- Email: [email protected]
Arthroscopic Debridement and Capsular Release for the Treatment of Shoulder Osteoarthritis
Arthroscopic Debridement and Capsular Release for the Treatment of Shoulder Osteoarthritis Nathan W. Skelley, M.D., Surena Namdari, M.D., Aaron M. Chamberlain, M.D., Jay D. Keener, M.D., Leesa M. Galatz, M.D., and Ken Yamaguchi, M.D.
Purpose: The purpose of this study was to evaluate patients who underwent isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis to determine clinical and functional outcomes and time until conversion to shoulder arthroplasty. Methods: We performed a retrospective review of 33 patients who underwent arthroscopic debridement and capsular release for shoulder osteoarthritis at our institution between 2006 and 2011. All procedures were performed by a single surgeon (K.Y.). Patients were evaluated for intraoperative arthritis grade, preoperative and postoperative range or motion, American Shoulder and Elbow Surgeons (ASES) score, pain score self-assessments, radiographic evaluation, and conversion to total shoulder arthroplasty. Clinical follow-up was on average 40.3 weeks postoperatively and telephone interview follow-up was performed at a minimum of 2 years postoperatively in all patients. Results: There was an initial improvement in range of motion and pain scores; however, patients in our study returned to preoperative levels approximately 3.8 months after debridement and capsular release. Twenty patients (60.6%) reported that they were not satisfied with the outcome of the procedure. Total shoulder arthroplasty was undertaken in 14 (42.4%) patients an average of 8.8 months after arthroscopy. Among the 19 (57.6%) patients who did not go on to have total shoulder arthroplasty, ASES scores (42.2 to 50.8; P ¼ .41) and visual analog scale pain scores (7.8 to 7.4; P ¼ .59) were similar preoperatively and at final telephone follow-up. Conclusions: Isolated arthroscopic debridement and capsular release without any other procedures were associated with only temporary pain relief and improvement in motion. Although there are limited nonarthroplasty surgical options available for glenohumeral arthritis, isolated arthroscopic debridement and capsular release may not provide substantial benefit to justify its use in most patients. Level of Evidence: Level IV, therapeutic case series.
A
lthough total shoulder arthroplasty provides the most reliable pain relief and restoration of function for most patients with arthritis involving the glenohumeral joint,1 many physicians would prefer to defer the procedure as long as possible in young patients and those with high activity levels. The concern with performing a total shoulder arthroplasty in these individuals is implant longevity, particularly of the glenoid component.2 Several nonarthroplasty procedures exist to treat
From the Department of Orthopaedic Surgery, Washington University in St. Louis, St. Louis, Missouri, U.S.A. The authors report the following potential conflict of interest or source of funding: K.Y. receives support from Zimmer and Tornier. Received January 28, 2014; accepted August 26, 2014. Address correspondence to Nathan W. Skelley, M.D., 660 South Euclid Ave, Mailbox 8233, St. Louis, MO 63110, U.S.A. E-mail: skelleyn@wudosis. wustl.edu Ó 2015 by the Arthroscopy Association of North America 0749-8063/1469/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.08.025
494
these patients, among them arthroscopic capsular release and joint debridement; however, the utility of these procedures is unclear. Among the different minimally invasive procedures used, arthroscopic capsular release and joint debridement have been described as achieving pain relief and restoring motion in some patients with glenohumeral arthritis.3-7 The goal of these procedures has been to provide temporary improvement in symptoms until a time when shoulder arthroplasty is considered a more appropriate surgical option.1,3 Previous studies have concluded that performing arthroscopic debridement and capsular release in conjunction with other procedures may benefit patients with glenohumeral arthritis.4-9 However, these studies could not make conclusions about the use of isolated arthroscopic debridement and capsular release without any other procedures. Concomitant procedures in other studies included subacromial decompression, rotator cuff repair, labral repair, distal clavicle excision, microfracture, inferior osteophyte excision, and acromioplasty.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 3 (March), 2015: pp 494-500
495
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA Table 1. Excluded Patients Reason for Exclusion Concomitant diagnoses Adhesive capsulitis Previous rotator cuff repair Impingement Previous fracture Total shoulder arthroplasty before release Removal of painful hardware with release Previous capsulorrhaphy Previous SLAP repair Debridement for osteoarthritis Concomitant procedures Subacromial decompression/bursectomy Rotator cuff repair Microfracture arthroplasty Tendinitis debridement and decompression Labral tear repair and debridement Teres minor neurolysis and decompression Acromioclavicular joint arthritis with clavicle resection
No. of Patients Excluded 60 8 7 4 3 3 1 1 1 14 13 9 8 6 4 2
NOTE. This table describes how 122 of the 155 patients available for review were excluded from our analysis for either concomitant diagnoses, concomitant procedures, or both performed at the same time as the debridement and capsular release.
The purpose of this study was to evaluate patients who underwent isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis to determine clinical and functional outcomes and time until conversion to shoulder arthroplasty. We hypothesized that isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis would provide significant long-term (>2 years) improvements in pain and function for patients with shoulder osteoarthritis.
Methods Study Sample We retrospectively reviewed a consecutive series of 155 patients who underwent arthroscopic debridement and capsular release for glenohumeral arthritis. Our hospital’s institutional review board approved the study (IRB No. 201201126). All surgical procedures were performed by the senior author (K.Y.) between 2006 and 2011. Patients of any age were considered candidates for inclusion in the study if they underwent arthroscopic glenohumeral capsular release for a primary diagnosis of glenohumeral osteoarthritis. Exclusion criteria included concomitant procedures (rotator cuff repair, distal clavicle excision, microfracture, subacromial decompression, labral repair, acromioplasty, and inferior osteophyte excision) and confounding diagnoses such as adhesive capsulitis (Table 1). A history of slowly progressive shoulder pain that was worse with
activity and without acute changes in range of motion was used to distinguish osteoarthritis with radiographic evaluation. Similarly, the operative reports were valuable because the inflammatory and capsular effects were noted intraoperatively for the patients with adhesive capsulitis and not for the patients with arthritis. All patients considered for surgery had failed nonoperative treatment with activity modification, corticosteroid injections, pain medication, and 6 weeks of a home-based patient-directed physical therapy exercise program. We identified 33 patients who satisfied the inclusion and exclusion criteria. Preoperative Variables Preoperative demographic informationdincluding age, sex, hand dominance, body mass index, smoking, and medical comorbiditiesdwas collected (Table 2). All patients underwent a standardized preoperative physical examination by the senior author (K.Y.) and the following parameters were consistently recorded: pain (visual analog scale), American Shoulder and Elbow Surgeons (ASES) score,10 active forward elevation (AFE), and active external rotation (AER). All patients underwent standardized radiography, including true anteroposterior, scapular Y, and axillary views, at the same medical center and were reviewed for quality by the senior author (K.Y.). Available radiographs were independently reviewed by 2 authors (N.W.S., S.N.) who were blinded to the patient’s outcome. Three studies and one axillary view were interpreted by the operating surgeon (K.Y.) at the time of treatment and were not available for study review. The following parameters were recorded: Samilson-Prieto arthritis grade11,12 (on true anteroposterior view; classified by the size of an inferior osteophyte; kappa ¼ 0.86), posterior glenohumeral subluxation (yes/no; axis of glenoid different from that of humeral head; kappa ¼ 0.97), and glenoid morphologic characteristics (concentric/nonconcentric; kappa ¼ 1.0). Any discrepancies on available studies were resolved by discussion and consensus review. Arthritis grade was evaluated intraoperatively during diagnostic arthroscopy (0 ¼ no cartilage defects, I ¼ cartilage softening, II ¼ partial-
Table 2. Demographic Data of Patient Population Age Body mass index Sex Hand dominance Smoking (yes/no) Alcohol use (yes/no) Hypertension (yes/no) Gastroesophageal reflux disease (yes/no) Diabetes mellitus (yes/no)
55.2 9.2 SD 28.74 5.0 SD 24 male/9 female 30 right/3 left 11 (33)/22 (67) 18 (55)/15 (45) 18 (55)/15 (45) 10 (30)/23 (70) 7 (21)/26 (79)
NOTE. Data presented as n (%) unless otherwise indicated. SD, standard deviation.
496
N. W. SKELLEY ET AL.
thickness fissures, III ¼ full-thickness cartilage defect, and IV ¼ subchondral bone exposed). Surgical Technique/Rehabilitation Protocol The surgical technique for debridement and capsular release has been previously described and was very similar to that performed for adhesive capsulitis of the shoulder.13 The technique used evaluation of cartilage and standard debridement of loose cartilage, frayed labrum, and other degenerative tissues but also emphasized complete release of the rotator interval, middle, and inferior glenohumeral ligaments. Arthroscopy was performed with the patient in the beach chair position under general anesthesia. Standard posterolateral and anterior rotator interval portals were used, and an arthroscopic shaver was applied to perform glenohumeral joint debridement. Loose bodies were removed if present. Subsequently, the rotator interval was released with shaver and electrocautery until there was clear visualization of the coracoacromial ligament, and then an arthroscopic capsular biter was used to release the anterior and inferior capsule. A needle was inserted into the joint through the rotator interval, confirmed through the arthroscope, and used to inject 10 mL of 0.5% bupivacaine hydrochloride with 40 mg of triamcinolone after the wounds were closed. Postoperatively, all arms were placed in slings for comfort. A patient-directed home exercise program was initiated immediately for stretching and range of motion therapy. Follow-up Data Patients were asked to return to the clinic 3 times after surgerydat approximately 2 weeks, 6 weeks, and 3 months. Patients were also seen for continued problems after these time points. Patient charts were retrospectively reviewed to determine visual analog scale pain scores and range of motion parameters (AFE, AER) which were recorded at each visit. Patients were contacted by telephone at a minimum of 2 years postoperatively for final re-evaluation. Subsequent corticosteroid injections and reoperations, including total shoulder arthroplasty, were recorded. Outcome instruments again included the visual analog scale pain scores and ASES scores. Patients were also asked, “If you could go back in time and make the decision again, would you have undergone an arthroscopic capsular release for your shoulder condition/problem? (definitely not, probably not, neutral, probably yes, definitely yes).” Finally, patients were asked the question, “Are you satisfied with the outcome of your arthroscopic surgery? (yes/no).” Statistical Analysis All results were analyzed by statistical testing comparing preoperative measures with corresponding
postoperative measures. Paired t tests were performed for continuous data, and c-square tests were performed for categorical data. All results were considered statistically significant at P < .05. Kaplan-Meier survival analysis was used to determine the postoperative interval until an arthroplasty was performed. Multiple regression analyses were performed to evaluate preoperative demographic and examination findings that may be predictive of outcome scores or conversion to total shoulder arthroplasty.
Results Demographics The study group consisted of 33 patients, 24 (73%) men and 9 (27%) women (Table 2). The actual average follow-up for the postoperative visits was 1.8 weeks, 10.6 weeks, and 16.5 weeks. As patients continued to have pain and loss of function, the final average followup in clinic, eliminating those who went on to arthroplasty, was 40.3 weeks (range, 20 to 95 weeks). Average final telephone follow-up was 43.4 months (range, 25 to 71 months). The average age was 55.2 years (range, 34 to 71 years) at the time of operation. All patients underwent glenohumeral joint debridement and capsular release as the surgical procedure. At the time of operation, it was determined that 4 (12.1%) patients had grade II cartilage defects (3 glenoid lesions and one humeral lesion): 17 (51.5%) were grade III (7 glenoid and 6 humeral lesions and 4 lesions occurring on both sides), and 12 (36.4%) were grade IV (4 glenoid and 2 humeral lesions and 6 lesions occurring on both sides). Pain Relief and Motion Trends At the first clinic visit, patients had significant improvement in visual analog scale pain scores from 7.8 preoperatively to 1.0 postoperatively (P < .001; n ¼ 33) (Fig 1); however, patients trended back toward preoperative pain levels by the third postoperative visit and achieved a mean visual analog scale score of 7.4 at final follow-up (P ¼ .59) (Fig 1). Patients showed significant improvement in range of motion at the first postoperative visit (mean, 12.5 days). AFE improved on average from 121.8 preoperatively to 140.2 postoperatively (P < .001) and AER improved from 21.9 to 47.6 (P < .001); however, AFE and AER trended toward preoperative levels, with a mean AFE of 122.2 (Fig 2) and a mean AER of 28.4 (Fig 3) at the third (approximately 3 months) postoperative visit (P ¼ .27 and P ¼ .14, respectively; n ¼ 31). Final Outcomes and Satisfaction Of the 19 patients who did not undergo a total shoulder arthroplasty, 18 (94.7%) reported continued pain in the affected shoulder. The final in-clinic follow-up
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA
Fig 1. Patient-reported pain scores throughout the preoperative and postoperative periods. Patients who underwent total shoulder arthroplasty were removed from this analysis after undergoing the arthroplasty procedure. (Avg., average.)
for these 19 patients was 40.3 weeks (range, 20 to 95 weeks). The matched pain (P ¼ .67), forward elevation (P ¼ .56), and external rotation (P ¼ .61) scores were statistically unchanged at 40.3 weeks from the postoperative visit at 16.5 weeks. All 33 patients were available for final follow-up telephone evaluation interviews at a minimum of 2 years after the debridement and capsular release. The mean ASES score was 42.2 (range, 21 to 80) preoperatively compared with 50.8 (range, 16 to 90) (P ¼ .41) postoperatively through telephone interviews in the 19 patients who did not undergo conversion to total shoulder arthroplasty.
497
Fig 3. External rotation range of motion throughout the preoperative and postoperative periods. (Avg., average.)
