Outcomes after the conversion of metal-on-metal hip resurfacing arthroplasty to ceramic-on-ceramic total hip arthroplasty: A retrospective observational study with a mean 10-year follow-up CURRENT STATUS: POSTED

Background The purpose of this retrospective study was to assess clinical outcomes following failed metal-on-metal hip resurfacing arthroplasty (MoM-HRA) revised using ceramic-on-ceramic total hip arthroplasty (CoC-THA) via the direct anterior approach (DAA). Methods Data involving 112 patients (112 hips) with failed primary MoM-HRA that was revised using CoC-THA via the DAA during 2006 - 2018 were retrospectively analysed. The mean age was 54.6 years (45–63 years). Frequent surgical indications for conversion were aseptic loosening and femoral neck fracture. The primary endpoint was the Postel-Merle d’Aubigne functional score (PMA). Secondary endpoints were the major orthopaedic complication rate and serum metal ion levels (chromium and cobalt). The mean follow-up time was 10 years (range, 4 - 13 years). The mean PMA improved from 9 (4-14) to 16 (13-18) (p = 0.001). Six patients had undergone a re-revision intervention. Fourteen incidents of aseptic loosening and four periprosthetic fractures were observed. A consistent decline in mean serum metal ion levels was detected [chromium decreased from 36.6 μg/L (12.5-76.5 μg/L) prior to conversion to 2.6 μg/L (0.1-13.5 μg/L) at final follow-up (p = 0.001), cobalt decreased from 37.5 μg/L (6.7-93.2 μg/L) to 2.2 μg/L (1.2-18.4 μg/L) (p =0.003)]. Failed primary MoM-HRA converted to CoC-THA using the DAA yielded acceptable clinical outcomes.


Background
The purpose of this retrospective study was to assess clinical outcomes following failed metal-onmetal hip resurfacing arthroplasty (MoM-HRA) revised using ceramic-on-ceramic total hip arthroplasty (CoC-THA) via the direct anterior approach (DAA).

Methods
Data involving 112 patients (112 hips) with failed primary MoM-HRA that was revised using CoC-THA via the DAA during 2006 -2018 were retrospectively analysed. The mean age was 54.6 years (45-63 years). Frequent surgical indications for conversion were aseptic loosening and femoral neck fracture.
The primary endpoint was the Postel-Merle d'Aubigne functional score (PMA). Secondary endpoints were the major orthopaedic complication rate and serum metal ion levels (chromium and cobalt).

Conclusion
Failed primary MoM-HRA converted to CoC-THA using the DAA yielded acceptable clinical outcomes.

Background
Failure of standard total hip arthroplasty (THA) triggered by an adverse reaction to metal debris (ARMD)-induced osteolysis can be prevented by metal-on-metal hip resurfacing arthroplasty (MoM-HRA) [1][2][3]. However, the use of MoM-HRA has frequently been conditioned by bone conditions [4,5].
Recurrent short-term failures are well recognized for MoM-HRA, particularly MoM bearings in individuals with small femoral components, and they are commonly addressed with a bipolar revision with a conversion to ceramic-on-ceramic THA (CoC-THA) [6,7]. Failure secondary to MoM wear has become a concern [2,8]. ARMD-induced osteolysis is overwhelmingly implicated in the pathophysiology of MoM-HRA failure, contributing to as high as a 10 to 27% revision rate and quick time-to-failure [9][10][11].
Limited studies have assessed complications from the conversion of MoM-HRA to CoC-THA. Interest in CoC-THA has increased over the last decade [6,7]. Numerous reports have demonstrated that the outcomes of conversion CoC-THA are mostly influenced by the reason for conversion [7,2,12].
Additionally, there remain concerns regarding whether MoM-HRA conversion to CoC-THA is associated with poor clinical outcomes [6,9].
To date, consensus remains lacking regarding long-term outcomes after MoM-HRA conversion to CoC-THA. To gain a comprehensive understanding of the long-term results of these conversions, we performed this retrospective study to evaluate the long-term clinical outcomes of failed primary MoM-HRA conversion to CoC-THA using the direct anterior approach (DAA).

