Shockwaves for myofascial pain: pooled evidence with reporting concerns · Review · 2022
Does shockwave ease myofascial pain more than other treatments?
In this review of 10 trials with 570 people, mostly with neck and upper-back muscle pain, pain just after treatment was about 1.3 points lower with shockwave than with other treatments, on average. Against dry needling, results were close.
Possible short-term help for myofascial pain- trials in the pain analysis
- 10
- people, mostly with neck and upper-back pain
- 570
- results pooled: pain, pressure tolerance, neck disability
- 3
Good to know: The trials varied widely in treatments and comparisons, and most followed people for a short time.
Wu T, Li S, Ren J, et al. Annals of Translational Medicine, 2022 · Read the paper ↗
How this study worked
What happened
- studies
- 10
- people
- 570
- Compared with
- Trigger-point injections, dry needling, ultrasound, taping and other active treatments
Pain compared with other treatments
Pooled difference in pain score · end of treatment or a week later
Show the numbers
| Difference (pain score points) | likely range | |
|---|---|---|
| Pain, 10 trials | −1.34 | −1.87 to −0.81 |
What the trials used
- Protocol availability
- ESWT protocols varied across the included studies; Table 1 names treatments but does not provide a reproducible device, energy, pulse and session schedule for each. The review cannot define one recommended dose.
Who took part
The authors report 571 people (541 with neck or upper-back pain, 30 with low back pain), but their tables add up to 570.
Side effects
The review did not count side effects across the trials; the authors, citing earlier research, describe shockwave as having no adverse effects.
All the numbers and the full appraisal
The results, in numbers.
The authors report 571 participants, 541 with neck/upper-back pain and 30 with lumbar pain. However, Table 1 and the pain forest plot sum to 570 participants (283 experimental, 287 control). These conflicting source totals must remain visible; the pooled sample is not one new cohort.
Pain intensity1 reported outcome
Pain intensity
Pain intensity Source scale units; pooled MD
ESWT minus heterogeneous active comparators
10 studies · 570 participants in this estimate · I²=85%
Reported pooled pain MD was −1.34 (95% CI −1.87 to −0.81; I²85%). Pressure threshold MD was +0.90 (0.73 to 1.07; I²98%) and NDI MD −1.79 (−2.48 to −1.10; I²92%). Figures 5 and 6 label the latter two models fixed-effect, contradicting the random-effects description. Individual comparisons with dry needling crossed the null; the review does not justify superiority to every comparator.
Analysis details and source
Model: random-effects. Source-reported estimate preserved without recalculation; participant count from forest plot, not conflicting abstract total.
Pressure pain threshold1 reported outcome
Pressure pain threshold
Pressure pain threshold Source scale units; pooled MD
ESWT minus heterogeneous active comparators
6 studies · 364 participants in this estimate · I²=98%
Reported pooled pain MD was −1.34 (95% CI −1.87 to −0.81; I²85%). Pressure threshold MD was +0.90 (0.73 to 1.07; I²98%) and NDI MD −1.79 (−2.48 to −1.10; I²92%). Figures 5 and 6 label the latter two models fixed-effect, contradicting the random-effects description. Individual comparisons with dry needling crossed the null; the review does not justify superiority to every comparator.
Analysis details and source
Model: fixed-effect in figure, random-effects in prose. Source-reported estimate preserved without recalculation; participant count from forest plot, not conflicting abstract total.
Neck disability1 reported outcome
Neck disability
Neck disability Source scale units; pooled MD
ESWT minus heterogeneous active comparators
3 studies · 264 participants in this estimate · I²=92%
Reported pooled pain MD was −1.34 (95% CI −1.87 to −0.81; I²85%). Pressure threshold MD was +0.90 (0.73 to 1.07; I²98%) and NDI MD −1.79 (−2.48 to −1.10; I²92%). Figures 5 and 6 label the latter two models fixed-effect, contradicting the random-effects description. Individual comparisons with dry needling crossed the null; the review does not justify superiority to every comparator.
Analysis details and source
Model: fixed-effect in figure, random-effects in prose. Source-reported estimate preserved without recalculation; participant count from forest plot, not conflicting abstract total.
Dry-needling comparisons1 reported outcome
Dry-needling comparisons
Dry-needling comparisons Narrative finding
Figure 4: Luan pain MD −0.18 (95% CI −0.66 to 0.30), Toghtamesh −0.25 (−0.93 to 0.43); both include no difference. Figure5 Luan pressure MD −0.04 (−0.29 to 0.21).
