➢ Proximal interphalangeal joint arthroplasty is indicated primarily for pain reduction in patients with inflammatory, noninflammatory, or traumatic arthritides. Despite an initial improvement in the active range of motion of the affected joint, long-term postoperative follow-up frequently demonstrates no significant change in the total arc of joint motion.
➢ Selection of the techniques and implants of proximal interphalangeal joint arthroplasty should be led by a patient-specific approach; however, pyrocarbon implants have been associated with a higher rate of complications and the subsequent need for revision compared with silicone implants.
➢ Primary arthrodesis is generally preferred over implant arthroplasty for the index finger proximal interphalangeal joint because of increased valgus stress on the digit in comparison with the middle, ring, or small fingers and the eventual instability and failure of the arthroplasty.
Small joint arthroplasty in the hand is undertaken with the goal of pain amelioration and optimization of the available active arc of motion in patients with noninflammatory, inflammatory, or traumatic arthritides. With this goal in mind, numerous methods of joint reconstruction utilizing constrained, semi-constrained, and nonconstrained implants have been reported. In those instances in which the arthroplasty constraint is minimal, competent collateral ligaments are required to ensure a stable and functional joint.
However, with any implant-based procedure, arthroplasty materials are prone to wear, fatigue, and eventual failure. Particulate matter, ranging in size from angstroms to millimeters, incites subsequent systemic and local responses. Additionally, chronic osteopenia from inflammatory arthritis or disuse can create very thin and fragile phalangeal cortical bone, resulting in potential implant malposition, cortical breach, or fracture. In comparison with large joint arthroplasties, the small joints of the fingers are subjected to repetitive varus and valgus stresses in addition to sagittal flexion and extension loads.
Proximal interphalangeal joint arthroplasty is a demanding procedure. It poses challenges from the standpoints of appropriate clinical evaluation and patient selection and of understanding the optimization of the appropriate implant and operative technique. This review details the various types of proximal interphalangeal joint arthroplasty, focusing on the available evidence for indications, outcomes, and the need for eventual revision.
History of Proximal Interphalangeal Joint Arthroplasty
Small joint arthroplasty of the hand was first described more than 100 years ago, with one of the earliest reports describing resection and interposition arthroplasty1. Although the proximal interphalangeal joint is not a true hinged joint, the anatomic constraint of this joint is maintained by an intimate relationship among the collateral ligaments, the joint capsule, the volar plate, and the central slip of the extensor apparatus. The proximal interphalangeal joint has a range of motion of 110°, with the functional range of motion from 36° to 86°2,3. The ideal joint replacement would recreate the native articular anatomy while providing pain control and stable functional range of motion.
In the 1940s, Burman and Amrahamson introduced the concept of cap arthroplasties, using Vitallium and Lucite to recreate the joint surface4. Although this was successful in resurfacing the eburnated joint surface, lateral instability remained a major pitfall of this technique. In response to this problem, Brannon and Klein developed a hinged metal prosthesis to combat the problem of a nonconstrained implant that was placed in coronal stress5. Flatt6 further optimized the procedure by addressing the rotational instability inherent in the spherical stem design by creating a two-pronged press-fit implant. This arthroplasty was impacted into the medullary canal with the expectation of osseous ingrowth. Although patient outcomes were satisfactory, this design was also fraught with complications such as infection and implant failure6. In 1950, Leibolt described a joint resection arthroplasty with the interposition of fascia7. Carroll and Taber later reported on the first large series of resection arthroplasties, but that series did not include any tissue or prosthetic interposition8.
Largely unsuccessful when these implants were first introduced, proximal interphalangeal joint implant arthroplasties were considerably advanced through the development of the constrained, but pliable, Swanson silicone arthroplasty9. Swanson’s initial clinical series in 1972 showed a 35° increase in the arc of motion from preoperatively to the postoperative follow-up of up to five years. Unlike the metals used in previous implants, silicone was heat stable and showed excellent durability, and patients reported excellent outcomes in terms of pain relief and improved motion9. However, silicone arthroplasty was complicated by implant breakage, loosening, silicone synovitis, and joint dislocation10,11.
Again, in an attempt to address the shortcomings of the previous arthroplasty design, resurfacing implant materials such as pyrocarbon (also known as pyrolytic carbon) were introduced in the United States as a way to improve on silicone arthroplasty. These implants were designed with minimal constraint, relying on intact collateral ligaments, and had a modulus of elasticity similar to bone to decrease the risk of cortical breach or subsidence12. Manufacturers have subsequently created metal-on-polyethylene implants with a cementing option.
Operative Indications for Proximal Interphalangeal Joint Arthroplasty
Arthritis of the proximal interphalangeal joint is common, and this joint is the third most common site for osteoarthritis of the hand13-16 (Fig. 1). In addition to primary osteoarthritis, posttraumatic arthritis, post-infectious arthritis, or inflammatory arthritis (rheumatoid, systemic lupus erythematous, psoriatic, or gouty) can affect the proximal interphalangeal joints. Patients can present with pain, stiffness, and instability that can be quite functionally limiting. Pain or instability of the index or middle finger proximal interphalangeal joint results in decreased fine and key pinch. Instability or decreased arc of motion in the ring finger and small finger proximal interphalangeal joints can markedly decrease grip strength. Proximal interphalangeal joint arthroplasty offers a reasonable alternative to joint arthrodesis. However, relative contraindications include young and physically active patients, patients seeking improvements in range of motion, and patients with active infection, poor soft-tissue coverage, and loss of extensor or flexor function.
The first-line treatment for proximal interphalangeal joint arthritis is nonoperative treatment with nonsteroidal anti-inflammatory drugs, physical therapy, corticosteroid injections, and bracing. In patients with persistent pain and dysfunction who have failed conservative treatment, proximal interphalangeal joint arthroplasty may be a reasonable alternative to joint arthrodesis.
The primary indication for arthroplasty is pain relief. It should be emphasized that most clinical series demonstrate minimal improvement in the overall total arc of motion compared with preoperative values. However, with the removal of osteophytes and the improvement of joint articulation, patients have a shift in the arc of motion, with improved extension, which leads to improvement in motion and use of the hand for activities of daily living. Overall, patients have shown consistently high satisfaction with this procedure and would not only choose it again but would also recommend it to others17. The index finger proximal interphalangeal joint is uniquely subjected to increased lateral stress compared with the other fingers, which may contribute to a potentially higher arthroplasty failure rate. Therefore, the patient’s goals and activity levels should be carefully considered against a joint arthrodesis for pain relief.
Regardless of the implant chosen, the operative techniques of implantation are similar, with the most commonly reported approaches being either dorsal or volar. The dorsal approach should be considered when extensive dorsal osteophytes require resection, as they may otherwise be difficult to remove or may result in damage to the central extensor slip insertion. Additionally, with a dorsal approach, the central extensor slip must be either longitudinally split or retracted with use of a distally based extensor flap (Chamay extensor approach18), which introduces the potential for rupture of the extensor tendon and extensor lag and risks boutonniere deformity18. In comparison, the volar approach requires release of the volar plate and collateral ligaments to hyperextend the joint. Although this provides excellent visualization and allows for immediate postoperative motion due to an intact extensor tendon, care is required to completely visualize and to release, as necessary, the neurovascular bundles to prevent postoperative neurapraxia or vascular injury.
Outcomes of Implant Arthroplasty
Silicone arthroplasty functions both as a constrained arthroplasty and as a spacer, allowing the formation of a hypertrophied joint capsule (Fig. 2). This capsule has been theorized to add to the stability of the joint in the coronal plane, especially in the setting of incompetent collateral ligaments17,19. The intramedullary stem is tapered to provide stability while still promoting flexion and extension9,11,16,20-25. Swanson et al. reported on 148 proximal interphalangeal joints treated with silicone arthroplasty for rheumatoid arthritis, osteoarthritis, or posttraumatic arthritis. The follow-up range was six months to five years, and a 35° increase in arc of motion was reported. The best improvement in total arc of motion occurred in patients with rheumatoid arthritis. In this cohort, the average range of motion was 4° of extension lag to 70° of flexion. For posttraumatic arthritis and osteoarthritis, the mean arc of motion was 9° of extension lag to 59° of flexion9. A subsequent study was performed with a cohort of 424 patients demonstrating a 10° improvement in range of motion and 98% pain relief after silicone arthroplasty26.
Similar outcomes have been reported by others, with pain relief, increase in grip strength, and improvement in extensor lag22. At an average follow-up duration of 3.4 years, in sixty-seven (97%) of sixty-nine silicone arthroplasties, Lin et al. showed that patients had a decrease in pain, an improved ability to pick up and grasp objects, no improvement in total range of motion, and an improvement in the extension deficit from 17° preoperatively to 8° postoperatively16. Similarly, in their series of seventy proximal interphalangeal joint arthroplasties, Takigawa et al. demonstrated that patients had improvement in extension deficit but no improvement in total arc of motion25. In a series of 138 joints with an average follow-up duration of 5.8 years, Ashworth et al. showed that their cohort of patients with rheumatoid arthritis had excellent pain relief but had a decrease in the average arc of motion (38° preoperatively to 29° postoperatively)27.
Despite the benefits of silicone proximal interphalangeal joint arthroplasty, complications of the procedure, including implant fracture, particulate synovitis, and lateral instability, have been well documented in the literature. In a 1972 study on 222 proximal interphalangeal joint arthroplasties9, Swanson reported five (2.25%) that failed because of implant fractures, whereas in a 1985 study, Swanson et al. showed a 5% implant fracture rate26. Takigawa et al. reported a 15% fracture rate and a suspected fracture in an additional sixteen (23%) of seventy replacements25.
Radiographic changes and concerns for particulate synovitis are also well described. In a large meta-analysis of 15,556 small joints, Foliart showed the rate of osseous change to be near 4% and the rate of particulate synovitis to be 0.06%11. Pellegrini and Burton reported that 35% of their silicone arthroplasties had erosion at two years, 20% had intramedullary absorption at four years, four implant dislocations had occurred, 50% of silicone arthroplasties had subsidence, and cystic changes had occurred in thirty-two (46%) of the seventy joints28.
Suboptimal lateral stability has also complicated silicone arthroplasty. Minamikawa et al. showed that silicone spacers had significantly more lateral angulation when subject to torque loading compared with native joints in a cadaver model29. Additionally, despite the stability afforded to the joint through the capsular formation around the silicone spacer, loss of lateral joint stability can occur following fracture of the implant17,19.
Unlike the constrained silicone replacements, pyrocarbon arthroplasties attempt to maintain the congruency of the joint through semi-constrained components (Fig. 3). Due to the congruent articulating nature of the components, there is some inherent stability through the arc of motion. However, the replacements are largely dependent on intact collateral ligaments for lateral stability. Pyrocarbon is a material with a graphite core coated with pure carbon that has been shown to be biocompatible with human tissue; it has good wear characteristics and is less tissue-reactive than silicone and stainless steel30. In addition, it has an elastic modulus similar to bone31. To counter the lateral instability associated with silicone spacers, the insertion of pyrocarbon resurfacing implants relies on maintaining intact native collateral ligaments, while the material triggers osseous ingrowth and improved implant stability over time (Fig. 3)32,33.
There have been many studies showing the efficacy of pyrocarbon implants. Bravo et al. performed a retrospective review on fifty pyrocarbon resurfacing arthroplasties for osteoarthritis, posttraumatic arthritis, or rheumatoid arthritis with an average follow-up duration of thirty-seven months12. They demonstrated an increase in the arc of motion from 42° to 49° for the osteoarthritis group, 34° to 39° for the posttraumatic arthritis group, and 45° to 52° for the rheumatoid arthritis group, although the increase in range of motion was not significant for any of the three groups. Bravo et al. also identified improvements in grip and pinch strength from 21 to 28 kg for the osteoarthritis group and 24 to 31 kg for the posttraumatic arthritis group. There was an 80% overall rate of patient satisfaction and improved pain relief12. Ono et al. obtained long-term follow-up on thirteen pyrocarbon arthroplasties at forty-four months, showing improved pain relief and patient satisfaction without sacrificing the active arc of motion34. Additionally, the authors utilized the Michigan Hand Outcomes Questionnaire35, and all six domains of the questionnaire improved greatly with long-term follow-up. Meier et al. reported on twenty-five pyrocarbon proximal interphalangeal replacements for posttraumatic and idiopathic arthritis, with favorable results; at an average follow-up duration of fifteen months, there was an 80% patient satisfaction rate with an average range of motion of 50°36. Wijk et al. examined fifty-three pyrocarbon implantations performed over a four-year period and also demonstrated a significant improvement in terms of pain and patient satisfaction for patients with osteoarthritis, rheumatoid arthritis, or posttraumatic arthritis. With use of the visual analog scale for pain, they found an improvement from 3.1 points preoperatively to 0.4 point postoperatively for pain at rest and an improvement from 6.2 points preoperatively to 2.0 points postoperatively for pain with activity37.
However, in much the same way as silicone implants, pyrocarbon implants are associated with osseous subsidence, fracture, and instability. Sweets and Stern highlighted many of the complications associated with pyrocarbon proximal interphalangeal joint replacements; in their study, there were a total of sixty complications for twenty-eight joint replacements: five implant dislocations, four subluxations, three implant fractures, two wound-healing problems, and twenty joint contractures, all with significantly decreased arc of motion; fifteen cases of loosening confirmed by radiography; and eleven cases of squeaking15. The average range of motion for the entire sample was 57° preoperatively, 67° at their best measurement at 11.6 months postoperatively, and 31° at the time of the latest follow-up of more than two years. In addition, the authors found that, at the time of follow-up, pain in the involved digit was greater than pain in the uninvolved digits, and they made the recommendation of abandoning the pyrocarbon implant for proximal interphalangeal arthroplasty, given the nature and frequency of complications. The complications highlighted in their report have been corroborated in several studies.
Chung et al. reported dislocations from instability in three (14%) of twenty-one pyrocarbon proximal interphalangeal implants38, and Meier et al. similarly reported two dislocations that required arthrodesis36. Mashhadi et al. reported that, of their twenty-four patients, five (20.8%) had a flexion contracture, four (16.6%) reported joint squeaking, and three (12.5%) with stiffness and joint contracture underwent tenoarthrolysis3. In a series of fifty-three joints, Wijk et al. reported that seven joints (13%) required a reoperation and concluded that pyrocarbon replacements are successful for pain relief but not for improved joint range of motion37.
In addition to pyrocarbon arthroplasty, a cobalt-chromium implant is available with a similar semi-constrained design. In comparison with the pyrocarbon implant, the cobalt-chromium implant has a two-piece bicondylar design with a distal polyethylene component, is commonly referred to as a surface replacement arthroplasty, and is typically cemented at the proximal and distal components. Stoecklein et al. reported on six proximal interphalangeal surface replacement arthroplasties in five patients utilizing a volar approach39. At an average follow-up duration of thirty-five months, the preoperative arc of motion of 33° had improved significantly (p = 0.03) to 60°, in addition to improvements in grip and key pinch strength. Murray reported on sixty-seven surface replacement arthroplasties with an average follow-up duration of 8.8 years, resulting in an average arc of motion of 40° and a failure rate of 12%40. Similarly, Jennings and Livingstone retrospectively reported on forty-three proximal interphalangeal surface replacement arthroplasties with an average follow-up duration of thirty-seven months41. In comparison with Stoecklein’s findings, motion did not significantly change from preoperatively (57°) to postoperatively (58°), and eleven (26%) of the arthroplasties failed, requiring major revision. Importantly, patients did not universally see an improvement in joint pain postoperatively: 77% had decreased pain, 7% were unchanged, and 16% had worsened pain. Overall, 60% of patients rated the outcome as “very satisfactory”41. Cementing the components has been proposed to decrease the risk of implant subsidence and loosening39,41. Johnstone et al. retrospectively compared cemented and uncemented surface replacement proximal interphalangeal joint arthroplasties in forty-eight fingers42. The average duration of follow-up was four years for the uncemented group and six years for the cemented group. Despite no significant differences between the two cohorts for the postoperative decline in the pain score or improvement of the arc of motion (average, 13°), there was a significantly higher (p < 0.001) rate of loosening and subsidence in the uncemented group (68%) compared with the cemented group (4%), resulting in a revision rate of 26% for the uncemented group and 8% for the cemented group42. However, opponents of cementing have reported the increased rate of bone loss from stress shielding, as well as the larger nidus for potential infection21,40,41.
Postoperative Rehabilitation Protocol
As with most procedures, the postoperative protocol can vary from surgeon to surgeon. However, with regard to proximal interphalangeal joint arthroplasty, postoperative therapy is governed by the operative approach and implant technique with the goal of facilitating motion while protecting the repair of the extensor mechanism and the collateral ligament complex. Most frequently, the proximal interphalangeal joint is splinted postoperatively in full extension, both to allow the soft tissues to heal and to prevent early loss of extension. At the one-week postoperative visit, hand therapy is started with the short-arc protocol, based on weekly or gradual increases of 10° to 15° in the allowed arc of motion, again to protect from the loss of terminal extension37. If the patient starts to develop an extensor lag, static extension splints are used for prolonged periods. Patients treated with a daytime dynamic splint and a nighttime static extension splint by a certified hand therapist have been shown to have good range of motion postoperatively, avoiding range of motion deficits such as hyperextension, extension lag, or a flexion contracture12. Of note, for those patients with a volar arthroplasty exposure, immediate postoperative flexion and extension may be considered appropriate because of the lack of an extensor repair. Regardless of the protocol chosen, patients usually return to full function between six and twelve weeks postoperatively.
Comparison of Silicone Arthroplasty and Pyrocarbon Arthroplasty
There has not been a consensus in the literature as to which implant is superior, as very few studies have compared the results of silicone replacements with those of pyrocarbon replacements for proximal interphalangeal joint arthroplasty. In a comprehensive review, Chan et al. reported on 1882 proximal interphalangeal joint replacements (1430 silicone and 452 pyrocarbon) for all types of arthritides, including idiopathic, inflammatory, and posttraumatic. They showed that pain relief (76% for silicone arthroplasty and 64% for pyrocarbon arthroplasty), joint mobility (an 8.2° improvement for silicone arthroplasty and an 8° improvement for pyrocarbon arthroplasty), and grip strength were equivalent for both procedures. Although both procedures offered pain relief and a similar arc of motion, the pyrocarbon group had more complications, requiring approximately four times more frequent revision or salvage. Of the 452 relevant pyrocarbon replacements, Chan et al. identified thirty-five joints (7.7%) with nonoperative complications, fifty-nine joints (13%) that required a revision surgery, and thirty-four joints (7.5%) that underwent a salvage procedure such as arthrodesis43.
Squitieri and Chung performed a systematic review comparing the two different implants in patients who underwent arthroplasties for posttraumatic arthritis44. Like Chan et al., they found similar range of motion and complication rates associated with pyrocarbon replacements. They reported on 194 silicone arthroplasties and eighteen pyrocarbon arthroplasties. The data demonstrated an active postoperative arc of motion of 44° for silicone replacements and 43° for pyrocarbon replacements. Pyrocarbon replacements had a 38% complication rate requiring a reoperation, and silicone replacements had an 18% reoperation rate44.
Daecke et al. compared silicone replacements with pyrocarbon and titanium replacements. With a total of sixty-two proximal interphalangeal joint replacements, eighteen fingers underwent silicone arthroplasty and eighteen fingers underwent pyrocarbon arthroplasty. Daecke et al. found no significant difference in the postoperative average arc of motion, with 56° for the silicone arthroplasty and 61° for the pyrocarbon arthroplasty, and they also concluded that both implants provided adequate pain relief. However, as in the other studies, pyrocarbon replacements were associated with a higher rate of complications. In the pyrocarbon group, seven implants (39%) were explanted because of aseptic loosening, restricted range of motion, or dislocation compared with two implants (11%) in the silicone group45.
The indications for proximal interphalangeal joint arthrodesis are the same as for arthroplasty, with the goal of a pain-free joint. Arthrodesis is an excellent option for a painful, arthritic joint in the setting of failed arthroplasties or when arthroplasty is not an option secondary to substantial lateral instability or abnormal patient anatomy. The optimal flexion angle for arthrodesis of the proximal interphalangeal joints of the fingers varies by the finger being fused: 40° for the index finger, 45° for the middle finger, 50° for the ring finger, and 55° for the small finger46.
There are many described techniques of proximal interphalangeal joint arthrodesis, including Kirschner wire fixation, tension banding, plate fixation, and compression screws, with similar efficacy rates in the literature47-50. The simplest form of arthrodesis involves placing multiple Kirschner wires across the joint in the position of function. Ijsselstein et al. showed a less successful arthrodesis rate for Kirschner wire fixation (85%) compared with tension band wiring (95%) and showed that Kirschner wire fixation was also associated with a higher risk of complications, such as infection48. Jones and Stern reported on the results of 290 joints arthrodesed via tension banding and found a fusion rate of 97% with minimal complications47.
Despite the benefits of pain amelioration, loss of motion at the proximal interphalangeal joint can impact the function and strength of the hand51. Woodworth et al. evaluated the impact of fused proximal interphalangeal joints on overall hand function and found that fusions in 40° of flexion did not impact activities on the Jebsen test or with activities of daily living. For the most part, patients were able to successfully adapt to a fused joint. However, they did demonstrate an impact on activities requiring fine motor skills such as buttoning and writing52. Domalain et al. looked specifically at motion and pinch restriction with a fused index finger proximal interphalangeal joint. They found that this fusion reduced the maximal amount of pinch aperture; however, a proximal interphalangeal joint fused in the natural angle for pinching did not affect the motion through other joints53.
Conclusions and Clinical Care Recommendations
Proximal interphalangeal joint arthroplasty can provide good pain relief without a clinical increase in the active arc of motion in patients with posttraumatic, inflammatory, or degenerative arthritis. However, the improved position of the arc of motion and decreased pain may provide improved function for fine tasks and activities of daily living. Surgical techniques can include constrained or semi-constrained implants.
Silicone replacement arthroplasty has consistently been shown to reduce pain in patients with painful joints secondary to arthritis of various etiologies. There is established long-term follow-up in the literature showing the benefits of silicone arthroplasty, with relatively few complications compared with semi-constrained implants like pyrocarbon17,25. With use of the available literature and referenced studies (Table I), we recommend a level-of-care rating of B for silicone replacement arthroplasty (Table II).
Semi-constrained replacements such as pyrocarbon became popular in clinical practice as an alternative to silicone replacement arthroplasty. Early results with short-term follow-up were promising, showing similar pain relief and patient satisfaction compared with silicone replacements3. With encouraging results in the literature, there was initial enthusiasm for surface replacement arthroplasty. However, a growing body of literature became concerning, with recent evidence for increased complications in long-term studies evaluating semi-constrained pyrocarbon resurfacing implants15,43,54,55. In addition to concerns with complications, there is no clear evidence in the literature that pyrocarbon arthroplasty has superior results compared with silicone arthroplasty. At best, results are equivocal43. Given the evidence in the literature, we caution surgeons on using pyrocarbon replacement arthroplasty for proximal interphalangeal joint osteoarthritis and have assigned it a clinical care recommendation level of C (Table II).
Despite this, complications have been noted in both silicone and semi-constrained implants. The implant and surgical exposure should be based on the individual patient needs and on the degree of osteophytosis and joint instability. In addition, postoperative hand therapy has been demonstrated to improve the functional results of the arthroplasty, and a short arc of motion may be beneficial following repair of the extensor mechanism to prevent postoperative extensor lag.
With the available evidence, we recommend silicone arthroplasty as the treatment of choice for patients with proximal interphalangeal joint arthritis. In accordance with other systematic reviews, pyrocarbon replacements have not been shown to be superior to silicone, and recent evidence is concerning with respect to an increased complications rate.
Source of Funding: There was no source of external funding for this study.
Investigation performed at the Department of Orthopaedics and Sports Medicine, University of Washington Medical Center, Seattle, Washington
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated