies evaluating patients treated with MoM devices. This may be the result of the relatively short followup period and the fact that MARS-MRI was not used, thus making it likely that some ALTRs were missed. This same study also reported that ALTR is more prevalent in women; we found no association between ALTR and sex. In general, our findings agree with most others that have studied MoM HRA and found an association between blood metal ion levels and ALTR, but not between symptom state and ALTR [3, 18, 25]. Two studies have shown that larger ALTRs were associated with symptoms [3, 25]. We found no such association. In general, our findings suggest that there are patients with ASR MoM HRA and THA who have symptoms despite not having an ALTR, and therefore, the clinician should be vigilant to other potential causes of symptoms.
Numerous reports have shown a strong connection between blood Co and Cr levels and ALTR formation in MoM THA and HRA [3, 4, 19, 25, 27]. However, some studies dispute the fact that metal ion levels are reliable predictors of ALTRs [14, 16]. Our findings indicate that blood Co is closely associated with ALTR. Moreover, we found that although Co and Cr tend to be colinear, Co is more closely associated with ALTR risk than Cr. Based on this finding, we calculated thresholds for blood Co. Although these thresholds had substantial sensitivity and specificity in both the ASR MoM THA and MoM HRA cohorts, they are by no means a perfect diagnostic measure of ALTR. It is important to note that the only way to identify all patients with ALTR is by conducting MARS-MRI on all patients who received a MoM implant. The Co thresholds that we have identified may be useful in stratifying patients with MoM implants by ALTR risk, especially patients with the ASR MoM prostheses, because blood metal ion thresholds have been recently suggested to be implant-specific .
We found that for patients treated with ASR MoM THA, ALTR presence is associated with patient symptoms. Furthermore, ALTR severity (quantified by synovial thickness) is associated with an increased risk of having symptoms in patients who underwent ASR MoM THA. In addition to symptomaticity, a Co threshold of 3.2 ppb may be used to stratify such patients for ALTR risk. For patients treated with ASR MoM HRA, a blood Co threshold level of 2.9 ppb may be used to screen for ALTR, but there was no association between symptoms and ALTR in patients who underwent MoM HRA. Our findings highlight the need for regular clinical followup for patients who have undergone ASR MoM hip arthroplasty regardless of symptom state.
We thank Mr Slav Lerner and Dr Charles Bragdon for their work in the management of and logistic support for the study. In addition, we acknowledge Dr Hollis Potter for her MARS-MRI analysis training efforts.
1. Anderson H, Toms AP, Cahir JG, Goodwin RW, Wimhurst J, Nolan JF. Grading the severity of soft tissue changes associated with metal-on-metal hip replacements: reliability of an MR grading system. Skeletal Radiol. 2011;40:303–307.
2. Bayley N, Khan H, Grosso P, Hupel T, Stevens D, Snider M, Schemitsch E, Kuzyk P. What are the predictors and prevalence of pseudotumor and elevated metal ions after large-diameter metal-on-metal THA? Clin Orthop Relat Res. 2014;473:477–484.
3. Bisschop R, Boomsma MF, Van Raay JJAM, Tiebosch ATMG, Maas M, Gerritsma CLE. High prevalence of pseudotumors in patients with a birmingham hip resurfacing prosthesis. J Bone Joint Surg Am. 2013;95:1554–1560.
4. Bosker BH, Ettema HB, Boomsma MF, Kollen BJ, Maas M, Verheyen CCPM. High incidence of pseudotumour formation after large-diameter metal-on-metal total hip replacement: a prospective cohort study. J Bone Joint Surg Br. 2012;94:755–761.
5. Bosker BH, Ettema HB, van Rossum M, Boomsma MF, Kollen BJ, Maas M, Verheyen CCPM. Pseudotumor formation and serum ions after large head metal-on-metal stemmed total hip replacement. Risk factors, time course and revisions in 706 hips. Arch Orthop Trauma Surg. 2015;135:417–425.
6. Bozic KJ, Kurtz SM, Lau E, Ong K, Vail TP, Berry DJ. The epidemiology of revision total hip arthroplasty in the United States. J Bone Joint Surg Am. 2009;91:128–133.
7. Campbell P, Ebramzadeh E, Nelson S, Takamura K, De Smet K, Amstutz HC. Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin Orthop Relat Res. 2010;468:2321–2327.
8. Chahal J, Van Thiel GS, Mather RC, Lee S, Song SH, Davis AM, Salata M, Nho SJ. The patient acceptable symptomatic state for the modified harris hip score and hip outcome score among patients undergoing surgical treatment for femoroacetabular impingement. Am J Sports Med. 2015;43:1844–1849.
9. Chang EY, McAnally JL, Van Horne JR, Van Horne JG, Wolfson T, Gamst A, Chung CB. Relationship of plasma metal ions and clinical and imaging findings in patients with ASR XL metal-on-metal total hip replacements. J Bone Joint Surg Am. 2013;95:2015–2020.
10. Cohen D. Out of joint: the story of the ASR. BMJ. 2011;342:d2905.
11. Duggan PJ, Burke CJ, Saha S, Moonim M, George M, Desai A, Houghton R. Current literature and imaging techniques of aseptic lymphocyte-dominated vasculitis-associated lesions (ALVAL). Clin Radiol. 2013;68:1089–1096.
12. FAA. Finnish Arthroplasty Association’s recommendations for the follow up of patients with metal on metal hip arthroplasty. 2017;2017. Available at: https://thl.fi/far/#index
. Accessed January 15, 2018.
13. Fehring TK, Odum S, Sproul R, Weathersbee J. High frequency of adverse local tissue reactions in asymptomatic patients with metal-on-metal THA. Clin Orthop Relat Res. 2014;472:517–522.
14. Griffin WL, Fehring TK, Kudrna JC, Schmidt RH, Christie MJ, Odum SM, Dennos AC. Are metal ion levels a useful trigger for surgical intervention? J Arthroplasty. 2012;27:32–36.
15. Hasegawa M, Naito Y, Yamaguchi T, Miyazaki S, Wakabayashi H, Sudo A. Factors associated with symptomatic pseudotumors following metal-on-metal total hip arthroplasty. BMC Musculoskelet Disord. 2016;17:1–5.
16. Hjorth MH, Stilling M, Soballe K, Bolvig LH, Thyssen JP, Mechlenburg I, Jakobsen SS. No association between pseudotumors, high serum metal-ion levels and metal hypersensitivity in large-head metal-on-metal total hip arthroplasty at 5–7-year follow-up. Skeletal Radiol. 2016;45:115–125.
17. Kwon YM, Lombardi AV, Jacobs JJ, Fehring TK, Lewis CG, Cabanela ME. Risk stratification algorithm for management of patients with metal-on-metal hip arthroplasty: consensus statement of the American Association of Hip and Knee Surgeons, the American Academy of Orthopaedic Surgeons, and the Hip Society. J Bone Joint Surg Am. 2014;96:e4.
18. Kwon YM, Ostlere SJ, McLardy-Smith P, Athanasou NA, Gill HS, Murray DW. 'Asymptomatic' pseudotumors after metal-on-metal hip resurfacing arthroplasty. Prevalence and metal ion study. J Arthroplasty. 2011;26:511–518.
19. Laaksonen I, Donahue GS, Madanat R, Makela KT, Malchau H. Outcomes of the recalled Articular Surface Replacement metal-on-metal hip implant system: a systematic review. J Arthroplasty. 2017;32:341–346.
20. Matharu GS, Berryman F, Brash L, Pynsent PB, Treacy RBC, Dunlop DJ. The effectiveness of blood metal ions in identifying patients with unilateral Birmingham hip resurfacing and Corail-Pinnacle metal-on-metal hip implants at risk of adverse reactions to metal debris. J Bone Joint Surg Am. 2016;98:617–626.
21. Matharu GS, Berryman F, Judge A, Reito A, McConnell J, Lainiala O, Young S, Eskelinen A, Pandit HG, Murray DW. Blood metal ion thresholds to identify patients with metal-on-metal hip implants at risk of adverse reactions to metal debris: an external multicenter validation study of Birmingham Hip Resurfacing and Corail-Pinnacle Implants. J Bone Joint Surg Am. 2017;99:1532–1539.
22. Matharu GS, Pandit HG, Murray DW. Poor survivorship and frequent complications at a median of 10 years after metal-on-metal hip resurfacing revision. Clin Orthop Relat Res. 2017;475:304–314.
23. MHRA. All metal-on-metal (MoM) hip replacements: updated advice for follow-up of patients. Med Healthc Prod Regul Agency. 2017:1–7. Available at: https://www.gov.uk/drug-device-alerts/all-metal-on-metal-mom-hip-replacements-updated-advice-for-follow-up-of-patients
. Accessed January 15, 2018.
24. Nawabi DH, Gold S, Lyman S, Fields K, Padgett DE, Potter HG. MRI predicts ALVAL and tissue damage in metal-on-metal hip arthroplasty. Clin Orthop Relat Res. 2014;472:471–481.
25. Nawabi DH, Hayter CL, Su EP, Koff MF, Perino G, Gold SL, Koch KM, Potter HG. Magnetic resonance imaging findings in symptomatic versus asymptomatic subjects following metal-on-metal hip resurfacing arthroplasty. J Bone Joint Surg Am. 2013;95:895–902.
26. Smith AJ, Dieppe P, Vernon K, Porter M, Blom AW. Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry of England and Wales. Lancet. 2012;379:1199–1204.
27. Sutphen SA, MacLaughlin LH, Madsen AA, Russell JH, McShane MA. Prevalence of pseudotumor in patients after metal-on-metal hip arthroplasty evaluated with metal ion analysis and MARS-MRI. J Arthroplasty. 2016;31:260–263.
28. Tubach F, Ravaud P, Baron G, Falissard B, Logeart I, Bellamy N, Bombardier C, Felson D, Hochberg M, Van Der Heijde D, Dougados M. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis. 2005;64:34–37.
29. Van Der Straeten C, Grammatopoulos G, Gill HS, Calistri A, Campbell P, De Smet KA. The 2012 Otto Aufranc Award: The interpretation of metal ion levels in unilateral and bilateral hip resurfacing. Clin Orthop Relat Res. 2013;471:377–385.
30. Van Der Weegen W, Brakel K, Horn RJ, Wullems JA, Das HP, Pilot P, Nelissen RG. Comparison of different pseudotumor grading systems in a single cohort of metal-on-metal hip arthroplasty patients. Skeletal Radiol. 2014;43:149–155.
31. Wiley KF, Ding K, Stoner JA, Teague DC, Yousuf KM. Incidence of pseudotumor and acute lymphocytic vasculitis associated lesion (ALVAL) reactions in metal-on-metal hip articulations: a meta-analysis. J Arthroplasty. 2013;28:1238–1245.
32. Wynn-Jones H, MacNair R, Wimhurst J, Chirodian N, Derbyshire B, Toms A, Cahir J. Silent soft tissue pathology is common with a modern metal-on-metal hip arthroplasty. Acta Orthop. 2011;82:301–307.
33. Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3:32–35.