Abstract
This systematic review reports the clinical, patient-reported, radiographic outcomes, complications and survivorship of dual-mobility trapeziometacarpal (TMC) joint arthroplasties including the MAÏA®, Moovis® and Touch® prostheses. Thirteen studies were included and outcomes reported for 1421 such arthroplasties. Overall, improvements in strength, range of motion, pain, function and satisfaction were demonstrated with a complication incidence of 13%. The risk of dislocation was 0.6%. The follow-up period was not long enough to draw conclusions about the long-term survival of the implants. Also, due to the low quality of the evidence, it was not possible to conclude that dual-mobility TMC arthroplasty was superior to trapeziectomy. This review highlights the need for Level I evidence with adequate long-term follow-up for TMC joint arthroplasty.
Keywords
Introduction
Trapeziectomy is considered the reference standard surgical procedure for trapeziometacarpal (TMC) joint osteoarthritis, although the most recent Cochrane review was unable to recommend the superiority of any surgical treatment for this condition (Wajon et al., 2015). The overall incidence of success for trapeziectomy and its variations is about 80% (Gangopadhyay et al., 2012). Concerns regarding subsidence, impingement and duration of recovery have led to the development of alternative options to improve outcomes. Reissner et al. (2016) found a 54% loss of metacarpal height in their series, whereas Mattila and Waris (2019) found carpometacarpal impingement to be the most common reason for revision after trapeziectomy. A qualitative study by Stepan et al. (2022) highlighted patients’ surprise at the length of recovery following this procedure.
TMC joint implant arthroplasty was developed to address some of these concerns. Currently, three third generation, uncemented, dual-mobility implants are available including the Moovis® (Stryker, Pusignan, France), Touch® (KeriMedical, Les Acacias, Switzerland) and MAÏA® (Groupe Lépine, Genay, France) prostheses. The Moovis® prosthesis consists of a hydroxyapatite (HA)-coated titanium metacarpal stem, a modular stainless-steel neck and head, a highly cross-linked polyethylene liner, and a trapezial cup that is cobalt-chromium on the articular surface and HA-coated titanium on the convex surface. The Touch® has similar design features to the Moovis® except that the cup articulation is stainless steel. Both implants were launched in 2012. They offer conical and hemispherical cup options. The MAÏA® was introduced in 2015 and is similar in design except that it has only a hemispherical cup option, which is made of titanium alloy. All three implants are modular with options for cup size, stem size, neck length and offset. The MAÏA® also offers a titanium neck and head for use if the patient has a nickel allergy.
This systematic review sought to answer questions about the clinical and patient-reported outcomes, complications and survivorship for the three dual-mobility TMC prostheses. The ultimate aim is to determine whether or not this procedure is superior to trapeziectomy.
Methods
This systematic review was conducted according to the Preferred Reporting Items for Systematic Review framework (Page et al., 2021) and registered in the PROSPERO International Prospective Register of Systematic Reviews (registration number: CRD42024517380).
Inclusion criteria were articles published in English, articles involving a TMC joint prosthesis with dual-mobility, level I–IV studies according to the Jovell and Navarro-Rubio (1995) classification, and articles published since 2012. There was no minimum follow-up. Studies were categorized as interventional (subdivided into randomized and non-randomized) and non-interventional. For interventional studies, any comparator was accepted, provided that at least one group included a dual-mobility TMC arthroplasty.
Exclusion criteria were: articles not available in English, articles on single-mobility TMC joint prostheses only, Jovell and Navarro-Rubio classification level V studies (including case reports and review articles), cadaver studies, biomechanical studies and animal studies.
The MEDLINE, Embase and Cochrane CENTRAL databases were searched on 31 January 2024 by one author (LM) with the assistance of a librarian. All peer-reviewed journals published since 2012 were considered.
Search strategy
The following keywords were used in combination to obtain search results: (‘trapeziometacarpal’ or ‘TMC’ or ‘CMC’ or ‘carpometacarpal’) and (‘prosthesis’ or ‘replacement’ or ‘arthroplasty’) and (‘Maïa’ or ‘Moovis’ or ‘Touch’ or ‘dual-mobility’ or ‘double mobility’). A filter was applied to exclude studies published prior to 2012.
Titles and abstracts were independently screened for relevance by two reviewers (AR and LM). Potentially relevant articles were then retrieved with the assistance of a librarian. These full texts were screened for inclusion and exclusion criteria by AR and LM independently. There was unanimous agreement on the inclusion or exclusion of each article. No automated tools were used (Figure 1).
Preferred Reporting Items for Systematic Review flowchart depicting study selection.
Data collection
Study characteristics including study type, study size, implant type, study period, journal, year of publication and country of origin were collected by two reviewers (AR and LM) with data on number of patients, implants and surgeons, level of surgical expertise, age, sex, and length of follow-up. The following outcomes were assessed: clinical outcomes, including range of motion, grip strength and key/tip pinch strength; patient-reported outcomes, including pain scores, functional scores, satisfaction and return to activities of daily living (ADL); complications; revision; and reoperation.
Validated outcome measures were sought. Range of motion was defined by the Kapandji Opposition and Counter Opposition Score. Patient-reported outcome measures included were the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire, the Michigan Hand Outcomes Questionnaire (MHQ), the Shortened Disabilities of the Arm, Shoulder and Hand Questionnaire (QuickDASH), the 36-item Short Form Health Survey (SF-36), the Short Form McGill Pain Questionnaire (SF-McGill) and a visual analogue scale. For studies with missing or unclear information, attempts were made to contact the author by e-mail. Studies without methodology or validated outcome data were excluded.
Risk-of-bias assessment
A Coleman methodology score was assigned to each study by two reviewers (AR and LM) (Tallon et al., 2001). A score of 85–100 was considered excellent, 70–84 was good, 50–69 was fair and <50 was poor. The Cochrane risk-of-bias tool for randomized trials, version 2 (RoB-2) and Cochrane risk-of-bias in non-randomized studies of interventions (ROBINS-I) tools were used to assess the risk of bias in randomized control trials (RCTs) and non-randomized intervention studies (Sterne et al., 2016; Sterne et al., 2019).
Effect measures
Means were reported for continuous outcomes and, where available, a range was also reported.
Synthesis methods
Study characteristics were assessed for clinical homogeneity to determine whether meta-analysis was feasible. Individual study results were summarized to describe study characteristics, Coleman methodology score, clinical outcomes, patient-reported outcomes and complications. Survival was presented graphically for each study.
Assessment of certainty
The certainty of the evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool (Guyatt et al., 2008).
Results
Fifteen studies met the eligibility criteria. Two were excluded because they included duplicate cases, leaving thirteen studies for review (Dreant and Poumellac, 2019; Dremstrup et al., 2021; Falkner et al., 2023; Farkash et al., 2023; Froschauer et al., 2021; Gonzalez-Espino et al., 2021; Guzzini et al., 2023; Klim et al., 2023; Lussiez et al., 2021; Martins et al., 2020; Reischenböck et al., 2024; Tchurukdichian et al., 2021; Theyskens and Vandesande, 2022). These included two randomized-controlled trials, six prospective cohort studies and five retrospective cohort studies.
Study characteristics are shown in Table 1. Farkash et al. (2023) reported on both the Touch® and the MAÏA® dual-mobility implants in addition to single-mobility implants. Only data that reliably related to the dual-mobility implants were extracted from this study. Three studies reported on the same patient cohort. To avoid duplication of results, only the most recent study was included (Reischenböck et al., 2024) and the other two were excluded (Herren et al., 2023a; Herren et al., 2023b). Another study was excluded (Duché and Trabelsi, 2022) because it focused on the technical aspects of TMC arthroplasty and did not provide sufficient methodological information to draw meaningful conclusions about the outcomes of the patient cohort.
Study characteristics
Risk of bias in studies
Each study was assigned a Coleman methodology score (see Supplemental Table S1). One study was of excellent quality, five were good, three were fair and four were poor. Both RCTs were assessed for bias using the RoB-2 tool. Both authors (LM and AR) raised some concerns of bias for each RCT using this tool. One non-randomized comparative study had a serious risk of bias according to the ROBINS-I tool.
Results of individual studies
Four studies described the postoperative rehabilitation protocol, which ranged from immobilization in a plaster cast for 3 weeks to immediate unrestricted use of the thumb. Hand therapy was not used or not clearly described in any study. Ten studies reported clinical outcomes, including range of motion, Kapandji opposition score, grip strength, key and tip pinch strength. Six reported radiological outcomes. Eleven studies included patient-reported outcomes, including pain and function scores. Four reported return to ADL. Seven included satisfaction scores and two included a general health or quality of life measure. All studies described complications, but, in one study, these were indistinguishable from those associated with single-mobility implants and were therefore not included in this review.
Clinical outcomes
Table 2 summarizes the clinical outcomes. Mean Kapandji opposition scores and grip, key and tip pinch increased postoperatively in all studies that assessed these parameters. Seven studies included radiographic assessments. Radiolucency around the stem and the cup was noted in 15 cases, but the clinical importance of this was not confirmed.
Postoperative clinical outcomes.
kgf: kilogram-force. NR: not reported. *Clinical outcomes omitted because it is impossible to differentiate from single mobility implants in this study.
Patient-reported outcomes
Table 3 summarizes the patient-reported outcomes. Measures included the MHQ, DASH and QuickDASH. Two studies found improvements in general quality of life measures after dual-mobility TMC arthroplasty (Falkner et al., 2023; Klim et al., 2023).
Postoperative patient reported outcomes.
ADL: activities of daily living. DASH: Disabilities of the Arm, Shoulder and Hand Questionnaire. MHQ: Michigan Hand Outcomes Questionnaire. NR: not reported. PROM: patient reported outcome measure. QuickDASH: Shortened Disabilities of the Arm, Shoulder and Hand Questionnaire. VAS: visual analogue scale. *Patient reported outcomes omitted because it is impossible to differentiate from single mobility implants in this study.
Return to ADL
Four studies included information on return to ADL or work. Four patients did not return to work, and two patients changed their type of work. The mean time to return to work ranged from 38 days to three months.
Complications
Complications are summarized in Table 4. Of those reporting implant dislocation, the majority described dislocation of the head from the polyethylene liner, rather than dislocation of the liner from the cup. The earliest dislocation occurred on the second postoperative day and the latest at 24 months postoperatively. Reischenböck et al. (2024) reported one dislocation in their cohort of patients who received capsular resection, but none in their cohort who received capsular suture. The authors did not attribute the dislocation to capsular resection, but many potentially confounding factors make interpretation of this event difficult.
Complications.
NR: not reported. *Other complications and reoperation omitted because it is impossible to differentiate from single mobility implants in this study.
Revision
Twenty-five cases underwent revision surgery for dislocation, infection, loosening, polyethylene wear, or trapezial fracture. Of these, ten were revised to a different prosthesis, seven had the prosthesis removed with trapeziectomy, and eight were not described. Studies describing the revision arthroplasty after dislocation detailed the use of a longer neck, a larger cup, and/or cementation of the cup. There were no reports of redislocation.
Reoperation
Twenty-six cases were re-operated with implant retention. The most common reason for reoperation was De Quervain’s tenosynovitis. Other indications for reoperation were rupture of the extensor pollicis longus (requiring extensor indicis proprius transfer), trigger thumb and deep infection.
Survivorship
Eleven studies looked at implant survival. Failure incidence ranged from 0% to 4% for the Moovis® and 0% to 5% for the Touch®. It was not possible to separate the failure incidence for the dual-mobility Maïa® from the other implants included in one study. Survival incidence for the studies are shown in Figure 2.
Trapeziometacarpal arthroplasty survivorship for individual studies. TMC: trapeziometacarpal.
Results of syntheses
Meta-analysis or other synthesis of results was not possible due to heterogeneity.
Certainty of evidence
Using the GRADE tool, ten studies were deemed to be of ‘very low quality’, which means that any effect estimate is very uncertain. Two studies were of ‘low quality’, meaning that further research would very likely have an important effect on our confidence in the effect estimate and is likely to change the estimate. One study was of ‘moderate quality’, which means that further research would likely have an important effect on our confidence in the effect estimate and may change the estimate.
Discussion
The studies included in this systematic review describe good clinical and patient-reported outcomes for TMC arthroplasty. One comparative study demonstrated a minimally clinically important difference (MCID) in favour of TMC arthroplasty, with the greatest difference seen in the early postoperative period (Bohannon, 2019; Guzzini et al., 2023; Villafañe et al., 2017). In this review, strength values also reached MCID for cohort studies comparing pre- and postoperative values. Minimal clinical important differences were found in favour of dual-mobility arthroplasty for pain and function, although these differences decreased over time (Guzzini et al., 2023; Klim et al., 2023; Myles et al., 2017; Sorensen et al., 2013). Other studies focusing on single-mobility implants also showed early PROM improvements (Jager et al., 2013; Toffoli and Teissier, 2017).
Although the studies included in this review describe positive clinical and patient-reported outcomes, the limited strength of the evidence base is of concern and does not allow formation of solid conclusions. TMC arthroplasty needs to be shown to have superior outcomes if it is to replace, or even coexist with, the cheaper and technically simpler trapeziectomy (Huang et al., 2015). The additional risks of such a procedure must also be considered, including loosening (0.3%) and dislocation (0.6%). The risk of dislocation with dual-mobility TMC arthroplasty is considerably lower than the 5% seen with single-mobility implants (Chiche et al., 2023). A greater arc of motion may help explain this (Lussiez et al., 2021). In their series, Chiche et al. (2023) found that half of their dislocations occurred within the first year and attributed these to cup malposition. This review did not examine cup positioning, but, given the low incidence of dislocation, one could speculate that this may be less important with dual-mobility implants.
Proponents of TMC arthroplasty cite an improvement in recovery time compared with trapeziectomy. TMC osteoarthritis affects a diverse group of patients with varying levels of normal activity (Wajon et al., 2015). This may explain the ambiguous or absent reporting of return to ADL or work in many of the studies evaluated in this review. In the non-randomized studies, confounding factors may have influenced return to work. For example, a manual worker may have been less likely to be offered TMC arthroplasty than a retired patient with lower activity demands. One would expect an earlier return to light work than heavy manual work, which could have biased the results. To date, an earlier return to ADL has not been proven for TMC arthroplasty, but this warrants further investigation (Andrzejewski and Ledoux, 2019; Chiche et al., 2023).
Despite the lower risk of dislocation with dual-mobility implants, TMC arthroplasty remains a technically demanding procedure with a steep learning curve. In a registry analysis of 43,076 thumb base procedures, arthroplasty was associated with a 2.5 times higher risk of subsequent procedures than trapeziectomy (Lane et al., 2021). Other hand and wrist arthroplasties have been shown to have higher incidences of complication with surgical inexperience, and it seems logical that TMC arthroplasty would follow this pattern (Brown et al., 2024). A limitation of this review is the range of expertise of the surgeons involved (Tang and Giddins, 2016). It was not possible to draw conclusions about the influence of surgical expertise on outcomes or complications.
It is possible that subtle differences in implant design can influence results. For example, the Moovis® and Touch® both have two different cup geometry options. There are differences in polyethylene composition and manufacture between implant companies that may influence wear properties (Newton and Talwalkar, 2022). In vitro differences in oxidative resistance and mechanical behaviour have been demonstrated between total hip arthroplasty polyethylene from different manufacturers (Gómez-Barrena et al., 2009). In addition, each company’s trapezial cup has a different metal bearing surface. Comparison of individual implants was beyond the scope of this review, but this should be revisited when the evidence base is sufficient to allow for subgroup analysis.
A meta-analysis was not possible due to the heterogeneity of the studies. Although limited conclusions can be drawn to support or refute the superiority of arthroplasty over other surgical procedures for TMC osteoarthritis, this review provides reassuring results for dual-mobility prostheses in terms of dislocation and the incidence of early failure. However, the overall complication risk of 13% is higher than for trapeziectomy, suggesting that this procedure is not benign (Huang et al., 2015; Lane et al., 2021). Heterogeneous and often inadequate follow-up times have made interpretation of TMC arthroplasty survival unreliable (Tang et al., 2019). It is important to assess how the arthroplasty performs in the long term, benchmarking against the sustained results seen with trapeziectomy (Gangopadhyay et al., 2012). This review highlights the need for high quality, level I evidence with appropriate long-term follow-up for TMC arthroplasty.
Supplemental Material
sj-pdf-1-jhs-10.1177_17531934241292249 - Supplemental material for Outcomes of dual-mobility trapeziometacarpal arthroplasties: a systematic review
Supplemental material, sj-pdf-1-jhs-10.1177_17531934241292249 for Outcomes of dual-mobility trapeziometacarpal arthroplasties: a systematic review by Lucy Maling and Aaron Rooney in Journal of Hand Surgery (European Volume)
Footnotes
Registration
This review is registered on the PROSPERO International Prospective Register of Systematic Reviews (registration number: CRD42024517380).
Data availability
All relevant data are presented in this review. Further data can be made available on request, including template data collection form and data extracted from included studies.
Acknowledgements
We thank the librarians at the University of Cambridge Medical Library for their efforts in sourcing articles for this review. LM is the guarantor of this study.
Declaration of conflicting interests
The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
Ethical approval
Ethical approval was not required for this systematic review of existing literature
Informed consent
Informed consent was not required for this systematic review of existing literature. No patient identifiable information was included.
Supplementary material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
