Introduction

According to reports from several national registries, revision hip dislocation necessitating re-revision is on the increase. The Kaiser-Permanente Register reports instability as the most common indication for re-revision (49.8%) [1]. Bozic reports in a review of the United States National Inpatients Sample Database that 22.5% of all revisions were performed because of instability, which in this database also was the most common cause of revision [2]. In Australia, the National Joint Replacement Register Annual Report 2015 describes dislocation and instability as the reason for first revision in 25.1% and 31.1% in re-revisions [3]. In the annual report of the Swedish Hip Arthroplasty Register (SHAR) of 2015, revisions due to dislocation tended to increase and amounted to 14.6% in first-time revisions and 25.6% in multiply revised hips [4].

Several strategies have been adopted to manage stability and prevent recurrent dislocation, with one being the use of a constrained liner. Studies [5–7] have shown unsatisfactory results after these procedures. Larger femoral-head size has been suggested as another possible solution, but there remains uncertainty about the influence of the big head on trunnion load, possibly leading to increased risk of corrosion and the effect of thinner polyethylene when this type of articulation is used [8–10]. The biomechanical advantage of an increase in head size to avoid dislocations has, however, been studied extensively [11–13].

The dual-mobility cup (DMC) was introduced in 1974 [14], with the original intention of combining the benefits of Charnley’s low-friction arthroplasty and reduced volumetric wear with the increased head size to improve stability. There are promising reports on the use of DMCs in both revision and primary surgery [15–19]. The few studies reporting on DMC outcomes following revision surgery [17, 18, 20–24] cover approximately 200 cases combined and report an early to intermediate re-revision rate due to dislocation of <5%. All previous reports analysing DMCs except one [25] are case series from single centres. Since the study by Hailer et al. [25], the use of DMCs in Sweden has increased continuously. To our knowledge, there are few studies focusing on revisions performed due to dislocation [26]. We therefore analysed outcomes of the most frequently used DMCs reported to the SHAR. Our primary aim was to compare implant survival of cemented DMCs with traditional cemented revision cups used in first-time revisions with the same indication and operated upon during the same time period.

Material and methods

In Sweden, all orthopaedic units performing revision THA report to the SHAR. The completeness of revision procedures has been reported as 90% [27]. Baseline data is completed on standard forms by a local secretary at each hospital. The operation notes are sent to the SHAR, where the register coordinators extract further data for the procedure. Data collection and validation is well accepted and has been described in SHAR annual reports.

The first DMC used in revision surgery was reported to the SHAR in 2005. From 2005 to 2015, 14,863 first-time cup revisions with or without concomitant stem revision were reported, of which 984 were performed due to dislocation with a cemented cup. A DMC had been used in about half of the cases (n = 523). In most cases (n = 436), a cemented Avantage™ cup (Biomet, Warsaw, IN, USA) had been inserted. The Avantage™ group (DMC) was then compared with a group operated using a standard cemented cup in which femoral head size were either 28, 32 or 36 mm (non-DMC, n = 355). In the non-DMC group, 14 different designs had been used, with the six most frequently used designs inserted in 85% of revisions. The flowchart for data extraction is illustrated in Fig. 1.

Fig. 1
figure 1

Data extraction from the Swedish Hip Arthroplasty Register (SHAR)

Full size image

The primary outcome was any re-revision due to all causes. We also analysed cup re-revision due to all causes and any kind of re-revision performed due to dislocation. Nonparametric testing was used to compare groups. Demographic data are presented as mean ± standard deviation (SD) or number and percentage. Kaplan–Meier statistics and log–rank test were used to compare the risk for a re-revision. Survival data are presented as percentage not re-revised and SD. Cox regression analyses adjusted for age and gender were applied. Data from the regression analysis is presented with hazard ratio (HR) and 95% confidence interval (CI). P value <0.05 was considered as significant.

Results

Patients operated upon with a DMC were older (75 ± 9 years) compared with those operated with a non-DMC (73 ± 10 years, p = 0.005). There was no significant difference regarding gender between groups (p = 0.30). The distribution of diagnostic indications for primary surgery did not differ between groups; 25% had been operated upon for secondary arthritis in both groups (p = 0.82) (Table 1).

Table 1 Baseline demographics and surgical data
Full size table

The most frequently used incision approach during revision surgery was posterolateral (DMC 65%, non-DMC 61%), followed by anterolateral (DMC 32%, non-DMC 37%). There was no significant difference (p = 0.08) regarding type of incision between groups. In the DMC group, concomitant stem revision was less common (16% vs 35%, p < 0.001) and mean follow-up (3.1 ± 2.4) shorter compared with the non-DMC group (4.4 ± 3.1) (p < 0.001) (Table 1). During the study period, 6.4% of the Avantage™ cups and 14% of non-DMC cups had been re-revised. The most common reason for a re-revision in the DMC group was infection (3.4%, n = 15), followed by dislocation (1.6%, n = 7). In the non-DMC group, 24 cases (6.8%) were re-revised due to dislocation and 13 (3.7%) due to infection (Table 1).

For up to four years there was a lower risk for re-revision due to all reasons when an Avantage™ cup had been inserted (91% ± 3.7% vs 86% ± 4.1%, p = 0.02) (Fig. 2a). Re-revision cup was less common in the DMC group (97% ± 2.0% vs. 93% ± 3.1%) (p = 0.02) (Fig. 2b). Using re-revision due to dislocation as end-point, the Avantage™ cup performed better than a standard cemented cup (96% ± 3.0% vs 92% ± 3.3%, p = 0.001) (Fig. 2c). After adjusting for age and gender the non-DMC group had an increased risk for revision due all causes (HR 1.7, CI 1.1–2.8) and revision due to dislocation (HR 3.6, CI 1.5–8.4). The risk for re-revision of the acetabular component for any reason was also higher for the non-DMC group (HR 2.1, CI 1.0–4.5), albeit not significant (p = 0.05).

Fig. 2
figure 2

Survival using Kaplan–Meier statistics. Percentage not revised due to all causes (top); percentage without re-revised cup (middle); percentage not re-revised due to dislocation (bottom)

Full size image

Discussion

In patients with first-time revision hip arthroplasty due to dislocation, we found favourable short to mid-term implant survival for the Avantage™ DMC compared with cemented polyethylene cups up to four years. In particular, the Avantage™ DMC was associated with a considerably lower risk of re-revision due to dislocation.

In revision surgery, the type of complication leading to the first revision is also the most common reason for further revision surgeries [4]. Besides periprosthetic infection, dislocation is a leading cause of early failure following primary THA. The aetiology of dislocation is often multifactorial; patient factors such as neurological comorbidities and soft tissue condition around the hip, as well as factors related to surgical technique, implant positioning and implant characteristics may all individually or in combination explain dislocation [28]. However, how the different factors contribute in any given patient with instability may be cumbersome to determine. The main challenge in revision due to dislocation is to achieve a stable situation based on careful preoperative planning and optimum surgical technique. Choice of an implant with reliable protective effect against dislocation is also desirable, as long as this implant is not associated with other negative effects such as increased wear or increased risk of loosening.

The Avantage™ DMC concept, with its large mobile polyethylene liner, constrained head, irregular-shaped shell design to reduce impingement and high jump distance, theoretically protects from dislocation. Many retrospective studies with limited numbers show acceptable dislocation rates in both primary and revision hip replacement using DMC [29]. To our knowledge, there are no randomised trials providing thorough evidence of the efficacy of DMC to prevent instability. Remaining concerns about fixation in cemented metal-backed polyethylene cups and speculation regarding increased polyethylene wear add to the uncertainty of the long-term outcomes [30, 31].

Guyen et al. investigated the DMC in unstable revision hip arthroplasty [18]. They retrospectively reviewed 54 patients with revision due to instability using DMC and reported no further revision due to dislocation at a mean follow-up of four years. In an analysis of 228 revisions performed due to dislocation, where a DMC had been used the authors report a survival of 99% at two years with re-revision due to dislocation as the endpoint [25]. Saragaglia et al. [32] analysed a group of 29 revisions performed due to dislocation. With a mean follow-up of four years, only one patient (3.6%) had redislocated.

In our analysis, including 436 revisions performed with a DMC cup, cup with revision due to dislocation at four years was 96%. Regardless of outcome being studied, survival for DMC cups was higher than when a traditional cemented cup was used.

We acknowledge that our study has some limitations. The mean follow-up in the DMC and control groups were only 3.1 and 4.4 years, respectively. However, dislocation is a complication that generally occurs early after revision surgery, suggesting a sufficient follow-up period for the primary outcome measure. A source of bias is the possible nonrandom variation of individual surgeons’ preferences to use cemented DMC or cemented non-DMC revision implants. Patient-related factors, such as severity of instability pre-operatively, anatomical capacity to accommodate a cup of desired inner diameter, possible differences in bone loss severity between groups and need for bone grafting may all contribute to bias that cannot be controlled for in this study. Finally, our results are only valid for one single design. There are no studies indicating that the protective effect of a DMC cup should vary between different manufacturers and designs. In SHAR, there are reports on other DMC designs, both cemented and uncemented, but so far, numbers and follow-up are limited, not allowing further analysis of these designs.

The strengths of this observational study are the relatively large number of cases included and the use of high-quality, continuously validated data of the SHAR. The results are representative on a national level comparing two different acetabular revision concepts used in routine health care performed by different surgeons at many different hospitals.

In conclusion, cemented DMC used in hip revision surgery due to dislocation appear to protect from further dislocations. Although long-term overall implant survival is not known, these favourable short- to mid-term results suggest an increased use of DMC in acetabular revisions for dislocation when the cemented technique is an option.