This article reviews the current indications and clinical outcomes of total wrist arthroplasty. The section on indications reviews both rheumatoid and nonrheumatoid arthritic conditions. The section on clinical outcomes examines the data regarding the 3 current total wrist implants approved by the Food and Drug Administration.
Key points
- •
Total wrist arthroplasty, an alternative to wrist arthrodesis, is indicated in patients with painful, debilitating, panarthritis with or without progressive wrist deformity, and may be essential in patients with bilateral wrist and/or concomitant upper extremity disease.
- •
Though classically performed in the rheumatoid patient, advances have led present-day application of total wrist arthroplasty in patients with rheumatoid and nonrheumatoid inflammatory arthritis, osteoarthritis, and posttraumatic arthritis.
- •
Modern designs have significantly reduced rates of instability and prosthetic dislocations; however, carpal component loosening remains an obstacle to good and predictable long-term survival.
- •
Published outcomes are limited to early and/or midterm results for the 3 prostheses approved by the Food and Drug Administration.
Although Gluck performed the first total wrist arthroplasty (TWA) in 1890, it was not until the 1970s that interest in wrist arthroplasty gained momentum. In 1967, Swanson designed the first wrist implant with widespread commercial distribution. The Swanson double-stemmed, flexible-hinge, silicone implant functioned as a joint spacer, rather than a joint replacement. Despite early positive results, longer follow-up revealed implant subsidence and prosthetic fracture. Since this first design, TWA has undergone considerable improvements in material and design that have had a considerable impact on implant survival. Research in joint arthroplasty and, in particular, wrist arthroplasty has contributed to implant longevity, patient satisfaction, and improved clinical outcomes. This article reviews the current indications and clinical outcomes of TWA.
Indications and contraindications
TWA is an alternative to wrist arthrodesis, and is indicated in patients with painful, debilitating, panarthritis with or without progressive deformity of the wrist. Its ideal candidate is the low-demand patient with painful, bilateral wrist disease and relatively good wrist alignment and motion despite arthritis. It is also ideal for patients with concomitant upper extremity arthritis. Where involvement of ipsilateral joints would complicate positioning of the hand in space, arthroplasty helps the patient compensate for lack of motion in neighboring joints. As with other joint replacements in the upper extremity, it is well described in patients afflicted with rheumatoid arthritis. As research in wrist arthroplasty has undergone expansion and refinement, so too have its indications. Such advances have led to its present-day application in patients with rheumatoid and nonrheumatoid inflammatory arthritis, osteoarthritis, and posttraumatic arthritis. Wrist arthroplasty is also applicable in failed limited wrist fusion and in advanced avascular necrosis of the carpal bones.
As with other joint replacements, TWA is contraindicated in the nonfunctioning hand lacking neurologic function, as well as in high-demand patients, laborers, and those with a previous history of sepsis or deep local infection. Specific to TWA, it is also contraindicated in those that require regular use of walking aids for support during ambulation or transfers. Relative contraindications include those conditions with potentially inadequate carpal bone stock, such as a previous surgical complete fusion or a proximal row carpectomy.
Rheumatoid Arthritis
Earlier implant designs were uniquely reported in rheumatoid patients. As understanding of wrist arthroplasty improved, indications and contraindications within the rheumatoid population were further defined. The high rate of complications and failures associated with earlier implants led to its contraindication in rheumatoid patients with poor bone stock or severe osteopenia. In these patients, this could result in component loosening owing to inability to support the prosthetic components, especially the carpal component. Severe ligamentous laxity in the form of chronic or severe volar and ulnar subluxation is also contraindicated because of prosthesis instability and ultimate dislocation. Rheumatoid patients with highly active synovitis are prone to severe bony erosions and/or joint hyperlaxity, resulting in a higher risk of subsequent implant instability and loosening. In end-stage rheumatoid arthritis characterized by soft-tissue imbalance and volar subluxation of the wrist, ruptured or attenuated radial wrist extensors are also a contraindication for wrist replacement, and result in prosthesis instability and subluxation.
In treating the rheumatoid patient with multiple joint involvement, lower extremity replacements such as the hip and/or knee should be performed before wrist replacement, to avoid weight bearing on the wrist replacement during rehabilitation. By contrast, procedures on the digits should be performed after wrist arthroplasty to optimize joint alignment and tendon tension in the hand.
