Ceramic failure is still a major concern of ceramic bearing surfaces in total hip arthroplasty (THA),1, 2 despite the advancement made in the quality of third generation alumina ceramic bearings since 1994.3, 4 Recently, impingement has been recognized as a common occurrence in THA,5–8 and it can not only limit range of motion, but also increase stress on the liner rim, which can result in damage or rim wear of the polyethylene liner.7–9 Impingement may be more problematic for ceramic-on-ceramic THA prostheses, because ceramic liners are brittle, and thus, have no potential to deform before breakage contrary to polyethylene liners. Impingement between the femoral metallic neck and a ceramic acetabular liner causes a V-shaped macroscopic notch in the contact region of the neck.10–12 However, no report has presented in vivo microscopic evidence of damage to the ceramic liner, although impingement has been suspected to be the cause of ceramic liner failure.10, 13
During revision surgery, we encountered evidence of impingement between the metallic neck and the ceramic liner. We hypothesized that in vivo microscopic evidence of liner damage induced by impingement, which is known to lead to ceramic fracture, would exist on the contact surface of ceramic liners.
MATERIALS AND METHODS
Between November 1999 and July 2008, we performed 18 reoperations on 18 patients who had undergone third-generation ceramic-on-ceramic THA. Excluding two patients who were reoperated upon within 1 year, 16 patients were included in the final study. Ten were originally treated by us; the remainder had undergone their primary THA at other hospitals. This study was approved by our Institutional Review Board, and patients were informed that data concerning their cases might be submitted for publication.
All bearings were hot isostatic pressed, laser-marked, and proof-tested third-generation BIOLOX Forte alumina (CeramTec, Plochingen, Germany) components, and 15 bearing couples (head and liner) were retrieved during reoperation. All ceramic heads had a diameter of 28 mm. The acetabular components were PLASMACUP SC® (Aseculap, Tuttlingen, Germany; n = 10), Trilogy AB® (Zimmer, Warsaw, IN, n = 1), Duraloc Option® (Depuy Orthopaedics, Warsaw, IN, n = 1), SPH CONTACT® (Lima, Udine, Italy; n = 2), and Trident® (Stryker, Mahwah, NJ, n = 2).
Fourteen acetabular liners, including two sandwich-type liners with a layer of polyethylene between the ceramic liner and metal shell, were flush-mounted inside the metal back. The remaining two liners were inserted into protecting type acetabular cups with metal fencing. The median outside diameter of the acetabular component in our 10 cases was 50 mm (range: 46–56 mm). The femoral components were Bicontact® (Aseculap; n = 10), VerSys CT® (Zimmer, n = 1), AML® (Depuy Orthopaedics, n = 1), C2 Stem® (Lima; n = 1), H-Moos® (Lima; n = 1), and Omnifit® (Stryker, n = 2).
All retrieved components and retained implants were examined visually and photographed. Evidence of metal neck to ceramic impingement was defined as macroscopic notching in the neck of the femoral component with corresponding black staining of the ceramic liner rim.
Demographic and clinical data, including gender, age, body-mass index (BMI), and reason for reoperation, were obtained by reviewing medical records. There were nine men and seven women. Mean age at reoperation was 50 years (range: 22–72 years), and the mean interval between THA and reoperation was 51.8 months (range: 12–99 months). The principal reasons for reoperation included ceramic head fracture in four hips, ceramic liner fracture in three, acetabular component loosening in two, implant malposition in one, infection in three, dislocation in one, squeaking in one, and osteolysis in one (Table 1).
Table 1. Patients Demographics
Time In Vivo (Months)
Reason for Reoperation
Inclination of Cup (°)
Anteversion of Cup (°)
Cases 1–4 showed the evidence of the metal–ceramic impingement during reoperation.
Immediate postoperative cup position is not available.
Radiographs before reoperation were used to determined acetabular component positions (inclination and anteversion). Inclination angle was measured between the horizontal teardrop line and the line through the long axis of the ellipse projected by the margin of the metallic backing.14 Anteversion was calculated as described by Widmer et al.15 We also reviewed serial radiographs for evidence of notching of the femoral neck.
The retrieved ceramic liners with evidence of impingement were coated with Au-Pd after removing the metallic stain with acid treatment,16 and examined by scanning electron microscopy (SEM; SM-500, Topcon, Tokyo, Japan). SEM analysis was performed with an accelerating voltage of 20 kV at a working distance of 8–15 mm, and SEM findings of black stained areas were compared with those of nonstained areas.
Statistical analysis was performed using the Mann–Whitney U-test and the Fisher's exact test. Differences were considered significant at the p < 0.05 level.
Four of the 16 cases showed evidence of impingement. In these cases, causes of reoperation were ceramic head fracture, squeaking, loosening of acetabular component, and periprosthetic infection. All four patients were able to sit on the floor in the tailor's position before reoperation, and reoperations were performed at 35, 46, and 70 and 99 months, respectively. No evidence of impingement was observed in any metal-backed cups, including Trident® acetabular cups, which had a metallic elevated rim.
