What Is the Long-term Survival for Primary THA With... : Clinical Orthopaedics and Related Research®

• Updated on: 2018-05-24
• Read original article here

Background Large-head metal-on-metal (MoM) bearing hip replacements have been shown to have a much higher rate of revision than other bearing surfaces. However, small-head (≤ 32 mm) MoM bearing surfaces have been in use for many years with several reports of satisfactory mid- to long-term survivorship. It is unclear whether the long-term survival of small-head MoM devices will continue to be satisfactory or whether the same concerns seen with the large-head MoM devices will ultimately become more prevalent.

Questions/purposes We analyzed a large national registry to ask: (1) What is the 15-year Kaplan-Meier survivorship of primary conventional THA using small-head (≤ 32 mm) MoM bearing surfaces compared with large-head MoM bearing surfaces in primary THA? (2) Is there an increased rate of revision for adverse reaction to metal debris (ARMD) in this group of patients over time?

Methods The Australian Orthopaedic Association National Joint Replacement Registry longitudinally maintains data on all primary and revision joint arthroplasties with nearly 100% capture. We analyzed all conventional primary THAs performed from Registry inception in September 1999 until December 31, 2015, in patients with a diagnosis of osteoarthritis and using MoM bearing surfaces ≤ 32 mm in diameter, defined as small-head MoM. The study group included 4838 primary THA with ≤ 32-mm MoM bearing surfaces. There were 2506 (51.8%) male patients and the median age of patients undergoing THA with a small-head MoM bearing surface was 64 years (range, 20-92 years of age). The outcome measure was the cumulative percent revision defined as the time to first revision using Kaplan-Meier estimates of survivorship at 15 years; reasons for revision and type of revision were also examined. We specifically investigated whether there was an increased risk of revision for ARMD in this MoM group compared with all other bearing surfaces. We compared these results with large-head MoM THAs (femoral head size > 32 mm).

Results The cumulative percent revision for small-head MoM designs at 15 years was 8.5% (95% confidence interval [CI], 7.3-9.9). The cumulative percent revision for large-head MoM at 14 years was 27.4% (95% CI, 24.8-30.2). Prostheses with a large-head MoM articulation have a higher rate of revision than small-head MoM bearing surfaces (hazard ratio after 6 years, 5.14; 95% CI, 4.1-6.5; p < 0.001). Over time, there was a gradual increase in the diagnosis of ARMD for small-head MoM and the cumulative incidence of revision for ARMD was 0.8% at 15 years.

Conclusions Despite survival that is substantially greater than that of large-head MoM THAs, there has been a marked decrease in the use of small-head MoM designs in our registry. Although the reasons for this are likely multifactorial, the increasing incidence of revisions for ARMD among small-head MoM THAs is concerning.

Each author certifies that he or she, or a member of his or her immediate family, has no funding or commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article.

Metal-on-metal (MoM) bearing surfaces have been used for conventional primary THA for > 50 years [ ]. With the advent of the Charnley low-friction arthroplasty, metal-on-polyethylene became the dominant bearing surface for THA and the use of MoM prostheses declined in the mid-1970s. However, concerns with polyethylene wear-related issues, especially in younger patients, resulted in renewed interest in alternative bearings, including MoM bearing surfaces. Anecdotal observation of satisfactory long-term results of the McKee-Farrar THA [] led to further research to develop second-generation MoM bearing articulations. One small-head MoM (defined in our registry as those metal bearings having a femoral head of ≤ 32 mm in diameter) design, the Metasul™ (Zimmer-Biomet, Winterthur, Switzerland) bearing couple [ ], was introduced in 1983 after improvements in metallurgy, tribology, frictional moments, and mechanical testing. There have been several reports of good midterm and longer term results of Metasul THA with small-head MoM bearings [ ]. However, there have also been reports of osteolysis in patients with small-head MoM implants [ ] thought to be an immunologic response to metal wear debris particles. The Registry has recorded a steady decline in the use of prostheses with a small-head MoM bearing surface over the past few years with surgeons opting instead to use designs with ceramic-on-ceramic or ceramic- or metal-on-crosslinked polyethylene bearing surfaces.

Large-head diameter MoM conventional primary THA (here defined as femoral head size > 32 mm) was introduced to aid in revision of femoral neck fractures after hip resurfacing arthroplasty and to provide the potential benefits of an increased head size with regard to ROM, impingement, and increased stability. Unfortunately, as a result of higher than expected rates of revision reported by national registries, and increasing reports of adverse reaction to metal debris (ARMD) [ ], regulatory authorities in the USA, United Kingdom, and Australia issued Medical Device Alerts [ ], certain devices were recalled, and the use of large-head MoM THA prostheses has largely been discontinued.

It is unclear whether the long-term survival of the small-head MoM devices will continue to be satisfactory or whether the same concerns seen with the large-head MoM devices will ultimately become more prevalent. We therefore analyzed the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) database to ask: (1) What is the long-term survival rate of primary conventional THA using small-head (≤ 32 mm) MoM bearing surfaces compared with large-head MoM bearing surfaces in primary THA? (2) Is there an increased rate of revision for ARMD metal-related pathology in this group of patients over time?

The AOANJRR started data collection on September 1, 1999, and achieved complete national implementation by mid-2002. Since then, the AOANJRR has collected data on almost 100% of hip arthroplasties performed in Australia. AOANJRR data are externally validated against patient-level data provided by all Australian state and territory health departments. A sequential, multilevel matching process is used to identify any missing data that is subsequently retrieved by contacting the relevant hospital. Each month, in conjunction with internal validation and data quality checks, all primary procedures are linked to any subsequent revision involving the same patient, same joint, and same side. Data are also matched biannually with the Australian Government’s National Death Index to obtain information on the date of death. Linking revision and death to the primary procedure enables revision rates to be determined.

The study population was all primary THAs performed for osteoarthritis reported to the AOANJRR between September 1999 and December 31, 2015, that had a small-head (defined as ≤ 32 mm) MoM articulation. All procedures using exchangeable (modular) femoral necks were excluded. The cumulative percent revision of procedures with a primary THA with a small-head MoM bearing was determined and reasons for revision and types of revision were examined. The study group included 4838 primary THAs with ≤ 32-mm MoM bearing surfaces and 94% were inserted with cementless fixation. There were 2506 (52%) male patients and the median age of patients undergoing THA with a small-head MoM bearing surface was 64 years (range, 20-92 years). The Metasul bearing couple (Zimmer) comprised 78% of all the THA with a small-head MoM bearing surface (Table 1).

The study group was compared with 13,403 primary THAs with large-head MoM bearing surfaces of which 88% were inserted with cementless fixation. There were 7787 (58%) male patients and the median age of the patients was also 64 years (range, 13–98 years). The median followup of the small-head MoM group was 9.7 years compared with 7.6 years for the large-head group.

We used Kaplan-Meier estimates of survivorship at 15 years to describe the time to first revision of an arthroplasty with censoring at the time of death or closure of the database at the time of analysis. When death is not a major competing risk, as was true with this study population, we prefer to use Kaplan-Meier rather than competing risks analysis [ ]. The Registry defines revisions as reoperations of previous hip replacements where one or more of the prosthetic components are replaced, removed, or one or more components are added. The unadjusted cumulative percent revision with an accompanying 95% confidence interval (CI) was calculated with use of unadjusted pointwise Greenwood estimates. Hazard ratios using Cox proportional hazard models adjusting for age and sex were used to compare the rate of revision among bearing surface groups. The assumption of proportional hazards was checked analytically for each model; if the interaction between the predictor and the log of the postoperative time was significant in the standard Cox model, then a time-varying model was used. Where possible, the cumulative percent revision was performed at the longest time period with available data. For small-head MoM, this was 15 years, but for large-head MoM, this was 14 years as the number at risk at 15 years was too small (n = 17 procedures). Therefore, for comparative analysis between these two groups, the longest followup (14 years) was used. All tests were two-tailed at the 5% level of significance. Statistical analysis was performed using SAS software, Version 9.3 (SAS Institute Inc, Cary, NC, USA).

The cumulative percent revision of the small-head MoM designs at 15 years was 8.5% (95% CI, 7.3-9.9). Prostheses with a large-head MoM articulation have a higher rate of revision than small-head MoM bearing surfaces in a similar timeframe. The cumulative percent revision at 14 years was 27% for large-head compared with 8% for small-head MoM bearing surfaces in the same time period (hazard ratio after 6 years, 5.14; 95% CI, 4.1-6.5; p < 0.001; Fig. 1).

The most common reasons for revision for small-head MoM were loosening, dislocation, infection, and fracture (Table 2). There was, however, a gradual increase with time for the diagnosis of ARMD (Fig. 2) and the cumulative incidence of revision for metal-related pathology was 0.8% (95% CI, 0.8-1.4) at 15 years. This was not observed with other bearing surfaces with head size ≤ 32 mm (Table 3). By contrast, the cumulative incidence revision for ARMD for large-head MoM at the longest followup of 14 years was 12.4% (95% CI, 11.4-13.7).

This study examined the long-term cumulative percent revision of prostheses with a small-head MoM bearing surface and compared this with the results of prostheses with large-head MoM designs using data from a large national joint replacement registry. The reasons for revision for small-head MoM prostheses were also assessed to determine whether the same concerns with regard to ARMD seen with large-head MoM were becoming more prevalent. This topic is of major importance because there has been a large number of THAs with a MoM bearing surface implanted worldwide with significant differences in the rates of revision between small and large femoral head devices [ ]. The cumulative percent revision at 14 years was 8% for small-head MoM bearing surfaces compared with 27% for large-head MoM bearing over the same time period. There was an increase in revisions for ARMD for small-head MoM over time, which was not seen with other bearing surface combinations.

There are several limitations for this study. We included all MoM bearing surfaces regardless of the stem or cup used with the procedures. Because time to first revision for all causes is the primary outcome measure, there are revision indications such as dislocation and fracture that may not be attributable to the bearing surface. The Metasul bearing couple comprised 78% of all THAs with a small-head MoM bearing surface and there may be some differences in the survival of other individual small-head MoM bearing designs; we did not have enough numbers of such designs to make a comparison. There are also potential differences in the performance of both low and high carbide femoral heads [], but, because the majority of the prostheses in this analysis used Metasul (high-carbide heads), a comparison between the relative performance of low- and high-carbide heads was also not possible.

Additionally, the Registry does not collect imaging or laboratory data and so we cannot comment on implant position or bony and soft tissue reactions in patients with MoM bearings. There have been studies that have demonstrated that the position of the acetabular cup is important with relationship to ARMD with large-head MoM [ ], but the authors are unaware of such issues with small-head MoM designs. It seems unlikely, however, that acetabular cup position would differ by the size of the metal insert. The Registry also does not collect information on patient activity levels, a factor that has been implicated in increased polyethylene wear []. Some surgeons may have preferentially chosen MoM bearings for more active patients, although the authors do not believe that this would have led to either marked differences in patient selection for small- versus large-head MoM designs or the marked difference seen in the outcomes of small- and large-head MoM THAs. The Registry is also aware that surgeons may have more likely recorded revisions for ARMD for small-head MoM in later years as awareness of this pathology became widely recognized. However, the Registry has not identified this problem with non-MoM bearing surfaces over a similar time period.

The survivorship of small-head MoM THAs in our Registry was similar to other, smaller cohort and registry studies with mid- to long-term followup [ ]. There was a lower rate of revision of small-head MoM bearing THAs compared with large-head MoM THAs in our Registry and there may be several reasons for this. In addition to wear of the bearing surface, there is evidence of metal pathology related to the taper junction as a result of crevice corrosion [ ]. When compared with hip resurfacing, there is a higher rate of revision and higher cobalt and chromium levels for large-head primary hip arthroplasty. This supports the concept of an additional metal ion burden with a femoral head/trunnion couple [ ]. There are also some increasing reports of revision for ARMD in THA with non-MoM bearings with larger diameter femoral heads, which is one of the risk factors implicated [ ]. Although the mechanism of failure may be ultimately the same with small- and large-head MoM THAs, the volume of metal particles released by both the bearing surface and taper junction may be less in small-head MoM bearing surfaces, which might account for the marked differences in rates of revision seen here.

This study has demonstrated that there is an increasing risk of revision over time for the diagnosis of ARMD in small-head MoM articulations. The registry uses the term “metal-related pathology” (MRP) and here the synonymous term “adverse reaction to metal debris” (ARMD) to encompass any diagnosis that the surgeon feels to be related to metal-bearing surface debris, which could include hypersensitivity reactions, crevice corrosion, pseudotumors, adverse local tissue reactions, and osteolysis. The Registry, however, has no information on metal ion levels or histopathology associated with THA revisions, although there was no correlation with metal ion levels and revision for Metasul THA in a study by Malek et al. []. Although revisions for loosening, dislocation, infection, and periprosthetic fracture are all more commonly performed for small-head MoM designs, the Registry has recorded 19 (0.4%) of all revisions that were directly attributable to ARMD. Because the Registry is unable to distinguish between the two, this could have been from either bearing surface or trunnion-related issues or a combination of the two. There were also 16 revisions for osteolysis, which also could have been a result of ARMD, but the surgeon did not directly attribute this to the bearing surface. There have been several studies on longer term followup of Metasul THA with variable reports of revision for ARMD from zero to 2% [ ] (Table 4). There are, however, revisions recorded for persistent, unexplained pain, osteolysis, and infection with no bacterial growth, and these revisions may potentially be linked to ARMD. Willert et al. first described the pathology associated with a series of 19 consecutive revisions on patients who had undergone primary small-head MoM THA []. The patients who had revisions to either a metal-on-polyethylene or ceramic-on-ceramic articulation improved, whereas the five patients who were revised to another MoM bearing did not. In our study, 19 patients with small-head MoM bearings were revised for ARMD and all but one was revised to non-MoM bearings. There have been no further revisions of any of these procedures. We would recommend that patients with a painful THA with a small-head MoM bearing surface be investigated along similar guidelines as large-head MoM [] and, if revision surgery is performed, a bearing surface other than MoM be used.

In contrast to small-head MoM, ARMD is the most common reason for revision recorded by the Registry for THA with large-head MoM.

There has been a dramatic reduction in the use of small MoM bearings in the AOANJRR over the past 10 years, which may be related to the relatively poor survival of large-head MoM THA bearings and associated device recalls. In Australia, there were only 18 small-head MoM bearing THAs implanted in 2015 compared with > 500 per year during peak use in the Registry > 10 years ago. Although small-head MoM THAs have performed well at long-term followup, reports of osteolysis, unexplained pain, and ARMD give some cause for concern with regard to their ongoing use. The success of other alternative bearing surfaces, most notably ceramic-on-ceramic and metal- or ceramic-on-highly crosslinked polyethylene, has likely also accelerated the trend away from all MoM designs. We believe that in light of present evidence, there is no indication for the use of large-head MoM with THA and limited, if any, indication for the use of any MoM bearing surface.

We thank the AOANJRR and the hospitals, orthopaedic surgeons, and patients whose data made this work possible. The Australian Government funds the AOANJRR through the Department of Health and Ageing.