Ten-year follow-up study of three alternative bearing surfaces used in total hip arthroplasty in young patients
Abstract
Aims
We present the ten-year data of a cohort of patients, aged between 18 and 65 years (mean age 52.7 years; 19 to 64), who underwent total hip arthroplasty. Patients were randomised to be treated with a cobalt-chrome (CoCr) femoral head with an ultra-high molecular weight polyethylene (UHMWPE), highly cross-linked polyethylene (XLPE) or ceramic-on-ceramic (CoC) bearing surface.
Patients and Methods
A total of 102 hips (91 patients) were randomised into the three groups. At ten years, 97 hips were available for radiological and functional follow-up. Two hips (two patients) had been revised (one with deep infection and one for periprosthetic fracture) and three were lost to follow-up. Radiological analysis was performed using a validated digital assessment programme to give linear, directional and volumetric wear of the two polyethylene groups.
Results
There was a significantly reduced rate of steady-state linear wear with XLPE (0.07 mm/yr) compared with UHMWPE (0.37 mm/yr) (p = 0.001). Volumetric wear was also significantly reduced in the XLPE group (29.29 mm3/yr) compared with the UHMWPE group (100.75mm3/yr) (p = 0.0001). There were six patients with UHMWPE who had non-progressive osteolysis and none in the XLPE group. All three bearing groups had significant improvements in 12-item short form health survey scores, Western Ontario and McMaster Universities Osteoarthritis Index score and Harris Hip Score. However, the improvement in HSS was significantly less in the UHMWPE group (p = 0.0188) than in the other two groups.
At ten years, the rates of volumetric and linear wear in the XLPE group remain low and predominantly below the estimated threshold for osteolysis (1 mm/yr). The rate of linear wear in the XLPE group was three times less than in the UHMWPE group at five-year follow-up and five times less at ten years. The rate of volumetric wear was also three times less in the XLPE group at ten years.
Conclusion
While CoC also performs well, XLPE at ten years remains a safe and excellent bearing option in young patients, with low rates of wear and no evidence of osteolysis.
Cite this article: Bone Joint J 2017;99-B:1590–5.
Wear debris is an inevitable by-product of any artificial hip joint.1,2 The particles are thought to be the principal factor in periprosthetic osteolysis, the loosening of components3,4 and the requirement for revision.5-7
In the early 1990s, highly cross-linked polyethylene (XLPE) was developed to minimise particulate wear compared with ultra-high molecular weight polyethylene (UHMWPE).8-11 Several authors have reported reduced volumetric wear with XLPE compared with UHMWPE five years post-operatively.12-15 However, a potential decrease in toughness, tensile strength and resistance to the propagation of fatigue cracks has been reported with XLPE, and little is known about its survivorship, function and wear profile into the second decade post-operatively.16-18
Our aim was to present the functional and radiological outcomes and overall implant survivorship of three bearing couples in a prospective randomised controlled trial (RCT) with a minimum ten-year follow-up. The results five years post-operatively have previously been reported.13
Patients and Methods
Between February 2003 and March 2005, 91 patients (102 hips) with a mean age of 52.7 years (19 to 64) underwent surgery at a single centre by one of three surgeons (JPW, EHS or EB). The techniques of randomisation and surgical approach have been previously described.13 In summary, cementless femoral (Synergy; Smith & Nephew, Memphis, Tennessee) and acetabular components (Reflection; Smith & Nephew) were used in all patients. They were randomised using opaque envelopes and a computer-generated block randomisation scheme (block size = 4), provided by the study co-ordinator on the day of surgery, to one of three combinations of bearing surfaces: 28 mm diameter cobalt-chrome head (CoCr) with UHMWPE liner, (n = 36); 28 mm diameter CoCr head with XLPE liner (n = 32) and a ceramic-on-ceramic (CoC) bearing, with 36 mm heads (n = 34). A standardised pre-operative evaluation involving the history, demographics and comorbidities and completion of the 12-item short form health (SF-12, physical component),19,20 the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)21 and the Harris Hip Score (HHS)22 self-reported questionnaires, was performed between one and two weeks pre-operatively.
