US Cochrane PDF

Therapeutic ultrasound for osteoarthritis of the knee or hip (Review)
Rutjes AWS, Nesch E, Sterchi R, Jni P
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2010, Issue 1 http://www.thecochranelibrary.com
Therapeutic ultrasound for osteoarthritis of the knee or hip (Review) Copyright 2010 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. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 1 Pain. . . . . . . . Analysis 1.2. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 2 Function. . . . . . . Analysis 1.3. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 3 Number of patients experiencing any adverse event. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.4. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 4 Number of patients withdrawn or dropped out due to adverse events. . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.5. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 5 Number of patients experiencing any serious adverse event. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 2 6 6 6 8 9 11 12 13 16 17 19 21 21 21 24 36 36 37 37 38 38 38 41 41 41 42 42 42 42
[Intervention Review]
Therapeutic ultrasound for osteoarthritis of the knee or hip

Anne WS Rutjes1 , Eveline Nesch1 , Rebekka Sterchi1 , Peter Jni1
1 Division
of Clinical Epidemiology and Biostatistics, Institute of Social and Preventive Medicine, University of Bern, Bern, Switzerland
Contact address: Anne WS Rutjes, Division of Clinical Epidemiology and Biostatistics, Institute of Social and Preventive Medicine, University of Bern, Finkenhubelweg 11, Bern, 3012, Switzerland. arutjes@ispm.unibe.ch. a.rutjes@crc-cesi.org. Editorial group: Cochrane Musculoskeletal Group. Publication status and date: New search for studies and content updated (conclusions changed), published in Issue 1, 2010. Review content assessed as up-to-date: 22 July 2009. Citation: Rutjes AWS, Nesch E, Sterchi R, Jni P. Therapeutic ultrasound for osteoarthritis of the knee or hip. Cochrane Database of Systematic Reviews 2010, Issue 1. Art. No.: CD003132. DOI: 10.1002/14651858.CD003132.pub2. Copyright 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT Background Osteoarthritis is the most common form of joint disease and the leading cause of pain and physical disability in the elderly. Therapeutic ultrasound is one of several physical therapy modalities suggested for the management of pain and loss of function due to osteoarthritis (OA). Objectives To compare therapeutic ultrasound with sham or no specic intervention in terms of effects on pain and function safety outcomes in patients with knee or hip OA. Search strategy We updated the search in CENTRAL, CINAHL, EMBASE, MEDLINE and PEDro up to 23 July 2009, checked conference proceedings, reference lists, and contacted authors. Selection criteria Studies were included if they were randomised or quasi-randomised controlled trials that compared therapeutic ultrasound with a sham intervention or no intervention in patients with osteoarthritis of the knee or hip. Data collection and analysis Two independent review authors extracted data using standardized forms. Investigators were contacted to obtain missing outcome information. Standardised mean differences (SMDs) were calculated for pain and function, relative risks for safety outcomes. Trials were combined using inverse-variance random-effects meta-analysis. Main results Compared to the previous version of the review, four additional trials were identied resulting in the inclusion of ve small sized trials in a total of 341 patients with knee OA. No trial included patients with hip OA. Two evaluated pulsed ultrasound, two continuous and one evaluated both pulsed and continuous ultrasound as the active treatment. The methodological quality and the quality of reporting was poor and a high degree of heterogeneity among the trials was revealed for function (88%). For pain, there was an effect in favour of ultrasound therapy, which corresponded to a difference in pain scores between ultrasound and control of -1.2 cm on a 10-cm VAS (95% CI -1.9 to -0.6 cm). For function, we found a trend in favour of ultrasound, which corresponded to a difference in function scores of -1.3 units on a standardised WOMAC disability scale ranging from 0 to 10 (95% CI -3.0 to 0.3). Safety was evaluated in two trials including up to 136 patients; no adverse event, serious adverse event or withdrawals due to adverse events occurred in either trial.
Therapeutic ultrasound for osteoarthritis of the knee or hip (Review) Copyright 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 1
Authors conclusions In contrast to the previous version of this review, our results suggest that therapeutic ultrasound may be benecial for patients with osteoarthritis of the knee. Because of the low quality of the evidence, we are uncertain about the magnitude of the effects on pain relief and function, however. Therapeutic ultrasound is widely used for its potential benets on both knee pain and function, which may be clinically relevant. Appropriately designed trials of adequate power are therefore warranted.
PLAIN LANGUAGE SUMMARY Therapeutic ultrasound for osteoarthritis This summary of a Cochrane review presents what we know from research about the effect of therapeutic ultrasound on knee or hip osteoarthritis. The previous version of this review concluded that therapeutic ultrasound had no benet over fake therapeutic ultrasound in pain relief and functional status. The updated review shows that in people with osteoarthritis, -Therapeutic ultrasound may be benecial for people with osteoarthritis of the knee. -Therapeutic ultrasound may improve your physical function but this nding could be the result of chance. - We are uncertain about the magnitude of effects on pain relief or the ability to use your knee, because of the low quality of the evidence. -Therapeutic ultrasound may not have any side effects: no side effects were reported, but we do not have precise information about side effects. This is particularly true for rare but serious side effects. There are no studies that address the benets of therapeutic ultrasound in people with hip osteoarthritis. What is osteoarthritis and what is therapeutic ultrasound? Osteoarthritis is a disease of the joints, such as your knee or hip. When the joint loses cartilage, the bone grows to try and repair the damage. Instead of making things better, however, the bone grows abnormally and makes things worse. For example, the bone can become misshapen and make the joint painful and unstable. This can affect your physical function or ability to use your knee. Therapeutic ultrasound means using sound waves to try and relieve pain or disability. Your doctor or physiotherapist will use a roundheaded wand or probe on the skin of the painful area. Ultrasound gel is used on the wand and on your skin to make it more comfortable and help the sound waves reach the affected area. Best estimate of what happens to people with osteoarthritis who have had therapeutic ultrasound for a duration of 2-8 weeks: Pain -People who used therapeutic ultrasound had an improvement in their pain of about 3 on a scale from 0 (no pain) to 10 (extreme pain) after using it up to 2 months. -People who used a fake therapeutic ultrasound had an improvement in their pain of about 2 on a scale from 0 to 10 after using it up to 2 months. Another way of saying this is: - 37 people out of 100 who use therapeutic ultrasound respond to treatment (37%). - 31 people out of 100 who use fake therapeutic ultrasound respond to treatment (31%). - 6 more people respond to treatment with therapeutic ultrasound than with fake therapeutic ultrasound (difference of 6%).
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]
Therapeutic ultrasound compared with sham or no specific intervention for patients with knee or hip osteoarthritis Patient or population: Patients with knee osteoarthritis (none of the trials included patients with hip osteoarthritis) Settings: Private physical therapy practice in USA (1 trial) and medical university hospital outpatient clinic in Taiwan (2 trials), and outpatient clinic of department of physical medicine and rehabilitation in Turkey (2 trials) Intervention: Therapeutic ultrasound Comparison: Sham or no specific intervention Outcomes Illustrative comparative risks* (95% CI) Relative effect (95% CI) No of Participants (studies) Quality of the evidence (GRADE) Comments
Assumed risk
Corresponding risk
Sham or no specific in- Therapeutic ultrasound tervention Pain VAS pain scales Follow-up: 2 to 8 weeks -1.8 cm change on 10 -3.0 cm change SMD -0.49 (-0.76 to - 320 cm VAS1 ( -1.2 cm, -1.9 to -0.6 0.23) (5) cm)2 29% improvement 50% improvement ( 21%, 10% to 32%)3 -2.6 units on WOMAC ( -1.3, -3.0 to 0.3)6 46% improvement ( 25%, -5% to 54%)7 SMD -0.64 (-1.42 to 251 0.14) (4) ++OO low4 NNT: 6 (95% CI 5 to 12)5
Function -1.2 units on WOMAC Various validated function (range 0 to 10)1 scales 21% improvement Follow-up: 2 to 8 weeks
++OO low8
Little evidence for a beneficial effect [NNT: not statistically significant] 0 events reported NNH not estimable
Number of patients ex- 150 per 1000 patient- See comment periencing any adverse years1 event Follow-up: 2 weeks Number of patients who 17 per 1000 patient-years See comment withdrew because of ad- 1 verse events Follow-up: 2 weeks
not estimable
67 (1)
++OO low9
not estimable
67 (1)
++OO low9
0 events reported NNH not estimable
Number of patients ex- 4 per 1000 patient-years See comment periencing any serious 1 adverse event Follow-up: 2 and 15 weeks
not estimable
136 (2)
++OO low10
0 events reported NNH not estimable
*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; GRADE: GRADE Working Group grades of evidence (see explanations); SMD: standardised mean difference; NNT: number needed to treat; NNH; number needed to harm GRADE Working Group grades of evidence High quality (++++): Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality (+++O): Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality (++OO): 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 (+OOO): We are very uncertain about the estimate. 1 Median reduction as observed across placebo groups in large osteoarthritis trials (see methods section, Nesch 2009). 2 Standardised mean differences (SMDs) were back-transformed onto a 10 cm visual analogue scale (VAS) on the basis of a typical pooled SD of 2.5 cm in large trials that assessed pain using a VAS and expressed as change based on an assumed standardised reduction of 0.72 standard deviation units in the control group. 3 Percentage of improvement was calculated based on median observed pain at baseline across control groups of large osteoarthritis trials of 6.1 cm on 10 cm VAS (Nesch 2009). 4 Downgraded (2 levels) because included trials were generally of low quality and small sample size: analyses was not according to intention to treat principle in at least 3 trials; concealment of allocation methods was unclear in all trials; and the possible indirectness of evidence (indirect population) in 1 trial. 5 Absolute response risks for pain in the control groups were assumed 31% (see methods section). 6 Standardised mean differences (SMDs) were back-transformed onto a standardised WOMAC disability score ranging from 0 to 10 on the basis of a typical pooled SD of 2.1 in trials that assessed function using WOMAC disability scores and expressed as change based on an assumed standardised reduction of 0.58 standard deviation units in the control group. 7 Percentage of improvement was calculated based on median observed WOMAC function scores at baseline across control groups of large osteoarthritis trials of 5.6 units (Nesch 2009). 8 Downgraded (2 levels) because included trials were generally of low quality and small sample size: analyses was not according to intention to treat principle in at least 2 trials; concealment of allocation methods was unclear in all trials; inadequate patient blinding in 3 trials, possible indirectness of evidence (indirect population) in 1 trial, the presence of large between trial heterogeneity and the confidence interval crossed no difference.
Downgraded (2 levels) because of: small sample size, due to absence of events, no relative risk could be calculated, 1 out of 5 studies reported this outcome, possibly leading to selective outcome reporting bias. 10 Downgraded (2 levels) because of: small sample size, due to absence of events, no relative risk could be calculated, 2 out of 5 studies reported this outcome, possibly leading to selective outcome reporting bias.
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
BACKGROUND
Osteoarthritis (OA) is an age related condition, occurring more frequently in women than in men. Its prevalence, causal associations and outcomes vary markedly according to the joint site affected (Jni 2006). OA is characterized by focal areas of loss of articular cartilage in synovial joints, accompanied by subchondral bone changes, osteophyte formation at the joint margins, thickening of the joint capsule and mild synovitis (Solomon 1997). The objectives of management of knee and hip OA are to relieve pain and to maintain or improve function. Therapeutic ultrasound is one of many physical therapy modalities and can be used as part of an overall rehabilitation program (Rand 2007). Therapeutic ultrasound consists of high frequency vibrations (Hartley 1993; Nelson 1999) that can be pulsed or continuous. Pulsed ultrasound produces non-thermal effects and is generally recommended for acute pain and inammation. Continuous ultrasound generates thermal effects (Rand 2007). Therapeutic ultrasound that penetrates deeply enough to increase collagen elasticity may be useful in the early stages of a exibility program (Hicks 1990). The 2008 OARSI recommendations for the management of hip and knee OA state that the optimal management of OA requires a combination of non-pharmacological and pharmacological modalities, including physical therapy (Zhang 2008), but does not specically mention ultrasound as treatment adjunct. Only 1 out of 5 existing guidelines considering ultrasound actually recommended its use, but the recommendation was not based on evidence from randomised trials.
Types of participants At least 75% of patients with clinically and/or radiologically conrmed osteoarthritis of the knee or hip.
Types of interventions Any type of continuous or pulsed therapeutic ultrasound. Phonophoresis (the use of ultrasound to enhance the delivery of topically applied drugs) or extracorporeal ultrasonic treatment (derivative of lithotripsy, the mechanical breaking up of renal stones with sound waves) were not considered. Eligible control interventions were placebo or a non-intervention control (usual care). Concurrent therapy was accepted providing it was provided to both, experimental and control groups.
Types of outcome measures
Main outcomes
OBJECTIVES
We set out to compare therapeutic ultrasound with sham or no specic intervention in terms of effects on pain and function and safety outcomes in patients with knee or hip OA and to explore whether potential variation between trials could be explained by characteristics of the ultrasound, by biases affecting individual trials or by publication bias.
METHODS
Criteria for considering studies for this review
Types of studies Randomised or quasi randomised controlled trials with a control group receiving a sham intervention or no intervention.
Main outcomes were pain and function, as currently recommended for osteoarthritis trials (Altman 1996; Pham 2004). If data on more than one pain scale were provided for a trial, we referred to a previously described hierarchy of pain-related outcomes (Jni 2006; Reichenbach 2007) and extracted only data on the pain scale that was highest on this hierarchy: 1. Global pain 2. Pain on walking 3. WOMAC osteoarthritis index pain subscore 4. Composite pain scores other than WOMAC 5. Pain on activities other than walking 6. Rest pain or pain during the night 7. WOMAC global algofunctional score 8. Lequesne osteoarthritis index global score 9. Other algofunctional scale 10. Patients global assessment 11. Physicians global assessment If data on more than one function scale were provided for a trial, we extracted data according to hierarchy presented below. 1. Global disability score 2. Walking disability 3. WOMAC disability subscore 4. Composite disability scores other than WOMAC 5. Disability other than walking 6. WOMAC global scale 7. Lequesne osteoarthritis index global score 8. Other algofunctional scale 9. Patients global assessment 10. Physicians global assessment If pain or function outcomes were reported at several time-points, we extracted the estimate at the end of the treatment period.
6
Secondary outcomes
Secondary outcomes were the number of patients experiencing any adverse event, patients who were withdrawn or dropped out because of adverse events, and patients experiencing any serious adverse events between randomisation and end of follow-up. Serious adverse events were dened as events resulting in hospitalisation, prolongation of hospitalisation, persistent or signicant disability, congenital abnormality/birth defect of offspring, lifethreatening events or death.
Search methods for identication of studies

