Chronic wounds and graft wounds · Review · 2025
Does shockwave ease pain from chronic wounds?
This review looked at 30 trials of drug-free treatments for chronic wound pain; three added shockwave to standard wound care. Pain results in those trials varied widely, and overall the review did not find a clear difference in pain compared with standard wound care alone.
- trials of drug-free treatments
- 30
- trials added shockwave
- 3
- kinds of drug-free treatment compared
- 10
Good to know: Trials covered many kinds of wounds and used different pain scales; side effects were not reviewed.
Fu CY, Li N, Li MJ, et al. Zhonghua Shao Shang Yu Chuang Mian Xiu Fu Za Zhi, 2025 · Read the paper ↗
How this study worked
Conceptual animation: a focused front converges, then spreads. Not to scale. Tap the field to send a pulse.
What happened
- studies
- 3
- Compared with
- Standard wound care alone
Wound pain with shockwave
3 trials pooled · standardised difference (no units)
Show the numbers
| Difference (standardised difference (no units)) | likely range | |
|---|---|---|
| Wound pain | −1.73 | −4.09 to 0.62 |
What the trials used
- ESWT protocols represented
- Three studies combined ESWT with standard wound care: Jeppesen twice weekly for 3 weeks, Taheri twice weekly for 3 weeks, Zhou Jianying weekly for 5 weeks. Device, radial/focused type, impulses, energy, frequency, anesthesia, and technique were not reported in this review.
- Comparators
- Across the review, control arms received standard wound care, placebo, or a different non-drug intervention; in all three ESWT trials the control arm received standard wound care alone.
- Outcome scales
- Eligible pain scales: VAS, NRS, EQ-5D, or Lattinen; the meta-analysis standardized effects across scales.
Who took part
Adults with chronic wounds of many kinds in 30 trials (1,929 people); three trials tested shockwave.
Side effects
The review did not look at side effects.
All the numbers and the full appraisal
Every reported result
The results, in numbers.
Adults with chronic wounds (≥18 years) across multiple etiologies; 30 RCTs, 1,929 participants overall. The review did not stratify by wound etiology.
Pain score; a visual analogue scale in all three ESWT trials (Table 1 note)1 reported outcome
ESWT pain versus control (pairwise synthesis)
Pain score; a visual analogue scale in all three ESWT trials (Table 1 note) standardized mean difference
ESWT plus standard wound care compared with standard wound care alone, the control arm in all three contributing trials (Table 1).
3 studies · I² = 96.8%
Default random-effects pairwise model; SMD direction is intervention minus comparator as described in the table note.
Reported estimate and confidence interval.
The 95% interval includes zero; this synthesis does not show a statistically clear pain difference. It does not establish equivalence.
Analysis details and source
Table 1 contains three ESWT studies: Jeppesen (11 control, 10 ESWT; post-treatment pain 14.5±10.4 vs 22.8±7.4; twice weekly for 3 weeks), Taheri (22/22; 3.8±1.6 vs 2.6±1.4; twice weekly for 3 weeks), and Zhou Jianying (40/40; 4.13±0.62 vs 1.82±0.46; weekly for 5 weeks). All three used a visual analogue scale (Table 1 note), but Jeppesen's values are on a much larger range than the other two; do not pool these displayed means directly.
Pain score; instruments varied across trials1 reported outcome
ESWT versus standard wound care (network contrast)
Pain score; instruments varied across trials standardized mean difference
Network-matrix contrast between ESWT and standard wound care; the direction is ambiguous in the source (see note).
Bayesian consistency model after node-splitting assessment; numeric model priors and convergence diagnostics are not reported in this article extract.
Reported estimate and confidence interval.
The 95% interval includes zero; this synthesis does not show a statistically clear pain difference. It does not establish equivalence.
Analysis details and source
The reported matrix cell (row SWC, column ESWT) is −1.20 (95% CI −2.54 to 0.44); the interval crosses zero. The source prose identifies FREMS, CAP, NCLFU, and LLLT as significant versus standard care, but does not identify ESWT as significant. Direction: the Table 3 note says the matrix is read row versus column (以行比列表示), which taken literally would make this cell standard care versus ESWT; the authors, however, read the same SWC row as each treatment versus standard care (FREMS −3.13, CAP −1.75, NCLFU −1.22 and LLLT −1.11 reported as lower pain than standard care), under which −1.20 means numerically lower pain with ESWT. The interval crosses zero under either reading.
How treatment was delivered
Three studies combined ESWT with standard wound care: Jeppesen twice weekly for 3 weeks, Taheri twice weekly for 3 weeks, Zhou Jianying weekly for 5 weeks. Device, radial/focused type, impulses, energy, frequency, anesthesia, and technique were not reported in this review.
Across the review, control arms received standard wound care, placebo, or a different non-drug intervention; in all three ESWT trials the control arm received standard wound care alone.
Eligible pain scales: VAS, NRS, EQ-5D, or Lattinen; the meta-analysis standardized effects across scales.
Adverse events were not synthesized; authors list adverse events and costs among outcomes not included. This is not evidence of no harms.
Overall risk of bias was uncertain; 24/30 trials did not blind participants/providers, and randomization and concealment reporting were often unclear.