Nonrheumatoid Arthritis
With regard to nonrheumatoid inflammatory arthritides, systemic lupus erythematosus is a relative contraindication because of its predisposition to joint laxity. Although severely underpowered in the literature, early and midterm results after arthroplasty have been reported for the treatment of arthritis secondary to gout, psoriasis, lupus, undifferentiated spondyloarthropathy, and chondrocalcinosis.
In a recent study, Nydick and colleagues reported on early outcomes in 22 patients treated with the Maestro TWA system (Biomet, Warsaw, IN). In this study the etiology included scapholunate advanced collapse (n = 8), scaphoid nonunion advanced collapse (n = 5), primary osteoarthritis (n = 2), Kienböck disease (n = 2), and rheumatoid arthritis (n = 5). The data demonstrated efficacy for arthroplasty in nonrheumatoid conditions, likely attributable to better bone stock and better radiocarpal alignment, and potentially resulting in future low long-term failure rates. Patients with noninflammatory, nonrheumatoid arthritis tend to be healthier and have higher demands, theoretically leading to more rapid implant failure. With strict adherence to activity limitations, this patient population may be at a lower risk of implant loosening because they lack the bony erosions and/or joint hyperlaxity typically associated with synovitis.
Indications and contraindications
TWA is an alternative to wrist arthrodesis, and is indicated in patients with painful, debilitating, panarthritis with or without progressive deformity of the wrist. Its ideal candidate is the low-demand patient with painful, bilateral wrist disease and relatively good wrist alignment and motion despite arthritis. It is also ideal for patients with concomitant upper extremity arthritis. Where involvement of ipsilateral joints would complicate positioning of the hand in space, arthroplasty helps the patient compensate for lack of motion in neighboring joints. As with other joint replacements in the upper extremity, it is well described in patients afflicted with rheumatoid arthritis. As research in wrist arthroplasty has undergone expansion and refinement, so too have its indications. Such advances have led to its present-day application in patients with rheumatoid and nonrheumatoid inflammatory arthritis, osteoarthritis, and posttraumatic arthritis. Wrist arthroplasty is also applicable in failed limited wrist fusion and in advanced avascular necrosis of the carpal bones.
As with other joint replacements, TWA is contraindicated in the nonfunctioning hand lacking neurologic function, as well as in high-demand patients, laborers, and those with a previous history of sepsis or deep local infection. Specific to TWA, it is also contraindicated in those that require regular use of walking aids for support during ambulation or transfers. Relative contraindications include those conditions with potentially inadequate carpal bone stock, such as a previous surgical complete fusion or a proximal row carpectomy.
Rheumatoid Arthritis
Earlier implant designs were uniquely reported in rheumatoid patients. As understanding of wrist arthroplasty improved, indications and contraindications within the rheumatoid population were further defined. The high rate of complications and failures associated with earlier implants led to its contraindication in rheumatoid patients with poor bone stock or severe osteopenia. In these patients, this could result in component loosening owing to inability to support the prosthetic components, especially the carpal component. Severe ligamentous laxity in the form of chronic or severe volar and ulnar subluxation is also contraindicated because of prosthesis instability and ultimate dislocation. Rheumatoid patients with highly active synovitis are prone to severe bony erosions and/or joint hyperlaxity, resulting in a higher risk of subsequent implant instability and loosening. In end-stage rheumatoid arthritis characterized by soft-tissue imbalance and volar subluxation of the wrist, ruptured or attenuated radial wrist extensors are also a contraindication for wrist replacement, and result in prosthesis instability and subluxation.
In treating the rheumatoid patient with multiple joint involvement, lower extremity replacements such as the hip and/or knee should be performed before wrist replacement, to avoid weight bearing on the wrist replacement during rehabilitation. By contrast, procedures on the digits should be performed after wrist arthroplasty to optimize joint alignment and tendon tension in the hand.
Nonrheumatoid Arthritis
With regard to nonrheumatoid inflammatory arthritides, systemic lupus erythematosus is a relative contraindication because of its predisposition to joint laxity. Although severely underpowered in the literature, early and midterm results after arthroplasty have been reported for the treatment of arthritis secondary to gout, psoriasis, lupus, undifferentiated spondyloarthropathy, and chondrocalcinosis.
In a recent study, Nydick and colleagues reported on early outcomes in 22 patients treated with the Maestro TWA system (Biomet, Warsaw, IN). In this study the etiology included scapholunate advanced collapse (n = 8), scaphoid nonunion advanced collapse (n = 5), primary osteoarthritis (n = 2), Kienböck disease (n = 2), and rheumatoid arthritis (n = 5). The data demonstrated efficacy for arthroplasty in nonrheumatoid conditions, likely attributable to better bone stock and better radiocarpal alignment, and potentially resulting in future low long-term failure rates. Patients with noninflammatory, nonrheumatoid arthritis tend to be healthier and have higher demands, theoretically leading to more rapid implant failure. With strict adherence to activity limitations, this patient population may be at a lower risk of implant loosening because they lack the bony erosions and/or joint hyperlaxity typically associated with synovitis.
Arthroplasty versus arthrodesis
The main advantage of arthroplasty over arthrodesis is preservation or improvement of some degree of wrist mobility. Although arthrodesis effectively relieves pain and corrects deformity, the resulting lack of motion may significantly impair function, especially when the shoulder, elbow, and hand are also affected by arthritis. At present, there are no prospective randomized trials comparing arthroplasty with arthrodesis. In a comparative, retrospective study of 51 operated wrists in 46 rheumatoid patients undergoing 24 arthrodeses and 27 arthroplasties, Murphy and colleagues found no difference in Disability of the Arm, Shoulder, and Hand (DASH) or Patient-Related Wrist Evaluation (PRWE) scores, nor complications. Patients in the arthroplasty group did, however, report a trend toward greater ease with personal hygiene and fine dexterity requiring wrist flexion. The investigators concluded that although rheumatoid patients can and do accommodate arthrodesis, it should not be construed that these patients would prefer arthrodesis over arthroplasty or that they would not obtain greater benefit from arthroplasty.
In 2008, Cavaliere and Chung performed a systematic review comparing 18 TWA studies representing approximately 500 procedures and 20 total wrist fusion studies, representing approximately 800 procedures. Based on their review of the existing literature, they reported more reliable pain relief with total wrist fusion, and higher complication and revision rates with TWA (30% arthroplasty group, 17% arthrodesis group). High satisfaction rates were seen in both groups. Of the 14 studies providing mean flexion, extension, and radial and ulnar deviations, only 3 studies showed an average active arc of motion within functional range as described by Palmer and colleagues. From this systematic review, the investigators concluded that in the rheumatoid patient, outcomes for total wrist fusion were comparable and possibly better than those for TWA, and that current evidence did not support the widespread implementation of arthroplasty. The cases reviewed consisted of second-generation and early third-generation metal-polyethylene TWA prostheses, and did not include newer third-generation implants.
A subsequent study by Cavaliere and Chung using a cost-utility analysis model showed that arthroplasty had only an incrementally higher cost in comparison with wrist fusion. Moreover, rheumatoid patients place emphasis on maintaining wrist motion, and for that reason would prefer arthroplasty to fusion and nonoperative management.
Outcomes
From the double-stemmed silicone spacer to the more current anatomic metal-polyethylene implants, TWA implants have undergone considerable changes aimed at increasing implant survivorship. In long-term follow-up, the Swanson flexible-hinge, silicone joint spacer (first-generation implant) was complicated by implant subsidence and prosthetic fracture. Often considered the first true arthroplasty of the wrist, second-generation implants, consisting of an articulating metal-polyethylene design (Meuli and Volz) had an unacceptably high complication rate characterized by wrist imbalance and implant loosening. In a failed attempt to reduce wrist imbalance and distal component loosening, early third-generation implants (Biaxial, Trispherical, revised Meuli, and revised Volz) were designed to better reproduce normal wrist kinematics. Nowadays, third-generation metal-polyethylene implants are available. At present there are only 3 TWA designs approved by the US Food and Drug Administration (FDA): Universal 2 (KMI, San Diego, CA), ReMotion (SBI, Morrisville, PA), and Maestro (Biomet, Warsaw, IN).
Universal Total Wrist Arthroplasty
Although the Universal 2 TWA is FDA-approved, it is difficult to discuss the evolution of this implant without discussion of the Universal 1 TWA. Before the development of the Universal 1 TWA, third-generation metal-polyethylene implants were plagued with persistent implant loosening and wrist imbalance. Developed by Menon in the 1990s, the Universal 1 TWA introduced a new approach to distal fixation and the concept of anatomic components. A short central stem cemented into the capitate and 2 deep-threaded osteointegrative screws into the radial and ulnar aspects of the carpus were designed to provide superior distal fixation. Intercarpal fusion was also performed to provide long-term solid bony support. The radial component was inclined 20° to replicate the slope of the normal distal radius. The polyethylene insert, toroidal in shape, consisted of a broad articulation closely matching that of the proximal carpal row. Soft-tissue balancing was adjusted by variations in polyethylene size.
In 1998, Menon published midterm results using the Universal 1 TWA implant. Thirty-seven procedures were performed in 31 patients, of whom 23 suffered from rheumatoid arthritis and the remaining 8 from osteoarthritis. The average follow-up was 6.7 years, ranging from 4 to 10 years. All patients were given the option to receive wrist fusion. Those without active wrist extension were not given the option of arthroplasty. After excluding 3 patients from the study, 34 were available for review. Of the 34 wrists, 88% (30 patients) were pain free while the remaining patients required periodic analgesics and occasionally wore a wrist brace for support. Although all ranges of motion increased postoperatively, only dorsiflexion and radial deviation increased to statistical significance. In this series the complication rate was 32%, occurring in 12 wrists. The most common complication was volar dislocation of the wrist, occurring in 5 wrists. No case demonstrated radiographic evidence of distal component loosening. However, radial component loosening was observed in 2 patients at 2 and 3 years postoperatively, requiring reoperation with extremely good results. Of the 3 patients eliminated from final review, 2 were converted to wrist fusion secondary to persistent flexion deformity interfering with activities of daily living and recurrent dislocation in a known alcoholic. The third patient received removal of hardware because of staphylococcal pyarthrosis of the wrist.
In a further follow-up study by Menon that included 57 implants, carpal component loosening was not reported. Subsidence of the radial component was observed, but was not progressive or symptomatic. Pain relief and functional range of motion remained constant while dislocation occurred in 6 of the 57 patients, with 5 occurring in the first 37 cases. The initial higher incidence of dislocation was attributed to the lack of availability of different implant sizes and the thickness of polyethylene inserts at the time. Menon also criticized himself for excessive bone resection in his early cases. Other investigators reported similar success with shorter follow-up, indicating that with a proper design, wrist replacement can provide suitable function and durability.
In 2002, Divelbiss and colleagues sought to duplicate Menon’s results. In 22 rheumatoid wrists (19 patients), clinical results similar to those of Menon were achieved with regard to pain relief, range of motion, and complications. All ranges of motions improved postoperatively, although only extension, radial deviation, and supination increased to statistical significance. Average DASH scores improved from 46 points preoperatively to 32.1 points at 1-year follow-up, and continued to improve to 22.4 points at 2-year follow-up. The latter did not reach statistical significance given the lack of power (n = 8). Early volar dislocations were seen in 3 patients (14%) who were found to have highly active rheumatoid disease and marked joint laxity at the time of initial surgery. Of the dislocations only 1 patient was converted to wrist fusion for persistent instability, after failed revision of carpal baseplate and polyethylene exchange, followed by failed open reduction, capsule imbrication, and external fixation, followed by failed closed reduction and splinting. The patient suffered traumatic dislocations before her second and third revision surgeries. No cases demonstrated radiographic evidence of distal component loosening. Nonprogressive subsidence of the radial component was seen in 1 patient.
In a more recent study, Ward and colleagues reviewed the long-term results associated with the Universal 1 TWA in 25 rheumatoid wrists (21 patients) with a minimum follow-up of 5 years. Clinical and radiographic review was available for 19 wrists (15 patients) with an average follow-up of 7.3 years (range 5–10.8 years). At a minimum of 5 years, the Universal 1 wrist prosthesis had a high failure rate of 50% (10 of 20 wrists). This failure rate includes revision of 1 patient who died before the minimum 5-year follow-up. In this patient, conversion to wrist fusion was performed secondary to recurrent implant instability at 2.2 years after the index procedure. Carpal component loosening was the most common cause of revisions, occurring in 45% of wrists (9 wrists in 8 patients). The mean time to revision surgery for carpal component loosening was 5.5 years (range 2.7–9.34 years). Of the 9 wrists revised because of carpal loosening, only 5 had radiographic evidence of intercarpal union. An additional 2 patients had radiographic evidence of carpal component subsidence. Revision surgery revealed polyethylene wear, metallosis, and carpal component loosening in all wrists. There was no evidence of gross radial component loosening in any of these wrists.
