Comparison of ceramic-on-ceramic bearing vs ceramic-on-highly cross-linked polyethylene-bearing surfaces in total hip arthroplasty for avascular necrosis of femoral head: a prospective cohort study with a mid-term follow-up

Total hip arthroplasty (THA) has been considered a successful solution for disabling hip conditions after end-stage avascular necrosis of the femoral head (AVNF). While the number of overall total joint arthroplasties continues to increase, the greatest rise is projected to be in the young patient population, with 52% of all joint replacements predicted to be in patients younger than 65 years old [1]. Major limitation affecting THA survivorship in younger patient has been polyethylene (PE) wear and particle-induced osteolysis resulting in aseptic loosening and late failure of the implant [2]. Traditional metal-on-polyethylene (MoP) has the risk of polyethylene wear and mechanically activated corrosion which many increase the risk of revision [3]. Ceramic-on-ceramic (CoC) has the advantage of low levels of wear and is a popular choice of bearing surface for younger patients [4]. Nevertheless, CoC has its own recognized complication risks (insertional and delayed fracture [5, 6] and squeaking [7, 8]). A ceramic-on-highly cross-linked polyethylene (CoXPE) articulation may decrease long-term wear [9] and potentially have a longer survivorship with less ceramic fracture and squeaking than CoC-bearing surface and is increased use in the younger population (less than 65 years old) over the last 5 years as demonstrated in the National Joint Registry [10]. The previous study has reported the comparative result between CoC and CoP [11], but the study focused on conventional polyethylene. There is limited data on the comparative result between CoC and CoXPE. Similarly, health-related quality of life (HRQL) after THA has been reported only in a handful of studies [12, 13]. In this retrospective study, we aim to evaluate the clinical outcomes, HRQL, complications, and wear rate of the bearing surface between CoC and CoXPE for THA patients with AVNF with a minimum follow-up of 5 years.

Although joint registries demonstrate excellent survival using polyethylene [10]. The incidence of osteolysis in conventional polyethylene was almost 18% [20]. Concerns about wear and osteolysis with polyethylene have led to an increased focus on the use of hard-on-hard articulations. There is a trend for the hard-on-hard bearing surface transition from metal-on-metal to CoC articulation [10]. Ceramics offer the best wear resistance, wettability, scratch resistance, and scratch profile [21]. CoC had a lower incidence of osteolysis than metal-on-metal in primary THAs [22]. CoC bearing was considered to be the ideal bearing surface in the younger, active patient with increased use in younger patients [23]. However, ceramic articulations have the drawbacks of fracture, squeaking, and the risk of acetabular torque [11]. Moreover, head and liner options are limited, such as the lack of a lip liner to improve the stability of the hip. Highly crosslinked-polyethylene also demonstrated good wear resistance [9, 24]. The new highly cross-linked polyethylene inserts may reduce the risks associated with ceramic implants while retaining their longevity with less wear than conventional poly [25]. According to literature, both CoC and ceramic-on-plastic bearing showed better implant survival compared with the metal-on-plastic bearing [13].

In this study, there were no failures of bearing surfaces in either the CoC or CoXPE group in AVN patients who underwent primary THA at an average 7 years follow-up. CoC THAs demonstrated good wear resistance within a 5–9-year follow-up. Participants reported significant improvements in the postoperative Harris hip score. This study also demonstrates that with an average 7 years follow-up, there is no difference in patient-reported outcome measures with SF-36 except for the sub-item of general health, suggesting that the CoC performs just as well as the CoXPE. The difference in general health may be related to the elder age of the CoXPE group.

There are many comparative studies that have examined the outcome comparing CoC-bearing surface to ceramic-on-poly surfaces [11, 12, 26, 27]. Atrey et al. reported comparable survivorship and function after 15 years follow-up between CoC and ceramic-on-conventional polyethylene [11]. The hard-on-soft articulation had higher mean annual wear than hard-on-hard articulation. The author concluded that polyethylene wear and osteolysis may represent issues in the future [11]. The limitation was the cross-linked polyethylene was not adopted and the BIOLOX delta ceramic was not widely available. Beaupre et al. reported CoC and ceramic-on-crossfire-polyethylene had a similar health-related quality of life after postoperative 10 years follow-up. There were no failures of loss of fixation related to the bearing surface in either group. The study was limited to radiographic review with plain radiographs and no data of annual wear rate [12]. In those that had a radiographic follow-up at 10 years (n = 57; 66%), no osteolysis was noted in either group [12]. The incidence of osteolysis in both bearings was lower than the 18% rate reported for conventional polyethylene in literature [20]. The annual wear rate in this study was similar to the wear rate in literature [11, 28, 29].

The annual wear rate in the current study of the CoC group was 0.0096 mm/year, supporting the excellent behavior and wear resistance of the ceramic bearing [25]. The annual wear rate of CoXPE in the current study was lower than the result of ceramic-on-conventional poly in literature [11]. The result also supported the good wear resistance of highly crosslinked-polyethylene [24]. Although the mean annual wear in the CoXPE group was significantly higher than the CoC group (0.047 vs 0.011 mm), the osteolysis is uncommon with a wear rate of < 0.1 mm/year according to the literature [30]. As a result, the current study shows that there was no difference of HHS, SF-36 scores and osteolysis between the two groups, suggesting that the CoC performed as well as the CoXPE.

Our study has strengths including the aim to study on the same disease and use more accurate CT data to measure the wear rate and radiological outcome. As we have known, the clinical outcome and implant survival tend to differ based on primary diagnosis [13]. Swarup et al. reported that patient diagnosis is predictive of implant survival with juvenile inflammatory arthritis patients having the lowest implant survival [13]. This study focused on THA for osteonecrosis which is one of the common etiologies for THA in our country. The study can more accurately discover the difference of outcome owing to the different bearing surface selection. CT offers a three-dimensional assessment of the liner wear in contrast to the conventional radiographic examination which may underestimate polyethylene wear. It was reported that CT can detect osteolysis and wear earlier than with conventional radiographic examination before loosening occurs [31].

The reported rate of squeaking and other noises for CoC bearing ranged from < 1 to 21%, and the etiology is multifactorial [32]. In our study, audible noise occurred in 8.6% in the CoC group and 2.6% in the CoXPE group. Although the difference was not statistically significant, there was a higher rate of audible noise for the CoC-bearing surface than the CoXPE-bearing surface. We found no reported postoperative squeaking in our study and no difference in postoperative Harris hip scores between the noise and silent hip groups. Furthermore, no revision was required for the noise.

However, there are some limitations to our study. Firstly, this study is limited by its nature of observation study with a relatively small number of cases. However, we conducted a power analysis to determine the sample size to detect the difference of wear between the 2 groups and found that 52 cases were needed in each group. Additionally, the analysis was based on consecutive cases, with no randomization. In this study, patients who received a CoC THA were younger than those who received a CoXPE THA. This might have been caused by selection bias and may have influenced the clinical outcome results. Secondly, our study is a midterm follow-up result. Longer-term studies are needed to determine the wear rate and incidence of osteolysis for implant longevity between the two groups. Thirdly, we did not obtain a full clinical and radiographic follow-up for all of the respondents. Fourthly, we used different sized femoral heads; this may have influenced the complication of postoperative dislocation and clinical results. Finally, the ceramic insert would cost more than poly. In the current study, no cost-effective analysis has been studied, which is also important for healthcare decision-making [12].

In summary, our results showed other than greater wear in the cross-link polyethylene group, there was no significant difference in outcome between the CoC and CoXPE THA for osteonecrosis after midterm follow-up in terms of functional outcomes, bearing surface-related complication, radiographic results, and HRQL outcome. Both CoC and CoXPE bearings can behave excellently. As for the longevity of the two bearing surfacing, long-term follow-up study will be needed.