- Email: [email protected]
Arthroscopic Shoulder Synovectomy in Patients With Rheumatoid Arthritis
Arthroscopic Shoulder Synovectomy in Patients With Rheumatoid Arthritis Adam M. Smith, M.D., John W. Sperling, M.D., Shawn W. O’Driscoll, Ph.D., M.D., and Robert H. Cofield, M.D.
Purpose: Currently, there is little information available concerning the results of shoulder synovectomy in patients with rheumatoid arthritis. Furthermore, it remains difficult to assess the success of shoulder synovectomy because of a high association of rotator cuff tears in rheumatoid patients. We hypothesized that synovectomy in patients with a functionally intact rotator cuff would provide durable pain relief. Type of Study: Case series. Methods: Sixteen shoulders in 13 patients with rheumatoid arthritis were treated with arthroscopic shoulder synovectomy from 1988 to 2002 with a mean follow-up of 5.5 years. Patients with full-thickness rotator cuff tears or partial tears that required repair were excluded. Two of the 13 patients had been diagnosed with juvenile rheumatoid arthritis and 11 had adult-onset disease. The medical records of the patients were reviewed and patients were assessed clinically and by questionnaire to assess pain, satisfaction, range of motion, radiographic outcomes, and occurrence of complications. Results: There was an improvement in pain at long-term follow-up in 13 of 16 patients (P ⬍ .001). Active shoulder elevation improved from a median of 145° to 160° but was not significant (P ⫽ .14). External rotation improved significantly (P ⫽ .01) from a median of 45° (range, 0° to 90°) to 60° (range, 0° to 100°). According to the Neer rating system, there were 5 excellent, 8 satisfactory, and 3 unsatisfactory results. Seven of 8 shoulders followed-up radiographically for more than 1 year showed radiographic progression of disease. All 3 patients reporting pain that was no better or worse than before surgery had radiographic arthrosis at last follow-up with advancing periarticular erosions and reduction of glenoid articular space. Conclusions: Arthroscopic synovectomy of the shoulder in patients with rheumatoid arthritis with an intact rotator cuff offers a reliable decrease in pain with less predictable improvements in range of motion. Limitations in predicting final results based on preoperative radiographs should be discussed with patients; those with more advanced radiographic changes may not benefit from the procedure. Level of Evidence: Level IV, therapeutic study, case series, no control group. Key Words: Rheumatoid arthritis—Synovectomy.
C
urrently there is very little information in the literature regarding synovectomy of the shoulder in patients with rheumatoid arthritis.1-8 Although most of these studies report good pain relief, concerns exist regarding the efficacy and long-term results of syno-
From the Department of Orthopedic Surgery (J.W.S., S.W.O., R.H.C.), Mayo Clinic, Rochester, Minnesota, U.S.A.; and Kentucky Sports Medicine (A.M.S.), Lexington, Kentucky, U.S.A. Address correspondence and reprint requests to John W. Sperling, M.D., Mayo Clinic, 200 First St SW, Rochester, MN 55905, U.S.A. E-mail: [email protected] E-mail: sperling.john@ mayo.edu © 2006 by the Arthroscopy Association of North America 0749-8063/06/2201-4522$32.00/0 doi:10.1016/j.arthro.2005.10.011
50
vectomy in these patients especially when glenohumeral changes are present on radiographs.4,9 Furthermore, rotator cuff abnormalities in this patient population have ranged from 80% to less than 25% in patients undergoing other reconstructive procedures,2,10-14 which can make interpreting success of synovectomy difficult.6 Nonarthroplasty surgical intervention for patients with rheumatoid arthritis can be technically challenging. Many patients have pre-existing shoulder pathology15,16 including limited range of motion16-18 and radiographic changes including osteopenia, proximal humeral migration, and glenohumeral and acromioclavicular erosions.14,19-21 Currently, there is no published information to guide
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 22, No 1 (January), 2006: pp 50-56
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA clinical decision-making about arthroscopic synovectomy in patients with rheumatoid arthritis with a functionally intact rotator cuff.22 We hypothesized that shoulder synovectomy would provide durable pain relief and functional gains in patients with rheumatoid arthritis. This study examines our results, risk factors for an unsatisfactory outcome, and the rates of failed treatment of synovectomy of the shoulder in patients with rheumatoid arthritis. METHODS Patient Demographics A computer-assisted search of the surgical database at this institution was performed from January 1988 to December 2002 to identify patients who had a preoperative diagnosis of rheumatoid arthritis and underwent arthroscopic synovectomy of the shoulder. Thirtysix such shoulders were identified. Approval for this study was obtained from the Institutional Review Board for human research. Seventeen shoulders in 17 patients had a full-thickness rotator cuff tear or a significant partial-thickness tear requiring repair and were excluded. Three shoulders in 3 patients had less than 1 year of follow-up and were excluded. This left 16 shoulders in 13 patients who underwent arthroscopic synovectomy by 3 surgeons who specialize in shoulder surgery. All patients had failed multiple trials of medical management including changes in oral medications, rest, ice, and therapy. All patients had undergone at least 2 intra-articular or subacromial steroid injections. Indications for surgery included a failed nonoperative management program with continued moderate or severe pain with shoulder activity. Fifteen of the 16 patients identified night pain that limited sleep. All patients had crepitus, and pain with overhead activity. A total of 16 shoulders in 13 patients (3 patients had bilateral operations) were identified who met the defined criteria with a mean follow-up of 5.5 years (range, 1 to 10 years). The mean age of the 13 patients at the time of the 16 synovectomies was 49 years (range, 28 to 71 years); there were 7 women and 6 men. Nine procedures were performed in patients less than 50 years of age and 7 that were older than age 50. Eleven patients had adult-onset rheumatoid arthritis, and 2 patients had been diagnosed with juvenile-onset rheumatoid arthritis before surgical intervention. All patients met criteria for rheumatoid arthritis established by the American Rheumatism Association.23 Four classes of medications (nonsteroidal, disease-
51
modifying, oral corticosteroid, and biological medications) were used in patients preoperatively to control the manifestations of the rheumatoid process, with all 13 patients taking oral nonsteroidal medications. Twelve patients were taking disease-modifying medications including methotrexate, sulfasalazine, minocycline, hydroxychloroquine sulfate, or azathioprine, and 12 patients were taking oral corticosteroids. Three patients were taking newer biologic agents such as etanercept. All patients were taking medications from at least 2 classes, 11 were taking medications from 3 classes, and 3 were taking medications from all 4 drug classes. A review of the patients’ medical records, including associated surgical procedures and available imaging studies, was performed. Data were collected from the last office visit on operative and postoperative complications, preoperative and postoperative pain, shoulder function, range of motion, strength, and overall patient function. Shoulder pain was graded as: 1 ⫽ no pain, 2 ⫽ slight pain, 3 ⫽ pain after unusual activity, 4 ⫽ moderate pain, and 5 ⫽ severe pain. All patients were mailed a questionnaire to evaluate shoulder pain, satisfaction with the operative shoulder, and overall patient function with the Simple Shoulder Test (SST), and the patient section of the American Shoulder and Elbow Surgeons (ASES) assessment instrument.24,25 There was 1 patient who chose not to participate in the questionnaire portion of the study, and 1 patient who underwent revision surgery, which left 14 shoulders that had outcome assessment with questionnaire. Radiograph Assessment Perioperative radiographs of the affected shoulder were examined, and a consensus was reached by 3 orthopaedic surgeons with specialty training in surgery of the shoulder. Each shoulder was staged according to the system defined by Crossan and Vallance.3,4,26-28 Shoulders with stage 1 disease had no abnormal features except for mild osteopenia. In stage 2, there is a spherical humeral head with a normal glenohumeral joint, but with humeral head erosions (Fig 1). Stage 3 is characterized by proximal humeral migration with a maintained glenohumeral joint space. Stage 4 is characterized by reduced overall joint space and a spherical humeral head (Fig 2). Usually, there is proximal humeral migration and periarticular marginal erosions. Patients with stage 5 changes have distortion of the glenohumeral joint with glenoid erosions and medial humeral displacement. Acromiohumeral distance was measured on the an-
52
A. M. SMITH ET AL. postoperative day with active-assisted motion and progressed as tolerated. Grading of Results The patient self-report section of the ASES was used24 in addition to the SST to assess patient function,25 and the results were rated according to the criteria by Neer and Cofield.10,30 In order to have excellent result rating, patients had to have at least 55° of external rotation, active elevation of at least 145°, and no pain. Patients were considered to have a satisfactory result if they had no pain, slight pain, or occasionally moderate pain, with external rotation to 30° or greater, and active elevation between 100° and 145°. If these criteria were not met, or the patient underwent further surgery, the result was considered unsatisfactory. Statistical Analysis
FIGURE 1. Anteroposterior view of a left shoulder with stage II Crossan and Vallance changes with a spherical humeral head, slight narrowing of the glenohumeral and acromiohumeral distance, and periarticular humeral head erosions.
