If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address correspondence and reprint requests to: Adalberto Loyola-Sánchez, School of Rehabilitation Science, IAHS – 403, 1400 Main Street West, Hamilton, ON, Canada L8S 1C7. Tel: 1-905-525-9140x26410.
To assess the efficacy of ultrasound therapy (US) for decreasing pain and improving physical function, patient-perception of disease severity, and cartilage repair in people with knee osteoarthritis (OA).
Methods
We conducted a systematic review (to February 2009) without language limits in MEDLINE, EMBASE, Cochrane Library, LILACS, MEDCARIB, CINAHL, PEDro, SPORT-discus, REHABDATA, and World Health Organization Clinical Trial Registry. We included randomized controlled trials of people with knee OA comparing the outcomes of interest for those receiving US with those receiving no US. Two reviewers independently selected studies, extracted relevant data and assessed quality. Pooled analyses were conducted using inverse-variance random effects models.
Main results
Six small trials (378 patients) were included. US improves pain [Standardized Mean Difference (SMD) (95% confidence interval (CI))=−0.49 (−0.79, −0.18), P=0.002], and tends to improve self-reported physical function [SMD (CI)=−0.54 (−1.19, 0.12), P=0.11] along with walking performance [SMD (CI)=0.81 (−0.09, 1.72), P=0.08]. Results from two trials (128 patients), conducted by the same group, show a positive effect of US on pain [SMD (CI)=−0.77 (−1.15, −0.39), P<0.001], self-reported physical function [SMD (CI)=−1.25 (−1.69, −0.81), P<0.001], and walking performance [SMD (CI)=1.47 (1.06, 1.88), P<0.001] at 10 months after the intervention concluded. Heterogeneity observed between studies regarding the effect of US on pain was explained by US dose, mode and intensity. The quality of evidence supporting these effect estimates was rated as low.
Conclusions
US could be efficacious for decreasing pain and may improve physical function in patients with knee OA. The findings of this review should be confirmed using methodologically rigorous and adequately powered clinical trials.
. Therefore, US could be an effective intervention in the management of pain and disability in people with knee OA.
The Osteoarthritis Research Society International (OARSI) struck a committee to complete a systematic review of existing treatment guidelines (2007) in order to develop recommendations for the management of knee and hip OA
. Ultrasound was not identified as a core treatment modality based on the results of a systematic review published in 2001 by the Cochrane Collaboration
while more recent trials evaluating the effectiveness of US in the management of knee OA were not reviewed. Since there is a need for effective conservative treatment options for people with knee OA, it is important to confirm or change current clinical practice guidelines based on best available evidence. Therefore, the objective of this systematic review and meta-analysis was to determine the efficacy of US in decreasing pain and improving physical function in people with knee OA. Further, we extended the scope of previous systematic reviews on the topic by evaluating the efficacy of US on patient-perception of disease severity, and cartilage repair. We followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) recommendations
The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
The Cochrane Central Register of Controlled Trials (fourth Quarter 2008), MEDLINE (1950 to January week 4, 2009), MEDLINE Daily Update (Feb 5, 2009), MEDLINE In Process & Other non-indexed citations (February 5, 2009), EMBASE (1980–2009 week 6), LILACS (February 6, 2009), MEDCARIB (February 6, 2009), CINAHL (February 8, 2009), pre CINAHL (February 8, 2009), PEDro (last updated February 2, 2009), AMED (1985 to January 2009), SPORTdiscus (1830 to February 9, 2009), REHABDATA (1956 to February 9, 2009), and World Health Organization Clinical Trial Registry (February 8, 2009) databases were searched by one of the authors (AL). In addition, published literature with restricted distribution was searched through the ISI Web of Knowledge, Papers First, Proceedings First, and ProQuest for Dissertations and Theses. Authors of an unpublished study reported as a conference proceeding in the Proceedings First database were contacted via e-mail. A detailed example of the full electronic search strategy for Ovid MEDLINE is provided in Appendix A. Briefly, the following medical subject headings (MeSH) were used: Osteoarthritis, Arthritis, Ultrasonic Therapy, Ultrasonics, Sonication, Diathermy, Cartilage, and Wound Healing; keywords were osteoarthritis, arthritis experimental, ultrasound therapy, low intensity pulsed ultrasound, low intensity ultrasound, and cartilage repair. To increase the search sensitivity, no date, language, or design limits were included. Duplicates were removed after all databases were searched.
Study selection and eligibility criteria
Two reviewers (AL, JR) independently screened all citations obtained and retrieved all parallel group randomized controlled trials involving patients with knee OA that compared US with placebo or no intervention. Trials that compared US in combination with another intervention to which the comparison group was exposed were also included. Studies were excluded from the review if phonophoresis was the only ultrasonic intervention, US was combined with another intervention not provided to the comparison group, and samples included subjects having other diagnoses and results for the subjects with knee OA were not reported separately. Cohen’s unweighted Kappa (κ) was used to measure agreement between reviewers
. Disagreement was solved by consensus including a third reviewer (NM). Colleagues translated non-English articles (n=9) written in their first language (a physical therapist, an engineer, an occupational therapist, a rheumatologist, and a physiatrist).
Data extraction and management
Two reviewers, using a pre-tested data collection form, followed a double extraction method. The reviewers independently extracted data related to the study population (clinical setting, diagnostic criteria, joint involvement, sex, age and OA severity), study design, US intervention (device, frequency, mode, intensity and dose), co-interventions, outcomes (pain, physical function, participant’s perception of disease severity, and cartilage repair), and the monitoring/reporting of adverse events. The US dose was calculated using the following formula
We defined cartilage repair as all measures that directly or indirectly assess the cartilage formation–degradation process (i.e., imaging, arthroscopy, fluid biomarkers). Primary authors were contacted in the case of missing data or unclear reporting. A physiatrist with expertise in OA and US (AL) and a physical therapist with expertise in research methodology and OA (NM) reviewed the papers and extracted the data.
Risk of bias and quality assessment
The risk of bias was assessed for each study by evaluating the rigor of the randomization process, the treatment allocation concealment, the blinding process, the completeness of the data, and the reporting of results following the Cochrane Collaboration recommendations
. To judge the completeness of the data as adequate, a dropout rate of less than 15% was required.
Given the objective of this systematic review, we identified randomization, treatment allocation concealment, blinding, and completeness of data as key domains for establishing risk of bias. For each included study, risk of bias was determined to be low when all key domains were performed adequately, unclear when one or more key domains were not clearly described, and high when one or more key domains were inadequate. Across studies, the risk of bias was considered low if all studies had low risk, unclear when more than 75% of the studies had unclear or low risk and less than 25% of the studies had high risk of bias, and high when more than 25% of the studies had high risk of bias. Observed agreement was fair (κ=0.56)
and disagreement was resolved by consensus and inclusion of a third reviewer, who is a physical therapist with research methodology expertise (JR).
The quality of the evidence for each outcome was determined by considering the risk of bias, the heterogeneity of the findings, the use of surrogate measurements for outcome assessment, and the precision of the effect estimates as recommended by the Grading of Recommendations, Assessment Development and Evaluation (GRADE) working group
was used to calculate pooled standardized mean differences (SMDs) using Review Manager (RevMan [computer program] Version 5.0 Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2008, Oxford, UK). The SMD is a ratio between the differences observed between groups and the standard deviation (SD) of the outcomes among participants (Hedges’ adjusted g). The 95% confidence interval (CI) was calculated and a Z test was performed with significance set at P<0.05. Heterogeneity among studies was assessed with a χ2 test with significance set at P<0.10 and an inconsistency test (I2) which represents the percentage variability in the effect estimates that is due to heterogeneity rather than chance. Inconsistency (I2)>40% was considered significantly high
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
reported the results of pain outcomes by knee rather than by patient. Therefore, we included the number of patients (n=35 instead of n=70) in each group during the pooled analysis, in order to account for the intercorrelation of measurements taken from the knees of the same patient.
In order to express the effect estimates as percentage change relative to the control groups, the SMDs were back transformed to mean differences (MD) using SDs reported in observational studies
Relative efficiency and validity properties of a visual analogue vs a categorical scaled version of the western ontario and McMaster universities osteoarthritis (WOMAC) index: Spanish versions.
(Appendix B). Heterogeneity was explored through subgroup analyses following a priori hypotheses that considered factors such as disease severity, US mode/intensity/dose, co-interventions, number of sessions provided, and methodological adequacy.
Results
Twenty-three studies, out of 1119 citations identified, fulfilled the inclusion criteria and were retrieved for full text review (see Fig. 1). Of these, only six studies were included and considered in the final analysis. The remaining 17 studies were excluded because the samples included persons with diagnoses other than knee OA
Bersani F. Mechanical Stimulation of Cartilage by Ultrasound. Electricity and Magnetism in Biology and Medicine. Kluwer,
Bologna, Italy1997 Jun: 947-950
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
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.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
, unpublished data were provided by the primary author. All trials included people with knee OA who met the American College of Rheumatology diagnostic criteria
Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and therapeutic criteria committee of the american rheumatism association.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
conducted by the same group reported a peak intensity value followed by the statement: “The intensity of sonication was adjusted to the level at which the patient felt a warm sensation or a mild sting”. It can be inferred from this statement that the output intensity was modified for each US application and the calculated dose would be inaccurate. Communication with the primary author confirmed that the intensity output was fixed and only the speed of the sound head varied during the US application. Thus, US dose could be calculated.
Table ISummary of included randomized controlled trials evaluating the effectiveness of US in people with knee OA
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.
The ultrasonic mode was not reported, we assumed it was continuous based on the way the intensity of the energy was delivered (“from 0W/cm2 to maximal tolerable dose not exceeding 2.5W/cm2 ”).
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
Sample size of subjects whose data were included in the present meta−analysis.
=30; Pulsed mode (duty cycle: 25%), 2.5 W/cm2 intensity, 112.5 J/cm2 dose. Both groups used a Sonopulus 590 US device, 24 fifteen-min sessions in 8 weeks. Also: standardized warm up, isokinetic exercises and hotpack (home exercise program following 8 weeks).
‡ Visual Analogue Scale (VAS) from 0 cm (no pain) to 10 cm (most intense pain).
§ LSI scale from 0 (better) to 26 (worst).
‖ Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) physical function subscale from 0 (better) to 68 (worst).
¶ The ultrasonic mode was not reported, we assumed it was continuous based on the way the intensity of the energy was delivered (“from 0 W/cm2 to maximal tolerable dose not exceeding 2.5 W/cm2 ”).
# 99mTechnetium uptake was measured by bone scan (TOSHIBA GCA-90 γ-camera) 3 h after administrating the radioisotope.
∗∗ Same number of sessions and period of time as in the intervention group.
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.
, which were conducted by the same research group, reported outcomes of pain and physical function at 12 months (10 months after completing the interventions) and this information was analyzed separately.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
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.
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.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
The quality of evidence is low for pain and physical function outcomes because of the high risk of bias (Table II), and the heterogeneity observed in results across trials [I2=51–92%, Fig. 2, Fig. 3, Fig. 4] (Appendix B). The quality of evidence is considered low for cartilage repair because the only trial
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
that reported this outcome had an unclear risk of bias (Table II) and used a surrogate measure (99mTechnetium uptake) to assess cartilage status (Appendix B).
Fig. 2Meta-analyses of ultrasound effect on pain (cm-VAS). A: SMDs at the end of the intervention. B: Mode/intensity and dose subgroup analysis. C: SMDs at 12 months (10 months after completing US).
Fig. 3Meta-analyses of ultrasound effect on self-reported physical function (LSI scores and WOMAC physical function subscale scores). A: SMDs at the end of the intervention. B: SMDs of LSI scores at 12 months (10 months after completing US).
Fig. 4Meta-analyses of ultrasound effect on walking performance (time to walk 50 m in minutes and walking speed in m/min). A: SMDs at the end of the intervention. B: SMDs of walking speed (m/min) at 12 months (10 months after completing US).
All included trials assessed pain using a Visual Analog Scale (VAS) measured in centimeters. Overall, the application of US resulted in decreased pain [SMD (CI)=−0.49 (−0.79, −0.18), P=0.002] [Fig. 2(A)]. However, high heterogeneity was found (χ2=10.26, P=0.07, I2=51%), therefore predefined subgroup analyses were conducted. Ultrasound mode, intensity, and therapeutic dose completely explained the inconsistency between the groups [Fig. 2(B)]. In all subgroups, effect estimates favored US therapy; however, the differences were statistically significant only in the low intensity/pulsed US and US dose<150 J/cm2 subgroups [SMD (CI)=−0.85 (−1.16, −0.54)]. Overall, trials
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
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.
(Table I). The effect estimate observed favored the US intervention, however the difference between groups was not statistically significant [SMD (CI)=−0.54 (−1.19, 0.12), P=0.11] [Fig. 3(A)]. Heterogeneity was high (χ2=29.11, P<0.001, I2=86%), and predefined subgroup analyses were performed. Inconsistency was not explained satisfactorily by any of the subgroup analyses performed. Data pooled from two trials
showed an improvement at 12 months (10 months after completing the interventions) in the self-reported physical function of people who received US [SMD (CI)=−1.25 (−1.69, −0.81), P<0.001] [Fig. 3(B)].
Walking performance
Five studies reported walking performance: two studies measured the time taken to walk 50 m in minutes
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.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
(Table I). There was no significant improvement in the walking performance in the US group [SMD (CI)=0.81 (−0.09, 1.72), P=0.08] [Fig. 4(A)]. High heterogeneity was observed (χ2=52.2, P<0.001, I2=92%), and the subgroup analyses did not explain this inconsistency. Pooling the results of two studies
, by the same research group, showed that walking speed at 12 months (10 months after completing the interventions) was improved in the US group [SMD (CI)=1.47 (1.06, 1.88), P<0.001] [Fig. 4(B)].
Effect of US on patient-perception of disease severity
No studies were identified which reported an outcome related to patient-perception of disease severity.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
reported an outcome related to knee joint structure. The authors measured 99mTechnetium uptake on bone scans in order to determine an ‘index of arthritis severity’, 99mTechnetium uptake in the knee divided by the 99mTechnetium uptake in the middle third of the ipsilateral femur. The authors validated this outcome measure on animal models of cartilage injury
. The results of this study showed a significant decrease in the ‘index of arthritis severity’ measured after an 8 week US intervention for patients in the lowest and middle tertile for baseline measures of ‘index of arthritis severity’ [MD (CI)=0.8 (0.32, 1.28), P<0.001 and 1.8 (0.85, 2.75), P<0.001, respectively] but not for those in the highest tertile at baseline [MD (CI)=0.10 (−1.06, 1.26), P=0.87]. We considered this outcome an indirect measurement of cartilage status/repair. Our search did not yield any trial reporting a direct measurement of cartilage repair.
described the intention to monitor the incidence of adverse events related to the application of US and reported the absence of major complications. Another study
reported that no adverse events occurred either during or after the interventions. Since the number of adverse events reported in these two trials was zero, and adverse events were not reported in the rest of the included trials, an estimate of US safety could not be calculated.
Discussion
This systematic review provides a meta-analysis of the efficacy of US for decreasing pain and improving physical function in people with knee OA. New evidence was found (Table I) which shows that US can reduce pain by 21%, compared to a control group (Appendix B). The clinical importance of this finding can be appreciated in terms of the number of patients it is necessary to treat in order to observe improvement in pain in one patient. To calculate the number needed to treat (NNT), we used the formula proposed by Chinn et al.
. The SMD was transformed to an odds ratio and the proportion of subjects in the control group that will experience improvement in pain (29%) was determined from a prospective study involving a comparable patient sample
. Using this approach, the NNT is 7. It also seems that US applied using low intensity (<1 W/cm2), pulsed mode, and a therapeutic dose<150 J/cm2 could be more effective at reducing pain than US applied using high intensity (≥1 W/cm2), continuous mode and a therapeutic dose>150 J/cm2. In general, a non-significant positive effect (19.68% lower Lequesne Severity Index (LSI) score than the control group) was observed on physical function with the use of US (Appendix B). Based on the findings from two trials by one research group
, beneficial effects may last for 10 months after the US treatment is completed [Fig. 2, Fig. 3, Fig. 4].
Limited evidence found in this review prevents conclusive statements regarding dose effects and the effectiveness of US on cartilage repair. Our findings which suggest a threshold US dosage for pain reduction in persons with knee OA are consistent with previous findings that higher US doses are less effective for tissue repair in humans
. Only one small trial with unclear risk of bias assessed the cartilage repair process. This study used an indirect measurement of cartilage status in people with knee OA and reported that low intensity pulsed US may help to enhance the cartilage repair process on this population. These findings together with animal studies which permit direct measurement of cartilage tissue response to US
are consistent with the “mechanotransduction theory”. This theory proposes that mechanical stimuli increase the chondrocyte production of proteoglycans and anti-inflammatory proteins
. Further studies are warranted to determine if low dose pulsed US is optimal for pain relief and stimulation of cartilage repair.
Our findings suggest that US delivered using a pulsed mode and at low intensities have a pronounced effect on pain reduction. In theory, the thermal effect of US is proposed to reduce pain
. However, our findings suggest a non-thermal mechanism for pain reduction which may be related to a threshold dose effect, as stated previously, or attenuation of the nociceptive signals via mechanical stimuli. Nevertheless, the limited number and quality of the trials included in the subgroup analyses limit further inferences.
None of the subgroup analyses that we conducted explained the heterogeneity observed for physical function and walking performance outcomes. Heterogeneity was reduced for self-reported physical function (I2 reduced from 86% to 30%) and walking performance (I2 reduced from 94% to 0%) when groups with a similar severity of knee OA were compared and the US group was favoured. However, these subgroups were comprised of subjects with mild disease severity (Altman II), so it is unclear if these observations are related to the comparison of homogeneous groups or to the mild disease severity. Thus, consideration of disease status may be important when assessing the effects of US in the knee OA population.
The main results of this review support the findings of a recently updated Cochrane review
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
related to our objective to evaluate the effectiveness of US on cartilage repair. This trial was not considered in the updated Cochrane review. Finally, we decided to analyze the data from two trials
that reported outcomes at 12 months (10 months after the interventions were completed) to explore the longer term US effects. These long-term effects were not considered in the updated Cochrane review. Despite these differences, our results are compatible.
A limitation of our review is that the findings are based on evidence which has a high risk of bias and low quality. When we synthesized only the trials that adequately blinded the participants
the effect size decreased considerably and was no longer statistically significant [SMD (CI)=−0.24 (−0.63, 0.14)]. This suggests that the effect sizes for pain found in this review could be partly inflated by the methodological limitations of the included studies.
A further limitation is the decision to pool the results of the studies that compared US and placebo with studies that included co-interventions. We assume that no interaction between US and the isokinetic exercises occurred, however we cannot test this assumption. It is well known that exercise is beneficial in relieving pain and improving physical function for people with knee OA
. Therefore, a positive interaction between US and exercise cannot be ruled out. Finally, because of the limited number of small trials identified, the risk of publication bias could not be determined. Hence, the possibility exists that only positive trials were published while negative trials were not. Overall, the methodological limitations described reduce the confidence in the effect estimates observed in the present meta-analyses.
Implications for practice
US (10–24 sessions) appears to be efficacious for decreasing pain, and may improve physical function in patients with knee OA. It is possible that the mode, intensity, and dose of US all influence the effect on pain. It is also possible that pain reduction may be sustained for 10 months after US is discontinued. However, these results are currently supported by low quality evidence and definitive trials are needed.
Implications for research
Trials that are methodologically rigorous and adequately powered are needed to confirm the effectiveness of US to reduce pain, and improve physical function in people with knee OA. Outcome measures in trials should include cartilage repair and patient-perception of knee OA severity to provide insight into potential synergistic action mechanisms. Careful consideration of ultrasound prescription and disease stage is required to assess the optimal therapeutic parameters and the subgroup(s) of people who will benefit most. Long-term effects of the US intervention should be assessed as well. Finally, the mechanism by which therapeutic ultrasound reduces pain in knee OA needs to be explored further.
Author contributions
All authors made substantial contributions to the conceptualization, design, data collection, analysis, interpretation, drafting and revisions; and approved the final version.
Conflict of interest
None of the authors has any financial and personal relationships with other people or organizations that could potentially and inappropriately influence this work and its conclusions.
Acknowledgement
Role of funding sources: AL was supported by the Consejo Nacional de Ciencia y Tecnologia (CONACYT) of Mexico scholarship (number 209621), and by a McMaster University School of Graduate Studies International Excellence Award. These sponsors had no involvement in the design, conduct or publication of this study.
Appendix A. Medical Subject Headings and keyword search strategy performed in Ovid MEDLINE (1950 to January week 4, 2009)
GRADE Working Group grades of evidence15. High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.
Comments
Pain VAS from 0 (no pain) to 10 (intense pain) Follow up: 2–8 weeks
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.21 [Lequesne severity index=2.06] and Huang et al.22 [walking speed 5.73].
Higher (7.07–14.71 higher)
13% Higher (8%, 18%)
Low
Cartilage repair Follow-up: 8 weeks Arthritis Severity Index (smaller values mean less severity)
4.9
1.8 Lower (2.47–1.13 lower)
36% Lower (50%, 23%)
Low
High risk of bias and indirectness of the outcome measure (we consider bone scan as a surrogate measurement of cartilage repair).
∗ We considered that the probability of having a “publication bias” in this review is low.
. High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.
‡ The final score mean for the control groups was calculated by pooling the means and standard errors through a generic inverse-variance method.
§ All patients included in these studies had mild knee OA (Altman Grade II) and completed a home-exercise program after 2 months of treatment/sham US.
‖ .
¶ MD were calculated through a back transformation of the SMD using the SD reported in Villanueva et al.
Relative efficiency and validity properties of a visual analogue vs a categorical scaled version of the western ontario and McMaster universities osteoarthritis (WOMAC) index: Spanish versions.
The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
Peek W.J. Lankhorst G.J. Effects of Pulse Sonication on Functional Status of Patients with Knee Osteoarthritis. International Society of Physical and Rehabilitation Medicine. Monduzzi,
Amsterdam, The Netherlands2001 Jul: 297-300
Relative efficiency and validity properties of a visual analogue vs a categorical scaled version of the western ontario and McMaster universities osteoarthritis (WOMAC) index: Spanish versions.
Bersani F. Mechanical Stimulation of Cartilage by Ultrasound. Electricity and Magnetism in Biology and Medicine. Kluwer,
Bologna, Italy1997 Jun: 947-950
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.
Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and therapeutic criteria committee of the american rheumatism association.