- Email: [email protected]
Long-Term Results of Silicone Wrist Arthroplasty in Patients With Rheumatoid Arthritis
Long-Term Results of Silicone Wrist Arthroplasty in Patients With Rheumatoid Arthritis Urs Kistler, MD, Zurich, Switzerland, Arnold-Peter C. Weiss, MD, Providence, RI, Beat R. Simmen, MD, Daniel B. Herren, MD, Zurich, Switzerland
Purpose: The surgical treatment of the rheumatoid wrist is key in managing the affected hand. Wrist fusion is often the treatment of choice in cases of severe destruction and deformation although most patients would prefer a motion-preserving procedure. The implantation of a wrist prosthesis might be an alternative to partial arthrodesis for selected cases. In this series we analyzed the long-term results (minimum follow-up period, 10 y) of the Swanson silicone spacer for the wrist in patients with rheumatoid arthritis. Methods: Sixteen patients with rheumatoid arthritis with 18 silicone spacers for the wrists were reviewed after a minimum follow-up period of 10 years (average, 15 y). Subjective evaluation, clinical examination, and radiographic analysis were included. An additional 9 patients (9 wrists) were interviewed by telephone. Results: In 12 of the patients the subjective result was good or very good, mostly because of adequate pain relief. The average range of motion for flexion (average, 28°)/extension (average, 15°) was 43° with a wide variation within the series. Radiologically all wrists had diminished residual carpal height at follow-up evaluation and 9 of the wrists had evidence of osteolysis and foreign-body granuloma. The initial good correction of the ulnar translation of the wrist was lost partially in the follow-up period (1.1 vs 4.0 mm). Three of the patients needed surgical revision within the follow-up period; all were converted to wrist fusion. Conclusions: These long-term results suggest that the silicone wrist spacer still may be considered as an alternative to wrist fusion or more complex wrist joint prostheses in patients with rheumatoid arthritis, especially in severe cases and in patients with low demands. In the long term osteolysis caused by foreign-body granulation is to be expected and has to be considered. (J Hand Surg 2005;30A:1282–1287. Copyright © 2005 by the American Society for Surgery of the Hand.) Type of study/level of evidence: Therapeutic, Level IV. Key words: Arthroplasty, wrist, Silastic, rheumatoid arthritis.
From the Department of Hand Surgery, Schulthess Clinic, Zurich, Switzerland; and University Orthopedics Inc., Providence, RI. Received for publication September 1, 2004; accepted in revised form July 6, 2005. 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. Corresponding author: Daniel B. Herren, MD, Schulthess Clinic, Lengghalde 2, 8008 Zurich, Switzerland; e-mail: [email protected]. Copyright © 2005 by the American Society for Surgery of the Hand 0363-5023/05/30A06-0026$30.00/0 doi:10.1016/j.jhsa.2005.07.009
1282
The Journal of Hand Surgery
Kistler et al / Silicone Wrist Arthroplasty
The goals of surgical intervention in the rheumatoid wrist are to obtain a pain-free, stable, functional wrist in the context of the entire upper extremity. In patients with advanced destruction and deformity wrist fusion (partial or total) or wrist arthroplasty are the most common treatment options.1–10 Wrist fusion offers excellent stability with the obvious disadvantage of loss of motion, which restricts certain activities of daily living. Patients prefer wrist arthroplasty to fusion because of a preserved arc of motion.7 Complex 2-component wrist implants show clinically good results although complications can be severe.1,2,6,9,11 Early loosening or dislocation of these implants are observed in up to 30% of cases. Silicone wrist implant arthroplasty—introduced by Swanson in the late 1960s—preceded these 2-component implants, and the silicone implants were relatively easy to implant. Despite the relatively good biocompatibility of silicone the implants are prone to fracture, particulate wear, debris generation (which often leads to silicone synovitis), and granuloma formation.12 This study reviewed the long-term results both clinically and radiographically of Swanson silicone wrist arthroplasty in patients with rheumatoid arthritis. Only implants with a minimum follow-up period of 10 years were included.
Materials and Methods Between 1978 and 1987 there were 99 wrists in 85 patients with rheumatoid arthritis that had replacement with a silicone implant (Wright Medical, Memphis, TN). In all patients the original prosthesis made of conventional elastomer was implanted. No titanium grommets were used. Thirty-six patients with 44 replaced wrists died before this study and 24 patients with 28 surgically treated wrists could not be located. Most of these patients live in foreign countries and could not be traced after such a long follow-up period. Sixteen patients (18 implants) were available for clinical and radiologic examination and another 9 patients (9 implants) were interviewed by telephone. The minimum follow-up period was 10 years (average, 15.2 y; range, 10 –19 y). The average duration of rheumatoid arthritis before surgery was 16 years (range, 5–33 y), the average patient age was 50 years (range, 24 – 64 y), and the female-to-male ratio was 5 to 1. The main indication for surgery was pain in all patients; in addition 8 patients complained of weakness and another 16 patients complained of wrist instability.
1283
Figure 1. Resection for the Swanson wrist implant. The center of rotation should be located in the head of the capitate.
Surgical Technique The surgical technique followed that described by Swanson et al.8 A dorsal approach to the wrist was used and extensor synovectomy and distal ulnar resection were performed. The radiocarpal joint capsule is elevated with a T-shaped incision. The distal radius is resected in a straight line perpendicular to the long axis of the radius while the proximal carpal row, if present, is resected through the head of the capitate. The implant is placed with the flexionextension axis approximately in the head of the capitate (Fig. 1). The proximal and distal medullary canals are rasped to accept the implant. The size of the prosthesis is chosen according to bone stock and the desired tension. The joint should not be overstuffed. The joint capsule is closed and if needed is reinforced with the distal third of the retinaculum, which is passed under the extensor tendons and sutured to the residual joint capsule to cover the implant. The standardized postoperative care included wrist immobilization for 5 to 6 weeks. After this time gentle active mobilization was started. The resting splint was worn for a total of 8 weeks during the day and 3 months at night.
Subjective Evaluation Subjective patient satisfaction with the surgical treatment of the wrist was rated as excellent, good, sat-
1284
The Journal of Hand Surgery / Vol. 30A No. 6 November 2005
isfactory, or poor. The subjective grading of residual pain and strength of the wrist and hand before and after surgery and at the follow-up examination was evaluated on a visual analog scale similar to the criteria of Huskisson13 for pain measurement from 1 to 10, with 10 representing the maximum level of pain. Objective strength measurement was not conclusive because most patients were disabled at the level of the fingers and hand strength was determined mainly by the finger deformation and not by the function of the wrist. Most patients were unable to hold a dynamometer.
creased power. Average strength values on the visual analog scale were 6.6 before surgery and 2.75 after surgery (p ⬍ .001).
Objective Evaluation
Radiologic Examination
In all patients seen clinically (n ⫽ 18) the wrist range of motion was recorded for flexion, extension, and ulnar and radial deviation.
The assessment for possible implant fracture was extremely difficult because all the implants had substantial subsidence. In 8 wrists implant fracture was seen easily although it is likely that, with rare exceptions, all implants suffered fragmentation. The fractures of the implants occurred at the junction of the silicone body and the distal stem. Residual carpal height, measured as the distance between the base of the third metacarpal to the distal part of the radius, was 22 mm (⫾6 mm) before surgery compared with 23 mm (⫾6 mm) just after surgery (p ⫽ .15). The carpal height at the final follow-up evaluation was 14 mm (⫾7 mm), representing a statistically significant loss compared with the first postoperative control (p ⬍ .001). The ulnar translation (distance between the center of the third metacarpal and the center of the radius) averaged 8.5 mm before surgery (⫾3.6 mm) compared with 1.1 mm (⫾3.3 mm) after surgery (p ⬍ .001). At the final follow-up evaluation ulnar translation was 4.0 mm (⫾6.8 mm), indicating a significant ulnar translation (p ⫽ .04). Four wrists showed massive cystic changes (Figs. 2, 3) with periprothetic bone loss, with 5 cases showing moderate cystic lesions. From the 99 silicone wrist implants 19 wrists required subsequent surgery; all wrists were converted to wrist fusion. Table 1 summarizes the study results.
Radiologic Evaluation The radiologic analysis included measurement of the absolute carpal height, ulnar translation both before and after surgery, signs of implant fracture, and cystic changes. The carpal height had to be measured absolutely in millimeters because old preoperative x-rays did not include the complete third metacarpal and no carpal height ratio could be determined. Because all x-rays were performed at the same institution with standardized film-object distance, the magnification factor was constant and an absolute height measurement could be obtained. The statistical analysis included the paired Student t test. The statistical significance level was set at p ⬍ .05.
Results Subjective Evaluation In the subjective evaluation all patients with bilateral wrist treatment could not differentiate between one wrist or the other. The results therefore are reported according to the number of patients. From a total of 27 patients (clinical and telephone follow-up interviews) 6 patients rated their results as excellent, 13 as good, 6 as satisfactory, and 2 as poor. The 2 patients with poor results complained about stiffness and weakness of the wrist. Twenty-five patients had less pain after surgery than before the intervention. In 2 patients pain did not improve after surgery. The pain level ratings were 6.4 before surgery and 1.7 after surgery (p ⬍ .001). None of the patients had increased pain after arthroplasty. Nineteen patients had subjectively less power in their hands after surgery, 4 patients had identical power, and 4 patients had in-
Objective Evaluation The average range-of-motion arc in flexion-extension for the 18 wrists was 44° (range, 0°–130°), with an average flexion of 28° (range, 0°–70°) and extension of 15° (range, 0°– 60°). The average range of ulnarradial deviation arc was 21° (range, 0°–55°), with 7° (range, –30°–30°) of radial deviation and 14° (range, 0°–50°) of ulnar deviation.
Discussion This study encompasses one of the longest follow-up reports of silicone wrist arthroplasty. In 1998 Lluch and Proubasta14 reported on 22 wrists with the conventional elastomer compared with another 11 wrists with the new elastomer: the new design with a shorter, wider stem used together with grommets. The first group with the first generation of implants, similar to the implant used in the current series,
Kistler et al / Silicone Wrist Arthroplasty
Figure 2. Postoperative x-ray after implantation of a wrist implant (Silastic, Wright Medical). The wrist is well balanced with good restoration of the carpal height.
showed a 75% fracture rate with 40% foreign-body reaction around the prosthesis. The fractures all occurred at the base of the distal stem, similar to the fracture pattern in the current series. In contrast the 11 implants with the new design and the different elastomer had only 1 fracture; however, the follow-up period was significantly shorter, with an average of a little more than 3 years (range, 18 mo to 6 y 8 mo) compared with 11.5 years (range, 6 –20 y) with the old design. For both groups no range of motion was reported. Schill et al15 (72 patients with 82 silicone wrist implants) had 51% good or excellent results after 10 years (range, 5–14 y) with a relatively low revision rate of 13%. Stanley and Tolat,16 in their series of 50 Swanson wrist implants with an average follow-up period of 8 years (range, 6.0 –11.8 y), found that two thirds of patients had a good to excellent result judged with a summary score including pain, patient satisfaction, function, and wrist motion. These contrast with the reports of Fatti et al4,17 who had an average follow-up period of 5.8 years (minimum, 3 y) with good to excellent results in only 25% of patients. Similar results have been reported by Comstock et al18 after 6 years of follow-up study (range, 3–9 y). On closer examination it is apparent that patients
1285
with rheumatoid arthritis have better results from silicone arthroplasty compared with patients with posttraumatic conditions. This observation is supported by the study of Jolly et al,19 which showed good to excellent results for silicone wrist implants in rheumatoid arthritis in 8 patients of patients but with a short follow-up period of only 6 years (range, 7 mo to 13.1 y). The fate of those implants over time still is unclear. Fatti et al4,17 had an initial favorable clinical course in their series followed by a deterioration of the results with longer-term follow-up study. Our study did not confirm this observation. Although there appears to be a high number of implant fractures over time there does not appear to be any correlation between clinical results and implant integrity. There is some evidence that greater mobility of the wrist after surgery has a positive correlation with the frequency of implant fracture. On the other hand there does not appear to be any link between postoperative range of motion and cystic lesions. It also remains unclear why some wrists move extremely well after surgery whereas others remain stiff. There must be multiple factors contributing to this observation, which include the disease course, hand use, and characteristics of individual scar for-
Figure 3. Same wrist as in Figure 2 after a follow-up period of 16 years. There is a notable cystic reaction with granuloma formation. The implant is difficult to identify and probably dissolved.
1286
The Journal of Hand Surgery / Vol. 30A No. 6 November 2005
Table 1. Synopsis of Statistical Analysis Results Criteria
Results
Age (y) Follow-up period (mo) Preoperative pain (VAS 1–10) Postoperative pain (VAS 1–10) Preoperative grip power (VAS 1–10) Postoperative grip power (VAS 1–10) Flexion (°) Extension (°) ROM flexion–extension (°) Radial deviation (°) Ulnar deviation (°) ROM radial–ulnar (°) Preoperative carpal height (mm) Postoperative carpal height (mm) Follow-up carpal height (mm) Preoperative ulnar translation (mm) Postoperative ulnar translation (mm) Follow-up ulnar translation (mm) Patient satisfaction, n (%) Excellent Good Satisfactory Poor
65.9 ⫾ 9.9 182 6.5 ⫾ 2.5 1.8 ⫾ 2.5 6.6 ⫾ 2.4 2.7 ⫾ 1.7 28.2 ⫾ 21.4 15.3 ⫾ 16.6 43.5 ⫾ 33.3 6.5 ⫾ 13.2 14.4 ⫾ 17.2 20.9 ⫾ 18.2 22.0 ⫾ 6.0 23.0 ⫾ 5.5 14.0 ⫾ 7.0 8.5 ⫾ 3.6 1.1 ⫾ 3.3 4.0 ⫾ 6.8 5 12 6 2
Statistics
p ⬍ .001 p ⬍ .001
p ⫽ .15, n.s.
p ⬍ 0.001
p ⬍ 0.001 p 0.040
(20) (48) (24) (8)
ROM, range of motion; VAS, visual analog scale.
mation. It seems logical that the load on an implant is more important for the survival of the implant than other factors. Brase and Millender20 already have described this association. Although the average range of motion in our series was 55°—slightly above the recorded ranges of the other series— only 2 of the patients reached functional mobility as defined by Palmer et al21 with 5° of flexion and 30° of extension. This definition of functional range of motion, however, was determined in healthy volunteers and has limited validity for patients with a disease that affects the entire upper extremity. A wrist without pain, even if the range of motion is limited, is a positive factor in this group of patients. Despite the radiologic appearance of the implant, pain control is quite good and predictable. The low revision rate may be caused partly by consistent clinical results, limited expectations, and a high threshold of tolerating functional deficits. Revision surgery of failed silicone wrist implants has its difficulties. The usual approach is conversion to a total wrist fusion but the bone defect may require substantial bone grafting.22,23 Hardware fixation often is difficult in these osteoporotic patients. There may be cases in which a simple implant exchange is the easier and more practical solution. The present
series shows numbers of wrists with extensive cystic changes and therefore difficult revision situations. Our study has clear limitations. Its retrospective character with remembered preoperative pain and function data and the loss of patients over the long follow-up period restrict the conclusions. Most other published series with long follow-up times, however, have similar limitations. Nevertheless it is important to analyze the efficacy of prosthetic implants in the long term and to compare them with modern treatment concepts. Newer, complex, 2-component wrist implants show clinically promising results although complications can be important.1,2,6,9,11 Early loosening and dislocation of these implants are observed in up to 30% of cases. One of the most important indications for wrist arthroplasty is rheumatoid arthritis. Complex wrist deformity and difficult bone stock conditions, however, limit the use of modern wrist prostheses in these patients. Silicone wrist implants might be an alternative to wrist fusion in this difficult-to-treat patient group. Cystic bone changes are the main complication of this device; these depend mainly on the silicone wear with reactive granulomatosis and are linked to implant load and deformation. Silicone wrist arthroplasty therefore might be recommended only in low-demand patients with
Kistler et al / Silicone Wrist Arthroplasty
rheumatoid arthritis who require some residual mobility at the level of the wrist joint. Wrists with gross deformity and mutilans type of destruction should be treated with a wrist fusion procedure. Swanson silicone wrist implants in patients with rheumatoid arthritis showed good clinical results with high patient satisfaction in a long-term follow-up study. The radiologic appearance of the implants shows substantial deterioration with significant cystic changes in the bone. This observation, however, has little direct correlation to residual pain or mobility. Silicone wrist arthroplasty remains a possible alternative to wrist fusion in low-demand patients with rheumatoid arthritis.
10.
11.
12.
13. 14. 15.
References 1. Cobb TK, Beckenbaugh RD. Biaxial total-wrist arthroplasty. J Hand Surg 1996;21A:1011–1021. 2. Costi J, Krishnan J, Pearcy M. Total wrist arthroplasty: a quantitative review of the last 30 years. J Rheumatol 1998; 25:451– 458. 3. Divelbiss BJ, Sollerman C, Adams BD. Early results of the Universal total wrist arthroplasty in rheumatoid arthritis. J Hand Surg 2002;27A:195–204. 4. Fatti JF, Palmer AK, Greenky S, Mosher JF. Long-term results of Swanson interpositional wrist arthroplasty: part II. J Hand Surg 1991;16A:432– 437. 5. Lundkvist L, Barfred T. Total wrist arthroplasty. Experience with Swanson flexible silicone implants, 1982–1988. Scand J Plast Reconstr Surg 1992;26:97–100. 6. Meuli HC, Fernandez DL. Uncemented total wrist arthroplasty. J Hand Surg 1995;20A:115–122. 7. Murphy DM, Khoury JG, Imbriglia JE, Adams BD. Comparison of arthroplasty and arthrodesis for the rheumatoid wrist. J Hand Surg 2003;28A:570 –576. 8. Swanson AB, de Groot Swanson G, Maupin BK. Flexible implant arthroplasty of the radiocarpal joint. Surgical technique and long-term study. Clin Orthop 1984;187:94 –106 9. Takwale VJ, Nuttall D, Trail IA, Stanley JK. Biaxial total wrist replacement in patients with rheumatoid arthritis. Clin-
16.
17.
18.
19.
20.
21.
22. 23.
1287
ical review, survivorship and radiological analysis. J Bone Joint Surg 2002;84B:692– 699. Vicar AJ, Burton RI. Surgical management of the rheumatoid wrist—fusion or arthroplasty. J Hand Surg 1986;11A: 790 –797. Summers B, Hubbard MJS. Wrist joint arthroplasty in rheumatoid arthritis: a comparison between the Meuli and Swanson prostheses. J Hand Surg 1984;9B:171–176. DeHeer DH, Owens SR, Swanson AB. The host response to silicone elastomer implants for small joint arthroplasty. J Hand Surg 1995;20A(Suppl):S101–S109. Huskisson EC. Measurement of pain. Lancet 1974;9:1127– 1131. Lluch A, Proubasta L. Les implants radio-carpeins de Swanson. Resultat a longterm. La Main 1998;3:176 –184. Schill S, Thabe H, Mohr W. Langzeitergebnisse nach Swanson-Prosthesenversorgung des rhematischen Handgelenks. [Long-term outcome of Swanson prosthesis management of the rheumatic wrist joint]. Handchir Mikrochir Plast Chir 2001;33:198 –206. Stanley JK, Tolat AR. Long-term results of Swanson Silastic arthroplasty in the rheumatoid wrist. J Hand Surg 1993;18B: 381–388. Fatti JF, Palmer AK, Mosher JF. The long-term results of Swanson silicone rubber interpositional wrist arthroplasty. J Hand Surg 1986;11A:166 –175. Comstock CP, Louis DS, Eckenrode JF. Silicone wrist implant: long-term follow-up study. J Hand Surg 1988;13A: 201–205. Jolly SL, Ferlic DC, Clayton ML, Dennis DA, Stringer EA. Swanson silicone arthroplasty of the wrist in rheumatoid arthritis: a long-term follow-up. J Hand Surg 1992;17A:142– 149. Brase DW, Millender LH. Failure of silicone rubber wrist arthroplasty in rheumatoid arthritis. J Hand Surg 1986;11A: 175–183. Palmer AK, Werner FW, Murphy D, Glisson BS. Functional wrist motion: a biomechanical study. J Hand Surg 1985; 10A:39 – 46. Carlson JR, Simmons BP. Wrist arthrodesis after failed wrist implant arthroplasty. J Hand Surg 1988;23A:893– 898. Vögelin E, Nagy L. Fate of failed Meuli total wrist arthroplasty. J Hand Surg 2003;28B:61– 68.
Purpose: The surgical treatment of the rheumatoid wrist is key in managing the affected hand. Wrist fusion is often the treatment of choice in cases of severe destruction and deformation although most patients would prefer a motion-preserving procedure. The implantation of a wrist prosthesis might be an alternative to partial arthrodesis for selected cases. In this series we analyzed the long-term results (minimum follow-up period, 10 y) of the Swanson silicone spacer for the wrist in patients with rheumatoid arthritis. Methods: Sixteen patients with rheumatoid arthritis with 18 silicone spacers for the wrists were reviewed after a minimum follow-up period of 10 years (average, 15 y). Subjective evaluation, clinical examination, and radiographic analysis were included. An additional 9 patients (9 wrists) were interviewed by telephone. Results: In 12 of the patients the subjective result was good or very good, mostly because of adequate pain relief. The average range of motion for flexion (average, 28°)/extension (average, 15°) was 43° with a wide variation within the series. Radiologically all wrists had diminished residual carpal height at follow-up evaluation and 9 of the wrists had evidence of osteolysis and foreign-body granuloma. The initial good correction of the ulnar translation of the wrist was lost partially in the follow-up period (1.1 vs 4.0 mm). Three of the patients needed surgical revision within the follow-up period; all were converted to wrist fusion. Conclusions: These long-term results suggest that the silicone wrist spacer still may be considered as an alternative to wrist fusion or more complex wrist joint prostheses in patients with rheumatoid arthritis, especially in severe cases and in patients with low demands. In the long term osteolysis caused by foreign-body granulation is to be expected and has to be considered. (J Hand Surg 2005;30A:1282–1287. Copyright © 2005 by the American Society for Surgery of the Hand.) Type of study/level of evidence: Therapeutic, Level IV. Key words: Arthroplasty, wrist, Silastic, rheumatoid arthritis.
From the Department of Hand Surgery, Schulthess Clinic, Zurich, Switzerland; and University Orthopedics Inc., Providence, RI. Received for publication September 1, 2004; accepted in revised form July 6, 2005. 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. Corresponding author: Daniel B. Herren, MD, Schulthess Clinic, Lengghalde 2, 8008 Zurich, Switzerland; e-mail: [email protected]. Copyright © 2005 by the American Society for Surgery of the Hand 0363-5023/05/30A06-0026$30.00/0 doi:10.1016/j.jhsa.2005.07.009
1282
The Journal of Hand Surgery
Kistler et al / Silicone Wrist Arthroplasty
The goals of surgical intervention in the rheumatoid wrist are to obtain a pain-free, stable, functional wrist in the context of the entire upper extremity. In patients with advanced destruction and deformity wrist fusion (partial or total) or wrist arthroplasty are the most common treatment options.1–10 Wrist fusion offers excellent stability with the obvious disadvantage of loss of motion, which restricts certain activities of daily living. Patients prefer wrist arthroplasty to fusion because of a preserved arc of motion.7 Complex 2-component wrist implants show clinically good results although complications can be severe.1,2,6,9,11 Early loosening or dislocation of these implants are observed in up to 30% of cases. Silicone wrist implant arthroplasty—introduced by Swanson in the late 1960s—preceded these 2-component implants, and the silicone implants were relatively easy to implant. Despite the relatively good biocompatibility of silicone the implants are prone to fracture, particulate wear, debris generation (which often leads to silicone synovitis), and granuloma formation.12 This study reviewed the long-term results both clinically and radiographically of Swanson silicone wrist arthroplasty in patients with rheumatoid arthritis. Only implants with a minimum follow-up period of 10 years were included.
Materials and Methods Between 1978 and 1987 there were 99 wrists in 85 patients with rheumatoid arthritis that had replacement with a silicone implant (Wright Medical, Memphis, TN). In all patients the original prosthesis made of conventional elastomer was implanted. No titanium grommets were used. Thirty-six patients with 44 replaced wrists died before this study and 24 patients with 28 surgically treated wrists could not be located. Most of these patients live in foreign countries and could not be traced after such a long follow-up period. Sixteen patients (18 implants) were available for clinical and radiologic examination and another 9 patients (9 implants) were interviewed by telephone. The minimum follow-up period was 10 years (average, 15.2 y; range, 10 –19 y). The average duration of rheumatoid arthritis before surgery was 16 years (range, 5–33 y), the average patient age was 50 years (range, 24 – 64 y), and the female-to-male ratio was 5 to 1. The main indication for surgery was pain in all patients; in addition 8 patients complained of weakness and another 16 patients complained of wrist instability.
1283
Figure 1. Resection for the Swanson wrist implant. The center of rotation should be located in the head of the capitate.
