The first metal-on-metal articulation for total hip arthroplasty was introduced by Wiles1 in 1958, and in the 1960s McKee2,3 and Watson-Farrar4 popularized a cobalt-chromium-molybdenum metal-on-metal articulation. These total hip arthroplasties featured femoral head sizes of 38 to 42 mm. The McKee-Farrar femoral stem had a short, wide neck that facilitated impingement of the femoral neck against the socket, a design flaw that has been implicated in the failure of that implant.
Metal-on-metal articulation was reintroduced in the early 1990s, theoretically to reduce the rate of failure of total hip arthroplasties associated with polyethylene wear debris. Recently, several authors5-9 reported periprosthetic osteolysis and aseptic failure following contemporary total hip arthroplasties with second-generation metal-on-metal articulations, and metal hypersensitivity has been proposed as a mechanism of aseptic loosening of these components7,9-11. Most published series have included patients who were followed for a short time12-17. In this study, we report the results at a minimum of five years following use of a cementless SL-Plus stem (Plus Orthopedics, Rotkreuz, Switzerland) and Bicon-Plus cup (Plus Orthopedics) with a Sikomet SM21 metal-on-metal articulation. Our analysis included the findings of histologic examination of retrieved periprosthetic soft tissues from revised hips.
Materials and Methods
Between 1994 and 1999, 194 consecutive patients (217 hips) underwent total hip arthroplasty with a Sikomet SM21 metal-on-metal articulation. All procedures were performed by the senior author (P.K.). There were 143 women and fifty-one men, and the mean age (and standard deviation) at the time of the index surgery was 55 ± 9 years (range, twenty-five to seventy years). The indications for implantation of the prosthesis were osteoarthritis in 104 hips (48%), developmental dysplasia in ninety-six (44%), osteonecrosis in eleven hips (5%), and miscellaneous disorders in six (3%). Criteria precluding implantation of this prosthesis were an age of older than seventy years, a history of septic arthritis, or known hypersensitivity to cobalt or chromium. All 194 patients (217 hips) were available for follow-up at a minimum of five years.
The SL-Plus stem is a cementless, grit-blasted rectangular tapered femoral component made of a titanium alloy containing niobium (Fig. 1). The microporous surface roughness is 4 to 6 μm. The stem was introduced for clinical use in 1993.
The cementless Bicon-Plus cup is a double-cone cup made of forged pure titanium (ISO [International Organization for Standardization] 5832.2) (Fig. 2). The outer surface of the cup is threaded for screw-in fixation (Fig. 2).
Sikomet SM21 is a low-carbide, wrought cobalt-chromium-molybdenum alloy (ASTM [American Society for Testing and Materials] F799 and 1537, ISO 5832-12). The metal articulating cup has a thickness of 3 mm and is assembled into the polyethylene cup liner by the manufacturer to minimize the possibility of micromotion between the cup and the polyethylene. The polyethylene and articulating cup construct is inserted by the surgeon into the metal shell with a snap-fit coupling (Fig. 2).
The total hip arthroplasties were performed through a modified anterolateral Watson-Jones approach. The desired acetabular position was 45° of abduction and 10° of anteversion. After reaming and broaching of the femoral canal, the stem was impacted into a position of 0° of anteversion. Bone graft taken from the resected femoral head was then impacted between the stem and the femoral cortices to increase bone-stem contact and to diminish the potential space for distal migration of wear particles. All patients were treated with a 28-mm-diameter head and one of three neck lengths (short, medium, or long) to restore limb length and hip stability.
All patients received subcutaneous tinzaparin sodium for prophylaxis against thrombosis for four weeks following the surgery. On the day before the surgery and for three days after it, all patients received a second-generation cephalosporin (cefamandole) intravenously. No patient received indomethacin or any other medication for prophylaxis against heterotopic ossification. The patients were discharged from the hospital at a mean of four days (range, three to six days) after the index surgery.
