Dual Mobility Total Hip Replacements in Young Patients- A Systematic Review

Abstract

Background

There is little consensus on outcomes of dual Mobility total hip replacement (DM-THR) in younger patients. We performed a systematic review to examine this.

Methods

A systematic review of all observational studies and randomised controlled trials of patients under 70 with primary DM-THR in MEDLINE, EMBASE, Pubmed, Cochrane was performed according to PRISMA guidelines. Collected data included demographics, revision, dislocation and infection rates and functional scores.

Results

Twelve papers studying 738 DM-THR in 664 patients were included. The mean patient age was 46.9 years (13.5–69.0), with 319 (48.0%) being female and 419 (52.0%) being male. After an average follow-up of 90.0 months (1–371 months), the revision rate was 12.7% (84 hips), 0 dislocations were reported but 2.5% (15 hips) underwent intra-prosthetic dislocation. 32.1% hip revisions were due to aseptic loosening.

Conclusion

Limited literature surrounding the use of DM-THR suggests favourable stability but higher revision rates than conventional THR.

Introduction

Total hip replacement (THR) remains one of the most effective treatment options for end-stage osteoarthritis of the hip, but outcomes are persistently affected by hip dislocation, accounting for 17.4% THR revisions in the United Kingdom (UK) in 2020 [1]. An attempt to increase the stability of the THR using two articulations was first devised by Bousquet and Rambert in 1974 [2]. This dual mobility total hip replacement (DM-THR) yielded an improved biomechanical profile, increasing hip stability.

The DM-THR is comprised of a 22 or 28 mm femoral head which lies within a polyethylene (PE) liner, and is freely mobile within it. Furthermore, the PE liner articulates with a metal shell, creating two articulations within the same implant. This firstly confers increased stability by permitting a large difference in radius between the femoral head and shell, and therefore lower frictional torque and wear as per Charnley’s principle [3]. Secondly, the addition of the PE liner allows for an effectively larger femoral head to be used, thus increasing both the jump distance and femoral head-to-neck ratio. This confers additional implant stability [4–6]. Evidence has conclusively supported the use of DM-THR as a means to improve hip stability and its efficacy has been widely proven in patients with a high risk of hip dislocation [7].

However, although DM-THR has shown efficacy in reducing dislocation, there remain concerns over higher rates of loosening, wear and the potential for adverse reaction to metal debris (ARMD) [8]. There’s also a concern about accelerated wear due to the presence of multiple articulations, although this has not been conclusively proven. Furthermore, wear at a third articulation between the PE liner and the femoral neck can risk intra-prosthetic dislocation (IPD) [8].

ARMD first arose with the use of modular dual mobility cups, comprised of a titanium acetabular shell with a cobalt-chrome liner. This modular design facilitates placement of screws and augments in the porous cup, however fretting corrosion at this interface releases metal ions, increasing the risk of ARMD [9].

Despite this, contemporary modifications of the DM-THR have proven to be a well-recognised viable option in patients at high risk of dislocation such as in older patients, following proximal femoral fracture and neuromuscular disorders [7, 10]. Despite its efficacy in the older age group, relatively little is known about its use in the younger patients. This systematic review aimed to summarise the literature on the use of the dual mobility total hip replacement in younger patients, assessing implant revision rates, hip function and complications.

Methods

Search Strategy

A systematic review of Medline, EMBASE, Cochrane and Pubmed was taken according to PRISMA guidelines on 6th February 2022 [11]. In addition Allied and Complementary Medicine (AMED), Cumulative Index to Nursing and Allied Health Literature (CINAHL) was searched, supplemented by manual searching (Fig. 1). The search strategy was devised according to the PICOS structure (Table 1). Inclusion criteria were all studies regarding the use of dual mobility total hip replacement in adult patients under the age of 70. Opinion-based reports and surgical technique papers were excluded. The primary outcome studied was implant survivorship (assessed through revision rate), with secondary outcomes including complications such as infection and dislocation, functional hip scores and return to work. The review was registered on PROSPERO (ID CRD42022300386).

Fig. 1.

PRISMA flow diagram

Table 1.

PICOS structure used to frame question for Systematic review on DM-THR in younger patients

PICO	Query
Population	‘Young’
Intervention	‘Total Hip arthroplasty’ OR ‘Total hip replacement’ AND ‘Dual mobility’
Control
Outcome	‘Revis*’ OR ‘Dislocation’ OR ‘Infection’
Studies	Clinical studies randomised controlled trials, prospective or retrospective case series, case control studies,

Data Extraction and Collection

Data extraction was performed by two independent authors (SW and MO), with conflicts resolved by a third (SS). Following finalisation of included studies, data extracted included patient demographics, follow-up duration and outcomes included functional scores, revision rate, dislocation and infection. Furthermore, data was extracted on ARMD and radiological appearances. In two cases, studies involved the same cohort of patients [12–16] and data was therefore summarised using the results of these combined papers.

