Absence of instabilities and intra-prosthetic dislocations at 7 to 11 years following THA using a fourth-generation cementless dual mobility acetabular cup

Purpose Dual-mobility (DM) cups are increasingly used in total hip arthroplasty (THA) but there lacks literature on their long-term results. We aimed to investigate outcomes of a fourth-generation cementless DM acetabular cup at 7–11 years. Methods We retrospectively evaluated 240 consecutive hips that received cementless THA using the same dual mobility cup (Novae Sunfit TH) and femoral stem (Corail). Patients were recalled at ≥7 years to collect Oxford hip scores (OHS), Harris hip scores (HHS), and inspect for radiolucent lines and granulomas. Multi-variable analyses were performed to determine whether HHS or OHS were associated with pre- or intra-operative variables. Results At 8.4 ± 0.8 years (range, 7–11), 6 hips were revised (2.5%), 54 deceased (22.5%), and 14 could not be reached (5.8%). Four revisions (2 cup+stem, 2 liners only) were due to sepsis (1.7%), one (cup and stem) for trauma (0.4%), and one (stem) due to aseptic loosening (0.4%). For the remaining 166 hips, HHS was 83.6 ± 13.2 and OHS was 20.3 ± 6.7. Multi-variable analysis confirmed that HHS (β = − 0.38; p = 0.039) and OHS (β = 0.36; p < 0.001) worsened with age, and that OHS was worse for Charnley C patients (β = 3.17; p = 0.009). Neither granulomas nor radiolucenies were observed around any cups, but radiolucenies were seen around 25 stems (20.3%). Conclusions This fourth-generation DM cup demonstrated satisfactory outcomes at 7–11 years, with no instabilities or cup revisions due to aseptic loosening. Better OHS was observed for younger patients and those presenting higher Charnley grade. Level of evidence Level IV, retrospective case study.

The primary goal of this study was to report revision rates, clinical scores and radiologic findings of a fourthgeneration DM acetabular cup, in a sizeable multi-centre series with up to ten years of follow-up. The secondary goal was to identify demographic and operative factors that could compromise clinical scores and hence optimise future patient selection and surgical choices.
All patients were recalled for clinical and radiographic evaluation, and their case notes were used to document implant materials, models and diameters. From the initial cohort of 240 THAs, 3 had stem and cup revisions, 2 had liner and/or head replacement, and 1 had an isolated stem revision. In addition, 50 patients (54 hips) deceased, none of which had revision surgery, and 14 patients (14 hips) could not be contacted. The remaining 166 hips were assessed clinically at 8.4 ± 0.8 years (range, 7-11), of which 123 hips were also assessed radiographically. The clinical scores collected included the Oxford hip score (OHS, best = 12; worst = 60) [13], Harris hip score (HHS, best = 100; worst = 0) (best) [5], pain on visual analogic scale (pVAS, best = 0; worst = 10). The radiographic assessment included frontal weight-bearing pelvic radiographs that were inspected for radiolucent lines (> 2 mm wide) and granulomas in the 7 femoral Gruen zones [19] and 3 acetabular DeLee-Charnley zones [6,14], and for Brooker heterotopic ossifications [4]. All patients provided written informed consent for their participation in the study.

Statistical analysis
Normality of distributions was verified using the Shapiro-Wilk test. In case of non-parametric quantitative data, significance of differences between groups was assessed by the Mann-Whitney U test (Wilcoxon ranksum test). Uni-and multi-variable linear regression analyses were performed after identification of relevant   2). Four of the revisions were due to deep infection (1.7%), 2 of which required cup and stem exchange (0.8%) while 2 required only PE liner exchange (0.8%). One of the revisions was due to periprosthetic femoral fracture secondary to trauma which required cup and stem exchange (0.4%), and only one revision was due to aseptic loosening and required isolated stem revision (0.4%). There were no dislocations recorded.

Discussion
This study demonstrated satisfactory clinical and radiographic outcomes of cementless THA using a fourthgeneration DM acetabular cup, with a cumulative revision rate of 2.5% at a mean follow-up of 8.4 years. It is important to note, however, that only one revision (0.4%) was due to aseptic loosening, and required femoral component exchange, but that there were no cup revisions, due to either instability or aseptic loosening. Deep infection remained the principal cause of revision (1.7%) and only one hip was revised for periprosthetic fracture (0.4%) secondary to trauma. Our cumulative revision rate is within the range reported for fourthgeneration DM acetabular cups [7, 8, 16-18, 21, 22, 31, 32, 51, 52]. While numerous smaller series (40-104 hips) [31,32,51,52] had no revisions of any kind at 5 to 10 years of follow-up, larger cohorts (167-3474 hips) [7, 8, 16-18, 21, 22] had overall revision rates between 0.5% and 3.6%, at 5 to 13 years of follow-up. Dual mobility acetabular cups proved increasingly popular in recent years as they allow improved range of    motion and prevent instabilities [25,47]. Fourthgeneration DM acetabular cups, with optimized bearing surfaces, liner materials and coatings, have reduced the risks of intra-prosthetic dislocations and the need for subsequent revisions [37,40,42,46]. However, there still remains a lack of published studies concerning their mid-to long-term outcomes. Recent studies indicated that deep infection is the most common cause for revision of DM acetabular cups [29], which may be explained by the infirmity and comorbidities of the older population in which they are implanted [3]. In this series, the cumulative rate of revision for infection was 1.7%, at 8.4 years which is slightly higher than the rate of 1.0% at 5 years, reported for all hip arthroplasty infections in the Danish hip registery [20].
This study revealed no intra-prosthetic instabilities at either the liner-cup junction or at the liner-head junction, which proves that 28 mm heads are compatible with this stem and cup combination [11]. According to the current literature, it is clear that DM is the best option to prevent instabilities after THA, particularly in women, elderly and obese patients, as well as those with elevated ASA scores or neuromuscular deficits [3,35,47]. Moreover, it is still debateable whether larger femoral head sizes should be used, as they are associated with lower risks of dislocations but increased PE wear [26,30,48]. Our study revealed no dislocations using 28 mm heads. Using larger femoral head sizes could exacerbate PE wear, debris and osteolysis [15], whereas using 22 mm heads, would increase the risk of intra-prosthetic instabilities by reducing the neck to head ratio, which causes earlier impingement between the stem neck and the retentive cup rim [12,44]. Psoas impingement was found in 3 hips (1.3%), all of which had intraoperative femoral cracks fixed using cerclage wires, which likely exacerbated tendon contact against implanted components. In a landmark anatomic study, Vandenbussche et al. [50] described the acetabular zone of psoas impingement, and warned that prosthetic overhang is more frequent with DM acetabular cups, because they are designed with a more protrusive rim.
For the present series, the median HHS and OHS at 7 to 11 years were 86 and 19 points respectively, and patient-repoted pVAS was 0. These outcomes compare favourably to scores repoted in recent studies on fourthgeneration DM acetabular cups [7,8,18,21,32,51,52]. Our multi-variable analysis revealed significant influence of preoperative Charnley disability index and age on The main limitations of the present study are its retrospective design, and hence considerable proportion of patients lost to follow-up (8%) or missing radiographic images (26%). The advanced age of many of the patients may have contributed to the high numbers that were lost to follow-up, but they shared the same standard demographics and surgical parameters as the rest of the series. It is noteworthy that the clinical follow-up was longer than the radiographic follow-up. Despite the size of the initial cohort and follow-up at 7 to 11 years, the present data may be insufficient to confirm elimination of rare complications such as instabilities and intra-prosthetic dislocations, which require larger cohorts with prospective follow-up. National registries provide larger datasets for more robust conclusions on complications and survival but the heterogeneity of implant models and surgical techniques, as well as the paucity of preoperative and surgical data do not enable identification of risk factors. Furthermore, this study is not comparative and cannot therefore decide on the relative functional or cost benefits as compared with unipolar cups. The principal strength of the study is the sizeable cohort, which includes patients susceptible to instabilities, and relatively extended followup for a fourth-generation generation of DM acetabular cups. Although two stem head materials were used and two different surgical approaches applied, the same DM acetabular cup design was used throughout the study which allows the authors to draw clear conclusions.

Conclusion
This study presented satisfactory radiographic and clinical mid-term outcomes of cementless THA using a fourthgeneration DM acetabular cup, with no instabilities or revisions due to aseptic loosening. Better HHS and OHS were observed for younger patients and those with preoperative Charnley grade A. Further studies should consider tribologic aspects of DM acetabular cups to confirm the best bearing couples that would minimize wear and metal ion release in the long-term.