Surgical Technique The surgical technique followed that described by Swanson et al.8 A dorsal approach to the wrist was used and extensor synovectomy and distal ulnar resection were performed. The radiocarpal joint capsule is elevated with a T-shaped incision. The distal radius is resected in a straight line perpendicular to the long axis of the radius while the proximal carpal row, if present, is resected through the head of the capitate. The implant is placed with the flexionextension axis approximately in the head of the capitate (Fig. 1). The proximal and distal medullary canals are rasped to accept the implant. The size of the prosthesis is chosen according to bone stock and the desired tension. The joint should not be overstuffed. The joint capsule is closed and if needed is reinforced with the distal third of the retinaculum, which is passed under the extensor tendons and sutured to the residual joint capsule to cover the implant. The standardized postoperative care included wrist immobilization for 5 to 6 weeks. After this time gentle active mobilization was started. The resting splint was worn for a total of 8 weeks during the day and 3 months at night.
Subjective Evaluation Subjective patient satisfaction with the surgical treatment of the wrist was rated as excellent, good, sat-
1284
The Journal of Hand Surgery / Vol. 30A No. 6 November 2005
isfactory, or poor. The subjective grading of residual pain and strength of the wrist and hand before and after surgery and at the follow-up examination was evaluated on a visual analog scale similar to the criteria of Huskisson13 for pain measurement from 1 to 10, with 10 representing the maximum level of pain. Objective strength measurement was not conclusive because most patients were disabled at the level of the fingers and hand strength was determined mainly by the finger deformation and not by the function of the wrist. Most patients were unable to hold a dynamometer.
creased power. Average strength values on the visual analog scale were 6.6 before surgery and 2.75 after surgery (p ⬍ .001).
Objective Evaluation
Radiologic Examination
In all patients seen clinically (n ⫽ 18) the wrist range of motion was recorded for flexion, extension, and ulnar and radial deviation.
The assessment for possible implant fracture was extremely difficult because all the implants had substantial subsidence. In 8 wrists implant fracture was seen easily although it is likely that, with rare exceptions, all implants suffered fragmentation. The fractures of the implants occurred at the junction of the silicone body and the distal stem. Residual carpal height, measured as the distance between the base of the third metacarpal to the distal part of the radius, was 22 mm (⫾6 mm) before surgery compared with 23 mm (⫾6 mm) just after surgery (p ⫽ .15). The carpal height at the final follow-up evaluation was 14 mm (⫾7 mm), representing a statistically significant loss compared with the first postoperative control (p ⬍ .001). The ulnar translation (distance between the center of the third metacarpal and the center of the radius) averaged 8.5 mm before surgery (⫾3.6 mm) compared with 1.1 mm (⫾3.3 mm) after surgery (p ⬍ .001). At the final follow-up evaluation ulnar translation was 4.0 mm (⫾6.8 mm), indicating a significant ulnar translation (p ⫽ .04). Four wrists showed massive cystic changes (Figs. 2, 3) with periprothetic bone loss, with 5 cases showing moderate cystic lesions. From the 99 silicone wrist implants 19 wrists required subsequent surgery; all wrists were converted to wrist fusion. Table 1 summarizes the study results.
Radiologic Evaluation The radiologic analysis included measurement of the absolute carpal height, ulnar translation both before and after surgery, signs of implant fracture, and cystic changes. The carpal height had to be measured absolutely in millimeters because old preoperative x-rays did not include the complete third metacarpal and no carpal height ratio could be determined. Because all x-rays were performed at the same institution with standardized film-object distance, the magnification factor was constant and an absolute height measurement could be obtained. The statistical analysis included the paired Student t test. The statistical significance level was set at p ⬍ .05.
Results Subjective Evaluation In the subjective evaluation all patients with bilateral wrist treatment could not differentiate between one wrist or the other. The results therefore are reported according to the number of patients. From a total of 27 patients (clinical and telephone follow-up interviews) 6 patients rated their results as excellent, 13 as good, 6 as satisfactory, and 2 as poor. The 2 patients with poor results complained about stiffness and weakness of the wrist. Twenty-five patients had less pain after surgery than before the intervention. In 2 patients pain did not improve after surgery. The pain level ratings were 6.4 before surgery and 1.7 after surgery (p ⬍ .001). None of the patients had increased pain after arthroplasty. Nineteen patients had subjectively less power in their hands after surgery, 4 patients had identical power, and 4 patients had in-
Objective Evaluation The average range-of-motion arc in flexion-extension for the 18 wrists was 44° (range, 0°–130°), with an average flexion of 28° (range, 0°–70°) and extension of 15° (range, 0°– 60°). The average range of ulnarradial deviation arc was 21° (range, 0°–55°), with 7° (range, –30°–30°) of radial deviation and 14° (range, 0°–50°) of ulnar deviation.
Discussion This study encompasses one of the longest follow-up reports of silicone wrist arthroplasty. In 1998 Lluch and Proubasta14 reported on 22 wrists with the conventional elastomer compared with another 11 wrists with the new elastomer: the new design with a shorter, wider stem used together with grommets. The first group with the first generation of implants, similar to the implant used in the current series,
Kistler et al / Silicone Wrist Arthroplasty
Figure 2. Postoperative x-ray after implantation of a wrist implant (Silastic, Wright Medical). The wrist is well balanced with good restoration of the carpal height.