Meta-Analysis

A meta-analysis was performed of Harris Hip scores (HHS) and Modified HHS stated in 4 studies using Revman (Version 5.4, The Cochrane Collaboration). A random effects model was used. This continuous variable was expressed in terms of mean difference and standard deviation (SD), with 95% confidence interval (CI) stated. Graphical representation of this was performed using forest plots.

Risk of Bias

Quality analysis of included papers was performed for cohort and case control studies using the Risk Of Bias In Non-randomized Studies—of Interventions (ROBINS-I tool) [17].

Results

Patient Demographics

Twelve studies were included, studying 738 DM-THR in 664 patients. Eight were cohort studies and 4 were case control studies. Of the case control studies, 2 compared DM-THR with fixed bearing THR in 153 patients [18, 19] whilst 2 compared outcomes in younger patients with 769 older patients [20, 21] (Table 2). The mean patient age of young patients with DM-THR was 46.9 (13.5–69), with 319 (48.0%) being female and 419 (52.0%) being male. Body mass index (BMI) was reported in 5 studies [12, 13, 18–20], in which the mean was 27.6 (17.7–44.7). In reported studies, 54% DM-THR was performed on the right hip [12–15, 22]. Implant type was stated in 586 papers, of which 481 were conventional [18, 20, 21, 23, 24] and 105 were modular [12, 18, 25]. 427 (86.6%) were performed through a posterior, with the remaining 66 (13.4%) performed through a lateral or anterolateral approach.

Table 2.

Summary of included studies

Author	Year	Study design	No of DM-THR patients	No of DM-THR hips	Cemented/Uncemented/Hybrid	Comparison	Mean Pre-op functional score (SD)	Mean Post-op functional score (SD)	Outcome
Lawrie et al.	2021	Prospective cohort study	43	43	Uncemented	–	UCLA 7.6 (1.8) mHHS 54.8 (SD 19)	UCLA 7.9 (2.3) mHHS 93.08 (10.5)	1 Revision 0 Loosening 0 Failure
Nam et al.	2019	Prospective cohort study	43	43	Uncemented	–	UCLA 8.3 (2.2) mHHS 54.8 (SD 19)	UCLA 7.1 (2.3) mHHS 91.2 (10.8)	0 Revision 0 Dislocation
Assi, Kheir et al.	2018	Retrospective cohort study	26	30	Hybrid	–		mHHS 98.7 (2.7)	1 Peri-prosthetic# 0 Dislocation
Philippot et al.	2017	Retrospective cohort study	114	137	Uncemented	–	HHS 48.30 PMA 11.5	HHS 85.1 PMA 17.1	44 Revision 0 Dislocation 1Peri-prosthetic# 15 Intra-prosthetic dislocation 1 Infection
Assi El-Najjar et al.	2017	Retrospective cohort study	75	85	Hybrid	–		mHHS 97.1 (7.0)	1 Infection 0 Dislocation
Martz et al.	2017	Retrospective cohort study	31	40	24 Uncemented 16 Hybrid	–	HHS 50.8 (15.5) PMA 11 (3.3)	PMA 17.4 (1.12) HHS 95.7 (6.9) HOOS 90.5 (7.6) EQ5D 0.97 (0.06)	1 Infection 0 Dislocation
Puch et al.	2017	Case control	119		Uncemented	444 > 55yo	mHHS 39.5 PMA 8.8 Oxford 45.6	mHHS 98.0 PMA 17.0 Oxford 13.5	1 Infection 4 Revision 0 Dislocation
Morin et al.	2016	Retrospective cohort study	33	40	Hybrid		GMFCS V 33	GMFCS V 33	3 Mortality 2 Infection 6 Revision
Assi, Barakat et al.	2021	Case control	60		Hybrid	28 Osteonecrosis 127 NOF#		mHHS 96.6 (7.4)	1 Dislocation * 1 Periprosthetic# 1 Infection 1 Revision
Epinette et al.	2014	Case control	112		110 Uncemented 2 Hybrid	325 > 70yo	HHS 41.8	HHS 97.55 MDP 17.57 WOMAC 2.69	0 Dislocation 2 Revision 0 Infection
Rowan et al.	2017	Case control	117	136	Uncemented	136 Non-DM THA		mHHS 87.2 (16.6)	0 Dislocation 27 Revision 1 Periprosthetic#
Nam Salih et al.	2016	Case control	26	26	Uncemented	17 Non-DM THA	HHS 56.9 (20.8) SF-12 MCS 51.7 (13.0) SF-12 PCS 39.4 (10.8)	HHS 84.6 (23.9) SF-12 MCS 55.3 (8.3) SF-12 PCS 53.1 ± 8.7

HHS: Harris hip score

mHHS: Modified harris hip score

HOOS: hip disability and osteoarthritis outcome score

UCLA: University of California, los angeles activity score

PMA: Postel-Merle d’Aubigne scores

GMFCS V: Gross motor function classification system

WOMAC: Western ontario and mcmaster universities arthritis index

MDP: Merle d’aubigné et postel

SF-12 MCS: Short-Form 12 mental component score

SF-12 PCS: Short-Form 12 physical component score

*Describes complications in the whole patient cohort, not just young DM-THR patients

Outcomes

Five hundred 92 (89.2%) patients were followed up for an average of 90.0 months (1–371 months). Throughout the duration of follow-up, 84 (14.2%) patients required a revision (Fig. 2). In one paper by Rowan et al., 24 patients underwent revision due to implant recall (with this excluded, 60 (10.1%) patients had a revision). There were 0 dislocations but 15 (2.5%) IPD. 6 (1.0%) patients developed a prosthetic joint infection. Other complications which did not require implant revision included symptomatic heterotopic ossification, leg length discrepancy [24], osteoma, acetabular loosening, peri-prosthetic fractures in 2 patients and cortex resorption [22].

Fig. 2.

Reasons for DM-THR revision. Graphical representation of listed reasons for revision of DM-THR in young patients

When comparing outcomes between modular and conventional DM-THR, modular DM-THR had a lower revision rate (1.3 vs. 13.3%, respectively), with comparable rates of infection (0 vs. 0.8%), with no dislocations reported in the groups.

Functional Scores

Eleven studies undertook functional assessment using the Harris Hip Score (HHS) [12–16, 18–21, 23, 24]. HHS displayed an improvement from a mean score of 47.5 pre-operatively to 92.3 post-operatively. Furthermore 3 studies utilised the Postel Merle d’Aubigne (PMA) score [20, 23, 24], reporting an increase in the score from a mean of 9.9 pre-operatively, to 17.2 post-operatively. Functional outcome scores are summarised in Table 2 and Fig. 3.

Fig. 3.

Forest plot of harris hip scores (HHS) before and after dual mobility THR. Forest plot of harris hip scores (HHS) amalgamated with modified HHS before and after dual mobility total hip replacement in young patients

Adverse Reaction to Metal Debris

Metal ion levels and ARMD was reported in 4 studies [12, 13, 19, 21]. Lawrie et al. [12] reported significant increases in Cobalt (0.065–0.08 µg/l) and Titanium (0.35–0.78 µg/l) levels at the 5 year follow-up mark, with a non-significant increase in chromium (0.18–0.23 µg/l) levels, also seen in the same cohort after 2 years of follow-up [13]. Statistically significant increases in serum Cobalt (0.18 vs. 0.015 µg/l) were found by Nam et al. [19] in young DM-THR cohort when compared with conventional THR after 1 year. Furthermore, Epinette et al. [21] reported a single case of stem revision secondary to aseptic lymphocytic vasculitis associated lesion (ALVAL). 24 femoral stems were revised in Rowan et al. cohort due to recall of stems due to ARMD [18].

Radiographic Appearances

Radiographic appearances were described for 512 patients in 9 studies [12–16, 20, 21, 23, 24]. Of these, osteolysis was described in 51 (10.0%) of patients. Acetabular osteolysis was noted in 46 patients, femoral osteolysis in 3 with bipolar osteolysis in 2 patients. Loosening was noted in 20 (3.9%) patients 19 in the acetabulum with 1 in both the femoral and acetabular components. Further, Lawrie et al. described a reduction in Gruen Zone 1, with a reduction then increase in Gruen Zone 2 over a 5 year period, but no change in Gruen Zones 3–7 [12]. Philippot et al. [23] observed a Gruen Zone 7 granuloma in 17% cases.

Risk of Bias

All studies had an acceptable risk of bias using the ROBINS-I tool (Table 3).

Table 3.

ROBINS-I tool assessment of risk of bias of included studies

	Lawrie et al.	Nam et al.	Assi, Kheir et al.	Philippot et al.	Assi, El-Najjar et al.	Martz et al.	Puch et al.	Morin et al.	Assi, Barakat et al.	Epinette et al.	Rowan et al.	Nam, Salih et al.
Bias due to confounding	Mod	Mod	Mod	Mod	Mod	Mod	Serious	Mod	Mod	Mod	Mod	Mod
Bias in selection of participants into the study	Mod	Mod	Mod	Mod	Mod	Mod	Mod	Mod	Mod	Mod	Mod	Mod
Bias in classification of interventions	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low
Bias due to deviations from intended interventions	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Serious	Low
Bias due to missing data	Serious	Mod	Low	Low	Mod	Low	Mod	Mod	Mod	Mod	Mod	Mod
Bias in measurement of outcomes	Mod	Mod	Low	Low	Low	Low	Low	Low	Low	Low	Low	Mod
Bias in selection of the reported result	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low	Low

Discussion

Whilst dual mobility has traditionally been reserved as a salvage for patients at a high risk of dislocation, enhanced implant biomechanics and design have led surgeons to question its indication in young, active patients [8]. This systematic review shows that although DM-THR has favourable outcomes in terms of function, rates of infection and stability, there are concerns when considering the revision rate and the potential for ARMD in these patients.

This systematic review revealed a mean revision rate of 12.7% (84 patients), which exceeds rates quoted in the literature of 2.0–3.5% [9, 10, 26]. However, a number of factors may be confounding this. Firstly, this value is confounded by the high revision rate in Rowan et al.’s cohort (24 patients) due to recall of the femoral stem [18]. It is further skewed by the over 10% loss to follow-up in included studies. Secondly, Philippot et al. study reports a 32.1% revision rate over 20 years, however included patients underwent DM-THR between 1985 to 95. Since this period, vast modifications have been made to the DM-THR to improve polyethylene wear and resistance to loosening of the acetabular component [8], and therefore this higher revision rate may not be generalisable to modern day practice. It is theorised that younger patients will have a fuller range of motion of the hip, predisposing to more impingement of the femoral neck on the shell and loosening of the cup [27, 28]. This is corroborated in this review, in which 32.1% revisions were due to aseptic loosening. This increased risk of revision due to loosening in these patients must be factored into the surgeon’s decision making when choosing the most appropriate implant in younger patients.

Although the revision rate due to loosening and wear was high in this cohort, this must be balanced with the 0% dislocation rate observed in younger patients. With instability being one of the most prevalent causes of THR revision [29], the impact of an implant with higher biomechanical stability in younger patients (with higher requirement for implant longevity) is significant. The low dislocation rate in DM-THR when compared to conventional THR is well described [30], with reported dislocation rates from 2.2 to 18% [26]. Furthermore, there are crucial cost saving applications to this, particularly in younger patients. Khoshbin et al. [31] found DM-THR to be cost saving in patients under 75 years of age, with an incremental cost effectiveness ratio (ICER) of £3006 per quality adjusted life years in patients under 55. Furthermore, this cohort demonstrated an overall prevalence of 2.5% of IPD, which in keeping with reported ranges in the literature [5]. Furthermore, 14 of these dislocations were from a single cohort studied by Philippot, which may confound this. The justification for this high IPD rate was due to wear of the collar secondary to friction of the roughened femoral neck and insert [23]. The findings of this study suggest that benefits from increased hip stability must be balanced with a potentially higher risk of loosening in this population. Large patient sample study or a randomised controlled trial involving younger patients is required to ascertain contemporary implant survival, dislocation and loosening over a longer follow-up period.

An additional caution should be taken with regard to ARMD in these patients. This review has shown significant increases in cobalt and titanium serum levels in younger patients after a maximum of 5 years of follow-up, a finding replicated in mixed age cohorts [9, 25]. The effect of this remains unclear, with patients with raised metal ions in our cohort showing no significant difference in functional scores [12, 13, 19]. The impact of raised serum metal ions on function remains unclear. French et al. performed a systematic review of metal debris following DH-THR, finding no correlation between hip function scores and raised serum ions [9]. However, Civinini et al. [25] found that higher metal cobalt and chromium levels were significantly associated with a lower UCLA score, and Ko et al. noted metal artifact reduction magnetic resonance imaging (MARS MRI) evidence of ARMD in 22.2% patients with elevated serum cobalt levels [32]. In addition, retrieval studies on DM-THR have demonstrated evidence of clinically significant corrosion[33]. Further caution should be taken in younger patients, as they will likely spend many decades with the DM-THR in situ, which could potentially increase the risk of raised serum metal ion levels and therefore risk of ARMD. The maximum duration of follow-up for metal ion monitoring was 5 years, and studies with longer periods of monitoring of metal ion levels and signs of ARMD are required to give generalisable evidence for young patients undergoing DM-THR.

This study is limited by the relatively small number of included studies with low levels of evidence, the majority of which included a small patient cohort, with evidence of bias and confounding in reported outcomes. Further patient study with a larger sample size is required to truly ascertain the safety and efficacy of DM-THR in these patients.

Conclusion

Although DM-THR provides an effective stability profile in younger patients, further study is required to ascertain if rates of aseptic loosening and wear are higher in these patients. Furthermore, the true extent of ARMD secondary to fretting corrosion of the cup needs to be clarified. Younger patients require increased implant longevity, and may therefore be more susceptible to the effects of loosening, wear and ARMD. Caution should be taken with careful patient counselling when considering DM-THR.in younger patients.

Funding

The author(s) received no financial or material support for the research, authorship, and/or publication of this article.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.