We obtained institutional review board approval to perform this study. Patients were followed clinically, with a physical examination, with use of the Harris hip score, and on the basis of the patient's subjective evaluation of the result of the surgery, by one of the authors (G.P.), who had not participated in the surgery. Follow-up evaluations were performed at three, six, and twelve months postoperatively and then once a year. Each patient was asked to choose the most appropriate answer to the question of whether he or she was “very satisfied,” “satisfied,” or “disappointed” with the result of the operation. The location of pain (hip or femur, or both) as well as disability before and after the operation were recorded. The function of the operatively treated hip was determined at the final observation according to a modification of the functional classification system described by Perry et al.18.
The circulating hematocrit, red blood-cell sedimentation rate, and C-reactive protein level were determined at each follow-up evaluation. Serum metallic ion (cobalt-chromium) levels were not determined.
Anteroposterior supine radiographs of the pelvis and anteroposterior and lateral supine radiographs of the hip and femur were made preoperatively, immediately postoperatively (anteroposterior only), three and six months postoperatively (anteroposterior and lateral), and annually thereafter. The radiographs were made under the supervision of a senior orthopaedic musculoskeletal radiologist with use of a method described previously19,20. All radiographs were evaluated by two independent observers (a senior musculoskeletal radiologist and one of us [G.P.]), neither of whom was involved with patient care during the perioperative period.
Osteolysis was defined, according to the classification described by Zicat et al.21, as a focal area of bone resorption that was ≥2 mm wide at the time of final follow-up and had not been evident on the immediate postoperative radiograph. Osteolytic areas adjacent to the femoral stem were recorded, on anteroposterior and lateral radiographs, according to the zones described by Gruen et al.22 (Fig. 3). Osteolytic areas around the cup were recorded, on anteroposterior and lateral radiographs, with use of the zones described by DeLee and Charnley23.
We defined definite radiographic loosening of the cup as a rotational change of alignment of ≥3° in relation to the biischial line on the anteroposterior radiograph, continuous osteolysis of >2 mm in at least two DeLee-Charnley zones19,20,23, or axial migration of the cup of ≥3 mm.
Definite radiographic loosening of the stem was defined as axial subsidence of >2 mm in relation to the tip of the greater trochanter, new varus inclination of the stem of >3°, or continuous periprosthetic osteolysis of ≥2 mm in two or more adjacent Gruen zones and/or continuous radiolucent lines of >2 mm in thickness in more than two Gruen zones19,20,22. When infection was suspected, a technetium bone scan was obtained; a complete blood-cell count, the erythrocyte sedimentation rate, and the C-reactive protein level were measured; and a hip aspiration was performed. A final diagnosis of septic loosening was based on operative cultures and histologic quantification of the number of white blood cells per high-power field of pericapsular tissue.
Periarticular ossification was classified postoperatively according to the scale of Brooker et al.24 (ranging from I through IV) and was correlated with local pain and disability.
During the revision procedures, all of which were performed by the senior author (P.K.), specimens were retrieved from the hip capsule. Samples of the periprosthetic tissue from the bone-implant interface were fixed in neutral buffered formalin for twelve to twenty-four hours, and paraffin sections were stained with hematoxylin and eosin and examined with a light microscope and under polarized light for the presence of metallosis. Metallosis was graded according to the modified Mirra classification25 (Table I). Under light microscopy, metallosis due to titanium debris is seen as black-stained particles, whereas cobalt and chromium particles are stained green.
Paired t tests were used to compare the same variables, and unpaired t tests were used to compare different variables. Kaplan-Meier survivorship analysis26 was used to determine survival rates, with 95% confidence intervals, according to three end points: (1) revision for any cause, (2) revision because of aseptic lvoosening of the stem, and (3) revision because of aseptic loosening of the cup. A p value of 0.05 was considered to be significant. Survivorship was calculated at the final follow-up evaluation at a mean of 77 ± 17 months postoperatively.
The Pearson correlation coefficient (r) was used to analyze the relationship among histologic evidence of metallosis, lymphocytic and plasma cell concentration, and the prevalence of periprosthetic osteolysis (Table II).
At the time of final follow-up, at a mean of seventy-seven months (range, sixty to 112 months), 159 patients (168 hips) were unlimited community walkers, nineteen patients (twenty-four hips) were active community walkers, eleven patients (sixteen hips) were limited community walkers, four patients (seven hips) were household walkers, and one patient (two hips) was confined to a wheelchair18. The mean Harris hip score improved significantly from 45 points (range, 15 to 74 points) preoperatively to 88 ± 4 points at the final postoperative evaluation (p < 0.0001). Of the 194 patients, 184 (95%) were “satisfied” or “very satisfied” with the result of the operation.
Periprosthetic osteolytic lesions around the stem were confined to the proximal part of the femur (Fig. 3). There were nine stems with aseptic loosening, and all nine were associated with an osteolytic lesion of ≥2 mm. Two cups had aseptic loosening but were not associated with an osteolytic lesion. Periarticular ossification was seen in thirty-six hips (17%), of which only eleven (5% of the 217 hips in the series) showed Brooker grade-III or IV ossification24; these hips were not associated with disability.
Of the components that were seen to be definitively loose radiographically, twelve (nine stems and three cups) were associated with a continuous osteolytic lesion, five (three stems and two cups) had migrated, and two (cups) had rotated (Table II, Fig. 3).
Fourteen patients (fourteen hips, 6.5%) underwent revision. Their mean age at the time of the index revision was 59 ± 12 years (range, thirty-three to seventy years). Nine hips were revised because of aseptic loosening; two, because of technical failure; and three, because of deep infection (Table II). None of the articular surfaces in the fourteen hips showed macroscopic damage such as scratches or dull or flattened areas. The technical failures included one cup implanted in an extreme valgus position in a patient (Case 7, Table II) who subsequently had recurrent anterior dislocations and underwent revision of the cup with no further problems. The other patient with a technical failure (Case 3, Table II) had developmental dysplasia of the hip associated with poor acetabular coverage. The initial cup fixation was suboptimal, and the cup became progressively loose, causing symptoms. It was revised nine months following the primary implantation. At revision, both cups had macroscopic evidence of titanium metallosis on their fixation surface without signs of osseointegration.
Titanium metal particles were observed macroscopically in the periprosthetic tissue around the stem in one hip (Case 8, Table II) of the nine that were revised because of aseptic loosening.
The three deep infections that led to revision (Cases 12, 13, and 14; Table II) were caused by Staphylococcus aureus. In one of these hips (Case 13), both components were stable but were removed. In the second hip (Case 12), which was revised twelve months following primary implantation, the cup was stable but the stem was found to be loose without osseointegration. In the third hip (Case 14), both components were loose at five months following primary implantation.
One femoral and two trochanteric fractures occurred intraoperatively and were stabilized with titanium cerclage wires. Transient nerve paresis occurred in two patients. One patient sustained a hip dislocation; she underwent a closed reduction and had no more dislocations.
There was evidence of cobalt-chromium metallosis in the capsule and the periprosthetic tissues from all eleven hips that were revised because of either aseptic loosening or technical failure and in two of the three hips that underwent revision because of deep infection (Fig. 4). The metallosis was graded 1+ or 2+ according to the modified Mirra classification25 (Table II). Extensive necrosis within the new synovial membrane was observed in all fourteen revised hips. Moreover, a predominantly perivascular lymphoplasmacytic infiltrate was observed in the periprosthetic tissues retrieved during the revisions for aseptic indications (Fig. 5). The diffuse lymphoplasmacytic infiltrate was not clearly seen in the tissues retrieved during the revisions due to infection because of the simultaneous presence of inflammatory cells and abscess formation. No significant relationships were found among histologic evidence of metallosis, lymphocytic and plasma cell concentration, and the prevalence of periprosthetic osteolysis (Table II).