Analysis details and source
Qualitative finding; no invented effect estimate.
How treatment was delivered
ESWT protocols varied across the included studies; Table 1 names treatments but does not provide a reproducible device, energy, pulse and session schedule for each. The review cannot define one recommended dose.
No systematic pooled adverse-event outcome or event denominator is supplied. The narrative claim of no adverse effects cannot establish safety. No funding or conflicts were declared.
High heterogeneity, nonrandomized studies, English-only inclusion and many inaccessible/excluded full texts weaken causal and coverage claims. Model labels disagree with Methods for pressure threshold and NDI; study and participant totals disagree internally. A visually symmetric funnel with few studies cannot exclude reporting bias. The stated PEDro scoring framework also needs caution.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- High heterogeneity, nonrandomized studies, English-only inclusion and many inaccessible/excluded full texts weaken causal and coverage claims. Model labels disagree with Methods for pressure threshold and NDI; study and participant totals disagree internally. A visually symmetric funnel with few studies cannot exclude reporting bias. The stated PEDro scoring framework also needs caution.
Source-linked critical reading
Systematic review with meta-analysis of controlled clinical studies, searching four databases to October 2021 in English. Nonrandomized and allocation-unclear studies were eligible. The abstract says eight studies, whereas the flowchart, Table 1 and Discussion identify ten.
Technical source record
Full local PDF read; All 9 PDF pages; Methods and eligibility; Table 1; Figures 1–7 and captions; Forest plots Figures 4–6 visually inspected on pages6–7; Discussion; disclosures; references
The review suggests possible short-term improvement, but important reporting and analysis inconsistencies make its precise pooled numbers uncertain. It does not prove shockwaves are better than every other treatment.
Does ESWT improve short-term pain, pressure sensitivity and neck disability more than other therapies for myofascial pain?
Systematic review with meta-analysis of controlled clinical studies, searching four databases to October 2021 in English. Nonrandomized and allocation-unclear studies were eligible. The abstract says eight studies, whereas the flowchart, Table 1 and Discussion identify ten.
The authors report 571 participants, 541 with neck/upper-back pain and 30 with lumbar pain. However, Table 1 and the pain forest plot sum to 570 participants (283 experimental, 287 control). These conflicting source totals must remain visible; the pooled sample is not one new cohort.
ESWT protocols varied across the included studies; Table 1 names treatments but does not provide a reproducible device, energy, pulse and session schedule for each. The review cannot define one recommended dose.
Controls included trigger-point injection with/without TENS, dry needling, ultrasound with/without other modalities, phonophoresis and kinesiology taping. Pooling these comparators does not estimate efficacy against any single treatment. The Discussion incorrectly conflates ultrasound with ultrasound-guided pulsed radiofrequency.
Prespecified extraction targeted end-of-treatment or one-week post-treatment VAS, pressure pain threshold and Neck Disability Index. Table 1 lists study follow-up of four to six weeks, without fully resolving the selected assessment origin. Forest plots provide the reported MDs and 95% CIs.
Reported pooled pain MD was −1.34 (95% CI −1.87 to −0.81; I²85%). Pressure threshold MD was +0.90 (0.73 to 1.07; I²98%) and NDI MD −1.79 (−2.48 to −1.10; I²92%). Figures 5 and 6 label the latter two models fixed-effect, contradicting the random-effects description. Individual comparisons with dry needling crossed the null; the review does not justify superiority to every comparator.
No systematic pooled adverse-event outcome or event denominator is supplied. The narrative claim of no adverse effects cannot establish safety. No funding or conflicts were declared.
High heterogeneity, nonrandomized studies, English-only inclusion and many inaccessible/excluded full texts weaken causal and coverage claims. Model labels disagree with Methods for pressure threshold and NDI; study and participant totals disagree internally. A visually symmetric funnel with few studies cannot exclude reporting bias. The stated PEDro scoring framework also needs caution.
Evidence mostly concerns neck/upper-back MPS and short follow-up. It cannot be generalized to lumbar disease on the basis of one 30-person comparison, nor used to select a device or replace exercise.
Reconcile the extraction with primary reports, correct study totals and comparator labels, rerun appropriate heterogeneous-effect models and provide comparator-specific, longer-term estimates and harms.
Read the original article.
Efficacy of extracorporeal shock waves in the treatment of myofascial pain syndrome: a systematic review and meta-analysis of controlled clinical studies.
Wu T, Li S, Ren J, Wang D, Ai Y. · Annals of translational medicine · 2022
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →