Dual mobility trapeziometacarpal prosthesis: a prospective study of 107 cases with a follow-up of more than 3 years

Abstract

We report the results of a prospective study using a dual mobility trapeziometacarpal prosthesis (Touch®) in 107 patients with a minimum follow-up of 3 years. One-hundred and two patients (95%) were very satisfied or satisfied with the functional outcomes and the mean pain intensity in visual analogue scale decreased from 7.4 to 0.8 (p < 0.001). Thumb opposition (Kapandji score) index increased from an average of 8.0 to 9.4, while the mean QuickDASH score improved from 38 preoperatively to 20 at follow-up (p < 0.01). Key-pinch strength improved from 3.5 kg (range 0.5–9.5) to 5.5 kg (range 3.0–11.5). There was a 4.6% rate of complications, including cup loosening and wear of polyethylene, which required revision, but no cases of prosthetic dislocation were seen. Applying the dual mobility principle to trapeziometacarpal arthroplasty may significatively improve the stability of these prostheses. Radiolucent zones around the components of the prostheses are not systematic predictors of future loosening.

Level of evidence: IV

Keywords

Trapeziometacarpal osteoarthritis arthroplasty prosthesis dual mobility

Introduction

While total trapeziectomy (Gervis, 1949) remains the operation of choice in trapeziometacarpal (TMC) osteoarthritis for many surgeons, total TMC prostheses have emerged as one of the surgical options over the last 20 years. The first model was developed in the early 1970s (de la Caffinière, 1973) based on the principle of a ball-and-socket joint, although preliminary results were disappointing due to a high complication rate. A new generation of total prostheses became available in the 1990s (Badia and Sambandam, 2006; Comtet, 2000; Klahn et al., 2012), incorporating several modifications, including an improved design of the metacarpal stem, modular nature of the trapezial, metacarpal and intermediate implants, various cup shapes and also various modes of fixation. These changes have resulted in an improvement of the clinical and radiological outcomes, thereby expanding the indications for these prostheses (Semere et al., 2015; Seng et al., 2013; Teissier and Alkar, 2011).

Of the various complications that have resulted from the use of TMC prostheses, prosthetic dislocation remains a reason of failure of up to 10% (Andrzejewski and Ledoux, 2019; Dehl et al., 2017). To improve the stability of the prosthesis, the principle of dual mobility, as developed for total hip prostheses in 1975 (Farizon et al., 1998), was applied to TMC prostheses. With an increased diameter of the pre-assembled neck and insert, the articular stability of such prostheses is improved. The aim of this study was to report our experience with the use of dual mobility TMC prosthesis, including the clinical outcomes and rate of complications at a minimum of 3-year follow-up.

Methods

Study design

We prospectively followed-up a continuous series of 310 patients operated since July 2014 at three hospitals by three Level 4 surgeons (Tang and Giddins, 2016). In this study, the results of consecutive cases with a follow-up of more than 3 years were included. From these, the inclusion criteria were patients with TMC arthritis who were operated by one of the three surgeons, who received the Touch® prosthesis, and who agreed to a clinical and radiological follow-up at 1, 3, 5, 10 and 15 years. Exclusion criteria included patients who did not have the capacity to understand and/or consent to the study. This multicentre observational study was approved by an independent ethics committee in October 2018. Clinical and radiological data were entered into a clinical review database (OpenClinica®). All data were analysed by an independent observer.

Preoperative data

The following clinical parameters were assessed: pain using a visual analogue scale (VAS), amount of thumb opposition using the Kapandji's score, TMC range of motion (radial abduction, antepulsion and retropulsion), key pinch strength (JAMAR® dynamometer, White Plains, CT, USA), upper limb function using the Quick Disabilities of the Arm, Shoulder, and Hand (QuickDASH) questionnaire (Beaton et al., 2005), noting the presence of any metacarpophalangeal (MCP) joint hyperextension or Z-deformity of the thumb and finally, work activity of the patient (classified as light, moderate and heavy work). Patient satisfaction was classified as very satisfied, satisfied or disappointed. Dell's radiological classification (Dell et al., 1978) was used to record the presence of osteoarthritis of the scaphotrapeziotrapezoidal (STT) and MCP joints. The M1/M2 ratio (length of first metacarpal + trapezium/length of second metacarpal + trapezoid; normal value 0.73 SD 0.015) (Ledoux, 2017), was used to evaluate preoperative thumb shortening and postoperative restoration of normal thumb length.

Postoperative data

The above preoperative findings were compared with postoperative ones, and the presence of any complications including loosening, implant impaction or dislocation, polyethylene wear and peri-prosthetic calcifications were recorded. Postoperative radiographs were examined for the presence of metacarpal and trapezial radiolucent zones and recorded according to a standard scoring protocol (Figure 1).

The TMC Touch® prosthesis

The Touch® prosthesis (Kerimedical®, Route des Acacias, Les Acacias, Switzerland) reproduces the original concept of dual mobility as defined by Gilles Bousquet for a total hip prosthesis (Farizon et al., 1998), consisting of a mobile prosthetic head in a polyethylene insert that is freely moveable within a metal cup. The prosthesis was developed in 2013, based on the principle of a ball-and-socket joint with three degrees of freedom. It comprises two concentric articulations, a smaller articulation between the head and polyethylene insert, and a larger articulation between the polyethylene insert and metal cup, with metal-on-polyethylene bearings (Figure 2).

The prosthesis is modular, composed of a titanium alloy stem (Ti–6Al–4 V) (five sizes), an M30 stainless steel intermediate implant, with a 4-mm diameter head covered by a pre-assembled high-density 7-mm diameter polyethylene insert (three lengths and two angles) and an M30 stainless steel hemispheric cup, available in two sizes (9 and 10 mm in diameter). Implant fixation is both primary by a press-fit effect, and secondary by means of a double-layer porous titanium and hydroxyapatite coating, which promotes and accelerates osteointegration. The total in vitro range of motion of the prosthesis is 117°.

Surgical technique

A dorsolateral incision was made over the TMC joint, between the extensor pollicis brevis and extensor pollicis longus tendons. This approach avoids damage to nerve fibres of the superficial branch of the radial nerve and the radial artery by allowing identification of these structures. An arthrotomy was performed using a capsuloperiosteal flap, usually with an L-shaped proximal pedicle. The modular prosthesis was inserted with an instrumentation set provided by the manufacturer of the prosthesis. The arthroplasty was closed by suturing the capsuloperiosteal flap (2-0 Ethicon®, Johnson and Johnson, Cincinnati, OH, USA). Skin closure was performed without drainage (4-0 Ethilon®, Johnson and Johnson). Postoperatively, the thumb was immobilized in a splint for 2 to 3 weeks, followed by active mobilization of the thumb with intermittent use of a splint for another 2 to 4 weeks.

Statistical analysis

Parametric data were presented as the means (standard deviation, SD). Differences between the pre- and postoperative M1/M2 index were assessed using Student t-test. A Wilcoxon test was performed when parameters were not normally distributed (pain, range of motion, strength, Quick-DASH score). P values less than 0.05 are considered statistically significant. Data about the cohort of patients are summarized in Table 1. Most of the patients were women and retired. Most common aetiology was osteoarthritis (only two cases were rheumatoid arthritis). Concurrent STT osteoarthritis was observed in 16% of cases.

Table 1.

Characteristics of the patient cohort.

Parameter	Patients (n = 107)
Age (years)	64.5 (64–84)
Gender: male/female (%)	20/80
Dominant hand involved	66%
Follow-up (months)	40 (36–52)
Working /retired (%)	34/66
Manual activity (heavy/light) (%)	12/88
Dell Stages (2/3/4) (%)	31/56/13
Indication (idiopathic/ rheumatoid arthritis)	105/2

Results

Clinical outcomes

One-hundred and eighteen patients met our initial inclusion criteria, although 11 were subsequently lost to follow-up (one patient died, and ten patients could not be reached). Patient demographics of the remaining 107 patients are as shown in Table 1. One-hundred and two patients (95%) were very satisfied or satisfied with the functional outcomes, while five patients (5%) were disappointed. The mean pain intensity (VAS) decreased from 7.4 to 0.8 (p < 0.001) and the postoperative pain score was 0.90 when no radiological lesions were visible at the STT joint, and 1.3 when STT lesions are present. Comparing pre- and postoperative range of motion, a significant improvement of radial abduction was noted: 29° (range 15–45) to 42° (range 20–55), thumb antepulsion 33° (range 20–50) to 42° (range 20–55) and retropulsion –10° (range –30–0) to 10° (range 0–20). Thumb opposition (Kapandji score) index increased from an average of 8.0 to 9.4, while the mean QuickDASH score improved from 38 (24 to 51) preoperatively to 20 (11 to 39) at follow-up (p < 0.01). Key-pinch strength improved from 3.5 kg (range 0.5–9.5) to 5.5 kg (range 3.0–11.5).

Of the 34 patients who were working prior to their operation, 30 returned to work, with 29 in their original occupation; the one patient who did not return was previously a heavy manual labourer. The time period to resume working was less than 3 months for the 30 patients who return to their professional activity. MCP joint hyperextension was present in 42 patients (39%) preoperatively versus 23 (22%) at follow-up, although no surgical procedure was performed at this joint.

Complications

There were no cases of extraprosthetic or intraprosthetic dislocation observed. However, five complications (4.6%) required surgical revision, including the following.

One case of early cup tilting due to inappropriate preparation of the trapezium, requiring the cup to be changed (Figure 3).

Two cases of painful cup loosening after 25 and 36 months, requiring a change of cup in both cases, with autologous bone graft added in one case (Figure 4).

Two cases of polyethylene insert wear after 4 years of follow-up in two men who had returned to heavy manual work. Total trapeziectomy was performed with ligamentoplasty using a strip of the abductor pollicis longus tendon. The final outcomes were considered satisfactory by the patients.

Ten minor complications occurred, including four cases of De Quervain's tenosynovitis and six cases of trigger thumb between 3 and 12 months after the arthroplasty. No cases of deep infection or complex regional pain syndrome was observed.

Radiological findings

Radiographs were inspected for any evidence of implant impaction or radiolucent zones around the prostheses (Table 2). Modifications of periprosthetic radiolucent lines (extension, thickness) were noted at each follow-up visit, including the last one. There were no cases of stem deepening (metacarpal migration Stage 0). Twelve cups showed moderate non-progressive proximal migration (cup migration Stage 1) compared with the immediate postoperative radiograph, but all cups were completely integrated into bone at the last follow-up. The comparative mean pre- and postoperative M1/M2 ratios remained stable with subsequent follow-up, confirming the absence of stem or cup impaction over time (Table 3). No radiological changes around the stem were observed in 93 of the 107 cases (87%), but in the other 14 cases (13%), a linear radiolucent zone was observed around the proximal part of the stem (metacarpal loosening Stage 1) with no extension of this area at the last review.

Table 2.

Postoperative radiological analysis of stem and cup: presence of radiolucent lines (RLL) and implants migration.

Stages	Cup RLL (%)	Cup migration (%)	Stem RLL (%)	Stem migration (%)
0	91.5	88.9	87.3	100
1	8.5	9.5	12.7	0
2	0	1.6	0	0
3	0	0	0	0

Table 3.

Change in M1/M2 ratio between preoperative and postoperative follow-up visits.

	Follow-up (months)
	Preoperative	3	4–6	7–12	40
Average	0.70	0.75	0.75	0.75	0.74
Min	0.55	0.67	0.67	0.6	0.65
Max	0.83	0.9	0.90	0.92	0.92
SD	0.044	0.040	0.040	0.040	0.040

The M1/M2 ratio is the length of M1 + trapezium/M2 + trapezoid (Ledoux, 2017).

No radiological changes were observed around the cup in 98 cases (91%), but a linear radiolucent zone around the cup (cup loosening Stage 1) was noted in nine cases (8%) (Figure 5) that did not compromise the cup’s stability. At the last review, the radiolucent zone had not increased in seven of these nine cases, but in the two other cases, the radiolucent zone around the cup has expanded leading to the development of painful loosening, requiring revision. Postoperative calcifications were detected in 25% of cases: minor in 22% of cases and major in 3% of cases; none of these calcifications had any impact on the outcomes.

Discussion

In this study, we reported the use of the Touch® prosthesis with a metal-on-polyethylene bearing and double-mobility design. The metal-on-polyethylene bearing configuration was preferred to metal-on-metal or ceramic bearings (Completo et al., 2016; Hernandez-Cortés et al., 2011; Thillemann et al., 2016) because a higher rate of complications had been noted with the latter, such as trapezial component loosening due to metallosis and high serum chrome and cobalt levels (Hansen et al., 2013; Huang et al., 2015; Klahn et al., 2012).

The Touch® prosthesis was designed to achieve several objectives, of which the primary one was to decrease the dislocation rate, a complication feared by both surgeons and patients (Cootjans et al., 2017; Goubau et al., 2011; Martin-Ferrero, 2014). Dislocation occurs when the head is displaced from point A to point B (Figure 6) (Cooney and Chao, 1977) and by increasing the diameter of the preassembled insert/neck (7 mm) in the dual mobility prosthesis design, this distance is increased, thereby decreasing the risk of dislocation. The dual mobility design also increases the range of motion of the joint between the small (34°) and large articulations (83°) to a total of 117°, thus decreasing the risk of neck–cup contact if the orientation of the cup was not perfect (Brauns et al., 2019). The absence of dislocations in our first 107 cases confirmed the biomechanical stability of this dual mobility principle.

Radiological modifications around the prosthetic components were analysed extensively in our study. The 12 moderate proximal cup migrations did not compromise its anchorage in the trapezium; this migration may be due to an overly deep trapezial cavity preparation, or it may be due to mediocre bone quality. After the early migration, a new mechanical balance is established, which allows the cup to be fixed in the trapezium and in our opinion, therefore, these observed cup migrations should not be considered as complications or failures. Proximal radiolucent lines were observed in 14 cases around the stem and seven limited radiolucent lines around the cup (radiological Stage 1). These are considered classical radiological findings in non-cemented implants during the first months (Sillwater et al., 2017), due to periprosthetic remodelling and was noted in 13% to 53% of cases (Johnston et al., 2011; Lussiez, 2011; Martin-Ferrero, 2014; Semere et al., 2015; Seng and Chantelot, 2013). When intramedullary stems are implanted, these modifications are due to a stress-shielding effect and most of the radiolucent lines around the stem did not worsen over time. However, the radiolucent zone around the cup was worsened in two cases, leading to loosening and revision at 25 and 36 months. Implant loosening remains the main cause of surgical revision, with a rate ranging between 2.6% and 26% in the literature (Bricout and Rezzouk, 2016; Johnston et al., 2011; Seng and Chantelot, 2013; Toledo-Romero and De Torres-Urrea, 2011), mostly involving the trapezial component (Lussiez, 2011). Two cases of polyethylene wear were observed in our series, which occurred in two men who did heavy manual work after a period of about 44 to 48 months. Consequently, we now discouraged patients who are involved in heavy manual work to undergo surgery with a TMC prosthesis.

Several limitations exist in our study. While we have not reached the threshold for a medium-term follow-up, defined as 5 years (Tang et al., 2019), the minimum follow-up of 3 years in our cohort exceeds the 2 years typically needed for new implants and only a small number of the patients (less of 10%) were lost to follow-up. The analysis of this cohort will be continued to gather long-term outcomes as defined in the protocol. In addition, we did not perform a comparison between our prosthesis and alternative procedures, such as trapeziectomy, which may have yielded useful information about the use of prostheses as compared with more established methods.

In conclusion, our preliminary cohort of 107 cases using a dual mobility TMC prosthesis with more than 3 years of follow-up have shown promising results, with no incidence of early prosthetic dislocations, rapid recovery of strength and range of motion (Badia and Sambandam, 2006; De Smet et al., 2004; Jager et al., 2013). A high satisfaction rate was achieved in 95% in our cohort, with rapid return to work for 30 of 34 patients. Survival of prostheses in our cohort are similar to published series of metal-on-polyethylene prostheses with long-term follow-up (Martin-Ferrero, 2014; Semere et al., 2015; Teissier and Alkar, 2011). Minor radiological modifications have been noted around the cup and stem, without clinical consequences after 3 years. These modifications justify close monitoring of the prosthesis over the longer term.

Footnotes

Declaration of conflicting interests

The authors declare the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Bruno Lussiez: Consultant and Royalties (Kerimedical®); Cyril Falaise: Consultant and Royalties (Kerimedical®); Pascal Ledoux: Consultant and Royalties (Kerimedical®).

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The study was approved by the ‘Comité de protection des personnes’ RCB 2018-A01675-50. Clinicaltrials.gov: NCT03611023.

References

Andrzejewski

Ledoux

. Maia® trapeziometacarpal joint arthroplasty: survival and clinical outcomes at 5 year’s follow-up. Hand Surg Rehabil. 2019, 38: 169–73.

Badia

Sambandam

. Total joint arthroplasty in the treatment of advanced stages of thumb carpometacarpal joint osteoarthritis. J Hand Surg Am. 2006, 31: 1605–13.

Brauns

Caekebeke

Duerinckx

. The effect of cup orientation on stability of trapeziometacarpal total joint arthroplasty: a biomechanical cadaver study. J Hand Surg Eur. 2019, 44: 708–13.

Beaton

Wright

Katz

. Development of the QuickDASH: comparison of three item-reduction approaches. J Bone Joint Surg Am. 2005, 87: 1038–46.

Bricout

Rezzouk

. Complications and failures of the trapeziometacarpal Maia® prosthesis: a series of 156 cases. Hand Surg Rehabil. 2016, 35: 190–8.

Completo

Nascimento

Neto

. Total arthroplasty of basal thumb joint with Elektra prosthesis: an in vitro analysis. J Hand Surg Eur. 2016, 41: 930–8.

Comtet

2000 ARPE prostheses. In: Simmen

Allieu

Lluch

, et al.(eds) Hand arthroplasties, London: Martin Dunitz, pp. 249–56.

Cooney

3rd Chao

. Biomechanical analysis of static strengths in the thumb during hand function. J Bone Joint Surg Am. 1977, 59: 27–36.

Cootjans

Vanhaecke

Dezillie

Barth

Pottel

Stockmans

. Joint survival analysis and clinical outcome of total joint arthroplasties with the ARPE implant in the treatment of trapeziometacarpal osteoarthritis with a minimal follow-up of 5 years. J Hand Surg Am. 2017, 42: 630–8.

10.

Dehl

Chelli

Lippmann

Benaissa

V Rotari

Moughabghab

. Results of 115 Rubis II reverse thumb carpometacarpal joint prostheses with a mean follow-up of 10 years. J Hand Surg Eur. 2017, 42: 592–8.

11.

De la Caffinière

. Prothèse totale trapézo-métacarpienne. Rev Chir Orthop. 1973, 59: 299–308.

12.

Dell

Brushart

Smith

. Treatment of trapeziometacarpal arthritis: results of resection arthroplasty. J Hand Surg Am. 1978, 3: 243–9.

13.

De Smet

Sioen

Spaepen

. Changes in key pinch strength after excision of the trapezium and total joint arthroplasty. J Hand Surg Br. 2004, 29: 40–1.

14.

Farizon

De Lavison

Azoulay

, et al.Results with a cementless alumina-coated cup with dual mobility. A twelve-year follow-up study. Int Orthop. 1998, 22: 219–24.

15.

Gervis

. Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. J Bone Joint Surg Br. 1949, 31: 537–9.

16.

Goubau

Benis

Tchurukdichian

, et al.La prothèse Ivory à plus de cinq ans de recul. Etude multicentrique. Chir Main. 2011, 30: 55–8.

17.

Hansen

Dremstrup

Stilling

. Complications related to metal-on-metal articulation in trapeziometacarpal total joint arthroplasty may have elevated serum chrome and cobalt. J Hand Surg Eur. 2013, 38: 860–5.

18.

Hernandez-Cortés

Pajarres-Lopez

Robles-Molina

Gómez-Sánchez

Toledo-Romero

De Torres-Urrea

. Two-year outcomes of Elektra prosthesis for trapeziometacarpal osteoarthritis: a longitudinal cohort study. J Hand Surg Eur. 2011, 37: 130–6.

19.

Huang

Hollevoet

Giddins

. Thumb carpometacarpal joint total arthroplasty: a systematic review. J Hand Surg Eur. 2015, 40: 338–50.

20.

Jager

Barbary

Dap

Dautel

. Evaluation of postoperative pain and early functional results in the treatment of carpometacarpal joint arthritis. Comparative prospective study of trapeziectomy vs. MAIA® prosthesis in 74 female patients. Chir Main. 2013, 32: 55–62.

21.

Johnston

Getgood

Larson

Chojnowski

Chakrabarti

Chapman

. De la Caffinière thumb trapeziometacrpal joint arthroplasty: 16-26 year follow up. J Hand Surg Eur. 2011, 37: 621–4.

22.

Klahn

Nygaard

Gvozdenovic

Boeckstyns

MEH

. Elektra prosthesis for trapeziometacarpal osteoarthritis: a follow-up of 39 consecutive cases. J Hand Surg Eur. 2012, 37: 605–9.

23.

Ledoux

. M1/M2 ratio for radiological follow-up of trapeziometacarpal surgery. Hand Surg Rehabil. 2017, 36: 146–7.

24.

Lussiez

. Analyse radiologique de deux types de cupules trapéziennes. A propos de 50 cas. Chir Main. 2011, 30: 86–90.

25.

Martin-Ferrero

. Ten-year long-term results of total joint arthroplasties with ARPE® implant in the treatment of trapeziometacarpal osteoarthritis. J Hand Surg Eur. 2014, 39: 826–32.

26.

Semere

Viuillerme

Corcella

Forli

Moutet

. Results with the Roseland® HAC trapeziometacarpal prosthesis after more than 10 years. Chir Main. 2015, 34: 59–66.

27.

Seng

Chantelot

. La prothèse trapézométacarpienne Isis® dans la rhizarthrose: à propos de 30 cas à 30 mois de recul moyen. Chir Main. 2013, 32: 8–16.

28.

Sillwater

Memauri

Ratanshi

Islur

Amaratunga

. Radiographic interpretation of carpometacarpal arthroplasty: correlation between radiographic loosening and clinical outcome. Skeletal Radiol. 2017, 46: 1057–62.

29.

Tang

Tonkin

Boeckstyns

Hooper

. The minimum length of follow-up in hand surgery reports. J Hand Surg Eur. 2019, 44: 330–1.

30.

Tang

Giddins

. Why and how to report surgeons’ levels of expertise. J Hand Surg Eur. 2016, 41: 365–6.

31.

Teissier

Alkar

. La prothèse trapézométacarpienne Maia® dans la rhizarthrose. Revue de 100 prothèses à plus de 3 ans. Chir Main. 2011, 30: 77–82.

32.

Thillemann

Munk

Krøner

. High revision rate with the metal-on-metal carpometacarpal joint prosthesis. J Hand Surg Eur. 2016, 41: 322–7.

33.

Toledo-Romero

De Torres-Urrea

. Two-year outcomes of Elektra prosthesis for trapeziometacarpal osteoarthritis: a longitudinal cohort study. J Hand Surg Eur. 2011, 37: 130–7.