The survivorship rate (and 95% confidence interval) at nine years, with removal of the component for any cause as the end point, was 97% (71% to 97%) for the cups and 92% (73% to 98%) for the stems. When the end point was removal of the component for aseptic reasons (loosening or technical errors), the survivorship rate was 93% (74% to 98%) for the stems (Fig. 6) and 98% (92% to 100%) for the cups (Fig. 7).
Total hip arthroplasty with a metal-on-metal articulation was reintroduced as an alternative to the use of metal-on-polyethylene bearings because of the theoretical advantages of reduced wear and a lower prevalence of osteolysis. Recent clinical studies on the early outcomes of contemporary total hip arthroplasties with metal-on-metal bearings have demonstrated mostly favorable results without osteolysis12,17,27,28; however, some investigators have raised concerns about the use of metal-on-metal articulations5-9.
Retrieval studies of patients with metal-on-metal bearings have shown that the serum levels of cobalt and chromium ions were substantially higher than those in normal individuals without implants29-31. Released metal ions can activate the immune system by forming metal-protein complexes that are considered to be candidate antigens for eliciting hypersensitivity responses30,32-34. Although many studies have demonstrated loosening of total hip prostheses with first-generation metal-on-metal articulations in association with hypersensitivities to cobalt, nickel, and chromium34-36, it has not been demonstrated whether the devices failed because the patients had a preexisting metal hypersensitivity or whether the patients became hypersensitive to metal as a result of the failed implants.
Willert et al.9 examined periprosthetic tissues retrieved from patients who had undergone revision because of periprosthetic osteolysis after a primary total hip arthroplasty (Centerpulse, Winterthur, Switzerland) with a second-generation metal-on-metal articulation (Metasul; Zimmer, Warsaw, Indiana). They found an immunological response that was characterized by a diffuse and perivascularly oriented infiltration of lymphocytes accompanied by plasma cells and eosinophilic granulocytes, swelling of the vascular endothelium, localized bleeding, fibrin exudation, necrosis, and macrophages with drop-like inclusions.
In our series, all of the hips revised for aseptic reasons and two of the three hips revised in the presence of infection had histologic evidence of metallosis. The metallosis was associated with a predominantly perivascular lymphoplasmacytic infiltrate, periprosthetic osteolysis, and loosening. The metallosis was graded 1+ or 2+ according to the modified Mirra classification25, a finding that resembles the observations by Willert et al.9.
Periprosthetic osteolysis in hips with a contemporary metal-on-metal total hip prosthesis has not always been associated with loosening of the components7,9,37,38. Periprosthetic osteolysis was the most common radiographic finding in the short-term follow-up study by Park et al.7, who observed osteolytic lesions localized solely within the greater trochanter but reported no loosening of the stem. However, they did report perivascular infiltration of lymphocytes and mononuclear phagocytes in two revised hips. Thus, in the study by Willert et al.9, the study by Park et al.7, and our report, in which a similar contemporary cementless metal-on-metal total hip arthroplasty was used, the histologic findings associated with loosening suggest that hypersensitivity to the cobalt-chromium metal of the articulation was involved in the pathogenesis of periprosthetic osteolysis and subsequent aseptic loosening.
The source of metal debris has not been clearly defined in previous reports. The bearing surfaces of the prostheses that were retrieved in the study by Park et al.7 were inspected with a three-dimensional scanner, and no visible areas of wear were detected. This might be attributable to the so-called selfpolishing capacity of the metal surfaces, in which scratches can be worn smooth by additional joint movement27. Similarly, in our series, neither notch nor groove formation was apparent histologically in the neck of the titanium-alloy femoral component and there were no visible scratches on the surfaces of the cobalt-chromium femoral head or the cup. We speculated that the cobalt-chromium wear particles originated from the metal-on-metal articulation surfaces despite our inability to detect damage to those surfaces.
Willert et al.9 compared the quantity of metal particles in the periarticular tissues with the intensity of the lymphocytic infiltrate and could not establish an association between the magnitude of the immunological response and the quantity of the metallosis. In our series, the quantity of cobalt-chromium metal particles in the periprosthetic tissues also did not appear to be associated with the intensity of the lymphocytic infiltration or the degree of periprosthetic osteolysis.
