Small Joint Arthroplasty of the Hand

Small joint arthroplasty in the hand means the metacarpophalangeal (MCP), proximal interphalangeal (PIP) and thumb carpometacarpal (CMC) joints. The disease, the implant and the evidence differ at each, so the page takes them joint by joint.

The Swanson principle. The Swanson silicone implant is a flexible spacer, not an anatomic joint. It is made of medical-grade silicone elastomer (polydimethylsiloxane) in a hinged design that allows flexion and extension, and it works as an encapsulation arthroplasty: the capsule that forms around the implant provides the stability, not the implant. Clinical function is often maintained despite implant fracture because the capsule remains intact.

The rheumatoid order. The rheumatoid hand presents with ulnar drift and MCP subluxation. Reconstruction runs proximal to distal: address wrist pathology before the MCP joints, and correct the MCP joints before the PIP joints.

MCP arthroplasty. Rheumatoid arthritis with painful joint destruction and ulnar drift is the primary indication. The others are:

• Post-traumatic arthritis with preserved bone stock
• Osteoarthritis (less common, generally good bone stock)
• Failed previous arthroplasty requiring revision

PIP arthroplasty. Results are better in osteoarthritis than in rheumatoid arthritis, which gives stiffer results and more guarded outcomes than at the MCP joint. The indications are:

• Osteoarthritis with stiffness greater than 30 degrees flexion contracture, where motion is desired
• Rheumatoid arthritis
• Post-traumatic arthritis
• Select cases of psoriatic arthritis with adequate bone stock

Choosing the PIP patient. Arthroplasty rather than fusion suits the patient who will gain the most from motion:

• Middle and ring fingers are the best candidates, because motion is valuable for grip
• Osteoarthritis is preferred to rheumatoid arthritis
• Intact collateral ligaments and extensor mechanism
• Adequate bone stock for implant fixation
• No fixed severe deformity: contractures correctable at surgery

Thumb CMC arthroplasty. The indications are Eaton stage II-IV thumb CMC osteoarthritis refractory to conservative management, post-traumatic arthritis of the first CMC joint, and failed previous CMC surgery. Pantrapezial arthritis is also an indication but may require STT arthrodesis in addition.

Contraindications. Arthroplasty is absolutely contraindicated by:

• Active infection, local or systemic
• Absent or severely deficient extensor mechanism
• Severe bone loss precluding implant fixation
• Fixed deformity not correctable at surgery (consider arthrodesis)
• A non-functional hand, where there is no reconstructive benefit

It is relatively contraindicated by:

• Young, high-demand manual labourers (consider arthrodesis)
• The index finger PIP joint (above)
• Severe soft-tissue contractures
• Inadequately controlled inflammatory arthritis
• A non-compliant patient unable to follow the postoperative protocol

The MCP joint. A condyloid (ellipsoid) joint permitting flexion, extension, abduction, adduction and limited rotation. The collateral ligaments are taut in flexion and lax in extension, which is critical for rehabilitation. Its other key structures are the accessory collateral ligaments, the volar plate and the dorsal extensor hood mechanism.

How rheumatoid disease deforms it. Synovitis leads to progressive destruction with stretching of the radial sagittal band and ulnar subluxation of the extensor tendons, producing the characteristic ulnar drift. The intrinsic muscles (interossei and lumbricals) contribute to the deformity through their altered line of pull.

The PIP joint. A bicondylar hinge whose primary motion is flexion and extension in the sagittal plane. Stability comes from the proper and accessory collateral ligaments, the volar plate and the central slip of the extensor mechanism. The joint withstands forces up to 3 times body weight during pinch grip.

How arthritis deforms it. In osteoarthritis, marginal osteophytes (Bouchard nodes) form dorsally and laterally. In rheumatoid arthritis, progressive synovitis leads to a boutonniere or a swan-neck deformity:

• Boutonniere - central slip attenuation, PIP flexion, DIP hyperextension
• Swan neck - PIP hyperextension, DIP flexion

The thumb CMC joint. A saddle (sellar) joint between the trapezium and the base of the first metacarpal, permitting opposition, flexion, extension, abduction and adduction. It is stabilised by the anterior oblique (beak) ligament, the most important for stability, together with the dorsoradial, posterior oblique and intermetacarpal ligaments.

Basal joint arthritis. Primary osteoarthritis preferentially affects this joint in postmenopausal women, due to ligamentous laxity and repetitive loading. Pantrapezial arthritis extends to the scaphotrapezial and trapeziotrapezoid joints.

