Botulinum Toxin For Shoulder Pain (Review) : Singh JA, Fitzgerald PM

Botulinum toxin for shoulder pain (Review)
Singh JA, Fitzgerald PM
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2011, Issue 1 http://www.thecochranelibrary.com
Botulinum toxin for shoulder pain (Review) Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUMMARY OF FINDINGS FOR THE MAIN COMPARISON . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 1 Pain (0-10 cm visual analog scale or verbal rating scale). . . . . . . . . . . . . . . . . . . . Analysis 1.2. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 2 Spasticity on Modied Ashworth Scale (0-5 scale; higher=worse spasticity). . . . . . . . . . . . . . Analysis 1.3. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 3 Passive shoulder exion (0-180; higher=better function). . . . . . . . . . . . . . . . . . . . Analysis 1.4. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 4 Passive shoulder abduction (0-180; higher=better function). . . . . . . . . . . . . . . . . . . Analysis 1.5. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 5 Shoulder external rotation (0-90; higher=better function). . . . . . . . . . . . . . . . . . . Analysis 1.6. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 6 Number of adverse events. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.1. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 1 Change in pain on 0-10 Numeric Rating Scale (higher=more pain). . . . . . . . . Analysis 2.2. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 2 Change in physician global rating scale (0-4; 4=maximum improvement). . . . . . . Analysis 2.3. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 3 Change in passive shoulder exion(0-180; higher=better function). . . . . . . . . . Analysis 2.4. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 4 Change in passive shoulder abduction (0-180; higher=better function). . . . . . . . Analysis 2.5. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 5 Change in passive shoulder external rotation (0-90; higher=better function). . . . . . Analysis 2.6. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 6 Change in passive shoulder internal rotation (0-90; higher=better function). . . . . . Analysis 2.7. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 7 Change in Muscle spasticity on Modied Ashworth Scale. . . . . . . . . . . . . Analysis 3.1. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 1 Pain on 0-10 Visual Analog Scale (0-10 cm; higher=worse pain). . . . . . . . . . . . . Analysis 3.2. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 2 SPADI Disability Subscale Score (0-100; higher=worse function). . . . . . . . . . . . . Analysis 3.3. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 3 SPADI Pain Subscale Score (0-100; higher=worse pain). . . . . . . . . . . . . . . . Analysis 3.4. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 4 Active Shoulder exion (0-180; higher=better function). . . . . . . . . . . . . . . . Analysis 3.5. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 5 Active shoulder abduction (0-180; higher=better function). . . . . . . . . . . . . . .
1 1 2 2 5 5 6 8 9 11 12 13 13 16 28 30 31 31 32 33 34 34 35 35 36 36 37 37 38 38 39 39 40
i
Analysis 3.6. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 6 McGill Total Pain Score (0-45; higher=worse pain). . . . . . . . . . . . . . . . . Analysis 3.7. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 7 McGill Sensory Dimension Pain Score (0-33; higher=worse). . . . . . . . . . . . . . Analysis 3.8. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 8 McGill Affective Dimension Pain Score (0-12; higher=worse pain). . . . . . . . . . . . Analysis 3.9. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 9 Serious adverse events. . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.10. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 10 Number of patients with 1 or more serious adverse events. . . . . . . . . . . . . . . Analysis 3.11. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 11 Death. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
40 41 41 42 42 43 43 46 46 47 47 47 48 48 48
ii
[Intervention Review]
Botulinum toxin for shoulder pain

Jasvinder A Singh1 , Patrick M Fitzgerald2
1 Department
of Medicine, Birmingham VA Medical Center, Birmingham, USA. 2 Medicine, Minneapolis VA Medical Center, Minneapolis, Minnesota, USA Contact address: Jasvinder A Singh, Department of Medicine, Birmingham VA Medical Center, Faculty Ofce Tower 805B, 510 20th Street South, Birmingham, AL 35294, USA. jasvinder.md@gmail.com. Editorial group: Cochrane Musculoskeletal Group. Publication status and date: Edited (no change to conclusions), published in Issue 1, 2011. Review content assessed as up-to-date: 21 January 2010. Citation: Singh JA, Fitzgerald PM. Botulinum toxin for shoulder pain. Cochrane Database of Systematic Reviews 2010, Issue 9. Art. No.: CD008271. DOI: 10.1002/14651858.CD008271.pub2. Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT Background Recent evidence suggests an anti-nociceptive effect of botulinum toxin. Objectives To compare the efcacy and safety of botulinum toxin in comparison to placebo or other treatment options for shoulder pain. Search strategy We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library), Ovid MEDLINE, CINAHL (via EBSCOhost), Ovid SPORTDiscus, EMBASE and Science Citation Index. Selection criteria Randomized controlled trials (RCTs) comparing botulinum toxin with placebo or active treatment in people with shoulder pain were included. Data collection and analysis For continuous measures we calculated mean difference (MD), and for categorical measures risk ratio (RR) (with 95% condence interval (CI)). Main results Six RCTs with 164 patients were included. Five RCTs in participants with post-stroke shoulder pain indicated that compared with placebo, a single intramuscular injection of botulinum toxin A signicantly reduced pain at three to six months post-injection (MD 1.2 points, 95% CI -2.4 to -0.07; 0 to 10 point scale) but not at one month (MD -1.1 points, 95% CI -2.9 to 0.7). Shoulder external rotation was increased at one month (MD 9.8, 95% CI 0.2 to 19.4) but not at three to six months. Shoulder abduction, external rotation or spasticity did not differ between groups, nor did the number of adverse events (RR 1.46, 95% CI 0.6 to 24.3). One RCT in arthritis-related shoulder pain indicated that botulinum toxin reduced pain severity (MD -2.0, 95% CI -3.7 to -0.3; 10 point scale) and shoulder disability with a reduction in Shoulder Pain and Disability Index score (MD -13.4, 95% CI -24.9 to -1.9; 100 point scale) when compared with placebo. Shoulder abduction was improved (MD 13.8 degrees, 95% CI 3.2 to 44.0). Serious adverse events did not differ between groups (RR 0.35, 95% CI: 0.11, 1.12).
Botulinum toxin for shoulder pain (Review) Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 1
Authors conclusions The results should be interpreted with caution due to few studies with small sample sizes and high risk of bias. Botulinum toxin A injections seem to reduce pain severity and improve shoulder function and range of motion when compared with placebo in patients with shoulder pain due to spastic hemiplegia or arthritis. It is unclear if the benet of pain relief in post-stroke shoulder pain at three to six months but not at one month is due to limitations of the evidence, which includes small sample sizes with imprecise estimates, or a delayed onset of action. More studies with safety data are needed.
PLAIN LANGUAGE SUMMARY Botulinum toxin for shoulder pain This summary of a Cochrane review presents what we know from research about the effects of botulinum toxin on shoulder pain. The review shows that in people with shoulder pain and muscle spasms, botulinum toxin: -may improve pain; -may lead to slightly more adverse events; -effects on physical function and disability were not reported. We often do not have precise information about side effects and complications. This is particularly true for rare but serious side effects of botulinum toxin injections such as difculty swallowing, muscle soreness, respiratory infections, or a rash. What is botulinum toxin and what is shoulder pain? Shoulder pain is a common disorder affecting adults around the world. It can be caused by several conditions such as rotator cuff tendinitis, adhesive capsulitis, osteoarthritis of the shoulder and acromio-clavicular joint disease. Although shoulder pain can often go away on its own, sometimes it can result in chronic pain and disability. Botulinum toxin is produced by the bacteria, Clostridium botulinum. In people with pain and muscle spasms in their shoulder, it may work by blocking muscular nerve signals to the brain and consequently decrease the feeling of pain in the shoulder. Botulinum toxin is given as an injection in the doctors ofce or clinic. Best estimate of what happens to people with shoulder pain and muscle spasms after 6 to 24 weeks Pain (after 6 to 12 weeks) -people who had botulinum toxin injections rated their pain to be 4 on a scale of 0 to 10; -people who had a fake injection (placebo) rated their pain to be 5 on a scale of 0 to 10; -people who had botulinum toxin injections rated their pain to be just over 1 point better after treatment (absolute difference: 12%). Side effects (after 24 weeks) -33 people out of 100 who had botulinum toxin injections had side effects; -24 people out of 100 who had a fake injection (placebo) had side effects; -9 more people out of 100 had side effects with botulinum toxin injections (9% absolute difference).
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Intramuscular botulinum toxin versus placebo for shoulder pain in patients with shoulder spasticity: Patient or population: patients with shoulder pain Settings: Inpatients and outpatients Intervention: Intramuscular Botulinum toxin versus placebo in patients with shoulder spasticity Outcomes Illustrative comparative risks* (95% CI) Relative effect (95% CI) No of Participants (studies) Quality of the evidence (GRADE) Comments
Assumed risk Control
Corresponding risk Intramuscular Botulinum toxin versus placebo in patients with shoulder spasticity The mean Pain (0-10 cm visual analog scale)- Pain at 6-12 weeks in the intervention groups was 1.22 lower (2.37 to 0.07 lower) 76 (3) low1,2 Absolute risk difference = 12% (95% CI 1% to 24%); relative percent change = 25% (95% CI 1% to 49%); NNT benefit = 4 (95% CI 2 to 102)
Pain (0-10 cm visual analog scale)- Pain at 612 weeks Visual Analog or Verbal Rating Scale. Scale from: 0 to 10. (follow-up: 6-12 weeks)
The mean pain (0-10 cm visual analog scale)- pain at 6-12 weeks in the control groups was 4.8
Number of Adverse 235 per 1000 Events (AEs) (follow-up: 4-24 weeks)
343 per 1000 (150 to 790)
RR 1.46 (0.64 to 3.36)
65 (3)
low1,2
Absolute risk difference = 9% (95% CI -22% to 39%); relative percent change = 46% (95% CI 36% to 226%); NNT harm = not estimable3 see comment4
Disability/function - not See comment reported
See comment
Not estimable
See comment
WIthdrawals due to Ad- See comment verse Events - not reported Serious Adverse Events See comment - not reported
See comment
Not estimable
See comment
see comment4
See comment
Not estimable
See comment
see comment4
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio; GRADE Working Group grades of evidance High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.
1 2 3 4
Allocation sequence generation and allocation concealment was not described in some studies Sample sizes were small for all included studies, making estimates liable to error 95% CI includes both positive and negative numbers, therefore NNT harm is not estimable Not reported in any study
BACKGROUND
emerging evidence on the use of botulinum toxin for pain relief in musculoskeletal disorders (Ranoux 2008).
Description of the condition

Shoulder pain is one of the most common musculoskeletal disorders affecting the adult population, worldwide. An episode of persistent shoulder pain lasting longer than one month affects over 5% of adult Americans each year (Bjelle 1989; Miller 1993). Symptomatic shoulder disorders were seen in 21% of the population older than 70 years in a community survey (Chard 1991). Other epidemiologic studies have an estimated prevalence of 7% to 25% in Western general populations (Bergenudd 1988; Bjelle 1989; Breivik 2006; Chakravarty 1990). Although many episodes of shoulder pain are self-limiting, a signicant proportion result in chronic pain and disability. Shoulder pain has a signicant impact on the ability to function in daily life and to enjoy an optimal quality of life (Chard 1991; Croft 1996; Pope 1997). Shoulder pain is associated with shoulder-related disability in more than half of patients reporting pain (Nygren 1995); as well as increased healthcare utilization (Wofford 2005) and disability claims. The common causes of shoulder pain include rotator cuff tendinitis, adhesive capsulitis, osteoarthritis of the shoulder and acromioclavicular joint disease. In the elderly, shoulder pain can follow hemiplegia that results from a stroke. Shoulder pain after stroke is a common complication that is associated with reduction in quality of life (Chae 2007; Joynt 1992). In the post-stroke period, shoulder pain and limited external rotation of the arm (humerus) are common. In addition, post-stroke patients may have adhesive capsulitis, shoulder subluxation, rotator cuff injury and spasticity. Conservative treatment options for shoulder pain include analgesics and anti-inammatory medications, oral steroids ( Buchbinder 2006), local steroid injections (Buchbinder 2003), arthrographic distension (Buchbinder 2008), physiotherapy ( Green 2003; van der Windt 2003), other interventions such as acupuncture (Green 2005) and, more recently, novel interventions such as topical glyceryl trinitrate (Cumpston 2009).
How the intervention might work

Laboratory observations suggest that botulinum neurotoxin A can directly inhibit peripheral nerve nociceptor activation and sensitization by local neurotransmitter release and may indirectly reduce central sensitization in spinal cord neurons. Distal injection of botulinum neurotoxin A just below the skin in rat paws reduced spinal cord activity and paw inammation produced with induced inammatory pain (Cui 2004). In vitro studies of embryonic rat dorsal neurons conrmed that botulinum neurotoxin A inhibited release of substance P, a neuropeptide that acts as a neurotransmitter and is associated with inammatory processes and pain (Purkiss 2000). The pain sensory inhibition effect of botulinum neurotoxin is supported by several clinical observations. These are (1) improvement of neck pain before reduction in muscle spasm following injection of botulinum neurotoxin A for cervical dystonia (Jankovic 1990); (2) pain relief that typically outweighed the degree of spasm reduction in treatment of painful spasticity in the extremities (Wissel 2000); and (3) efcacy of botulinum toxin Type A in reduction of severity or frequency of migraine and tension headaches (Gobel 2001) and pain relief in patients with myofascial pain (Cherkin 1998; Porta 2000; Smith 2002) and chronic tennis elbow (Hayton 2005; Keizer 2002; Wong 2005). These observations suggest an anti-nociceptive action for botulinum toxin which seems to be independent of its well-described muscle paralysing action based on neuromuscular junction blockade.
Why it is important to do this review

Current non-surgical treatments for shoulder pain have some limitations. The effectiveness of corticosteroid injections or physiotherapy for shoulder pain demonstrate varying short-term successes while long-term efcacy requires further evaluations (Buchbinder 2003; van der Windt 2003). Pharmacological approaches such as non-steroidal anti-inammatory drugs (NSAIDs) are frequently used for treatment of pain and stiffness but hold the potential for serious gastrointestinal side effects (Shamoon 2000). Botulinum toxin may potentially be a novel treatment option for patients with refractory pain, as discussed above. A review of studies of botulinum toxin for shoulder pain provides readers with a synthesis of randomized controlled trial data for botulinum toxin use in patients with shoulder pain.
Description of the intervention

Botulinum toxin is one of the neurotoxins produced by Clostridium botulinum bacteria. Botulinum neurotoxins are zinc dependent enzymes (Wissel 2000; Setler 2002; Simpson 2002) which reversibly block neurotransmission (Setler 2002) by inhibiting release of neurotransmitters (the chemical signals) and disrupting neuronal communication. There are seven botulinum serotypes (A to G), all of which inhibit acetylcholine release at the neuromuscular junction to prevent the muscle from contracting. This property enables the use of botulinum toxin in treating spasticity associated with stroke, multiple sclerosis, Parkinsons disease and cerebral palsy. Botulinum toxin is also used cosmetically to prevent formation of wrinkles by paralyzing facial muscles. There is
OBJECTIVES
To assess the benets and safety of botulinum toxin compared with placebo or alternative treatments in adults with shoulder pain.
5
METHODS
Criteria for considering studies for this review
6. Stroke patient disability instruments, including Barthel Index (BI) (Sulter 1999), motor component of the Functional Independence Measure (M-FIM) and the Modied Ranking Scale (Kwon 2004).
Types of studies Randomized clinical trials (RCTs) and quasi-randomized clinical trials were eligible for inclusion. Types of participants Adults, older than 18 years of age, with shoulder pain due to any cause including any type of arthritis in the glenohumeral or acromio-clavicular joint (osteoarthritis, rheumatoid arthritis and other inammatory arthritides), rotator cuff tendinitis, adhesive capsulitis or post-stroke shoulder pain. Types of interventions Botulinum toxin injection by any route, including but not limited to intramuscular, subcutaneous, intradermal or intra-articular routes, compared with placebo injection or another active treatment. Types of outcome measures
Search methods for identication of studies
Electronic searches We searched the following databases for randomized or quasirandomized clinical trials. 1. Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2009, Issue 3) on Wiley InterScience (www.thecochranelibrary.com). 2. Ovid MEDLINE (1966 to August week 4, 2009). 3. CINAHL (via EBSCOhost) (1982 to September 2009). 4. Dissertation abstracts. 5. EMBASE (1980 to Week 36, 2009). 6. Science Citation Index (Web of Science) (1945 to August 2009). 7. Current Controlled Trials. The search strategies for the database searches are provided in Appendix 1. The initial search was performed in September 2009 and updated on 22 January 2010. Searching other resources We realized that the randomized controlled trials included in this review would be limited in their ability to assess safety. We searched the U.S. Food and Drug Administration (FDA) website for labels and warnings (www.fda.gov/) and other similar regulatory agency (European Medicines Agency (EMEA)) websites (www.ema.europa.eu/) to summarize warnings related to botulinum toxin injections.
Primary outcomes
Primary outcomes included the following. 1. Pain, measured on a visual analog scale (VAS), numeric rating scale (NRS) or semi-quantitative descriptive scale. 2. Adverse effects, total and serious; number of withdrawals due to adverse events; and deaths.
Secondary outcomes
Data collection and analysis

Selection of studies Two review authors (JS, PF) reviewed trials identied for potential inclusion, based on predetermined criteria (see Criteria for selecting studies). Data extraction and management Two review authors (JS, PF) extracted data from included trials: source of funding, study population, number of centers, number and experience of surgeons in each trial, duration of operation, intervention, analyses and outcomes. We extracted raw data (means and standard deviations for continuous outcomes and number of events for dichotomous outcomes)
6
1. Function or disability, measured using validated shoulderspecic instruments (e.g. Constant score, University of California and Los Angeles Shoulder scale (UCLA) or American Shoulder and Elbow Surgeons Shoulder Score (ASES), Western Ontario Osteoarthritis of the shoulder (WOOS)); general disability measures such as Health Assessment Questionnaire (HAQ) (Fries 1980). 2. Joint range of motion in exion and extension, and abduction and adduction. 3. Quality of life, assessed by validated instruments such as Short-Form 36 (SF-36). 4. Patient or physician evaluated success of treatment. 5. Radiographic progression for patients with shoulder arthritis, assessed by validated scales such as Kellgren-Lawrence grade, Larsen and Sharp scores (Larsen 1977; Sharp 1971).
for the outcomes of interest in order to assess the effects of treatment. We also extracted number of patients actually treated in each study.
Assessment of heterogeneity We assessed studies for clinical homogeneity based on disease type (arthritis versus muscle disease versus post-stroke etc.), duration of the disorder, control group and outcomes. Trials determined to be clinically heterogeneous were included for discussion but they were not included in the meta-analysis. For studies judged as clinically homogeneous, we tested statistical heterogeneity using the I2 statistic. Interpretation of I2 was carried out in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2008) using the following: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90%, substantial heterogeneity; 75% to 100%, considerable heterogeneity. We used a random-effects model as the default option independent of the I2 statistic.
Assessment of risk of bias in included studies For each included study, two review authors (JS, PF) independently assessed risk of bias against key criteria: random sequence generation; allocation concealment; blinding of participants, personnel and outcome assessors; incomplete outcome data; selective outcome reporting; and other sources of bias, in accordance with methods recommended by The Cochrane Collaboration (Higgins 2008). These criteria were labelled as: Yes (low risk of bias); No (high risk of bias); or Unclear (either a lack of information or uncertainty over the potential for bias). Divergent assessments by the two authors for risk of bias were resolved by consensus.
Assessment of reporting biases Measures of treatment effect To expedite rapid and easier updating of the review, we extracted all outcomes from the included trials. RCT results were plotted as point estimates with means and standard deviations (SDs) for continuous outcomes and risk ratios (RRs) with corresponding 95% condence intervals (CIs) for dichotomous outcomes. We used post-treatment scores for continuous outcomes or change scores. We used mean differences (MDs) for continuous measures, when possible, for ease of interpretation. We had planned to use standardized mean difference if different outcome measures of the same construct (that is different scales were used across trials or we were unable to convert data into the same scale) had been used in studies. Since we did not nd this with any of the measures, all differences were presented as MDs. For studies reporting only median scores, without measures of variation, we used the median as the mean and calculated the SD by dividing the inter-quartile range by 1.35 as recommended in the Cochrane Handbook. We planned to assess publication bias using a funnel plot if 10 or more studies were available for meta-analysis. Due to insufcient data, funnel plots could not be constructed.
Data synthesis Outcomes of clinically homogeneous studies were pooled for meta-analysis using the random-effects model.
Subgroup analysis and investigation of heterogeneity We presented results separately based on underlying disease, namely post-stroke shoulder pain versus arthritis shoulder pain; and by the comparator, placebo versus a comparator medication. We had planned to analyze studies that only described these populations presenting with shoulder pain. We planned subgroup analyses, where data were available, to determine if outcomes differed for the following factors. 1. Different routes of administration, i.e., subcutaneous versus intradermal. 2. Different underlying diagnoses (rotator cuff disease, adhesive capsulitis, osteoarthritis, rheumatoid arthritis). 3. Different types or doses of botulinum toxins. Due to limitations of available data, we were unable to perform any of the the proposed subgroup analyses.
Unit of analysis issues The unit of analysis was the patient. If analyses were presented only by shoulders, we planned to perform additional analysis considering each shoulder joint as the unit of analysis. Should eligible studies present joints as the unit of analysis (and a unit of analysis error) we plan to record the data as if from a cluster-randomized trial (Higgins 2009). All included studies except one presented their data considering patients as the unit of analysis.
Sensitivity analysis Dealing with missing data Missing data were addressed in the Characteristics of included studies table and summarized in the Risk of bias tables. We did not make any assumptions with respect to missing data, nor did we perform imputations. We planned to perform sensitivity analyses to investigate the effect of including quasi-randomized studies and to assess the impact of bias attributed to inadequate allocation concealment, but due to the lack of studies we were unable to do these analyses. We assessed the presence of small-study bias by performing a xed-effect metaanalysis.
7
Presentation of results
For the comparison of botulinum toxin and placebo in participants with shoulder spasticity due to hemiplegia, we presented the main outcomes in summary of ndings tables that contain key information including the quality of the evidence, the magnitude of effect of the interventions examined, and the sum of available data; as recommended by The Cochrane Collaboration (Schnemann 2008a). We included the following main outcomes that were selected a priori in the summary of ndings tables: pain; disability or function; adverse events, total; number of withdrawals due to adverse events; adverse events, serious (as dened in Types of outcome measures above). In addition to the absolute and relative magnitude of effect provided in the summary of ndings table, we calculated the absolute risk difference and relative per cent change for each outcome. For dichotomous outcomes, the absolute risk difference is the pooled risk difference statistic in RevMan, and the relative per cent change is the risk ratio-1. For dichotmous outcomes we calculated the number needed to treat (NNT) for statistically signicant outcomes from the control group event rate and the risk ratio using the Visual Rx NNT calculator (Cates 2004). For the continuous outcome, pain, which was pooled using the MD in RevMan, we calculated the absolute risk difference as MD/points on the scale (10 points for pain) as a percentage. Relative per cent change was calculated as the absolute benet divided by the mean of the control group; we used the mean pain from the placebo group of Marco 2007 as this trial contributed the largest weighting to the pooled MD. We calculated the NNT using the Wells calculator software available at the Cochrane Musculoskeletal Group
(CMSG) editorial ofce (www.cochranemsk.org), assuming the minimal clinical important difference for pain is 1.5 points on a 10 point scale and using the placebo group SD from Marco 2007. The summary of ndings table included a grading of the bias of the overall evidence related to each of the main outcomes using the GRADE approach (Atkins 2004a; Atkins 2004b; Schnemann 2008b).
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies. The studies are described in the sections below. Results of the search The search strategy is provided in Appendix 1. The original search (September 2009) identied 160 titles and abstracts (Figure 1). Of these 21 qualied for review of the full article. After review of these studies by two independent review authors (JS, PF), six studies were identied as tting the inclusion criteria. An updated search was performed (22 January 2010), which identied an additional seven titles and abstracts of which none qualied for full article review (Figure 1). Therefore, six studies were included in this systematic review.
Figure 1. Study selection ow chart
Included studies
Excluded studies Of the 21 studies reviewed 15 were excluded, as summarized in Figure 1 and detailed in the Excluded studies table. The two most common reasons for exclusion were that the studies were for myofacial pain (for which another Cochrane review is in progress) or were reviews.
Six studies qualied for full review (de Boer 2008; Kong 2007; Lim 2008; Marco 2007; Singh 2009; Yelnik 2007). All were doubleblind RCTs. The sample size ranged from 17 patients (Kong 2007) to 43 patients (Singh 2009). In ve studies (de Boer 2008; Kong 2007; Lim 2008; Marco 2007; Yelnik 2007) botulinum toxin was injected intramuscularly in patients with spasticity after stroke or hemiplegia. In one study (Singh 2009) botulinum toxin was injected intra-articularly in patients with refractory pain due to osteoarthritis or rheumatoid arthritis. Botulinum toxin was compared to placebo in all but one study ( Lim 2008), where it was compared to triamcinolone. In all studies, botulinum toxin type A was used and in each case a single injection was given (at single or multiple sites). The preparation was 500 units of Dysport (manufactured by Ipsen Inc, UK) in three studies (Kong 2007; Marco 2007; Yelnik 2007) and 100 units of Botox (manufactured by Allergan pharmaceuticals, Inc) in three studies (de Boer 2008; Lim 2008; Singh 2009).
Risk of bias in included studies

For most studies the risk of bias was low, except the de Boer 2008 study where it was difcult to assess the risk of bias for most criteria due to lack of details in the method section. The risk of bias tables under the section Assessment of risk of bias in included studies provide details for each criterion for each study.
Allocation Allocation sequence generation: there was no bias in sequence generation in any study except for the de Boer 2008 study, in which it was unclear.
Allocation concealment: allocation was concealed in three studies (Lim 2008; Marco 2007; Singh 2009); it was not concealed in the Kong 2007 study and was unclear in two studies (de Boer 2008; Yelnik 2007). Blinding In the methods or title, all six studies were labelled as doubleblind. However, blinding was described well in only three studies (Lim 2008; Marco 2007; Singh 2009); it was unclear in two studies due to lack of details (Kong 2007; Yelnik 2007) and was only for the outcome assessor in de Boer 2008. Incomplete outcome data Four of the six studies seemed to present all outcome data (Kong 2007; Marco 2007; Singh 2009; Yelnik 2007). Two studies did not account for either missing patients or missing data (de Boer 2008; Lim 2008).
Selective reporting The study protocol for Singh 2009 was available on clinicaltrials.gov and most outcome measures were reported in the study report. For other studies, we could not make a determination in the absence of publicly available protocols.
Other potential sources of bias Marco 2007 explicitly stated that there was no pharmaceutical funding for the study and de Boer 2008 declared no competing interest. Three studies were partially funded by the makers of botulinum toxin (Kong 2007; Lim 2008; Singh 2009) and in one study authors declared having received funding from Ipsen, Inc, a maker of botulinum toxin (Yelnik 2007).
Effects of interventions
See: Summary of ndings for the main comparison The results were described separately based on the underlying condition, that is shoulder spasticity due to hemiplegia or shoulder arthritis, and by comparison. Shoulder spasticity due to hemiplegia: effect of intramuscular botulinum toxin compared with placebo Four studies provided data (de Boer 2008; Kong 2007; Marco 2007; Yelnik 2007). The details of each outcome at 4 to 6 weeks and at 12 to 24 weeks are provided in the data and analyses section. 1. Pain: four studies provided data (de Boer 2008; Kong 2007; Marco 2007; Yelnik 2007), one provided data on change in pain
(Kong 2007) while the other three studies provided scores at end points. There was a signicantly greater reduction in pain severity in the botulinum toxin group compared to the placebo group at 12 to 24 weeks but not at 4 to 6 weeks. Surprisingly, the magnitude of the mean difference (MD) between the botulinum toxin and placebo groups at both time points was similar (MD -1.1, 95% CI -2.9 to 0.7 at 4 to 6 weeks; MD -1.2, 95% CI -2.37 to -0.07 at 12 to 24 weeks) (Analysis 1.1). The I2 statistic was 76% at 4 to 6 weeks, the time point when the difference was not signicant; and 0% at the 12 to 24 week end point when the difference was signicant. The heterogeneity at 4 to 6 weeks was due to a single study (Kong 2007) in which the magnitude of change favored placebo; removal of this study reduced the heterogeneity to 0%. Sensitivity analyses assuming a xed-effect instead of a random effects model had no effect on the 12 to 24 week estimate but made the 4 to 6 week estimate signicant with a MD of -0.9 favoring botulinum toxin over placebo (95% CI -1.8 to -0.1; P = 0.03). 2. Shoulder range of motion: evidence from one study (Marco 2007) indicated that there was a no signicant difference between groups in exion (Analysis 1.3); and evidence from three studies (Kong 2007; Marco 2007; Yelnik 2007) suggested no difference in abduction of the shoulder at 4 to 6 weeks or 12 to 24 weeks (Analysis 1.4). External rotation improved signicantly more in the botulinum toxin group compared to the placebo group in three RCTs (de Boer 2008; Marco 2007; Yelnik 2007) at 4 to 6 weeks (MD 9.79, 95% CI 0.20 to 19.39); but this difference was not signicant at 12 to 24 week follow up (Analysis 1.5). Sensitivity analyses using a xed-effect model did not change estimates or signicance for any outcome except shoulder abduction at 4 to 6 weeks, which changed from being insignicant in the randomeffects model to being signicant in the xed-effect model (MD 9.6, 95% CI 1.6 to 17.5; P = 0.02) . The I2 statistic was 36% at 4 to 6 weeks and 67% at 6 to 12 weeks for shoulder abduction. 3. Spasticity on the Modied Ashworth Scale (0-5 score; higher is more rigidity, i.e., worse function): there were no signicant differences between groups at 4 to 6 weeks or 12 to 24 weeks using data from two RCTs (Analysis 1.2). Sensitivity analyses using the xed-effect model did not change the estimates or their signicance. 4. Number of adverse events: three RCTs reported on adverse events (Kong 2007; Marco 2007; Yelnik 2007). There were no signicant differences in frequency of adverse events between botulinum toxin (9/31; 29%) and placebo groups (8/34; 24%) (Analysis 1.6). Sensitivity analyses using the xed-effect model did not change the estimates or their signicance. Summary of ndings table 1 provides a summary of the main outcomes in patients with spasticity-associated shoulder pain comparing botulinum toxin to placebo. Three of the ve prespecied outcomes were not presented in any study: serious adverse events, withdrawals due to adverse events and disability assessment. Shoulder spasticity due to hemiplegia: effect of
10
intramuscular botulinum toxin compared with placebo One study (25 participants) provided data (Lim 2008). There were no signicant differences in pain severity (Analysis 2.1), physician global rating (Analysis 2.2), shoulder range of motion (Analysis 2.3; Analysis 2.4; Analysis 2.5; Analysis 2.6) or muscle spasticity (Analysis 2.7) between botulinum toxin and placebo groups.
Shoulder arthritis due to osteoarthritis or rheumatoid arthritis: effect of intra-articular botulinum toxin compared to placebo One study provided data (Singh 2009). The efcacy data were presented by joints, not patients. The safety data were presented by patient as the unit of analysis. 1. Pain: pain severity on a VAS was signicantly lower in the botulinum toxin compared to the placebo group at one month, (MD -2.0 points on 10 point scale, 95% CI -3.71 to -0.29) ( Analysis 3.1). 2. Shoulder disability: signicantly greater improvements in patient-reported shoulder disability (i.e., lower disability score) were noted in the botulinum toxin group than the placebo group at one month follow up (MD -13.40 points on 100 point scale, 95% CI -24.93 to -1.87; higher score is worse) (Analysis 3.2). 3. Shoulder range of motion: there was signicant improvement in in active shoulder abduction (MD 23.60 , 95% CI 3.25 to 43.95; score range 0-180; higher range is better) (Analysis 3.5) one month after botulinum toxin injection compared with placebo but no differences between treatment groups in shoulder exion (Analysis 3.4). 4. McGill pain score: the McGill total score (MD -7.20 points on 0-45 scale; higher score=worse pain). scale, 95% CI -13.06 to -1.34) (Analysis 3.6), McGill sensory score (MD -4.50 points, 95% CI -8.96 to -0.04) (Analysis 3.7) and McGill affective score (MD -2.70 points, 95% CI -4.39 to -1.01) (Analysis 3.8) were signicantly improved (i.e., lower score) in the botulinum toxin group compared with the placebo group one month following treatment. 5. Adverse events: there were no signicant differences in serious events (Analysis 3.9), number of patients with one or more adverse events (Analysis 3.10) and deaths (Analysis 3.11) between the two groups.
and breathing difculties and even death. These symptoms have mostly been reported in children with cerebral palsy being treated with botulinum toxin for muscle spasticity, a use of the drugs that has not been approved by FDA. Symptoms have also been reported in adults treated both for approved and unapproved uses. In an earlier warning (2nd August 2008) the FDA commented: FDA has received reports of systemic adverse reactions including respiratory compromise and death following the use of botulinum toxins types A and B for both FDA-approved and unapproved uses. The reactions reported are suggestive of botulism, which occurs when botulinum toxin spreads in the body beyond the site where it was injected. The most serious cases had outcomes that included hospitalization and death, and occurred mostly in children treated for cerebral palsy-associated limb spasticity. The EMEA website was searched using the term botulinum toxin and documents were reviewed. In a patient information document related to NeuroBloc, botulinum toxin type B, the agency noted: The most common side effects with NeuroBloc (seen in more than 1 patient in 10) are dry mouth, headache (in patients new to treatment with botulinum toxins), dysphagia (difculty swallowing) and reactions at the site of the injection (in patients previously treated with botulinum toxins). For the full list of all side effects reported with NeuroBloc, see the Package Leaet. NeuroBloc should not be used in people who may be hypersensitive (allergic) to botulinum toxin or any of the other ingredients. NeuroBloc should not be used in patients with other neuromuscular (nerve and muscle) disorders. In another document related to Botox, Botulinum toxin type A, the agency noted: An anaphylactic reaction may occur very rarely after injection of botulinum toxin.......There have been spontaneous reports of death, sometimes associated with dysphagia, pneumonia and/or other significant disability, after treatment with botulinum toxin type A. Patients or caregivers should be advised to seek immediate medical care if swallowing, speech or respiratory disorders arise. Planned subgroup analyses by dose or route were not done due to lack of studies using other routes of administration or different doses of botulinum toxin in patients with shoulder pain.
DISCUSSION
Summary of main results

Summary of warnings from the FDA and other regulatory websites We searched the FDA website using the keyword botulinum toxin. The US FDA has issued recent warnings regarding botulinum toxin (FDA 2009): The boxed warning cautions that the effects of the botulinum toxin may spread from the area of injection to other areas of the body, causing symptoms similar to those of botulism. Those symptoms include potentially life-threatening swallowing In this study we reviewed six RCTs of botulinum toxin in patients with shoulder pain due to hemiplegia (ve RCTs) and end-stage arthritis (one RCT). The included studies were small in size and had signicant bias in an assessment of quality. In patients with shoulder pain and spasticity after hemiplegia, a single injection of botulinum toxin A was associated with statistically signicantly greater reduction in pain severity at three to six months compared to placebo. Signicantly greater improvements in shoulder
11
external rotation were noted at one month but not at three to six months. No signicant differences were noted in muscle spasticity, shoulder exion or shoulder abduction. Adverse events were reported by only three studies and were not signicantly different between placebo and botulinum toxin. In patients with end-stage arthritis, signicant improvements were noted in pain severity, shoulder-related disability and shoulder abduction in participants receiving botulinum toxin injection compared with placebo at one month follow up. Quality of life improvements were also reported to be signicantly greater in the botulinum toxin group. Again, the outcomes were at one month follow up and the sample size was small. The improvements in pain severity with botulinum toxin A was not only statistically signicant in both patient populations (hemiplegia and arthritis) but also exceeded the threshold for clinically important changes in pain severity of two points or a 33% reduction (Farrar 2000). This is an important observation since the ndings are replicated by multiple studies with a similar magnitude of pain reduction and ndings were consistent across two disease conditions that are likely to have somewhat different mechanisms of shoulder pain. Even though the pain reduction at one month did not achieve statistical signicance, the magnitude of effect (mean difference of -1.1 points from placebo on a 0 to 10 pain scale) was similar to the -1.3 point difference noted at three to six months. There was greater heterogeneity (I2 of 68%) at one month compared to that noted at three to six months (I2 of 0%). The heterogeneity at one month was due to one study (Kong 2007) that reported an effect in the opposite direction, favoring placebo. In conjunction with the absence of a statistically significant improvement in spasticity in these studies, this observation supports the notion that the anti-nociceptive action of botulinum toxin type A is likely to be independent of its muscle spasmolytic action, as discussed in detail in the Background section of this review. Most studies were small and the overall level of evidence is rated as low to mediocre. More evidence is needed from large multicenter studies to conrm these ndings.
unclear. Only a single injection of botulinum toxin A was tested in these studies, it is unknown if different dosing schedules may have had better efcacy. The duration of pain relief following a single injection is unknown with the current evidence.
Quality of the evidence

Despite the fact that most studies had allocation concealment, a comparison group (placebo or control), blinding and description of drop-outs, the risk of bias was assessed to be medium, as highlighted by the summary of ndings (SoF) table. Three studies were partially supported by pharmaceutical funding and the authors of one study had received grants from these companies in the past.
Potential biases in the review process

Our review is limited to published data and therefore we may be missing unpublished data showing absence of effect. The overall quality of evidence was low for the key outcomes in the SoF table. None of the included studies had more than 50 patients and all were single center trials. Therefore results need to be replicated by larger, multicenter studies. In addition, several studies received at least partial funding support by the pharmaceutical companies that produce and market botulinum toxin, so potential nancial conict issues remain. The strength of the evidence synthesis is limited by the quality of the included studies, some of which were were not of the highest quality.
Agreements and disagreements with other studies or reviews

The anti-nociceptive action of botulinum toxin A for shoulder pain noted in this review is very similar to the anti-nociceptive action of botulinum toxin A noted for neck pain in patients with torticollis (or wry neck) (Herskowitz 2004). Other studies of botulinum toxin have demonstrated a potential anti-nociceptive action in patients with migraine and tension headaches (Gobel 2001), myofascial pain (Cherkin 1998; Porta 2000; Smith 2002) and chronic tennis elbow (Hayton 2005; Keizer 2002; Wong 2005).
Overall completeness and applicability of evidence

Data were extracted independently by two authors so it is unlikely that there were errors in abstraction. The search was designed by a Cochrane librarian and was across six databases, therefore it is very comprehensive and was updated to January 2010. Most studies were performed at tertiary medical centers, therefore there are issues of generalizability to all patients with these conditions. In studies of patients with hemiplegia, patients had both spasticity and pain in the shoulder and the ndings may only be applicable to hemiplegic patients with both spasticity and pain. None of the studies performed a work-up for shoulder pain and therefore the exact reason for shoulder pain in these patients is
AUTHORS CONCLUSIONS Implications for practice

Shoulder pain is a common medical problem in the general population, due to tendinitis, arthritis and bursitis. It is also very common following stroke (Turner-Stokes 2002). Very few treatment options have been shown to be more effective than placebo for
12
relief of shoulder pain, in general, and post-stroke shoulder pain in particular. This review summarizes data from ve RCTs of shoulder pain associated with shoulder spasticity after stroke. The interpretation of these ndings is limited since it is based on few studies with small sample sizes and moderate to high risks of bias. A single intramuscular injection of botulinum toxin A decreased shoulder pain severity three to six months post-injection, with no signicant decrease in spasticity. The main limitation is the small sample size of these studies. We also found in one RCT of patients with end-stage shoulder arthritis pain that a single injection of botulinum toxin A improved pain, function and quality of life at one month. Large scale multicenter RCTs are needed to conrm these ndings and to assess safety. Botulinum toxin A is not approved by the FDA for use in poststroke or arthritis shoulder pain.
Implications for research

More studies of large sample size and longer follow up are needed to conrm these ndings before the clinical use of botulinum toxin can be considered for patients with shoulder pain. Further studies should test different dosages and schedules to determine the schedule or dose that is most effective and safe.
ACKNOWLEDGEMENTS
We thank Kelly McMaken for checking the extracted data, Louise Falzon for performing and updating the comprehensive literature search and Tamara Rader for help with the lay language summary.
REFERENCES
References to studies included in this review

de Boer 2008 {published data only} de Boer KS, Arwert HJ, de Groot JH, Meskers CG, Mishre AD, Arendzen JH, et al.Shoulder pain and external rotation in spastic hemiplegia do not improve by injection of botulinum toxin A into the subscapular muscle. Journal of Neurology, Neurosurgery, and Psychiatry 2008;79(5):5813. Kong 2007 {published data only} Kong KH, Neo JJ, Chua KS. A randomized controlled study of botulinum toxin A in the treatment of hemiplegic shoulder pain associated with spasticity. Clinical Rehabilitation. 2007;21(1): 2835. Lim 2008 {published data only} Lim JY, Koh JH, Paik NJ. Intramuscular botulinum toxin-A reduces hemiplegic shoulder pain: a randomized, double-blind, comparative study versus intraarticular triamcinolone acetonide. Stroke 2008;39(1):12631. Marco 2007 {published data only} Marco E, Duarte E, Vila J, Tejero M, Guillen A, Boza R, et al.Is botulinum toxin type A effective in the treatment of spastic shoulder pain in patients after stroke? A double-blind randomized clinical trial. Journal of Rehabilitation Medicine 2007;39(6):4407. Singh 2009 {published data only} Singh JA, Mahowald ML, Noorbaloochi S. Intra-articular botulinum toxin A for refractory shoulder pain: a randomized, double-blinded, placebo-controlled trial. Translational Research: The Journal Of Laboratory and Clinical Medicine 2009;153(5): 20516. Yelnik 2007 {published data only} Yelnik AP, Colle FM, Bonan IV, Vicaut E, et al.Treatment of shoulder pain in spastic hemiplegia by reducing spasticity of the subscapular muscle: a randomised, double blind, placebo controlled study of botulinum toxin A. Journal of Neurology, Neurosurgery, and Psychiatry 2007;78(8):8458.
References to studies excluded from this review

Cummings 2001 {published data only} Cummings TM, White AR. Needling therapies in the management of myofascial trigger point pain: A systematic review. Archives of Physical Medicine and Rehabilitation 2001;82(7):98692. Ferrante 2005 {published data only} Ferrante FM, Bearn L, Rothrock R, King L. Evidence against trigger point injection technique for the treatment of cervicothoracic myofascial pain with botulinum toxin type A. Anesthesiology 2005;103(2):37783. Gobel 2006 {published data only} Gobel H, Heinze A, Reichel G, Hefter H, Benecke R, Dysport Myofascial Pain Study Group, et al.Efcacy and safety of a single botulinum type A toxin complex treatment (Dysport) for the relief of upper back myofascial pain syndrome: results from a randomized double-blind placebo-controlled multicentre study. Pain 2006;125(1-2):828. Herskowitz 2004 {published data only} Herskowitz A, Herskowitz B. Treatment of neck and shoulder pain with botulinum neurotoxins. Pain Practice 2004;4 Suppl 1:2737. Jankovic 2009 {published data only} Jankovic J. Disease-oriented approach to botulinum toxin use. Toxicon 2009;54(5):61423. Lew 2002 {published data only} Lew MF. Review of the FDA-approved uses of botulinum toxins, including data suggesting efcacy in pain reduction. Clinical Journal of Pain 2002;18 Suppl(6):1426. Mahowald 2009 {published data only} Mahowald ML, Krug HE, Singh JA, Dykstra D. Intra-articular botulinum toxin type A: a new approach to treat arthritis joint pain. Toxicon 2009;54(5):65867. Ojala 2006 {published data only} Ojala T, Arokoski JP, Partanen J. The effect of small doses of botulinum toxin a on neck-shoulder myofascial pain syndrome: a
13
double-blind, randomized, and controlled crossover trial. Clinical Journal of Pain 2006;22(1):906. Paik 2006 {published data only} Paik N, Lim J, Koh J. Intramuscular botulinum toxin-A reduces hemiplegic shoulder pain: a comparison study with intra-articular triamcinolone acetonide. American Journal of Physical Medicine & Rehabilitation 2006;85(3):2523. Pedreira 2008 {published data only} Pedreira G, Cardoso E, Melo A. Botulinum toxin type A for refractory post-stroke shoulder pain. Arquivos de Neuro-Psiquiatria 2008;66(2A):2135. Qerama 2006 {published data only} Qerama E, Fuglsang-Frederiksen A, Kasch H, Bach FW, Jensen TS. A double-blind, controlled study of botulinum toxin A in chronic myofascial pain. Neurology 2006;67(2):2415. Richards 2007 {published data only} Richards BA. A double-blind, controlled study of botulinum toxin A in chronic myofascial pain. Neurology 2007;68(12):963. Rodgers 2008 {published data only} Rodgers H, Shaw L, Price C, Van Wijck F, Barnes M, Graham L, et al.Study design and methods of the BoTULS trial: A randomised controlled trial to evaluate the clinical effect and cost effectiveness of treating upper limb spasticity due to stroke with botulinum toxin type A. Trials 2008;9:59. Turner-Stokes 2002 {published data only} Turner-Stokes L, Jackson D. Shoulder pain after stroke: A review of the evidence base to inform the development of an integrated care pathway. Clinical Rehabilitation 2002;16(3):27698. van Kuijk 2002 {published data only} van Kuijk AA, Geurts ACH, Bevaart BJW, van Limbeek J. Treatment of upper extremity spasticity in stroke patients by focal neuronal or neuromuscular blockade: A systematic review of the literature. Journal of Rehabilitation Medicine 2002;34(2):5161.
Boutron 2008 Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P. Extending the CONSORT statement to randomized trials of nonpharmacologic treatment: explanation and elaboration. Annals of Internal Medicine 2008;148(4):295309. Breivik 2006 Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. European Journal of Pain 2006;10(4):287333. Buchbinder 2003 Buchbinder R, Green S, Youd JM. Corticosteroid injections for shoulder pain. Cochrane Database of Systematic Reviews 2003, Issue 1. [Art. No.: CD004016. DOI: 10.1002/14651858.CD004016] Buchbinder 2006 Buchbinder R, Green S, Youd JM, Johnston RV. Oral steroids for adhesive capsulitis. Cochrane Database of Systematic Reviews 2006, Issue 4. [Art. No.: CD006189. DOI: 10.1002/ 14651858.CD006189] Buchbinder 2008 Buchbinder R, Green S, Youd JM, Johnston RV, Cumpston M. Arthrographic distension for adhesive capsulitis (frozen shoulder). Cochrane Database of Systematic Reviews 2008, Issue 1. [Art. No.: CD007005. DOI: 10.1002/14651858.CD007005] Cates 2004 Cates C. EBM Web Site. Visual Rx Version 3 (accessed 6 October 2009) [Computer program]. Cates C. Available from: www.nntonline.net/visualrx/ 2004. Chae 2007 Chae J, Ng A, Yu DT, Kirsteins A, Elovic EP, Flanagan SR, et al.Intramuscular electrical stimulation for shoulder pain in hemiplegia: does time from stroke onset predict treatment success?. Neurorehabilitation and Neural Repair 2007;21(6):5617. Chakravarty 1990 Chakravarty KK, Webley M. Disorders of the shoulder: an often unrecognised cause of disability in elderly people. BMJ 1990;300 (6728):8489. Chard 1991 Chard MD, Hazleman R, Hazleman BL, King RH, Reiss BB. Shoulder disorders in the elderly: a community survey. Arthritis and Rheumatism 1991;34(6):7669. Cherkin 1998 Cherkin D, Wheeler K, Barlow W, Deyo R. Medication use for low back pain in primary care. Spine 1998;23:60714. Croft 1996 Croft P, Pope D, Silman A. The clinical course of shoulder pain: prospective cohort study in primary care. Primary Care Rheumatology Society Shoulder Study Group. BMJ 1996;313 (7057):6012. Cui 2004 Cui M, Khanijou S, Rubino J, Aoki R. Subcutaneous administration of botulinum toxin A reduces formalin-induced pain. Pain 2004;107:12533. Cumpston 2009 Cumpston M, Johnston RV, Wengier L, Buchbinder R. Topical glyceryl trinitrate for rotator cuff disease. Cochrane Database of
14
Additional references
Atkins 2004a Atkins D, Eccles M, Flottorp S, Guyatt GH, Henry D, Hill S, et al.Systems for grading the quality of evidence and the strength of recommendations I: critical appraisal of existing approaches The GRADE Working Group. BMC Health Services Research 2004;4(1): 38. Atkins 2004b Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al.Grading quality of evidence and strength of recommendations. BMJ 2004;328(7454):1490. Bergenudd 1988 Bergenudd H, Lindgarde F, Nilsson B, Petersson CJ. Shoulder pain in middle age. A study of prevalence and relation to occupational work load and psychosocial factors. Clinical Orthopaedics and Related Research 1988;231:2348. Bjelle 1989 Bjelle A. Epidemiology of shoulder problems. Baillires Clinical Rheumatology 1989;3(3):43751.
Systematic Reviews 2009, Issue 3. [Art. No.: CD006355. DOI: 10.1002/14651858.CD006355.pub2] Deeks 2008 Deeks J, Higgins J, Altman D. Chapter 9: Analysing data and undertaking meta-analyses, Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 (updated February 2008). The Cochrane Collaboration. Available from: www.cochranehandbook.org. 2008. Farrar 2000 Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Dening the clinically important difference in pain outcome measures. Pain 2000;88(3):28794. FDA 2009 FDA US Food, Drug Administration. FDA gives update on Botulinum toxin safety warnings; Established names of drugs changed. www.fda.gov/NewsEvents/Newsroom/ PressAnnouncements/2009/ucm175013.htm. Department of Health and Human Services, August 3, 2009. Fries 1980 Fries JF, Spitz PW, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis and Rheumatism 1980;23: 13745. Gobel 2001 Gobel H, Heinze A, Heinze-Kuhn K, Jost W. Evidence based medicine: Botulinum toxin A in migraine and tension type headache. Journal of Neurology 2001;248 Suppl:348. Green 2003 Green S, Buchbinder R, Hetrick SE. Physiotherapy interventions for shoulder pain. Cochrane Database of Systematic Reviews 2003, Issue 2. [Art. No.: CD004258. DOI: 10.1002/ 14651858.CD004258] Green 2005 Green S, Buchbinder R, Hetrick SE. Acupuncture for shoulder pain. Cochrane Database of Systematic Reviews 2005, Issue 2. [Art. No.: CD005319. DOI: 10.1002/14651858.CD005319] Hayton 2005 Hayton MJ, Santini AJ, Hughes PJ, Frostick SP, Trail IA, Stanley JK. Botulinum toxin injection in the treatment of tennis elbow. A double-blind, randomized, controlled, pilot study. The Journal of Bone and Joint Surgery. American volume 2005;87(3):5037. Higgins 2008 Higgins JP, Altman DG. Chapter 8: Assessing risk of bias in included studies. In: Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.1 (updated September 2008). The Cochrane Collaboration, 2008. Available from www.cochrane-handbook.org. The Cochrane Collaboration. Higgins 2009 Higgins JPT, Deeks JJ, Altman DG (editors). Chapter 16: Special topics in statistic. In: Higgins JPT, Green S (editors), Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 (updated September 2009). The Cochrane Collaboration, 2009. Available from www.cochrane-handbook.org.
Jankovic 1990 Jankovic J, Schwartz K. Botulinum toxin injections for cervical dystonia. Neurology 1990;40:27780. Joynt 1992 Joynt RL. The source of shoulder pain in hemiplegia. Archives of Physical Medicine and Rehabilitation 1992;73(5):40913. Keizer 2002 Keizer S, Rutten H, Pilot P, Morre H, v Os J, Verburg A. Botulinum toxin injection versus surgical treatment for tennis elbow: a randomized pilot study. Clinical Orthopaedics and Related Research 2002;401:12531. Kwon 2004 Kwon S, Hartzema AG, Duncan PW, Min-Lai, S. Disability measures in stroke. Relationship among the Barthel Index, the Functional Independence Measure, and the Modied Rankin Scale. Stroke 2004;35:918. Larsen 1977 Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference lms. Acta Radiologic Diagnosis (Stockholm) 1977;18(4):481-91. Miller 1993 Miller H. Plan and Operation of the Health and Nutrition Examination Survey: United States 1971-1973. Vital and Health Statistics. Public Health Service, US Department of Health Service 1993; Vol. Series 1. No.10. Nygren 1995 Nygren A, Berglund A, von Koch M. Neck-and-shoulder pain, an increasing problem. Strategies for using insurance material to follow trends. Scandinavian Journal of Rehabilitation Medicine 1995;32 Suppl:10712. Pope 1997 Pope DP, Croft PR, Pritchard CM, Silman AJ, Macfarlane GJ. Occupational factors related to shoulder pain and disability. Occupational and Environmental Medicine 1997;54(5):31621. Porta 2000 Porta M. A comparative trial of botulinum toxin thye A and methylprednisolone for the treatment of myofascial pain syndrome and pain from chronic muscle spasm. Pain 2000;85:1015. Purkiss 2000 Purkiss J, Welch M, Doward S, Foster K. Capsicin stimulated release of substance P from cultured dorsal root ganglion neurons: Involvement of two distinct mechanisms. Biochemical Pharmacology 2000;59:14036. Ranoux 2008 Ranoux D, Attal N, Morain F, Bouhassira D. Botulinum toxin type A induces direct analgesic effects in chronic neuropathic pain. Annals of Neurology 2008;64(3):27483. Schnemann 2008a Schnemann HJ, Oxman AD, Higgins JP, Vist GE, Glasziou P, Guyatt GH. Chapter 11: Presenting results and Summary of ndings tables. In: Higgins JP, Green S editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 500 [updated February 2008]. The Cochrane Collaboration. www.cochrane-handbook.org, 2008.
15
Schnemann 2008b Schnemann HJ, Oxman AD, Higgins JP, Vist GE, Glasziou P, Guyatt GH. Chapter 12: Interpreting results and drawing conclusions. In: Higgins JP, Green S editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 500 [updated February 2008]. The Cochrane Collaboration. www.cochranehandbook.org, 2008. Setler 2002 Setler P. Therapeutic use of botulinum toxins: background and history. The Clinical Journal of Pain 2002;18 Suppl:11924. Shamoon 2000 Shamoon M, Hochberg MC. Treatment of osteoarthritis with acetaminophen: efcacy, safety, and comparison with nonsteroidal anti-inammatory drugs. Current Rheumatology Reports 2000;2(6): 4548. Sharp 1971 Sharp JT, Lidsky MD, Collins LC, Moreland J. Methods of scoring the progression of radiologic changes in rheumatoid arthritis. Correlation of radiologic, clinical and laboratory abnormalities. Arthritis and Rheumatism 1971;14(6):706-20. Simpson 2002 Simpson KH. Individual choice of opioids and formulations: strategies to achieve the optimum for the patient. Clinical Rheumatology 2002;21 Suppl 1:58. Smith 2002 Smith HS, Audette J, Royal MA. Botulinum toxin in pain management of soft tissue syndromes. The Clinical Journal of Pain
2002;18 Suppl(6):14754.. Sulter 1999 Sulter G, Steen C, De Keyser J. Use of the Barthel index and modied Rankin scale in acute stroke trials. Stroke 1999;30(8): 153841. van der Windt 2003 van der Windt DA, Bouter LM. Physiotherapy or corticosteroid injection for shoulder pain?. Annals of the Rheumatic Diseases 2003; 62(5):3857. Wissel 2000 Wissel J, Muller J, Dressnandt J, Heinen F, Naumann M, Topka H, Poewe W. Management of spasticity associated pain with botulinum toxin A. Journal of Pain and Symptom Management 2000;20(1):449. Wofford 2005 Wofford JL, Manseld RJ, Watkins RS. Patient characteristics and clinical management of patients with shoulder pain in U.S. primary care settings: secondary data analysis of the National Ambulatory Medical Care Survey. BMC Musculoskeletal Disorders (available from www.biomedcentral.com/1471-2474/6/4) 2005;6:4. [DOI: 10.1186/1471-2474-6-4] Wong 2005 Wong SM, Hui AC, Tong PY, Poon DW, Yu E, Wong LK. Treatment of lateral epicondylitis with botulinum toxin: a randomized, double-blind, placebo-controlled trial. Annals of Internal Medicine 2005;143(11):7937. Indicates the major publication for the study
16
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

de Boer 2008 Methods We performed a randomised, double blind, placebo controlled study. Between October 2003 and November 2006, 22 adult stroke patients, aged 18 years or over, were included. Inclusion criteria were: (1) signicant shoulder pain with a minimal score of 40 mm on a pain visual analogue scale (VAS), lasting for at least 1 week (2) restricted passive external rotation of the humerus >50% relative to the unaffected arm; (3) an Ashworth Score15 at the elbow of >1. Exclusion criteria were: (1) unable to reliably ll in the VAS score (checked beforehand by means of dummy questions); (2) an International Normalised Ratio >3; (3) glenohumeral inltration in the last 4 weeks; and (4) shoulder pathology in the medical history (affected or unaffected side). 50 units of Botox dissolved in 1 ml of saline 0.9% or placebo (1 ml 0.9% of saline only, control group) at each of the two injection locations, into the subscapular muscle site. Primary outcomes: 1. pain - a vertical pain VAS score was used instead of a horizontal one to exclude bias from hemi-inattention; 2. humeral external rotation - recorded by means of an electrical goniometer xed to the wrist. The study declared no competing interest.
Participants
Interventions
Outcomes
Notes Risk of bias Item Adequate sequence generation?
Authors judgement Unclear
Description No mention of the method of sequence generation. No mention of any method of allocation concealment. Study did not address blinding method, although labelled as ..double-blind, placebo-controlled study. Insufcient reporting of exclusions to permit a judgement of Yes or No. It was stated that, 22 stroke patients
17
Allocation concealment?
Unclear
Blinding? All outcomes
Unclear
Incomplete outcome data addressed? All outcomes
Unclear
de Boer 2008
(Continued)
with....participated in a randomised, double blind, placebo controlled effect study and 21 patients completed the study without mentioning which group, treatment or control, lost a participant. Free of selective reporting? Unclear Insufcient reporting of exclusions to permit a judgement of Yes or No. Study appeared to be free of other sources of bias.
Free of other bias?
Yes
Kong 2007 Methods This was a single-centre randomized, double-blind study comparing the effects of one set of intramuscular injections of BT-A (Dysport; Ipsen, Maidenhead, UK) with placebo in patients with hemiplegic shoulder pain associated with shoulder spasticity. Patients were recruited from the outpatient clinic of a tertiary rehabilitation centre according to the following selection criteria. 1) Age 21-80 years. 2) More than three months after onset of stroke. 3) Hemiplegic shoulder pain of two weeks duration or more, with pain severity score of 4 or higher when assessed on a visual analogue scale, with 0 indicating no pain and 10, severe pain. 4) Shoulder adductor and elbow exor spasticity of at least 2 or higher on the 5 point Ashworth Scale. They were excluded if they had any of the following. 1) Previous history of shoulder pain/shoulder surgery before stroke. 2) Signicant aphasia or cognitive impairment that precluded accurate clinical assessment of visual analogue scale scores. 3) Reex sympathetic dystrophy. 4) Poststroke central pain. 5) History of neuromuscular disease (e.g. myasthenia gravis or previous BT-A injection) . Subjects were block-randomized in block sizes of four into either the BT-A or placebo group using a computer-generated sequence. In the BT-A group, 500 units of BT-A were diluted with 2.5 mL of normal saline, and 250 units of BT-A were injected into the pectoralis major and biceps brachii respectively using anatomical landmarks. In the placebo group, 2.5 mL of normal saline without BT-A was used. Primary outcome measure: 1) shoulder pain: the intensity of shoulder pain over the previous two weeks was rated on the visual analogue scale. Secondary outcomes: 1) muscle tone: shoulder adductor and elbow exor tone was evaluated using the Ashworth Scale; 2) passive range of shoulder abduction measured with a goniometer with the patient
18
Participants
Interventions
Outcomes
Kong 2007
(Continued)
seated, dened as the maximum passive range of shoulder abduction achieved till onset pain. Notes Risk of bias Item Adequate sequence generation? Authors judgement Yes Description The investigators described a random component in the sequence generation process referring to a computer random number generator. Subjects were block randomized in block sizes of four into either the BT-A or placebo group using a computer generated sequence. Participants and investigators enrolling participants could not forsee assignment because, Treatment allocation was revealed only after eligible subjects had consented to participate in the study. Some key study personnel were not blinded; the outcome assessment was blinded. Lack of blinding of others could introduce bias. No missing outcome data.
No
Unclear
Incomplete outcome data addressed? All outcomes Free of selective reporting?
Yes
Unclear
Insufcient reporting of exclusions to permit a judgement of Yes or No. The study was funded in part by Ipsen Beaufor, the makers of botulinum toxin.
Free of other bias?
No
Lim 2008 Methods The present study was a prospective randomized, double-blind, clinical trial which compared intramuscular BoNT-A and intraarticular triamcinolone acetonide (TA). Subjects were recruited from a single center, both from inpatients and outpatient clinic, between May 2004 and February 2006. Inclusion criteria were: (1) hemiplegia in an arm after stroke (maximum time interval between BoNT-A treatment and stroke 24 months and duration of pain 12 months), (2) a pain level in the hemiplegic shoulder of 6 (on a numeric scale of 0 to 10) as rated
19
Participants
Lim 2008
(Continued)
by the patient during passive ROM during at least 2 of 3 visits before enrolment, (3) limitation of passive external rotation of the hemiplegic shoulder of at least 20 compared with the unaffected side. Exclusion criteria were: (1) an intraarticular injection into the affected shoulder during the previous 6 months or use of systemic corticosteroids during the previous 4 months, (2) the presence of another obvious explanation for the pain (eg, fracture, radiculopathy) , (3) prior surgery to either the shoulder or neck region, (4) patient immobility involving connement to bed for 50% of daytime hours, (5) any medical condition that might increase the risk to the subject with exposure to BoNT-A (eg, diagnosed myasthenia gravis, Eaton-Lambert syndrome, amyotrophic lateral sclerosis, or any other disorder that might interfere with neuromuscular function) , (6) a known allergy or sensitivity to any component of the medication, (7) evidence of recent alcohol or drug abuse or severe depression, (8) the presence of an unstable medical condition or a known uncontrolled systemic disease, (9) concurrent participation in another drug or device study or participation in such a study during the 30 days before enrolment, (10) prior treatment with BoNT-A, (11) the use of aminoglycoside antibiotics, curare-like agents, or any other agent that might interfere with neuromuscular function, (12) any condition or situation that might place the subject at signicant risk. Interventions Intramuscular BoNT-A injections and intraarticular saline injection were administered to the BoNT-A group, whereas intramuscular saline injections and intraarticular TA injections were administered to the TA group. BoNT-A (Botox, Allergen) was injected into the infraspinatus, subscapularis and pectoralis muscles, which are believed to be responsible for shoulder pain using 27 gauge - monopolar needle under electromyographic guidance. One vial of Botox (100 U) was reconstituted with 4.0 mL of saline at a concentration of 25 U/mL. Primary outcome measures 1. Pain: measured using a numeric rating scale of 0-10 (0=no pain and 10=highest pain level) during passive ROM of the shoulder in 4 planes (forward exion, abduction, external and internal rotation). 2. Physician global rating scale: (range 0-4: 0=no change; 4=improvement clearly exceeding the dened therapeutic goal). 3. Passive ROM of the shoulder in four planes: using goniometry: forward exion, abduction, external rotation, and internal rotation. Secondary outcome measures 1. Arm function: measured using Fugl-Meyer scores (range 0-66, 0=no function; 66= normal function). 2. Spasticity: measured at the external rotator muscles of the shoulder using the modied Ashworth scale (range 0-5, 0=no spasticity; joint is rigid in exion or extension). 3. Adverse effects.
Outcomes
Notes
Lim 2008
(Continued)
Risk of bias Item Adequate sequence generation? Authors judgement Yes Description The investigators described a random component in the sequence generation process referring to a computer random number generator. Patients were randomized into 2 groups using a stratied randomization procedure with permuted block size of 4 using a computer that balanced ages and sexes in 2 groups before the trial. Participants and investigators enrolling participants could not forsee assignments and thus introduce selection bias. For the treatment allocation, numbered sealed envelopes were used. Blinding of participants and key study personnel ensured, and unlikely that the blinding could have been broken. Randomization codes were kept by one physician, and injection materials were prepared by this physician out of sight of patients. Syringes were sealed with plaster before injection to blind patients. Injection and evaluation were performed by separate physicians. One physician evaluated the outcome measures, and he was blinded to group allocation throughout the study. Therefore, the patients and all other people involved, except for the injecting physicians, were blinded for the type of treatment. Four of the initial 29 participants (2 from the BoNT-A group, 2 from the TA group) were lost to follow up because of admission to other hospitals (n3) or poor general condition (n1). After rst follow up (2 weeks after the injection), 3 other patients also dropped out (1 patient in week 6 and 2 during week 12). Insufcient data to permit a judgement of Yes or No.
Yes
Yes
Incomplete outcome data addressed? All outcomes
Yes
Free of selective reporting?
Unclear
21
Lim 2008
(Continued)
Free of other bias?
No
An unrestricted educational grant to cover the cost of the Botox in this study was provided by Allergan Korea.
Marco 2007 Methods A prospective, double-blind, randomized controlled clinical trial was established to determine the efcacy of BTA for spastic shoulder pain in rehabilitation patients after stroke, and as a secondary end-point, to assess its effect on spasticity, shoulder range of motion and disability. Inclusion criteria: 1. age over 18 years, of either sex; 2. having spastic hemiparesis due to CVA of 3 or more months of evolution; 3. moderate-severe spastic shoulder pain; 4. visual analogue scale (VAS) for pain equal to or greater than 40 mm; 5. spasticity of 3 or more points as determined by the Modied Ashworth Scale (MAS); 6. ability to understand and accept the trial procedures and to sign an informed consent form in accordance with national legislation. Exclusion criteria: 1. mild hemiparesis (dened as Brunnstrom stage 6); 2. previous concomitant shoulder pathology; 3. peripheral nervous system disease; 4. hypersensitivity to botulinum toxin; 5. pregnant. Patients were distributed between 2 treatment groups: Group I received treatment with TENS + BTA inltration; Group II received treatment with TENS + placebo inltration (2.5 ml of physiological saline solution). Primary outcome 1. Pain: measured on 0 to 100 mm VAS when mobilizing the shoulder. The treatment was considered to have yielded good results concerning pain when the patients VAS score was below 33.3 mm (mild pain) or when the score was less than half the initial score. Secondary outcome measures 1.Spasticity measured with the MAS (ranging from 0 to 5). 2. Shoulder range of motion expressed in degrees: exion, abduction and external rotation.
Participants
Interventions
Outcomes
Notes Risk of bias Item Authors judgement Description
22
Marco 2007
(Continued)
Adequate sequence generation?
Yes
Computer random number generator, Treatment blindingand randomization were carried out in the hospitals pharmacy service using a program to generate random numbers. Central allocation (pharmacy controlled), The injections were also prepared by the pharmacy service. At the time of BTA inltration, neither the patients, the doctors (rehabilitation specialist and electromyographist) nor the physiotherapists were aware of which group they had been allocated to. At one month follow up, clinical and electromyographic controls were conducted separately, since at that phase it might be possible to differentiate between patients injected with BTA or placebo. From then on, clinical assessment was conducted exclusively by the same trained rehabilitation specialist who had no knowledge of the electromyography measurements, in order to ensure the blind status of the trial. No missing outcomes.
Yes
Yes
Yes
Unclear
The main measure outcomes under study (intra-subjects) were analysed by intentionto-treat. The trial received no funding from the pharmaceutical industry. The Institut Municipal dInvestigacio Mdica in Barcelona provided a grant to fund this study. The funding source has no role in the study design, data collection, data analysis, data interpretations or writing of the report. The corresponding author had full access to all the data in the study and had nal responsibility for the decision to submit for publication.
Free of other bias?
Yes
23
Singh 2009 Methods This was a randomized controlled trial to evaluate the efcacy of IA-BoNT/A compared with placebo for refractory shoulder pain. Inclusion criteria: (1) shoulder pain of 6 months or more; (2) pain score of 4.5 or greater on a 0-10 numeric rating scale; (3) inadequate (efcacy, 2 months) or no response to IA corticosteroid injections; (4) radiographic changes of osteoarthritis or rheumatoid arthritis involving the glenohumeral joint, as is evident by the presence of glenohumeral joint space narrowing, osteophytes, cysts, or erosions; and (5) signicant joint line tenderness in the glenohumeral joint. Exclusion criteria: (1) shoulder joint malignancy, prosthetic shoulder joint, or planned shoulder joint surgery in the next 6 months; (2) prior botulinum toxin A injection into the joint; (3) evidence of a full-thickness rotator cuff tear on clinical examination or previous magnetic resonance imaging report; (4) clinical examination ndings indicating that acromio-clavicular disease or rotator cuff disease was the primary source of shoulder pain; (5) history of myasthenia gravis, Eaton-Lambert syndrome, amyotrophic lateral sclerosis, or other known diseases of the neuromuscular junction or motor neuron disease; (6) concomitant use of aminoglycoside or agents that interfere with neuromuscular junction transmission; (7) known allergy to lidocaine or botulinum toxin; (8) infection at or near the joint injection site or systemic infection; (9) pregnancy or lactation; (10) concurrent participation in other drug study; (11) unstable and serious medical condition or psychiatric disorder. Joints were randomized in a 1:1 ratio to experimental treatment, which consisted 2 mL 1% lidocaine and 100 units BoNT/A injection (BoNT/A group) or placebo treatment, which consisted of 2 mL 1% lidocaine and 1 mL sterile normal saline injection. Primary efcacy outcome 1. Change in pain severity at 1 month, assessed on a 0-10 cm VAS (0= no pain to 10 = worst possible pain), Secondary efcacy outcomes 1. Drop-out due to treatment failure at 1 month: per cent of patients who ended the study at 1 month because of inadequate pain relief 2. Shoulder disability and function: Shoulder Pain and Disability Index (SPADI) is a validated self-administered measure of shoulder disability and activity-related pain (0100; higher score is worse) 3. Global assessment of change: excellent - very good improvement on global scale 4. Quality of life: Short-form 36 subscales 5. Safety, adverse events 6. McGill pain score (0-45; higher score is worse)
Participants
Interventions
Outcomes
Notes
24
Singh 2009
(Continued)
Risk of bias Item Adequate sequence generation? Authors judgement Yes Description The investigators described a random component in the sequence generation. The research pharmacist prepared computerized randomization permutated blocks of 4 patients each. Central allocation (pharmacy controlled): The treatment and placebo syringes were prepared by the research pharmacist following a strict standardized protocol. Both placebo and botulinum toxin injections were transparent and could not be differentiated. Blinding of participants and key study personnnel ensured, and unlikely that the blinding was broken. No missing outcome data.
Yes
Yes
Yes
Yes
Study protocol is available and all of the studys prespecied (primary and secondary) outcomes including WOOS, and Simple Shoulder Test (SST). Supported by the Arthritis Foundation North Central Chapter grant, the VA Scholar grant from the Center for Epidemiological and Clinical Research, the Minneapolis VA Medical Center, the NIH CTSA Award 1 KL2 RR024151-01 (Mayo Clinic Center for Clinical and Translational Research), and a vial grant from Allergan Pharmaceuticals, Inc.
Free of other bias?
No
Yelnik 2007 Methods The study was conducted according to a randomised, double blind, placebo controlled, parallel group design in hemiplegic patients of either sex presenting with upper limb spasticity related to cerebral stroke.
25
Yelnik 2007
(Continued)
Participants
Inclusion criteria: 1. post-stroke spasticity, score of at least 1+ on the Modied Ashworth Scale (MAS) for the medial rotators and elbow exors, with limited range of passive motion of the shoulder: external rotation 10or <30 related to the opposite side. Exclusion criteria: 1. previous traumatic or neurological disease of the hemiplegic shoulder; 2. retraction of at least one muscle of the elbow, wrist or ngers in the hemiplegic upper limb; 3. previous treatment with botulinum toxin A or alcohol in the subscapularis muscle of the hemiplegic shoulder; 4. neuromuscular disease such as myasthenia gravis; 5. pregnant or lactating female patients. Before the intramuscular injection, the needle was used as a stimulation electrode to detect the motor point where minimal stimulation induces maximum internal rotation, and then botulinum toxin A (Dysport, 500 Speywood units) or placebo (all constituents of Dysport solvent) was injected while pulling back the needle by 1-2 cm. In addition, all patients received after treatment, on weekdays - non-standardised physical therapy for stretching, spasticity inhibition and increasing active motion when possible. 1. Pain: using a 10 point verbal scale or, for aphasic patients only (one in the placebo group, three in the toxin group), a visual analogue scale. 2. Subscapularis muscle spasticity: assessed, with the patient sitting, by a range of motion inducing a strong resistance after slow stretching in passive lateral rotation and abduction measured in degrees. 3. Change in consumption of analgesics: categorized as increased (increasing the dosage or changing to another analgesic), no change or decreased. 4. Change in shoulder range of motion. 5. Assessment of the spasticity of the whole upper limb: with the patient sitting, using the Modied Ashworth Scale for shoulder medial rotators, elbow exors, wrist exors and ngers exors.
Interventions
Outcomes
Notes Risk of bias Item Adequate sequence generation? Authors judgement Yes Description The investigators described a random component in the sequence generation. Treatment was allocated by computerized randomization. Method of allocation concealment not described. Insufcient reporting of exclusions to permit a judgement of Yes or No.
26
Unclear
Unclear
Yelnik 2007
(Continued)
Yes
No missing outcome data.
Unclear
Insufcient reporting of exclusions to permit a judgement of Yes or No. Competing interests: In the past 5 years, AY and FC have been reimbursed by Ipsen Pharma, the manufacturer of Dysport, for attending several conferences.
Free of other bias?
No
Characteristics of excluded studies [ordered by study ID]
Study Cummings 2001 Ferrante 2005 Gobel 2006 Herskowitz 2004 Jankovic 2009 Lew 2002 Mahowald 2009 Ojala 2006 Paik 2006 Pedreira 2008 Qerama 2006 Richards 2007 Rodgers 2008 Turner-Stokes 2002 van Kuijk 2002
Reason for exclusion Myofacial pain Myofacial pain Myofacial pain Review Review Review Review Myofacial pain Abstract No control group Myofacial pain Letter Protocol only, no data Review Review
27
DATA AND ANALYSES
Comparison 1. Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity
Outcome or subgroup title 1 Pain (0-10 cm visual analog scale or verbal rating scale) 1.1 Pain 4-6 weeks 1.2 Pain at 12-24 weeks 2 Spasticity on Modied Ashworth Scale (0-5 scale; higher=worse spasticity) 2.1 Spasticity Score at 4-6 weeks 2.2 Spasticity score at 12-24 weeks 3 Passive shoulder exion (0-180; higher=better function) 3.1 Passive exion at 4-6 weeks 3.2 Passive exion at 12-24 weeks 4 Passive shoulder abduction (0-180; higher=better function) 4.1 Passive Abduction at 4-6 weeks 4.2 Passive Abduction at 12-24 weeks 5 Shoulder external rotation (0-90; higher=better function) 5.1 External Rotation at 4-6 weeks 5.2 External Rotation at 12-24 weeks 6 Number of adverse events 6.1 Number of AEs at 4-6 weeks 6.2 Number of AEs at 12-24 weeks
No. of studies 4 4 3 2
No. of participants
Statistical method Mean Difference (IV, Random, 95% CI)
Effect size Subtotals only -1.12 [-2.89, 0.66] -1.22 [-2.37, -0.07] Subtotals only
86 66
Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)
2 2 1 1 1 3 3 2 3 3 2 3 1 2
45 45
Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)
-0.62 [-1.40, 0.17] -0.13 [-0.65, 0.38] Totals not selected Not estimable Not estimable Subtotals only 8.49 [-2.40, 19.39] 17.72 [-9.61, 45.04] Subtotals only 9.84 [0.20, 19.49] 11.86 [-0.61, 24.33] 1.46 [0.64, 3.36] 3.00 [0.37, 24.17] 0.84 [0.10, 7.10]
65 45
Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)
70 50 65 20 45
Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI)
28
Comparison 2. weeks
Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12
Outcome or subgroup title 1 Change in pain on 0-10 Numeric Rating Scale (higher=more pain) 2 Change in physician global rating scale (0-4; 4=maximum improvement) 3 Change in passive shoulder exion(0-180; higher=better function) 4 Change in passive shoulder abduction (0-180; higher=better function) 5 Change in passive shoulder external rotation (0-90; higher=better function) 6 Change in passive shoulder internal rotation (0-90; higher=better function) 7 Change in Muscle spasticity on Modied Ashworth Scale
No. of studies 1
No. of participants
Statistical method Mean Difference (IV, Random, 95% CI)
Effect size Totals not selected
Mean Difference (IV, Random, 95% CI)
Totals not selected
Totals not selected
Totals not selected
Totals not selected
Totals not selected
Totals not selected
Comparison 3. Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks
Outcome or subgroup title 1 Pain on 0-10 Visual Analog Scale (0-10 cm; higher=worse pain) 2 SPADI Disability Subscale Score (0-100; higher=worse function) 3 SPADI Pain Subscale Score (0-100; higher=worse pain) 4 Active Shoulder exion (0-180; higher=better function) 5 Active shoulder abduction (0-180; higher=better function) 6 McGill Total Pain Score (0-45; higher=worse pain) 7 McGill Sensory Dimension Pain Score (0-33; higher=worse)
No. of studies 1 1 1 1 1 1 1
No. of participants
Statistical method Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI)
Effect size Totals not selected Totals not selected Totals not selected Totals not selected Totals not selected Totals not selected Totals not selected
29
8 McGill Affective Dimension Pain Score (0-12; higher=worse pain) 9 Serious adverse events 10 Number of patients with 1 or more serious adverse events 11 Death
Totals not selected
1 1 1
Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI)
Totals not selected Totals not selected Totals not selected
Analysis 1.1. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 1 Pain (0-10 cm visual analog scale or verbal rating scale).
Review: Botulinum toxin for shoulder pain
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 1 Pain (0-10 cm visual analog scale or verbal rating scale)
Study or subgroup
Favours botulinum toxin N Mean(SD)
Placebo N Mean(SD)
Mean Difference IV,Random,95% CI
Weight
1 Pain 4-6 weeks de Boer 2008 Kong 2007 Marco 2007 Yelnik 2007 10 7 14 10 4.2 (2.2) -2 (0.7) 3.9 (2.5) 1.5 (1.5) 11 9 15 10 5.2 (3.2) -3 (2) 6 (2.2) 4 (2.2) 21.2 % 27.6 % 25.4 % 25.9 % -1.00 [ -3.33, 1.33 ] 1.00 [ -0.41, 2.41 ] -2.10 [ -3.82, -0.38 ] -2.50 [ -4.15, -0.85 ]
Subtotal (95% CI)

Test for overall effect: Z = 1.23 (P = 0.22) 2 Pain at 12-24 weeks de Boer 2008 Kong 2007 Marco 2007
41
45
100.0 % -1.12 [ -2.89, 0.66 ]
Heterogeneity: Tau2 = 2.45; Chi2 = 12.50, df = 3 (P = 0.01); I2 =76%
10 7 14
3.8 (1.8) -3 (2.2) 3 (2.7)
11 9 15
4.7 (2.8) -2 (1.6) 4.8 (2.9)
33.1 % 35.2 % 31.7 %
-0.90 [ -2.90, 1.10 ] -1.00 [ -2.94, 0.94 ] -1.80 [ -3.84, 0.24 ]
Subtotal (95% CI)

Test for overall effect: Z = 2.08 (P = 0.037)
31
35
100.0 % -1.22 [ -2.37, -0.07 ]
Heterogeneity: Tau2 = 0.0; Chi2 = 0.46, df = 2 (P = 0.79); I2 =0.0%
-2
-1
Favours botulinum toxin
Favours placebo
30
Analysis 1.2. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 2 Spasticity on Modied Ashworth Scale (0-5 scale; higher=worse spasticity).
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 2 Spasticity on Modied Ashworth Scale (0-5 scale; higher=worse spasticity)
Study or subgroup
Botulinum toxin A N Mean(SD)
Placebo N Mean(SD)
Weight
1 Spasticity Score at 4-6 weeks Kong 2007 Marco 2007 7 14 -1 (0.7) 2.9 (1) 9 15 0 (0.4) 3.1 (0.8) 52.1 % 47.9 % -1.00 [ -1.58, -0.42 ] -0.20 [ -0.86, 0.46 ]
Subtotal (95% CI)

Test for overall effect: Z = 1.54 (P = 0.12) 2 Spasticity score at 12-24 weeks Kong 2007 Marco 2007
21
24
100.0 % -0.62 [ -1.40, 0.17 ]
7 14
-1 (0.7) 2.9 (1.2)
9 15
-1 (0.7) 3.2 (0.9)
55.8 % 44.2 %
0.0 [ -0.69, 0.69 ] -0.30 [ -1.08, 0.48 ]
Subtotal (95% CI)

21
24
100.0 % -0.13 [ -0.65, 0.38 ]
Heterogeneity: Tau2 = 0.0; Chi2 = 0.32, df = 1 (P = 0.57); I2 =0.0%
-4
-2
Favours Botulinum toxin
Favours control
Analysis 1.3. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 3 Passive shoulder exion (0-180; higher=better function).
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 3 Passive shoulder exion (0-180; higher=better function)
Study or subgroup
Placebo N Mean(SD)
1 Passive exion at 4-6 weeks Marco 2007 2 Passive exion at 12-24 weeks Marco 2007 14 98 (31) 15 97 (19) 1.00 [ -17.87, 19.87 ] 14 95 (29) 15 92 (21) 3.00 [ -15.54, 21.54 ]
-20
-10
10
20
Favours placebo
31
Analysis 1.4. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 4 Passive shoulder abduction (0-180; higher=better function).
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 4 Passive shoulder abduction (0-180; higher=better function)
Study or subgroup
Placebo N Mean(SD)
Weight
1 Passive Abduction at 4-6 weeks Kong 2007 Marco 2007 Yelnik 2007 7 14 10 22 (32) 81 (11) 73 (7) 9 15 10 26 (31) 65 (18) 70 (19) 10.8 % 48.1 % 41.2 % -4.00 [ -35.18, 27.18 ] 16.00 [ 5.22, 26.78 ] 3.00 [ -9.55, 15.55 ]
Subtotal (95% CI)

Test for overall effect: Z = 1.53 (P = 0.13) 2 Passive Abduction at 12-24 weeks Kong 2007 Marco 2007
31
34
100.0 %
8.49 [ -2.40, 19.39 ]
7 14
48 (30) 78 (28)
9 15
15 (20) 73 (26)
45.4 % 54.6 %
33.00 [ 7.22, 58.78 ] 5.00 [ -14.70, 24.70 ]
Subtotal (95% CI)

21
24
100.0 % 17.72 [ -9.61, 45.04 ]
-100
-50
50
100
Favours control
32
Analysis 1.5. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 5 Shoulder external rotation (0-90; higher=better function).
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 5 Shoulder external rotation (0-90; higher=better function)
Study or subgroup
Placebo N Mean(SD)
Weight
1 External Rotation at 4-6 weeks de Boer 2008 Marco 2007 Yelnik 2007 10 14 10 20 (18) 21 (29) 12.5 (15) 11 15 10 15 (24) 14 (19) -2.5 (18.5) 28.6 % 28.8 % 42.7 % 5.00 [ -13.04, 23.04 ] 7.00 [ -10.98, 24.98 ] 15.00 [ 0.24, 29.76 ]
Subtotal (95% CI)
34
36
100.0 %
9.84 [ 0.20, 19.49 ]
Heterogeneity: Tau2 = 0.0; Chi2 = 0.84, df = 2 (P = 0.66); I2 =0.0% Test for overall effect: Z = 2.00 (P = 0.045) 2 External Rotation at 12-24 weeks de Boer 2008 Marco 2007 10 14 32 (14) 39 (35) 11 15 24 (21) 19 (24) 67.8 % 32.2 % 8.00 [ -7.14, 23.14 ] 20.00 [ -1.99, 41.99 ]
Subtotal (95% CI)
24
26
100.0 % 11.86 [ -0.61, 24.33 ]
Heterogeneity: Tau2 = 0.0; Chi2 = 0.78, df = 1 (P = 0.38); I2 =0.0% Test for overall effect: Z = 1.86 (P = 0.062)
-20
-10
10
20
Favours control
33
Analysis 1.6. Comparison 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity, Outcome 6 Number of adverse events.
Comparison: 1 Intramuscular botulinum toxin versus placebo in patients with shoulder spasticity Outcome: 6 Number of adverse events Study or subgroup Botulinum toxin A n/N 1 Number of AEs at 4-6 weeks Yelnik 2007 3/10 1/10 14.4 % 3.00 [ 0.37, 24.17 ] Placebo n/N Risk Ratio M-H,Random,95% CI Weight Risk Ratio M-H,Random,95% CI
Subtotal (95% CI)

Total events: 3 (Botulinum toxin A), 1 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.03 (P = 0.30) 2 Number of AEs at 12-24 weeks Kong 2007 Marco 2007
10
10
14.4 %
3.00 [ 0.37, 24.17 ]
6/7 0/14
5/9 2/15
78.1 % 7.5 %
1.54 [ 0.80, 2.98 ] 0.21 [ 0.01, 4.09 ]
Subtotal (95% CI)

Total events: 6 (Botulinum toxin A), 7 (Placebo)
21
24
85.6 %
0.84 [ 0.10, 7.10 ]
Heterogeneity: Tau2 = 1.59; Chi2 = 2.34, df = 1 (P = 0.13); I2 =57% Test for overall effect: Z = 0.16 (P = 0.87)
Total (95% CI)