Willert et al.9 and Jasty et al.38 reported that lymphocytic infiltration may also occur in the periprosthetic tissues around other cobalt-chromium-molybdenum-containing implants; therefore, the reaction that we reported with the second-generation metal-on-metal articulations is not a new phenomenon.
Although we did not perform hypersensitivity tests (e.g., the patch test7), we agree with Willert et al.9 that patients should be informed about the risk of becoming sensitive to a metal-on-metal articulation.
Prospective comparative randomized studies with longer observation periods will be necessary to show whether modern metal-on-metal articulations are advantageous and whether the association of histologically detected articular wear debris is related to the histologic findings in retrieved periprosthetic tissues. ▪
The authors did not receive grants or outside funding in support of their research for or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
Investigation performed at Orthopaedic Department, General Hospital “Agios Andreas,” Patras, Greece
1. . The surgery of the osteoarthritic hip. Br J Surg. 1958;45: 488-97.
2. , Chen SC. The statistics of the McKee-Farrar method of total hip replacement. Clin Orthop Relat Res. 1973;95: 26-33.
3. . Development of total prosthetic replacement of the hip. Clin Orthop Relat Res. 1970;72: 85-103.
4. , Watson-Farrar J. Replacement of arthritic hips by the McKee-Farrar prosthesis. J Bone Joint Surg Br. 1966;48: 245-59.
5. . Metal sensitivities and orthopaedic implants revisited: the potential for metal allergy with the new metal-on-metal joint prostheses. Br J Dermatol. 2003;148: 1089-93.
6. , Holden CA. Metal allergy resurfaces in failed hip endoprostheses. Contact Dermatitis. 2003;48: 49-50.
7. , Moon YW, Lim SJ, Yang JM, Ahn G, Choi YL. Early osteolysis following second-generation metal-on-metal replacement. J Bone Joint Surg Am. 2005;87: 1515-21.
8. , Buchhorn GH, Fayyazi A, Lohmann CH. Histopathologische Veränderungen bei Metall / Metall - Gelenken geben Hinweise auf eine zellvermittelte Überempfindlichkeit-vorläufige Untersuchungsergebnisse von 14 Fällen. Osteologie. 2000;9: 165-79.
9. , Buchhorn GH, Fayyazi A, Flury R, Windler M, Koster G, Lohmann CH. Metal-on-metal bearings and hypersensitivity in patients with artificial hip joints. A clinical and histomorphological study. J Bone Joint Surg Am. 2005;87: 28-36.
10. , Graham J, Pruitt L, Ries MD. Failure of a metal-on-metal total hip arthroplasty from progressive osteolysis. J Arthroplasty. 1999;14: 877-81.
11. , Pukkala E, Paavolainen P, Pulkkinen P, Riska EB. Cancer risk after metal on metal and polyethylene on metal total hip arthroplasty. Clin Orthop Relat Res. 1996;(329 Suppl): S280-9.
12. , Wan Z, Longjohn DB, Dubois B, Murken R. Total hip arthroplasty with use of the Metasul metal-on-metal articulation. Four to seven-year results. J Bone Joint Surg Am. 2000;82: 789-98.
13. , Gordiev K. Metal-on-metal articulation in total hip arthroplasty: preliminary results in 57 cases. Aust NZ J Surg. 1997;67: 634-6.
14. , Blum GW, Wait ME, Walker PS. Size and shape of metal particles from metal-on-metal total hip replacements. Trans Orthop Res Soc. 1996;21: 462.
15. . Indications, technical considerations, and early results with modern metal-on-metal couple in total hip arthroplasty. Semin Arthroplasty. 1998;9: 143-56.
16. . [Metal-metal total prosthesis of the hip joint: back to the future]. Z Orthop Ihre Grenzgeb. 1992;130: 306-9. German.