Before committing to arthroplasty, confirm the correct diagnosis and joint. Several conditions mimic primary degenerative or rheumatoid joint destruction and change the operation entirely.

The implant follows the joint. The table sets the options side by side; the technique for each is in the next section, and what each has achieved is under Outcomes.

Silicone implant fracture. Silicone implants fracture in 50-70% at 10-year follow-up, and roughly 60-67% are radiographically broken by long-term follow-up. It is often asymptomatic because capsular integrity is maintained, so treat it only if it is symptomatic, with pain, instability or crepitus. Observe the asymptomatic fracture; offer revision arthroplasty or arthrodesis for the symptomatic one.

Silicone synovitis. A foreign-body reaction to silicone particles, in 5-10% of silicone arthroplasties. It presents with swelling, pain and osteolysis on radiographs, and may mimic infection. Treat it with synovectomy and implant exchange or conversion to arthrodesis; titanium grommets may reduce wear-particle generation (below).

Subsidence and loosening. More common with the anatomic designs (SRA, pyrocarbon) than with silicone. Poor bone quality, malalignment and high activity level are the risk factors, and it presents with progressive pain and loss of motion. Management is revision with bone grafting or conversion to arthrodesis.

Recurrent ulnar drift. At the MCP joint, drift recurs through:

• Inadequate soft-tissue balancing at the index operation
• Progressive stretching of the radial sagittal band repair
• Continued rheumatoid disease activity

Prevention is meticulous soft-tissue technique and disease control. Established recurrence is managed with a revision soft-tissue procedure or revision arthroplasty.

Boutonniere or swan-neck recurrence. At the PIP joint, central slip attenuation leads to a recurrent boutonniere and lateral band subluxation causes a swan neck. Either may require extensor reconstruction or conversion to arthrodesis.

Infection. Rheumatoid patients are at higher risk because of immunosuppression. Early wound infection is treated with antibiotics and wound care; deep or chronic infection needs implant removal, debridement and delayed reconstruction.

Stiffness. Common, and more so at the PIP joint than at the MCP joint. Prevention is early protected motion in a compliant patient. Treatment is aggressive therapy; manipulation under anaesthesia is rarely helpful.

Instability. The result of ligament insufficiency or bone loss, which may require revision with a constrained implant or arthrodesis.

MCP arthroplasty: early gains. In rheumatoid arthritis the arc of motion improves (around 30 to 46 degrees), and ulnar drift and extensor lag correct substantially in the first year.

MCP arthroplasty: the long view. The gains deteriorate with time. By 14 years in Goldfarb and Stern's series the arc had fallen back toward 36 degrees, the extension deficit had recurred to around 23 degrees and ulnar drift to a mean of 16 degrees; patients were satisfied with only 38% of hands, and 27% were pain-free. Implant survival in Trail's series was 63% at 17 years, with failure defined as revision or radiographic fracture.

Surgery against medical therapy. Despite this, prospective comparison with medical therapy alone shows that surgery still improves patient-reported hand function and corrects deformity better than non-operative management (Chung, level II).

Predictors of a good MCP result. These are:

• Rheumatoid arthritis (better than osteoarthritis or post-traumatic arthritis)
• Adequate bone stock
• Good soft-tissue quality
• Compliance with the therapy protocol
• Well-controlled systemic disease

PIP arthroplasty: silicone. In rheumatoid and osteoarthritic joints the motion gain is modest, an average arc of 40-50 degrees, and pain relief is achieved in 75-80%.

PIP arthroplasty: surface replacement and pyrocarbon. In osteoarthritis the outcomes are inconsistent and series-dependent: pain relief can be good, but motion often does not improve. In Sweets and Stern's single-surgeon pyrocarbon series the arc of motion actually fell from 57 to 31 degrees, with frequent loosening, migration, dislocation and squeaking, and the authors abandoned the implant.

What that means for the patient. These designs have a higher reoperation and loosening rate than silicone and are best reserved for osteoarthritis with intact collateral ligaments and good bone stock. Counsel patients that motion preservation is not guaranteed and that arthrodesis is a reliable fallback.

Thumb CMC: trapeziectomy with or without LRTI. Good pain relief in roughly 78% of patients at long-term RCT follow-up (Gangopadhyay), and higher in observational series. Grip and pinch strength recover, although pinch may decline slowly over years irrespective of technique. Subsidence (proximal metacarpal migration) is near-universal but correlates poorly with symptoms.

Thumb CMC: implants. Implant arthroplasty (pyrocarbon, total joint) has a higher complication rate. Total joint (dual-mobility) prostheses show faster early recovery in some series but lack the long-track-record durability data of trapeziectomy.

The question. The long-term series paint a pessimistic radiographic picture, which raises a fair examiner question: is silicone MCP arthroplasty worth doing and worth funding? The health-economic evidence answers more favourably than the survivorship data alone would suggest, and the answer depends heavily on the outcome instrument. A hand-specific, patient-rated measure captures the real benefit far better than a whole-body arthritis-impact scale.

Durable when measured well. In the Squitieri prospective cohort, the functional gains of surgery were maintained across the full 5-year follow-up rather than eroding, when measured with the hand-specific Michigan Hand Outcomes Questionnaire (MHQ).

Cheap on one instrument, dear on the other. Improved hand-specific outcomes were achieved at an incremental cost of $787 to $1,150 per unit of MHQ benefit, so the operation looks cost-effective on a hand-specific measure even after accounting for complications and revision. The same improvement cost $49,843 to $149,530 per unit of benefit on the generic Arthritis Impact Measurement Scale: a gap of two orders of magnitude, and a range that straddles and mostly exceeds conventional willingness-to-pay thresholds.

Two readings. The usual reading is that rebuilding a few finger joints barely moves a systemic disease score, so a whole-patient measure understates a focal operation. Be ready for the opposite reading in a viva, because it is equally available: if four reconstructed joints cannot shift a whole-patient score, the whole-patient value may genuinely be modest. Neither instrument is a QALY, so neither figure is a true cost-utility ratio.

Revision does not sink the economics. The favourable cost-effectiveness held even when the observed and previously published revision rates (around 5 to 6 percent) were built into the model.

The reconciliation. This is how the gloomy long-term radiographic series and continued use of the procedure fit together. For the severely deformed, painful rheumatoid hand, silicone MCP arthroplasty delivers durable, patient-valued functional improvement at acceptable cost, provided it is judged by what the hand can do, not by a systemic disease score.

Global Epidemiology

• Thumb CMC (trapeziometacarpal) osteoarthritis is the most common small-joint indication worldwide, with a strong female and post-menopausal predominance; radiographic basal-joint OA is found in roughly a third of post-menopausal women.
• Rheumatoid arthritis affects approximately 0.5-1% of adults globally. The classic severely deformed rheumatoid MCP hand requiring arthroplasty is becoming less common as early biologic and conventional DMARD therapy controls disease before joint destruction.
• Erosive/inflammatory hand OA and post-traumatic arthritis make up most remaining PIP and MCP arthroplasty caseload.

Side-by-Side Guidance

Registry and Evidence Notes

• Unlike hip and knee arthroplasty, small joint arthroplasty is poorly captured by national joint registries; most outcome data come from single-centre series and a small number of RCTs (Gangopadhyay; Chung). This limits implant-level comparative survivorship data.
• The consistent registry-independent message: trapeziectomy-based procedures are durable and additional ligament reconstruction adds morbidity without benefit; silicone MCP implants fracture frequently but the encapsulation arthroplasty continues to function.

High- vs Limited-Resource Practice Variation

• In well-resourced settings, certified hand therapists and dynamic splinting protocols are integral to MCP arthroplasty outcomes, and pyrocarbon or total joint implants are available at higher cost for selected motion-preserving indications.
• In limited-resource settings, arthrodesis and simple trapeziectomy are favoured because they are cheap, reliable, and require less specialised post-operative therapy and follow-up. The reduced dependence on hand-therapy infrastructure is a legitimate driver of procedure choice, not a compromise of standards.

Related pages: Rheumatoid Hand and Wrist is the disease context for every MCP card here, and the page that carries the reconstructive sequence and the effect of modern disease-modifying therapy on how often these operations are now needed; MCP Joint Arthritis for the degenerative rather than inflammatory MCP joint, in which silicone behaves differently and the evidence above does not transfer; Thumb CMC Arthritis for the basal joint in full, including the Eaton staging and the trapeziectomy literature summarised here; Swan Neck Deformity and Boutonniere Deformity for the PIP-level deformities that must be corrected or accounted for before a small-joint arthroplasty will balance - an unaddressed swan neck will recur through a new implant.