Study population
Between February 2006 and December 2018, an original study cohort of 186 patients (212 hips) who underwent conversion of a failed primary MoM-HRA to a CoC-THA was obtained from our joint surgery practice database (Sun Yat-sun University, Guangzhou, China), which was established in 2002 and contains data derived from computerized records. A detailed description of the database is available in our previous report [13]. The inclusion criteria were as follows: patients who underwent conversion of an original MoM-HRA (uncemented cup: Dynacup, Tornier, Montbonnot, France; a PE insert: Selene, Tornier) to a CoC-THA (ceramic-on-ceramic, Stryker, Mahwah, New Jersey); and patients who had no exposure to other known sources of chromium or cobalt or impaired renal function. The main exclusion criteria were deficient clinical outcome data, dyskinesia prior to conversion, joint-related diseases, lack of ability to follow instructions, active systemic or local infection, cancer, an injuryseverity score (ISS) of more than 10, serious organ complications (i.e., brain, pulmonary, cardiovascular, renal), cognitive impairment or psychosis, drug abuse, body mass index (BMI) > 40 kg/m 2 , and an American Society of Anesthesiologists (ASA) score of IV or V.
The revision surgery was performed by the 3 high-volume surgeons (WY, XZ, MZ) using the DAA in the supine position on a traction table, as previously described [6]. The indications for conversion involved aseptic loosening, femoral neck fracture, instability, and intolerable hip pain. A consistent protocol of antibiotic application was performed perioperatively for each patient. An anticoagulant regimen was applied to all patients without contraindications for one month after conversion. The same rehabilitation program was used for all patients. Full weight-bearing was allowed from the day after conversion.
The primary outcome was the Postel-Merle d'Aubigne functional score (PMA). Secondary outcomes were the rates of major orthopaedic complications (re-revision, aseptic loosening, periprosthetic fracture, dislocation, deep infection, intolerable hip pain) and serum metal ion levels (chromium and cobalt). Data were collected at the baseline evaluation and the final clinical review by WY and XZ. All patients who underwent CoC-THA were evaluated individually using the PMA at each follow-up.
Measurements of the PMA occurred 1 year, 2 years, and every 2 years after conversion. Image data obtained at these same time points consisted of an anteroposterior (AP) pelvis view and AP and lateral views of the hip and were assessed by experienced radiologists. The abduction angle of the cup, measured as reported previously, was assessed on the pelvic side. Analyses of blood trace elements were performed at the Central Laboratory of Physics and Chemistry, Sun Yat-sen University (Guangzhou, China). Measurements of serum metal ion levels occurred 1 year, 2 years, and every 1 year after conversion. The procedures for the measurements are based on previous descriptions [6,[14][15][16][17] and our laboratory trace element detection guide.
To ensure the validity of these data, all study data were audited by the two co-authors (WY and GH).  Figure 1, and the patient characteristics are described in Table 1.