17. , Petsinis G, Repanti M. Zweymueller with metal-on-metal articulation: clinical, radiological and histological analysis of short-term results. Arch Orthop Trauma Surg. 2003;123: 5-11.
18. , Garrett M, Gronley JK, Mulroy SJ. Classification of walking handicap in the stroke population. Stroke. 1995;26: 982-9.
19. , Stamatakis M, Baikousis A, Petsinis G. Treatment of dysplastic and congenitally dislocated hips with the Zweymueller total hip prosthesis. Orthopedics. 2001;24: 465-71.
20. , Droutsas P, Piperos G, Michael A, Baikousis A, Stamatakis M. Course of bone mineral content changes around cementless Zweymueller total hip arthroplasty. A 4-year follow-up study. Arch Orthop Trauma Surg. 1997;116: 60-5.
21. , Engh CA, Gokcen E. Patterns of osteolysis around total hip components inserted with and without cement. J Bone Joint Surg Am. 1995;77: 432-9.
22. , McNeice GM, Amstutz HC. “Modes of failure” of cemented stemtype femoral components: a radiographic analysis of loosening. Clin Orthop Relat Res. 1979;141: 17-27.
23. , Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res. 1976;121: 20-32.
24. , Bowerman JW, Robinson RA, Riley LH Jr. Ectopic ossification following total hip replacement. Incidence and a method of classification. J Bone Joint Surg Am. 1973;55: 1629-32.
25. , Mirra JM, Campbell PA, Amstutz HC. Tissue reaction to metal on metal total hip prostheses. Clin Orthop Relat Res. 1996;(329 Suppl): S187-205.
26. , Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53: 457-81.
27. , Rieker CB, Kottig P. Analysis of 118 second-generation metal-on-metal retrieved hip implants. J Bone Joint Surg Br. 1999;81: 46-50.
28. , Wagner H. Medium-term results of a modern metal-on-metal system in total hip replacement. Clin Orthop Relat Res. 2000;379: 123-33.
29. , Skipor AK, Doorn PF, Campbell P, Schmalzried TP, Black J, Amstutz HC. Cobalt and chromium concentrations in patients with metal on metal total hip replacements. Clin Orthop Relat Res. 1996;(329 Suppl): S256-63.
30. , Lee PT, Arora A, Villar RN. Levels of metal ions after small- and large-diameter metal-on-metal hip arthroplasty. J Bone Joint Surg Br. 2003;85: 913-7.
31. , Bitzan P, Meisinger V, Kaider A, Gottsauner-Wolf F, Kotz R. Serum cobalt levels after metal-on-metal total hip arthroplasty. J Bone Joint Surg Am. 2003;85: 2168-73.
32. . T lymphocyte-mediated immune responses to chemical haptens and metal ions: implications for allergic and autoimmune disease. Int Arch Allergy Immunol. 2004;134: 186-98.
33. , Merritt K, Jacobs JJ. Metal sensitivity in patients with orthopaedic implants. J Bone Joint Surg Am. 2001;83: 428-36.
34. , Goodwin PG, Brostoff J. Metal sensitivity in patients with joint replacement arthroplasties. Br Med J. 1975;4: 374-5.
35. , Wilson JN, Scales JT, Kemp HB. Incidence of metal sensitivity in patients with total joint replacements. Br Med J. 1975;4: 376-8.
36. , Freeman MAR, Miller AJ, Vernon-Roberts B. Metal sensitivity as a cause of bone necrosis and loosening of the prosthesis in total joint replacement. J Bone Joint Surg Br. 1974;56: 626-42.
37. , Campbell P, Mirra J, Hooper JC, Schmalzried TP. Osteolysis in a cementless, second generation metal-on-metal hip replacement. Clin Orthop Relat Res. 2001;386: 159-65.
Copyright 2006 by The Journal of Bone and Joint Surgery, Incorporated
38. , Bragdon C, Jiranek W, Chandler H, Maloney W, Harris WH. Etiology of osteolysis around porous-coated cementless total hip arthroplasties. Clin Orthop Relat Res. 1994;308: 111-26.