showed a 75% fracture rate with 40% foreign-body reaction around the prosthesis. The fractures all occurred at the base of the distal stem, similar to the fracture pattern in the current series. In contrast the 11 implants with the new design and the different elastomer had only 1 fracture; however, the follow-up period was significantly shorter, with an average of a little more than 3 years (range, 18 mo to 6 y 8 mo) compared with 11.5 years (range, 6 –20 y) with the old design. For both groups no range of motion was reported. Schill et al15 (72 patients with 82 silicone wrist implants) had 51% good or excellent results after 10 years (range, 5–14 y) with a relatively low revision rate of 13%. Stanley and Tolat,16 in their series of 50 Swanson wrist implants with an average follow-up period of 8 years (range, 6.0 –11.8 y), found that two thirds of patients had a good to excellent result judged with a summary score including pain, patient satisfaction, function, and wrist motion. These contrast with the reports of Fatti et al4,17 who had an average follow-up period of 5.8 years (minimum, 3 y) with good to excellent results in only 25% of patients. Similar results have been reported by Comstock et al18 after 6 years of follow-up study (range, 3–9 y). On closer examination it is apparent that patients
1285
with rheumatoid arthritis have better results from silicone arthroplasty compared with patients with posttraumatic conditions. This observation is supported by the study of Jolly et al,19 which showed good to excellent results for silicone wrist implants in rheumatoid arthritis in 8 patients of patients but with a short follow-up period of only 6 years (range, 7 mo to 13.1 y). The fate of those implants over time still is unclear. Fatti et al4,17 had an initial favorable clinical course in their series followed by a deterioration of the results with longer-term follow-up study. Our study did not confirm this observation. Although there appears to be a high number of implant fractures over time there does not appear to be any correlation between clinical results and implant integrity. There is some evidence that greater mobility of the wrist after surgery has a positive correlation with the frequency of implant fracture. On the other hand there does not appear to be any link between postoperative range of motion and cystic lesions. It also remains unclear why some wrists move extremely well after surgery whereas others remain stiff. There must be multiple factors contributing to this observation, which include the disease course, hand use, and characteristics of individual scar for-
Figure 3. Same wrist as in Figure 2 after a follow-up period of 16 years. There is a notable cystic reaction with granuloma formation. The implant is difficult to identify and probably dissolved.
1286
The Journal of Hand Surgery / Vol. 30A No. 6 November 2005
Table 1. Synopsis of Statistical Analysis Results Criteria
Results
Age (y) Follow-up period (mo) Preoperative pain (VAS 1–10) Postoperative pain (VAS 1–10) Preoperative grip power (VAS 1–10) Postoperative grip power (VAS 1–10) Flexion (°) Extension (°) ROM flexion–extension (°) Radial deviation (°) Ulnar deviation (°) ROM radial–ulnar (°) Preoperative carpal height (mm) Postoperative carpal height (mm) Follow-up carpal height (mm) Preoperative ulnar translation (mm) Postoperative ulnar translation (mm) Follow-up ulnar translation (mm) Patient satisfaction, n (%) Excellent Good Satisfactory Poor
65.9 ⫾ 9.9 182 6.5 ⫾ 2.5 1.8 ⫾ 2.5 6.6 ⫾ 2.4 2.7 ⫾ 1.7 28.2 ⫾ 21.4 15.3 ⫾ 16.6 43.5 ⫾ 33.3 6.5 ⫾ 13.2 14.4 ⫾ 17.2 20.9 ⫾ 18.2 22.0 ⫾ 6.0 23.0 ⫾ 5.5 14.0 ⫾ 7.0 8.5 ⫾ 3.6 1.1 ⫾ 3.3 4.0 ⫾ 6.8 5 12 6 2
Statistics
p ⬍ .001 p ⬍ .001
p ⫽ .15, n.s.
p ⬍ 0.001
p ⬍ 0.001 p 0.040
(20) (48) (24) (8)
ROM, range of motion; VAS, visual analog scale.
mation. It seems logical that the load on an implant is more important for the survival of the implant than other factors. Brase and Millender20 already have described this association. Although the average range of motion in our series was 55°—slightly above the recorded ranges of the other series— only 2 of the patients reached functional mobility as defined by Palmer et al21 with 5° of flexion and 30° of extension. This definition of functional range of motion, however, was determined in healthy volunteers and has limited validity for patients with a disease that affects the entire upper extremity. A wrist without pain, even if the range of motion is limited, is a positive factor in this group of patients. Despite the radiologic appearance of the implant, pain control is quite good and predictable. The low revision rate may be caused partly by consistent clinical results, limited expectations, and a high threshold of tolerating functional deficits. Revision surgery of failed silicone wrist implants has its difficulties. The usual approach is conversion to a total wrist fusion but the bone defect may require substantial bone grafting.22,23 Hardware fixation often is difficult in these osteoporotic patients. There may be cases in which a simple implant exchange is the easier and more practical solution. The present
series shows numbers of wrists with extensive cystic changes and therefore difficult revision situations. Our study has clear limitations. Its retrospective character with remembered preoperative pain and function data and the loss of patients over the long follow-up period restrict the conclusions. Most other published series with long follow-up times, however, have similar limitations. Nevertheless it is important to analyze the efficacy of prosthetic implants in the long term and to compare them with modern treatment concepts. Newer, complex, 2-component wrist implants show clinically promising results although complications can be important.1,2,6,9,11 Early loosening and dislocation of these implants are observed in up to 30% of cases. One of the most important indications for wrist arthroplasty is rheumatoid arthritis. Complex wrist deformity and difficult bone stock conditions, however, limit the use of modern wrist prostheses in these patients. Silicone wrist implants might be an alternative to wrist fusion in this difficult-to-treat patient group. Cystic bone changes are the main complication of this device; these depend mainly on the silicone wear with reactive granulomatosis and are linked to implant load and deformation. Silicone wrist arthroplasty therefore might be recommended only in low-demand patients with
Kistler et al / Silicone Wrist Arthroplasty
rheumatoid arthritis who require some residual mobility at the level of the wrist joint. Wrists with gross deformity and mutilans type of destruction should be treated with a wrist fusion procedure. Swanson silicone wrist implants in patients with rheumatoid arthritis showed good clinical results with high patient satisfaction in a long-term follow-up study. The radiologic appearance of the implants shows substantial deterioration with significant cystic changes in the bone. This observation, however, has little direct correlation to residual pain or mobility. Silicone wrist arthroplasty remains a possible alternative to wrist fusion in low-demand patients with rheumatoid arthritis.
