|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Synovitis of the wrist induced by particles of silicone debris is a destructive inflammatory process. Many silicone-rubber carpal implants remain in place, and there are few reports regarding the treatment of this condition. The purpose of the present study was to examine the results of treatment of synovitis induced by particles of silicone debris. Twenty-eight patients were identified, with use of computerized indexing, as having been evaluated for silicone-induced synovitis between 1972 and 1992. Seventeen of the twenty-eight patients were included in the study. At the time of the latest follow-up, twelve of the seventeen patients had pain, thirteen of the fourteen patients for whom radiographs were available had evidence of osteolysis typical of that associated with debris-induced synovitis, and eight of the seventeen patients reported difficulty with activities of daily living because of problems with the wrist. Seven patients had been treated non-operatively, and ten had been treated operatively. With the small number of patients available for study, we could not detect a significant difference between the two groups with respect to pain, perceived limitation of motion, difficulty with activities of daily living, grip strength, or the total range of motion of the wrist. There was no significant difference between the two groups with regard to the age at the time of the initial procedure, the time to the diagnosis of the synovitis, and the duration of follow-up after treatment. There was no clear advantage to removal of the implant and débridement with or without arthrodesis of the wrist or other reconstructive procedures. We recommend caution when a reconstructive or salvage procedure in the wrist is selected for a patient who has synovitis induced by particles of silicone debris.
The element silica is most commonly found in the form of silicone dioxide or sand. In addition to silicone dioxide, there are more than 60,000 chemical compounds that contain silica15. Polydimethylsiloxane, a chain of carbon, oxygen, hydrogen, and silica, is the most common silicone polymer produced15. The length of the polymetric chain determines if the compound is in a liquid phase15,62,69. Crosslinking of polydimethylsiloxane results in the production of silicone rubber15,54. Silastic is a proprietary name and registered trademark of the medical grade of silicone rubber that was first produced by Dow Corning (Midland, Michigan)17. Silicone chemicals were first prepared15 by Kipping in 1907, and silicone rubber was first used medically, as a lubricant for glass syringes, in World War II13,73. In 1963, Swanson developed a silicone implant to cap the distal end of the tibia after below-the-knee amputation in children62. Between 1963 and 1973, he also developed silicone implants for replacement of the metacarpophalangeal joint, carpal trapezium, scaphoid, lunate, radial head, wrist joint, and elbow joint as well as for use as a cap for the distal end of the ulna53,62-68,70. Although silicone initially was thought to be biologically inert, it was subsequently recognized that particles of debris from wear of silicone implants can evoke an intensive inflammatory response2,6,7,17,49,61,71,74,76. The clinical features of synovitis of the wrist joint induced by particles of silicone debris include pain with motion, joint tenderness, dramatic loss of wrist motion, and soft-tissue swelling58. Radiographs may show well defined marginal erosions with subchondral cysts on the osseous surfaces adjacent to the implant1,61, whereas bone mineralization is typically normal in the absence of rheumatoid arthritis or osteoporosis. The purpose of the present study was to evaluate the long-term results of treatment of synovitis of the wrist induced by particles of silicone debris. In addition, we compared the outcomes of operative and non-operative management.
Twenty-eight patients were identified, with use of computerized indexing, as having been managed at our institution, between 1972 and 1992, for synovitis of the wrist induced by particles of silicone debris. We reviewed the records and original radiographs of these patients and included only those who had disease attributable to a carpal bone implant, radiographic evidence of periprosthetic destruction, and documentation of clinical synovitis of the wrist in the record. Patients who had rheumatoid arthritis were excluded. Histological reports were reviewed to confirm the diagnosis of the synovitis in patients who had been managed operatively. Nine of the twenty-eight patients who did not fulfill our criteria for inclusion in the study were excluded because it was impossible to confirm the diagnosis of silicone-induced synovitis. Two patients (one of whom had been managed operatively) had died, and they were excluded as well. The remaining seventeen patients were included in the study. After the investigational protocol had been reviewed and approved by the Mayo Foundation Institutional Review Board and informed consent had been obtained, each patient had a detailed physical examination of the hand and wrist. The examination included neurological evaluation, testing of grip strength with use of a Jamar dynamometer (Asimow Engineering, Los Angeles, California), testing of key pinch strength and chuck pinch strength, and determination of the range of motion of the wrist. Posteroanterior and lateral radiographs of the wrist were made at the time of the latest follow-up for all but three patients. The status of the patient at the time of the follow-up was determined with use of a questionnaire, with questions pertaining to work status, pain, level of function, occupation, swelling about the wrist, difficulty with activities of daily living because of problems with the wrist, overall satisfaction with the result of treatment, and perceived level of wrist function. The patient responded to each question on a visual-analog scale, which has been shown to be a sensitive method of assessment38. The scale was a ten-centimeter-long grid that was divided equally into one-centimeter squares that corresponded to levels of intensity. The patients were divided into two groups according to whether they had been managed non-operatively or operatively. The initial diagnosis and associated procedures were recorded for each patient. The patients who were managed operatively had removal of the implant, débridement, and a variety of reconstructive or salvage procedures. Non-operative treatment consisted of observation and immobilization of the wrist in a splint. Statistical analysis was used to compare the results of non-operative and operative treatment. Comparisons between the non-operative and operative groups for variables with a non-gaussian distribution, such as pain, perceived limitation of motion, and difficulty with activities of daily living, were carried out with use of the exact Wilcoxon rank sum test36. The sample sizes of seven patients in the non-operative group and ten patients in the operative group clearly limited the statistical power of the tests. Additionally, it must be recognized that it is difficult to quantify the exact power of non-parametric tests. However, the power of the Wilcoxon test has been shown to be very close to that of the two-sample t test. In fact, for distributions with very long tails, such as pain and activities of daily living, it may be greater than 100 per cent of the power of the t test22. With the sample sizes of seven and ten, there was at least an 80 per cent chance of detecting an effect size of 1.3that is, a difference in the means of at least 1.3 standard deviations. All tests were two-sided, and a p value of less than 0.05 was considered to be significant.
The mean age of the patients at the time of the initial procedure was fifty-seven years (range, thirty-five to seventy years), the mean interval between the initial procedure and the diagnosis of silicone-induced synovitis was fifty-eight months (range, eight to 151 months), and the mean duration of follow-up after treatment of the synovitis was eighty months (range, thirty-nine to 135 months). Ten dominant and seven non-dominant hands were involved. The initial diagnosis was osteoarthrosis in ten patients, non-union of the scaphoid in three, Kienböck disease in three, and fracture of the scaphoid in one. A scaphoid implant was used in eight patients (Figs. 1-A and 1-B); a trapezium implant, in five; a lunate implant, in three (Figs. 2-A and 2-B); and a scapholunate implant, in one. Three patients were farmers, three were skilled laborers, two were homemakers, two were blue-collar workers, and seven were retired (Table I).
Overall, twelve of the seventeen patients had pain at the time of the latest follow-up, with a mean score of 5.0 on the ten-centimeter visual-analog scale for these twelve patients. Findings characteristic of active debris-induced synovitis were seen on thirteen of fourteen radiographs of the wrist. (Three patients did not wish to have follow-up radiographs.) Fourteen patients thought that the motion of the wrist was limited, and eight reported difficulty with activities of daily living because of problems with the wrist (Table I). Seven patients were managed non-operatively, and ten were managed operatively. A review of the medical records did not reveal a clear reason for the choice of either type of treatment. Six of the seven patients who were managed non-operatively and four of the ten patients who were managed operatively had involvement of the dominant hand. The patients who had been managed non-operatively were a mean of fifty-five years and eight months old (range, thirty-five to sixty-nine years old). The mean time to the diagnosis of the synovitis was fifty-nine months (range, eight to eighty-five months), and the mean duration of follow-up after treatment was ninety-two months (range, fifty-four to 135 months). The initial diagnosis was osteoarthrosis in three patients, non-union of the scaphoid in two, fracture of the scaphoid in one, and Kienböck disease in one. Three patients were retired, two were blue-collar workers, one was a skilled laborer, and one was a farmer. Three patients were managed with observation only, and four were managed with immobilization of the wrist in a splint. The ten patients who were managed operatively had a mean age of fifty-eight years and six months (range, thirty-nine to seventy years). The mean time to the diagnosis of the synovitis was sixty-five months (range, twelve to 151 months), and the mean duration of follow-up was seventy-two months (range, thirty-nine to 126 months). Seven patients had osteoarthrosis, two had Kienböck disease, and one had non-union of the scaphoid. Four patients were retired, two were homemakers, two were skilled laborers, and two were farmers. The operative procedures included removal of the implant with limited synovectomy and débridement (four patients), removal of the implant with proximal row carpectomy (excision of the scaphoid, lunate, and triquetrum) as well as limited synovectomy and débridement (one patient), four-corner (lunate-triquetrum-hamate-capitate) arthrodesis with débridement (one patient), panarthrodesis of the wrist with débridement (two patients), scaphotrapezium-trapezoid arthrodesis77 with débridement (one patient), and thumb-sling suspensionplasty with débridement (one patient). With the numbers available for study, we could detect no significant difference between the non-operative and operative groups with regard to the mean scores on the visual-analog scale for pain (2.7 compared with 4.1; p = 0.5116), perceived limitation of wrist motion (2.4 compared with 3.9; p = 0.4260), or difficulty with activities of daily living (0.9 compared with 2.7; p = 0.2391) (Table II). The difference in the total motion of the wrist (92 degrees for the non-operative group and 64 degrees for the operative group) also could not be shown to be significant (p = 0.2391). It should be recognized, however, that two of the patients who were managed operatively had a panarthrodesis and two had a partial arthrodesis of the wrist. The mean grip strength (94 per cent that of the contralateral side for the patients who had non-operative treatment and 70 per cent for the patients who had operative treatment) was not found to be significantly different either, but there was a trend toward a greater grip strength in the non-operative group (p = 0.08). No patient had regional lymphadenopathy at the time of the most recent follow-up.
With the numbers available, we could detect no significant difference (p = 0.2125) between the two groups of patients with regard to evidence of osteolysis typical of debris-induced synovitis on follow-up radiographs. There was evidence of progressive osteolysis with formation of subchondral cysts in thirteen of the fourteen patients for whom follow-up radiographs were available. Only two of these thirteen patients were asymptomatic at the time of the latest follow-up.
In the 1950s and 1960s, there were several reports of a granulomatous reaction after the use of silicone-rubber implants in mice as well as in tissue cultures37,47,51. In 1971, Donahue et al.10 reported on intra-articular injections of silicone fluid in rabbits. As the time after the injection increased, the amount of silicone retrieved from the joint diminished. At three months, the animals were killed and silicone material was found in the synovial tissues, where it had evoked a granulomatous inflammatory response. It was deduced that the silicone fluid had gradually left the knee and deposited in the synovial tissue. Later, a series of case reports by Aptekar et al.2, Christie et al.7, Gordon and Bullough17, and Worsing et al.82 contained descriptions of an inflammatory reaction around silicone implants, with histological evidence of a foreign-body reaction. Aptekar et al. are credited with the first description, in 1974, of clinical silicone-induced synovitis, which occurred around a metacarpophalangeal joint. Silicone-induced synovitis of the wrist is caused by wear of the implant2,6,17,49,61,76. Silicone-rubber debris in the synovial tissue surrounding a silicone-rubber implant has been shown by mass spectrophotometry, light microscopy, and electron microscopy17-47,79. The relationship between the amount of debris and the length of time that the silicone-rubber implant has been in place also has been established46,47,76. The size of the particles generally ranges from twenty to 100 micrometers1,3,27, but larger particles may be found11. These larger particles are phagocytized by multinucleated giant cells in association with a generalized foreign-body response4,5,9,17. The smaller particles can be directly phagocytized by microphages and can be transported a substantial distance from the implant7,8,11. This phenomenon may account for the lytic areas in bone far from the implant12. The smaller particles seem to initiate the most destructive foreign-body inflammatory process in silicone-induced synovitis17,75,82, and this seems to be true for any foreign implantable material. Worsing et al.82 demonstrated this foreign-body response in a rabbit model, in which inflammatory changes developed in response to silicone material in the synovial tissue. The process is perhaps analogous to the foreign-body osteolysis that has been seen in association with total hip arthroplasty52. The polymetric wear that has been postulated to cause lytic reactions around hip prostheses results in particles of debris that are similar in size to those found in association with silicone-induced synovitis33,39,40,51,78,81. Similarly, macrophages and histiocytes have been implicated in the foreign-body reaction around total hip components33,38,40,51,78,81. It has also been shown, by Mirra et al.39,40, that the amount of wear debris corresponds to the intensity of the giant-cell response, the degree of osteolysis, and eventual loosening of the prosthesis. The amount of wear debris is probably a function of the load that is borne by the silicone-rubber implant1,11. Debris-induced synovitis is more prevalent in young patients who have high activity demands. In addition, radiographs of implants in such patients show a greater degree of deformity than do radiographs of implants in elderly patients1,6. The longer the implant is in place, the more wear debris it produces1,49. Specifically, the lunate is at higher risk in younger, more active patients because it transmits a high amount of load across the wrist6,11. Others have suggested that an autoimmune response may cause the destructive process of silicone-induced synovitis. Studies have demonstrated encapsulation of silicone-rubber implants17,21,23,50,59, which is believed to be a cellular immune response to foreign material and was originally recognized around breast implants14,20,43,56,60,75. The intact silicone-rubber implant, for example, is too large to be phagocytized. Therefore, it has been postulated, an autoimmune response, which is IgG-mediated, produces an infiltration of fibrocytes, histiocytes, and lymphocytes, eventually forming a collagenous capsule around the implant60. Furthermore, as the particles of silicone debris accumulate, a point may be reached at which the debris can no longer be phagocytized and the overabundance of debris gradually accumulates in the lymph nodes, creating a mononuclear hyperplasia and a giant-cell response. Indeed, in several reports8,18,27,29, patients who had lymphadenopathy were found on biopsy to have particles of silicone debris in the regional lymph nodes of the extremity containing the silicone-rubber implant. Lazaro et al.29 found particles of silicone-rubber debris at the sites of regional lymphadenopathy in three of twenty-three patients who had had a silicone-rubber implant in place for variable durations. Smith et al.60 observed a passive cutaneous anaphylactic reaction in animals that had been previously sensitized to a silicone solution by repeated subnuchal injections over a period of six weeks. There was