After arthroscopic release, 12 (36.4%) patients reported undergoing steroid injection for pain control in the postoperative setting. When asked if they would undergo the procedure again knowing what they know now, 10 (30.3%) said “definitely not” or “probably not,” 11 (33.3%) said “definitely yes” or “probably yes,” and 12 (36.4%) were “neutral.” When asked if they were satisfied with the arthroscopic procedure, 13 (39.4%) said they were satisfied and 20 (60.6%) were not satisfied. No major complications were associated with the procedure. One patient had a superficial portal site infection that resolved with oral antibiotic therapy. Radiographic Analysis Preoperatively, the Samilson-Prieto score was grade 1 in 11 of 33 (33.3%) patients, grade 2 in 9 of 33 (27.3%) patients, and grade 3 in 13 of 33 (39.4%) patients. Posterior glenohumeral subluxation was present in 14 of 33 (42.4%) patients. Nine of 33 (27.2%) patients had nonconcentric glenoid morphologic characteristics.
Fig 2. Range of motion in forward elevation throughout the preoperative and postoperative periods. (Avg., average.)
Survival Analysis Fourteen of the 33 (42.4%) shoulders underwent total shoulder arthroplasty at an average of 38.4 weeks (range, 8 to 92 weeks) after arthroscopic debridement and capsular release (Fig 4). The average age of the patients who underwent total shoulder arthroplasty was 57.1 years (range, 43.7 to 70.2 years) at the time of the arthroscopic procedure compared with 53.7 years (range, 34.0 to 71.0 years) for those who did not undergo total shoulder arthroplasty (P ¼ .31). Only 7 (50%) patients who eventually underwent total shoulder arthroplasty were classified as Samilson-Prieto grade 3 at the time of the arthroscopic release. Six
498
N. W. SKELLEY ET AL.
(75.0%) patients and grade IV lesion occurred on both sides in 6 (85.7%) patients who underwent arthroplasty (P ¼ .002). The ASES and visual analog scale pain scores for patients who underwent total shoulder arthroplasty after the arthroscopy procedure were 70 (range, 43.8 to 96.4) and 4 (range, 0 to 7), respectively, at the most recent telephone follow-up. No added difficulties were noted in the arthroplasty operative reports related to the previous arthroscopic surgery.
Fig 4. Survival analysis of the 14 patients who eventually went on to total shoulder arthroplasty after arthroscopic debridement and capsular release. The average time to total shoulder arthroplasty for these 14 patients was 38.4 weeks or 269 days.
(31.6%) patients who did not undergo total shoulder arthroplasty were also classified as Samilson-Prieto grade 3 at the time of arthroscopic release (P ¼ .71). Of the remaining patients who eventually went on to total shoulder arthroplasty, 4 (28.6%) patients were classified as grade 2 and 3 (21.4%) were classified as grade 1. Of the patients who did not undergo arthroplasty, 5 (26.3%) were classified as grade 2 and 8 (42.1%) were classified as grade 1. Patients undergoing arthroplasty had intraoperative arthritis grades of III (6 of 17 [35.3%]) and IV (8 of 12 [66.6%]). Patients who underwent arthroplasty were more likely to have lesions on both sides at the time of arthroscopy compared with patients who did not undergo arthroplasty (Table 3). Grade III lesions occurred on both sides in 3
Regression Analysis With the numbers observed in this study, multiple linear regression analysis of the data found no significant correlations between preoperative radiographic parameters relating to visual analog scale pain scores (P ¼ .58), ASES score (P ¼ .48), or satisfaction score (P ¼ .14) at final follow-up. Similarly, no significant correlation was found between intraoperative arthritis grade relating to visual analog scale pain scores (P ¼ .43), ASES score (P ¼ .50), or satisfaction score (P ¼ .23) at final follow-up. On multiple logistic regression, age, sex, dominant side of surgery, body mass index, tobacco use, alcohol use, presence of subluxation, glenoid morphologic characteristics, cartilage grade, and Samilson-Prieto grade preoperatively were not significant predictors for conversion to total shoulder arthroplasty (P ¼ .53). Additionally, the univariate analysis of radiographic subluxation and arthritis severity preoperatively did not predict total shoulder arthroplasty (P ¼ .13 and P ¼ .35, respectively). The presence of nonconcentric glenoid morphologic features preoperatively was moderately predictive (r ¼ 0.41) of total shoulder arthroplasty (P ¼ .04).
Discussion The results in this study did not support the use of isolated arthroscopic debridement and capsular release for the treatment of glenohumeral osteoarthritis because these procedures did not provide long-term relief of pain and improved function. A recent systematic review14 of outcomes of arthroscopic debridement for glenohumeral arthritis found that Level IVetype evidence suggested improved pain level and patient satisfaction postoperatively. This review highlighted the
Table 3. Intraoperative Arthritis Grade at the Time of Arthroscopy Intraoperative Arthritis Grade Nonarthroplasty Group (n ¼ 19) Grade II Grade III Grade IV Total
Glenoid Side 3 (15.8) 5 (26.3) 2 (10.5) 10 (30.3)
NOTE. Data presented as n (%).
Humeral Side 1 (5.3) 5 (26.3) 1 (5.3) 7 (21.2)
Arthroplasty Group (n ¼ 14) Both e 1 (5.3) 1 (5.3) 2 (6.1)
Glenoid Side e 2 (14.3) 1 (7.1) 3 (9.1)
Humeral Side e 1 (7.1) 1 (7.1) 2 (6.1)
Both e 3 (21.4) 6 (42.9) 9 (27.2)
Total (N ¼ 33) 4 (12.1) 17 (51.5) 12 (36.4) 33 (100)
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA
deficits in the literature and suggested that future studies comprehensively define patient characteristics and trend results over time.14 The current review noted the difficulty in interpreting debridement and capsular release based on confounding variables. We had a known cohort of young patients with isolated arthritis who were not affected by concomitant pathologic processes or procedures. These patients underwent isolated debridement and capsular release as a less invasive alternative to arthroplasty. During the study period, we found a rapid return of pain and stiffness after surgery and low satisfaction and survival rates, calling into question the benefit of isolated capsular release and debridement for the treatment of shoulder osteoarthritis. These findings led us to reject our study hypothesis that these procedures in isolation would provide significant long-term (>2 years) improvement in pain and function for patients with shoulder osteoarthritis. Functional outcomes and satisfaction scores in our population were lower than those in other reports in the literature that used concomitant procedures such as biceps tenotomy, distal clavicle excision, and acromioplasty, among others.3-8,14 These studies found better results in patients by combining procedures. Most of our patients (60.6%) were not satisfied with the isolated procedure, and only 33.3% said that they would have the capsular release procedure again. A large proportion, 14 (42.4%) patients, were dissatisfied with the procedure as evidenced by the high rate of conversion to total shoulder arthroplasty in a relatively short time (8.8 months) after the index arthroscopic debridement and capsular release. This high conversion rate was believed to largely result from the lack of longer pain relief provided by the arthroscopic procedure. Our study was consistent with other smaller studies that did not find an association between demographic factors, preoperative radiographic grade of arthritis, or arthroscopic grade and functional outcome.6,7,15,16 We did, however, find that nonconcentric glenoid morphologic characteristics preoperatively were moderately predictive (r ¼ 0.41) of the patient undergoing a total shoulder arthroplasty in the postoperative period, and the presence of lesions on both sides was significantly associated with undergoing total shoulder arthroplasty. Previous studies have shown benefit from concomitant procedures.3-8,14 Our results from isolated debridement and capsular release, however, showed that this procedure may be unnecessary and may provide little to no benefit when combined with other procedures. Van Thiel et al.7 reported that 22% of their patients underwent arthroplasty at a mean of 10.1 months after capsular release, debridement, and other concomitant procedures. As expected, the patients who underwent conversion to total shoulder arthroplasty had better
499
ASES scores, greater pain relief, and improved range of motion compared with the patients who underwent only an arthroscopic debridement and capsular release. Van Thiel et al. also reported a higher Samilson-Prieto arthritis score as a risk factor for conversion to arthroplasty during the postoperative period. We were unable to find an association between demographic variables or radiographic arthritis grade and the need for conversion to total shoulder arthroplasty. Our results were unique within the literature and could be related to several factors. We focused only on isolated arthroscopic debridement and capsular release as a treatment for glenohumeral arthritis and thus eliminated the effects of additional procedures on clinical and functional outcomes. Our patients were seen at a large referral center, which may have influenced treatment with arthroplasty. The patients in our series were young but had an older average age compared with other similar studies in the literature.4,6,7,9 They may have had more severe arthritis at the time of surgery: many other studies did not characterize osteoarthritis grade. Other investigations used different techniques to perform the capsular release because there is no standardized procedure.4-7 Whether the technique for capsular release had a significant effect on the maintenance of motion or pain relief, or both, after arthroscopic debridement and capsular release is unclear and could be the basis for future study. Similarly, postoperative rehabilitation protocols were not well established or defined in the literature and may influence outcome comparisons. Interestingly, the anesthetic and corticosteroid injection at the end of the procedure may have been a significant reason for the temporary improvement in pain and motion observed at the first postoperative visit. This would explain the rapid return to preoperative pain and range of motion in our study. Although complications are rare from arthroscopic procedures, this lack of persistent improvement after arthroscopic debridement and capsular release may place the patient at an unnecessary surgical risk.13,17,18 Limitations Our study is limited by the inherent design factors associated with a retrospective analysis without a control group. Patients were contacted at different times after the procedure and this could affect patient satisfaction. The study is strengthened by complete followup through telephone interviews but is limited by the minimum length of telephone follow-up at 2 years. Only one surgeon’s standardized technique for arthroscopic capsular release was analyzed, and this could influence generalizability. Preoperative and postoperative range of motion was conducted in an unblinded fashion by the operating surgeon and was at risk for bias. We also recognize that there could have
500
N. W. SKELLEY ET AL.
been operative bias to pursue total shoulder arthroplasty in the postoperative period by the same surgeon; however, an argument could be made that the bias may have been more likely to be toward success of the arthroscopic procedure and thus avoidance of additional surgery.
Conclusions Isolated arthroscopic debridement and capsular release without any other procedures were associated with only temporary pain relief and improvement in motion. Although there are limited nonarthroplasty surgical options available for glenohumeral arthritis, isolated arthroscopic debridement and capsular release may not provide substantial benefit to justify use in most patients. Patients should be carefully counseled on the expectations associated with isolated arthroscopic debridement and capsular release.
References 1. Carter MJ, Mikuls TR, Nayak S, Fehringer EV, Michaud K. Impact of total shoulder arthroplasty on generic and shoulder-specific health-related quality-of-life measures: A systematic literature review and meta-analysis. J Bone Joint Surg Am 2012;94:e1271-e1279. 2. Matsen FA III. Early effectiveness of shoulder arthroplasty for patients who have primary glenohumeral degenerative joint disease. J Bone Joint Surg Am 1996;78:260-264. 3. Ogilvie-Harris DJ, Wiley AM. Arthroscopic surgery of the shoulder. A general appraisal. J Bone Joint Surg Br 1986;68:201-207. 4. Millett PJ, Gaskill TR. Arthroscopic management of glenohumeral arthrosis: humeral osteoplasty, capsular release, and arthroscopic axillary nerve release as a jointpreserving approach. Arthroscopy 2011;27:1296-1303. 5. Richards DP, Burkhart SS. Arthroscopic debridement and capsular release for glenohumeral osteoarthritis. Arthroscopy 2007;23:1019-1022. 6. Cameron BD, Galatz LM, Ramsey ML, Williams GR, Iannotti JP. Non-prosthetic management of grade IV
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17. 18.
osteochondral lesions of the glenohumeral joint. J Shoulder Elbow Surg 2002;11:25-32. Van Thiel GS, Sheehan S, Frank RM, et al. Retrospective analysis of arthroscopic management of glenohumeral degenerative disease. Arthroscopy 2010;26:1451-1455. Ogilvie-Harris DJ, Gordon R, MacKay M. Arthroscopic treatment for posterior impingement in degenerative arthritis of the elbow. Arthroscopy 1995;11:437-443. Weinstein DM, Bucchieri JS, Pollock RG, Flatow EL, Bigliani LU. Arthroscopic debridement of the shoulder for osteoarthritis. Arthroscopy 2000;16:471-476. Richards RR, An KN, Bigliani LU, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3:347-352. Samilson RL, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 1983;65:456-460. Ilg A, Bankes M, Emery R. The intra- and inter-observer reliability of the Samilson and Prieto grading system of glenohumeral arthropathy. Knee Surg Sports Traumatol Arthrosc 2001;9:187-190. Yamaguchi K, Sethi N, Bauer GS. Postoperative pain control following arthroscopic release of adhesive capsulitis. Arthroscopy 2002;18:359-365. Namdari S, Skelley N, Keener JD, Galatz LM, Yamaguchi K. What Is the role of arthroscopic debridement for glenohumeral arthritis? A critical examination of the literature. Arthroscopy 2013;29:1392-1398. Guyette TM, Bae H, Warren RF, Craig E, Wickiewicz TL. Results of arthroscopic subacromial decompression in patients with subacromial impingement and glenohumeral degenerative joint disease. J Shoulder Elbow Surg 2002;11:299-304. Kerr BJ, McCarty EC. Outcome of arthroscopic débridement is worse for patients with glenohumeral arthritis of both sides of the joint. Clin Orthop Relat Res 2008;466: 634-638. Marecek GS, Saltzman MD. Complications in shoulder arthroscopy. Orthopedics 2010;33:492-497. De Wachter J, Van Glabbeek F, van Riet R, Van Leemput W, Vermeyen K, Somville J. Surrounding soft tissue pressure during shoulder arthroscopy. Acta Orthop Belg 2005;71:521-527.
Purpose: The purpose of this study was to evaluate patients who underwent isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis to determine clinical and functional outcomes and time until conversion to shoulder arthroplasty. Methods: We performed a retrospective review of 33 patients who underwent arthroscopic debridement and capsular release for shoulder osteoarthritis at our institution between 2006 and 2011. All procedures were performed by a single surgeon (K.Y.). Patients were evaluated for intraoperative arthritis grade, preoperative and postoperative range or motion, American Shoulder and Elbow Surgeons (ASES) score, pain score self-assessments, radiographic evaluation, and conversion to total shoulder arthroplasty. Clinical follow-up was on average 40.3 weeks postoperatively and telephone interview follow-up was performed at a minimum of 2 years postoperatively in all patients. Results: There was an initial improvement in range of motion and pain scores; however, patients in our study returned to preoperative levels approximately 3.8 months after debridement and capsular release. Twenty patients (60.6%) reported that they were not satisfied with the outcome of the procedure. Total shoulder arthroplasty was undertaken in 14 (42.4%) patients an average of 8.8 months after arthroscopy. Among the 19 (57.6%) patients who did not go on to have total shoulder arthroplasty, ASES scores (42.2 to 50.8; P ¼ .41) and visual analog scale pain scores (7.8 to 7.4; P ¼ .59) were similar preoperatively and at final telephone follow-up. Conclusions: Isolated arthroscopic debridement and capsular release without any other procedures were associated with only temporary pain relief and improvement in motion. Although there are limited nonarthroplasty surgical options available for glenohumeral arthritis, isolated arthroscopic debridement and capsular release may not provide substantial benefit to justify its use in most patients. Level of Evidence: Level IV, therapeutic case series.
A
lthough total shoulder arthroplasty provides the most reliable pain relief and restoration of function for most patients with arthritis involving the glenohumeral joint,1 many physicians would prefer to defer the procedure as long as possible in young patients and those with high activity levels. The concern with performing a total shoulder arthroplasty in these individuals is implant longevity, particularly of the glenoid component.2 Several nonarthroplasty procedures exist to treat
From the Department of Orthopaedic Surgery, Washington University in St. Louis, St. Louis, Missouri, U.S.A. The authors report the following potential conflict of interest or source of funding: K.Y. receives support from Zimmer and Tornier. Received January 28, 2014; accepted August 26, 2014. Address correspondence to Nathan W. Skelley, M.D., 660 South Euclid Ave, Mailbox 8233, St. Louis, MO 63110, U.S.A. E-mail: skelleyn@wudosis. wustl.edu Ó 2015 by the Arthroscopy Association of North America 0749-8063/1469/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.08.025
494
these patients, among them arthroscopic capsular release and joint debridement; however, the utility of these procedures is unclear. Among the different minimally invasive procedures used, arthroscopic capsular release and joint debridement have been described as achieving pain relief and restoring motion in some patients with glenohumeral arthritis.3-7 The goal of these procedures has been to provide temporary improvement in symptoms until a time when shoulder arthroplasty is considered a more appropriate surgical option.1,3 Previous studies have concluded that performing arthroscopic debridement and capsular release in conjunction with other procedures may benefit patients with glenohumeral arthritis.4-9 However, these studies could not make conclusions about the use of isolated arthroscopic debridement and capsular release without any other procedures. Concomitant procedures in other studies included subacromial decompression, rotator cuff repair, labral repair, distal clavicle excision, microfracture, inferior osteophyte excision, and acromioplasty.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 3 (March), 2015: pp 494-500
495
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA Table 1. Excluded Patients Reason for Exclusion Concomitant diagnoses Adhesive capsulitis Previous rotator cuff repair Impingement Previous fracture Total shoulder arthroplasty before release Removal of painful hardware with release Previous capsulorrhaphy Previous SLAP repair Debridement for osteoarthritis Concomitant procedures Subacromial decompression/bursectomy Rotator cuff repair Microfracture arthroplasty Tendinitis debridement and decompression Labral tear repair and debridement Teres minor neurolysis and decompression Acromioclavicular joint arthritis with clavicle resection
No. of Patients Excluded 60 8 7 4 3 3 1 1 1 14 13 9 8 6 4 2
NOTE. This table describes how 122 of the 155 patients available for review were excluded from our analysis for either concomitant diagnoses, concomitant procedures, or both performed at the same time as the debridement and capsular release.
The purpose of this study was to evaluate patients who underwent isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis to determine clinical and functional outcomes and time until conversion to shoulder arthroplasty. We hypothesized that isolated arthroscopic debridement and capsular release without any other procedures for primary glenohumeral osteoarthritis would provide significant long-term (>2 years) improvements in pain and function for patients with shoulder osteoarthritis.
Methods Study Sample We retrospectively reviewed a consecutive series of 155 patients who underwent arthroscopic debridement and capsular release for glenohumeral arthritis. Our hospital’s institutional review board approved the study (IRB No. 201201126). All surgical procedures were performed by the senior author (K.Y.) between 2006 and 2011. Patients of any age were considered candidates for inclusion in the study if they underwent arthroscopic glenohumeral capsular release for a primary diagnosis of glenohumeral osteoarthritis. Exclusion criteria included concomitant procedures (rotator cuff repair, distal clavicle excision, microfracture, subacromial decompression, labral repair, acromioplasty, and inferior osteophyte excision) and confounding diagnoses such as adhesive capsulitis (Table 1). A history of slowly progressive shoulder pain that was worse with
activity and without acute changes in range of motion was used to distinguish osteoarthritis with radiographic evaluation. Similarly, the operative reports were valuable because the inflammatory and capsular effects were noted intraoperatively for the patients with adhesive capsulitis and not for the patients with arthritis. All patients considered for surgery had failed nonoperative treatment with activity modification, corticosteroid injections, pain medication, and 6 weeks of a home-based patient-directed physical therapy exercise program. We identified 33 patients who satisfied the inclusion and exclusion criteria. Preoperative Variables Preoperative demographic informationdincluding age, sex, hand dominance, body mass index, smoking, and medical comorbiditiesdwas collected (Table 2). All patients underwent a standardized preoperative physical examination by the senior author (K.Y.) and the following parameters were consistently recorded: pain (visual analog scale), American Shoulder and Elbow Surgeons (ASES) score,10 active forward elevation (AFE), and active external rotation (AER). All patients underwent standardized radiography, including true anteroposterior, scapular Y, and axillary views, at the same medical center and were reviewed for quality by the senior author (K.Y.). Available radiographs were independently reviewed by 2 authors (N.W.S., S.N.) who were blinded to the patient’s outcome. Three studies and one axillary view were interpreted by the operating surgeon (K.Y.) at the time of treatment and were not available for study review. The following parameters were recorded: Samilson-Prieto arthritis grade11,12 (on true anteroposterior view; classified by the size of an inferior osteophyte; kappa ¼ 0.86), posterior glenohumeral subluxation (yes/no; axis of glenoid different from that of humeral head; kappa ¼ 0.97), and glenoid morphologic characteristics (concentric/nonconcentric; kappa ¼ 1.0). Any discrepancies on available studies were resolved by discussion and consensus review. Arthritis grade was evaluated intraoperatively during diagnostic arthroscopy (0 ¼ no cartilage defects, I ¼ cartilage softening, II ¼ partial-
Table 2. Demographic Data of Patient Population Age Body mass index Sex Hand dominance Smoking (yes/no) Alcohol use (yes/no) Hypertension (yes/no) Gastroesophageal reflux disease (yes/no) Diabetes mellitus (yes/no)
55.2 9.2 SD 28.74 5.0 SD 24 male/9 female 30 right/3 left 11 (33)/22 (67) 18 (55)/15 (45) 18 (55)/15 (45) 10 (30)/23 (70) 7 (21)/26 (79)
NOTE. Data presented as n (%) unless otherwise indicated. SD, standard deviation.
496
N. W. SKELLEY ET AL.
thickness fissures, III ¼ full-thickness cartilage defect, and IV ¼ subchondral bone exposed). Surgical Technique/Rehabilitation Protocol The surgical technique for debridement and capsular release has been previously described and was very similar to that performed for adhesive capsulitis of the shoulder.13 The technique used evaluation of cartilage and standard debridement of loose cartilage, frayed labrum, and other degenerative tissues but also emphasized complete release of the rotator interval, middle, and inferior glenohumeral ligaments. Arthroscopy was performed with the patient in the beach chair position under general anesthesia. Standard posterolateral and anterior rotator interval portals were used, and an arthroscopic shaver was applied to perform glenohumeral joint debridement. Loose bodies were removed if present. Subsequently, the rotator interval was released with shaver and electrocautery until there was clear visualization of the coracoacromial ligament, and then an arthroscopic capsular biter was used to release the anterior and inferior capsule. A needle was inserted into the joint through the rotator interval, confirmed through the arthroscope, and used to inject 10 mL of 0.5% bupivacaine hydrochloride with 40 mg of triamcinolone after the wounds were closed. Postoperatively, all arms were placed in slings for comfort. A patient-directed home exercise program was initiated immediately for stretching and range of motion therapy. Follow-up Data Patients were asked to return to the clinic 3 times after surgerydat approximately 2 weeks, 6 weeks, and 3 months. Patients were also seen for continued problems after these time points. Patient charts were retrospectively reviewed to determine visual analog scale pain scores and range of motion parameters (AFE, AER) which were recorded at each visit. Patients were contacted by telephone at a minimum of 2 years postoperatively for final re-evaluation. Subsequent corticosteroid injections and reoperations, including total shoulder arthroplasty, were recorded. Outcome instruments again included the visual analog scale pain scores and ASES scores. Patients were also asked, “If you could go back in time and make the decision again, would you have undergone an arthroscopic capsular release for your shoulder condition/problem? (definitely not, probably not, neutral, probably yes, definitely yes).” Finally, patients were asked the question, “Are you satisfied with the outcome of your arthroscopic surgery? (yes/no).” Statistical Analysis All results were analyzed by statistical testing comparing preoperative measures with corresponding
postoperative measures. Paired t tests were performed for continuous data, and c-square tests were performed for categorical data. All results were considered statistically significant at P < .05. Kaplan-Meier survival analysis was used to determine the postoperative interval until an arthroplasty was performed. Multiple regression analyses were performed to evaluate preoperative demographic and examination findings that may be predictive of outcome scores or conversion to total shoulder arthroplasty.
Results Demographics The study group consisted of 33 patients, 24 (73%) men and 9 (27%) women (Table 2). The actual average follow-up for the postoperative visits was 1.8 weeks, 10.6 weeks, and 16.5 weeks. As patients continued to have pain and loss of function, the final average followup in clinic, eliminating those who went on to arthroplasty, was 40.3 weeks (range, 20 to 95 weeks). Average final telephone follow-up was 43.4 months (range, 25 to 71 months). The average age was 55.2 years (range, 34 to 71 years) at the time of operation. All patients underwent glenohumeral joint debridement and capsular release as the surgical procedure. At the time of operation, it was determined that 4 (12.1%) patients had grade II cartilage defects (3 glenoid lesions and one humeral lesion): 17 (51.5%) were grade III (7 glenoid and 6 humeral lesions and 4 lesions occurring on both sides), and 12 (36.4%) were grade IV (4 glenoid and 2 humeral lesions and 6 lesions occurring on both sides). Pain Relief and Motion Trends At the first clinic visit, patients had significant improvement in visual analog scale pain scores from 7.8 preoperatively to 1.0 postoperatively (P < .001; n ¼ 33) (Fig 1); however, patients trended back toward preoperative pain levels by the third postoperative visit and achieved a mean visual analog scale score of 7.4 at final follow-up (P ¼ .59) (Fig 1). Patients showed significant improvement in range of motion at the first postoperative visit (mean, 12.5 days). AFE improved on average from 121.8 preoperatively to 140.2 postoperatively (P < .001) and AER improved from 21.9 to 47.6 (P < .001); however, AFE and AER trended toward preoperative levels, with a mean AFE of 122.2 (Fig 2) and a mean AER of 28.4 (Fig 3) at the third (approximately 3 months) postoperative visit (P ¼ .27 and P ¼ .14, respectively; n ¼ 31). Final Outcomes and Satisfaction Of the 19 patients who did not undergo a total shoulder arthroplasty, 18 (94.7%) reported continued pain in the affected shoulder. The final in-clinic follow-up
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA
Fig 1. Patient-reported pain scores throughout the preoperative and postoperative periods. Patients who underwent total shoulder arthroplasty were removed from this analysis after undergoing the arthroplasty procedure. (Avg., average.)
for these 19 patients was 40.3 weeks (range, 20 to 95 weeks). The matched pain (P ¼ .67), forward elevation (P ¼ .56), and external rotation (P ¼ .61) scores were statistically unchanged at 40.3 weeks from the postoperative visit at 16.5 weeks. All 33 patients were available for final follow-up telephone evaluation interviews at a minimum of 2 years after the debridement and capsular release. The mean ASES score was 42.2 (range, 21 to 80) preoperatively compared with 50.8 (range, 16 to 90) (P ¼ .41) postoperatively through telephone interviews in the 19 patients who did not undergo conversion to total shoulder arthroplasty.
497
Fig 3. External rotation range of motion throughout the preoperative and postoperative periods. (Avg., average.)
After arthroscopic release, 12 (36.4%) patients reported undergoing steroid injection for pain control in the postoperative setting. When asked if they would undergo the procedure again knowing what they know now, 10 (30.3%) said “definitely not” or “probably not,” 11 (33.3%) said “definitely yes” or “probably yes,” and 12 (36.4%) were “neutral.” When asked if they were satisfied with the arthroscopic procedure, 13 (39.4%) said they were satisfied and 20 (60.6%) were not satisfied. No major complications were associated with the procedure. One patient had a superficial portal site infection that resolved with oral antibiotic therapy. Radiographic Analysis Preoperatively, the Samilson-Prieto score was grade 1 in 11 of 33 (33.3%) patients, grade 2 in 9 of 33 (27.3%) patients, and grade 3 in 13 of 33 (39.4%) patients. Posterior glenohumeral subluxation was present in 14 of 33 (42.4%) patients. Nine of 33 (27.2%) patients had nonconcentric glenoid morphologic characteristics.
Fig 2. Range of motion in forward elevation throughout the preoperative and postoperative periods. (Avg., average.)
Survival Analysis Fourteen of the 33 (42.4%) shoulders underwent total shoulder arthroplasty at an average of 38.4 weeks (range, 8 to 92 weeks) after arthroscopic debridement and capsular release (Fig 4). The average age of the patients who underwent total shoulder arthroplasty was 57.1 years (range, 43.7 to 70.2 years) at the time of the arthroscopic procedure compared with 53.7 years (range, 34.0 to 71.0 years) for those who did not undergo total shoulder arthroplasty (P ¼ .31). Only 7 (50%) patients who eventually underwent total shoulder arthroplasty were classified as Samilson-Prieto grade 3 at the time of the arthroscopic release. Six
498
N. W. SKELLEY ET AL.
(75.0%) patients and grade IV lesion occurred on both sides in 6 (85.7%) patients who underwent arthroplasty (P ¼ .002). The ASES and visual analog scale pain scores for patients who underwent total shoulder arthroplasty after the arthroscopy procedure were 70 (range, 43.8 to 96.4) and 4 (range, 0 to 7), respectively, at the most recent telephone follow-up. No added difficulties were noted in the arthroplasty operative reports related to the previous arthroscopic surgery.
Fig 4. Survival analysis of the 14 patients who eventually went on to total shoulder arthroplasty after arthroscopic debridement and capsular release. The average time to total shoulder arthroplasty for these 14 patients was 38.4 weeks or 269 days.
(31.6%) patients who did not undergo total shoulder arthroplasty were also classified as Samilson-Prieto grade 3 at the time of arthroscopic release (P ¼ .71). Of the remaining patients who eventually went on to total shoulder arthroplasty, 4 (28.6%) patients were classified as grade 2 and 3 (21.4%) were classified as grade 1. Of the patients who did not undergo arthroplasty, 5 (26.3%) were classified as grade 2 and 8 (42.1%) were classified as grade 1. Patients undergoing arthroplasty had intraoperative arthritis grades of III (6 of 17 [35.3%]) and IV (8 of 12 [66.6%]). Patients who underwent arthroplasty were more likely to have lesions on both sides at the time of arthroscopy compared with patients who did not undergo arthroplasty (Table 3). Grade III lesions occurred on both sides in 3
Regression Analysis With the numbers observed in this study, multiple linear regression analysis of the data found no significant correlations between preoperative radiographic parameters relating to visual analog scale pain scores (P ¼ .58), ASES score (P ¼ .48), or satisfaction score (P ¼ .14) at final follow-up. Similarly, no significant correlation was found between intraoperative arthritis grade relating to visual analog scale pain scores (P ¼ .43), ASES score (P ¼ .50), or satisfaction score (P ¼ .23) at final follow-up. On multiple logistic regression, age, sex, dominant side of surgery, body mass index, tobacco use, alcohol use, presence of subluxation, glenoid morphologic characteristics, cartilage grade, and Samilson-Prieto grade preoperatively were not significant predictors for conversion to total shoulder arthroplasty (P ¼ .53). Additionally, the univariate analysis of radiographic subluxation and arthritis severity preoperatively did not predict total shoulder arthroplasty (P ¼ .13 and P ¼ .35, respectively). The presence of nonconcentric glenoid morphologic features preoperatively was moderately predictive (r ¼ 0.41) of total shoulder arthroplasty (P ¼ .04).
Discussion The results in this study did not support the use of isolated arthroscopic debridement and capsular release for the treatment of glenohumeral osteoarthritis because these procedures did not provide long-term relief of pain and improved function. A recent systematic review14 of outcomes of arthroscopic debridement for glenohumeral arthritis found that Level IVetype evidence suggested improved pain level and patient satisfaction postoperatively. This review highlighted the
Table 3. Intraoperative Arthritis Grade at the Time of Arthroscopy Intraoperative Arthritis Grade Nonarthroplasty Group (n ¼ 19) Grade II Grade III Grade IV Total
Glenoid Side 3 (15.8) 5 (26.3) 2 (10.5) 10 (30.3)
NOTE. Data presented as n (%).
Humeral Side 1 (5.3) 5 (26.3) 1 (5.3) 7 (21.2)
Arthroplasty Group (n ¼ 14) Both e 1 (5.3) 1 (5.3) 2 (6.1)
Glenoid Side e 2 (14.3) 1 (7.1) 3 (9.1)
Humeral Side e 1 (7.1) 1 (7.1) 2 (6.1)
Both e 3 (21.4) 6 (42.9) 9 (27.2)
Total (N ¼ 33) 4 (12.1) 17 (51.5) 12 (36.4) 33 (100)
DEBRIDEMENT AND CAPSULAR RELEASE IN SHOULDER OA
deficits in the literature and suggested that future studies comprehensively define patient characteristics and trend results over time.14 The current review noted the difficulty in interpreting debridement and capsular release based on confounding variables. We had a known cohort of young patients with isolated arthritis who were not affected by concomitant pathologic processes or procedures. These patients underwent isolated debridement and capsular release as a less invasive alternative to arthroplasty. During the study period, we found a rapid return of pain and stiffness after surgery and low satisfaction and survival rates, calling into question the benefit of isolated capsular release and debridement for the treatment of shoulder osteoarthritis. These findings led us to reject our study hypothesis that these procedures in isolation would provide significant long-term (>2 years) improvement in pain and function for patients with shoulder osteoarthritis. Functional outcomes and satisfaction scores in our population were lower than those in other reports in the literature that used concomitant procedures such as biceps tenotomy, distal clavicle excision, and acromioplasty, among others.3-8,14 These studies found better results in patients by combining procedures. Most of our patients (60.6%) were not satisfied with the isolated procedure, and only 33.3% said that they would have the capsular release procedure again. A large proportion, 14 (42.4%) patients, were dissatisfied with the procedure as evidenced by the high rate of conversion to total shoulder arthroplasty in a relatively short time (8.8 months) after the index arthroscopic debridement and capsular release. This high conversion rate was believed to largely result from the lack of longer pain relief provided by the arthroscopic procedure. Our study was consistent with other smaller studies that did not find an association between demographic factors, preoperative radiographic grade of arthritis, or arthroscopic grade and functional outcome.6,7,15,16 We did, however, find that nonconcentric glenoid morphologic characteristics preoperatively were moderately predictive (r ¼ 0.41) of the patient undergoing a total shoulder arthroplasty in the postoperative period, and the presence of lesions on both sides was significantly associated with undergoing total shoulder arthroplasty. Previous studies have shown benefit from concomitant procedures.3-8,14 Our results from isolated debridement and capsular release, however, showed that this procedure may be unnecessary and may provide little to no benefit when combined with other procedures. Van Thiel et al.7 reported that 22% of their patients underwent arthroplasty at a mean of 10.1 months after capsular release, debridement, and other concomitant procedures. As expected, the patients who underwent conversion to total shoulder arthroplasty had better
499
ASES scores, greater pain relief, and improved range of motion compared with the patients who underwent only an arthroscopic debridement and capsular release. Van Thiel et al. also reported a higher Samilson-Prieto arthritis score as a risk factor for conversion to arthroplasty during the postoperative period. We were unable to find an association between demographic variables or radiographic arthritis grade and the need for conversion to total shoulder arthroplasty. Our results were unique within the literature and could be related to several factors. We focused only on isolated arthroscopic debridement and capsular release as a treatment for glenohumeral arthritis and thus eliminated the effects of additional procedures on clinical and functional outcomes. Our patients were seen at a large referral center, which may have influenced treatment with arthroplasty. The patients in our series were young but had an older average age compared with other similar studies in the literature.4,6,7,9 They may have had more severe arthritis at the time of surgery: many other studies did not characterize osteoarthritis grade. Other investigations used different techniques to perform the capsular release because there is no standardized procedure.4-7 Whether the technique for capsular release had a significant effect on the maintenance of motion or pain relief, or both, after arthroscopic debridement and capsular release is unclear and could be the basis for future study. Similarly, postoperative rehabilitation protocols were not well established or defined in the literature and may influence outcome comparisons. Interestingly, the anesthetic and corticosteroid injection at the end of the procedure may have been a significant reason for the temporary improvement in pain and motion observed at the first postoperative visit. This would explain the rapid return to preoperative pain and range of motion in our study. Although complications are rare from arthroscopic procedures, this lack of persistent improvement after arthroscopic debridement and capsular release may place the patient at an unnecessary surgical risk.13,17,18 Limitations Our study is limited by the inherent design factors associated with a retrospective analysis without a control group. Patients were contacted at different times after the procedure and this could affect patient satisfaction. The study is strengthened by complete followup through telephone interviews but is limited by the minimum length of telephone follow-up at 2 years. Only one surgeon’s standardized technique for arthroscopic capsular release was analyzed, and this could influence generalizability. Preoperative and postoperative range of motion was conducted in an unblinded fashion by the operating surgeon and was at risk for bias. We also recognize that there could have
500
N. W. SKELLEY ET AL.
been operative bias to pursue total shoulder arthroplasty in the postoperative period by the same surgeon; however, an argument could be made that the bias may have been more likely to be toward success of the arthroscopic procedure and thus avoidance of additional surgery.
Conclusions Isolated arthroscopic debridement and capsular release without any other procedures were associated with only temporary pain relief and improvement in motion. Although there are limited nonarthroplasty surgical options available for glenohumeral arthritis, isolated arthroscopic debridement and capsular release may not provide substantial benefit to justify use in most patients. Patients should be carefully counseled on the expectations associated with isolated arthroscopic debridement and capsular release.
References 1. Carter MJ, Mikuls TR, Nayak S, Fehringer EV, Michaud K. Impact of total shoulder arthroplasty on generic and shoulder-specific health-related quality-of-life measures: A systematic literature review and meta-analysis. J Bone Joint Surg Am 2012;94:e1271-e1279. 2. Matsen FA III. Early effectiveness of shoulder arthroplasty for patients who have primary glenohumeral degenerative joint disease. J Bone Joint Surg Am 1996;78:260-264. 3. Ogilvie-Harris DJ, Wiley AM. Arthroscopic surgery of the shoulder. A general appraisal. J Bone Joint Surg Br 1986;68:201-207. 4. Millett PJ, Gaskill TR. Arthroscopic management of glenohumeral arthrosis: humeral osteoplasty, capsular release, and arthroscopic axillary nerve release as a jointpreserving approach. Arthroscopy 2011;27:1296-1303. 5. Richards DP, Burkhart SS. Arthroscopic debridement and capsular release for glenohumeral osteoarthritis. Arthroscopy 2007;23:1019-1022. 6. Cameron BD, Galatz LM, Ramsey ML, Williams GR, Iannotti JP. Non-prosthetic management of grade IV
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17. 18.
osteochondral lesions of the glenohumeral joint. J Shoulder Elbow Surg 2002;11:25-32. Van Thiel GS, Sheehan S, Frank RM, et al. Retrospective analysis of arthroscopic management of glenohumeral degenerative disease. Arthroscopy 2010;26:1451-1455. Ogilvie-Harris DJ, Gordon R, MacKay M. Arthroscopic treatment for posterior impingement in degenerative arthritis of the elbow. Arthroscopy 1995;11:437-443. Weinstein DM, Bucchieri JS, Pollock RG, Flatow EL, Bigliani LU. Arthroscopic debridement of the shoulder for osteoarthritis. Arthroscopy 2000;16:471-476. Richards RR, An KN, Bigliani LU, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3:347-352. Samilson RL, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 1983;65:456-460. Ilg A, Bankes M, Emery R. The intra- and inter-observer reliability of the Samilson and Prieto grading system of glenohumeral arthropathy. Knee Surg Sports Traumatol Arthrosc 2001;9:187-190. Yamaguchi K, Sethi N, Bauer GS. Postoperative pain control following arthroscopic release of adhesive capsulitis. Arthroscopy 2002;18:359-365. Namdari S, Skelley N, Keener JD, Galatz LM, Yamaguchi K. What Is the role of arthroscopic debridement for glenohumeral arthritis? A critical examination of the literature. Arthroscopy 2013;29:1392-1398. Guyette TM, Bae H, Warren RF, Craig E, Wickiewicz TL. Results of arthroscopic subacromial decompression in patients with subacromial impingement and glenohumeral degenerative joint disease. J Shoulder Elbow Surg 2002;11:299-304. Kerr BJ, McCarty EC. Outcome of arthroscopic débridement is worse for patients with glenohumeral arthritis of both sides of the joint. Clin Orthop Relat Res 2008;466: 634-638. Marecek GS, Saltzman MD. Complications in shoulder arthroscopy. Orthopedics 2010;33:492-497. De Wachter J, Van Glabbeek F, van Riet R, Van Leemput W, Vermeyen K, Somville J. Surrounding soft tissue pressure during shoulder arthroscopy. Acta Orthop Belg 2005;71:521-527.