Statistical analysis
Follow-up was defined as the interval from the date of MoM-HRA revision to the date of death or the last follow-up, whichever came first. Aseptic loosening was defined as in previous reports [18,19]. Rerevision was defined as the removal or exchange of any component. Osteolysis was defined as expanding areas of bone resorption surrounding the prosthesis (cup and/or stem). Continuous variables are presented as the means with standard deviations (SD) and ranges and were analysed using Student's t tests. Categorical data are presented as frequencies and percentages and were analysed by Pearson's χ² test or Fisher's exact test, as appropriate. Statistical analyses were performed using SPSS, version 24.0 (IBM, Armonk, NY). A 2-sided p value < 0.05 was considered significant.
Prior to the 3 years after revision, no fracture around the prosthesis or dysesthesia on the affected limb was reported for any patient. Thirty-one major orthopaedic complications were detected in the cohort (Table 3). Of these complications, 6 (5.3%) required re-revision surgery, 14 (12.5%) suffered aseptic loosening, 4 (3.5%) suffered periprosthetic fracture, 3 (2.6%) suffered dislocation, 2 (1.7%) suffered deep infection, and 2 (1.7%) suffered intolerable hip pain. Nearly 82.1% of re-revisions were attributed to aseptic loosening. The mean concentration of serum chromium decreased from 36. with CoC-THA [9,22]. However, there remain concerns about the suitability of CoC-THA [23]. Several studies have verified that conversion CoC-THA for failed MoM-HRA is associated with acceptable functional outcomes [12,20]. However, an increasing but still exceedingly limited body of literature has assessed the role of conversion CoC-THA following a failed primary MoM-HRA and demonstrated that the functional outcomes for CoC-THA were susceptible to patient indications [8].
A low re-revision rate was documented after conversion CoC-THA following a failed MoM-HRA.
Matharu et al. [24] noticed that 6% of cases required re-revision in a series of 48 failed MoM-HRAs converted to CoC-THAs. This may partly be attributed to the DAA because it potentially improves dynamic hip stability and has the ability to decrease the risk of dislocation after conversion [6].
Additionally, a reduced rate of complications was observed in a study by Migaud et al. [25] that assessed CoC-THA via the DAA. Notably, no dislocation was observed in their cohort at a mid-term follow-up. Combined with the current findings, the results confirmed by these studies appear to verify that the DAA is the preferred surgical approach for converting a failed MoM-HRA. Metal ions can decrease the number of osteoblast cells by inhibiting their gene expression, theoretically increasing the risk of MoM-HRA failure and creating unnecessary obstacles for a future conversion [26,27]. The present results for failed MoM-HRA conversion to CoC-THA have also shown low rates of re-revision or aseptic loosening. In Europe [28], the five-year and ten-year revision rates for MoM-HRA and stemmed large head MoM implants were higher than for CoC-THA. Hip stability following cemented CoC-THA is superior to that reported for uncemented CoC-THA because of the enhanced stability of bone and cemented components. The mechanical wear of the acetabular component tends to be related to uncommonly high wear initiated by encumbered edge loading [4,29]. A prospective study [20]  The risk associated with conversion of MoM-HRA to CoC-THA is still an extensive concern [30,8].
Nevertheless, the available studies on outcomes of the conversion remain deficient [31]. Savarino et al. [26] reported the role of CoC-THA and revealed noteworthy dissimilarities regarding conclusions about it. Attention has been paid to whether CoC-THA has a low rate of either re-revision or aseptic loosening according to long-term outcomes [32]. Bouveau et al. [6] assessed 13 conversions of MoM-HRA to CoC-THA and showed a low re-revision rate (6%), primarily due to the aseptic loosening of acetabular components. Using an analogous premise, Stryker et al. [33] reported 65 conversions of MoM-HRA to CoC-THA and demonstrated a low re-revision rate (7%), mainly ascribed to aseptic loosening (76%) and deep infection (16%). Recently, Rahman et al. [8] reported on 25 conversions and showed that none of the patients suffered aseptic loosening. In brief, documented results of MoM-HRA revised using CoC-THA have exhibited an exceedingly low rate of re-revision, and this conversion tends to be a safe choice for decreasing the re-revision rate.
Earlier studies have shown that the serum metal ion levels, mainly for chromium ions, remain at moderate to high levels after conversion CoC-THA in the elderly population [34,35]. The return to exceptionally low or unnoticeable levels could take more than 3 years [36], which is in accordance with our follow-up outcomes that serum chromium ion levels remain moderately high in a majority of patients 3 years after conversion. Urinary excretion of chromium has a significant effect on decreasing serum chromium ion levels [35]. Regrettably, several reports have shown that the amount of chromium ions excreted in urine is extremely small and that the accumulation of chromium ions predominantly occurs in tissues and erythrocytes [26,16]. Consequently, significant reductions in serum chromium ion levels will be extremely challenging to achieve following an increase [16].
Amstutz et al. [36] assessed 18 patients who were treated with MoM-HRA in a prospective study and concluded that serum chromium ions escalated within the 1st year and then declined and stabilized.
Additionally, they pointed out that serum chromium ions in a well-functioning endoprosthesis appear to be low and fail to increase with time. Furthermore, a sharp increase in serum metal ions has been shown to be related to prosthesis failure [37]. Several studies [20,36,38] report the toxic effects of prolonged elevated systemic metal ion exposure; therefore, whole blood metal ion testing should have a place in follow-up. Persistent low-dose exposure to chromium ions in patients with conversion of a failed MoM-HRA to a CoC-THA may have systemic effects and to date, long-term epidemiological studies do not seem to have been performed [36].

Conclusions
The long-term results reported in this study support an increasing body of evidence that conversion CoC-THA for a failed primary MoM-HRA was associated with a noteworthy improvement in clinical outcome, along with PMAs, re-revision rates, and rates of aseptic loosening, as well as significantly reduced serum metal ion levels. However, based on the data presented here, to determine other clinical benefits to, for example, the orthopaedic complication rate and prosthesis survival curve, future prospective multi-centre studies are needed.