teroposterior view of the proximal humerus in external rotation from the midportion of the acromion to the highest portion of the humeral head. Measurements were placed in 3 groups: ⱖ6 mm, less than 6 to 3 mm, or less than 3 mm.
Among the entire group of 16 shoulders, improvement from preoperative to postoperative measures was estimated for overall pain, active elevation, external rotation, Neer rating, and patient satisfaction, using the Wilcoxon signed-rank test. Patients considered to have minimal radiographic changes (Crossan and Vallance stages 1 and 2) were compared with those con-
Operative Technique Patients were positioned in the beach-chair or lateral position with the affected shoulder widely prepped. Standard arthroscopic portals were used to gain access to the shoulder and subacromial space.29 Debridement of the glenohumeral joint and hypertrophic synovial tissue was carried out using an oscillating shaver and a cautery/ablation device. Attempts were made to perform a complete synovectomy of all pathologic appearing tissue, and hemostasis was achieved prior to completion of the procedure. Twelve of 16 shoulders underwent arthroscopic bursectomy. Additional procedures included biceps tendon debridement (6 patients), acromioplasty (3 patients), distal clavicle excision (1 patient), labrum debridement for degenerative fraying (1 patient), and complete capsulectomy (1 patient). Six shoulders with a partial-thickness supraspinatus tear deemed to not require repair had minimal debridement of the frayed tendon edges. Postoperatively, patients were placed in a sling for comfort and rehabilitation began on the first
FIGURE 2. Anteroposterior view of a right shoulder with stage IV Crossan and Vallance changes with a spherical humeral head, moderate upward humeral subluxation, and absent glenohumeral joint space.
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA TABLE 1. Clinical Outcomes
Preop (min, max)
Overall pain scored 1-5 (16 shoulders) Active elevation (degrees) (14 shoulders) External rotation (degrees) (14 shoulders) ASES (14 shoulders) SST (14 shoulders) Satisfaction (14 shoulders)
4 (4, 5)
53
Clinical Outcomes
Postop (min, max) 2 (1, 3)
Change (postop, preop)
Wilcoxon Signed-Rank P Value
⫺2 (⫺4, ⫺1)
⬍.01*
145 (110, 160)
160 (120, 180)
25 (0, 40)
.14
45 (20, 50) — — 1 (1, 3)
60 (30, 80) 62 (47, 67) 8 (6, 11) 7 (5, 8)
20 (10, 30) — — 5 (3, 7)
.01* — — ⬍.01*
NOTE. Values are expressed as median (25th and 75th percentile). *Statistically significant.
sidered to have more advanced radiographic changes (Crossan and Vallance stages 3 and 4) for postoperative active elevation (preoperative to postoperative) and change in external rotation (preoperative to postoperative) using the Wilcoxon rank sum test; P ⬍ .05 was used to determine significance. RESULTS
examination with the ASES and SST), the mean ASES score was 62 points (out of 100 possible) and the mean number of “yes” responses to the SST was 8 (out of 12 possible) (Table 1). Satisfaction was assessed on a 10-point scale for the operative shoulder before surgery and by questionnaire. Except for the 1 patient requiring revision surgery with arthroplasty, all patients expressed a higher level of satisfaction with their shoulder than before surgery (P ⬍ .001).
Clinical Results At last follow-up, 13 of the 16 shoulders had improved pain (P ⬍ .001) with the median change from preoperative to postoperative of 2 points. There were 4 patients with no pain, 6 with slight pain, 3 with pain only after unusual activity, and 3 with severe pain (Table 1). Of the 3 patients with severe pain, 2 patients complained of pain that was worse than before the initial procedure (1 at 1 year postoperative, and 1 at 4 years). One other patient had no improvement of pain after synovectomy 7.5 years after the procedure. Range of motion was assessed in 14 patients (1 patient declined participation in the follow-up portion of the study and 1 patient had undergone revision surgery with arthroplasty). Median active elevation improved from 145° to a median of 160°, but was not significantly different (P ⫽ .14). External rotation did significantly improve (P ⫽ .01) from a preoperative median of 45° to a median of 60° postoperatively (Table 1). Postoperative active elevation and external rotation was the same or diminished in 4 and 2 shoulders, respectively. For all 16 shoulders at last follow-up, there were 5 excellent, 8 satisfactory, and 3 unsatisfactory results. Patients with 14 of the 16 shoulders participated in the questionnaire portion of the study (1 patient did not participate and 1 patient had revision surgery before
Radiographic Results During the perioperative period, according to radiographic assessment with the Crossan and Vallance staging system, there were 7 patients with stage 1, 1 with stage 2, 1 with stage 3, and 7 with stage 4. Shoulders were evaluated based on these radiographs and were separated into groups with 8 patients having minimal radiographic changes (Crossan and Vallance stage 1 or 2) and 8 patients having more advanced radiographic changes (Crossan and Vallance stage 3 and 4). The 2 groups were evaluated for range of motion, ASES, and SST scores (Table 2). Statistical analysis was not performed because of the small number of patients. The patients were also evaluated by the Neer rating system.30 In patients with minimal radiographic changes, there were 3 excellent, 3 satisfactory, and 2 unsatisfactory results. In patients with more advanced radiographic changes, there were 2 excellent, 5 satisfactory, and 1 unsatisfactory result. The acromiohumeral distance was greater than 6 mm in 11 shoulders, narrowed from 3 to 6 mm in 4 shoulders, and was less than 3 mm in 1 shoulder. Of the 5 patients with radiographic findings of decreased acromiohumeral distance, 1 had minimal partial-thickness rotator cuff tearing and 4 had cuff thinning at arthroscopy.
