Silicone implant arthroplasty of the metacarpophalangeal and proximal interphalangeal joints has dominated joint replacement in the fingers for decades. Despite their generally positive clinical results, issues remain with Silicone rubber (silastic) implants, the most important being fracture. Although there are reported fracture rates of up to 65% at 15 years, by the time of failure a fibrous pseudojoint has formed, so that replacement of the implant is not always necessary.
The three main designs are the Swanson, the Sutter, and the NeuFlex. For Swanson and Sutter MCP joint prostheses, fracture generally occurs at the junction of the distal stem and the body of the implant. For the NeuFlex, it can break there, but most commonly fractures across the hinge. For Sutter MCP joint prostheses in particular, it has been shown that fracture is initiated on the dorsal aspect of the implant and travels in a palmar direction, likely due to the dominance of subluxing forces in rheumatoid metacarpophalangeal joints. While it has been claimed for many years that hard bearing two-piece implants more akin to the anatomy of finger joints will supersede the nonanatomical single-piece silicone designs, this has not been achieved clinically, despite many clever designs being proposed. It may be that improvements to current silicone designs will be an important next step for finger joint replacement.
Key wordsmetacarpophalangeal – proximal interphalangeal – prosthesis – silicone – Swanson – Sutter – NeuFlex – fracture – failure – implant
7 Failure Analysis of Silicone Implants in Metacarpophalangeal and Proximal Interphalangeal Joints
Single-piece silicone implants are the commonest metacarpophalangeal (MCP) and proximal interphalangeal (PIP) joint replacements 1 ; MCP joints tend to be replaced far more often than the PIP joints 1 and there is more data on their outcomes. While a number of different designs of silicone implant have been proposed, all share a central hinge section intended to keep the ends of the finger bones separated and two stems which fit into holes reamed in the bones on each side of the joint (Fig. 7.1). The most commonly implanted design is the Swanson prosthesis. 2 This implant has been supplied with titanium grommets which slip over the stems and are intended to protect the relatively soft silicone material from damage by bone although are not typically inserted. The second main design is the Sutter which is also known as the Avanta. 3 More recently, the NeuFlex design from De Puy has been available, with pre-flexed stems, 4 in contrast to the colinear stems of the Swanson and Sutter implants (Fig. 7.1).
Despite the successes of single-piece silicone implants, a key concern is implant fracture. 2 , 3 , 4 , 5 It has been reported that approximately two-thirds of silicone finger implants fracture after 14 to 17 years in the body. 6 , 7 It is also known that fracture rates increase with duration in vivo; radiographic fracture rates of 7% at 5 years, 42% at 10 years, and 65% at 15 years have been reported. 8 Nevertheless, once these implants fracture, a fibrous pseudojoint is formed, providing some stability so that a revision operation may not be necessary. 9 Boe et al 8 reported on 325 silicone MCP joint implants. Of these 325 implants, 22 were revised; the majority of these (14/22) were fractured. Of 214 MCP joint implants with radiographic follow-up greater than 2 years, 80 (37%) were fractured. The authors stated that implant fracture was significantly associated with increasing age, female gender, increased body mass index, osteoarthritis, posttraumatic arthritis, and diabetes mellitus. 8 Although less information is available for silicone PIP joint implants, there are common features with silicone MCP joint implants. For Swanson PIP joint implants, a high incidence of in vivo fracture was noted but as for the MCP joints these do not necessarily lead to revision. 10 At a mean follow-up of 10 years, 21 of 38 (55%) Swanson PIP joint implants had fractured. Four implants from this cohort were removed, three for implant fracture. This high rate of in vivo fracture is similar to that seen in the MCP joint implants and this commonality should be recognized. For NeuFlex PIP joint implants, at a mean follow-up of 39 months, a 10% fracture rate was reported. 11 Herren et al stated that 44% of silicone PIP implant revision operations were associated with implant breakage. 12 The authors also reviewed 70 published papers and stated that the most common reason for revision (in 40% of cases) was implant breakage. They did not define what type of silicone implant they used. Fracture rates are summarized in Table 7.1.
7.2 Failure Mechanisms
For Swanson and Sutter MCP joint prostheses, fracture generally occurs at the junction of the distal stem and the body of the implant, 3 , 13 but can occur at the proximal stem and body. In contrast a recent study has shown that the NeuFlex implants appear to fracture primarily across the hinge, 4 but can occur at the stem-body junction or both (Fig. 7.2).
It was presumed previously that once an implant fracture occurred the stresses were dissipated and further fracture would not occur but this recent study 4 challenges that idea. The mechanism for this is not understood. If the prevalence of fractures could be reduced it would be expected that silicone finger implants would function better for longer and thus provide greater patient benefit. Before cases of fracture can be reduced, the origins of fracture need to be understood. An analysis of 12 explanted Sutter MCP joint silicone implants 14 determined that the fracture started on the dorsal aspect of the distal stem and travelled in a palmar direction. This fracture initiation point may have been caused by bone impingement from the cortical bone of the proximal phalanx, due to the subluxing forces which dominate in rheumatoid MCP joints or simply the direction of the forces (typically into ulnar deviation and flexion) causing the greatest tearing moment at the dorsoradial corner of the proximal stem insertion into the implant body. In contrast, there was relatively little damage on the palmar aspect of the stems of the implants. In addition, classic features of fatigue fractures were seen on the explants. 14 These features included “radial marks” which indicated the origin of fracture on the dorsal aspect of the distal stem, toward the radial side (Fig. 7.3). Next, “beach marks” showed fatigue crack propagation in an arc from radial to ulnar and from dorsal to palmar (Fig. 7.3). Lastly, a “shear lip” on the palmar aspect indicated the final area of fracture of the silicone implant (Fig. 7.3). Fatigue fractures are the typical failure mode with silicone finger implants.
Possible causes of crack initiation include poor surgical handling damaging the implant at operation and damage from sharp bone edges. Poor handling rarely leads to overt implant tears but there may be microscopic damage later leading to failure. Sharp bone edges can clearly damage the implant surfaces. Because of these worries grommets were introduced but did not seem to affect failure rates and are largely not used. 6 Eccentric loading appears to be the commonest cause of crack progression as shown in Fig. 7.3. These forces will eventually tear the dorsoulnar corner of the insertion of the distal stem into the body (or the hinge for NeuFlex implants). The tears then propagate in response to the continuing deforming forces. Plastic deformation occurs with prolonged ulnar deviation as shown in intact explants and in simulated jigs. 15