Classification

Various classification systems exist, of which the Eaton-Glickel classification of 1987 is, by far, the most commonly used ( Box 28.1 ). This classification is based on plain radiographic findings in a lateral view. However, its reliability is inconsistent and its clinical relevance doubtful. The opinion of the authors of a more recent article is that the best view of the TMC joint is provided by a combination of a “Roberts” view and a lateral view of the TMC joint ( Fig. 28.2 ). There is no clear correlation between the radiographic severity and the clinical severity of TMC osteoarthritis.

(A and B) Trapeziometacarpal (TMC) joint osteoarthritis, Eaton-Glicker stage 3. (A) Robert’s hyperpronation view. This is an anteroposterior view of thumb taken with forearm in maximal pronation and dorsum of thumb resting on the x-ray cassette. (B) True lateral view of the TMC joint and the first metacarpal. This view is taken with the forearm pronated and the radial aspect of the thumb flat on the x-ray cassette and x-ray tube angled 10 degrees from vertical. (C) Artwork illustrating bony changes in Eaton-Glicker Stage I-4 (also see Box 28.1)

(C. Courtesy Martin Langer.)

Clinical presentations and diagnosis

The patients are usually more than 50 years old. Typically, a patient with symptomatic TMC osteoarthritis presents with pain in the thenar region when gripping and pinching, and when putting weight on the outstretched hand. The patient may also complain of weakness. Most are postmenopausal women, but the condition may also affect men and younger persons. There is also a small subgroup of younger patients with a history of previous malunited Bennett’s fracture.

In early cases, there is some swelling of the joint, and crepitus is elicited when moving the joint passively under axial compression—the “grind” test ( Box 28.2 ). An alternative and similar test is the “lever” test, in which the examiner grasps the first metacarpal close to the TMC joint and moves the extended thumb radially and ulnarly. In advanced cases, the joint is subluxed, with the metacarpal base deviating in adduction so that it protrudes laterally ( Fig. 28.3 ), with the consequence that the first web space opening is limited. In longstanding cases, the metacarpophalangeal (MCP) joint hyperextends progressively to compensate for the narrowing of the web space as a consequence of increasing limitation of TMC joint movement. Point tenderness at this joint is usually present from a very early stage. The “shoulder” sign, viz. prominence of the base of the thumb with dorsal subluxation secondary to ligament laxity and pulling of the APL tendon on the metacarpal base, first metacarpal adduction, and MCP hyperextension, are commonly seen in advanced cases. In the elderly, longstanding disease may result in ankylosis of the joint, which may then become pain-free.

Severe trapeziometacarpal joint osteoarthritis with adduction contracture of the first metacarpal.

The differential diagnoses include stenosing tenosynovitis of the flexor pollicis longus (FPL) tendon (with or without triggering), de Quervain tenosynovitis, and scaphotrapezotrapezoideal (STT) osteoarthritis. All of these conditions may coexist with TMC osteoarthritis and must be identified and treated accordingly.

Trapeziectomy: Evidence-based surgical choice

Many series of results after trapeziectomy, with or without TI or LRTI, have been published, and these procedures are generally considered to give excellent or good results in at least 85% of cases. Data extracted from eight randomized studies show no or little difference in outcomes with or without TI or LRTI. Davis et al published preliminary results in a randomized study that, later, was extended in both the number of patients and the follow-up time. Davis et al ^, compared trapeziectomy alone with trapeziectomy plus TI or trapeziectomy plus LRTI. The found that at 3 months and 1 year follow-up the results of the three procedures were indistinguishable in terms of pain relief, hand function and thumb strength. No benefit of tendon interposition or ligament reconstruction was seen at longer follow-up. Similar findings are presented by Gangopadhyay et al. Similarly, Davis and Pace and Salem et al reported that pain, function, and strength improved significantly after both trapeziectomy and trapeziectomy plus TI or LRTI, and the results seem long-lasting, although the thumb key and tip pinch strengths declined from 1 year to 5 years. This was also the case for the contralateral thumbs and was probably due to age-related strength deterioration.

The procedures consistently resulted in shortening of the thumb, but the amount of shortening did not correlate with a reduction of any subjective or objective measure of outcome and did not reduce total hand-span. The dogma that trapeziectomy causes loss of strength because of thumb shortening is contradicted by these and other clinical studies. A randomized study by Field and Buchanan compared trapeziectomy with a Burton-Pellegrini-like LRTI. In this study, sham incisions were made in the “simple” trapeziectomy group to mimic those used for harvesting the FCR tendon for the sake of blinding the trial. Sixty-four of the 65 patients said they would undergo the procedure again. The one patient who would not was a patient who developed chronic regional pain syndrome (CRPS) Type 1. There was a greater increase of range of movement in the trapeziectomy alone group, but no difference in grip or pinch strength. Corain et al compared trapeziectomy with joint distraction by means of a Kirschner (K) wire between the first and second metacarpal for 4 weeks, with trapeziectomy plus LRTI (with APL tendon) in a randomized study. After 7 years, there was no difference between the two techniques in terms of objective measurements and DASH score and visual analog scale (VAS) score for pain. On lateral unloaded radiographs, the height of the space between the base of the thumb metacarpal and the scaphoid was similar in the two groups.

The surgical complications noted in a systematic review are listed in Table 28.1 . They include sensory nerve symptoms affecting the superficial branch of the radial nerve and the palmar cutaneous branch of the median nerve, superficial wound infections, and pain-related problems such as persistent pain and complex regional pain syndrome. Tendon-related problems included tendon ruptures (extensor pollicis brevis (EPB), FCR) ( Fig. 28.4 ), tendonitis (FCR and de Quervains), and tendon “pulling sensation” (FCR, FPL). One trapeziectomy underwent further surgery with re-resection. The lead author (MB) considers transient paraesthesia in the territory of the superficial radial nerve as the most common complication after “simple” trapeziectomy, whereas tendon problems are virtually nonexistent in his practice, as no tendon adjunct is used and the pulley of the first extensor compartment is routinely released. Instability of the arthroplasty and CRPS occurs very rarely.

A ruptured FCR tendon after Welby’s suspension arthroplasty. (A) Painful swelling after trapeziectomy plus tendon interposition (Weilby procedure through an anterior approach). (B) Operative view: the blue arrow indicates the Weilby knot: a strip of the FCR tendon wrapped around the APL tendon. The FCR tendon has ruptured.

Evidence reported on “simple trapeziectomy” versus trapeziectomy with tendon adjunct

In the controlled randomized studies by Davis et al ^, as stated above, they compared trapeziectomy alone with trapeziectomy plus TI or trapeziectomy plus LRTI. This study showed no benefit of tendon interposition or ligament reconstruction at 3 months, 1 year, and also at longer follow-up. Other randomized studies and at least 11 systematic reviews have reported similarly.

Ritchie and Belcher compared the anterior and posterior approaches for trapeziectomy and found that the anterior approach group had significantly better subjective power, motion, and satisfaction, and better objective key pinch and scar tenderness. A review by Li et al, based on two systematic reviews, ^, three randomized studies, ^{, ,} and one comparative study, concluded that there is no difference in key and tip pinch strength, grip strength, postoperative adverse events, pain score, DASH score, and between trapeziectomy and trapeziectomy with ligament reconstruction and tendon interposition.

Ceruso et al published a systematic review in the Federation of European Societies for Surgery of the Hand (FESSH) 2017 instructional course book, available online at https://fessh.com/library/2017_icbook.pdf . The procedures were simple trapeziectomy with or without hematoma distraction, trapeziectomy with ligament reconstruction, tendon interposition or both, tight-rope suspension, and arthrodesis with or without an associated bone graft. Comparing trapeziectomy with LRTI to simple trapeziectomy, the analysis was not able to ascertain any difference in terms of strength, patient’s global assessment, or range of motion (ROM). Trapeziectomy achieved a clinically irrelevant, even if statistically significant, improvement in pain score. Overall trapeziectomy alone had a lower complication incidence than trapeziectomy with LRTI or TI ( Table 28.1 ).

Knightly and Sullivan concluded that “low-moderate quality evidence suggests that trapeziectomy plus LRTI yields better range of movement (palmar abduction) when compared with trapeziectomy alone” but found no other differences. Saheb et al summarized 15 randomized studies, of which three compared trapeziectomy with trapeziectomy plus LRTI or TI. ^{, ,} Besides comparing trapeziectomy with trapeziectomy plus LRTI/TI, some of these papers also considered denervation, prosthetic replacement, and arthrodesis. The authors concluded that no procedure was superior to any other in terms of pain, physical function, patient global assessment, range of motion, or strength. Other studies have reached similar conclusions. ^{, , ,} De Smet et al, in a nonrandomized study, compared patients with trapeziectomy alone with patients who underwent trapeziectomy with an LRTI procedure using the entire FRC tendon passed through a hole in the base of the metacarpal. They found an increased height of the joint space and stronger key pinch grip after LRTI but, otherwise, no differences between the two procedures.

Saab and Chick reviewed all of the studies that reported on trapeziectomy in the context of osteoarthritis, regardless of the additional procedures. They also included a nonrandomized study comparing “simple” trapeziectomy with trapeziectomy plus TI (with APL) and concluded that it is unnecessary to perform a ligamentoplasty when a trapeziectomy is indicated. Furthermore, trapeziectomy plus LRTI has more iatrogenic injury, more short-term postoperative complications, and higher surgical costs. They consider if there is no special strength requirement, they recommend simple trapeziectomy. Rosales et al included seven randomized trials in a systematic review ^{, ,} and concluded that “The outcomes of trapeziectomy with LRTI are not superior to trapeziectomy or trapeziectomy with tendon interposition in the treatment of OA of the thumb TMC joint.”

Surgical technique of “simple” trapeziectomy

It is worthy of note that “simple” trapeziectomy does not necessarily mean simply excising the trapezium. Some stabilization of the neo-arthrosis is necessary to prevent subluxation of the first metacarpal. Davis et al describe their technique as:

After ensuring that the trapezium had been completely excised, the palmar and dorsal periosteal/capsular flaps were sutured together. If the dorsal and palmar flaps were incomplete due to the presence of a thin periosteum, great care was taken to ensure that the capsular remnants attached to the base of the thumb metacarpal were sutured to those attached to the distal pole of the scaphoid: it was felt that this would prevent dorsal subluxation of the pseudarthrosis during the postoperative period.

The method of the lead author (MB) is similar ( Fig. 28.5 ). I use it in the vast majority of my patients with Eaton-Glickel stage 3 or 4 disease and in the very few patients with stage 2 disease in whom nonsurgical treatment has failed. The operation is done under a lateral infraclavicular block or general anesthesia. Wide-awake local anesthesia without tourniquet is also possible, although I find it slightly more difficult to identify the subcutaneous nerves with this anesthetic technique.

A method to stabilize the neo-arthrosis. (A) After trapeziectomy, the dosral capsular flap is raised. (B and C) Interposing and suturing a dorsal capsular flap. The flap need not reach the bottom of the cavity.

(Courtesy Michel Boecktyns .)

The approach is dorsal, through a curved incision from the base of the first metacarpal to the anatomical snuff-box. The skin is lifted with a retractor proximally, the dorsal branches of the radial nerve are identified, and the first extensor compartment is released, paying attention to the possibility of the EPB tendon running through a separate canal. The nerve branches are further visualized distally and carefully retracted: This is best done by pulling on the tendons (APL and EPB) rather than on the nerves themselves to reduce the risk of postoperative paraesthesia. Next, the radial artery is identified and its small branches running distally toward the TMC joint are cauterized. The capsule is opened creating a distally based U-shaped flap that extends proximally to the scaphotrapezial joint ( Fig. 28.5 A).

As much as possible of the capsule and ligaments are released from the trapezium, and the joint between the trapezium and the trapezoid is identified. This joint is usually very tight and may be difficult to identify. Then, the trapezium is split in two parts with an osteotome or a rongeur and removed in small bites, taking care not to injure the FCR tendon lying in the bottom of the cavity. Alternatively, a corkscrew can be used to pull the bone out of the surgical cavity while gradually releasing the anterior ligament attachments. Now it is possible to examine the joint between the scaphoid and the trapezoid. If osteoarthritis is present, approximately 2 to 3 mm of the trapezoid is resected.

The capsule is closed while the assistant holds the metacarpal in abduction. At this point, it is important to interpose the proximal part of the flap in the trapezial space ( Fig. 28.5 B). No attempt is made to suture the free end of the flap to the soft tissue at the bottom of the cavity, nor is any tissue interposed between the scaphoid and the trapezoid. Stability is checked by pushing the metacarpal toward the scaphoid, longitudinally and sidewards. If there is any bony contact, the capsular closure is adjusted with additional stitches ( Fig. 28.5 C). When stability and mobility are satisfactory, hemostasis is checked, optionally after tourniquet release, and the skin is closed with stitches or a running suture.

Postoperative care after “simple” trapeziectomy

Padding and a dorsoradial cast are applied, leaving the wrist and thumb interphalangeal joints free. Care must be taken not to position the metacarpal in adduction. Most patients will experience quite a lot of pain during the first week or two, needing painkillers. It is important that the patient keep their hand elevated and perform active exercises of the fingers for as long as there is a tendency toward swelling. The hand may be used for light activities of daily living. The cast and stitches are removed after 3 weeks and the use of the hand increased within reasonable pain limits. Usually, no further splinting is necessary. In the experience of the lead author (MB), hand therapy is required in half of the patients. Efforts to flatten out the hand and place the tip of the thumb near the base of the little finger are encouraged. Most patients return to work and sports at 3 months, but the need for pain relief may continue for 6 months.

In the experience of the lead author (MB), this surgery gives predictable outcomes and is sufficient for the majority of patients. Very few of the lead author’s patients need LRTI, which is used mainly in patients with a failed, unstable trapeziectomy or patients with pronounced hyperlaxity. The lead author’s tips for selection of surgical options are given in Boxes 28.4 and 28.5 .

We have noted that despite there is substantial evidence to support that “simple” trapeziectomy and trapeziectomy with TI or LRTI yield comparable results, both in the short and in the longer term, many surgeons have still prefered the more complex procedures using tendon adjuncts. Two surveys were undertaken in the United States a decade ago. ^, According to the first survey, less than 3% of respondents in the survey performed trapeziectomy alone, and according to the second survey, 62% perform trapeziectomy with LRTI for “common Eaton stage III arthritis.” Similarly, a report from the Swedish national health care quality registry for hand surgery, HAKIR, revealed that no more than 59 out of 650 (9%) procedures for TMC osteoarthritis performed by specialized hand surgery units were “simple” trapeziectomies. It seems that tradition and habits play an important role in the decision-making of surgical treatment of TMC osteoarthritis.

Revision and salvage of failed trapeziectomy

Instability of the first metacarpal after trapeziectomy, with or without TI/LRTI, may cause significant symptoms. The metacarpal tends to adduct, and its base may sublux and impinge on the scaphoid if the space between the bones collapses ( Fig. 28.7 ). Revision surgery may be necessary for these patients. To date, there are no clear guidelines available on how to treat these patients. The most frequently used techniques are stabilization of the existing interposition with an autologous tendon ( Fig. 28.8 ), suspension with a Mini TightRope® (Arthrex, Naples, FL, USA) ( Fig. 28.9 ), and arthrodesis between the first and second metacarpal. Most surgeons would wait at least 1 year before performing revision surgery. The results and patient satisfaction may be unpredictable.

Impingement of the first metacarpal base on the scaphoid after suspension arthroplasty (Weilby) 1 year ago in a 64-year-old woman. Clinical examination reveals painful impingement and instability. VAS of pain: 70/100 during activity, 30/100 at rest. QDASH: 78.

Salvage of a failed trapeziectomy with FCR tendon interposition. (A) Dislocated FCR-APL knot (retracted) after a failed Weilby arthroplasty. (B) Harvest of an ECRL slip. (C) Passing the ECRL slip through a burr hole in the base of the first metacarpal. (D) The ECRB slip is wound around itself between the first and second metacarpals and secured with stitches, thus reconstructing the first intermetacarpal ligament. (E and F) The ECRL slip is further used to draw back the dislocated FCR/APL knot into the trapezial space. APL, Abductor pollicis longus tendon; ECRL, extensor carpi radialis longus tendon; FCR, flexor carpi radialis tendon.

Wrist radiograph of a 57-year-old man treated with a mini tightrope after previous trapeziectomy and tendon graft interposition. The aim was to correct the proximal migration of the first metacarpal, but in this patient the metacarpal height could not be completely restored.

Implants and surgical outcomes

Implants.

More than 5 decades ago, Alfred Swanson introduced replacement of the trapezium with a silicone spacer. ^, However, subluxation, luxation, and fragmentation of the implant have been recurrent complications, and despite modifications and the introduction of other similar implants, their use has been abandoned. At about the same time, de la Caffinière introduced a cemented total joint prosthesis converting the TMC saddle joint into a ball-and-socket joint with a cup in the trapezium and a stem in the first metacarpal ( Fig. 28.9 ). Underlying the concept of prosthetic TMC joint replacement was the aim of preserving the length of the thumb, believing that this would improve strength. Results seemed promising, and at least 20 ball-and-socket designs followed ^, and became popular, especially in continental Europe. A recent French survey showed that 23 of 36 respondents (64%) used prosthetic replacement for the surgical treatment of TMC osteoarthritis. According to the Belgian National Health Insurance and Social Security, about 80% of the procedures performed in Belgium are prosthetic replacements. However, implant loosening, especially of the cups, has been a concern. Dislocation of the prosthetic head is also a risk in nonconstrained ball-and-socket designs.

A new concept of “dual-mobility” prostheses seems to reduce this risk, but the long-term results of dual-mobility articulations are still unclear. TMC replacement remains a challenging surgery with a difficult learning curve. The most common implants are cementless metal-on-polyethylene ball-and-socket implants, including the Moovis (Stryker, Pusignan, France), the Ivory (Memometal, Stryker European Holdings I, LLC, Amsterdam, Netherlands), the Arpe (Zimmer-Biomet Warsaw, IN, USA), the Maia (Groupe Lépine, Genay, France), and the Touch (KeriMedical, Les Acacias, Switzerland). The Ivory and Arpe protheses have recently been withdrawn from the market. Basically, there are two cup designs: a conical and a spherical cup ( Figs. 28.10 and 28.11 ). There seems to be no difference in the long-term clinical results in terms of aseptic loosening and the need for revision between these two designs. The long-term survival has much improved, and the 10-year cumulated survival rates range from 85% to 95% ( Table 28.2 ). ^, The long-term survival of the Moovis, Touch, and Maia dual-mobility prostheses have not yet been published.

Explanted Caffinière ball-and-socket prosthesis.

Cup designs (from left to right): conical with single mobility, conical with double mobility, spherical with single mobility.

(From Hansen TB. Joint replacement for trapeziometacarpal osteoarthritis: implants and outcomes. J Hand Surg Eur Vol . 2021;46:115-119.)

TABLE 28.2

Long-Term Cumulated Survival of Currently Used Implants

Authors (year)	Implant	Number of Cases	Lost to Follow-Up	Follow-Up	Cumulated Implant Survival at 10 Years	Cumulated Implant Survival at 15 Years
Dumartinet-Gibaud et al (2020)	Arpe	80	27	138 months (range, 120–280)	85%	80%
De Smet et al (2020)	Arpe	63	13	10 years (range, 3–16)	80%	75%
Martin-Ferrero et al (2019)	Arpe	228	11	10 years	92%	-NA
Tchurukdichian et al (2020)	Ivory	110	7	10 years	95%	-NA
Vissers et al (2019)	Ivory	32	7	130 months (range, 120–142)	85%	-NA
Chiche et al (2022)	Maia	293	109	12 years (range, 17 days to 140 months)	88%	NA
Semere et al (2015)	Roseland	75	24	12.5 years (mean, SD 1.8)	91%*	-NA
Dehl et al (2017)	Rubis 2	253	138	10 years (range, 6–17)	89%	-NA

 

Outcomes.

We summarized the clinical long-term results of the seven above-mentioned articles, selecting five relevant and frequently reported parameters ( Table 28.3 ). This table is not meant for comparison of the results of the different implants but merely to give a general idea of the long-term results of total TMC joint replacement that have been reported. It appears that effective pain reduction was achieved. Quick Disabilities of Arm, Shoulder, and Hand (Q-DASH) scores at follow-up were in the range of <14 to 30, indicating no, or moderate, functional impairment. These scores are dependent on other upper-extremity-related conditions and are mainly useful if they can be compared with preoperative values. Thumb opposition was generally excellent. Key pinch strength improved in those studies that mentioned preoperative values. Patient satisfaction was 90% or higher.

TABLE 28.3

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Modern clinical settings and approaches Distal radius fractures Other carpal fractures and carpal disorders Fractures of the elbow and forearm Dupuytren disease Enhancing hand control in tetraplegia

Stay updated, free articles. Join our Telegram channel

Join