54
A. M. SMITH ET AL. TABLE 2.
Clinical Outcomes Based on Radiographs
Crossan and Vallance Stage 1 and 2 (n ⫽ 8)
Crossan and Vallance Stages 3 and 4 (n ⫽ 8)
Clinical Outcomes
Mean Preop (range)
Mean Postop (range)
Mean Preop (range)
Mean Postop (range)
Overall pain score Active elevation (degrees) External rotation (degrees) ASES SST* Satisfaction
4.3 (3–5) 145 (140–170) 52 (20–90) — — 1 (1–3)
2.8 (1–5) 169 (130–180) 69 (0–90) 67 (42–90) 9 (8–12) 7 (5–9)
4.5 (4–5) 122 (30–165) 29 (0–50) — — 2 (1–4)
2.3 (1–5) 134 (90–180) 45 (30–80) 58 (33–83) 8 (5–11) 7 (4–8)
*Simple Shoulder Test, number of “yes” responses.
Eight shoulders were followed-up radiographically for more than 1 year (mean, 4 years; range, 1 to 9 years). The remaining 8 shoulders did not have radiographs available for comparison. Seven of the 8 shoulders showed radiographic progression of disease. One patient with stage 1 initial radiographs did not have progressive articular changes at 4 years. Of the patients with minimal radiographic changes, 4 of the 5 (Crossan and Vallance stage 1 or 2 disease) had evidence of advancing radiographic changes. Two patients with stage 1 disease progressed to stage 2 disease (at 3 and 9 years). One patient with stage 1 progressed to stage 3 at 5 years, and 1 patient with stage 2 at the index operation progressed to stage 4. Two of the 5 patients with minimal radiographic changes (stage 1 or 2) showed progressive acromiohumeral narrowing from greater than 6 mm to 3 to 6 mm. All 3 remaining patients with more advanced Crossan and Vallance stage 4 disease at the index operation had disease progression at the last follow-up to stage 5 radiographic disease (2 patients at 1 year and 1 at 7 years). One of the 3 patients progressed to an acromiohumeral distance of less than 3 mm. Of these 7 shoulders with radiographic progression, the mean active elevation improved from 125° (range, 30° to 170°) to 159° (range, 120° to 180°). There were 2 excellent, 3 satisfactory, and 2 unsatisfactory results. Complications and Revisions One patient (age 40 years at the index procedure) required revision shoulder surgery with total shoulder arthroplasty because of continued pain 11 months after arthroscopic synovectomy. This patient’s radiographs revealed rapid progression (1 year from synovectomy) with loss of glenohumeral space, proximal humeral migration, and periarticular marginal erosions. Two other patients (age 28 and 42 years) are
currently considering arthroplasty at 5 and 9 years after the index procedure because of increasing discomfort and dysfunction after several years of excellent pain relief. Both patients had Crossan and Vallance stage 4 changes at the index procedure, with 1 of these progressing to stage 5 over a 7-year period. There were no infections, neurologic injuries, or hematomas requiring reoperation in this series. DISCUSSION Although shoulder synovectomy in patients with rheumatoid arthritis has been examined in a limited number of patients and reported in the literature,4-8 synovectomy of the knee has been examined more extensively in larger, more homogenous patient groups. Good early results were often tempered by concerns about waning pain relief and loss of functional improvement.9,31-34 These same issues have been raised with shoulder synovectomy. Pahle5,6 reported on the largest series of patients (54 shoulders) undergoing open surgical synovectomy, noting good pain relief even in advanced cases of arthritis with increased abduction and external rotation. He did caution, however, that radiographs continued to show advancing disease. Peterson7 noted similar findings with pain relief and functional gains, with 2 patients requiring later shoulder arthroplasty. However, both of these studies included patients with rotator cuff tearing, limiting the interpretation of outcome in the assessment of isolated shoulder synovectomy. The results in our patients though are similar to the findings of Pahle and Peterson in their use of open shoulder synovectomy.5-7 Pain relief and patient satisfaction were high. Improvement in function could occur and often did but was not as predictable. Synovectomy does not seem to prevent progression of the rheumatoid arthritis, with advancing radio-
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA graphic changes noted in 7 of our patients.35 Kelly4 recommended that synovectomy must be done before significant damage to articular cartilage has occurred. Although synovectomy would be ideally performed in rheumatoid patients with normal shoulder articular contour, this study shows that pain relief can be achieved despite moderately advanced radiographic changes. However, in the study by Wakitani et al.36 in patients with a wider spectrum of radiographic changes, those with severe arthritis had less favorable results with synovectomy compared with those undergoing prosthetic arthroplasty procedures. This study supports the work of Cruess17 in that acromiohumeral distance was not found to be a good predictor of rotator cuff tearing in this patient population. Only 1 of the 5 patients with diminished acromiohumeral distance on radiographs was found to have a partial-thickness tear under direct arthroscopic visualization. If rotator cuff tearing is suspected, we recommend further evaluation with magnetic resonance imaging so that appropriate preoperative planning may be performed. The technique of open shoulder synovectomy has been described as difficult.1,3,5-7,17,37 In this series of arthroscopic synovectomy, bleeding was encountered that was often difficult to control and made visualization difficult. Meticulous hemostasis is necessary intraoperatively to facilitate adequate viewing, and hemostasis should be achieved before termination of the procedure to avoid a significant hematoma. Caution must be taken with the use of cautery or ablation techniques in the inferior capsular recess regions because injury to the axillary nerve has been reported with other procedures.38 Several other authors have also cautioned against sacrificing the coracoacromial ligament, thus compromising the coracoacromial arch due to concerns about rheumatoid disease progression, increasing rotator cuff deficiency, and the development of severe upward and forward humeral subluxation.12,37 We do not divide the coracoacromial ligament or perform acromioplasty in this patient population unless impingement or wear is directly visualized on the rotator cuff tendons. In this study, only 3 shoulders had surgical manipulation of the coracoacromial arch. The limitations in this study are related to its retrospective nature, including the small number of patients, absence of a nonoperative comparison group, and the lack of a control group. Also, none of the patients in this study had the most severe radiographic changes associated with Crossan and Vallance stage 5 disease. Surgeon bias as to which patients were of-