deposition of IgG in the tissue adjacent to the silicone implant in the sensitized animals. It appears that an autoimmune response may play a role in the inflammatory reaction to particles of silicone debris. The osteolysis typically seen in association with synovitis induced by silicone debris may occur due to proteolytic enzymes released by macrophages engulfing particles of silicone. Goldring et al.16 described the production of prostaglandin E2 and collagenase by synovial pannus around femoral components used for total hip arthroplasty. This phenomenon may also explain the destruction of bone that is seen in association with silicone-induced synovitis48,55,57. There have been few clinical studies of the results of treatment for silicone-induced synovitis, and the follow-up has been short-term and incomplete70. Peimer et al.49, reporting on eighteen patients who had been managed for synovitis induced by particles of silicone debris, found that all of the symptoms were relieved after removal of the implant and débridement and all radiographic lesions were stable at twenty-nine months. However, the authors did not report the subjective results or the findings of a physical examination at the time of follow-up. Furthermore, they reported that non-operative treatment of this condition failed. Smith et al.61 reported that removal of the implant apparently was successful in nine patients, but they did not indicate the duration of follow-up. By contrast, Eiken et al.11 reported that osseous destruction progressed after the implant was removed. The mean duration of follow-up of eighty months in the present study is the longest, to our knowledge, of patients who have been managed for synovitis induced by particles of silicone debris. At the time of the latest follow-up, twelve of the seventeen patients had pain, with a mean score of 5.0 (as determined on the ten-centimeter visual-analog scale); thirteen of fourteen patients had radiographic changes that were consistent with silicone-induced synovitis; and eight of seventeen patients thought that problems with the wrist caused difficulty with activities of daily living. The patients who had been managed operatively had more pain (p = 0.5116) and more difficulty with activities of daily living (p = 0.2391) than the patients who had been managed non-operatively, but we could detect no significant differences, with the numbers available. Similarly, the operatively managed patients had less grip strength (p = 0.08) and a decreased total range of motion of the wrist (p = 0.2391), but the differences were not found to be significant. It is important to note that the two patients who had had panarthrodesis of the wrist had a pain score of 0. This may imply that the result is better after more extensive exploration and débridement. However, there were radiographic findings typical of progressive silicone-induced synovitis in these two patients, and painful deterioration of the wrist is possible. Our findings differ from those of previous reports49,61 that have suggested that synovitis induced by particles of silicone debris responds favorably to removal of the implant and débridement. We believe that this difference may be due to the more detailed, outcome-based nature of the present study and the longer duration of follow-up. Our results are more consistent with those of authors11,80 who found progressive destruction of bone, pain, and decreased motion after removal of the implant. Wilgis and Clark80 reported that, of fifteen patients followed for an average of 3.7 years after removal of a carpal implant and synovectomy because of synovitis, six still had pain and four still had synovitis. The discrepancy between the findings of these studies may be accounted for by an initial, perhaps short-lived, success of operative intervention, with the result diminishing over time. We recognize several limitations of the present study. First, the small number of patients precludes accurate statistical comparison of the two groups (those managed operatively and those managed non-operatively) with regard to the diagnosis, the type of implant, hand dominance, and the occupation of the patient. Thus, the results may be biased, with one group being more active and placing greater demands on the wrist or having less debilitating conditions than the other. Four dominant hands and six non-dominant hands were treated in the operative group, compared with six dominant hands and one non-dominant hand in the non-operative group, but this difference did not seem to affect the overall result, as the scores for all of the subjective parameters were better for the non-operative group. A second limitation of the study was that eleven of the patients in the original cohort were unavailable for follow-up, so complete retrieval of data was impossible. Finally, because the study was retrospective, there was a lack of uniformity with regard to the type of treatment. The long-term results of the treatment of synovitis induced by particles of silicone debris were generally poor in the present series, and we concluded that there was no clear benefit to removal of the implant, débridement, and reconstruction of the wrist. Currently, we recommend that patients who have this condition be carefully evaluated and that treatment be individualized. We recommend caution when removal of the implant, débridement, and a reconstructive or salvage procedure is selected for a patient, as the result may be unpredictable.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||