Does silicone MCP arthroplasty earn its place? The long-term series (Goldfarb, Trail) show deteriorating motion, recurrent deformity and high fracture rates, while the prospective Chung trial shows real functional gains over medical therapy. The honest position is that it is a reasonable option for the severely deformed, painful rheumatoid hand, but expectations must be measured, and it is being performed less often as biologic disease control improves.

LRTI or simple trapeziectomy? LRTI is the traditional gold standard and remains popular, yet the highest-level evidence (the Gangopadhyay RCT and the Vermeulen review) shows no benefit over trapeziectomy alone, and the Vermeulen review found a higher complication rate with LRTI. Outcomes are equivalent at one year and, in the Gangopadhyay trial, at a minimum of five years; ligament reconstruction adds operative time without proven clinical benefit.

Where the thumb debate is going. Many surgeons nonetheless retain a suspension step. The debate is now shifting toward whether modern total joint replacement offers faster recovery worth its revision risk.

Pyrocarbon and surface PIP implants. They are marketed for motion preservation, yet independent series (Sweets and Stern) report motion loss, loosening, migration and high reoperation. Whether better patient selection (intact collaterals, osteoarthritis rather than rheumatoid arthritis, a well-aligned joint) rescues these designs remains unresolved.

Arthrodesis or arthroplasty? For the index PIP, the border digits and young manual workers, fusion trades motion for durable, pain-free stability. There is no consensus threshold: the decision is individualised to the joint, the demand, and the value the patient places on motion versus reliability.

The rationale. Two classic failings of the silicone spacer, implant fracture and particulate silicone synovitis, were both attributed, in part, to abrasion of the flexible stem against the sharp cut edge of bone at the mouth of the medullary canal. Swanson's answer was the grommet: a small titanium eyelet seated in the canal entrance at the bone-implant junction, interposing a smooth, rigid collar so the stem no longer chafes against a bony edge. The theoretical case is attractive: protect the weakest part of the hinge and reduce the shedding of the silicone particles that provoke synovitis and osteolysis.

The evidence. The clinical evidence has not supported it. In Trail's 17-year survivorship series grommets did not protect the implant from fracture, and they add cost, technical steps and their own potential for malposition or loosening. The modifiable factors that prolonged survival in that series were soft-tissue balancing, crossed intrinsic transfer and wrist realignment, not the hardware.

Where they stand. Grommets may still be considered where the priority is limiting particulate wear and synovitis rather than fracture. They should never be relied on to prevent implant breakage, and many surgeons now omit them.

1. Swanson AB. Flexible implant resection arthroplasty. Hand. 1972;4(2):119-134.

2. Goldfarb CA, Stern PJ. Metacarpophalangeal joint arthroplasty in rheumatoid arthritis. A long-term assessment. J Bone Joint Surg Am. 2003;85(10):1869-1878.

3. Trail IA, Martin JA, Nuttall D, Stanley JK. Seventeen-year survivorship analysis of silastic metacarpophalangeal joint replacement. J Bone Joint Surg Br. 2004;86(7):1002-1006.

4. Vermeulen GM, Slijper H, Feitz R, et al. Surgical management of primary thumb carpometacarpal osteoarthritis: a systematic review. J Hand Surg Am. 2011;36(1):157-169.

5. Gangopadhyay S, McKenna H, Burke FD, Davis TRC. Five- to 18-year follow-up for treatment of trapeziometacarpal osteoarthritis. J Hand Surg Am. 2012;37(3):411-417.

6. Sweets TM, Stern PJ. Pyrolytic carbon resurfacing arthroplasty for osteoarthritis of the proximal interphalangeal joint of the finger. J Bone Joint Surg Am. 2011;93(15):1417-1425.

7. Chung KC, Burns PB, Wilgis EFS, et al. A multicenter clinical trial in rheumatoid arthritis comparing silicone metacarpophalangeal joint arthroplasty with medical treatment. J Hand Surg Am. 2009;34(5):815-823.

8. Eaton RG, Glickel SZ. Trapeziometacarpal osteoarthritis. Staging as a rationale for treatment. Hand Clin. 1987;3(4):455-471.

9. Squitieri L, Chung KC, Hutton DW, et al. A 5-year cost-effectiveness analysis of silicone metacarpophalangeal arthroplasty in patients with rheumatoid arthritis. Plast Reconstr Surg. 2015;136(2):305-314.

10. Burton RI, Pellegrini VD Jr. Surgical management of basal joint arthritis of the thumb. Part II. Ligament reconstruction with tendon interposition arthroplasty. J Hand Surg Am. 1986;11(3):324-332.