55
fered synovectomy should be considered when evaluating these results. The results presented should not be misinterpreted to include patients with full-thickness rotator cuff tearing. Although many rheumatoid patients with shoulder pain have rotator cuff tearing, the focus of treatment is usually directed at the rotator cuff pathology using synovectomy as an adjunctive procedure. This is the first study to examine arthroscopic synovectomy as a primary treatment for shoulder synovitis in patients with an intact rotator cuff. Patients with rheumatoid arthritis with shoulder involvement frequently have substantial associated pathologic problems such as capsular stiffness,16,18 more advanced cartilage loss than is apparent radiographically,7,14,19-21 and rotator cuff tearing that can be severe. Although 13 of the 16 shoulders in this study had a satisfactory outcome, the prognostic factors for an unsatisfactory outcome with arthroscopic synovectomy remain to be determined. The results presented represent the best scenario with an intact rotator cuff. When considering surgical outcomes, the limitations of arthroscopic synovectomy should be discussed with patients. There is a greater expectation for pain relief and patient satisfaction, less predictability for gains of motion, and no certainty that radiographic progression of the disease will be halted. REFERENCES 1. Bennett WF, Gerber C. Operative treatment of the rheumatoid shoulder. Curr Opin Rheumatol 1994;6:177-182. 2. Clayton ML, Ferlic DC. Surgery of the shoulder in rheumatoid arthritis: A report of nineteen patients. Clin Orthop 1975;106: 166-174. 3. Cuomo F, Greller MJ, Zuckerman JD. The rheumatoid shoulder. Rheum Dis Clin North Am 1998;24:67-82. 4. Kelly IG. Surgery of the rheumatoid shoulder. Ann Rheum Dis 1990;49:824-829 (suppl 2). 5. Pahle JA. The shoulder joint in rheumatoid arthritis: Synovectomy. Reconstr Surg Traumatol 1981;18:33-47. 6. Pahle JA, Kvarnes L. Shoulder synovectomy. Ann Chir Gynaecol Suppl 1985;198:37-39. 7. Petersson CJ. Shoulder surgery in rheumatoid arthritis. Acta Orthop Scand 1986;57:222-226. 8. Smith-Peterson MN, Aufranc OE, Larson CB. Useful surgical procedures for rheumatoid arthritis involving the upper extremity. Arch Surg 1943;46:764-770. 9. Ochi T, Iwase R, Kimura T, et al. Effect of early synovectomy on the course of rheumatoid arthritis. J Rheumatol 1991;18: 1794-1798. 10. Cofield RH. Total shoulder arthroplasty with Neer prosthesis. J Bone Joint Surg Am 1984;66:899-906. 11. Ennevarra K. Painful shoulder joint in rheumatoid arthritis: A clinical and radiologic study of 200 cases with special reference to arthrography of the glenohumeral joint. Acta Rheum Scand 1967;1-108 (suppl II). 12. Kelly IG, Foster RS, Fisher WD. Neer total shoulder replacement in rheumatoid arthritis. J Bone Joint Surg Br 1987;69: 723-726.
56
A. M. SMITH ET AL.
13. Rozing PM, Brand R. Rotator cuff repair during shoulder arthroplasty in rheumatoid arthritis. J Arthroplasty 1998;13: 311-319. 14. Weiss JJ, Thompson GR, Doust V, Burgener F. Rotator cuff tears in rheumatoid arthritis. Arch Intern Med 1975;135:521525. 15. Curran JF, Ellman MH, Brown NL. Rheumatologic aspects of painful conditions affecting the shoulder. Clin Orthop 1983; 173:27-37. 16. Laine VAI. Shoulder affections in rheumatoid arthritis. Ann Rheum Dis 1954;13:157-160. 17. Cruess RL. Rheumatoid arthritis of the shoulder. Orthop Clin North Am 1980;11:333-342. 18. Dabrowski W, Fonseka N, Ansell BM, Liyanage IS, Arden GP. Shoulder problems in juvenile chronic polyarthritis. Scand J Rheumatol 1979;8:49-53. 19. Hirooka A, Wakitani S, Yoneda M, Ochi T. Shoulder destruction in rheumatoid arthritis: Classification and prognostic signs in 83 patients followed 5-23 years. Acta Orthop Scand 1996; 67:258-263. 20. Lehtinen JT, Belt EA, Lyback CO, et al. Subacromial space in the rheumatoid shoulder: A radiographic 15-year follow-up study of 148 shoulders. J Shoulder Elbow Surg 2000;9:183187. 21. Petersson CJ. The acromioclavicular joint in rheumatoid arthritis. Clin Orthop 1987;223:86-93. 22. Witwity T, Uhlmann R, Nagy MH, Bhasin VB, Bahgat MM, Singh AK. Shoulder rheumatoid arthritis associated with chondromatosis, treated by arthroscopy. Arthroscopy 1991;7:233236. 23. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315324. 24. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-594. 25. Roddey TS, Olson SL, Cook KF, Gartsman GM, Hanten W. Comparison of the University of California-Los Angeles
26. 27. 28. 29. 30. 31. 32.
33. 34. 35. 36. 37. 38.
Shoulder Scale and the Simple Shoulder Test with the shoulder pain and disability index: Single administration reliability and validity. Phys Ther 2000;80:759-768. Crossan JF, Vallance R. Clinical and radiological features of the shoulder joint in rheumatoid arthritis. J Bone Joint Surg Br 1980;62:116. Crossan JF, Vallance R. The shoulder joint in rheumatoid arthritis. In: Bayley I, Keddel L, eds. Shoulder surgery. New York: Springer-Verlag, 1982;131-139. Souter WA. The surgical treatment of the rheumatoid shoulder. Ann Acad Med Singapore 1983;12:243-255. Jobe CM. Shoulder arthroscopy. In: Ianotti JP, Williams GP, eds. Disorders of the shoulder: Diagnosis and management. Philadelphia: Lippincott Wilkins & Williams, 1999;883-910. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: A preliminary report. J Bone Joint Surg Am 1972;54:41-50. Bryan RS, Peterson LF. Synovectomy of the knee. Orthop Clin North Am 1971;2:705-713. Geens S, Clayton ML, Leidholt JD, Smyth CJ, Bartholomew BA. Synovectomy and debridement of the knee in rheumatoid arthritis. II. Clinical and roentgenographic study of thirty-one cases. J Bone Joint Surg Am 1969;51:626-642. Paradies LH. Synovectomy for rheumatoid arthritis of the knee. J Bone Joint Surg Am 1975;57:95-100. Ranawat CS, Desai K. Role of early synovectomy of the knee joint in rheumatoid arthritis. Arthritis Rheum 1975;18:117121. Kaarela K, Kautiainen J. Continuous progression of radiologic destruction on seropositive rheumatoid arthritis. J Rheumatol 1997;24:1285-1287. Wakitani S, Imoto K, Saito M, et al. Evaluation of surgeries for rheumatoid shoulder based on the destruction pattern. J Rheumatol 1999;26:41-46. Chen AL, Joseph TN, Zuckerman JD. Rheumatoid arthritis of the shoulder. J Am Acad Orthop Surg 2003;11:12-24. Wong KL, Williams GR. Complications of thermal capsulorrhaphy of the shoulder. J Bone Joint Surg Am 2001;83:S151S155 (suppl).
