Journal Logo


Thumb Carpal Metacarpal Arthritis

Van Heest, Ann E. MD; Kallemeier, Patricia MD

Author Information
Journal of the American Academy of Orthopaedic Surgeons: March 2008 - Volume 16 - Issue 3 - p 140-151
  • Free


The thumb carpometacarpal (CMC) joint is the most common site of surgical reconstruction for osteoarthritis (OA) in the upper extremity. In persons older than age 75 years, the prevalence of radiographic CMC degeneration is 25% in men and 40% in women.1 A clear understanding of the anatomy and physiology of the joint is necessary to make well-informed decisions regarding diagnosis and treatment of thumb CMC arthritis.


Bony Anatomy

The thumb CMC joint is semiconstrained and relatively incongruent. In 1854, Fick2 coined the term “saddle joint” to describe this complex association3 (Figure 1). The CMC is made up of two saddles. The axes of the opposing saddles are perpendicular to each other, such that the distal saddle faces proximal and the proximal saddle is rotated 90° in relation to the distal upside-down saddle. Biomechanically, the CMC is referred to as a universal joint. The anatomy of the joint allows thumb motion in extension, flexion, adduction, and abduction (Figure 2). These motions can be combined to form the complex movements of opposition, retropulsion, palmar abduction, radial abduction, palmar adduction, and radial adduction.4 The thumb CMC joint is more congruent along the radial-ulnar axis than along the dorsal-volar axis.5 This bony architecture allows increased range of motion for opposition. The prevalence of ligamentous stability in the thumb CMC joint allows greater mobility than were the joint stability provided by bone.

Figure 1
Figure 1:
The thumb carpometacarpal joint is known as the “saddle joint” because its shape and configuration are similar to a saddle. (Adapted with permission from Kuczynski K: The thumb and the saddle. Hand 1975;7:120-122.)
Figure 2
Figure 2:
The axes of the saddle configuration of the thumb carpometacarpal joint. Minimal bone congruity allows for thumb carpometacarpal motion, which includes extension, flexion, adduction, and abduction. These four movements can be combined to form the complex movements of opposition and retropulsion, palmar abduction, radial abduction, palmar adduction, and radial adduction. (Adapted with permission from Ateshian GA, Rosenwasser MP, Mow VC: Curvature characteristics and congruence of the thumb carpometacarpal joint: Differences between female and male joints. J Biomech 1992;25:591-607.)

Ligamentous Stability

The ligamentous anatomy of the thumb CMC joint is important for stability. Several ligaments play a major role in the stabilization of the thumb CMC joint (Figure 3). The palmar ligament, also known as the oblique or beak ligament, acts as a static restraint by virtue of its intracapsular location. It originates from the palmar tubercle of the trapezium and inserts on the articular margin of the ulnar side of the metacarpal base.6 The palmar ligament resists abduction, extension, and pronation forces. Doerschuk et al7 showed in a cadaveric study that the degree of palmar (anterior oblique) ligament degeneration correlates with the stage of OA.

Figure 3 A,
Figure 3 A,:
Dorsal view of the thumb right trapeziometacarpal joint illustrating the posterior oblique ligament (POL), dorsoradial ligament (DRL), abductor pollicis longus (APL), first intermetacarpal ligament (IML), and extensor carporadialis (ECRL) tendon. B, Palmar view of the thumb right trapeziometacarpal joint illustrating the anterior oblique ligament (AOL), ulnar collateral ligament (UCL), first IML, APL tendon, transverse carpal ligament (TCL), and flexor carpi radialis (FCR) tendon. MI = 1st metacarpal, MII = 2nd metacarpal, MIII = 3rd metacarpal. (Reproduced with permission from Imaeda T, An KN, Cooney WP III: Functional anatomy and biomechanics of the thumb. Hand Clin 1992;8:9-15.)

The dorsoradial ligament is the primary stabilizer of dorsal and radial translation of the thumb metacarpal on the trapezium8 and is the primary restraint to dorsal dislocation.9 The dorsoradial ligament also has been shown to be the strongest and thickest ligament of the CMC joint.6,10

The intermetacarpal ligament attaches from the radial base of the 2nd metacarpal to the ulnar aspect of the base of the 1st metacarpal and constrains radial translation of the base of the 1st metacarpal. The dorsoradial and posterior oblique ligaments provide secondary stability.

Muscular Stability

Nine muscles provide dynamic stabilization of the thumb CMC joint. The volar muscles include the three thenar muscles (abductor pollicis brevis [APB], flexor pollicis brevis, opponens pollicis), the flexor pollicis longus, and the adductor pollicis. The dorsal muscles include the first compartment muscles (abductor pollicis longus [APL], extensor pollicis brevis [EPB]), the extensor pollicis longus, and the first dorsal interosseous muscle. The coordination of these muscles creates a balance of stability, allowing positioning of the thumb to provide the platform for thumb pinch activities.

Pathophysiology of Disease

The etiology of CMC joint arthritis is multifaceted and includes both intrinsic and posttraumatic causes. Intrinsic causes may include ligamentous hypermobility, ligament laxity related to hormonal influences, sex differences, and biochemical differences. Jónsson et al11 postulated that hypermobility is a cause of CMC arthritis. In their study of 50 patients, they correlated thumb CMC joint arthritis with joint hypermobility (passive extension of the 5th finger of ≥90°). Patients with Ehler-Danlos syndrome have been shown to have a higher incidence of CMC joint subluxation and exhibit radiographic degenerative changes at a mean age of 15 years.12 Compared with men, women exhibit a higher amount of joint laxity (Figure 4) and a higher incidence of thumb CMC joint arthritis. This difference has been attributed to the hormones prolactin, relaxin, and estrogen. The finding of increased joint laxity may be confounded by the subtle sex differences in trapezium geometry.

Figure 4
Figure 4:
Anteroposterior thumb radiograph demonstrating carpometacarpal joint subluxation in a female athlete with joint laxity and increased oblique orientation of the trapezium.

One group reported that CMC joints are less congruent in females than in males and that the trapezial surface is smaller in women than in men.5 In cadaveric studies, North and Rutledge13 found that the trapezial surface transformed from a saddle shape to a semicylindrical shape in patients with advanced arthritis. Furthermore, both female and male specimens with early degenerative changes exhibited a flatter trapezial surface, which appears to be correlated with increased arthritic changes.

In their Finnish population study, Haara et al14 established a direct correlation between increased body mass index (BMI) and increased prevalence of CMC arthritis. The authors offered two proposed mechanisms for increased incidence of CMC arthritis in obese patients. First, even though the thumb CMC is a non-weight bearing joint, patients with a higher BMI may have increased mechanical loading across the joint, causing increased wear. Second, patients with a higher BMI may have an altered biochemical environment in the joint, such as a change in circulating lipid levels, insulin-like growth factor, and sex hormones. These hormonal differences may provide local biochemical changes that promote joint degeneration.

In addition to intrinsic causes of CMC arthritis, a history of trauma predisposes to arthritic progression. A Bennett fracture resulting in joint incongruity can lead to arthritis. Additionally, injury to the anterior oblique ligament may cause altered joint mechanics and may predispose a patient to thumb CMC arthritis. In the presence of an incompetent anterior oblique ligament, abnormal translation of the thumb metacarpal on the trapezium occurs (Figure 4). The shear forces wear on the volar compartment of the CMC joint near the volar oblique ligament insertion. In late-stage arthrosis, articular wear begins on the radial quadrant of the metacarpal and progresses to the volar quadrants, while on the trapezium, wear starts on the dorsoradial surface and progresses to the volar quadrants.15 The mechanism for wear occurs mainly in a shear mechanism, secondary to ligament laxity. In anatomic studies, wear has been shown to occur in a 3:1 ratio of trapezial to metacarpal degeneration.

Additional evidence points to metacarpophalangeal (MCP) joint subluxation as a cause of CMC joint wear. In a cadaveric study, Moulton et al16 found that MCP joint flexion effectively unloaded the most palmar surfaces of the CMC joint, and MCP joint hyperextension loaded the most palmar surface of the CMC joint. This is problematic because the volar compartment shows the earliest signs of arthritic changes. This suggests that patients with hyperextension of the MCP and symptomatic CMC joint arthritis might benefit from splinting or surgical stabilization of the MCP joint in a flexed position. These treatments would effectively unload the volar compartment of the CMC joint.16


Symptoms of thumb CMC arthritis range from insignificant, occasional aching to severe pain with weakness and disability. Patients often describe the pain as a diffuse ache localized to the thumb abductor and thenar musculature area. Physical examination tests for CMC arthritis include tenderness to palpation over the dorsal or dorsoradial capsule of the CMC joint. Findings include localized swelling and warmth at the base of the thumb. The grind test includes axial compression of the thumb CMC, which produces crepitus and pain (Figure 5).

Figure 5
Figure 5:
The grind test for carpometacarpal arthritis. This test consists of axial compression and rotation of the thumb carpometacarpal joint. (Adapted with permission from Acquired deformities, in American Society for Surgery of the Hand: The Hand: Examination and Diagnosis, ed 3. New York, NY: Churchill-Livingstone, 1990, p 85.)

Secondary deformity related to thumb CMC arthritis occurs over time. As the patient persistently avoids painful thumb abduction, adduction deformity occurs within the first web space contracture. As the CMC joint becomes stiff and adducted, the thumb MCP joint may develop a hyperextension deformity to compensate for the loss of motion. The compromised thumb metacarpal cannot abduct adequately to grasp a sizable object, leading to MCP joint hyperextension with progressive attenuation of the volar plate. Late-stage secondary deformity includes a zigzag collapse pattern (Figure 6).

Figure 6
Figure 6:
As the carpometacarpal joint becomes stiff and adducted in the patient with osteoarthritis, the thumb metacarpophalangeal joint may develop a hyperextension deformity to compensate for the loss of motion, leading to a zigzag collapse pattern.

Other etiologies of pain at the base of the thumb include de Quervain tenosynovitis of the first dorsal compartment, flexor carpi radialis (FCR) tendinitis, and carpal tunnel syndrome. The patient with scaphoid pathology (eg, fracture, nonunion, osteonecrosis) may present with pain at the base of the thumb. Additionally, arthritis of the thumb MCP, radiocarpal joints, and scaphotrapeziotrapezoid (STT) joint may cause pain along the thumb ray. Differential injections into the thumb STT, CMC, and MCP joints can be helpful in determining the precise etiology and location of the pain.

Radiographic Classification

The most widely used classification system for thumb CMC arthritis was described by Eaton in 1973.17 This classification was later modified to include scaphotrapezial joint involvement.18 Radiographically, the patient with Eaton stage I thumb CMC arthritis appears to have normal articular cartilage. Often, there is a widened joint space, which may indicate joint effusion or synovitis (Figure 7). In Eaton stage II disease, slight narrowing can be seen at the thumb CMC joint space (<2 mm), along with maintained joint contours, minimal sclerosis, and joint debris. Lateral dorsal subluxation of the thumb metacarpal on the trapezium may be present. In Eaton stage III disease, radiographic evaluation indicates significant narrowing of the thumb CMC joint and arthritis with sclerosis, cyst formation, and osteophytes longer than 2mm. Joint subluxation is usually present. In Eaton stage IV OA, the patient presents with significant thumb CMC joint space deterioration with concomitant scaphotrapezial joint degeneration.

Figure 7
Figure 7:
Bett views of the thumb carpometacarpal (CMC) joint that also provide a view of the scaphotrapezial joint. A, Eaton stage I thumb CMC arthritis, with normal articular cartilage and a widened joint space. B, Eaton stage II thumb CMC arthritis. Note the slight narrowing at the thumb CMC joint space (<2 mm) along with maintained joint contours, minimal sclerosis, and joint debris. Lateral dorsal subluxation of the thumb metacarpal on the trapezium is seen on this radiograph, but is variably present in stage II disease. C, Eaton stage III thumb CMC arthritis showing significant narrowing of the thumb CMC joint arthritis with sclerosis, cyst formation, and the presence of osteophytes (>2 mm in length). Joint subluxation is usually present and is evident on this radiograph. D, Eaton stage IV thumb CMC osteoarthritis showing significant thumb CMC joint space deterioration with concomitant scaphotrapezial joint degeneration.

Several authors have studied the reliability of the Eaton classification for thumb CMC joint arthritis. Comparing observations by three hand surgeons and three orthopaedic residents, Kubik and Lubahn19 reported intrarater reliability of 0.657 and interrater reliability of 0.529. The hand surgeons had an interrater reliability of 0.601, and the chief residents had an interrater reliability of 0.487. Most chief residents had higher intrarater scores than did the junior residents, suggesting that the use of the system may improve with experience. Dela Rosa et al20 studied the use of a different radiographic view, called the Bett or Gedda view, which shows all four articulations of the trapezium without overlap from surrounding bones (ie, a view of the trapezial-1st metacarpal, trapezial-2nd metacarpal, trapezial-trapezoid, scaphotrapezial joint spaces). Forty sets of radiographs were evaluated by six hand surgeons in two sessions with three subgroups. The posteroanterior, lateral, and Bett views provided the best kappa intraobserver reliability (0.61). Use of these three radiographic views has increased the reliability of the Eaton classification. Despite its prevalent use in the literature, radiographic staging of CMC arthritis does not predict severity of symptoms in the clinical setting. In fact, the radiographic staging of the disease has been shown to underestimate the extent of degenerative pathology visualized at the time of surgery.21


Treatment of thumb CMC arthritis should be based on the severity of the clinical symptoms and not simply on the radiographic stage. Treatment goals include improved pain and disability as well as prevention of secondary deformity.


Nonsurgical treatment of CMC arthritis includes hand therapy (stretching and strengthening exercises), splinting, and injection. The goals of hand therapy are to increase range of motion of the thumb CMC, prevent secondary zigzag collapse deformity, and strengthen the thumb muscles to aid in providing joint stability. Stretching of the first web space helps prevent adduction contracture and subsequent MCP hyperextension deformity. Strengthening the first dorsal interosseous ligament can help provide medial joint stability in the presence of an incompetent anterior oblique ligament.

Splinting with the thumb in abduction has been shown to decrease pain. Options range from a prefabricated hand-based thumb splint to a custom forearm-based splint. Splinting should reduce subluxation at the CMC joint, and it counters adduction contracture. Some devices accommodate splinting of the MCP joint in flexion to reduce the loading across the CMC joint. The use of splints has been shown to reduce symptoms by an average of 55% to 60% within the first 6 months.22 In their 6-month retrospective study of 130 patients, Swigart et al22 reported a decrease in pain of 76% in patients with stage I and II disease following application of a support splint. They concluded that splinting was well tolerated and was an effective nonsurgical treatment option.

Hand therapists typically can provide assistive devices and modify household tools to help decrease the load across the thumb CMC joint. Berggren et al23 reported on 33 patients who were treated nonsurgically with adaptive devices, splints, and therapy while awaiting CMC joint surgery. After 7 months of nonsurgical management, 70% of the patients declined surgical treatment. During the subsequent 7 years, only 2 of the 19 patients who were still living required surgical treatment.

Corticosteroids have been considered the preferred injection therapy; however, evidence-based benefits of this treatment have been mixed. In a double-blind, randomized controlled trial, Meenagh et al24 treated 40 patients with either triamcinolone or saline intra-articular joint injection. At 24 weeks, the patients who received steroid injection showed no improvement in hand function and no difference in the visual analog pain scales versus patients injected with saline. No clinical benefit of corticosteroid injection was found in terms of joint stiffness or tenderness.

Other studies have shown corticosteroids to be effective when combined with splinting. Day et al25 found that steroids in conjunction with splinting was successful at an average follow-up of 18 months. Five of 6 patients with Eaton stage I disease had an average of 23 months of pain relief, and 6 of 17 patients with Eaton stage II and III disease had relief of symptoms at 18 months. However, only one of seven patients with Eaton stage IV disease had any pain relief. These studies suggest that steroid injection plus splinting may have a role in the treatment of patients with earlier stages of CMC arthritis.

Studies evaluating the use of viscosupplementation for the management of thumb CMC arthritis indicate that a series of viscosupplementation injections may be more effective than one corticosteroid injection.26,27 Long-term results of viscosupplement injection with regard to preventing surgery have not been reported, and viscosupplement injections for thumb arthritis are not FDA approved, but are experimental only.


Surgical treatment of CMC joint arthritis is selected based on the radiographic stage of disease, patient symptoms, patient age, and occupation. Joint preservation procedures are effective in the earlier stages of disease, whereas fusion or arthroplasty is more effective in the later stages of the disease process.

Volar Ligament Reconstruction

Volar ligament reconstruction is indicated in the patient who presents with prearthritic synovitis caused by CMC joint laxity, including Ehlers-Danlos syndrome. This syndrome is characterized radiographically as Eaton stage I or II disease and clinically with pain and weakness related to joint subluxation. The candidate for volar ligament reconstruction typically has had no success with nonsurgical treatment, including splinting techniques aimed at stabilizing the thumb CMC joint. Preoperative assessment includes ruling out other causes of pain. Ligament reconstruction is most effective when preoperative stress radiographs show radial subluxation of the thumb metacarpal on the trapezium. This finding correlates with physical examination findings of symptomatic joint subluxation.

One of the many surgical approaches for volar ligament reconstruction involves using a tendon autograft to render the CMC joint stable through a full arc of motion. Brunelli et al28 described the use of the APL tendon to recreate stability of the anterior oblique ligament. In this technique, a hole is drilled through the metacarpal base in a dorsal to volar direction, and the tendon (either the APL or one half of the FCR tendon) is passed from the ulnar to radial direction. The Littler-Eaton procedure, which reconstructs the intermetacarpal, radial, and anterior oblique ligaments, makes use of the FCR tendon, which is left attached to its insertion at the base of the index metacarpal. The FCR tendon is passed through the thumb metacarpal in a volar-ulnar to dorsal-radial direction and around the APL to simulate the dorsoradial ligament, then reattached back to itself to reproduce the anterior oblique ligament (Figure 8). Alternately, the FCR tendon can be looped around the APL tendon and then back beneath the intact FCR distal attachment. Before suturing the tendon graft in place, the thumb CMC joint is reduced in a position of palmar abduction and extension to allow a full arc of motion.

Figure 8
Figure 8:
Volar ligament reconstruction in a patient with thumb carpometacarpal joint arthritis. One half of the flexor carpi radialis (FCR) tendon is looped around the abductor pollicis longus tendon and then back beneath the intact FCR distal attachment. I = 1st metacarpal, II = 2nd metacarpal, III = 3rd metacarpal. (Adapted with permission from Eaton RG, Littler JW: Ligament reconstruction for the painful thumb carpometacarpal joint. J Bone Joint Surg Am 1973;55:1655-1666.)

Eaton et al29 performed volar ligament reconstruction using the FCR. Patients at all stages of disease were treated surgically and followed for an average of 7 years. Results were good and excellent in all patients with stage I disease and in 91% of patients with stage II disease. However, only 80% of patients with stage III disease and 66% of patients with stage IV disease had good and excellent results. Pain relief was correlated with good and excellent results. In 13 of 19 patients with stage I and II disease, pinch strength was greater than or equal to that on the opposite side. The authors emphasized the importance of intraoperative assessment of the joint surfaces to determine whether more advanced arthritis is present than was assessed on preoperative radiographs.

In another study, Freedman et al30 limited their indications for volar ligament reconstruction to stage I and stage II disease. At an average follow-up of 5.2 years, all patients with stage I disease had good or excellent results, with complete pain relief. Patients with stage II disease had 82% good to excellent results, with 70% pain relief. Follow-up radiographs showed no further degeneration at the CMC joint. At 15-year follow-up, all patients with stage I thumb CMC arthritis who were treated with volar oblique ligament reconstruction had high satisfaction rates; 65% of patients showed no further progression of radiographic CMC arthritis.

Metacarpal Extension Osteotomy

Correcting the alignment of the thumb CMC joint also can be effective in early CMC disease. First metacarpal osteotomy is done to unload the volar arthritic compartment of the CMC joint by changing the joint mechanics. Indications for metacarpal extension osteotomy include CMC joint pain in patients with Eaton stage I or II disease. Exclusion criteria include patients with more advanced stages of disease (assessed at the time of surgery). When cartilage wear extends beyond the volar one third of the compartment, the extension osteotomy will transfer little, if any, load to the dorsal trapezium metacarpal articular surface.31 Patients with hypermobility or fixed subluxation of the CMC joint are better treated with ligament reconstruction, soft-tissue arthroplasty, or arthrodesis. Osteotomy is also contraindicated in the patient with MCP joint hyperextension deformity >10°, as the volar aspect of the joint would still be overloaded.

First, a bone cut is made parallel to the metacarpal joint surface (Figure 9). A second bone cut is made 5 mm distal to the first cut at a 30° angle, allowing removal of a dorsal wedge of bone. The first metacarpal is then extended, and the bone gap is closed and fixed using Kirschner wires (K-wires), intraosseous wiring, or plate fixation. Hobby et al32 reported on 41 patients who were treated with first metacarpal extension osteotomy for Eaton stage I and II disease. All osteotomies healed within 7 weeks. At 6.8-year followup, 95% of patients had worthwhile pain relief. Grip and pinch strength increased to 90% of established norms. The authors concluded that metacarpal extension osteotomy is a useful treatment option for early stages of CMC arthritis of the thumb. Further studies have shown that in addition to unloading the volar articulation of the CMC joint, osteotomy may reduce joint laxity and provide dynamic stabilization, which also may contribute to its success.33,34

Figure 9
Figure 9:
Osteotomy of the thumb metacarpal to correct adduction contracture, restore the first web space, and reduce the tendency of the action of the flexor and extensor pollicis longus to subluxate the thumb carpometacarpal joint. A, The proximal cut is made parallel to the metacarpal joint surface. The distal cut is made at a 30° angle, allowing resection of a dorsal wedge of bone. B, The metacarpal is extended, and the bone gap is closed and held with internal fixation. (Reproduced with permission from Hobby JL, Lyall HA, Meggitt BF: First metacarpal osteotomy for trapeziometacarpal osteoarthritis. J Bone Joint Surg Br 1998;80:508-512.)

Carpometacarpal Joint Arthroscopy

Arthroscopy can be used to evaluate the CMC joint and treat early stages of arthritis with débridement, synovectomy, and/or electrothermal capsular shrinkage.35 In the later stages of disease (stages III and IV), arthroscopy can be combined with partial or complete trapeziectomy, with or without interpositional arthroplasty and percutaneous pinning. Arthroscopic surgery allows assessment of posttraumatic arthritis, which is especially applicable in patients with a previous Bennett fracture.

The portals used for CMC joint arthroscopy include a dorsal 1R portal (radial to the APL tendon) and a dorsal 1U portal (ulnar to the EPB tendon between the EPL and EPB tendons) (Figure 10). The joint is insufflated with saline. An incision is made with a no.11 blade, followed by blunt dissection carried down to the capsule to create the portals. A 1.9-mm arthroscope is inserted into the CMC joint, and a triangulation probe is inserted. The articular surfaces of the CMC joint are assessed diagnostically. Therapeutic intervention is determined based on the arthroscopic findings and stage of disease.

Figure 10
Figure 10:
Lateral (radial) view of the first carpometacarpal (CMC) joint region, illustrating the location of the 1R and 1U arthroscopic portals utilized during CMC joint arthroscopy. Note the proximity of the 1R portal to the radial artery (RA) and the relationship of the portals to the extrinsic tendons. APL = abductor pollicis longus tendon, EPB = extensor pollicis brevis tendon, EPL = extensor pollicis longus tendon, MI = 1st metacarpal, MII = 2nd metacarpal, MIII = 3rd metacarpal, SRN = superficial radial nerve, Tm = trapezium. (Reproduced with permission from the Mayo Foundation.)

Culp and Rekant36 reported on 24 thumbs following arthroscopy combined with various procedures, including hemitrapeziectomy, total trapeziectomy, and thermal capsular shrinkage. Good to excellent results were reported in 88% of patients at 1.2- to 4-year follow-up. Pinch strength improved by 22%.

Thumb Carpometacarpal Arthrodesis

Fusion of the thumb CMC joint has been shown to improve pain and increase strength. This procedure is indicated for the patient with painful instability secondary to systemic joint laxity, the patient younger than age 50 years in a high-demand occupation, and the patient with isolated CMC arthritis with Eaton stage II/III disease. The STT joint should be spared. The advantages of fusion include a stable thumb, improved strength, pain relief, and joint stability. Disadvantages include the risk of the development of adjacent joint degeneration (STT or MCP joints), a fairly high nonunion rate, and prolonged postoperative casting (up to 3 months). Additionally, the patient loses full mobility of the thumb and may have difficulty getting her or his hand into a flat position or into pockets.

The preferred position for fusion is thumb key pinch, which involves positioning the thumb in pronation so that the thumb pulp rests on the radial aspect of the index middle phalanx. This position generally is described as 30° to 40° of palmar abduction and 10° to 20° of radial abduction and extension. CMC arthrodesis is performed through an S-shaped incision between the APL and APB tendons. The capsule is incised, and the scaphotrapezial joint is inspected for signs of arthritis. Fusion can be performed in the absence of significant arthritis in this adjacent joint. The cartilaginous surfaces of the CMC joint are then removed down to a bleeding bone bed. The thumb is positioned in key pinch with the cut bone surfaces approximated and correctly aligned. Various fixation methods can be used, including K-wires for a tension band, plates and screws, condylar blade plates, and Herbert screws (Figure 11).

Figure 11
Figure 11:
Carpometacarpal arthrodesis using mini-fragment blade plate fixation.

Fulton and Stern37 studied CMC arthrodesis in 49 patients (average follow-up, 7 years). The authors used K-wire fixation and reported a 7% nonunion rate. Three of the four nonunions were asymptomatic; the one symptomatic patient was revised using blade-plate fixation. At 7-year follow-up, the average visual analog pain score was 1.5 (scale, 0 to 10). Seven patients went on to develop adjacent joint arthritis—three at the scaphotrapezial joint and four at the trapezial-index metacarpal articulation.

Implant Arthroplasty

Long-term complications associated with silicone implants have precluded their use in the thumb CMC joint. Complications include implant wear and silicone synovitis caused by wear-induced particles ≤15 μm. Wear generally occurs at approximately 2 years postoperatively as a result of shear and compression forces across the implant. Silicone synovitis is treated with removal with curettage and interpositional arthroplasty with satisfactory results.38

Athwal et al39 reported the results of the use of the Orthosphere (Wright Medical Technology, Arlington, TN), a zirconia implant, in seven patients. At a mean 33-month follow-up, six of the seven implants had subsided into the trapezium, with the patients reporting pain and weakness. Five of the seven were revised to ligament reconstruction and tendon interposition (LRTI).

Swanson et al40 reported on the results of 105 Swanson titanium condylar implants (Wright Medical Technology) at an average follow-up of 5 years. At 6 months postoperatively, improvements were seen in motion and strength, bone remodeling was seen radiographically, and the implant was stable. No signs of wear were noted at average 5-year follow-up. These results have not been reproduced by other authors. Badia and Sambandam41 reported 24 good and excellent results in 25 elderly patients following fixed ballsocket CMC arthroplasty (average follow-up, 59 months).

Resection Arthroplasty With or Without LRTI

Resection arthroplasty of the CMC joint is regarded as the gold standard for surgical treatment of thumb CMC arthritis. Indications include a symptomatic patient with Eaton stage III/IV CMC arthritis who has failed nonsurgical treatment.

Gervis42 first described simple trapeziectomy in 1949 to treat symptomatic thumb CMC arthritis and wrote a descriptive report of 15 patients who had undergone the procedure. Since then, trapeziectomy has become one of the most popular surgical treatments for CMC arthritis, and many variations have been developed. In 1970, Froimson43 described performing tendon interposition in the space created by the trapeziectomy in an effort to prevent metacarpal subsidence. He coined the term “anchovy operation” because of the similarity of the appearance of the interposed tendon to an anchovy. In 1986, Burton and Pellegrini44 described trapeziectomy combined with LRTI to augment stability after trapezial resection, thereby preventing painful impingement of the thumb metacarpal on the scaphoid. Trapezium excision can be performed with or without ligament reconstruction and with or without tendon interposition. Tendons used for ligament reconstruction include the APL, FCR, and palmaris longus.

Selection of the approach for resection arthroplasty is based on surgeon preference. Surgical approaches described include volar, midradial, and dorsal incisions. A midaxial lateral incision takes advantage of the internervous plane (radial and median) and includes dissection between the APL (radial innervation) and APB (median innervation). If encountered, branches of the superficial radial sensory nerve should be protected. Arthrotomy exposes the CMC joint to allow for resection of the trapezium. After a simple trapeziectomy is performed, the joint capsule, subcutaneous tissues, and the skin are closed; percutaneous pinning is optional.

Once the decision has been made to perform tendon interposition and/or ligament reconstruction,45 tendon harvest is the next step. Options for tendon used for ligament reconstruction include slips of the APL,46,47 part or all of the FCR, and the palmaris longus. The palmaris longus can be used for tendon interposition without ligament reconstruction. Other interpositional materials have been described in the literature, including hematoma48 and gel foam.49 When ligament reconstruction is performed, the FCR or APL tendon is placed through drill hole tunnels that are made through the base of the first metacarpal in a volar-ulnar to dorsoradial direction (Figure 12). A more recently introduced alternative to bone tunnels is the use of bone anchor sutures. Most commonly, the procedure is done as described above, after which the tendon is passed back around the FCR tendon insertion. Overtightening of the ligament reconstruction should be avoided. After the LRTI has been performed, the thumb is held in abduction and extension. Another option is to use K-wire insertion from the 1st metacarpal to the 2nd metacarpal after resection arthroplasty or LRTI procedure.

Figure 12
Figure 12:
Trapezium resection with flexor carpi radialis ligamentous interposition (ligament reconstruction and tendon interposition) to treat carpometacarpal arthritis. (Reproduced with permission from Burton R: Resection/suspension arthroplasty of the basal joint of the thumb for osteoarthritis, in Strickland JW [ed]: The Hand. Philadelphia, PA: Lippincott-Raven, 1998, p 455.)

Gerwin et al50 reported on 20 patients who were randomized to interpositional arthroplasty or simple trapeziectomy. At an average 23-month follow-up, the authors found no difference between the groups in ability to perform activities of daily living, pain relief, radiographic basilar joint height, or patient satisfaction.

Kriegs-Au et al51 published a prospective randomized study of 31 patients in which they compared trapeziectomy plus ligament reconstruction without tendon interposition with trapeziectomy plus LRTI (average follow-up, 48 months). The authors found no difference in strength or subjective scores for satisfaction and pain and found no difference in proximal migration at rest or on stress radiographs. No correlation was found between proximal migration of the metacarpal and maximal pinch strength.

Martou et al52 performed a metaanalysis of all techniques used for the treatment of Eaton stage II, III, and IV arthritis. Of the available 254 studies in the literature, only 2 were randomized controlled trials. The authors found great variability in the indications for surgery and the outcome measurement tools used in these studies. In their meta-analysis, the authors compared arthrodesis, trapeziectomy with or without interposition, trapeziectomy with or without LRTI, osteotomy, and joint arthroplasty. LRTI provided no additional benefit compared with arthrodesis or trapeziectomy alone. Thumb CMC arthrodesis provided the best strength and stability scores compared with tendon interposition treatments. Patients with silicone arthroplasty had higher complication rates, including joint subluxation implant fractures and silicone synovitis. All procedures resulted in pain relief, but the objective measurements did not always correlate with the patient satisfaction scores. Pain relief was the most consistent finding across all studies. In a recent Cochrane review, Wajon et al53 reported that the best and safest operation was a simple trapeziectomy. In that review, trapeziectomy alone had the lowest complication rate and provided the best pain relief.


The thumb CMC joint is very important for pinch and grasp function. Its bony incongruities and required ligamentous and muscular attachments predispose it to arthritis. When managing CMC joint arthritis, it is important to exclude other causes of symptoms and to obtain adequate radiographic assessment of the surrounding joints, including a Bett view. The Eaton classification is used in conjunction with severity of clinical symptoms to make treatment recommendations. For Eaton stages I and II thumb CMC arthritis, nonsurgical treatment should be the first line of treatment, including muscle strengthening/stabilization exercises, splinting, and cortisone injection.

When nonsurgical measures fail, surgical treatments should be considered based on patient symptoms and radiographic findings. An individualized approach is required. Volar oblique ligament reconstruction may be appropriate for Eaton stages I and II disease when associated with joint subluxation. Stable but symptomatic joints with Eaton stage I or II disease can be considered for metacarpal extension osteotomy, provided the disease is confined to the volar one third of the joint. Trapezial resection arthroplasty with or without tendon interposition can be done with or without ligamentous reconstruction, and with or without supplemental K-wire fixation. Each is effective treatment for symptomatic Eaton stage III and IV disease. Trapeziectomy seems to be the common denominator that is most important for pain relief. No significant differences have been shown between interpositional tendon techniques, ligament reconstruction, and pinning. Research is being done into new synthetic implants, resurfacing arthroplasty, and spacer materials. The role of arthroscopic treatment techniques for the CMC joint and of metacarpal extension osteotomy for earlier stages of arthritis has recently been defined.


Evidence-based Medicine: There are several level I (24-27, and 51) and level II (23, 48, and 52) studies. The remaining references are casecontrol level III studies or expert opinion.

Citation numbers printed in bold type indicate references published within the past 5 years.

1. Armstrong AL, Hunter JB, Davis TR: The prevalence of degenerative arthritis of the base of the thumb in postmenopausal women. J Hand Surg [Br] 1994;19:340-341.
2. Fick A: Die gelenke mit sattel formigen flach, in Winter Cf (ed): Zeitschrift fur Rationelle Medicin. Heidelberg, Germany: Adademische Verlagshandlung, 1854, pp 314-321.
3. Hollister A, Buford WL, Myers LM, Giurintano DJ, Novick A: The axes of rotation of the thumb carpometacarpal joint. J Orthop Res 1992;10:454-460.
4. Cooney WP III, Lucca MJ, Chao EY, Linscheid RL: The kinesiology of the thumb trapeziometacarpal joint. J Bone Joint Surg Am 1981;63:1371-1381.
5. Ateshian GA, Rosenwasser MP, Mow VC: Curvature characteristics and congruence of the thumb carpometacarpal joint: Differences between female and male joints. J Biomech 1992;25:591-607.
6. Bettinger PC, Linscheid RL, Berger RA, Cooney WP III, An KN: An anatomic study of the stabilizing ligaments of the trapezium and trapeziometacarpal joint. J Hand Surg [Am] 1999;24:786-798.
7. Doerschuk SH, Hicks DG, Chinchilli VM, Pellegrini VD Jr: Histopathology of the palmar beak ligament in trapeziometacarpal osteoarthritis. J Hand Surg [Am] 1999;24:496-504.
8. Van Brenk B, Richards RR, Mackay MB, Boynton EL: A biomechanical assessment of ligaments preventing dorsoradial subluxation of the trapeziometacarpal joint. J Hand Surg [Am] 1998;23:607-611.
9. Strauch RJ, Behrman MJ, Rosenwasser MP: Acute dislocation of the carpometacarpal joint of the thumb: An anatomic and cadaver study. J Hand Surg [Am] 1994;19:93-98.
10. Bettinger PC, Smutz WP, Linscheid RL, Cooney WP III, An KN: Material properties of the trapezial and trapeziometacarpal ligaments. J Hand Surg [Am] 2000;25:1085-1095.
11. Jónsson H, Valtýsdóttir ST, Kjartansson O, Brekkan A: Hypermobility associated with osteoarthritis of the thumb base: A clinical and radiological subset of hand osteoarthritis. Ann Rheum Dis 1996;55:540-543.
12. Gamble JG, Mochizuki C, Rinsky LA: Trapeziometacarpal abnormalities in Ehlers-Danlos syndrome. J Hand Surg [Am] 1989;14:89-94.
13. North ER, Rutledge WM: The trapezium-thumb metacarpal joint: The relationship of joint shape and degenerative joint disease. Hand 1983;15:201-206.
14. Haara MM, Heliövaara M, Kröger H, et al: Osteoarthritis in the carpometacarpal joint of the thumb: Prevalence and associations with disability and mortality. J Bone Joint Surg Am 2004;86:1452-1457.
15. Koff MF, Ugwonali OF, Strauchj RJ, Rosenwasser MP, Ateshian GA, Mow VC: Sequential wear patterns of the articular cartilage of the thumb carpometacarpal joint in osteoarthritis. J Hand Surg [Am] 2003;28:597-604.
16. Moulton MJR, Parentis MA, Kelly MJ, Jacobs C, Naidu SH, Pellegrini VD Jr: Influence of metacarpophalangeal joint position on basal joint-loading in the thumb. J Bone Joint Surg Am 2001;83:709-716.
17. Eaton RG, Littler JW: Ligament reconstruction for the painful thumb carpometacarpal joint. J Bone Joint Surg Am 1973;55:1655-1666.
18. Eaton RG, Glickel SZ: Trapeziometacarpal osteoarthritis: Staging as a rationale for treatment. Hand Clin 1987;3:455-471.
19. Kubik NJ III, Lubahn JD: Intrarater and interrater reliability of the Eaton classification of basal joint arthritis. J Hand Surg [Am] 2002;27:882-885.
20. Dela Rosa TL, Vance MC, Stern PJ: Radiographic optimization of the Eaton classification. J Hand Surg [Br] 2004;29:173-177.
21. Brown GD III, Roh MS, Strauch RJ, Rosenwasser MP, Ateshian GA, Mow VC: Radiography and visual pathology of the osteoarthritic scaphotrapeziotrapezoidal joint, and its relationship to trapeziometacarpal osteoarthritis. J Hand Surg [Am] 2003;28:739-743.
22. Swigart CR, Eaton RG, Glickel SZ, Johnson C: Splinting in the treatment of arthritis of the first carpometacarpal joint. J Hand Surg [Am] 1999;24:86-91.
23. Berggren M, Joost-Davidsson A, Lindstrand J, Nylander G, Povlsen B: Reduction in the need for operation after conservative treatment of osteoarthritis of the first carpometacarpal joint: A seven year prospective study. Scand J Plast Reconstr Surg Hand Surg 2001;35:415-417.
24. Meenagh GK, Patton J, Kynes C, Wright GD: A randomised controlled trial of intra-articular corticosteroid injection of the carpometacarpal joint of the thumb in osteoarthritis. Ann Rheum Dis 2004;63:1260-1263.
25. Day CS, Gelberman R, Patel AA, Vogt MT, Ditsios K, Boyer MI: Basal joint osteoarthritis of the thumb: A prospective trial of steroid injection and splinting. J Hand Surg [Am] 2004;29:247-251.
26. Stahl S, Karsh-Zafrir I, Ratzon N, Rosenberg N: Comparison of intraarticular injection of depot corticosteroid and hyaluronic acid for treatment of degenerative trapeziometacarpal joints. J Clin Rheumatol 2005;11:299-302.
27. Fuchs S, Mönikes R, Wohlmeiner A, Heyse T: Intra-articular hyaluronic acid compared with corticoid injections for the treatment of rhizarthrosis. Osteoarthritis Cartilage 2006;14:82-88.
28. Brunelli G, Monini L, Brunelli F: Stabilisation of the trapezio-metacarpal joint. J Hand Surg [Br] 1989;14:209-212.
29. Eaton RG, Lane LB, Littler JW, Keyser JJ: Ligament reconstruction for the painful thumb carpometacarpal joint: A long-term assessment. J Hand Surg [Am] 1984;9:692-699.
30. Freedman DM, Eaton RG, Glickel SZ: Long-term results of volar ligament reconstruction for symptomatic basal joint laxity. J Hand Surg [Am] 2000;25:297-304.
31. Pellegrini VD Jr, Parentis M, Judkins A, Olmstead J, Olcott C: Extension metacarpal osteotomy in the treatment of trapeziometacarpal osteoarthritis: A biomechanical study. J Hand Surg [Am] 1996;21:16-23.
32. Hobby JL, Lyall HA, Meggitt BF: First metacarpal osteotomy for trapeziometacarpal osteoarthritis. J Bone Joint Surg Br 1998;80:508-512.
33. Shrivastava N, Koff MF, Abbot AE, Mow VC, Rosenwasser MP, Strauch RJ: Simulated extension osteotomy of the thumb metacarpal reduces carpometacarpal joint laxity in lateral pinch. J Hand Surg [Am] 2003;28:733-738.
34. Koff MF, Shrivastava N, Gardner TR, Rosenwasser MP, Mow VC, Strauch RJ: An in vitro analysis of ligament reconstruction or extension osteotomy on trapeziometacarpal joint stability and contact area. J Hand Surg [Am] 2006;31:429-439.
35. Badia A: Arthroscopy of the trapeziometacarpal and metacarpophalangeal joints. J Hand Surg [Am] 2007;32:707-724.
36. Culp RW, Rekant MS: The role of arthroscopy in evaluating and treating trapeziometacarpal disease. Hand Clin 2001;17:315-319.
37. Fulton DB, Stern PJ: Trapeziometacarpal arthrodesis in primary osteoarthritis: A minimum two-year followup study. J Hand Surg [Am] 2001;26:109-114.
38. Peimer CA: Long-term complications of trapeziometacarpal silicone arthroplasty. Clin Orthop Relat Res 1987;220:86-98.
39. Athwal GS, Chenkin J, King GJ, Pichora DR: Early failures with a spheric interposition arthroplasty of the thumb basal joint. J Hand Surg [Am] 2004;29:1080-1084.
40. Swanson AB, de Groot Swanson G, DeHeer DH, et al: Carpal bone titanium implant arthroplasty: 10 years' experience. Clin Orthop Relat Res 1997;342:46-58.
41. Badia A, Sambandam SN: Total joint arthroplasty in the treatment of advanced stages of thumb carpometacarpal joint osteoarthritis. J Hand Surg [Am] 2006;31:1605-1614.
42. Gervis WH: Excision of the trapezium for osteoarthritis of the trapeziometacarpal joint. J Bone Joint Surg Br 1949;31:537-539.
43. Froimson AI: Tendon arthroplasty of the trapeziometacarpal joint. Clin Orthop Relat Res 1970;70:191-199.
44. Burton RI, Pellegrini VD Jr: Surgical management of basal joint arthritis of the thumb: II. Ligament reconstruction with tendon interposition arthroplasty. J Hand Surg [Am] 1986;11:324-332.
45. Tomaino MM, Pellegrini VD Jr, Burton RI: Arthroplasty of the basal joint of the thumb: Long-term follow-up after ligament reconstruction with tendon interposition. J Bone Joint Surg Am 1995;77:346-355.
46. Thompson JS: Suspensionplasty technique, in Atlas of the Hands Clinics. Philadelphia, PA: WB Saunders, 1997, pp 101-126.
47. Diao E: Trapezio-metacarpal arthritis: Trapezium excision and ligament reconstruction not including the LRTI arthroplasty. Hand Clin 2001;17:223-236.
48. Kuhns CA, Emerson ET, Meals RA: Hematoma and distraction arthroplasty for thumb basal joint osteoarthritis: A prospective, single-surgeon study including outcomes measures. J Hand Surg [Am] 2003;28:381-389.
49. Nusem I, Goodwin DR: Excision of the trapezium and interposition arthroplasty with gelfoam for the treatment of trapeziometacarpal osteoarthritis. J Hand Surg [Br] 2003;28:242-245.
50. Gerwin M, Griffith A, Weiland AJ, Hotchkiss RN, McCormack RR: Ligament reconstruction basal joint arthroplasty without tendon interposition. Clin Orthop Relat Res 1997;342:42-45.
51. Kriegs-Au G, Petje G, Fojtl E, Ganger R, Zachs I: Ligament reconstruction with or without tendon interposition to treat primary thumb carpometacarpal osteoarthritis: A prospective randomized study. J Bone Joint Surg Am 2004;86:209-218.
52. Martou G, Veltri K, Thoma A: Surgical treatment of osteoarthritis of the carpometacarpal joint of the thumb: A systematic review. Plast Reconstr Surg 2004;114:421-432.
53. Wajon A, Ada L, Edmunds I: Surgery for thumb (trapeziometacarpal joint) osteoarthritis. Cochrane Database Syst Rev 2005;4:CD004631.
© 2008 by American Academy of Orthopaedic Surgeons