10.
11.
12.
13. 14. 15.
References 1. Cobb TK, Beckenbaugh RD. Biaxial total-wrist arthroplasty. J Hand Surg 1996;21A:1011–1021. 2. Costi J, Krishnan J, Pearcy M. Total wrist arthroplasty: a quantitative review of the last 30 years. J Rheumatol 1998; 25:451– 458. 3. Divelbiss BJ, Sollerman C, Adams BD. Early results of the Universal total wrist arthroplasty in rheumatoid arthritis. J Hand Surg 2002;27A:195–204. 4. Fatti JF, Palmer AK, Greenky S, Mosher JF. Long-term results of Swanson interpositional wrist arthroplasty: part II. J Hand Surg 1991;16A:432– 437. 5. Lundkvist L, Barfred T. Total wrist arthroplasty. Experience with Swanson flexible silicone implants, 1982–1988. Scand J Plast Reconstr Surg 1992;26:97–100. 6. Meuli HC, Fernandez DL. Uncemented total wrist arthroplasty. J Hand Surg 1995;20A:115–122. 7. Murphy DM, Khoury JG, Imbriglia JE, Adams BD. Comparison of arthroplasty and arthrodesis for the rheumatoid wrist. J Hand Surg 2003;28A:570 –576. 8. Swanson AB, de Groot Swanson G, Maupin BK. Flexible implant arthroplasty of the radiocarpal joint. Surgical technique and long-term study. Clin Orthop 1984;187:94 –106 9. Takwale VJ, Nuttall D, Trail IA, Stanley JK. Biaxial total wrist replacement in patients with rheumatoid arthritis. Clin-
16.
17.
18.
19.
20.
21.
22. 23.
1287
ical review, survivorship and radiological analysis. J Bone Joint Surg 2002;84B:692– 699. Vicar AJ, Burton RI. Surgical management of the rheumatoid wrist—fusion or arthroplasty. J Hand Surg 1986;11A: 790 –797. Summers B, Hubbard MJS. Wrist joint arthroplasty in rheumatoid arthritis: a comparison between the Meuli and Swanson prostheses. J Hand Surg 1984;9B:171–176. DeHeer DH, Owens SR, Swanson AB. The host response to silicone elastomer implants for small joint arthroplasty. J Hand Surg 1995;20A(Suppl):S101–S109. Huskisson EC. Measurement of pain. Lancet 1974;9:1127– 1131. Lluch A, Proubasta L. Les implants radio-carpeins de Swanson. Resultat a longterm. La Main 1998;3:176 –184. Schill S, Thabe H, Mohr W. Langzeitergebnisse nach Swanson-Prosthesenversorgung des rhematischen Handgelenks. [Long-term outcome of Swanson prosthesis management of the rheumatic wrist joint]. Handchir Mikrochir Plast Chir 2001;33:198 –206. Stanley JK, Tolat AR. Long-term results of Swanson Silastic arthroplasty in the rheumatoid wrist. J Hand Surg 1993;18B: 381–388. Fatti JF, Palmer AK, Mosher JF. The long-term results of Swanson silicone rubber interpositional wrist arthroplasty. J Hand Surg 1986;11A:166 –175. Comstock CP, Louis DS, Eckenrode JF. Silicone wrist implant: long-term follow-up study. J Hand Surg 1988;13A: 201–205. Jolly SL, Ferlic DC, Clayton ML, Dennis DA, Stringer EA. Swanson silicone arthroplasty of the wrist in rheumatoid arthritis: a long-term follow-up. J Hand Surg 1992;17A:142– 149. Brase DW, Millender LH. Failure of silicone rubber wrist arthroplasty in rheumatoid arthritis. J Hand Surg 1986;11A: 175–183. Palmer AK, Werner FW, Murphy D, Glisson BS. Functional wrist motion: a biomechanical study. J Hand Surg 1985; 10A:39 – 46. Carlson JR, Simmons BP. Wrist arthrodesis after failed wrist implant arthroplasty. J Hand Surg 1988;23A:893– 898. Vögelin E, Nagy L. Fate of failed Meuli total wrist arthroplasty. J Hand Surg 2003;28B:61– 68.