Purpose: Currently, there is little information available concerning the results of shoulder synovectomy in patients with rheumatoid arthritis. Furthermore, it remains difficult to assess the success of shoulder synovectomy because of a high association of rotator cuff tears in rheumatoid patients. We hypothesized that synovectomy in patients with a functionally intact rotator cuff would provide durable pain relief. Type of Study: Case series. Methods: Sixteen shoulders in 13 patients with rheumatoid arthritis were treated with arthroscopic shoulder synovectomy from 1988 to 2002 with a mean follow-up of 5.5 years. Patients with full-thickness rotator cuff tears or partial tears that required repair were excluded. Two of the 13 patients had been diagnosed with juvenile rheumatoid arthritis and 11 had adult-onset disease. The medical records of the patients were reviewed and patients were assessed clinically and by questionnaire to assess pain, satisfaction, range of motion, radiographic outcomes, and occurrence of complications. Results: There was an improvement in pain at long-term follow-up in 13 of 16 patients (P ⬍ .001). Active shoulder elevation improved from a median of 145° to 160° but was not significant (P ⫽ .14). External rotation improved significantly (P ⫽ .01) from a median of 45° (range, 0° to 90°) to 60° (range, 0° to 100°). According to the Neer rating system, there were 5 excellent, 8 satisfactory, and 3 unsatisfactory results. Seven of 8 shoulders followed-up radiographically for more than 1 year showed radiographic progression of disease. All 3 patients reporting pain that was no better or worse than before surgery had radiographic arthrosis at last follow-up with advancing periarticular erosions and reduction of glenoid articular space. Conclusions: Arthroscopic synovectomy of the shoulder in patients with rheumatoid arthritis with an intact rotator cuff offers a reliable decrease in pain with less predictable improvements in range of motion. Limitations in predicting final results based on preoperative radiographs should be discussed with patients; those with more advanced radiographic changes may not benefit from the procedure. Level of Evidence: Level IV, therapeutic study, case series, no control group. Key Words: Rheumatoid arthritis—Synovectomy.
C
urrently there is very little information in the literature regarding synovectomy of the shoulder in patients with rheumatoid arthritis.1-8 Although most of these studies report good pain relief, concerns exist regarding the efficacy and long-term results of syno-
From the Department of Orthopedic Surgery (J.W.S., S.W.O., R.H.C.), Mayo Clinic, Rochester, Minnesota, U.S.A.; and Kentucky Sports Medicine (A.M.S.), Lexington, Kentucky, U.S.A. Address correspondence and reprint requests to John W. Sperling, M.D., Mayo Clinic, 200 First St SW, Rochester, MN 55905, U.S.A. E-mail: [email protected] E-mail: sperling.john@ mayo.edu © 2006 by the Arthroscopy Association of North America 0749-8063/06/2201-4522$32.00/0 doi:10.1016/j.arthro.2005.10.011
50
vectomy in these patients especially when glenohumeral changes are present on radiographs.4,9 Furthermore, rotator cuff abnormalities in this patient population have ranged from 80% to less than 25% in patients undergoing other reconstructive procedures,2,10-14 which can make interpreting success of synovectomy difficult.6 Nonarthroplasty surgical intervention for patients with rheumatoid arthritis can be technically challenging. Many patients have pre-existing shoulder pathology15,16 including limited range of motion16-18 and radiographic changes including osteopenia, proximal humeral migration, and glenohumeral and acromioclavicular erosions.14,19-21 Currently, there is no published information to guide
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 22, No 1 (January), 2006: pp 50-56
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA clinical decision-making about arthroscopic synovectomy in patients with rheumatoid arthritis with a functionally intact rotator cuff.22 We hypothesized that shoulder synovectomy would provide durable pain relief and functional gains in patients with rheumatoid arthritis. This study examines our results, risk factors for an unsatisfactory outcome, and the rates of failed treatment of synovectomy of the shoulder in patients with rheumatoid arthritis. METHODS Patient Demographics A computer-assisted search of the surgical database at this institution was performed from January 1988 to December 2002 to identify patients who had a preoperative diagnosis of rheumatoid arthritis and underwent arthroscopic synovectomy of the shoulder. Thirtysix such shoulders were identified. Approval for this study was obtained from the Institutional Review Board for human research. Seventeen shoulders in 17 patients had a full-thickness rotator cuff tear or a significant partial-thickness tear requiring repair and were excluded. Three shoulders in 3 patients had less than 1 year of follow-up and were excluded. This left 16 shoulders in 13 patients who underwent arthroscopic synovectomy by 3 surgeons who specialize in shoulder surgery. All patients had failed multiple trials of medical management including changes in oral medications, rest, ice, and therapy. All patients had undergone at least 2 intra-articular or subacromial steroid injections. Indications for surgery included a failed nonoperative management program with continued moderate or severe pain with shoulder activity. Fifteen of the 16 patients identified night pain that limited sleep. All patients had crepitus, and pain with overhead activity. A total of 16 shoulders in 13 patients (3 patients had bilateral operations) were identified who met the defined criteria with a mean follow-up of 5.5 years (range, 1 to 10 years). The mean age of the 13 patients at the time of the 16 synovectomies was 49 years (range, 28 to 71 years); there were 7 women and 6 men. Nine procedures were performed in patients less than 50 years of age and 7 that were older than age 50. Eleven patients had adult-onset rheumatoid arthritis, and 2 patients had been diagnosed with juvenile-onset rheumatoid arthritis before surgical intervention. All patients met criteria for rheumatoid arthritis established by the American Rheumatism Association.23 Four classes of medications (nonsteroidal, disease-
51
modifying, oral corticosteroid, and biological medications) were used in patients preoperatively to control the manifestations of the rheumatoid process, with all 13 patients taking oral nonsteroidal medications. Twelve patients were taking disease-modifying medications including methotrexate, sulfasalazine, minocycline, hydroxychloroquine sulfate, or azathioprine, and 12 patients were taking oral corticosteroids. Three patients were taking newer biologic agents such as etanercept. All patients were taking medications from at least 2 classes, 11 were taking medications from 3 classes, and 3 were taking medications from all 4 drug classes. A review of the patients’ medical records, including associated surgical procedures and available imaging studies, was performed. Data were collected from the last office visit on operative and postoperative complications, preoperative and postoperative pain, shoulder function, range of motion, strength, and overall patient function. Shoulder pain was graded as: 1 ⫽ no pain, 2 ⫽ slight pain, 3 ⫽ pain after unusual activity, 4 ⫽ moderate pain, and 5 ⫽ severe pain. All patients were mailed a questionnaire to evaluate shoulder pain, satisfaction with the operative shoulder, and overall patient function with the Simple Shoulder Test (SST), and the patient section of the American Shoulder and Elbow Surgeons (ASES) assessment instrument.24,25 There was 1 patient who chose not to participate in the questionnaire portion of the study, and 1 patient who underwent revision surgery, which left 14 shoulders that had outcome assessment with questionnaire. Radiograph Assessment Perioperative radiographs of the affected shoulder were examined, and a consensus was reached by 3 orthopaedic surgeons with specialty training in surgery of the shoulder. Each shoulder was staged according to the system defined by Crossan and Vallance.3,4,26-28 Shoulders with stage 1 disease had no abnormal features except for mild osteopenia. In stage 2, there is a spherical humeral head with a normal glenohumeral joint, but with humeral head erosions (Fig 1). Stage 3 is characterized by proximal humeral migration with a maintained glenohumeral joint space. Stage 4 is characterized by reduced overall joint space and a spherical humeral head (Fig 2). Usually, there is proximal humeral migration and periarticular marginal erosions. Patients with stage 5 changes have distortion of the glenohumeral joint with glenoid erosions and medial humeral displacement. Acromiohumeral distance was measured on the an-
52
A. M. SMITH ET AL. postoperative day with active-assisted motion and progressed as tolerated. Grading of Results The patient self-report section of the ASES was used24 in addition to the SST to assess patient function,25 and the results were rated according to the criteria by Neer and Cofield.10,30 In order to have excellent result rating, patients had to have at least 55° of external rotation, active elevation of at least 145°, and no pain. Patients were considered to have a satisfactory result if they had no pain, slight pain, or occasionally moderate pain, with external rotation to 30° or greater, and active elevation between 100° and 145°. If these criteria were not met, or the patient underwent further surgery, the result was considered unsatisfactory. Statistical Analysis
FIGURE 1. Anteroposterior view of a left shoulder with stage II Crossan and Vallance changes with a spherical humeral head, slight narrowing of the glenohumeral and acromiohumeral distance, and periarticular humeral head erosions.
teroposterior view of the proximal humerus in external rotation from the midportion of the acromion to the highest portion of the humeral head. Measurements were placed in 3 groups: ⱖ6 mm, less than 6 to 3 mm, or less than 3 mm.
Among the entire group of 16 shoulders, improvement from preoperative to postoperative measures was estimated for overall pain, active elevation, external rotation, Neer rating, and patient satisfaction, using the Wilcoxon signed-rank test. Patients considered to have minimal radiographic changes (Crossan and Vallance stages 1 and 2) were compared with those con-
Operative Technique Patients were positioned in the beach-chair or lateral position with the affected shoulder widely prepped. Standard arthroscopic portals were used to gain access to the shoulder and subacromial space.29 Debridement of the glenohumeral joint and hypertrophic synovial tissue was carried out using an oscillating shaver and a cautery/ablation device. Attempts were made to perform a complete synovectomy of all pathologic appearing tissue, and hemostasis was achieved prior to completion of the procedure. Twelve of 16 shoulders underwent arthroscopic bursectomy. Additional procedures included biceps tendon debridement (6 patients), acromioplasty (3 patients), distal clavicle excision (1 patient), labrum debridement for degenerative fraying (1 patient), and complete capsulectomy (1 patient). Six shoulders with a partial-thickness supraspinatus tear deemed to not require repair had minimal debridement of the frayed tendon edges. Postoperatively, patients were placed in a sling for comfort and rehabilitation began on the first
FIGURE 2. Anteroposterior view of a right shoulder with stage IV Crossan and Vallance changes with a spherical humeral head, moderate upward humeral subluxation, and absent glenohumeral joint space.
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA TABLE 1. Clinical Outcomes
Preop (min, max)
Overall pain scored 1-5 (16 shoulders) Active elevation (degrees) (14 shoulders) External rotation (degrees) (14 shoulders) ASES (14 shoulders) SST (14 shoulders) Satisfaction (14 shoulders)
4 (4, 5)
53
Clinical Outcomes
Postop (min, max) 2 (1, 3)
Change (postop, preop)
Wilcoxon Signed-Rank P Value
⫺2 (⫺4, ⫺1)
⬍.01*
145 (110, 160)
160 (120, 180)
25 (0, 40)
.14
45 (20, 50) — — 1 (1, 3)
60 (30, 80) 62 (47, 67) 8 (6, 11) 7 (5, 8)
20 (10, 30) — — 5 (3, 7)
.01* — — ⬍.01*
NOTE. Values are expressed as median (25th and 75th percentile). *Statistically significant.
sidered to have more advanced radiographic changes (Crossan and Vallance stages 3 and 4) for postoperative active elevation (preoperative to postoperative) and change in external rotation (preoperative to postoperative) using the Wilcoxon rank sum test; P ⬍ .05 was used to determine significance. RESULTS
examination with the ASES and SST), the mean ASES score was 62 points (out of 100 possible) and the mean number of “yes” responses to the SST was 8 (out of 12 possible) (Table 1). Satisfaction was assessed on a 10-point scale for the operative shoulder before surgery and by questionnaire. Except for the 1 patient requiring revision surgery with arthroplasty, all patients expressed a higher level of satisfaction with their shoulder than before surgery (P ⬍ .001).
Clinical Results At last follow-up, 13 of the 16 shoulders had improved pain (P ⬍ .001) with the median change from preoperative to postoperative of 2 points. There were 4 patients with no pain, 6 with slight pain, 3 with pain only after unusual activity, and 3 with severe pain (Table 1). Of the 3 patients with severe pain, 2 patients complained of pain that was worse than before the initial procedure (1 at 1 year postoperative, and 1 at 4 years). One other patient had no improvement of pain after synovectomy 7.5 years after the procedure. Range of motion was assessed in 14 patients (1 patient declined participation in the follow-up portion of the study and 1 patient had undergone revision surgery with arthroplasty). Median active elevation improved from 145° to a median of 160°, but was not significantly different (P ⫽ .14). External rotation did significantly improve (P ⫽ .01) from a preoperative median of 45° to a median of 60° postoperatively (Table 1). Postoperative active elevation and external rotation was the same or diminished in 4 and 2 shoulders, respectively. For all 16 shoulders at last follow-up, there were 5 excellent, 8 satisfactory, and 3 unsatisfactory results. Patients with 14 of the 16 shoulders participated in the questionnaire portion of the study (1 patient did not participate and 1 patient had revision surgery before
Radiographic Results During the perioperative period, according to radiographic assessment with the Crossan and Vallance staging system, there were 7 patients with stage 1, 1 with stage 2, 1 with stage 3, and 7 with stage 4. Shoulders were evaluated based on these radiographs and were separated into groups with 8 patients having minimal radiographic changes (Crossan and Vallance stage 1 or 2) and 8 patients having more advanced radiographic changes (Crossan and Vallance stage 3 and 4). The 2 groups were evaluated for range of motion, ASES, and SST scores (Table 2). Statistical analysis was not performed because of the small number of patients. The patients were also evaluated by the Neer rating system.30 In patients with minimal radiographic changes, there were 3 excellent, 3 satisfactory, and 2 unsatisfactory results. In patients with more advanced radiographic changes, there were 2 excellent, 5 satisfactory, and 1 unsatisfactory result. The acromiohumeral distance was greater than 6 mm in 11 shoulders, narrowed from 3 to 6 mm in 4 shoulders, and was less than 3 mm in 1 shoulder. Of the 5 patients with radiographic findings of decreased acromiohumeral distance, 1 had minimal partial-thickness rotator cuff tearing and 4 had cuff thinning at arthroscopy.
54
A. M. SMITH ET AL. TABLE 2.
Clinical Outcomes Based on Radiographs
Crossan and Vallance Stage 1 and 2 (n ⫽ 8)
Crossan and Vallance Stages 3 and 4 (n ⫽ 8)
Clinical Outcomes
Mean Preop (range)
Mean Postop (range)
Mean Preop (range)
Mean Postop (range)
Overall pain score Active elevation (degrees) External rotation (degrees) ASES SST* Satisfaction
4.3 (3–5) 145 (140–170) 52 (20–90) — — 1 (1–3)
2.8 (1–5) 169 (130–180) 69 (0–90) 67 (42–90) 9 (8–12) 7 (5–9)
4.5 (4–5) 122 (30–165) 29 (0–50) — — 2 (1–4)
2.3 (1–5) 134 (90–180) 45 (30–80) 58 (33–83) 8 (5–11) 7 (4–8)
*Simple Shoulder Test, number of “yes” responses.
Eight shoulders were followed-up radiographically for more than 1 year (mean, 4 years; range, 1 to 9 years). The remaining 8 shoulders did not have radiographs available for comparison. Seven of the 8 shoulders showed radiographic progression of disease. One patient with stage 1 initial radiographs did not have progressive articular changes at 4 years. Of the patients with minimal radiographic changes, 4 of the 5 (Crossan and Vallance stage 1 or 2 disease) had evidence of advancing radiographic changes. Two patients with stage 1 disease progressed to stage 2 disease (at 3 and 9 years). One patient with stage 1 progressed to stage 3 at 5 years, and 1 patient with stage 2 at the index operation progressed to stage 4. Two of the 5 patients with minimal radiographic changes (stage 1 or 2) showed progressive acromiohumeral narrowing from greater than 6 mm to 3 to 6 mm. All 3 remaining patients with more advanced Crossan and Vallance stage 4 disease at the index operation had disease progression at the last follow-up to stage 5 radiographic disease (2 patients at 1 year and 1 at 7 years). One of the 3 patients progressed to an acromiohumeral distance of less than 3 mm. Of these 7 shoulders with radiographic progression, the mean active elevation improved from 125° (range, 30° to 170°) to 159° (range, 120° to 180°). There were 2 excellent, 3 satisfactory, and 2 unsatisfactory results. Complications and Revisions One patient (age 40 years at the index procedure) required revision shoulder surgery with total shoulder arthroplasty because of continued pain 11 months after arthroscopic synovectomy. This patient’s radiographs revealed rapid progression (1 year from synovectomy) with loss of glenohumeral space, proximal humeral migration, and periarticular marginal erosions. Two other patients (age 28 and 42 years) are
currently considering arthroplasty at 5 and 9 years after the index procedure because of increasing discomfort and dysfunction after several years of excellent pain relief. Both patients had Crossan and Vallance stage 4 changes at the index procedure, with 1 of these progressing to stage 5 over a 7-year period. There were no infections, neurologic injuries, or hematomas requiring reoperation in this series. DISCUSSION Although shoulder synovectomy in patients with rheumatoid arthritis has been examined in a limited number of patients and reported in the literature,4-8 synovectomy of the knee has been examined more extensively in larger, more homogenous patient groups. Good early results were often tempered by concerns about waning pain relief and loss of functional improvement.9,31-34 These same issues have been raised with shoulder synovectomy. Pahle5,6 reported on the largest series of patients (54 shoulders) undergoing open surgical synovectomy, noting good pain relief even in advanced cases of arthritis with increased abduction and external rotation. He did caution, however, that radiographs continued to show advancing disease. Peterson7 noted similar findings with pain relief and functional gains, with 2 patients requiring later shoulder arthroplasty. However, both of these studies included patients with rotator cuff tearing, limiting the interpretation of outcome in the assessment of isolated shoulder synovectomy. The results in our patients though are similar to the findings of Pahle and Peterson in their use of open shoulder synovectomy.5-7 Pain relief and patient satisfaction were high. Improvement in function could occur and often did but was not as predictable. Synovectomy does not seem to prevent progression of the rheumatoid arthritis, with advancing radio-
SHOULDER SYNOVECTOMY IN PATIENTS WITH RA graphic changes noted in 7 of our patients.35 Kelly4 recommended that synovectomy must be done before significant damage to articular cartilage has occurred. Although synovectomy would be ideally performed in rheumatoid patients with normal shoulder articular contour, this study shows that pain relief can be achieved despite moderately advanced radiographic changes. However, in the study by Wakitani et al.36 in patients with a wider spectrum of radiographic changes, those with severe arthritis had less favorable results with synovectomy compared with those undergoing prosthetic arthroplasty procedures. This study supports the work of Cruess17 in that acromiohumeral distance was not found to be a good predictor of rotator cuff tearing in this patient population. Only 1 of the 5 patients with diminished acromiohumeral distance on radiographs was found to have a partial-thickness tear under direct arthroscopic visualization. If rotator cuff tearing is suspected, we recommend further evaluation with magnetic resonance imaging so that appropriate preoperative planning may be performed. The technique of open shoulder synovectomy has been described as difficult.1,3,5-7,17,37 In this series of arthroscopic synovectomy, bleeding was encountered that was often difficult to control and made visualization difficult. Meticulous hemostasis is necessary intraoperatively to facilitate adequate viewing, and hemostasis should be achieved before termination of the procedure to avoid a significant hematoma. Caution must be taken with the use of cautery or ablation techniques in the inferior capsular recess regions because injury to the axillary nerve has been reported with other procedures.38 Several other authors have also cautioned against sacrificing the coracoacromial ligament, thus compromising the coracoacromial arch due to concerns about rheumatoid disease progression, increasing rotator cuff deficiency, and the development of severe upward and forward humeral subluxation.12,37 We do not divide the coracoacromial ligament or perform acromioplasty in this patient population unless impingement or wear is directly visualized on the rotator cuff tendons. In this study, only 3 shoulders had surgical manipulation of the coracoacromial arch. The limitations in this study are related to its retrospective nature, including the small number of patients, absence of a nonoperative comparison group, and the lack of a control group. Also, none of the patients in this study had the most severe radiographic changes associated with Crossan and Vallance stage 5 disease. Surgeon bias as to which patients were of-
55
fered synovectomy should be considered when evaluating these results. The results presented should not be misinterpreted to include patients with full-thickness rotator cuff tearing. Although many rheumatoid patients with shoulder pain have rotator cuff tearing, the focus of treatment is usually directed at the rotator cuff pathology using synovectomy as an adjunctive procedure. This is the first study to examine arthroscopic synovectomy as a primary treatment for shoulder synovitis in patients with an intact rotator cuff. Patients with rheumatoid arthritis with shoulder involvement frequently have substantial associated pathologic problems such as capsular stiffness,16,18 more advanced cartilage loss than is apparent radiographically,7,14,19-21 and rotator cuff tearing that can be severe. Although 13 of the 16 shoulders in this study had a satisfactory outcome, the prognostic factors for an unsatisfactory outcome with arthroscopic synovectomy remain to be determined. The results presented represent the best scenario with an intact rotator cuff. When considering surgical outcomes, the limitations of arthroscopic synovectomy should be discussed with patients. There is a greater expectation for pain relief and patient satisfaction, less predictability for gains of motion, and no certainty that radiographic progression of the disease will be halted. REFERENCES 1. Bennett WF, Gerber C. Operative treatment of the rheumatoid shoulder. Curr Opin Rheumatol 1994;6:177-182. 2. Clayton ML, Ferlic DC. Surgery of the shoulder in rheumatoid arthritis: A report of nineteen patients. Clin Orthop 1975;106: 166-174. 3. Cuomo F, Greller MJ, Zuckerman JD. The rheumatoid shoulder. Rheum Dis Clin North Am 1998;24:67-82. 4. Kelly IG. Surgery of the rheumatoid shoulder. Ann Rheum Dis 1990;49:824-829 (suppl 2). 5. Pahle JA. The shoulder joint in rheumatoid arthritis: Synovectomy. Reconstr Surg Traumatol 1981;18:33-47. 6. Pahle JA, Kvarnes L. Shoulder synovectomy. Ann Chir Gynaecol Suppl 1985;198:37-39. 7. Petersson CJ. Shoulder surgery in rheumatoid arthritis. Acta Orthop Scand 1986;57:222-226. 8. Smith-Peterson MN, Aufranc OE, Larson CB. Useful surgical procedures for rheumatoid arthritis involving the upper extremity. Arch Surg 1943;46:764-770. 9. Ochi T, Iwase R, Kimura T, et al. Effect of early synovectomy on the course of rheumatoid arthritis. J Rheumatol 1991;18: 1794-1798. 10. Cofield RH. Total shoulder arthroplasty with Neer prosthesis. J Bone Joint Surg Am 1984;66:899-906. 11. Ennevarra K. Painful shoulder joint in rheumatoid arthritis: A clinical and radiologic study of 200 cases with special reference to arthrography of the glenohumeral joint. Acta Rheum Scand 1967;1-108 (suppl II). 12. Kelly IG, Foster RS, Fisher WD. Neer total shoulder replacement in rheumatoid arthritis. J Bone Joint Surg Br 1987;69: 723-726.
56
A. M. SMITH ET AL.
13. Rozing PM, Brand R. Rotator cuff repair during shoulder arthroplasty in rheumatoid arthritis. J Arthroplasty 1998;13: 311-319. 14. Weiss JJ, Thompson GR, Doust V, Burgener F. Rotator cuff tears in rheumatoid arthritis. Arch Intern Med 1975;135:521525. 15. Curran JF, Ellman MH, Brown NL. Rheumatologic aspects of painful conditions affecting the shoulder. Clin Orthop 1983; 173:27-37. 16. Laine VAI. Shoulder affections in rheumatoid arthritis. Ann Rheum Dis 1954;13:157-160. 17. Cruess RL. Rheumatoid arthritis of the shoulder. Orthop Clin North Am 1980;11:333-342. 18. Dabrowski W, Fonseka N, Ansell BM, Liyanage IS, Arden GP. Shoulder problems in juvenile chronic polyarthritis. Scand J Rheumatol 1979;8:49-53. 19. Hirooka A, Wakitani S, Yoneda M, Ochi T. Shoulder destruction in rheumatoid arthritis: Classification and prognostic signs in 83 patients followed 5-23 years. Acta Orthop Scand 1996; 67:258-263. 20. Lehtinen JT, Belt EA, Lyback CO, et al. Subacromial space in the rheumatoid shoulder: A radiographic 15-year follow-up study of 148 shoulders. J Shoulder Elbow Surg 2000;9:183187. 21. Petersson CJ. The acromioclavicular joint in rheumatoid arthritis. Clin Orthop 1987;223:86-93. 22. Witwity T, Uhlmann R, Nagy MH, Bhasin VB, Bahgat MM, Singh AK. Shoulder rheumatoid arthritis associated with chondromatosis, treated by arthroscopy. Arthroscopy 1991;7:233236. 23. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315324. 24. Michener LA, McClure PW, Sennett BJ. American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form, patient self-report section: reliability, validity, and responsiveness. J Shoulder Elbow Surg 2002;11:587-594. 25. Roddey TS, Olson SL, Cook KF, Gartsman GM, Hanten W. Comparison of the University of California-Los Angeles
26. 27. 28. 29. 30. 31. 32.
33. 34. 35. 36. 37. 38.
Shoulder Scale and the Simple Shoulder Test with the shoulder pain and disability index: Single administration reliability and validity. Phys Ther 2000;80:759-768. Crossan JF, Vallance R. Clinical and radiological features of the shoulder joint in rheumatoid arthritis. J Bone Joint Surg Br 1980;62:116. Crossan JF, Vallance R. The shoulder joint in rheumatoid arthritis. In: Bayley I, Keddel L, eds. Shoulder surgery. New York: Springer-Verlag, 1982;131-139. Souter WA. The surgical treatment of the rheumatoid shoulder. Ann Acad Med Singapore 1983;12:243-255. Jobe CM. Shoulder arthroscopy. In: Ianotti JP, Williams GP, eds. Disorders of the shoulder: Diagnosis and management. Philadelphia: Lippincott Wilkins & Williams, 1999;883-910. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: A preliminary report. J Bone Joint Surg Am 1972;54:41-50. Bryan RS, Peterson LF. Synovectomy of the knee. Orthop Clin North Am 1971;2:705-713. Geens S, Clayton ML, Leidholt JD, Smyth CJ, Bartholomew BA. Synovectomy and debridement of the knee in rheumatoid arthritis. II. Clinical and roentgenographic study of thirty-one cases. J Bone Joint Surg Am 1969;51:626-642. Paradies LH. Synovectomy for rheumatoid arthritis of the knee. J Bone Joint Surg Am 1975;57:95-100. Ranawat CS, Desai K. Role of early synovectomy of the knee joint in rheumatoid arthritis. Arthritis Rheum 1975;18:117121. Kaarela K, Kautiainen J. Continuous progression of radiologic destruction on seropositive rheumatoid arthritis. J Rheumatol 1997;24:1285-1287. Wakitani S, Imoto K, Saito M, et al. Evaluation of surgeries for rheumatoid shoulder based on the destruction pattern. J Rheumatol 1999;26:41-46. Chen AL, Joseph TN, Zuckerman JD. Rheumatoid arthritis of the shoulder. J Am Acad Orthop Surg 2003;11:12-24. Wong KL, Williams GR. Complications of thermal capsulorrhaphy of the shoulder. J Bone Joint Surg Am 2001;83:S151S155 (suppl).