Digital pre-operative radiological evaluation included an anteroposterior (AP) pelvic radiograph and AP and lateral hip radiographs. Each patient was reviewed clinically and radiologically at three months and at two, five and ten years post-operatively. AP pelvis and cross-table lateral views were obtained using a standardised protocol. The radiological assessment of wear was performed by a blinded independent company that had developed the Polyware digital analysis programme (Draftware Inc., Vevay, Indiana). This software creates a 3D model of the components based on the radiographs.13 The process has been validated,13 with an inter-observer error at a 95% confidence interval estimated to be ± 0.007679 mm and has been reported to be more reliable than 2D23 or stereoisometric analysis.13 Calibration is performed using standardised 28 mm femoral heads and the programme is compatible with the Smith & Nephew Reflection acetabular component, from which changes in volumetric and linear wear and directional analysis can be calculated. The accuracy of the measurements of linear wear of Polyware is ± 0.015 mm and the volumetric information is validated as ± 0.13mm3.14 We assessed the acetabular inclination using the automated programme as well as by manual assessment with reference to the trans-ischial line. Any osteolysis or radiolucent lines (RLLs) were assessed as described by Gruen et al24 for the femoral components and by DeLee and Charnley25 for the acetabular components. The digital technique is not, however, accurate when assessing non-polyethylene implants, so the CoC group was analysed only for function and survivorship.
Inclusion criteria included patients aged between 18 and 65 years undergoing total hip arthroplasty (THA) for primary or secondary osteoarthritis. Exclusion criteria included a history of sepsis affecting the hip, primary or secondary malignancy involving the hip, fracture of the femoral neck and bone deficiency requiring the use of graft. All patients were directed to follow standard precautions to avoid dislocation of the hip post-operatively and were encouraged to fully mobilise weight-bearing from the time of surgery.
This study was approved by the Research Ethics Boards of St. Michael’s Hospital and the University of Toronto.
Statistical analysis
The comparability of the baseline demographics, including age, gender, primary diagnosis, level of education, laterality, surgeon and pre-operative functional outcome measures (SF-12, WOMAC and HHS) of the three groups were confirmed using univariate techniques (including Mann-Whitney U and chi-squared tests). Analysis of variance was undertaken using the Kruskal-Wallis test to assess differences in the SF-12, WOMAC and HHS scores and the rates of linear wear between groups ten years post-operatively. A p-value of < 0.05 was considered significant for all analyses.
Results
Osteoarthritis was the indication for surgery in 67 hips (66%) (Table I). A total of 97 THAs (95%) in 87 patients were available for review at ten years (32 UHMWPE, 31 XLPE and 34 CoC). Two hips (two patients) had been revised, one for infection (UHMWPE) at 18 months and one for periprosthetic fracture (XLPE) four months post-operatively. Two patients in the UHMWPE group (three THAs) were lost to follow-up. There were no failures due to aseptic loosening in any group. Radiological assessment showed a non-progressive RLL in more than one acetabular zone in four patients all in the UHMWPE group.
| Diagnosis | XLPE | UHMWPE | CoC |
|---|---|---|---|
| Osteoarthritis | 19 | 23 | 25 |
| Avascular necrosis | 5 | 6 | 3 |
| Development dysplasia of the hip | 2 | 2 | 1 |
| Post-traumatic arthritis | 1 | 0 | 1 |
| Rheumatoid arthritis | 1 | 1 | 1 |
| Other | 4 | 4 | 3 |
| Total hips | 32 | 36 | 34 |
At ten years, the SF-12 (physical health component), WOMAC and HHS scores had significantly improved in all groups (Figs 1 to 4). However, the improvement in the HHS scores in the UHMWPE group was significantly less than in the other two groups (p < 0.038). (Table II).
Fig. 1 Graph showing pre- and post-operative 12-item short form (SF) health questionnaire scores with standard deviation in the three groups (XLPE, highly cross-linked polyethylene; UHMWPE, ultra-high molecular weight polyethylene; CoC, ceramic-on-ceramic).
Fig. 2 Graph showing pre- and post-operative Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores with standard deviation in the three groups (XLPE, highly cross-linked polyethylene; UHMWPE, ultra-high molecular weight polyethylene; CoC, ceramic-on-ceramic).
Fig. 3 Graph showing pre and post-operative Harris Hip Score (HHS) with standard deviation in the three bearing groups (XLPE, highly cross-linked polyethylene; UHMWPE, ultra-high molecular weight polyethylene; CoC, ceramic-on-ceramic).
Fig. 4 Graphic representation of pre- and post-operative outcome scores in the three groups (XLPE, highly cross-linked polyethylene; CoCr, cobalt-chrome; UHMWPE, ultra-high molecular weight polyethylene; CoC, ceramic-on-ceramic; pre-op, pre-operative; SF-12, 12-item short form health questionnaire; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index; HHS, Harris Hip Score).
| Bearing surfaces | n | SF-12 (pre-op) | SF-12 (10 yrs) | WOMAC (pre-op) | WOMAC (10 yrs) | HHS (pre-op) | HHS (10 yrs) |
|---|---|---|---|---|---|---|---|
| XLPE/CoCr | 29 | 32.35 (12.27) | 53.51 (7.99) | 59.00 (12.40) | 12.02 (16.50) | 49.13 (12.45) | 91.07 (9.87) |
| UHMWPE | 34 | 30.16 (7.42) | 50.87 (10.11) | 56.46 (12.12) | 21.76 (20.28) | 49.01 (10.43) | 81.91 (18.27) |
| CoC | 29 | 28.49 (7.16) | 52.93 (7.48) | 59.51 (12.30) | 12.72 (16.51) | 45.59 (10.95) | 86.27 (15.23) |
There were two early revisions, one due to deep infection at 18 months and one due to periprosthetic fracture at four months post-operatively.
At ten years, there were four patients in the UHMWPE group with evidence of acetabular osteolysis in more than one zone, all in DeLee and Charnley zones25 1 and 2, and two patients with femoral osteolysis in Gruen zones24 1 and 7. In the XLPE group, there were no patients with any evidence of osteolysis.
The mean total linear wear at ten years was 0.6 mm (sd 0.25) in the XLPE group and 1.57 mm (sd 0.77) in the UHMWPE group (p < 0.0001). Considering the bedding-in period (which occurs within the first six weeks and then slows down when the head has seated/caused the plastic to conform), this equated to a mean annual rate wear of 0.07 mm/year (sd 0.05) for the XLPE group and 0.37 mm/year (sd 0.47) for the UHMWPE group (p < 0.0014). The mean total volumetric wear at ten years was 217.81 mm3 (sd 143.56) in the XLPE group and 784.2 mm3 (sd 415.7) in the UHMWPE group (p < 0.0001). This equated to a mean annual volumetric wear rate of 29.29 mm3/year (sd 23.47) for the XLPE group and 100.75 mm3/year (sd 59.08) for the UHMWPE group (p < 0.0001, Table III).
| n* | XLPE | n† | UHMWPE | p-value | |
|---|---|---|---|---|---|
| Mean linear wear after 5 yrs, mm (sd) | 28 | 0.39 (0.22) | 32 | 0.73 (0.4) | 0.0002 |
| Mean linear wear after 10 yrs, mm (sd) | 26 | 0.6 (0.25) | 31 | 1.57 (0.77) | < 0.0001 |
| Mean linear wear, mm/yr (sd) | 27 | 0.07 (0.05) | 31 | 0.37 (0.47) | 0.0014 |
| Mean volumetric wear after 5 yrs, mm3 (sd) | 28 | 116.39 (111.82) | 32 | 346.75 (233.02) | < 0.0001 |
| Mean volumetric wear after 10 yrs, mm3 (sd) | 27 | 217.81 (143.56) | 31 | 784.16 (415.66) | < 0.0001 |
| Mean volumetric wear, mm3/yr (sd) | 27 | 29.29 (23.47) | 32 | 100.75 (59.08) | < 0.0001 |
| Acetabular component inclination angle (sd) | 27 | 45.68° (5.85°) | 32 | 45.37° (6.59°) | 0.8488 |
Regression analysis indicated that neither the mean inclination or anteversion of the acetabular component was related to the rate of wear in any group. The direction of wear was superior and lateral. Four hips (of 31, 12.9%) in the XLPE group had a rate of wear of > 0.1 mm/year compared with 26 (of 32, 81.2%) in the UHMWPE group (Figs 5 and 6). This is the threshold for osteolysis linked to the formation of debris.26
Fig. 5 Graph showing the comparison of cumulative rates of mean linear wear at two, five and ten years (XLPE, highly cross-linked polyethylene; UHMWPE, ultra-high molecular weight polyethylene).
Fig. 6 Graph showing the comparison of annual rates of mean linear wear for the two polyethylene bearing groups (XLPE, highly cross-linked polyethylene; UHMWPE, ultra-high molecular weight polyethylene).
While no wear analysis could be performed in the CoC group, no patients had osteolysis at ten years and the improvement in functional outcome was not statistically different from that in the XPLE group.
In the report of these patients published at five years,13 three in the CoC group had audible ‘squeaking’. These symptoms have persisted at the ten-year review.
The overall linear wear of ative steady state of wear over the ten-year period. However, the rate of wear in the UHMWPE group increased two-fold between five and ten years. After the initial ‘bedding-in period’, the annual linear wear rate of the XLPE group decreased, while that of the UHMWPE group remained in a steady state throughout the ten-year period (Fig. 6).
Discussion
We have previously reported a significant difference between the rates of linear wear of XLPE and UHMWPE in these patients at five years.13 In this review of the same cohort, using a more accurate method of analysis, we have shown persistent significant differences. While the annual wear rates of XLPE remained relatively stable, those of UHMWPE accelerated. In a systematic review of the wear rates of XLPE compared with conventional polyethylene, Kurtz et al27 reported a mean annual wear rate of 0.042 mm/year for XLPE at two years. The data for this review were collected from 28 studies in which there was heterogeneity of techniques of measurement, sizes and types of femoral head and manufacturers. The annual rates of linear wear of the XLPE group fitted the pattern previously described. There was an initial period of creep/‘bedding-in’ and then steady-state wear which continued up to ten years.28,29 Conversely, the linear rate of wear of the UHMWPE group increased after five years. While the volumetric wear also increased for the UHMWPE group after five years, the acceleration was less than the linear wear. Given the confidence intervals, the mean linear wear in the XLPE group of 0.07 mm/year is comparable with other studies; for example, 0.035 mm/year at 13 years,30 0.04 mm/year at 15 years31 and 0.03 mm/year at ten years.12 Rates of volumetric wear of > 100 mm3/year have been associated with implant failure,32 however at ten years, our XLPE group had rates of 29 mm3/yr.
All groups showed statistically significant improvements in functional outcome scores, although the improvement in the HHS in the UHMWPE group was unexpectedly less than in the other groups. This was an isolated finding though and the cohort was small so it could be due to a statistical error.
While the method of the analysis of wear could not be applied to the CoC group, there were no patients in this group who had osteolysis, nor were there any significant differences in outcome scores compared with the XLPE group.
The limitation of this study is the relatively small sample size, although a power calculation was performed prior to randomisation. However, it was a prospective RCT and included functional outcome data at ten years, used a validated method of analysis of wear which has been found to be accurate,13,23 and showed a statistically significant difference in the primary outcome measures.
In conclusion, early and mid-term rates of wear and clinical outcomes of CoCr femoral heads on XLPE acetabular liners in these active young patients are excellent and no new concerns or modes of failure have been observed. XLPE acetabular liners and CoCr femoral heads with a follow-up of ten years remains a safe and effective bearing option in young patients, in whom there was no evidence of osteolysis. However, following the poor results from using XLPE after ten years in one study comparing XLPE with UHMWPE in dysplastic hips,32 we will continue to monitor these patients closely.
Take home message:
- XLPE has a significantly lower rate of wear when compared with UHMWPE in the young population.
- XLPE remains safe to use into the second decade.
- CoC remains comparable with the metal versus XLPE in terms of survivorship.
References
- 1 . Mechanism and clinical significance of wear debris-induced osteolysis. Clin Orthop Relat Res 1992;276:7–18. Google Scholar
- 2 . The problem is osteolysis. Clin Orthop Relat Res 1995;311:46–53. Google Scholar
- 3 , Macrophage integrins modulate response to ultra-high molecular weight polyethylene particles and direct particle-induced osteolysis. Biomaterials 2017;115:128–140. Crossref, Medline, ISI, Google Scholar
- 4 , Periprosthetic osteolysis after total hip replacement: molecular pathology and clinical management. Inflammopharmacology 2013;21:389–396. Crossref, Medline, ISI, Google Scholar
- 5 , Reasons for revision of first-generation highly cross-linked polyethylenes. J Arthroplasty 2010;25(suppl):67–74. Crossref, ISI, Google Scholar
- 6 , Total hip arthroplasties: what are the reasons for revision? Int Orthop 2008;32:597–604. Crossref, Medline, ISI, Google Scholar
- 7 . Reasons for revision hip surgery: a retrospective review. Clin Orthop Relat Res 2004;429:188–192. Crossref, ISI, Google Scholar
- 8 . Development of an extremely wear-resistant UHMWPE for total hip replacements. J Orthop Res 1999;17:157–167. Crossref, Medline, ISI, Google Scholar
- 9 . Ex vivo wear of conventional and cross-linked polyethylene acetabular liners. Clin Orthop Relat Res 2005;438:158–164. Crossref, Medline, ISI, Google Scholar
- 10 . Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res 1999;17:157–167. Crossref, Medline, ISI, Google Scholar
- 11 , Unified wear model for highly crosslinked ultra-high molecular weight polyethylenes (UHMWPE). Biomaterials 1999;20:1463–1470. Crossref, Medline, ISI, Google Scholar
- 12 , The John Charnley Award: Highly crosslinked polyethylene in total hip arthroplasty decreases long-term wear: a double-blind randomized trial. Clin Orthop Relat Res 2015;473:432–438. Crossref, Medline, ISI, Google Scholar
- 13 . A prospective randomised controlled trial comparing three alternative bearing surfaces in primary total hip replacement. J Bone Joint Surg [Br] 2012;94-B:459–465. Link, Google Scholar
- 14 . Highly crosslinked polyethylene reduces wear in total hip arthroplasty at 5 years. Clin Orthop Relat Res 2010;468:3228–3233. Crossref, Medline, ISI, Google Scholar
- 15 . Low wear of a second-generation highly crosslinked polyethylene liner: a 5-year radiostereometric analysis study. Clin Orthop Relat Res 2013;471:3596–3600. Crossref, Medline, ISI, Google Scholar
- 16 , Comparison of cross-linked polyethylene materials for orthopaedic applications. Clin Orthop Relat Res 2003;414:289–304. Crossref, ISI, Google Scholar
- 17 , Tradeoffs amongst fatigue, wear, and oxidation resistance of cross-linked ultra-high molecular weight polyethylene. J Mech Behav Biomed Mater 2011;4:1033–1045. Crossref, Medline, ISI, Google Scholar
- 18 . Deformation, yielding, fracture and fatigue behavior of conventional and highly cross-linked ultra high molecular weight polyethylene. Biomaterials 2005;26:905–915. Crossref, Medline, ISI, Google Scholar
- 19 . A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220–223. Crossref, Medline, ISI, Google Scholar
- 20 . Assessment tools: functional health status and patient satisfaction. Am J Med Qual 1996;11:S50–S53. Medline, Google Scholar
- 21 . Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15:1833–1840. Medline, ISI, Google Scholar
- 22 . Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J Bone Joint Surg [Am] 1969;51-A:737–755. Crossref, ISI, Google Scholar
- 23 . Two-dimensional versus three-dimensional radiographic measurements of polyethylene wear. Clin Orthop Relat Res 1999;365:117–123. Crossref, Google Scholar
- 24 . “Modes of failure” of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop Relat Res 1979;141:17–27. Google Scholar
- 25 . Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976;121:20–32. Google Scholar
- 26 . A literature review of the association between wear rate and osteolysis in total hip arthroplasty. J Arthroplasty 2002;17:649–661. Crossref, Medline, ISI, Google Scholar
- 27 . History and systematic review of wear and osteolysis outcomes for first-generation highly crosslinked polyethylene. Clin Orthop Relat Res 2011;469:2262–2277. Crossref, Medline, ISI, Google Scholar
- 28 . Influence of polyethylene creep behavior on wear in total hip arthroplasty. J Orthop Res 2006;24:422–427. Crossref, Medline, ISI, Google Scholar
- 29 . Finite element simulation of early creep and wear in total hip arthroplasty. J Biomech 2005;38:2365–2374. Crossref, Medline, ISI, Google Scholar
- 30 . Highly cross-linked polyethylene decreases the rate of revision of total hip arthroplasty compared with conventional polyethylene at 13 years’ follow-up. Bone Joint J 2016;98-B:28–32. Link, ISI, Google Scholar
- 31 , Fifteen-year comparison of wear and osteolysis analysis for cross-linked or conventional polyethylene in cementless total hip arthroplasty for hip dysplasia: a retrospective cohort study. J Arthroplasty 2017;32:161–165. Crossref, Medline, ISI, Google Scholar
- 32 . The association between rates of wear in retrieved acetabular components and the radius of the femoral head. Proc Inst Mech Eng H 1998;212:321–326. Crossref, Medline, ISI, Google Scholar
Author contributions:
A. Atrey: Writing the paper.
S. Ward: Co-editing the paper, Data research.
A. Koshbin: Co-editing the paper, Data research.
N. Hussain: Co-editing the paper, Statistical analysis.
E. Bogoch: Co-editing the paper, Primary surgeon.
E. H. Schemitsch: Co-editing the paper, Primary surgeon.
J. P. Waddell: Primary editor of the paper, Primary surgeon.
The Orthopaedic Department at St Michael’s Hospital receives an annual stipend of $45 000 Ca from Smith & Nephew which is used to fund a fellowship post for a trainee. Additionally, a research grant of $1000 was received for this project.
Both E. H. Schemitsch and J. P. Waddell are paid experts and lecturers for Smith & Nephew.
None of the other authors have any conflicts of interest nor do they have other declarations.
The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other nonprofit organisation with which one or more of the authors are associated.
This article was primary edited by S. P. F. Hughes and first proof edited by J. Scott.