Total events: 9 (Botulinum toxin A), 8 (Placebo)
31
34
100.0 %
1.46 [ 0.64, 3.36 ]
Heterogeneity: Tau2 = 0.12; Chi2 = 2.28, df = 2 (P = 0.32); I2 =12% Test for overall effect: Z = 0.90 (P = 0.37)
0.005
0.1
10
200
Favours placebo
Analysis 2.1. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 1 Change in pain on 0-10 Numeric Rating Scale (higher=more pain).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 1 Change in pain on 0-10 Numeric Rating Scale (higher=more pain)
Study or subgroup
Botulinum toxin A N Mean(SD) 4.2 (1.5)
Triamcinolone N 11 Mean(SD) 2.5 (2.7)
Mean Difference IV,Random,95% CI 1.70 [ -0.08, 3.48 ]
Lim 2008
14
-4
-2
Favours triamcinolone
34
Analysis 2.2. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 2 Change in physician global rating scale (0-4; 4=maximum improvement).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 2 Change in physician global rating scale (0-4; 4=maximum improvement)
Study or subgroup
Triamcinolone N 11 Mean(SD) 0.2 (1)
Lim 2008
14
-1
-0.5
0.5
Analysis 2.3. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 3 Change in passive shoulder exion(0-180; higher=better function).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 3 Change in passive shoulder exion(0-180; higher=better function)
Study or subgroup
Lim 2008
14
-50
-25
25
50
35
Analysis 2.4. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 4 Change in passive shoulder abduction (0-180; higher=better function).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 4 Change in passive shoulder abduction (0-180; higher=better function)
Study or subgroup
Lim 2008
14
-20
-10
10
20
Favours experimental
Favours control
Analysis 2.5. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 5 Change in passive shoulder external rotation (0-90; higher=better function).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 5 Change in passive shoulder external rotation (0-90; higher=better function)
Study or subgroup
Lim 2008
14
-20
-10
10
20
Favours control
36
Analysis 2.6. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 6 Change in passive shoulder internal rotation (0-90; higher=better function).
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 6 Change in passive shoulder internal rotation (0-90; higher=better function)
Study or subgroup
Lim 2008
14
-20
-10
10
20
Favours placebo
Analysis 2.7. Comparison 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks, Outcome 7 Change in Muscle spasticity on Modied Ashworth Scale.
Comparison: 2 Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12 weeks Outcome: 7 Change in Muscle spasticity on Modied Ashworth Scale
Study or subgroup
Mean Difference IV,Random,95% CI -0.20 [ -1.00, 0.60 ]
Lim 2008
14
-2
-1
Favours placebo
37
Analysis 3.1. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 1 Pain on 0-10 Visual Analog Scale (0-10 cm; higher=worse pain).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 1 Pain on 0-10 Visual Analog Scale (0-10 cm; higher=worse pain)
Study or subgroup
Botulinum toxin N Mean(SD) 4.4 (3)
Placebo N 22 Mean(SD) 6.4 (2.4)
Mean Difference IV,Random,95% CI -2.00 [ -3.71, -0.29 ]
Singh 2009
18
-10
-5
10
Favours placebo
Analysis 3.2. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 2 SPADI Disability Subscale Score (0-100; higher=worse function).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 2 SPADI Disability Subscale Score (0-100; higher=worse function)
Study or subgroup
Singh 2009
20
-20
-10
10
20
Favours placebo
38
Analysis 3.3. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 3 SPADI Pain Subscale Score (0-100; higher=worse pain).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 3 SPADI Pain Subscale Score (0-100; higher=worse pain)
Study or subgroup
Mean Difference IV,Random,95% CI -9.10 [ -20.33, 2.13 ]
Singh 2009
20
-50
-25
25
50
Favours placebo
Analysis 3.4. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 4 Active Shoulder exion (0-180; higher=better function).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 4 Active Shoulder exion (0-180; higher=better function)
Study or subgroup
Singh 2009
20
-50
-25
25
50
Favours placebo
39
Analysis 3.5. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 5 Active shoulder abduction (0-180; higher=better function).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 5 Active shoulder abduction (0-180; higher=better function)
Study or subgroup
Mean Difference IV,Random,95% CI 23.60 [ 3.25, 43.95 ]
Singh 2009
20
-100
-50
50
100
Favours placebo
Analysis 3.6. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 6 McGill Total Pain Score (0-45; higher=worse pain).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 6 McGill Total Pain Score (0-45; higher=worse pain)
Study or subgroup
Singh 2009
18
-10
-5
10
Favours placebo
40
Analysis 3.7. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 7 McGill Sensory Dimension Pain Score (0-33; higher=worse).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 7 McGill Sensory Dimension Pain Score (0-33; higher=worse)
Study or subgroup
Placebo N 22 Mean(SD) 14.2 (8)
Singh 2009
18
-10
-5
10
Favours placebo
Analysis 3.8. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 8 McGill Affective Dimension Pain Score (0-12; higher=worse pain).
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 8 McGill Affective Dimension Pain Score (0-12; higher=worse pain)
Study or subgroup
Botulinum toxin A N Mean(SD) 2 (2.1)
Singh 2009
18
-10
-5
10
Favours placebo
41
Analysis 3.9. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 9 Serious adverse events.
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 9 Serious adverse events
Study or subgroup
Botulinum toxin A n/N
Placebo n/N 9/22
Risk Ratio M-H,Random,95% CI
Risk Ratio M-H,Random,95% CI 0.35 [ 0.11, 1.12 ]
Singh 2009
3/21
0.1 0.2
0.5
10
Favours placebo
Analysis 3.10. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 10 Number of patients with 1 or more serious adverse events.
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 10 Number of patients with 1 or more serious adverse events
Study or subgroup
Placebo n/N 4/22
Singh 2009
3/21
0.05
0.2
20
Favours placebo
42
Analysis 3.11. Comparison 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks, Outcome 11 Death.
Comparison: 3 Intra-articular botulinum toxin versus placebo in patients with end-stage arthritis: 4 weeks Outcome: 11 Death
Study or subgroup
Placebo n/N 0/22
Singh 2009
0/21
0.01
0.1
10
100
Favours control
APPENDICES Appendix 1. Search Strategy for the review
Ovid MEDLINE(R) In-Process and Other Non-Indexed Citations and Ovid MEDLINE(R) (1950 to August Week 4, 2009) 1. Shoulder Pain/ 2. Shoulder Impingement Syndrome/ 3. Rotator Cuff/ 4. exp Bursitis/ 5. Shoulder/ 6. Shoulder Joint/ 7. exp Pain/ 8. (5 or 6) and 7 9. ((should$ or rotator cuff ) adj5 (bursitis or frozen or impinge$ or tend?nitis or pain$)).tw. 10. or/1-4,8-9 11. exp Botulinum Toxins/ 12. botulin$.tw. 13. botox.tw. 14. OnabotulinumtoxinA.tw. 15. RimabotulinumtoxinB.tw. 16. AbobotulinumtoxinA.tw. 17. BTXA.tw. 18. dyslor.tw. 19. dysport.tw. 20. lanzox.tw. 21. myobloc.tw. 22. neurobloc.tw.
23. oculinum.tw. 24. prosigne.tw. 25. vistabel.tw. 26. vistabex.tw. 27. xeomin.tw. 28. or/11-27 29. 10 and 28
EMBASE 1980 to 2009 Week 36 1. Shoulder Pain/ 2. Shoulder Impingement Syndrome/ 3. rotator cuff/ 4. Bursitis/ 5. shoulder/ 6. exp Pain/ 7. 5 and 6 8. ((should$ or rotator cuff ) adj5 (bursitis or frozen or impinge$ or tend?nitis or pain$)).tw. 9. or/1-4,7-8 10. exp botulinum toxin/ 11. botulinum toxin E/ or botulinum toxin B/ or botulinum toxin A/ or botulinum toxin F/ 12. botulin$.tw. 13. botox.tw. 14. OnabotulinumtoxinA.tw. 15. RimabotulinumtoxinB.tw. 16. AbobotulinumtoxinA.tw. 17. BTXA.tw. 18. dyslor.tw. 19. dysport.tw. 20. lanzox.tw. 21. myobloc.tw. 22. neurobloc.tw. 23. oculinum.tw. 24. prosigne.tw. 25. vistabel.tw. 26. vistabex.tw. 27. xeomin.tw. 28. or/10-27 29. 9 and 28
The Cochrane Library, Issue 3, 2009 #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 MeSH descriptor Shoulder Pain explode all trees MeSH descriptor Shoulder Impingement Syndrome explode all trees MeSH descriptor Rotator Cuff explode all trees MeSH descriptor Bursitis explode all trees MeSH descriptor Shoulder explode all trees MeSH descriptor Shoulder Joint explode all trees MeSH descriptor Pain explode all trees (( #5 OR #6 ) AND #7) ((should* or rotator cuff ) Near/5 (bursitis or frozen or impinge* or tendinitis or pain*)):ti,ab (#1 OR #2 OR #3 OR #4 OR #8 OR #9) MeSH descriptor Botulinum Toxins explode all trees
44
#12 botulin*:ti,ab #13 botox:ti,ab #14 OnabotulinumtoxinA:ti,ab #15 RimabotulinumtoxinB:ti,ab #16 AbobotulinumtoxinA:ti,ab #17 BTXA:ti,ab #18 dyslor:ti,ab #19 dysport:ti,ab #20 lanzox:ti,ab #21 myobloc:ti,ab #22 neurobloc:ti,ab #23 oculinum:ti,ab #24 prosigne:ti,ab #25 vistabel:ti,ab #26 vistabex:ti,ab #27 xeomin:ti,ab #28 (#11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR # 25 OR #26 OR #27) #29 (#10 AND #28)
CINAHL S28 S12 and S27 S27 S13 or S14 or S15 or S16 or S17 or S18 or S19 or S20 or S21 or S22 or S23 or S24 or S25 or S26 S26 ti vistabex or ab vistabex S25 ti vistabel or ab vistabel S24 ti prosigne or ab prosigne S23 ti oculinum or ab oculinum S22 ti neurobloc or ab neurobloc S21 ti lanzox or ab lanzox S20 ti myobloc or ab myobloc S19 ti dyslor or ab dyslor S18 ti dysport or ab dysport S17 ti OnabotulinumtoxinA or ab OnabotulinumtoxinA or ti RimabotulinumtoxinB or ab RimabotulinumtoxinB or ti AbobotulinumtoxinA or ab AbobotulinumtoxinA S16 ti BTXA or ab BTXA S15 ti botox or ab botox S14 ti botulin* or ab botulin* S13 (MH Botulinum Toxins) Search S12 S1 or S2 or S3 or S4 or S9 or S10 or S11 S11 TI rotator cuff N5 bursitis or AB rotator cuff N5 bursitis or TI rotator cuff N5 frozen or AB rotator cuff N5 frozen or TI rotator cuff N5 impinge* or AB rotator cuff N5 impinge* or TI rotator cuff N5 tendonitis or AB rotator cuff N5 tendonitis or TI rotator cuff N5 tendinitis or AB rotator cuff N5 tendinitis or TI rotator cuff N5 pain* or AB rotator cuff N5 pain* Search modes S10 TI should* N5 bursitis or AB should* N5 bursitis or TI should* N5 frozen or AB should* N5 frozen or TI should* N5 impinge* or AB should* N5 impinge* or TI should* N5 tendonitis or AB should* N5 tendonitis or TI should* N5 tendinitis or AB should* N5 tendinitis or TI should* N5 pain* or AB should* N5 pain* S9 S7 and S8 S8 S5 or S6 S7 (MH Pain+) S6 (MH Shoulder Joint+) S5 (MH Shoulder) S4 (MH Bursitis+)
S3 (MH Rotator Cuff+) S2 (MH Shoulder Impingement Syndrome) S1 (MH Shoulder Pain) Web of Science #3 #2 AND #1 #2 Topic=(botulin* or botox or OnabotulinumtoxinA or RimabotulinumtoxinB or AbobotulinumtoxinA or BTXA or dyslor or dysport or lanzox or myobloc or neurobloc or oculinum or prosigne or vistabel or vistabex or xeomin) #1 Topic=(shoulder* AND (pain* or frozen or impinge* or tendonitis or tendinitis or bursitis)) OR Topic=(rotator cuff ) Dissertation Abstracts Shoulder* or rotator cuff in citation and abstract AND (botulin* or botox or OnabotulinumtoxinA or RimabotulinumtoxinB or AbobotulinumtoxinA or BTXA or dyslor or dysport or lanzox or myobloc or neurobloc or oculinum or prosigne or vistabel or vistabex or xeomin) in in citation and abstract Current Controlled Trials Search 1 botox and shoulder* Search 2 botox and rotator cuff Search 3 botulin* and shoulder* Search 3 BTXA Search 4 dyslor Search 5 dysport Search 6 lanzox Search 7 neurobloc Search 8 oculinum Search 9 prosigne Search 10 vistabel Search 11 vistabex Search 12
WHATS NEW
Last assessed as up-to-date: 21 January 2010.
Date 16 November 2010
Event Amended
Description Correct minor typing errors, and gure 1
46
HISTORY
Protocol rst published: Issue 1, 2010 Review rst published: Issue 9, 2010
Date 2 September 2009
Event Amended
Description CMSG ID A047-P
CONTRIBUTIONS OF AUTHORS
JS: title registration JS, PF: protocol development and editing JS, PF: review of titles and abstracts JS, PF: data abstraction JS: data entry into Revman JS: data analysis JS, PF: drafting and revising the review
DECLARATIONS OF INTEREST
JS: speaker honoraria from Abbott pharmaceuticals; research grants from Amgen, Allergan, Takeda and Savient pharmaceuticals; consultant fee from Savient and URL pharmaceuticals; travel grants from Allergan and Wyeth pharmaceuticals. PF: none. The review included one study by a review author (JS). Selection of studies for inclusion as well as abstraction of all data were done independently by the two review authors (JS, PF). In addition, the abstracted data were checked for accuracy by another individual (KM).
SOURCES OF SUPPORT Internal sources

Birmingham VA Medical Center, Minneapolis, USA, USA. Research time protection
47
External sources
NIH CTSA KL2 Award (Mayo Clinic), USA, Not specied. 75% time protection for research
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

We did not perform the preplanned assessment of trials against the modied CONSORT criteria for non-pharmacological treatment (Boutron 2008) as we decided this assessment would add only limited value to the review.
NOTES
None
INDEX TERMS Medical Subject Headings (MeSH)

Analgesics, Non-Narcotic [ therapeutic use]; Botulinum Toxins [ therapeutic use]; Randomized Controlled Trials as Topic; Range of Motion, Articular; Shoulder Pain [ drug therapy]
MeSH check words

Humans
48

Botulinum Toxin For Shoulder Pain (Review) : Singh JA, Fitzgerald PM

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Botulinum Toxin For Shoulder Pain (Review) : Singh JA, Fitzgerald PM

Uploaded by

Copyright:

Available Formats

Botulinum toxin for shoulder pain (Review)

Singh JA, Fitzgerald PM

Botulinum toxin for shoulder pain

Assumed risk Control

343 per 1000 (150 to 790)

RR 1.46 (0.64 to 3.36)

Disability/function - not See comment reported

Description of the condition

How the intervention might work

Why it is important to do this review

Description of the intervention

Criteria for considering studies for this review

Search methods for identication of studies

Data collection and analysis

Figure 1. Study selection ow chart

Risk of bias in included studies

Summary of main results

Quality of the evidence

Potential biases in the review process

Agreements and disagreements with other studies or reviews

Overall completeness and applicability of evidence

AUTHORS CONCLUSIONS Implications for practice

Implications for research

References to studies included in this review

References to studies excluded from this review

Characteristics of included studies [ordered by study ID]

Notes Risk of bias Item Adequate sequence generation?

Authors judgement Unclear

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes

Free of other bias?

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes Free of selective reporting?

Free of other bias?

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes

Free of selective reporting?

Free of other bias?

Notes Risk of bias Item Authors judgement Description

Adequate sequence generation?

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes Free of selective reporting?

Free of other bias?

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes Free of selective reporting?

Free of other bias?

Blinding? All outcomes

Incomplete outcome data addressed? All outcomes Free of selective reporting?

No missing outcome data.

Free of other bias?

Characteristics of excluded studies [ordered by study ID]

DATA AND ANALYSES

Statistical method Mean Difference (IV, Random, 95% CI)

Intramuscular botulinum toxin versus intra-articular triamcinolone in shoulder spasticity: 12

Statistical method Mean Difference (IV, Random, 95% CI)

Effect size Totals not selected

Mean Difference (IV, Random, 95% CI)

Totals not selected

Mean Difference (IV, Random, 95% CI)

Totals not selected

Mean Difference (IV, Random, 95% CI)

Totals not selected

Mean Difference (IV, Random, 95% CI)