The four patients with impingement did not differ from the other 12 in age (p = 0.808), gender (p = 1.000), BMI (p = 0.275), or the ability to sit in the tailor's position (p = 0.234). Similarly, inclination and anteversion of the acetabular component did not differ between the groups (p = 1.000, 0.953, respectively) (Table 1). In one hip with impingement, notching of the neck was visualized on a translateral radiograph (Fig. 1).
In the hips with impingement that was reoperated upon for uncontrolled periprosthetic infection, the ceramic liner was not available for evaluation, because the ceramic liner and the fractured ceramic head had been previously replaced with a polyethylene liner and a metal head at another hospital. However, macroscopic notching was found on the postero-superior aspect of the neck of the retrieved femoral component (Fig. 2A). In remaining three hips, macroscopic notching and a black-stained liner rim were present (Figs. 2 and 3). A V-shaped indented wear scar was located at the postero-superior aspect of the neck in three cases and at the anterosuperior aspect in the case that was reoperated upon for squeaking. Ceramic liners were stained black at the postero-superior edge in cases with postero-superior neck scar and at the anterosuperior edge in the case with anterosuperior neck scar (Figs. 2 and 3).
Almost identical SEM surface analysis findings were obtained for the three hips with a liner available for evaluation. Unlike nonstained areas, which were relatively well preserved with smooth, polished surfaces, the surfaces of the black-stained areas were markedly roughened by disruptive wear. Furthermore, surface integrity was lost, and the underlying granular alumina structure was exposed. In one case (Case 2), multiple microcracks were observed in the roughened surface area (Fig. 4), and when this liner was cross-sectioned at the area of a microcrack, the microcrack was found to propagate deep into the liner body (Fig. 5).
Impingement between the metallic neck and the ceramic liner rim is a common finding in THA. Recent retrieval studies demonstrated that a high percentage (39–60%) of conventional polyethylene bearing THAs show impingement.7, 8, 17 For ceramic-on-ceramic bearing THA, Sugano et al.10 observed impingement in 7 (6%) of 111 cases. Two of these cases were confirmed by revision surgery, and five were detected on follow-up radiographs. In our study, 4 (25%) of 16 revision cases showed impingement.
Several factors increase the possibility of impingement in ceramic-on-ceramic THA. Surgical factors, including cup malposition, have been suggested to be critical.10, 17–21 Previous experimental or finite element analysis studies presented that implant malposition such as low inclination of acetabular cup is more prone to impingement.18, 20–22 However, in our study, no significant difference was observed between hips with and without impingement in terms of acetabular component position. But our sample size was small, and we agree that surgeons should take care to avoid malpositioning the acetabular component.
Of the patients-related factors, repeated sitting in the tailor's position, squatting, and kneeling, which are more common in East Asian populations than among Caucasians, were probably responsible for the impingement observed in our patients,13, 23 and for microcrack propagation into ceramic liners.17 Additionally, implant design, including head size and neck geometry, be also influences impingement.21, 22, 24
Unlike impingement between the metal neck and a polyethylene liner, impingement in hard-on-hard bearing couples causes notching of the metallic neck,25, 26 which could potentially cause metal failure, although no such case has been reported. In our four impingement cases, neck failure did not occur over implantation periods ranging from 48 to 99 months.
Impingement between a metallic neck and polyethylene liner increases stress on the liner rim, which leads to polyethylene wear and early osteolysis.7–9, 27 Furthermore, impingement can cause liner cracking and oxidative wear, which can result in catastrophic failure.6 Mochida et al.28 observed increased ceramic particle production in a ceramic-on-ceramic THA case with evidence of impingement. In our study, surface damage was confirmed in the impingement area on the liner rim, and this impingement had produced microcracks that propagated deep into the liner. This finding suggests that impingement may be a potential for ceramic fractures. Conversely, this conclusion is different from that reached by Maher et al.29 who concluded that ceramic liner fracture was unlikely to occur as a result of impingement after conducting an in vitro simulation of the effect of impingement between a ceramic liner and the neck of the femoral component.
Historically, Nevelos et al.30–32 identified several risk factors associated with accelerated wear in ceramic bearings. Ceramic particles released during microcracking could accelerate ceramic wear, although we did not attempt to quantify the ceramic particles produced or the wear that occurred in our cases.
Our in vivo observations appear to be the first to support the microcrack propagating theory, which was developed based on in vitro findings.33, 34 Our observations suggest that impingement between a ceramic liner and a metallic neck causes not only metal transfer and liner surface damage, but also microcrack formation in ceramic liner.
No benefits in any form have been received or will be received from any commercial party related directly or indirectly to the subject of this article.