At the time of latest follow-up, 10 wrists in 8 patients remained functional. Despite failures, those with a stable prosthesis did maintain a functional range of motion and had improvement in patient-reported outcome measures. In this group of patients, average DASH scores improved from 62 points preoperatively to 40 points postoperatively. All ranges of motion increased postoperatively. There was no significant change in wrist flexion or extension, radial deviation, ulnar deviation, and forearm supination. At the time of latest follow-up, intercarpal fusion was seen in all 10 functional wrists.
In this series, patients with carpal component loosening were less likely to have radiographic evidence of intercarpal fusion than those who had stable prostheses, although this difference did not reach significance. Also in this series, the failure rate was higher than previously reported for this prosthesis. Implant survival rates at 5 and 7 years for the original prosthetic components were 75% and 60%, respectively.
In an attempt to better understand the mechanical behavior of implant instability and the high rate of dislocations associated with the Universal 1 TWA, Grosland and colleagues created a finite element analysis comparing toroidal and ellipsoidal carpal articulations. Using computer modeling and laboratory experiments, the elliptical distal articular surface provided consistent congruity and centralization of the contact area on the polyethylene over the entire range of prosthetic motion. This approach resulted in low stress and wear on the polyethylene and better articular stability without creating a fully constrained joint. The toroidal distal articulation maintained true congruency only during radial and ulnar deviations with the prosthesis in a neutral position.
Pioneered by Adams, the next generation of the Universal TWA prosthesis made use of an ellipsoidal rather than a toroidal articulation. Improving on fixation, a beaded porous coating was applied to both radial and carpal components to allow for fixation by osseous integration. Radial and ulnar variable-angle screws allow compression between the carpal baseplate and the remaining carpus.
In a recent study, Ferreres and colleagues reviewed midterm results using the Universal 2 TWA implant in 22 wrists (22 patients). Clinical and radiographic review was available for 21 wrists (21 patients) with average follow-up of 5.5 years (range 3–9 years). Although 2 wrists received the Universal 1 TWA implant and the remaining wrists the Universal 2 implant, the investigators did not exclude nor differentiate Universal 1 results from the final review. Fifteen patients had rheumatoid arthritis. Two patients had wrist destruction from grade IV Kienböck disease. The remaining 5 patients had nonrheumatoid inflammatory arthritis (lupus, psoriasis, undifferentiated spondyloarthropathy), chondrocalcinosis, and arthrosis. The PRWE score averaged 24 points, with an average of 11 points out of 50 for functional evaluation and 12 points out of 50 for pain. In terms of patient satisfaction, 10 patients were very satisfied with the procedure and 10 were satisfied. Only postoperative ranges of motion were available for review. The average range of flexion arc was 68°, with an average of 26° of extension and 42° of flexion. The average range of radial and ulnar deviation was 26°, with an average of 1° of radial deviation and 26° of ulnar deviation. Only average flexion and extension were within functional range of motion as described by Palmer. Pain during activities of daily living was absent or slight in 17 patients. Where discomfort was present, it was attributable to neighboring diseased joints.
At the time of latest follow-up, 18 wrists showed no signs of radiolucency or loosening around the carpal component. Nonprogressive osteolysis was observed in 2 patients around the tip of the radial screw into the index finger metacarpal. Progressive carpal component subsidence and osteolysis was seen in 1 patient at 1 and 7 years’ follow-up. The subsidence resulted in progressive radial inclination of the carpal component, yet the patient remained asymptomatic at the time of latest follow-up. Stress shielding of the radial styloid was also observed in 2 patients, with no signs of radial component loosening. In this series, there were no dislocations or surgical revisions of the components.
ReMotion TWA
The ReMotion TWA (SBI, Morrisville, PA) implant was designed to allow some degree of intercarpal rotation. The polyethylene ellipse snaps securely on a ball on the proximal side of the carpal plate, allowing 10° of relative rotation. Acting like a dampener, it avoids torque transmission to the carpal component. The relative axial motion between the polyethylene and the carpal plate is believed to preserve the complex “dart thrower’s” motion of the wrist. Distal fixation is achieved by a central press-fit stem into the capitate and flanking variable-angle screws that allow compression between the carpal plate and remaining carpus. The radial component has 10° of palmar tilt and radioulnar inclination to approximate the native wrist ( Fig. 1 ). Conferring further stability, bony resections are minimal, allowing preservation of the important ligaments of the wrist. Preservation of the rim of the distal radius is a key feature of the ReMotion implant.