Electronic searches We searched electronic databases of CENTRAL through The Cochrane Library (http://mrw.interscience.wiley.com/cochrane/ ), MEDLINE and EMBASE through the Ovid platform (www.ovid.com), CINAHL through EBSCOhost, Physiotherapy Evidence Database (PEDro, http://www.pedro.org.au/), all from implementation to July 23, 2009 using a combination of keywords and text words related to ultrasound combined with keywords and text words related to osteoarthritis and a validated lter for controlled clinical trials (Dickersin 1994). The search strategy is presented in Appendix 1 and Appendix 2. Searching other resources We manually searched conference proceedings, used Science Citation Index to retrieve reports citing relevant articles, contacted content experts and trialists and screened reference lists of all obtained articles, including related reviews. Finally, we searched several clinical trial registries (www.clinicaltrials.gov, www.controlledtrials.com, www.actr.org.au, www.umin.ac.jp/ctr) to identify ongoing trials. The last update of the manual search was on February 02, 2009.
by discussion. We extracted the characteristics of the therapeutic application, including type of device, head size of the probe, any specic skin or probe preparation, the frequency and duration of treatment and the total number of sessions. Other data extracted included the type of control intervention used, patient characteristics (gender, average age, duration of symptoms, type of joint), characteristics of pain, function, and safety outcomes, design, trial size, trial duration (dened as time from randomisation until end of follow-up), type and source of nancial support and publication status. When necessary, means and measures of dispersion were approximated from gures in the reports. For crossover trials, we extracted data from the rst period only. Whenever possible, we used results from an intention-to-treat analysis. If effect sizes could not be calculated, we contacted the authors for additional data.
Assessment of risk of bias in included studies Two review authors (AR and EN or RS) independently assessed randomisation, blinding and adequacy of analyses (Jni 2001). Disagreements were resolved by consensus. Two components of randomisation were assessed: generation of allocation sequences and concealment of allocation. Generation of sequences was considered adequate if it resulted in an unpredictable allocation schedule; mechanisms considered adequate included random-number tables, computer-generated random numbers, minimisation, coin tossing, shufing of cards, and drawing of lots. Trials using an unpredictable allocation sequence were considered randomised; trials using potentially predictable allocation mechanisms, such as alternation or the allocation of patients according to date of birth, were considered quasi-randomised. Allocation concealment was considered adequate, if the investigators responsible for patient selection were unable to suspect before allocation which treatment was next; methods considered adequate included central randomisation and sequentially numbered, sealed, opaque envelopes. Blinding of patients was considered adequate if a sham intervention was used that was identical in appearance to the experimental intervention. Handling of incomplete outcome data was considered adequate if all randomised patients were included in the analysis (intentionto-treat principle). Finally, we used GRADE to describe the quality of the overall body of evidence (Guyatt 2008; Higgins 2008), dened as the extent of condence into the estimated treatment benets and harms.
Data collection and analysis

Selection of studies Two review authors independently evaluated all titles and abstracts for eligibility. Disagreements were resolved by discussion. We applied no language restrictions. If multiple reports described the same trial, we considered all. Data extraction and management Two review authors (AR and EN or RS) extracted trial information independently using a standardised, piloted data extraction form accompanied by a codebook. Disagreements were resolved
Data synthesis Continuous outcomes were summarized using standardized mean differences (SMD), with the differences in mean values at the end of treatment across treatment groups divided by the pooled standard deviation. If differences in mean values at the end of the treatment were unavailable, differences in mean changes were used. If some of the required data were unavailable, we used approximations as previously described (Reichenbach 2007). Huang
7
2005 B had two experimental groups, which were combined using weighted averages of the point estimates and pooled standard deviations. In Huang 2005 A and Huang 2005 B the units of randomisation were patients, whereas the units of analysis were knees. We therefore inated the standard error with the square root of the ratio of number of knees divided by number of patients to account for the correlation of knees within patients. An SMD of -0.20 standard deviation units can be considered a small difference between experimental and control group, an SMD of -0.50 a moderate, and -0.80 a large difference (Cohen, 1988; Jni 2006). SMDs can also be interpreted in terms of the percent of overlap of the experimental groups scores with scores of the control group. An SMD of -0.20 indicates an overlap in the distributions of pain or function scores in about 85% of cases, an SMD of -0.50 in approximately 67%, and an SMD of -0.80 in about 53% of cases (Cohen, 1988; Jni 2006). On the basis of a median pooled SD of 2.5 cm found in large-scale osteoarthritis trials that assessed pain using a 10 cm visual analogue scale (VAS) (Nesch 2009), SMDs of -0.20, -0.50 and -0.80 correspond to approximate differences in pain scores between experimental and control groups of 0.5, 1.25 and 2.0 cm on a 10 cm VAS. SMDs for function were back transformed to a standardised WOMAC disability score (Bellamy 1995) ranging from 0 to 10 on the basis of a median pooled SD of 2.1 units observed in large-scale osteoarthritis trials (Nesch 2009). Binary outcomes were expressed as relative risks. We used standard inverse-variance random-effects meta-analysis ( DerSimonian 1986) to combine trials. We quantied heterogeneity between trials using the I2 statistic (Higgins 2003), which describes the percentage of variation across trials that is attributable to heterogeneity rather than to chance and the corresponding 2 test. I2 values of 25%, 50% and 75% may be interpreted as low, moderate, and high between-trial heterogeneity, although the interpretation of I2 depends on the size and number of trials included (Rcker 2008). The association between trial size and treatment effects was investigated in funnel plots, plotting effect sizes on the horizontal axis against their standard errors on the vertical axis. We assessed asymmetry by the Eggers test (Sterne 2001). For the main analysis on pain and function, we performed analyses stratied by the following trial characteristics: concealment of allocation, use of a sham intervention in the control group, blinding of patients, blinding of outcome assessors, analysis in accordance with the intention-to-treat principle, trial size, difference in the use of analgesic co-interventions, use of concurrent treatment, specic type of ultrasound therapy (pulsed versus continuous), duration of stimulation per session, total number of sessions, duration of treatment period and source of funding. A cut-off of 200 patients was used to distinguish between small and large trials; a sample
size of 100 patients per group will yield more than 80% power to detect a small to moderate SMD of -0.40 at a two-sided P value of 0.05. Then, we converted SMDs of pain intensity and function to odds ratios (Chinn 2000) to derive numbers needed to treat (NNT) to cause one additional treatment response on pain or function as compared with control, and numbers needed to harm (NNH) to cause one additional adverse outcome. We dened treatment response as a 50% improvement in scores (Clegg 2006), which corresponds to an average decrease of 1.2 standard deviation units (Nesch 2009). With a median standardised pain intensity at baseline of 2.4 standard deviation units and a median standardised decrease in pain scores of 0.72 standard deviation units observed in large osteoarthritis trials (Nesch 2009), we calculated that a median of 31% of patients in the control group would achieve an improvement of pain scores of 50% or more. This percentage was used as the control group response rate to calculate NNTs for treatment response on pain. Based on the median standardised WOMAC function score at baseline of 2.7 standard deviation units and the median standardised decrease in function scores of 0.58 standard deviation units (Nesch 2009), 26% of patients in the control group would achieve a reduction in function of 50% or more. Again, this percentage was used as the control group response rate to calculate NNTs for treatment response on function. Median risks of 150 patients with adverse events per 1000 patient-years, 4 patients with serious adverse events per 1000 patient-years and 17 dropouts due to adverse events per 1000 patient-years observed in placebo groups in large osteoarthritis trials (Nesch 2009) were used to calculate NNHs for safety outcomes. Analyses were performed in RevMan version 5 (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen) and STATA version 10.1 (StataCorp, College Station, Texas). All P values are twosided.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classication. We identied 2156 references to articles and considered 66 to be potentially eligible (Figure 1). Five randomised trials in 341 patients met our inclusion criteria. Two trials evaluated pulsed ultrasound (Falconer 1992; Huang 2005 A), two continuous ( Cetin 2008; zgnenel 2009) and one evaluated both pulsed and continuous ultrasound as the active treatment (Huang 2005 B).
Figure 1. Flow chart
All trials used a parallel group design, in which three used a sham intervention in the control group (Falconer 1992; Huang 2005 B; zgnenel 2009). All but one trial (zgnenel 2009) provided standardised concurrent treatment in both experimental and control groups. Cetin 2008 used hot packs and isokinetic exercise, Falconer 1992 isometric exercise, mobilising exercise, manual therapy and joint protection, Huang 2005 A and Huang 2005 B isokinetic exercise and hot packs. In addition, instruction in daily home exercises were given in the trials by Falconer 1992; Huang 2005 A and Huang 2005 B. In the trial by zgnenel 2009, it was unclear if concurrent treatment was given, with the authors stating that no physiotherapy was prescribed prior to ultrasound treatment to either of the groups, but referring to their intervention as physiotherapy program. Therefore, we could not exclude that other physiotherapy modalities were provided in addition to ultrasound treatment. Treatment parameters were insufciently described (see Characteristics of included
studies). All trials used ultrasound probes with direct skin contact at 1 MHz. 1 to 1.5 W/cm2 was the maximum intensity in trials using continuous ultrasound, 2.5 W/cm2 in trials using pulsed ultrasound. Two trials explicitly stated the mean intensity, which was 1.5 W/cm2 in Cetin 2008 and 1.7 W/cm2 in Falconer 1992. The two trials by Huang and colleagues reported that intensity was increased up to a level at which a warm sensation or a mild sting was felt, not exceeding 2.5 W/cm2 . Only two trials reported the size of the probe used (Falconer 1992; zgnenel 2009). The duration of treatment per lesion was reported to be 3, 5 and 5 minutes, the number of lesions treated per session four, three and three, respectively in Falconer 1992; Huang 2005 A; and Huang 2005 B. From these data, we estimated that the duration of ultrasound treatment per session was 12, 15 and 15 minutes in these trials. Cetin 2008 and zgnenel 2009 described that the duration of ultrasound per session was 10 and 5 minutes, respectively. The median number of treatment sessions per week was 3 (range 2 to 5), the median length of the treatment period
was 8 weeks (range 2 to 8). All trials explicitly included patients with knee OA only, with the diagnosis based on clinical and/or radiographic evidence. The majority of patients had a clinical severity requiring simple non-surgical treatments (Jni 2006). In one trial, 21 out of 40 patients had severity grade III or IV according to Kellgren and Lawrence, however (Cetin 2008). In another trial, 8 out of 69 patients had received total knee arthroplasty more than 6 months before the trial started, and all patients in this study had chronic knee contractures (Falconer 1992). One trial reported the average disease duration, which was 7 years (Falconer 1992), two reported the average body mass index, which was 28 and 32 kg/m2 in Cetin 2008 and zgnenel 2009, respectively. Five randomised trials were excluded because a concurrent therapy was given in the ultrasound group, which was not given in the control group: ultrasound therapy, exercise and hot packs versus exercise in Ones 2006, treatment according to Dutch physiotherapy guideline versus behavioral graded activity in Veenhof 2007, pulsed short wave diathermy, exercise and ultrasound therapy versus pulsed short wave diathermy with different parameters, exercise and ultrasound therapy in Tuzun 2003, ultrasound therapy, transcutaneous electrostimulation and exercise versus exercise in Dincer 2008 and glucosamine sulfate, NSAID and ultrasound therapy versus combined traditional Chinese medicine versus glucosamine sulfate, NSAID, ultrasound therapy and traditional Chinese medicine in Qin 2008. Three randomised trials were excluded because ultrasound was compared to active control interventions: ultrasound therapy versus short wave diathermy or ultrasound therapy with exercise or diathermy with exercise in Jan 1991, ultrasound therapy with deep relief gel versus ultrasound therapy with standard gel in Yang 2005, ultrasound therapy versus phonophoresis in Danova 1996 and Kozanoglu 2003, and ultrasound therapy versus shortwave diathermy or galvanic electrostimulation in Bansil 1975, Kalpakcioglu 2006 and Svarcova 1988. Bansil 1975 and Svarcova 1988 were included in the previous version of this review. The Characteristics of excluded studies table presents the main reasons for exclusion of other studies.
Risk of bias in included studies

Figure 2 summarises the methodological characteristics and source of funding of included trials. No trial reported adequate sequence
generation or adequate concealment of allocation. Both trials by Huang et al reported the use of sequentially numbered, sealed and opaque envelopes, but they reported that envelopes were sequentially numbered I to IV, likely to represent the four comparison groups patients were allocated to in both of these trials. The envelopes should have been numbered 1 to 140 and 1 to 120 to cover the numbers of patient randomised in Huang 2005 A and Huang 2005 B, respectively. As we had no evidence to suggest that the envelopes were appropriately numbered in a sequential manner, random allocation could have been undermined. Two trials were described as double-blind (Falconer 1992, zgnenel 2009). Three trials used sham interventions, using identical devices in experimental and control groups (Falconer 1992, Huang 2005 B, zgnenel 2009). In Falconer 1992, the sham device was fully functional, dials were lit, audible beeps were given, but no energy was delivered to the tissue. In Huang 2005 B, the investigators stated that the absence of a warm or stinging sensation in the control group resulted in a failure of patient blinding. zgnenel 2009 stated that the probe was disconnected on the back of a working ultrasound machine and the patient was unable to determine whether the cable was disconnected or not. Only the sham device used in Falconer 1992 and zgnenel 2009 2009 were therefore considered to have adequate patient blinding. None of the trials was considered to have performed an intentionto-treat analysis for any of the primary outcomes of this review. Exclusions had explicitly occurred in all but two trials, with percentages of exclusions ranging from 0% to 10% in experimental groups and from 0% to 17% in control groups. In two trials (Cetin 2008; zgnenel 2009) it was unclear whether exclusions from the analysis had occurred. Three trials explicitly specied primary outcomes, although it was unclear whether these were specied a priori (Falconer 1992; Huang 2005 B; zgnenel 2009). Only Falconer 1992 and zgnenel 2009 reported a sample size calculation. None of the trials had a sample size of approximately 200 patients to achieve sufcient power for detecting a small to moderate SMD. Three trials were supported by a non-prot organisation (Falconer 1992; Huang 2005 A; Huang 2005 B), one was also supported by a commercial body (Huang 2005 B). The source of funding was unclear in the the other two trials (see Characteristics of included studies).
10
Figure 2. Methodological characteristics and source of funding of included trials. (+) indicates low risk of bias, (?) unclear and (-) a high risk of bias on a specic item.
11
For the effectiveness outcomes pain and function, the quality of the evidence (Guyatt 2008) was classied as low in view of the risk of bias in the included, predominantly small trials of questionable quality, the large heterogeneity between trials on function, the potential for selective reporting of function outcomes and the possible indirectness of one trial (Summary of ndings for the main comparison). For the safety outcomes, the quality of the evidence (Guyatt 2008) was classied as low, again because of the predominantly small trials of questionable quality, the small number of trials reporting the outcomes and the absence of events resulting in not estimable relative risks.
Effects of interventions
See: Summary of ndings for the main comparison Pain Five trials (320 patients, 464 knees) contributed to the meta-analysis of pain outcomes (Figure 3). The analysis suggested a SMD of -0.49 (95% CI -0.76 to -0.23), which corresponds to a difference in pain scores of 1.2 cm on a 10-cm VAS between ultrasound and control. Referring to a median pain intensity of 6.1 cm in placebo groups at baseline, this corresponds to a difference of 21% improvement (95% CI 10% to 32%) between electrostimulation and control (Summary of ndings for the main comparison). The NNT to cause one additional treatment response on pain as compared to control was 6 (95% CI 5 to 12). An I2 of 26% indicated a low degree of of between-trial heterogeneity (P value for heterogeneity = 0.25). The funnel plot did not appear asymmetrical ( Figure 4, P value for asymmetry = 0.23). Figure 3. Forest plot of 5 trials comparing the effects of therapeutic ultrasound and control (sham or no intervention) on knee pain. Values on x-axis denote standardised mean differences. Data relating to the 2 experimental arms in Huang 2005 B were pooled.
12
Figure 4. Funnel plot for effects on knee pain. Numbers on x-axis refer to standardised mean differences (SMDs), on y-axis to standard errors of SMDs.
Table 1 presents results from stratied analyses. There was little evidence that estimates of SMD increased with the duration of ultrasound therapy applied per session (P value for trend = 0.15). There was no evidence to suggest that SMDs depended on the type of ultrasound therapy used (P value for interaction = 0.48). SMDs did not seem to depend on type of control intervention, patient blinding, total number of sessions or treatment duration. The absence of sufcient (high quality) trials hampered the exploration of associations of SMDs with other trial characteristics. In all trials it was unclear whether the provision of analgesic cointerventions was similar between groups, none had used adequate concealment of allocation, performed analyses according to the intention-to-treat principle or achieved a sample size of at least 200 patients. Therefore, we could not evaluate the impact of these characteristics. Table 1. Results of stratied analyses of pain outcomes
Variable
No. of trials
No. of patients: No. of patients: experimental control n n
Pain-intensity
Heterogeneity
P for interaction
SMD (95% CI)
I2 (%)
13
Table 1. Results of stratied analyses of pain outcomes
(Continued)
All trials
177
143
-0.49 (-0.76 to - 26% 0.23) 0.8
Type of control intervention Sham interven- 3 tion No control inter- 2 vention Blinding of patients Adequate 2 68 68 -0.42 (-0.76 to - 0% 0.08) -0.52 (-1.01 to - 60% 0.04) 125 93 -0.52 (-0.80 to - 0% 0.24) -0.40 (-1.24 to 77% 0.44)
52
50
0.71
Inadequate unclear
or 3
109
75
Blinding of outcome assessor Adequate 2 68 68 -0.42 (-0.76 to - 0% 0.08) -0.52 (-1.01 to - 60% 0.04)
0.71
Inadequate unclear
or 3
109
75
Type of Ultrasound Continuous 3 81 65 -0.39 (-0.80 to 45% 0.02) -0.62 (-0.97 to - 0% 0.26)
0.48
Pulsed
96
78
Duration of US per session 5 minutes 1 34 33 -0.38 (-0.87 to N/A 0.11) 0.04 (-0.58 to N/A 0.66)
0.15
10 minutes
20
20
14
Table 1. Results of stratied analyses of pain outcomes
(Continued)
12 minutes
34
35
-0.45 (-0.93 to N/A 0.03) -0.77 (-1.11 to - 0% 0.41) 0.69
15 minutes
89
55
Total number of sessions 10 1 34 33 -0.38 (-0.87 to N/A 0.11) -0.45 (-0.93 to N/A 0.03) -0.52 (-1.01 to - 60% 0.04)
12
34
35
24
109
75
Duration of treatment period 2 weeks 1 34 33 -0.38 (-0.87 to N/A 0.11) -0.45 (-0.93 to N/A 0.03) -0.52 (-1.01 to - 60% 0.04)
0.69
6 weeks
34
35
8 weeks
109
75
Commercial funding Yes 1 57 25 -0.72 (-1.20 to - N/A 0.23) -0.43 (-0.74 to - 31% 0.11)
0.45
No or unclear
120
118
US: Ultrasound therapy; No.: number; N/A: not applicable; Huang 2005 B contributed to both strata, with the same 25 control patients displayed in each stratum; , P-value from test for trend.
15
Function Four trials (251 patients) contributed to the meta-analyses of function (Figure 5). The analysis suggested a SMD of -0.64 (95% CI -1.42 to 0.14, P value = 0.11), corresponding to a difference in function scores of 1.3 units on a standardised WOMAC disability scale ranging from 0 to 10, favouring ultrasound therapy. Referring to a median function score of 5.6 units in placebo groups at baseline, this corresponds to a difference of 25% improvement (95% CI -5% to 54%) between electrostimulation and control (Summary of ndings for the main comparison). The NNT to cause one additional treatment response on function as compared to control was not calculated as differences were not statistically signicant. An I2 of 88% indicated a large degree of betweentrial heterogeneity (P value for heterogeneity < 0.001). On visual inspection, the variation could be explained by the trial of Cetin 2008 that estimated an effect size favouring control, in contrast to the other trials. The trial by Huang 2005 A reported an unrealistically large effect in favour of ultrasound therapy. The funnel plot did not appear asymmetrical (Figure 6, P value for asymmetry = 0.93). Figure 5. Forest plot of 3 trials comparing the effects of therapeutic ultrasound and control (sham or no intervention) on function. Values on x-axis denote standardised mean differences. Data relating to the 2 experimental arms in Huang 2005 B were pooled.
16
Figure 6. Funnel plot for effects on functioning of the knee. Numbers on x-axis refer to standardised mean differences (SMDs), on y-axis to standard errors of SMDs.
Table 2 presents results from stratied analyses. There was little evidence that estimates of SMD increased with the duration of ultrasound therapy applied per session (P value for trend = 0.16), but a positive interaction test suggested that SMDs were associated with the type of ultrasound therapy used, with more favourable effects of pulsed ultrasound therapy (P value for interaction = 0.005).The absence of sufcient high quality trials hampered the exploration of potential effects of all other treatment and design characteristics. Condence intervals were either wide and tests of interaction not signicant or the impact of several characteristics could not be evaluated at all. Table 2. Results of stratied analyses of function
Variable
No. of trials
No. of patients: No. of patients: experimental control n 143 n 108
Pain-intensity
Heterogeneity
P for interaction
n All trials 4
SMD (95% CI) -0.64 to0.14)
I2 (%)
(-1.42 88%
17
Table 2. Results of stratied analyses of function
(Continued)
Type of control intervention Sham interven- 2 tion No control inter- 2 vention Blinding of patients Adequate 1 34 33 -0.27 (-0.76 to N/A 0.22) -0.76 (-1.83 to 94% 0.31) 91 58 -0.58 (-1.20 to 69% 0.03) -0.68 (-2.66 to 95% 1.29)
0.93
52
50
0.70
Inadequate unclear
or 3
109
75
Blinding of outcome assessor Adequate 2 66 63 -0.97 (-2.35 to 93% 0.42) -0.30 (-1.50 to 89% 0.90)
0.52
Inadequate unclear
or 2
77
45
Type of Ultrasound Continuous 3 81 65 -0.15 (-0.56 to 41% 0.26) -1.58 (-2.00 to - 0% 1.16)
0.005
Pulsed
62
43
Duration of US per session 5 minutes 1 34 33 -0.27 (-0.76 to N/A 0.22) 0.33 (-0.30 to N/A 0.95)
0.16
10 minutes
20
20
12 minutes 15 minutes
0 2 89 55 -1.28 (-2.05 to - 76% 0.51)

18
Table 2. Results of stratied analyses of function
(Continued)
Total number of sessions 10 1 34 33 -0.27 (-0.76 to N/A 0.22)
0.70
12 24
0 3 109 75 -0.76 (-1.83 to 91% 0.31) 0.70
Duration of treatment period 2 weeks 1 34 33 -0.27 (-0.76 to N/A 0.22)
6 weeks 8 weeks
0 3 109 75 -0.76 (-1.83 to 94% 0.31) 0.78
Commercial funding Yes 1 57 25 -0.90 (-1.39 to - N/A 0.41) -0.54 (-1.65 to 91% 0.56)
No or unclear
86
83
US: Ultrasound therapy; No.: number; N/A: not applicable; Huang 2005 B contributed to both strata, with the same 25 control patients displayed in each stratum; P-value from test for trend.
Safety One trial contributed to the analysis of patients experiencing any adverse event or being withdrawn or dropping out because of adverse events (zgnenel 2009). This trial reported that no adverse events or withdrawals or dropouts due to adverse events had occurred, neither in the experimental nor in the control group. Therefore, relative risks could not be estimated. Two trials (136 patients) contributed to the meta-analysis of patients experiencing any serious adverse event (Falconer 1992; zgnenel 2009). Again, the relative risk could not be calculated since both reports stated that no serious adverse events had occurred, neither in the
ultrasound nor in the control group. In general, there was no evidence to suggest that electrostimulation is unsafe.
DISCUSSION Summary of main results

Our systematic review of trials comparing any type of therapeutic ultrasound with a sham or non-intervention control revealed a lack of adequately sized, methodologically sound and appropriately
19
reported trials and a high degree of between trial heterogeneity for one of the two main outcomes, which made the interpretation of results difcult. Overall effect sizes for pain relief and improvement of function in patients with knee osteoarthritis seemed moderately in favour of ultrasound therapy. Only two trials reported safety outcomes; generally, there is no evidence to suggest that ultrasound therapy is unsafe. In view of its mechanisms of action, ultrasound therapy is rather unlikely to cause serious adverse events but active surveillance of harms with formal monitoring of potential adverse events is clearly desirable.
Quality of the evidence

The methodological quality and the quality of reporting was poor. All trials were small sized, including 82 patients at most. None of the trials described generation of allocation sequences or concealment of allocation, or reported whether primary outcomes were specied a priori. None of the trials was analysed according to the intention-to-treat principle, and only one trial described the occurrence of adverse events or withdrawals and dropouts because of adverse events.
benet of ultrasound. If we had used walking speed data rather than the Lequesne index as a measure of function, effect sizes in the trials by Huang and colleagues would have been even more extreme in favour of ultrasound (SMD -1.7, 95% CI -2.3 to 1.1 in Huang 2005 A and -2.1, 95% CI -2.64 to -1.50 in Huang 2005 B). Walking speed was also recorded in Falconer 1992, and although we were unable to obtain sufcient data to calculate the SMD for function, the reported results indicate that differences between groups were likely to be small, non-signicant and in contrast to Huang and colleagues results. The absence of significance can not be explained by sample size alone, as the number of patients in Falconer 1992 exceeded the numbers analysed in Huang 2005 A.
Agreements and disagreements with other studies or reviews

We are aware of 14 related reviews addressing the effectiveness of ultrasound therapy for knee or hip osteoarthritis (list available on request). Here we will focus mainly on the similarities with and differences to the previous version of this review (Welch 2001), which included three studies. We updated the search and used stricter selection criteria, which resulted in four additional trials. In contrast to the review of Welch 2001, we only included randomised trials. Welch and colleagues also included other design types and allowed active treatment in the control group. Here we focused on sham or non-intervention controls and excluded two previously included studies (Bansil 1975; Svarcova 1988). In this update, we performed a detailed quality assessment of component trials, followed by an exploration of sources of variation between trials, including concealment of allocation, blinding, intentionto-treat analysis, and characteristics of ultrasound therapy, which was not possible in the previous version, due to the low number of trials identied. To analyse continuous data, Welch and colleagues used weighted mean differences of the change from baseline scores, whereas we used SMDs of end of treatment scores. In addition, xed effect models were used in the previous version unless there was statistically signicant heterogeneity between trials based on 2 testing. Model selection based on the mechanistic application of heterogeneity tests should be avoided, however. Here, we used random effects models, which will generally be more conservative in terms of the estimated precision, but will be more affected by small study effects than a xed effect model, which makes an exploration of sources of variation including different types of bias mandatory. Results from the previous and current version are therefore not directly comparable. Welch 2001 concluded that ultrasound therapy appeared to have no benet over placebo or short wave diathermy for people with hip or knee OA. Our overall estimates for pain and function seemed moderate, favouring therapeutic ultrasound, but we share concerns expressed by Welch 2001, who stated that conclusions were limited by the poor reporting of the characteristics of the device, the population, the stage of OA, therapeutic appli20
Potential biases in the review process

Our review is based on a broad literature search, and it seems unlikely that we missed relevant trials (Egger 2003). Trial selection and data extraction, including quality assessment, were done independently by two authors to minimize bias, transcription errors and other sources of observer variation (Egger 2001; Tendal 2009). Components used for quality assessment are validated and reported to be associated with bias (Jni 2001; Wood 2008; Nesch 2009). As with any systematic review, our study is limited by the quality of included trials. As indicated above, trials generally suffered from poor methodological quality, inadequate reporting and small sample size. The low number of trials hampered our exploration of the potential effects of ultrasound modalities, suboptimal design choices and type of funding. One trial showed an unrealistically large SMD for function (Huang 2005 A). Conversely, Cetin 2008 reported a null effect for pain and a trend towards a harmful effect of ultrasound for function. The condence intervals of this trials effect size for pain overlapped with the condence intervals of all other trials, accordingly the estimated heterogeneity between trials was low. SMDs for function were discrepant, however: whereas the effect size in zgnenel 2009 were just about compatible with Cetin 2008, and Huang 2005 B, they were clearly incompatible with Huang 2005 A. These discrepancies may be due to differences in characteristics of the intervention, differences in patient characteristics, bias or chance. The discrepancies are unlikely to be explained by the type of scale used, however: the Lequesne algofunctional scale used by Cetin 2008 was also used in Huang 2005 A and Huang 2005 B, which showed a clearcut
cation of the ultrasound and overall low methodological quality of the trials included. Srbely 2008 recently published a systematic review evaluating therapeutic ultrasound in people with osteoarthritis. They included ve reviews and nine original studies, including animal and non randomised studies, and refrained from statistically summarising the data. Nevertheless, the paper seems to give a complete overview on the available evidence for therapeutic ultrasound. As in the present review, the authors acknowledged the low quality of evidence. They highlighted, however, that there is an accumulating body of research suggesting that ultrasound may have the potential to provide signicant benets in the treatment and management of the osteoarthritic complex. Following this statements, we think ultrasound should not be discarded as treatment adjunctive, however, health specialist should be aware of the lack of sound evidence and be reluctant to give ultrasound therapy as a routine option, until valid evidence supporting its use becomes available. Our ratings of methodological quality differ in some respect from those that can be found in PEDro (http://www.pedro.org.au/). As an example, for both trials of Huang 2005 A and Huang 2005 B our ratings concerning generation of allocation sequences and concealment of allocation differ. We rated these components as unclear, whereas PEDro assessors rated them as adequate. This difference can not be explained by the denitions used, which are similar in our review and in PEDro. PEDro does not allow for a category unclear, however, which may have driven the assessors to give credit to these two trials and classify randomisation methods as adequate rather than inadequate in the absence of evidence suggesting neither of the two mutually exclusive categories.
Implications for practice

Despite the widespread use of therapeutic ultrasound in physical therapy, evidence of its clinical effectiveness in people with hip or knee OA is of poor quality and therefore inconclusive. The effects of ultrasound therapy on both, knee pain and function, are potentially clinically relevant and deserve further clinical evaluation.
Implications for research

The current systematic review is inconclusive, hampered by the inclusion of only small trials of questionable quality. Adequately sized randomised parallel-group trials in approximately 2 x 100 patients with knee or hip osteoarthritis are necessary to determine whether either continuous or pulsed ultrasound therapy is indeed associated with a clinically relevant benet on pain and function. A sample size of 2 x 100 patients will yield more than 80% power to detect a small to moderate SMD of -0.40 at a two-sided P value of 0.05. The trials should enrol patients without prior experience of ultrasound therapy or evaluate success of blinding at the end of trial, use adequate concealment of allocation, experimental and sham interventions that are close to indistinguishable and an intention-to-treat analysis. Evaluation whether effects increase with longer stimulation times per session or with type of ultrasound is recommended.
ACKNOWLEDGEMENTS
Vivian Welch, Lucie Brosseau, Joan Peterson, Beverley Shea, Peter Tugwell and George A. Wells were authors on the original review. We thank the Cochrane Musculoskeletal editorial team for valuable comments and Malcolm Sturdy for database support. The authors are grateful to Judith Falconer for providing additional information concerning design and for trying to locate les of her trial published in 1992.
AUTHORS CONCLUSIONS
REFERENCES
References to studies included in this review

Cetin 2008 {published data only} Cetin N, Aytar A, Atalay A, Akman MN. Comparing hot pack, short-wave diathermy, ultrasound, and TENS on isokinetic strength, pain, and functional status of women with osteoarthritic knees: a single-blind, randomized, controlled trial. American Journal of Physical Medicine & Rehabilitation 2008;87:44351. Falconer 1992 {published data only} Falconer J, Hayes KW, Change RW. Effect of ultrasound on mobility in osteoarthritis of the knee. Arthritis Care and Research 1992;5(1):2935.
Huang 2005 A {published data only} Huang MH, Yang RC, Lee CL, Chen TW, Wang MC, Huang MH, et al.Preliminary results of integrated therapy for patients with knee osteoarthritis. Arthritis and rheumatism 2005;53:81220. Huang 2005 B {published data only} Huang MH, Lin YS, Lee CL, Yang RC, Huang MH, Lin YS, et al.Use of ultrasound to increase effectiveness of isokinetic exercise for knee osteoarthritis. Archives of Physical Medicine & Rehabilitation 2005;86:154551. zgnenel 2009 {published data only} zgnenel L, Aytekin E, Durmuolu G. A double-blind trial of clinical effects of therapeutic ultrasound in knee osteoarthritis. Ultrasound in medicine & biology 2009;35(1):449.
21
References to studies excluded from this review

Antich 1986 {published data only} Antich TJ, Randall CC, Westbrook RA, Morrissey MC, Brewster CE. Physical therapy treatment of knee extensor mechanism disorders: comparison of four treatment modalities. The Journal of Orthopaedic and Sports Physical Therapy 1986;8:25559. Bansil 1975 {published data only} Bansil CK, Joshi JB. Effectiveness of shortwave diathermy and ultrasound in the treatment of osteoarthritis of the knee joint. Medical journal of Zambia 1975;9(5):1389. Esmat 1975 {published data only} Esmat N. Treatment of arthrosis deformans by simultaneous application of interferential current and ultrasonic waves. The Journal of the Egyptian Medical Association 1975;58:32833. Jan 1991 {published data only} Jan MH, Lai JS. The effects of physiotherapy on osteoarthritic knees of females. Journal Formosan Medical Association 1991;90:100813. Jones 2004 {published data only} Jones JP, Bae YK. Ultrasonic imaging, characterization, and stimulation of acupuncture points. Journal of Alternative & Complementary Medicine. 2004; Vol. 10, issue 1:202-203 (O-9). Lindahl 1952 {published data only} Lindahl O. Clinical effect of ultrasonic waves in chondromalacic and osteoarthritic changes in the knee. Rheumatism 1952;8(2): 3637. Machalett 1975 {published data only} Machalett H. Questions of therapy and morbidity of the bony framework and the locomotor apparatus in a rural practice. Zeitschrift fr rztliche Fortbildung 1975;69:4345. Svarcova 1988 {published data only} Svarcova J, Trnavsky K, Zvarova J. The inuence of ultrasound, galvanic currents and shortwave diathermy on pain intensity in patients with osteoarthritis. Scandinavian journal of rheumatology 1988;Suppl. 67:8385. Zielke 1987 {published data only} Zielke A. Physical therapy in degenerative diseases of the joints. Zeitschrift fr die gesamte innere Medizin und ihre Grenzgebiete 1987;42:1058.
Bellamy 1995 Bellamy N. Outcome measurement in osteoarthritis clinical trials. J Rheumatol 1995;22(suppl.43):4951. Chinn 2000 Chinn S. A simple method for converting an odds ratio to effect size for use in meta-analysis. Statistics in Medicine 2000;19(22): 312731. Clegg 2006 Clegg DO, Reda DJ, Harris CL, Klein MA, ODell JR, Hooper MM, et al.Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. New England Journal of Medicine 2006;354(8):795808. Cohen, 1988 Cohen, J. (1988). Statistical power analysis for the behavioral sciences. 2nd Edition. Hillsdale, NJ: Lawrence Earlbaum Associates, 1988. Danova 1996 Danova Y, Boncheva I, Dikova M, Popov V, Lozanova M. Ultraphonophoretic application of voltaren emulgel in patients with knee-joint osteoarthrosis [Bulgarian]. Fizikalna Kurortna i Rekhabilitatsionna Meditsina 1996;35(1):1820. DerSimonian 1986 DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled clinical trials 1986;7(3):17788. [PUBMED: 3802833] Dickersin 1994 Dickersin K, Scherer R, Lefebvre C. Identifying relevant studies for systematic reviews. BMJ 1994;309(6964):128691. Dincer 2008 Dincer U, Cakar E, Ozdemir B, Kiralp MZ, Dursun H. Comparison of effects of combined physical therapy program and exercise on corrupted balance functions in patient with knee bilateral osteoarthritis [Turkish]. Romatizma 2008;23(1):913. Egger 2001 Egger M, Smith GD. Principles of and procedures for systematic reviews. In: Egger M, Smith GD, Altman DG editor(s). Systematic Reviews in Health Care: Meta-Analyis in Context. London: BMJ Books, 2001:2342. Egger 2003 Egger M, Juni P, Bartlett C, Holenstein F, Sterne J. How important are comprehensive literature searches and the assessment of trial quality in systematic reviews? Empirical study. Health technology assessment (Winchester, England) 2003;7(1):176. Guyatt 2008 Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, AlonsoCoello P, et al.GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336(7650): 9246. Hartley 1993 Hartley A. Therapeutic Ultrasound. Anne Hartley Agency. 2. Etobicoke, Ontario, Canada: Anne Hartley Agency, 1993. Hicks 1990 Hicks JE. Exercise in patients with inammatory arthritis and connective tissue disease. Rheumatic diseases clinics of North America 1990;16(4):84570.
22
References to studies awaiting assessment

Kostadinov 1978 {published data only} Kostadinov D, Mihailov S, Kalaidjiev G. Kurortologiya i. Comparative investigations on gonarthrosis treatment with some physical factors. [Bulgarian]. Fizioterapiya 1978;15:16974.
Additional references
Altman 1996 Altman R, Brandt K, Hochberg M, Moskowitz R, Bellamy N, Bloch DA, et al.Design and conduct of clinical trials in patients with osteoarthritis: recommendations from a task force of the Osteoarthritis Research Society. Results from a workshop. Osteoarthritis Cartilage 1996;4(4):21743.
Higgins 2003 Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327(7414):55760. Higgins 2008 Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.0.1 [updated September 2008]. The Cochrane Collaboration, 2008. Available from www.cochranehandbook.org. Jni 2001 Jni P, Altman DG, Egger M. Systematic reviews in health care: Assessing the quality of controlled clinical trials. BMJ 2001;323 (7303):426. [PUBMED: 11440947] Jni 2006 Jni P, Reichenbach S, Dieppe P. Osteoarthritis: rational approach to treating the individual. Best practice & research. Clinical rheumatology 2006;20(4):72140. Kalpakcioglu 2006 Kalpakcioglu BA, Cakmak B, Bahadir C. Comparison of ultrasound and short wave diathermy therapy in knee osteoarthritis [Turkish]. Turkiye Fiziksel Tip ve Rehabilitasyon Dergisi 2006;52(4):16873. Kozanoglu 2003 Kozanoglu E, Basaran S, Guzel R, Guler-Uysal F. Short term efcacy of ibuprofen phonophoresis versus continuous ultrasound therapy in knee osteoarthritis. Swiss Medical Weekly 2003;133: 3338. Nelson 1999 Nelson RM, Hayes KW, Currier DP. Clinical Electrotherapy. Third Edition. Appleton & Rouge, 1999. Nesch 2009 Nesch E, Trelle S, Reichenbach S, Rutjes AW, Brgi E, Scherer M, et al.The effects of the exclusion of patients from the analysis in randomised controlled trials: meta-epidemiological study. BMJ 2009;339:b3244. Ones 2006 Ones K, Tetik S, Tetik C, Ones N. The effects of heat on osteoarthritis of the knee. The Pain Clinic 2006;18(1):6775. Pham 2004 Pham T, van der Heijde D, Altman RD, Anderson JJ, Bellamy N, Hochberg M, et al.OMERACT-OARSI initiative: Osteoarthritis Research Society International set of responder criteria for osteoarthritis clinical trials revisited. Osteoarthritis Cartilage 2004; 12(5):38999. Qin 2008 Qin XY, Li XX, Berghea F, Suteanu S, Qin X-Y, Li X-X, et al.Comparative study on Chinese medicine and western medicine for treatment of osteoarthritis of the knee in Caucasian patients. Zhongguo Zhenjiu 2008;28(6):45962. Rand 2007 Rand SE, Goerlich C, Marchand K, Jablecki N. The physical therapy prescription. American family physician 2007;76(11): 16616. [PUBMED: 18092708] Reichenbach 2007 Reichenbach S, Sterchi R, Scherer M, Trelle S, Brgi E, Brgi U, et al.Meta-analysis: chondroitin for osteoarthritis of the knee or hip. Annals of internal medicine 2007;146(8):58090.
Rcker 2008 Rcker G, Schwarzer G, Carpenter JR, Schumacher M. Undue reliance on I2 in assessing heterogeneity may mislead. BMC medical research methodology 2008;8(1):79. [PUBMED: 19036172] Solomon 1997 Solomon L. Clinical features of osteoarthritis. In: Kelly WN, Harris ED Jr, Ruddy S, Sledge CB editor(s). Textbook of Rheumatology. 5th Edition. Vol. 2, Philadelphia: WB Saunders, 1997:138393. Srbely 2008 Srbely JZ. Ultrasound in the management of osteoarthritis: part I: a review of the current literature. Journal of the Canadian Chiropractic Association 2008;52:3038. Sterne 2001 Sterne JA, Egger M. Funnel plots for detecting bias in metaanalysis: guidelines on choice of axis. Journal of Clinical Epidemiology 2001;54(10):104655. Tendal 2009 Tendal B, Higgins JP, Jni P, Hrbjartsson A, Trelle S, Nesch E, et al.Disagreements in meta-analyses using outcomes measured on continuous or rating scales: observer agreement study. BMJ 2009; 339:b3128. Tuzun 2003 Tuzun EH, Otman S, Kirdi N. Comparison of different methods of pulsed shortwave diathermy in knee osteoarthritis. The Pain Clinic 2003;15:42127. Veenhof 2007 Veenhof C, Chm, Dekker J, Koke AJA, Oostendorp RA, Bijlsma JWJ. Which patients with osteoarthritis of hip and/or knee benet most from behavioral graded activity?. International Journal of Behavioral Medicine 2007;14:8691. Wood 2008 Wood L, Egger M, Gluud LL, Schulz KF, Jni P, Altman DG, et al.Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: metaepidemiological study. BMJ (Clinical research ed.) 2008;336(7644): 6015. [PUBMED: 18316340] Yang 2005 Yang DJ, Xu FY, Ma L, Gan JH. Ultrasonic wave in combination with quadriceps exercise for the treatment of senile knee osteoarthritis. Zhongguo Linchuang Kangfu [Chinese Journal of Clinical Rehabilitation] 2005;9:25254. Zhang 2008 Zhang W, Moskowitz RW, Nuki G, Abramson S, Altman RD, Arden N, et al.OARSI recommendations for the management of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis and cartilage / OARS, Osteoarthritis Research Society 2008;16(2):13762. [PUBMED: 18279766]
References to other published versions of this review

Welch 2001 Welch V, Brosseau L, Peterson J, Shea B, Tugwell P, Wells G. Therapeutic ultrasound for osteoarthritis of the knee. Cochrane Database of Systematic Reviews 2001, Issue 3. [DOI: 10.1002/ 14651858.CD003132]
23
Indicates the major publication for the study
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Cetin 2008 Methods Randomised controlled trial 5-arm parallel group design Trial duration: 8 weeks No power calculation reported Funding by non-prot organisation: no information provided 100 patients randomised* 100 patients with knee OA reported at baseline Study joints: 100 knees Number of females: 100 of 100 (100%) Average age: 60 years Average BMI: 28 kg/m2 Average disease duration: not reported Severity**: according to Kellgren and Lawrence: 3 patiens with grade I, 16 with II, 16 with grade III, and 5 with grade IV Experimental intervention: Ultrasound therapy + hot packs + isokinetic exercise, 3 times per week Control intervention: hot packs + isokinetic exercise, 3 times per week Duration of treatment period: 8 weeks Analgesics allowed, unclear whether intake was similar between groups. Device: Sonopuls 590 US (Enraf Nonius) Type: continuous US Duration of stimulation per session: 10 minutes, minutes per treated region not reported Total sessions: 24 Frequency: 1.00 MHz Intensity: 1.5 W/cm2 Head size (cm2): not reported Treated anatomical areas: around the knee, with lower extremity in maximal extension Area treated per session: not reported Skin preparation: not reported (direct contact) Safety precautions: not reported Extracted pain outcome: Pain on walking after 8 weeks (VAS), described as Knee pain severity after a 50-m walk Extracted function outcome: Lequesne OA index global score (Likert) after 8 weeks No primary outcome reported *Only 2 arms (40 patients) qualied for inclusion in this review. In total, 100 patients were equally randomised in 5 groups: 1) short wave diathermy + hot packs + isokinetic
Participants
Interventions
Outcomes
Notes
24
Cetin 2008
(Continued)
exercise, 2) transcutaneous electrical stimulation + hot packs + isokinetic exercise, 3) Ultrasound therapy + hot packs + isokinetic exercise, 4) hot packs + isokinetic exercise, 5) isokinetic exercise. ** numbers applicable to the 40 patients included in the review Risk of bias Item Adequate sequence generation? Allocation concealment? Adequate blinding of patients? Authors judgement Unclear Unclear No Description No information provided. No information provided. No sham intervention. For safety reasons, the treating health care provider must be aware of the type of ultrasound therapy given. Although the investigators wrote that Patients were evaluated at baseline and at the end of the treatment sessions by the physician, who was blinded with regard to the type of treatment the patients would receive, VAS and Lequesne are likely to be self reported by patients, who were not blinded. No information provided.
Adequate blinding of physicians/physical No therapists?
Adequate blinding of outcome assessors?
Unclear
Intention-to-treat analysis performed? Pain Intention-to-treat analysis performed? Function
Unclear
Unclear
No information provided.
Funding by commercial organisation Unclear avoided? Falconer 1992 Methods
No information provided.
Randomised, double-blind trial 2-arm parallel group design Trial duration: 14.6 weeks Power calculation reported Funding by non-prot organisation: Arthritis Foundation, Arthritis Health Prof. Association, NIAMS Grant number AM30692 74 patients randomised Number of patients with knee OA not reported at baseline
25
Participants
Falconer 1992
(Continued)
Study joints analysed: 69 Number of females: 50 of 69 (72%) Average age: 68 years Average BMI: not reported Average disease duration: 6.95 years (US: 6.4, Control: 7.5) Severity: moderate knee OA, Altman grade II, 8 patients had history of total knee replacement Interventions Experimental intervention: ultrasound therapy, 2-3 times per week Control intervention: sham ultrasound therapy, 2-3 times per week Concurrent Treatment (30 minutes): passive stretch during cooling period, manual therapy, active ROM and isometric strengthening exercises, instruction in daily home exercise, joint protection. Duration of treatment period: 6 weeks Unclear whether analgesics were allowed and the intake was assessed Device: Sonopulus 590 (Enraf Nonius) Type: not reported, but in light of the maximal intensity reported, most likely pulsed ultrasound therapy Duration of stimulation per session: 12 minutes, 3 minutes per treated region Total sessions: 12 Frequency: 1.00 MHz Intensity: mean 1,7 W/cm2, increased to maximum tolerable dosage, at which warmth without pain was felt, not exceeding 2.5 W/cm2 Head size (cm2): 10 (effective radiating surface 8.5) Treated anatomical areas: anterior, posterior, medial, lateral, with lower extremity in maximal exion or extension, depending on which end range presented with the most functional loss Area treated per session: 100cm2 Skin preparation: aqueous gel Safety precautions: power increased in 0.1W/cm2 increments to a maximum of 2.5W/cm2 Extracted pain outcome: Global pain at 6 weeks, described as pain, measured on 10 cm VAS. Extracted function outcome*: Walking disability at 6 weeks, described as gait velocity. Primary outcome: Pain and active range of motion *not enough data presented to allow pooling, it was only stated that no signicant differences were observed between the experimental and control group. Despite a great effort, Dr. Falconer was unable to locate the original data le and to provide additional data.
Outcomes
Notes
Risk of bias Item Adequate sequence generation? Authors judgement Unclear Description No information provided. In personal communication Dr Falconer indicated that permuted block randomisation was used.
26
Falconer 1992
(Continued)
Allocation concealment? Adequate blinding of patients?
Unclear Yes
No information provided. Sham device: identical in appearance, dials were lit and audible beeps were given at each sham increase in dosage. No energy delivered to tissue. For safety reasons, the treating health care provider must be aware of the type of ultrasound therapy given.
Adequate blinding of outcome assessors? Intention-to-treat analysis performed? Pain
Yes No 34 out of 37 (92%) randomised to experimental and 35 out of 37 (95%) randomised to control group were analysed. Not applicable, trial did not contribute to function meta-analyses. No information provided.
Intention-to-treat analysis performed? Function
No
Funding by commercial organisation Unclear avoided? Huang 2005 A Methods
Randomised trial 4-arm parallel group design Trial duration: 8 weeks No power calculation reported Funding by non-prot organisation: Supported by a project grant provided from the National Science Council of Taiwan 140 patients randomised* 140 patients with bilateral knee OA reported at baseline Study joints: 280 knees for pain outcome Number of females: 113 of 140 (81%) Average age: 65 years Average BMI: not reported Average disease duration: not reported, ranging from 5 months to 12 years Severity: moderate knee OA, Altman grade II Experimental intervention: Ultrasound therapy, isokinetic exercise, hot packs, three times per week Control intervention: isokinetic exercise and hot packs, three times per week Duration of treatment period: 8 weeks Unclear whether analgesics were allowed and the intake was assessed Device: Sonopulus 590 (Enraf Nonius) Duration of stimulation per session: unclear, 5 minutes per treated to each treated region
27
Participants
Interventions
Huang 2005 A
(Continued)
Total sessions: 24 planned Frequency: 1.00 MHz Intensity: up to a level at which a warm sensation or a mild sting was felt, not exceeding 2.5 W/cm2 Head size (cm2): not reported Treated anatomical areas: medial collateral ligament, anserine bursa, and the popliteal fossa tender points (5min/area), with lower extremity in 90 exion for rst 2 areas and maximal extension for popliteal fossa points Area treated per session: 25cm2 Skin prep: not reported Safety precautions: none reported Outcomes Extracted pain outcome: Pain on walking at 8 weeks, described as severity of knee pain while remaining in a weight-bearing position (walking or standing) for 5 minutes, measured on 10 cm VAS. Extracted function outcome: Lequesnes Index after 8 weeks, measured on 0 to 26 Likert scale. Primary outcome: not reported *2 arms (70 patients) were excluded from the review: no intervention arm (35 patients) and isokinetic exercise + pulsed ultrasound + intraarticular hyaluronan + hot packs arm (35 patients).
Notes
Risk of bias Item Adequate sequence generation? Allocation concealment? Authors judgement Unclear Unclear Description No information provided. Although authors wrote that they used sequentially numbered opaque sealed envelopes, we had no evidence to suggest that the envelopes were appropriately numbered in a sequential manner and random allocation could therefore have been undermined. Quote: and randomly assigned to 4 groups (groups I-IV) by a secure system of opaque sealed envelopes that were sequentially numbered I-IV. The doctor who assigned the patients was blinded to the treatment the patients would receive. With 140 patients randomised, the envelopes should have been sequentially numbered 1-140 and not I to IV. No sham intervention.
Adequate blinding of patients?
No
28
Huang 2005 A
(Continued)
Adequate blinding of physicians/physical No therapists? Adequate blinding of outcome assessors? Unclear
No sham intervention.
Quote: All the evaluations were performed by the same physiatrists who were also blinded to the treatment the patients received. We scored unclear for both pain and function, however, as a VAS was used for pain, and Lequesne for function that are usually scored by the patient, who was not blinded. 64 out of 70 (91%) randomised to experimental and 60 out of 70 (86%) randomised to control group were analysed (analysis on knee level). 32 out of 35 (91%) randomised to experimental and 30 out of 35 (86%) randomised to control group were analysed (analysis on patient level). No information provided.
Intention-to-treat analysis performed? Pain
No
No
Funding by commercial organisation Unclear avoided? Huang 2005 B Methods
Randomised trial 4-arm parallel group design Trial duration: 8 weeks No power calculation reported Funding by non-prot organisation: Supported by a project grant provided from the National Science Council of Taiwan (grant number NSC-92-2314-B-037-067) 120 patients randomised* 120 patients with bilateral knee OA reported at baseline Study joints: 240 knees for pain outcome Number of females: 97 of 120 (81%) Average age: 62 years Average BMI: not reported Average disease duration: not reported, ranging from 6 months to 11 years Severity: moderate knee OA, Altman grade II Comparison 1: Experimental intervention: continuous ultrasound therapy, isokinetic exercise, hot packs, three times per week Control intervention: sham ultrasound therapy, isokinetic exercise and hot packs, three times per week
29
Participants
Interventions
Huang 2005 B
(Continued)
Comparison 2: Experimental intervention: pulsed ultrasound therapy, isokinetic exercise, hot packs, three times per week Control intervention: sham ultrasound therapy, isokinetic exercise and hot packs, three times per week Duration of treatment period: 8 weeks Unclear whether analgesics were allowed and the intake was assessed Device: Sonopulus 590 (Enraf Nonius) Duration of stimulation per session: unclear, 5 minutes per treated to each treated region Total sessions: 24 planned Frequency: 1.00 MHz Intensity: up to a level at which a warm sensation or a mild sting was felt, not exceeding 2.5 W/cm2 Head size (cm2): not reported Treated anatomical areas: medial collateral ligament, anserine bursa, and the popliteal fossa tender points (5min/area), with lower extremity in 90 exion for rst 2 areas and maximal extension for popliteal fossa points Area treated per session: 25cm2 Skin prep: not reported Safety precautions: none reported Outcomes Extracted pain outcome: Pain on walking at 8 weeks, described as severity of knee pain while remaining in a weight-bearing position (walking or standing) for 5 minutes in the parallel bars of the treadmill, measured on 10 cm VAS. Extracted function outcome: Lequesnes Index after 8 weeks, measured on 0 to 26 Likert scale. Primary outcome: Change in ambulation speed and Lequesne index *1 arm (30 patients), in which no intervention was given, was excluded from the review
Notes Risk of bias Item Adequate sequence generation? Allocation concealment?
Authors judgement Unclear Unclear
Description No information provided. Although authors wrote that they used sequentially numbered opaque sealed envelopes, we had no evidence to suggest that the envelopes were appropriately numbered in a sequential manner and random allocation could therefore have been undermined. Quote: they were randomly assigned to 4 groups by a secure system of sequentially numbered I through IV opaque sealed envelopes. The physician who assigned the patients was blinded as to the treatment they would receive.
30
Huang 2005 B
(Continued)
With 120 patients randomised, the envelopes should have been sequentially numbered 1-120 and not I to IV. Adequate blinding of patients? No Sham intervention given in the control group, but the investigators stated (...) however, the absence of a warm or stinging sensation for the treated patients caused the failure of the sham US application. For safety reasons, the treating health care provider must be aware of the type of ultrasound therapy given. For Pain: VAS is usually scored by patients, who were not blinded. For function: no information provided. Comparison I: 54 out of 60 (90%) randomised to experimental and 50 out of 60 (83%) randomised to control group were analysed. Comparison II: 60 out of 60 (100%) randomised to experimental and 50 out of 60 (83%) randomised to control group were analysed. Analyses performed on knee level. Comparison I: 27 out of 30 (90%) randomised to experimental and 25 out of 30 (83%) randomised to control group were analysed. Comparison II: 30 out of 30 (100%) randomised to experimental and 25 out of 30 (83%) randomised to control group were analysed. Analyses performed on patient level. Treatment equipment supplied by Chattecx group, ATL ultrasound and Enraf Nonius.
No
No
No
Funding by commercial organisation Unclear avoided?
31
zgnenel 2009 Methods Randomised trial 2-arm parallel group design Trial duration: 2 weeks Power calculation reported Funding by non-prot organisation: No information provided 67 patients randomised 67 patients with newly diagnosed knee OA reported at baseline Study joints: 67 knees Number of females: 54 of 67 (81%) Average age: 55 years Average BMI: 32 kg/m2 Average disease duration: not reported, but all patients were newly diagnosed with OA Severity: knee pain and limitation on most days of the past 6 months and KellgrenLawrence score II (31 patients) to III (36 patients) Experimental intervention: continuous ultrasound therapy, ve times per week Control intervention: sham ultrasound therapy, ve times per week Duration of treatment period: 2 weeks Analgesics were not allowed, two patients were dropped out from the control group due to analgesic intake Device: Petson .250 ultrasound equipment Duration of stimulation per session: 5 minutes Total sessions: 10 Frequency: 1.00 MHz Intensity: 1 W/cm2 Head size: 4 cm Treated anatomical areas: patellofemoral and tibiofemoral borders of the target knee on both the lateral and medial margins, avoiding the patella Area treated per session: 25cm2 Skin prep: Aqueous gel Safety precautions: circular movements Extracted pain outcome: WOMAC pain subscale at 2 weeks, 2 days after end of last treatment, measured on 0 to 20 Likert scale Extracted function outcome: WOMAC physical function subscale at 2 weeks, 2 days after end of last treatment, measured on 0 to 68 Likert scale Primary outcome reported: knee pain on movement over the past week, measured on 10 cm VAS
Participants
Interventions
Outcomes
Notes Risk of bias Item Adequate sequence generation? Authors judgement Unclear Description No information provided.
32
zgnenel 2009
(Continued)
Allocation concealment?
Unclear
Authors merely state that An independent researcher not involved in the data assessment randomized the subjects.. Sham device: identical in appearance. Quotes: An applicator disconnected from the back to working ultrasound machine. Patient in position unable to see whether the cable was disconnected or not. For safety reasons, the treating health care provider must be aware of the type of ultrasound therapy given. Quote: Both the therapeutic US and the sham US application were performed by the same therapist and patients were assessed by three blinded researchers before and at the end of therapy program. Pain and function were measured on WOMAC subscales, that are usually scored by the patient. In this trial, patients were blinded. No information provided. It was only reported that In the placebo group, two patients used analgesics because of increased pain and were thus dropped out of the study.. No information provided. It was only reported that In the placebo group, two patients used analgesics because of increased pain and were thus dropped out of the study.. No information provided.
Adequate blinding of patients?
Yes
Yes
Unclear
Unclear
Funding by commercial organisation Unclear avoided?
33
Characteristics of excluded studies [ordered by study ID]
Antich 1986
Not osteoarthritis but several types of knee extensor mechanism disorders described as chondromalacia patella, infrapatellar tendinitis and peripatellar pain. Additional description: RCT with 4-arm parallel group design comparing ice, phonophoresis, iontophoresis and ultrasound therapy + ice. Use of active control group. Additional description: RCT with 2 arm parallel group design comparing ultrasound therapy with shortwave diathermy. Use of active control group. Additional description: 3 arm parallel group design comparing ultrasound therapy to interferential current stimulation or ultrasound therapy with interferential current stimulation. Unclear whether allocation of treatment was at random. Use of active control group. Additional description: RCT with 4 arm parallel group design comparing ultrasound therapy versus short wave diathermy or ultrasound therapy + exercise or diathermy + exercise. No relevant outcome reported. Additional description: study design unclear and unclear whether people with osteoarthritis were included. Unable to extract data for osteoarthritis. Additional description: RCT, parallel design, two arms comparing ultrasound therapy versus sham ultrasound therapy in people with chondromalacic and osteoarthritic knee conditions. Results presented jointly. No contact could be established with the author. Not a randomised study (historical case series) Use of active control groups. Additional description: controlled trial with groups receiving either Galvanic electrostimulation or ultrasound therapy or pulsed short wave diathermy. Not a randomised study. Additional description: report summarising percentages of any type of active or passive treatment option given to patients with osteoarthritis during a course of treatment in the state health resorts Bad Brambach--Bad Elster in 1979 to 1981.
Bansil 1975
Esmat 1975
Jan 1991
Jones 2004
Lindahl 1952
Machalett 1975 Svarcova 1988
Zielke 1987
Characteristics of studies awaiting assessment [ordered by study ID]

Kostadinov 1978 Methods Most likely comparative study with historical controls or cohort study. Quote: An attempt is made for comparative assessment of the most commonly used physical factors in the treatment of gonarthrosis patients. The objects of the investigation were the widely used (in Bulgaria) physical factors (...) 427 in- and outpatients with osteoarthritis of the knee Mineral-oil electrophoresis and low frequency impulse currents Device: Bipulsator, Magnetimpulsator and a bipolar electrostimulator Interferential current (electro)stimulation Therapeutic Ultrasound Phonophoresis with hydrocortisone
34
Participants Interventions
Kostadinov 1978
(Continued)
Electrophoresis with KJ and novocaine Outcomes Awaiting translation Function was described as locomotion Publication in Bulgarian. We have not yet been able to involve a Bulgarian translator.
Notes
35
DATA AND ANALYSES
Comparison 1. Any type of therapeutic ultrasound versus control

No. of studies 5 4 1 1 No. of participants 320 251 67 67
Outcome or subgroup title 1 Pain 2 Function 3 Number of patients experiencing any adverse event 4 Number of patients withdrawn or dropped out due to adverse events 5 Number of patients experiencing any serious adverse event
Statistical method Std. Mean Difference (Random, 95% CI) Std. Mean Difference (Random, 95% CI) Risk Ratio (IV, Fixed, 95% CI) Risk Ratio (IV, Fixed, 95% CI)
Effect size -0.49 [-0.76, -0.23] -0.64 [-1.42, 0.14] Not estimable Not estimable
136
Risk Ratio (IV, Random, 95% CI)
Not estimable
Analysis 1.1. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 1 Pain.
Review: Therapeutic ultrasound for osteoarthritis of the knee or hip
Comparison: 1 Any type of therapeutic ultrasound versus control Outcome: 1 Pain
Study or subgroup
Experimental N
Control N 20 35 30 25 33
Std. Mean Difference (SE)
Std. Mean Difference IV,Random,95% CI
Weight
Cetin 2008 Falconer 1992 Huang 2005 A Huang 2005 B zgnenel 2009
20 34 32 57 34
0.05 (0.32) -0.45 (0.24) -0.81 (0.26) -0.72 (0.25) -0.38 (0.25)
14.6 % 22.6 % 20.1 % 21.3 % 21.3 %
0.05 [ -0.58, 0.68 ] -0.45 [ -0.92, 0.02 ] -0.81 [ -1.32, -0.30 ] -0.72 [ -1.21, -0.23 ] -0.38 [ -0.87, 0.11 ]
Total (95% CI)

Heterogeneity: Tau2 = 0.02; Chi2 = 5.42, df = 4 (P = 0.25); I2 =26% Test for overall effect: Z = 3.62 (P = 0.00030)
100.0 %
-0.49 [ -0.76, -0.23 ]
-4
-2
Favours ultrasound
Favours control
36
Analysis 1.2. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 2 Function.
Comparison: 1 Any type of therapeutic ultrasound versus control Outcome: 2 Function
Study or subgroup
Ultrasound N
Control N 20 30 25 33
Std. Mean Difference (SE)
Weight
Cetin 2008 Huang 2005 A Huang 2005 B zgnenel 2009
20 32 57 34
0.33 (0.32) -1.69 (0.3) -0.9 (0.25) -0.27 (0.25)
24.1 % 24.6 % 25.7 % 25.7 %
0.33 [ -0.30, 0.96 ] -1.69 [ -2.28, -1.10 ] -0.90 [ -1.39, -0.41 ] -0.27 [ -0.76, 0.22 ]
Total (95% CI)

Heterogeneity: Tau2 = 0.56; Chi2 = 24.71, df = 3 (P = 0.00002); I2 =88% Test for overall effect: Z = 1.60 (P = 0.11)
100.0 %
-0.64 [ -1.42, 0.14 ]
-4
-2
Favours ultrasound
Favours control
Analysis 1.3. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 3 Number of patients experiencing any adverse event.
Comparison: 1 Any type of therapeutic ultrasound versus control Outcome: 3 Number of patients experiencing any adverse event
Study or subgroup
Experimental n/N
Control n/N 0/33
Risk Ratio IV,Fixed,95% CI
Risk Ratio IV,Fixed,95% CI 0.0 [ 0.0, 0.0 ]
zgnenel 2009
0/34
Total (95% CI)

Total events: 0 (Experimental), 0 (Control) Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P < 0.00001)
34
33
0.0 [ 0.0, 0.0 ]
0.01
0.1
10
100
Favours experimental
Favours control
37
Analysis 1.4. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 4 Number of patients withdrawn or dropped out due to adverse events.
Comparison: 1 Any type of therapeutic ultrasound versus control Outcome: 4 Number of patients withdrawn or dropped out due to adverse events
Study or subgroup
Experimental n/N
Control n/N 0/33
Risk Ratio IV,Fixed,95% CI
Risk Ratio IV,Fixed,95% CI 0.0 [ 0.0, 0.0 ]
zgnenel 2009
0/34
Total (95% CI)

Total events: 0 (Experimental), 0 (Control) Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P < 0.00001)
34
33
0.0 [ 0.0, 0.0 ]
0.01
0.1
10
100
Favours control
Analysis 1.5. Comparison 1 Any type of therapeutic ultrasound versus control, Outcome 5 Number of patients experiencing any serious adverse event.
Comparison: 1 Any type of therapeutic ultrasound versus control Outcome: 5 Number of patients experiencing any serious adverse event
Study or subgroup
Experimental n/N
Control n/N 0/35 0/33
Risk Ratio IV,Random,95% CI
Risk Ratio IV,Random,95% CI 0.0 [ 0.0, 0.0 ] 0.0 [ 0.0, 0.0 ]
Falconer 1992 zgnenel 2009
0/34 0/34
Total (95% CI)

Total events: 0 (Experimental), 0 (Control)
68
68
0.0 [ 0.0, 0.0 ]
Heterogeneity: Tau2 = ; Chi2 = 0.0, df = 0 (P<0.00001); I2 =0.0% Test for overall effect: Z = 0.0 (P < 0.00001)
0.01
0.1
10
100
Favours control
38
APPENDICES Appendix 1. MEDLINE, EMBASE and Cinahl search strategy
OVID MEDLINE Search terms for design 1. randomized controlled trial.pt. 2. controlled clinical trial.pt. 3. randomized controlled trial.sh. 4. random allocation.sh. 5. double blind method.sh. 6. single blind method.sh. 7. clinical trial.pt. 8. exp clinical trial/ 9. (clin$ adj25 trial$).ti,ab. 10. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. 11. placebos.sh. 12. placebo$.ti,ab. 13. random$.ti,ab. 14. research design.sh. 15. comparative study.sh. 16. exp evaluation studies/ 17. follow up studies.sh. 18. prospective studies.sh. 19. (control$ or prospectiv$ or volunteer$).ti,ab.
OVID EMBASE Search terms for design 1. randomized controlled trial.sh. 2. randomization.sh. 3. double blind procedure.sh. 4. single blind procedure.sh. 5. exp clinical trials/ 6. (clin$ adj25 trial$).ti,ab. 7. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. 8. placebo.sh. 9. placebo$.ti,ab. 10. random$.ti,ab. 11. methodology.sh. 12. comparative study.sh. 13. exp evaluation studies/ 14. follow up.sh. 15. prospective study.sh. 16. (control$ or prospectiv$ or volunteer$).ti,ab.
CINAHL through EBSCOhost Search terms for design 1. (MH Clinical Trials+) 2. (MH Random Assignment) 3. (MH Double-Blind Studies) or (MH Single-Blind Studies) 4. TX (clin$ n25 trial$) 5. TX (sing$ n25 blind$) 6. TX (sing$ n25 mask$) 7. TX (doubl$ n25 blind$) 8. TX (doubl$ n25 mask$) 9. TX (trebl$ n25 blind$) 10. TX (trebl$ n25 mask$) 11. TX (tripl$ n25 blind$) 12. TX (tripl$ n25 mask$) 13. (MH Placebos) 14. TX placebo$ 15. TX random$ 16. (MH Study Design+) 17. (MH Comparative Studies) 18. (MH Evaluation Research) 19. (MH Prospective Studies+) 20. TX (control$ or prospectiv$ or volunteer$) 21. S1 or S2 or (.) or S20 Search terms for Osteoarthritis 22. osteoarthriti$ 23. (MH Osteoarthritis) 24. TX osteoarthro$ 25. TX gonarthriti$ 26. TX gonarthro$ 27. TX coxarthriti$ 28. TX coxarthro$ 29. TX arthros$ 30. TX arthrot$ 31. TX knee$ n3 pain$ 32. TX hip$ n3 pain$ 33. TX joint$ n3 pain$ 34. TX knee$ n3 ach$ 35. TX hip$ n3 ach$ 36. TX joint$ n3 ach$ 37. TX knee$ n3 discomfort$ 38. TX hip$ n3 discomfort$
39
Search terms for Osteoarthritis 20. exp osteoarthritis/ 21. osteoarthriti$.ti,ab,sh. 22. osteoarthro$.ti,ab,sh. 23. gonarthriti$.ti,ab,sh. 24. gonarthro$.ti,ab,sh. 25. coxarthriti$.ti,ab,sh. 26. coxarthro$.ti,ab,sh. 27. arthros$.ti,ab. 28. arthrot$.ti,ab. 29. ((knee$ or hip$ or joint$) adj3 (pain$ or ach$ or discomfort$)).ti,ab. 30. ((knee$ or hip$ or joint$) adj3 stiff$) .ti,ab.
Search terms for Osteoarthritis 17. exp osteoarthritis/ 18. osteoarthriti$.ti,ab,sh. 19. osteoarthro$.ti,ab,sh. 20. gonarthriti$.ti,ab,sh. 21. gonarthro$.ti,ab,sh. 22. coxarthriti$.ti,ab,sh. 23. coxarthro$.ti,ab,sh. 24. arthros$.ti,ab. 25. arthrot$.ti,ab. 26. ((knee$ or hip$ or joint$) adj3 (pain$ or ach$ or discomfort$)).ti,ab. 27. ((knee$ or hip$ or joint$) adj3 stiff$).ti,ab.
(Continued)
39. TX joint$ n3 discomfort$ 40. TX knee$ n3 stiff$ 41. TX hip$ n3 stiff$ 42. TX joint$ n3 stiff$ 43. S22 or S23 or S24.or S42 Search terms for Therapeutic Ultrasound 31. exp ultrasonography/ 32. exp Ultrasonic Therapy/ 33. us.fs. 34. (ultrasound$ or ultrasonic$).tw. 35. short wave therapy.tw. 36. ultrasonograph$.tw. Combining terms 37. or/1-19 38. or/20-30 39. or/31-36 40. and/37-39 41. animal/ 42. animal/ and human/ 43. 41 not 42 44. 40 not 43 Search terms for Therapeutic Ultrasound 28. exp ultrasonography/ 29. exp Ultrasonic Therapy/ 30. (ultrasound$ or ultrasonic$).tw. 31. short wave therapy.tw. 32. ultrasonograph$.tw. Search terms for Therapeutic Ultrasound 44 exp ULTRASONOGRAPHY/ 45 exp Ultrasonic Therapy/ 46 (ultrasound$ or ultrasonic$).tw. 47 short wave therapy.tw. 48 ultrasonograph$.tw. 49 S44 or S45 or (...) or S48 Combining terms S21 and S43 and S49
Combining terms 33. or/1-16 34. or/17-27 35. or/28-32 36. and/33-35 37. animal/ 38. animal/ and human/ 39. 37 not 38 40. 36 not 39
Appendix 2. CENTRAL and PEDro search strategy
CENTRAL Search terms for Osteoarthritis #1. (osteoarthritis* OR osteoarthro* OR gonarthriti* OR gonarthro* OR coxarthriti* OR coxarthro* OR arthros* OR arthrot* OR ((knee* OR hip* OR joint*) near/3 (pain* OR ach* OR discomfort*)) OR ((knee* OR hip* OR joint*) near/3 stiff*)) in Clinical Trials #2. MeSH descriptor Osteoarthritis explode all trees Search terms for TENS #3 short wave therapy in Clinical Trials #4 ultrasonograph in Clinical Trials #5 (ultrasound* or ultrasonic*)in Clinical Trials #6 MeSH descriptor Ultrasonography explode all trees #7 MeSH descriptor Short-Wave Therapy explode all trees Combining terms #8. (#3 OR #4 OR #5 OR #6 OR #7) #9. (#1 OR #2)
PEDro
1. Ultra in title or abstract 2. Method: clinical trial 3. Body part: thigh or hip 4. Body part lower leg or knee Combination 1. and 2. and 3. Combination 1. and 2. and 4. 1. short wave therapy in title or abstract 2. Method: clinical trial 3. Body part: thigh or hip 4. Body part lower leg or knee Combination 1. and 2. and 3. Combination 1. and 2. and 4. Combine all
40
(Continued)
#10. (#8 AND #9) in Clinical Trials
WHATS NEW
Last assessed as up-to-date: 22 July 2009.
23 July 2009 23 July 2009
New search has been performed New citation required and conclusions have changed
4 additional trials included Change in authors and conclusions. We updated the search and applied more strict selection criteria, which resulted in 4 additional trials and the exclusion of 2 studies previously included. We used more detailed quality assessment of component trials; use of end of trial estimates to calculate SMDs; detailed exploration of sources of variation between trials, including concealment of allocation, blinding, intention-totreat analysis, characteristics of electrostimulation, and investigations of funnel plots; use of a random-effects model. The previous version included only one trial comparing therapeutic ultrasound with a sham intervention and found a statistically non-signicant benecial effects on pain and function, whereas this version shows a statistically signicant effect on pain, although we are uncertain about the exact magnitude of the effect. CMSG ID C087-R
1 May 2008
Amended
HISTORY
Review rst published: Issue 3, 2001
2 May 2001
New citation required and conclusions have changed
Substantive amendment
41
CONTRIBUTIONS OF AUTHORS
Study conception: Rutjes, Jni Protocol development: Rutjes, Nesch, Jni Acquisition of data: Rutjes, Nesch, Sterchi Analysis and interpretation of data: Rutjes, Nesch, Jni Drafting of the manuscript: Rutjes, Jni Critical revision of the manuscript for important intellectual content:: Rutjes, Nesch, Sterchi, Jni Statistical analysis. Nesch, Rutjes, Jni Obtained funding: Jni
DECLARATIONS OF INTEREST
None
SOURCES OF SUPPORT Internal sources

Institute of Social and Preventive Medicine, University or Bern, Switzerland. Intramural grants
External sources
Swiss National Science Foundation, Switzerland. National Research Program 53 on musculoskeletal health (grant numbers 4053-40-104762/3 and 3200-066378)
DIFFERENCES BETWEEN PROTOCOL AND REVIEW

Before embarking on this review, we generated a standard protocol for this, and all other Cochrane reviews performed by our group. The protocol was approved by the Editorial Board of the Cochrane Musculoskeletal Review Group (CMSG), but, as an update, did not result in a specic publication in the Cochrane database. Because of the low number of trials identied, we deviated from the standard protocol in terms of omitting several stratied analyses and univariable random-effects meta-regression analyses to determine whether treatment effects were affected by treatment or design characteristics. In addition, we used risk of bias tables to present the methodological quality of included trials and a summary of ndings table to present results.
42
INDEX TERMS Medical Subject Headings (MeSH)

Ultrasonic
Therapy; Clinical Trials as Topic; Osteoarthritis, Knee [ therapy]; Short-Wave Therapy
MeSH check words

Adult; Humans
43

US Cochrane PDF

Diunggah oleh

Informasi Dokumen

Deskripsi Asli:

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

US Cochrane PDF

Diunggah oleh

Hak Cipta:

Format Tersedia

Therapeutic ultrasound for osteoarthritis of the knee or hip (Review)

Rutjes AWS, Nesch E, Sterchi R, Jni P

Therapeutic ultrasound for osteoarthritis of the knee or hip

0 events reported NNH not estimable

0 events reported NNH not estimable

Types of outcome measures

Criteria for considering studies for this review

Search methods for identication of studies

Data collection and analysis

Figure 1. Flow chart

Risk of bias in included studies

No. of patients: No. of patients: experimental control n n

SMD (95% CI)

Table 1. Results of stratied analyses of pain outcomes

-0.49 (-0.76 to - 26% 0.23) 0.8

Table 1. Results of stratied analyses of pain outcomes

-0.45 (-0.93 to N/A 0.03) -0.77 (-1.11 to - 0% 0.41) 0.69

No. of patients: No. of patients: experimental control n 143 n 108

SMD (95% CI) -0.64 to0.14)

Table 2. Results of stratied analyses of function

0 2 89 55 -1.28 (-2.05 to - 76% 0.51)

Table 2. Results of stratied analyses of function

Total number of sessions 10 1 34 33 -0.27 (-0.76 to N/A 0.22)

0 3 109 75 -0.76 (-1.83 to 91% 0.31) 0.70

Duration of treatment period 2 weeks 1 34 33 -0.27 (-0.76 to N/A 0.22)

0 3 109 75 -0.76 (-1.83 to 94% 0.31) 0.78

DISCUSSION Summary of main results

Quality of the evidence

Agreements and disagreements with other studies or reviews

Potential biases in the review process

Implications for practice

Implications for research

References to studies included in this review

References to studies excluded from this review

References to studies awaiting assessment

References to other published versions of this review

Indicates the major publication for the study

Characteristics of included studies [ordered by study ID]

Adequate blinding of physicians/physical No therapists?

Adequate blinding of outcome assessors?

Intention-to-treat analysis performed? Pain Intention-to-treat analysis performed? Function

Funding by commercial organisation Unclear avoided? Falconer 1992 Methods

Allocation concealment? Adequate blinding of patients?

Adequate blinding of physicians/physical No therapists?

Adequate blinding of outcome assessors? Intention-to-treat analysis performed? Pain

Intention-to-treat analysis performed? Function

Funding by commercial organisation Unclear avoided? Huang 2005 A Methods

Adequate blinding of patients?

Adequate blinding of physicians/physical No therapists? Adequate blinding of outcome assessors? Unclear

Intention-to-treat analysis performed? Pain

Intention-to-treat analysis performed? Function

Funding by commercial organisation Unclear avoided? Huang 2005 B Methods

Notes Risk of bias Item Adequate sequence generation? Allocation concealment?

Authors judgement Unclear Unclear

Adequate blinding of physicians/physical No therapists?

Adequate blinding of outcome assessors?

Intention-to-treat analysis performed? Pain

Intention-to-treat analysis performed? Function

Funding by commercial organisation Unclear avoided?

Adequate blinding of patients?

Adequate blinding of physicians/physical No therapists?

Adequate blinding of outcome assessors?