 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. The Contribution of Frictional Torque to Loosening at the Cement-Bone Interface in Tharies Hip Replacements*  , LOS ANGELES, CALIFORNIAInvestigation performed at the Joint Replacement Institute, Orthopaedic Hospital, Los Angeles
We evaluated the results in a series of 170 osteoarthrotic hips (156 patients) at a median of 10.0 years (range, 3.0 to 16.4 years) after a Tharies resurfacing arthroplasty performed with cement. Our purpose was to assess the role of frictional torque in loosening at the cement-bone interface. The hips were divided into three groups on the basis of the diameter of the bearing surface: small (thirty-six, thirty-nine, or forty-three millimeters), medium (forty-seven millimeters), and large (fifty-one or fifty-four millimeters). Comparisons were made with Kaplan-Meier survivorship analysis, with stepwise discriminant analysis of risk factors for aseptic loosening. Revision for aseptic loosening was the end point for survivorship analysis. Despite higher frictional torques due to the increased diameter of the bearing surface and the increased average load, the larger prostheses survived significantly longer than the smaller ones (p = 0.002). Stepwise covariate discriminant analysis indicated that the size of the bearing surface was the only factor identified that significantly affected survival. In six patients who had had a bilateral surface replacement with a component of a different size in each hip, the smaller prosthesis always failed first (p = 0.001), at an average of seventy-seven months; the larger prosthesis failed at an average of 113 months. Nine patients had had a bilateral replacement with components of the same size, and there was no significant difference in the durations of survival of the two prostheses. Analysis of radiographs and retrieved specimens indicated that, regardless of the size of the component, the mechanism of loosening on both the acetabular and the femoral side of this so-called double-cup replacement was progressive resorption of bone induced by polyethylene wear particles that compromised the fixation of the components. More time was required for the process to disrupt the larger fixation area of the larger components. These data indicate that frictional torque was not the primary factor in the loosening of these prostheses with a large bearing surface and that higher friction and frictional torques can be tolerated if the generation of wear debris is sufficiently limited. These findings may be important as alternatives to polyethylene bearing surfaces are investigated.
Charnley's use of a small (twenty-two-millimeter-diameter) metal-on-plastic bearing surface for total hip arthroplasty was based on the premise that if frictional torque is minimized, the longevity of fixation of the implant increases. Experience has demonstrated a higher prevalence of loosening of hip prostheses with femoral heads of larger diameters, and this observation has been thought by some to be due to increased frictional torque7,17. However, the volumetric wear of polyethylene increases with the square of the radius of the bearing surfaces13, and polyethylene wear debris has been identified as a major factor in periprosthetic bone resorption and loosening of implants20-22. Detailed biomechanical, radiographic, and histological studies of retrieved cemented total hip replacements have not implicated frictional torque in the initiation of loosening of femoral or acetabular components12,16,21,22. To determine the contribution of frictional torque to loosening at the cement-bone interface, we studied the survival of Tharies (total hip articular replacement with internal eccentric shells) resurfacing components (Zimmer, Warsaw, Indiana) that had been inserted with cement. This implant is an appropriate clinical model for this determination because the large diameter of the resurfacing components results in increased frictional torque15 and higher volumetric wear of polyethylene13 compared with those associated with conventional total hip replacements with bearing surfaces of smaller diameters. Also, variables related to the design of the stem, the position of the stem within the femoral canal, debonding between the stem and the cement, the adequacy of the cement mantle around the femoral component19, and impingement of the neck on the acetabular rim25 are eliminated with the use of resurfacing arthroplasties.
Three hundred and thirty hips were resurfaced with use of Tharies components with cement fixation by one of us (H. C. A.), between July 1975 and October 1984. The Tharies system included a femoral component made of cobalt-chromium and an acetabular component made of ultra-high molecular weight polyethylene. The thickness of the femoral shell was two millimeters. The component was available in six sizes, based on the diameter of the bearing surface: size 1 (thirty-six millimeters), size 2 (thirty-nine millimeters), size 3 (forty-three millimeters), size 4 (forty-seven millimeters), size 5 (fifty-one millimeters), and size 6 (fifty-four millimeters) (Table I). The operative technique for Tharies surface replacement has been described previously1. The polyethylene acetabular component was eccentrically shaped and inserted with the maximum five-millimeter thickness in all sizes aligned with the weight-bearing axis. The smallest possible femoral component was chosen in order to conserve bone in the acetabulum. Instrumentation was designed to obtain a 1.5-millimeter-thick mantle of cement between the components and bone.
The largest single group based on diagnosis consisted of 170 hips (in 156 patients) that had been resurfaced as treatment for osteoarthrosis (Table I). For survivorship analysis, the component sizes were grouped together as small (sizes 1, 2, and 3; fifty-three hips), medium (size 4; sixty-eight hips), or large (sizes 5 and 6; forty-nine hips). The patients were evaluated clinically and radiographically at yearly intervals. The average age of the patients at the time of the resurfacing arthroplasty in all three groups was about sixty years (range, forty-two to eighty-three years for the patients who had a small component, twenty-seven to seventy-seven years for those who had a medium component, and forty-five to seventy-seven years for those who had a large component; Table II). As expected, larger components were associated with an increased average weight of the patients. The average weight (and standard deviation) was 69.2 ± 18.3 kilograms for the patients who had a small component, 81.2 ± 17.9 kilograms for those who had a medium component, and 91.4 ± 21.2 kilograms for those who had a large component (Table II). Forty-eight of the small components were in women, compared with five in men; seventeen of the medium components were in women, compared with fifty-one in men; and all forty-nine of the large components were in men.
In the over-all series of 330 hips in which a Tharies replacement had been performed with cement by one of us, fifteen patients had a bilateral replacement with later revision of both hips. This subset of patients is of interest for testing of the hypothesis because patient-dependent variables are controlled. In six patients, Tharies components of different sizes were implanted. The average age of these six patients was fifty-two years, and the average weight was sixty-eight kilograms. In two hips, the smaller component was implanted first; however, over-all the larger component was implanted an average of 15.5 months before the smaller one. The other nine patients had the same size of component in both hips. The average age of these patients was fifty years, and the average weight was seventy-nine kilograms. The second component was inserted an average of seven months after the first. The primary mode of failure of all of the revised hips was recorded according to the classification of Amstutz et al.2. The history, radiographs, findings at the revision operation, and retrieved components and tissues were all used to classify the mode of failure as primarily loosening of the acetabular component (1A), primarily loosening of the femoral component (1B), loosening of both components (1C), or another mode. The indications for revision were pain and a reduction in the function of the hip with radiographic evidence of loosening of the component, such as a change in the position of either the femoral or the acetabular component or extensive radiolucent lines about the acetabular component and resorption of bone. In the eighty revisions that were performed by us, both the femoral and the acetabular component were routinely retrieved as part of an ongoing implant-retrieval analysis program. Residual bone from the femoral head and neck as well as any interface membrane or other reactive tissue were retrieved with the femoral component. Pseudocapsular tissue and membrane from the acetabular interface were also retrieved. The specimens were processed for analysis with light microscopy with the use of routine histological techniques. Comparison of the groups based on the size of the bearing surface was performed with Kaplan-Meier survivorship analysis, with revision for aseptic loosening as the end point. The hips were censored for death of the patient, infection, and loss to follow-up. Statistical analyses for survivorship were performed with the Mantel-Cox generalized salvage model (BMDP-90; BMDP Statistical Software, University of California, Berkeley, California). Stepwise regression analysis with the Cox proportional hazards model (BMDP-90) was used to assess the influence of the age, weight, and sex of the patient; the date of the operation; and the size of the bearing surface on survival. A paired Student t test was used to compare the times to revision of bilateral Tharies components.
Sixty of the 170 hips treated for osteoarthrosis had not been revised and were still being followed at an average of eleven years (range, 5.5 to 16.0 years) after the operation. Eight hips were lost to follow-up, and twenty-two were in patients who had died. Thirty-two of the fifty-three hips with a small component were revised (Table II). The mode of failure was 1A for twelve, 1B for fourteen, 1C for two, and another mode for four. Thirty of the sixty-eight hips with a medium component were revised. The mode of failure was 1A for fourteen, 1B for ten, 1C for two, infection for one, and another mode for three. Eighteen of the forty-nine hips that had a large component were revised. The mode of failure was 1A for eleven, 1B for three, 1C for two, infection for one, and another mode for one. The median survival time (the time to 50 per cent failure due to aseptic loosening of either the femoral or the acetabular component) was 118 months for the small components, 146 months for the medium components, and 197 months for the large components. On the basis of the Kaplan-Meier survivorship curves, the large prostheses lasted significantly longer (p = 0.002) than the small ones, with an over-all twelve-year rate of survival of 59 per cent for the large prostheses compared with only 39 per cent for the smaller ones (Fig. 1). Stepwise regression analysis with the Cox proportional hazards model indicated that the size of the prosthesis was the only factor identified that significantly affected survival. Factors that did not significantly affect survival were the age (p = 0.19), weight (p = 0.50), and sex (p = 0.77) of the patient and the date of the operation (p = 0.15).
Analysis of hips with primary loosening of the femoral component indicated that the large components lasted significantly longer than the small ones, with a twelve-year rate of survival of 82 and 58 per cent, respectively (p = 0.003) (Fig. 2). The rate of loosening of the small acetabular components was significantly higher than that of the large acetabular components at ten years (p <0.01), but at twelve years and thereafter the difference was not significant (p > 0.1) (Fig. 3).
For the six patients who had had bilateral replacement and subsequent revision of Tharies components of different sizes, the smaller component lasted an average (and standard deviation) of 77 ± 36 months, compared with 113 ± 37 months for the larger component. This difference is highly significant (p = 0.001). In contrast, in the nine patients who had had bilateral replacement and subsequent revision of Tharies components of the same size, the two components had similar longevity (average, 104 ± 29 months for the right and 101 ± 37 months for the left), with the component that was inserted first lasting only slightly longer than the one inserted second (111 ± 33 compared with 94 ± 36 months). This difference is not significant (p = 0.12). Twenty-four femoral components and associated soft tissue retrieved at the time of the revision operation were analyzed. Independent of the size of the bearing surface, soft tissue that was grossly similar to the pseudocapsule was present at the intra-articular margins. This soft tissue appeared to have replaced the bone of the femoral neck, resulting in a reduction of the cross-sectional area of bone. As has been reported previously10, similar tissue was present proximally under the femoral shell, resulting in variable degrees of disruption of the cement-bone interface. This tissue consisted of variable amounts of fibroblasts and organized connective tissue and an inflammatory foreign-body response characterized by plump macrophages and occasional giant cells. The phagocytic cells had occasional round or oval vacuoles, often containing barium sulfate crystals, consistent with cement particles. Under polarized light, occasional large shards of polyethylene could be identified easily. Additionally, there was a very fine diffuse birefringence in the cytoplasm of the phagocytic cells, indicating the presence of multiple submicrometer particles of polyethylene8. Osteocytes in the remaining bone of the femoral head and neck indicated viability. Osteoclastic bone resorption was seen, as was direct bone resorption by debris-laden macrophages. The interface membranes from sixteen acetabula were also analyzed. The findings on histological examination were essentially identical to those associated with the femoral pseudocapsular tissues.
Charnley's principle of low-friction arthroplasty is based on the assumption that the reduction of frictional torque would lessen loosening of the implant. The frictional torque of a Charnley prosthesis with a load of 890 newtons has been reported to be 0.4 to 1.2 newton-meters3,24. Under similar conditions, the average frictional torque of a twenty-eight-millimeter prosthesis, a Tharies size-3 component (the largest component in our small group), and a Tharies size-5 component (the smaller component in our large group) was 1.3, 2.7, and 3.2 newton-meters, respectively15. Although these values are twenty to 100 times smaller than the reported static torques to failure for cemented acetabular components3,6,15,23, one of the concerns regarding surface replacement arthroplasty has been that the large diameter of the bearing surface is not consistent with the principle of low frictional torque. The reduced durability of resurfacing components inserted with cement as compared with conventional (stemmed) total hip replacements inserted with cement18 appears to support Charnley's premise of the importance of low frictional torque. However, if frictional torque is the main factor responsible for loosening of the implant, then resurfacing components with a large diameter should fail before those with a smaller one. We did not observe this. The large components survived significantly longer than the small components (p = 0.002), despite a higher frictional torque due to an approximately 45 per cent larger potential contact area of the bearing surface, a higher load (an average of an additional twenty-three kilograms of body weight), and a predominance of men with large components. Furthermore, the smaller prosthesis always failed significantly sooner (p = 0.001) in the patients who had had bilateral replacement with Tharies components of different sizes, while there was no difference in the survival of the prostheses when Tharies components of the same size had been used for the bilateral replacement (p = 0.12). Clinical experience with resurfacing arthroplasty of the hip has been characterized by relatively early loosening, which is often accompanied by substantial loss of acetabular bone stock4,9,11. In an analysis of aseptic failure, Bell et al. observed a marked foreign-body response to wear products that was associated with active bone resorption at the acetabular bone-membrane interface. Studies22 of well functioning total hip replacements retrieved at autopsy have subsequently shown that the degree of loosening of the polyethylene acetabular component was related to the volumetric wear of that component. Progressive disruption in three dimensions of the cement-acetabular bone interface, which can result in loosening of the component, is driven by the inflammatory response to polyethylene wear particles22. This process appears to be accelerated with surface replacements. The large bearing surface of these prostheses results in rates of volumetric wear that are four to ten times higher than those of conventional total hip replacements with bearing surfaces with a twenty-eight-millimeter diameter12. When the femoral side of a resurfacing arthroplasty is revised, the femoral neck is transected distal to the level of the resurfacing component and the specimen that is obtained includes the prosthesis, the cement-bone interface, and any associated reactive or inflammatory soft tissue. This specimen is similar to that retrieved at autopsy because the fixation of the component and the interface tissues are not disturbed by the retrieval process. Howie et al.10 analyzed seventy-two femoral components retrieved at revision and observed polyethylene wear particles and a macrophage response at the cement-bone interface without loosening of the component. Their findings indicate that wear particles can migrate along cement-bone interfaces that are macroscopically and microscopically solid and emphasize the important role of wear particles in loosening at the cement-bone interface. The results of our retrieval analyses agree. Thus, it appears that a similar mechanism of degradation of the cement-bone interface occurs on both sides of the articulation in a so-called double-cup hip replacement, and it is an example of the adverse effects of excessive polyethylene wear debris in the effective joint space20. Because of the completely intra-articular location of the implant-bone interfaces, the inflammatory foreign-body response to wear debris undermines femoral as well as acetabular fixation by progressive three-dimensional resorption of the bone at the cement-bone interface. The large prostheses have approximately 37 per cent more area for acetabular fixation and approximately 49 per cent more area for femoral fixation than the small ones (Table III). Our data indicate that more time is required, on the average, for the process to disrupt the larger fixation area of the larger components. Additional support for this mechanism comes from the analysis of the patients who had had a bilateral hip resurfacing arthroplasty and subsequent revision. In the patients who had two sizes of components, the smaller component always failed significantly sooner than the larger component (p = 0.001). There was no difference in the survival of the components in the patients who had had a bilateral resurfacing arthroplasty with components of the same size (p = 0.12).
It is likely that several factors are responsible for the superior survival of the femoral components compared with that of the acetabular components. These include the type and quality of bone, with the femur having more cortical and denser bone. Also, it is easier to keep the proximal aspect of the femur free of blood and to pressurize the cement into it at the time of implantation. Thus, the initial cement-bone interface and fixation of the femoral component are likely to be superior to those of the acetabular component. The time to failure for both the femoral and the acetabular components increased as their sizes increased, but the effect was greater for the femoral components. This may be related to the greater increase in the area of fixation on the femoral side for the larger components. The area of acetabular fixation also increases but not to the same extent; the estimated ratio of the area of fixation of the acetabulum to that of the femur is 7.8 for size-1 components and 5.5 for size-6 components (Table III). Our data and observations indicate that the frictional torques associated with these resurfacing components were well tolerated. While we believe that the reduction of frictional torque is correct in principle, in clinical practice other aspects of joint-replacement technology have set the limitations for durability of the implant. Perhaps the most striking example was Charnley's experience with acetabular components made of Teflon (polytetrafluoroethylene). Although the metal-on-Teflon bearing surface had the theoretical advantage of very low friction, the early failure of these implants was characterized by high rates of wear and extensive periprosthetic resorption of bone. Acetabular bearing surfaces of ultra-high molecular weight polyethylene offer better wear characteristics than Teflon but, with time, the detrimental effects of polyethylene wear debris have also been realized. Accumulating evidence, including the present experience with resurfacing arthroplasty, indicates that polyethylene wear has a greater effect on the durability of fixation of the implant than does frictional torque. From this perspective, the success of the Charnley low-friction arthroplasty is more a function of the low volumetric wear of the twenty-two-millimeter bearing surface. Our data indicate that higher friction at the bearing surface and higher frictional torques can be tolerated if the release of wear particles into the periprosthetic tissues is sufficiently limited. This finding may be important as alternatives to polyethylene bearing surfaces are investigated.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was The Los Angeles Orthopaedic Foundation, Los Angeles, California.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |