Archives of Physical Medicine and Rehabilitation. journal homepage:

Transcription

1 Archives of Physical Medicine and Rehabilitation journal homepage: Archives of Physical Medicine and Rehabilitation 2013;94: REVIEW ARTICLE (META-ANALYSIS) Subacromial Impingement Syndrome: Effectiveness of Pharmaceutical InterventionseNonsteroidal Anti-Inflammatory Drugs, Corticosteroid, or Other Injections: A Systematic Review Renske van der Sande, MD, a Willem D. Rinkel, MSc, b Lukas Gebremariam, MD, a Elaine M. Hay, FRCP, MD, c Bart W. Koes, PhD, a Bionka M. Huisstede, PhD a,b From the Departments of a General Practice and b Rehabilitation Medicine, Erasmus MC e University Medical Center, Rotterdam, The Netherlands; and c the Arthritis Research Campaign National Primary Care Centre, Keele University, Keele, United Kingdom. Abstract Objective: To present an evidence-based overview of the effectiveness of pharmaceutical interventions, including nonsteroidal anti-inflammatory drugs, corticosteroid injections, and other injections, used to treat the subacromial impingement syndrome (SIS). An overview can help physicians select the most appropriate pharmaceutical intervention, and it can identify gaps in scientific knowledge. Data Sources: The Cochrane Library, PubMed, Embase, PEDro, and CINAHL databases. Study Selection: Two reviewers independently selected relevant reviews and randomized clinical trials. Data Extraction: Two reviewers independently extracted the data and assessed the methodologic quality. Data Synthesis: A best evidence synthesis was used to summarize the results. Three reviews and 5 randomized clinical trials were included. Although we found limited evidence for effectiveness in favor of 2 sessions with corticosteroid injections versus 1 session, for the effectiveness of corticosteroid injections versus placebo, nonsteroidal anti-inflammatory drugs, or acupuncture, only conflicting and no evidence for effectiveness was found. Moderate evidence was found in favor of immediate release oral ibuprofen compared with sustained-released ibuprofen in the shortterm. Also, moderate evidence for effectiveness was found in favor of glyceryltrinitrate patches versus placebo patches in the short-term and mid term. Furthermore, injections with disodium ethylene diamine tetraacetic acid plus ultrasound with ethylene diamine tetraacetic acid gel were more effective (moderate evidence) than was placebo treatment in the short- and long-term. Conclusions: This article presents an overview of the effectiveness of pharmaceutical interventions for SIS. Some treatments seem to be promising (moderate evidence) to treat SIS, but more research is needed before firm conclusions can be drawn. Archives of Physical Medicine and Rehabilitation 2013;94: ª 2013 by the American Congress of Rehabilitation Medicine Musculoskeletal disorders of the shoulder, including tendinitis and bursitis, are difficult to differentiate in clinical practice. In the Complaints of the Arm, Neck, and/or Shoulder (CANS) model, the term subacromial impingement syndrome (SIS) is used for the rotator cuff syndrome, tendonosis of the Musculus infraspinatus, Musculus supraspinatus, and Musculus subscapularis, and bursitis in the shoulder area. 1 More than 50% of the patients suffering from chronic CANS reported complaints of the shoulder. 2 The relation No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. between shoulder complaints and work-related factors, such as repetitive work, working with the hand above the shoulder, and high psychosocial job demands, has been found positive by several authors. 3 In general practice, SIS is the most frequently reported diagnosis of the shoulder, with a cumulative incidence of 5 per 1000 patients per year. 4 Patients with SIS are characterized by pain localized in the shoulder that is exacerbated when performing overhead activities. 5 The first step in treatment for SIS by a general practitioner often includes an analgesic. 6 Also, corticosteroid injections are an often-used intervention in primary care. 7 New treatment modalities such as tenoxicam injections /13/$36 - see front matter ª 2013 by the American Congress of Rehabilitation Medicine

2 962 R. van der Sande et al have been studied for their effectiveness. A systematic review on this subject is necessary to give an evidence-based overview of (new) studies 8-12 and the effectiveness of pharmaceutical interventions, that is, nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroid injections, and other injections, to treat SIS. Methods Search strategy The Cochrane Library, PubMed, Embase, PEDro, and CINAHL were searched up to March Keywords related to SIS and interventions were included. The complete search strategy is available on request. Inclusion criteria Systematic reviews and randomized clinical trials (RCTs) were included if they fulfilled all of the following criteria: (a) acute or chronic SIS (grades IeIV), not caused by an acute trauma or any systemic disease as described in the definition of CANS, was studied; (b) an intervention for treating SIS was evaluated; (c) results on pain, function, or recovery were reported; and (d) a follow-up period of 2 weeks was reported. There were no language restrictions. After the full-text articles were included, we divided the included studies into different treatment groups for which separate reviews could be conducted. One of these groups is pharmaceutical intervention. For this review, only those studies were included in which pharmaceutical therapy (oral, injected, or patched) was compared with placebo, no treatment, or another nonsurgical treatment. Study selection Two reviewers (B.M.H., L.G.) independently applied the inclusion criteria to select potential relevant studies from the title, abstracts, and full-text articles, respectively. A consensus method was used to solve any disagreements concerning inclusion of studies, and a third reviewer (B.W.K.) was consulted if disagreement persisted. Categorization of the relevant literature Relevant articles are categorized under 3 headers: Systematic reviews describes all (Cochrane) reviews; Recent RCTs contains all RCTs published after the search date of the systematic review on the same intervention; Additional RCTs describes all RCTs concerning an intervention that has not yet been described in a systematic review. Data extraction Three authors (L.G., R.S., B.M.H.) independently extracted data from the included articles. Information was collected on the study List of abbreviations: CANS Complaints of the Arm, Neck, and/or Shoulder CI confidence interval EDTA ethylene diamine tetraacetic acid GTN glyceryltrinitrate NSAIDs nonsteroidal anti-inflammatory drugs RCT randomized clinical trial ROM range of motion SIS subacromial impingement syndrome population, interventions, and outcome measures. A consensus procedure was used to solve any disagreement between the authors. Results were reported in short-term (3mo), midterm (4e6mo), and long-term (>6mo). Methodologic quality assessment Two reviewers (L.G., M.S.R.) independently assessed the methodologic quality of each RCT using Furlan s 12 criteria (table 1). 13 Each item was scored as yes, no, or unclear. High-quality was defined as a yes score of 50%. A consensus procedure was used to solve disagreement between the reviewers. In a (Cochrane) review, the use of a methodologic quality assessment is a standard procedure. We describe the methodologic quality scale/criteria that were used in the review, and used the authors definitions of high and low quality for the included studies. Data synthesis A quantitative analysis of the studies was not possible due to heterogeneity of the outcome measures. Therefore, we summarized the results using a best-evidence synthesis. 14 The article was included in the best-evidence synthesis only if a comparison was made between the groups and the level of significance was reported. The results of the study were labeled significant if 1 of the 3 outcome measures on pain, function, or recovery reported significant results. The levels of evidence for effectiveness are given in table 2. Results Characteristics of the included studies The initial search resulted in 5 reviews from the Cochrane library. Via PubMed 5 reviews and 215 RCTs, via Embase 21 reviews and 193 RCTs, via CINAHL 141 reviews/rcts, and via PEDro 0 reviews and 13 RCTs were identified. Finally, 3 reviews and 5 RCTs reported on the effectiveness of pharmaceutical interventions and were included. The review of Green et al 15 that studied the effectiveness of physiotherapy for shoulder pain included 26 trials; 11 trials reported on SIS and were included in the present review. Another review of Green et al 16 studied the effectiveness of acupuncture for shoulder pain and included 9 trials; 2 trials reported on SIS and were included in the present review. The review of Buchbinder et al 17 studied the effectiveness of corticosteroid injections for shoulder pain. Seven of the 26 trials included in this review met our inclusion criteria and were included in the present review. The characteristics of the included studies are listed in appendices 1, 2, and 3. Methodologic quality assessment The results of the methodologic quality assessment are presented in table 3. Four (of the 5 included) RCTs were of high quality. Most prevalent methodologic flaws were (1) care provider not blinded and (2) unclear whether allocation was concealed. The 3 reviews used a methodologic scoring list of 11 items 16 or 5 items 17 : 18 of the 20 RCTs in these reviews scored 50% on the

3 Pharmaceutics for subacromial impingement syndrome 963 Table 1 Methodologic quality assessment: sources of risk bias A. 1. Was the method of randomization adequate? B. 2. Was the treatment allocation concealed? C. Was knowledge of the allocated interventions adequately prevented during the study? 3. Was the patient blinded to the intervention? 4. Was the care provider blinded to the intervention? 5. Was the outcome assessor blinded to the intervention? D. Were incomplete outcome data adequately addressed? 6. Was the dropout rate described and acceptable? 7. Were all randomized participants analyzed in the group to which they were allocated? E. 8. Are reports of the study free of suggestion of selective outcome reporting? F. Other sources of potential bias: 9. Were the groups similar at baseline regarding the most important prognostic indicators? 10. Were cointerventions avoided or similar? 11. Was the compliance acceptable in all groups? 12. Was the timing of the outcome assessment similar in all groups? quality assessment and were classified as high quality. The RCT by Berry et al 18 was included in all the 3 reviews. Effectiveness of pharmaceutical interventions for SIS Table 4 shows the interventions for which moderate or strong evidence for effectiveness for pharmaceutical interventions for SIS was found. Table 5 presents a complete overview. Effectiveness of analgesics: NSAIDs versus laser therapy In a systematic review, 1 high-quality RCT 19 (nz20) included in the review of Green et al 15 compared NSAIDs (naproxen sodium 550mg 2 times a day for 2wk) with laser therapy (902nm) for shoulder tendinitis. Significantly less pain (mean difference, 2; 95% confidence interval [CI]Z1e3.5) was reported for laser versus NSAIDs at 2-week follow-up. No significant results were found on function. There is moderate evidence for the effectiveness of laser therapy versus NSAIDs in the short-term (2wk). Effectiveness of analgesics: ibuprofen tablet versus sustainedrelease ibuprofen One high-quality RCT 11 (nz147) compared conventional tablets ibuprofen (600mg 4 times a day) with sustained-release tablets Table 2 Levels of evidence Strong evidence: consistent (ie, when 75% of the trials report the same findings) positive (significant) findings within multiple higher quality RCTs Moderate evidence: consistent positive (significant) findings within multiple lower quality RCTs and/or 1 high-quality RCT Limited evidence for effectiveness: positive (significant) findings within 1 low-quality RCT Conflicting evidence: provided by conflicting (significant) findings in the RCTs (<75% of the studies reported consistent findings) No evidence: RCT(s) available, but no (significant) differences between the intervention and control groups were reported ibuprofen (1200mg 2 times a day). At 3-week follow-up, significant differences between the groups were found in favor of conventional tablets ibuprofen on complete pain relief of shoulder tendonitis (7% vs 21%, PZ.02) and on patients global assessment on a 5-point scale (no data given, PZ.04); no significant differences were found on the patient s assessment concerning better or complete relief, or the physician s assessment. There is moderate evidence for effectiveness in favor of conventional ibuprofen tablets versus sustained-released ibuprofen tablets for SIS in the short-term (3wk). Effectiveness of analgesics: glyceryltrinitrate transdermal patches versus placebo A high-quality RCT 12 (nz53; 57 shoulders) compared glyceryltrinitrate (GTN) patches with placebo patches. Both groups also received a tendon rehabilitation program. Significant differences were found between the groups on decrease in pain at rest in favor of the GTN patches group at 12- (PZ.04) and 24-week (PZ.03) follow-up (no further data given). From baseline to 24 weeks, the range of motion (ROM) increased by 24% (17%e31%) in the GTN group versus 8% (4%e13%) in the placebo group (no P value given). Increases in force (supraspinatus, subscapularis, adduction, and external and internal rotation) were all significant in favor of the GTN group at 12- and 24-week follow-up (, P value ranged from.001 to.02). At 24-week follow-up, there were significant differences between the groups for the outcome asymptomatic shoulder in favor of the GTN group (64% vs 24%, PZ.007; weighted mean difference,.26; 95% CIZ.19e.32). There is moderate evidence for the effectiveness of GTN patches versus placebo for SIS in the short- (12wk) and midterm (24wk). Effectiveness of injections: coricosteroid injections versus placebo Seven RCTs in the systematic review of Buchbinder et al 17 compared corticosteroid injections with placebo injections. Five of the 7 trials compared subacromial steroid injection with placebo. 18,20-23 One trial compared intraarticular injection with placebo, 24 and another trial compared supraspinatus tendon injection with placebo. 25 Two high-quality trials 20,22 (nz90) reported a significant improvement between the groups in patients treated with corticosteroid injections with respect to pain (standard mean difference,.83; 95% CIZ.39e1.26), ROM (abduction) (standard mean difference,.63; 95% CIZ.20e1.06), and function (standard mean difference,.82; 95% CIZ.39e1.25) at 4-week follow-up. A low-quality study 21 (nz40) reported significantly better results () in the corticosteroid group versus placebo on pain and ROM at a mean of 33-week and 28-week follow-up, respectively. Furthermore, no significant differences between the groups were reported on improvement in daily activities (no exact follow-up time given). A high-quality trial 18 (nz24) found no significant differences on pain, ROM (abduction), or success rate between the 2 groups at 4-week follow-up. Another high-quality study 26 (nz57) also found no significant differences on pain and ROM (abduction) between the 2 groups (no further data were given) at 12-week follow-up. Strobel (low quality) 24 (nz40) reported significantly greater reduction in pain in the placebo group versus the treatment group at 90 and 360 days. Another high-quality study 25 (nz25) studied treatment with supraspinatus tendon injection versus placebo in patients with supraspinatus tendonitis. No significant differences were found on pain improvement or analgesic consumption between the 2 groups at 2- and 8-week follow-up.

4 964 R. van der Sande et al Table 3 Reference Methodologic quality scores of the included recent and additional RCTs Adequate Randomization? Allocation Concealment? Blinding? Patients? Blinding? Caregiver? Blinding? Outcome Assessors? Incomplete Outcome Data Addressed? Dropouts? Paoloni et al 12 þ þ þ þ þ þ Akgun et al 9 þ? þ þ þ þ Friis et al 11 þ? þ?? þ? Cacchio et al 10 þ? þ þ þ þ Itzkowitch et al 8?? þ?? þ Methodologic quality scores of Green et al 15 * Random Allocation? Concealed Allocation? Blind Subjects? Blind Therapists? Blind Assessors? Between-group Comparisons? Ebenbichler 27 þ þ þ þ þ þ þ Nykanen 28 þ þ þ þ þ þ Saunders 29 þ? þ þ þ þ Vecchio et al 23 þ þ þ þ þ þ Berry et al 18 þ þ þ þ Conroy 30 þ þ þ þ Dal Conte 31 þ þ þ þ þ England et al 19 þ þ þ þ þ Bang 32 þ þ þ Brox 33 þ þ þ Perron 34 þ þ þ þ Methodologic quality scores of Green et al 16y Berry et al 18 þ n.p. þ þ þ Methodologic quality scores of Buchbinder et al 17z Blind Participants? Blind Assessors? Acceptable Lost to Follow-up? x Appropriate Statistical Analysis? (ITT Analysis?) Adebajo et al 20 þ þ þ þ þ Petri et al 22 þ þ þ þ þ Withrington et al 25 þ þ þ þ? Berry et al 18? þ þ þ? Vecchio et al 23 þ þ þ? White et al 35 þ þ þ Blair et al 21 þ þ??? Strobel 24??? Allocation Concealed Incomplete Outcome Data? ITT Analysis? Adequate Follow-up? Abbreviations: þ, yes;, no;?, unclear/unsure; ITT, intention-to-treat; n.p., not possible. * Based on the PEDro scale (specifically designed and validated for the assessment of validity for trials of physiotherapy interventions) ( y Based on the Cochrane reviewer s handbook (Alderson P, Green S, Higgins JPT, editors. Cochrane reviewers handbook. Chichester, UK: John Wiley & Sons, Ltd; December 2003). z Adapted from the Cochrane reviewers handbook (Clarke M, Oxman AD, editors. Cochrane reviewers handbook 4.0 [updated July 1999]. In: The Cochrane Library [database on CDROM]. The Cochrane Collaboration. Oxford: Update Software; 2000, Issue 2, 1999). x Lost to follow-up: We defined this item as positive if the percentage of lost to follow-up is <20% for the short-term follow-up and <30% for the long-term follow-up. A recent high-quality RCT 9 (nz48) compared 3 groups according to 2 therapeutic injections given at a 10-day interval: group 1 received corticosteroid injections at the first and second sessions, group 2 received corticosteroid injections at the first session and placebo injections at the second session, and group 3 received placebo injections at both sessions. Only on pain causing sleep disturbance, significant differences between the groups were found in favor of group 1 a month after treatment (P<.001). However, at 3-month follow-up, this significant difference was not sustained. Furthermore, no significant differences were found on the Constant score between the 3 groups at 1- and 3-month follow-up. We found conflicting evidence for the effectiveness of corticosteroid injections versus placebo for SIS in the short- (4wk) and long-term (12wk). Effectiveness of injections: coricosteroid injections versus acupuncture In a systematic review, 1 high-quality RCT 18 (nz24) included in the review of Green et al 16 compared acupuncture with

5 Pharmaceutics for subacromial impingement syndrome 965 Table 3 Continued Free of Suggestions of Selective Outcome Reporting? Similarity of Baseline Characteristics? Cointerventions Avoided or Similar? Compliance Acceptable in all Groups? Timing of the Outcome Assessment Similar? Maximum Score Score Study þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ?? þ þ??? þ Methodologic quality scores of Green et al 15 * ITT Analysis? Baseline comparability? Availability of Point Estimates and Measures of Variability of Primary Outcome Measures? Specification of Eligibility Criteria? Score Maximum Score Study þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ Methodologic quality scores of Green et al 16y þ þ þ Methodologic quality scores of Buchbinder et al 17z Score Maximum Score Study Percentage Percentage Percentage moxibustion (a form of fire heat treatment that stimulates specific acupuncture points of the body) with corticosteroid injections. Postintervention pain, range of abduction, and success in the short-term showed no significant differences between the 2 groups. There is no evidence for the effectiveness of corticosteroid injections versus acupuncture plus moxibustion in the shortterm (4wk). Effectiveness of injections: coricosteroid injections versus NSAIDs In a systematic review, 3 high-quality trials 20,22,35 (nz120) included in the review of Buchbinder et al 17 compared subacromial corticosteroid injections with NSAIDs. No significant differences between the groups were found on pain and ROM at 4- to 6-wk follow-up, 20,22,35 on improvement in function at 4-week follow-up, 20,22 and on global assessment at 6-week follow-up. 35

6 966 R. van der Sande et al Table 4 Evidence for the effectiveness of pharmaceutical interventions for SIS SIS Conservative Oral Injection Other O b O c,f O a,d,e NOTE. O, strong or moderate evidence found; 0, RCT(s) found, but only limited, conflicting, or no evidence for the effectiveness of interventions was found; empty cells, no RCTs or reviews found. Short-term: a: Moderate evidence: NSAID vs laser therapy.* b: Moderate evidence: Ibuprofen* vs sustained-release ibuprofen. c: Moderate evidence: Disodium EDTA plus ultrasound* vs placebo. d: Moderate evidence: GTN patches* vs placebo. Midterm: e: Moderate evidence: GTN patches* vs placebo. Long-term: f: Moderate evidence: Disodium EDTA plus ultrasound* vs placebo. * In favor of. There is no evidence for the effectiveness of corticosteroid injections versus NSAIDs for SIS in the short-term (4e6wk). Effectiveness of injections: coricosteroid injection plus NSAID versus placebo In a systematic review, Berry et al 18 (nz24) compared corticosteroid injection plus NSAID versus placebo. No significant differences between the 2 groups were found on pain and success rate at 4-week follow-up. On the range of abduction, significant differences were found in favor of placebo (weighted mean difference, 27.60; 95% CIZ to 5.21). There is no evidence for the effectiveness of corticosteroid plus NSAID versus placebo for SIS in the short-term (4wk). Effectiveness of injections: coricosteroid injection plus NSAID versus NSAID In a systematic review, Petri et al 22 (nz50) studied the effectiveness of corticosteroid plus NSAID versus NSAID alone. No significant differences were reported on pain, improvement in function, and improvement in range of abduction between the groups at 4-wk follow-up. There is no evidence for the effectiveness of corticosteroid injection plus NSAID versus NSAID alone for SIS in the short-term (4wk). Effectiveness of injections: coricosteroid injection (plus placebo NSAID) versus acupuncture In a systematic review, 1 high-quality RCT 18 (nz24) compared steroid injection plus placebo NSAID and steroid injection plus NSAID with acupuncture. No significant differences were found for postintervention pain or range of abduction. There is no evidence for the effectiveness of corticosteroid injection (plus placebo NSAID) versus acupuncture for SIS in the short-term (4wk). Effectiveness of injections: tenoxicam injections versus placebo injections One low-quality RCT 8 compared periarticular tenoxicam injections (20mg/2mL) with placebo injections once a week for up to 4 weeks in 80 patients with rotator cuff tendinitis. Significant results (P<.001) were found in favor of tenoxicam injections on pain at rest and on pain during active movement. Also, significant results (P<.001) were found on improvement of the active shoulder mobility in favor of the treatment group. There is limited evidence for the effectiveness of tenoxicam injections versus placebo for rotator cuff tendonitis in the short-term. Effectiveness of injections: disodium ethylene diamine tetraacetic acid plus ultrasound versus placebo A high-quality study 10 compared disodium ethylene diamine tetraacetic acid (EDTA) injections plus ultrasound (using disodium EDTA gel) with placebo in patients with calcific tendinitis (nz80). Disodium EDTA could bind and remove calcium deposits on the tendon. The treatment group showed significantly (P<.01) more improvements on pain ( vs ) and on the Constant and Murley score ( vs ) at 1-week and 1-year follow-up ( vs and vs , respectively). The treatment group had a significantly (P<.01) smaller amount of calcification versus the placebo group after the treatment ( vs ). We found moderate evidence for the effectiveness of disodium EDTA plus ultrasound versus placebo in treating patients with calcific tendonitis in the short-term (1wk) and the long-term (52wk). Discussion In this review, we give an overview of the evidence for effectiveness of all pharmaceutical therapiesdnsaids, corticosteroid injections, and other injectionsdstudied in a randomized controlled setting in the treatment of SIS. Analgesics According to the results of this review, for physicians wishing to give analgesics to treat SIS we would recommend the use of oral ibuprofen rather than sustained-release ibuprofen (for 3wk; moderate evidence). However, our review has highlighted the lack of evidence regarding the use of simple analgesics (paracetamol/acetaminophen), mild opioids (eg, codeine phosphate), or other commonly used NSAIDs for SIS. Although we found moderate evidence for the short-term (3mo) and mid-term (6mo) benefits of GTN patches, the body s systemic responses on GTN or the pharmacokinetics are not entirely clear. Injections For other specific CANS, strong and moderate evidence was found in favor of the effectiveness of corticosteroid injections versus placebo in the short-term but not in the long-term. 36,37 In contrast to these findings, for SIS we found conflicting evidence in the short- and long-term. More research is needed to allow us to draw definite conclusions about the effectiveness of corticosteroid injections for treating SIS. Future research should also concentrate on the long-term benefits and disadvantages, as well as the accuracy and safety, of corticosteroid injections. A new treatment modality, an injection with disodium EDTA, seems to be promising in both the short-term (1wk) and the longterm. Disodium EDTA is a heavy metal and mineral chelator amino acid. EDTA can bind and remove calcium deposits on the rotator cuff tendon. 38 Cacchio et al 10 concluded that disodium EDTA administered through single needle mesotherpy and ionopheresis is safe and effective for treating patients with calcific tendinitis. However, this conclusion is based on 1 RCTonly and more research is needed to confirm the effectiveness and safety of disodium EDTA.

7 Pharmaceutics for subacromial impingement syndrome 967 Table 5 Oral Treatment CANS: Evidence for the effectiveness of pharmaceutical interventions for SIS Injections Corticosteroid Tenoxicam NSAID < NSAID vs laser therapy* < Corticosteroid injection vs placebo < Tenoxicam* injection vs placebo Short-term þþ Short-term Short-term þ Ibuprofen Long-term < Ibuprofen* vs sustained-release ibuprofen < Corticosteroid injection 1 session vs 2 sessions* Disodium EDTA < Disodium EDTA plus ultrasound (with EDTA gel) vs placebo Short-term þþ þ Short-term þþ Long-term þþ < Corticosteroid injection vs acupuncture plus moxibustion Other conservative Short-term NE GTN patches < GTN patches* vs placebo < Corticosteroid injection vs Short-term þþ NSAID Mid-term þþ Short-term < Corticosteroid injection plus NSAID vs placebo Short-term < Corticosteroid injection plus NSAID vs NSAID Short-term < Corticosteroid injection plus placebo NSAID vs acupuncture Short-term NOTE. þ, limited evidence found; þþ, moderate evidence found; þþþ, strong evidence found;, conflicting evidence for effectiveness; NE, no evidence found for effectiveness of the treatment: RCT(s) available, but no differences between intervention and control groups were found. * In favor of. NE NE NE NE Study limitations The present review has some limitations. First, the methodologic quality assessment of the RCTs included in the reviews and our methodologic quality assessment of the recent and additional RCTs differ greatly; for example, we defined a study as high quality when the study scored 50% on the quality assessment. However, in the Cochrane reviews, no clear definition is given as to whether the study is considered to be of high or low quality. Buchbinder et al 17 used only 5 items, in contrast to the 12 items of Furlan et al 13 that we used to assess the methodologic quality of the recent and additional RCTs. Therefore, a risk of bias can be introduced by categorizing some of the studies as high-quality studies. Second, some studies with a small sample size were categorized as highquality studies. Therefore, the results must be viewed with caution. Furthermore, the RCTs in which corticosteroids are studied must be interpreted with the notion that different types and dosages of corticosteroids are compared with each other. Conclusions Moderate evidence for effectiveness was found in the short-term in favor of oral ibuprofen versus sustained-release ibuprofen. Furthermore, promising results were found for the effectiveness of GTN and injections with disodium EDTA plus ultrasound with EDTA gel. GTN patches were more effective (moderate evidence) than placebo patches in the short- and midterm. Injections with disodium EDTA plus ultrasound with EDTA gel were more effective (moderate evidence) than placebo treatment in the shortand long-term. Keywords Drug therapy; Pain; Rehabilitation; Shoulder impingement syndrome; Treatment outcome Corresponding author Bionka M. Huisstede, PhD, Department of Rehabilitation, Erasmus MC e University Medical Center Rotterdam, Room H-016, PO Box 2040, 3000 CA Rotterdam, The Netherlands. address: BMA.Huisstede@gmail.com. Acknowledgment We thank M.S. Randsdorp (M.S.R.), MD, for her participation in the quality assessment.

8 Appendix 1 Author Buchbinder et al 17 Data extractiondsystematic reviews Total Number of Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size Corticosteroid injections Corticosteroid vs placebo injection Adebajo et al Subacromial steroid injection and Petri et al 22 (1mL of 80mg/mL triamcinolone hexacetomide) (nz45) Rotator cuff tendonitis Blair et al Subacromial steroid injection SIS (2mL of 40mg/mL triamcinolone acetonide) (nz19) Berry et al Intraarticular steroid injection with 40mg methylprednisolone (nz12) Rotator cuff lesion Vecchio et al Subacromial steroid injection Rotator cuff tendonitis (40mg methylprednisolone) (nz29) Strobel 24 Subacromial bursitis or 40 Subacromial steroid injection (20mg triamcinolone hexacetonide (nz20) supraspinatus tendinitis Withrington et al Supraspinatus injection with Supraspinatus tendonitis 80mg methylprednisolone (nz12) Berry et al 18 Rotator cuff lesion Corticosteroid vs acupuncture Intraarticular steroid injection (40mg methylprednisolone) (nz12) Placebo (nz45) Pain improvement (VAS) (4wk) SMD 0.83 (95% CIZ0.39e1.26) ifo injection Improvement in function (4wk) SMD 0.63 (95% CIZ0.20e1.06) ifo injection Improvement in ROM abduction (4wk) SMD 0.82 (95% CIZ0.39e1.25) ifo injection Placebo (nz21) No data were given At a mean of 33wk in corticosteroid group and 28wk in placebo group, the corticosteroid group was significantly better with respect to pain and ROM. No significant difference between the groups with respect to improvement (ADL) Placebo (nz12) Pain (VAS) (4wk) WMD 4.60 (95% CIZ to 25.19) Range of abduction (4wk) WMD (95% CIZ to 7.10) Success rate* RR 0.67 (95% CIZ0.35e1.28) Placebo (nz28) No data were given No significant differences with respect to pain and ROM between the treatment and placebo groups Placebo (nz20) No data were given Greater reduction in pain in the placebo group at 90 and 360d Saline (nz13) No data were given No difference in improvement in pain or analgesic consumption between the 2 groups at 2 and 8wk Acupuncture (once a week) (nz12) Pain (VAS) (4wk) WMD 7.50 (95% CIZ to 12.47) Range of abduction (4wk) WMD 2.90 (95% CIZ to 26.83) Success rate (4wk) RR 0.83 (95% CIZ0.35e2.00) (continued on next page) 968 R. van der Sande et al

9 Appendix 1 (continued) Author Adebajo et al, 20 Petri et al, 22 White et al 35 Total Number of Patients Treatment Placebo Control/Comparison Outcome Measures Effect Size Corticosteroid vs NSAID 200/120 Subacromial steroid injection (1mL of 80mg/mL triamcinolone hexacetomide) (nz45) NSAID (50mg diclofenac thrice a day for 28d) Pain improvement (VAS) (4e6wk) SMD 0.18 (95% CIZ 0.54 to 0.18) Improvement in SMD 0.03 (95% CIZ 0.39 to 0.44) function (4wk) (nz60) (nz60) Rotator cuff tendonitis (nz45) (nz45) Improvement in range of SMD 0.17 (95% CIZ 0.53 to 0.19) abduction (4e6wk) (nz60) (nz60) (nz15) (nz15) Global assessment score SMD 0.03 (95% CIZ 0.75 to 0.68) (sum of pain and motion deficit) (0e9 points) (6wk) Corticosteroid plus NSAID vs placebo Berry et al Intraarticular steroid injection plus Placebo (nz12) Pain (VAS) (4wk) WMD 7.20 (95% CIZ to 28.43) Rotator cuff lesion NSAID (40mg methylprednisolone plus 50mg diclofenac thrice a day (nz12) Range of abduction (4wk) WMD (95% CIZ to 5.21) ifo placebo Success rate* RR 0.56 (95% CIZ0.26e1.17) Corticosteroid plus NSAID vs NSAID Petri et al Subacromial steroid injection plus NSAID (500mg naproxen Pain improvement WMD 0.19 (95% CIZ 0.73 to 1.11) Rotator cuff tendonitis NSAID (1mL of 40mg/mL twice a day for 30d) (0e5 scale) (4wk) triamcinolone plus 500mg naproxen (nz25) Improvement in WMD 0.10 (95% CIZ 0.96 to 0.76) twice a day for 30d) (nz25) function (4wk) Improvement in range WMD 0.56 (95% CIZ 0.15 to 1.27) of abduction (4wk) Corticosteroid plus NSAID vs acupuncture Berry et al Steroid (400mg tolmetin sodium thrice Acupuncture once a week Pain (VAS) MD 7.50 (95% CIZ to 27.47) Rotator cuff lesion a day plus anterior injection of with moxibustion (nz12) (postintervention) 40mg methylprednisolone) (nz12) Range of abduction MD 2.90 (95% CIZ to 32.63) (postintervention) Success rate (short term) RR 0.83 (95% CIZ0.35e2.00) Abbreviations: ADL, activities of daily living; ifo, in favor of; MD, mean difference; RR, relative risk; SMD, standardized mean difference; VAS, visual analog scale; WMD, weighted mean difference. * Success rate Z success or failure of the treatment at the end of 4wk defined as the need for a steroid injection. Pharmaceutics for subacromial impingement syndrome 969

10 Appendix 2 Data extractiondrecent RCTs Author Treatment Placebo Control Comparison Corticosteroid injections Corticosteroid vs placebo Akgun et al 9 Group 1: 2 injections, Group 3: only SIS 10-d interval: 10cc of 1% lignocaine (nz16) lignocaine þ 40mg methylprednisolone for both injections (nz16) SIS Group 1: 2 injections, 10-d interval: 10cc of 1% lignocaine þ 40 mg methylprednisolone for both injections (nz16) Group 2: first injection 10cc of 1% lignocaine þ40mg methylprednisolone, second injection only lignocaine (nz16) Outcome Measures and FU Time ResultsdStatistical ResultsdWords Rest pain NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 3: from at baseline to at 1-mo FU NS (no P value given) At 3-mo FU: Group 1: vs group 3: Activity pain NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 3: from at baseline to at 1-mo FU NS (no P value given) At 3-mo FU: Group 1: vs group 3: Pain disturbing sleep Significant (no P value given) NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 3: from at baseline to at 1-mo FU ifo group 1 At 3-mo FU: Group 1: vs group 3: Constant score NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 3: from at baseline to at 1-mo FU NS (no P value given) At 3-mo FU: Group 1: vs group 3: Rest pain NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 2: from at baseline to at 1-mo FU NS (no P value given) At 3-mo FU: Group 1: vs group 2: Activity pain NS (no P value given) Group 1: from (continued on next page) 970 R. van der Sande et al

11 Appendix 2 (continued) Author Treatment Placebo Control Comparison Abbreviations: FU, follow-up; ifo, in favor of; NS, not significant; S, significant. Outcome Measures and FU Time ResultsdStatistical ResultsdWords Pain disturbing sleep NS (no P value given) S (no P value given) NS (no P value given) at baseline to at 1-mo FU vs group 2: from at baseline to at 1-mo FU At 3-mo FU: Group 1: vs group 2: Group 1: from at baseline to at 1-mo FU vs group 2: from at baseline to at 1-mo FU ifo group 1 At 3-mo FU: Group 1: vs group 2: Constant score NS (no P value given) Group 1: from at baseline to at 1-mo FU vs group 2: from at baseline to at 1-mo FU NS (no P value given) At 3-mo FU: Group 1: vs group 2: Pharmaceutics for subacromial impingement syndrome 971

12 Appendix 3 Data extractiondadditional RCTs Author Treatment Placebo Control/Comparison Friis et al 11 Shoulder tendonitis Paoloni et al 12 Chronic supraspinatus tendinopathy Analgesics Ibuprofen tablet vs sustained-released ibuprofen Conventional tablets ibuprofen (CTI) (600mg QID) plus 1 local injection with corticosteroid at day 0 (nz74) GTN transdermal vs placebo GTN patch plus 24 tablets paracetamol 500mg for use with possible headache plus tendon rehabilitation program (nz26 [28 shoulders]) Placebo patch plus 24 tablets paracetamol 500mg for use with possible headache plus tendon rehabilitation program (nz27 [29 shoulders]) Sustained-release tablets ibuprofen (SRTI) (1200mg BID) plus 1 local injection with corticosteroid at day 0 (nz73) Outcome Measures and FU Time Patients global assessment 5-point scale (3wk) Resultsd Statistical PZ.04 ResultsdWords SRTI vs CTI ifo CTI () Complete pain relief PZ.02 SRTI 7% (no 95% CI given) vs (3wk) CTI 21% (no 95% CI given) Improvement: better or complete relief (3wk).2<P<.3 SRTI 67% (95% CIZ55%e77%) vs CTI 77% (95% CIZ65%e86%) Improvement: (doctor s assessment) (3wk) Decrease in pain at rest (0e4) Decrease in pain with activity (0e4) Decrease in pain at night (0e4) Patient-related pain scores (at rest, at night, with activity) No P value given PZ.04 PZ.03 P>.05 PZ.01 PZ.03 PZ.01 No P value given SRTI 73% (no 95% CI given) vs CTI 79% (no 95% CI given) 12wk: GTN vs placebo ifo GTN, 24wk: GTN vs placebo ifo GTN, 12wk: GTN vs placebo, no exact data given 24wk: GTN vs placebo ifo of GTN, 12wk: GTN vs placebo ifo of GTN, 24wk: GTN vs placebo ifo GTN, GTN: from baseline to 24wk, a reduction of 65% (64%e67%) vs placebo: from baseline to 24wk, a reduction of 30% (27%e33%) (continued on next page) 972 R. van der Sande et al

13 Appendix 3 (continued ) Author Treatment Placebo Control/Comparison Outcome Measures and FU Time ROM Increase in supraspinatus force Increase in external rotation force Increase in internal rotation force Increase in subscapularis force Increase in adduction force Increase in dynanometermeasured shoulder force Impingement signs Resultsd Statistical No P value given PZ.001 PZ.001 PZ.01 PZ.01 PZ.01 PZ.01 PZ.02 PZ.02 PZ.003 PZ.003 No P value given No P value given ResultsdWords GTN: from baseline to 24wk, an increase of 24% (17%e31%) vs placebo: from baseline to 24wk, an increase of 8% (4%e13%) 12wk: GTN vs placebo ifo of GTN, 24wk: GTN vs placebo ifo of GTN, 12wk: GTN vs placebo ifo GTN, 24wk: GTN vs placebo ifo GTN, 12wk: GTN vs placebo ifo GTN, 24wk: GTN vs placebo ifo of GTN, 12wk: GTN vs placebo ifo GTN, 24wk: GTN vs placebo ifo GTN, 12wk: GTN vs placebo ifo GTN, 24wk: GTN vs placebo ifo GTN, GTN: from baseline to 24wk, an increase of 29% (10%e61%) vs placebo: from baseline to 24wk, an increase of 12% (5%e 32%) GTN: from baseline to 24wk, a decrease of 76% (74%e78%) vs placebo: from baseline to 24wk, a decrease of 43% (40%e47%) (continued on next page) Pharmaceutics for subacromial impingement syndrome 973

14 Appendix 3 (continued ) Author Treatment Placebo Control/Comparison Itzkowitch et al 8 Rotator cuff tendinitis Cacchio et al 10 Calcific tendinitis Other injections Tenoxicam injection vs placebo injection Tenoxicam injection 20mg/2mL once per week for up to 4wk (nz38) Disodium EDTA plus US vs placebo Study-group: injection with 1mL disodium EDTA, 1mL of 1% procaine and 3mL injectable H 2 O (once a week for 3wk) plus 15-min pulsed-mode US (1MHz) using 15% solution disodium EDTA gel (5 times a week for 3wk) (nz40) Placebo injection (nz39) Sham injection plus sham US (nz40) Outcome Measures and FU Time Resultsd Statistical ResultsdWords Asymptomatic at 24wk PZ wk: GTN 64% vs placebo 24%, no range given 24wk: mean estimated effect size GTN: 0.26 (95% CIZ0.19e0.32) Pain at rest (VAS) (2e4wk) Pain on active movement (VAS) (2-4wk) Active abduction (degrees) (2e4wk) NS P<.001 NS P<.001 NS P<.001 Baseline: Tenoxicam: vs placebo: Last available score: Tenoxicam: vs placebo: Baseline: Tenoxicam: vs placebo: Last available score: Tenoxicam: vs placebo: Baseline: Tenoxicam: vs placebo: Last available score: Tenoxicam: vs placebo: Study group vs placebo Pain (VAS) P<.43 Baseline: vs Total Constant and Murley score P<.01 P<.01 P<.53 1wk: vs y: vs Study group vs placebo Baseline: vs (continued on next page) 974 R. van der Sande et al

15 Pharmaceutics for subacromial impingement syndrome 975 Appendix 3 (continued ) Resultsd Statistical ResultsdWords Outcome Measures and FU Time Author Treatment Placebo Control/Comparison P<.01 1wk: vs P<.01 1y: vs Study group vs placebo PZ.08 Baseline: vs P<.01 After treatment: vs Size of calcification (mm) (radiography) Abbreviations: BID, 2 times a day; CTI, conventional tablets ibuprofen; FU, follow-up; ifo, in favor of; NS, not significant; S, significant; SRTI, sustained-release tablets ibuprofen; US, ultrasound; VAS, visual analog scale (range, 0e10); QID, 4 times a day. References 1. Huisstede BM, Miedema HS, Verhagen AP, Koes BW, Verhaar JA. Multidisciplinary consensus on the terminology and classification of complaints of the arm, neck and/or shoulder. Occup Environ Med 2007;64: Huisstede BM, Wijnhoven HA, Bierma-Zeinstra SM, Koes BW, Verhaar JA, Picavet S. Prevalence and characteristics of complaints of the arm, neck, and/or shoulder (CANS) in the open population. Clin J Pain 2008;24: van Rijn RM, Huisstede BM, Koes BW, Burdorf A. Associations between work-related factors and specific disorders of the shoulder e a systematic review of the literature. Scand J Work Environ Health 2010;36: van der Windt DA, Koes BW, de Jong BA, Bouter LM. Shoulder disorders in general practice: incidence, patient characteristics, and management. Ann Rheum Dis 1995;54: Koester MC, George MS, Kuhn JE. Shoulder impingement syndrome. Am J Med 2005;118: Nederlands Huisartsen Genootschap (NHG) Standaard Schouderklachten: the Netherlands; Rhon DI, Boyles RE, Cleland JA, Brown DL. A manual physical therapy approach versus subacromial corticosteroid injection for treatment of shoulder impingement syndrome: a protocol for a randomised clinical trial. BMJ Open Itzkowitch D, Ginsberg F, Leon M, Bernard V, Appelboom T. Periarticular injection of tenoxicam for painful shoulders: a double-blind, placebo controlled trial. Clin Rheumatol 1996;15: Akgun K, Birtane M, Akarirmak U. Is local subacromial corticosteroid injection beneficial in subacromial impingement syndrome? Clin Rheumatol 2004;23: Cacchio A, De Blasis E, Desiati P, Spacca G, Santilli V, De Paulis F. Effectiveness of treatment of calcific tendinitis of the shoulder by disodium EDTA. Arthritis Rheum 2009;61: Friis J, Jarner D, Toft B, et al. Comparison of two ibuprofen formulations in the treatment of shoulder tendonitis. Clin Rheumatol 1992; 11: Paoloni JA, Appleyard RC, Nelson J, Murrell GA. Topical glyceryl trinitrate application in the treatment of chronic supraspinatus tendinopathy: a randomized, double-blinded, placebo-controlled clinical trial. Am J Sports Med 2005;33: Furlan AD, Pennick V, Bombardier C, van Tulder M updated method guidelines for systematic reviews in the Cochrane Back Review Group. Spine (Phila Pa 1976) 2009;34: van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group. Spine (Phila Pa 1976) 2003;28: Green S, Buchbinder R, Hetrick S. Physiotherapy interventions for shoulder pain. Cochrane Database Syst Rev 2003;(2):CD Green S, Buchbinder R, Hetrick S. Acupuncture for shoulder pain. Cochrane Database Syst Rev 2005;(2):CD Buchbinder R, Green S, Youd JM. Corticosteroid injections for shoulder pain. Cochrane Database Syst Rev 2003;(1):CD Berry H, Fernandes L, Bloom B, Clark RJ, Hamilton EB. Clinical study comparing acupuncture, physiotherapy, injection and oral antiinflammatory therapy in shoulder-cuff lesions. Curr Med Res Opin 1980;7: England S, Farrell AJ, Coppock JS, Struthers G, Bacon PA. Low power laser therapy of shoulder tendonitis. Scand J Rheumatol 1989; 18: Adebajo AO, Nash P, Hazleman BL. A prospective double blind dummy placebo controlled study comparing triamcinolone hexacetonide injection with oral diclofenac 50 mg TDS in patients with rotator cuff tendinitis. J Rheumatol 1990;17: Blair B, Rokito AS, Cuomo F, Jarolem K, Zuckerman JD. Efficacy of injections of corticosteroids for subacromial impingement syndrome. J Bone Joint Surg Am 1996;78:

16 976 R. van der Sande et al 22. Petri M, Dobrow R, Neiman R, Whiting-O Keefe Q, Seaman WE. Randomized, double-blind, placebo-controlled study of the treatment of the painful shoulder. Arthritis Rheum 1987;30: Vecchio PC, Hazleman BL, King RH. A double-blind trial comparing subacromial methylprednisolone and lignocaine in acute rotator cuff tendinitis. Br J Rheumatol 1993;32: Strobel G. [Long-term therapeutic effect of different intra-articular injection treatments of the painful shouldereeffect on pain, mobility and work capacity] [German]. Rehabilitation (Stuttg) 1996;35: Withrington RH, Girgis FL, Seifert MH. A placebo-controlled trial of steroid injections in the treatment of supraspinatus tendonitis. Scand J Rheumatol 1985;14: Vecchio P, Cave M, King V, Adebajo AO, Smith M, Hazleman BL. A double-blind study of the effectiveness of low level laser treatment of rotator cuff tendinitis. Br J Rheumatol 1993;32: Ebenbichler GR, Erdogmus CB, Resch KL, et al. Ultrasound therapy for calcific tendinitis of the shoulder. N Eng J Med 1999;340: Nykanen M. Pulsed ultrasound treatment of the painful shoulder: a randomized, doubleeblind, placeboecontrolled study. Scand J Rehabil Med 1995;27: Saunders L. The efficacy of low level laser therapy in supraspinatus tendinitis. Clin Rehabil 1995;9: Conroy DE, Hayes KW. The effect of mobilization as a component of comprehensive treatment for primary shoulder impingement syndrome. J Orthop Sports Phys Ther 1998;28: Dal Conte G, Rivoltini P, Combi F. [Trattamento della periartrite calcarea di spalla con campi magnetici pulsanti: studio controllato] [Italian]. La Riabilitazione 1990;23: Bang M, Deyle G. Comparison of supervised exercise with and without manual physical therapy for patients with shoulder impingement syndrome. J Orthop Sports Phys Ther 2000;30: Brox J, Staff P, Ljunggren A, Brevik J. Arthroscopic surgery compared with supervised exercises in patients with rotator cuff disease. BMJ 1993;307: Perron M, Malouin F. Acetic acid iontophoresis and ultrasound for the treatment of calcifying tendinitis of the shoulder: a randomized control trial. Arch Phys Med Rehabil 1997;78: White RH, Paull DM, Fleming KW. Rotator cuff tendinitis: comparison of subacromial injection of a long acting corticosteroid versus oral indomethacin therapy. J Rheumatol 1986;13: Huisstede BM, Hoogvliet P, Randsdorp MS, Glerum S, van Middelkoop M, Koes BW. Carpal tunnel syndrome, part I: effectiveness of nonsurgical treatmentsea systematic review. Arch Phys Med Rehabil 2010;91: Huisstede BM, van Middelkoop M, Randsdorp MS, Glerum S, Koes BW. Effectiveness of interventions of specific complaints of the arm, neck, and/or shoulder: 3 musculoskeletal disorders of the hand. An update. Arch Phys Med Rehabil 2010;91: Holleman AF. In organic chemistry. San Diego: Academic Press;