The ESWT pairwise estimate is highly heterogeneous (I² 96.8%) and its interval includes zero.
Wound etiology was not stratified; pain was the only outcome, and harms/costs were not evaluated.
Direct head-to-head evidence between interventions was relatively sparse; network ranking probabilities should not be treated as proof of superiority.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Overall risk of bias was uncertain; 24/30 trials did not blind participants/providers, and randomization and concealment reporting were often unclear.
- The ESWT pairwise estimate is highly heterogeneous (I² 96.8%) and its interval includes zero.
- Wound etiology was not stratified; pain was the only outcome, and harms/costs were not evaluated.
- Direct head-to-head evidence between interventions was relatively sparse; network ranking probabilities should not be treated as proof of superiority.
Source-linked critical reading
Systematic review with network meta-analysis
Technical source record
Complete full-article HTML read, all eight sections, all three tables, both figure captions, and disclosures.
This review does not show that ESWT reliably reduces chronic wound pain compared with control: the pairwise estimate is very uncertain and highly heterogeneous, and the network estimate versus standard wound care also crosses no effect. The authors’ ranking of other treatments should not be treated as proof that one is best. Wound type, treatment dose, and harms remain uncertain.
Among adults with chronic wounds, how do non-pharmacological treatments compare for wound-related pain, including ESWT? The analysis combines direct pairwise trials and a Bayesian treatment network.
Systematic review and Bayesian network meta-analysis of 30 RCTs (1,929 participants): 27 found through eight Chinese and international databases searched through June 30, 2024, and 3 from review reference lists. The review used random-effects pairwise meta-analysis and a Bayesian consistency model after node-splitting assessment. It is a synthesis, not a new trial.
Adults with chronic wounds (≥18 years) were eligible. Ulcer types were not analyzed separately; the authors say there were too few studies per wound type to form connected networks. The paper reports 1,016 participants across intervention groups and 913 across controls, but does not give a pooled ESWT-specific total in Table 2.
ESWT was included as one of ten non-drug categories, generally alongside standard wound care. The review table identifies three contributing ESWT trials with different schedules: Jeppesen twice weekly for 3 weeks, Taheri twice weekly for 3 weeks, and Zhou Jianying once weekly for 5 weeks. The article does not report device, focused/radial modality, impulses, energy, frequency, anesthesia, or application technique for these studies.
Comparators varied across included RCTs: standard wound care, placebo, or a different non-drug treatment. Standard care could include debridement, dressings, and pressure management. All three ESWT trials compared ESWT plus standard care with standard wound care alone.
Pain score was the only network outcome; eligible scales included VAS, NRS, EQ-5D, or Lattinen. Timing varied among trials and was not analyzed as a common follow-up. Table 2 pairwise ESWT estimate was SMD −1.73 (95% CI −4.09 to 0.62; I² 96.8%), using a random-effects model. Table 3’s network cell (row SWC, column ESWT) was −1.20 (95% CI −2.54 to 0.44); its direction is ambiguous, because the table note says row versus column while the authors read this row as each treatment versus standard care.
Neither ESWT synthesis shows a clear pain difference: the pairwise CI includes zero and heterogeneity is very high; the network cell versus standard care also includes zero. The authors state ESWT did not significantly reduce pain versus control in pairwise analysis. In the network ranking, FREMS/CAP/NCLFU/LLLT were described as significant versus standard care; rank probabilities are not effect sizes. The review’s abstract claims these four are effective and ranks FREMS highest, but ranking is uncertain under the review’s own bias concerns.
The review did not synthesize adverse events. Its limitations section explicitly notes that adverse-event rates and costs were not included as outcomes. This does not mean the treatments had no harms.
The review judged overall risk of bias uncertain: randomization methods were unclear in 13 studies; allocation concealment was achieved in 8, absent in 7, and unclear in 15; 24 studies did not blind participants/providers; one study had incomplete data; selective reporting and other bias were unclear in all 30. ESWT pairwise heterogeneity was I² 96.8%. Wound types were not subgrouped, outcome was pain only, comparator regimens varied, and the network included relatively few direct head-to-head comparisons. Although a comparison-adjusted funnel plot was reported as nonsignificant (P>0.05), this does not rule out publication bias.
These pooled results apply to mixed adult chronic-wound populations and heterogeneous care settings. They cannot establish an ESWT protocol or a specific effect for diabetic foot ulcers, venous ulcers, pressure injuries, or another wound type. Directness to shockwave treatment is limited by few studies and very high heterogeneity.
Future wound-pain trials should use concealed randomization and blinded assessment where feasible, prespecify pain scale and timing, stratify wound etiology, report complete arm denominators and adverse events, and test ESWT against standardized wound care/sham with device and dose fully specified. Updated network reviews should report model diagnostics, uncertainty in rankings, and direct versus indirect evidence contributions.
Read the original article.
[A systematic review and Bayesian network meta-analysis of the effectiveness of non-pharmacological interventions in treating chronic wound pain in patients].
Fu CY, Li N, Li MJ, Li HS, Cheng L, Wang LH. · Zhonghua shao shang yu chuang mian xiu fu za zhi · 2025
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →