Trapezius trigger point: radial shockwaves versus dry needling · Randomized trial · 2019
How does shockwave compare with dry needling for a neck trigger point?
In this trial, pain fell in both groups after three weekly sessions: with shockwave from about 3.6 to 1.5 out of 10 at 3 months, and with dry needling from about 3.8 to 1.7. The groups ended close.
No clear difference between the groups- people with a trapezius trigger point
- 65
- sessions, one a week
- 3
- after the last session
- 3 mo
Good to know: Starting pain was mild, under 4 out of 10.
Luan S, Zhu ZM, Ruan JL, et al. Am J Phys Med Rehabil, 2019 · Read the paper ↗
How this study worked
What happened
Neck and shoulder pain
0–10 pain scale · start and 3 months after the last session
Start3 months after
Show the numbers
| Group | People | Start | 3 months after |
|---|---|---|---|
| Radial shockwave | — | 3.57 | 1.5 |
| Dry needling | — | 3.78 | 1.69 |
The treatment
Radial shockwave
Radial, once a week for 3 weeks
- Sessions
- 3, one a week
- Pulses
- 2,000 per session · 1,500 on the point, 500 around the tight band
- Strength
- 0.10mJ/mm²
- Numbing
- Not mentioned
- Device
- Swiss DolorClast
- Everyone also got
- No pain medicines or physiotherapy during the study
Compared with: Dry needling once a week for three weeks
Settings are what each study used, not a prescription.
Who took part
65 people aged 16–60 in Guangzhou, China, with one active trigger point in the upper trapezius for at least 6 months.
hollow squares: no final score
Side effects
No complications or soreness were reported after either treatment.
All the numbers and the full appraisal
The results, in numbers.
Outpatients aged 16–60 at a Guangzhou hospital rehabilitation department (recruited January–December 2016) with a single active trigger point in one upper trapezius and symptoms for at least 6 months. Completers: 30 ESWT (22 women, mean age 32.47) and 32 dry needling (21 women, mean age 33.09); mean baseline VAS 3.57 and 3.78 of 10 cm.
32 randomized; 2 dropped out after receiving other treatments; 30 completed and were analysed.
33 randomized; 1 dropped out after receiving other treatment; 32 completed and were analysed.
Pain intensity3 reported outcomes
Pain intensity
Visual analog scale (VAS) cm on a 10 cm line
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 3.57 (SD 1.04) | 2.93 (SD 0.94) |
| Dry needling | — | 3.78 (SD 1.18) | 2.78 (SD 1.07) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
VAS fell in both groups at every visit (P<0.01 versus baseline). Table 2 marks no between-group difference at any time point; the repeated-measures group effect was F=0.27, P=0.60 and group×time F=0.91, P=0.44.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect. Figure 1 instead places this assessment after the third treatment; the text and limitations indicate it followed the first session.
Pain intensity
Visual analog scale (VAS) cm on a 10 cm line
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 3.57 (SD 1.04) | 1.73 (SD 0.91) |
| Dry needling | — | 3.78 (SD 1.18) | 1.91 (SD 1) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
VAS fell in both groups at every visit (P<0.01 versus baseline). Table 2 marks no between-group difference at any time point; the repeated-measures group effect was F=0.27, P=0.60 and group×time F=0.91, P=0.44.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Pain intensity
Visual analog scale (VAS) cm on a 10 cm line
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 3.57 (SD 1.04) | 1.5 (SD 0.82) |
| Dry needling | — | 3.78 (SD 1.18) | 1.69 (SD 1.03) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus dry needling across all four assessments (F=0.27; group×time F=0.91, P=0.44)
Repeated-measures ANOVA, group main effect
VAS fell in both groups at every visit (P<0.01 versus baseline). Table 2 marks no between-group difference at any time point; the repeated-measures group effect was F=0.27, P=0.60 and group×time F=0.91, P=0.44.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Pressure pain threshold3 reported outcomes
Pressure pain threshold
Pressure algometer (PTH AF2) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 206.9 (SD 34.24) | 253.1 (SD 51.26) |
| Dry needling | — | 198.63 (SD 35.16) | 246.03 (SD 46.91) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Pressure pain threshold rose in both groups at every visit (P<0.01 versus baseline). Group effect F=0.15, P=0.70; group×time F=0.39, P=0.72; no between-group difference marked at any time point.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect. Figure 1 instead places this assessment after the third treatment; the text and limitations indicate it followed the first session.
Pressure pain threshold
Pressure algometer (PTH AF2) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 206.9 (SD 34.24) | 320.13 (SD 61.15) |
| Dry needling | — | 198.63 (SD 35.16) | 317.09 (SD 49.82) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Pressure pain threshold rose in both groups at every visit (P<0.01 versus baseline). Group effect F=0.15, P=0.70; group×time F=0.39, P=0.72; no between-group difference marked at any time point.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Pressure pain threshold
Pressure algometer (PTH AF2) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 206.9 (SD 34.24) | 316.97 (SD 52.37) |
| Dry needling | — | 198.63 (SD 35.16) | 320.84 (SD 46.08) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus dry needling across all four assessments (F=0.15; group×time F=0.39, P=0.72)
Repeated-measures ANOVA, group main effect
Pressure pain threshold rose in both groups at every visit (P<0.01 versus baseline). Group effect F=0.15, P=0.70; group×time F=0.39, P=0.72; no between-group difference marked at any time point.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Neck disability2 reported outcomes
Neck disability
Neck Disability Index (Chinese version) points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 15.37 (SD 2.04) | 9.57 (SD 1.77) |
| Dry needling | — | 15.31 (SD 2.44) | 9.38 (SD 2.46) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
NDI fell in both groups at 1 and 3 months (P<0.01 versus baseline); it was not measured immediately after the first session. Group effect F=0.02, P=0.90; group×time F=0.65, P=0.52.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Neck disability
Neck Disability Index (Chinese version) points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 15.37 (SD 2.04) | 9.07 (SD 1.7) |
| Dry needling | — | 15.31 (SD 2.44) | 9.47 (SD 1.87) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus dry needling at baseline, 1 and 3 months (F=0.02; group×time F=0.65, P=0.52)
Repeated-measures ANOVA, group main effect
NDI fell in both groups at 1 and 3 months (P<0.01 versus baseline); it was not measured immediately after the first session. Group effect F=0.02, P=0.90; group×time F=0.65, P=0.52.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect.
Trigger-point stiffness (shear modulus)3 reported outcomes
Trigger-point stiffness (shear modulus)
Shear wave elastography (Aixplorer) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 46.87 (SD 6.79) | 33.35 (SD 5.23) |
| Dry needling | — | 44.83 (SD 6.37) | 41.7 (SD 6.54) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Change in shear modulus over time, radial ESWT versus dry needling (F=16.61)
Repeated-measures ANOVA, group×time interaction
Shear modulus fell in both groups. The group×time interaction was significant (F=16.61, P<0.01), driven by the larger immediate drop after the first ESWT session; the text reports a between-group difference at that visit (P<0.01), but Table 2 prints no between-group marker. There was no significant between-group difference at 1 and 3 months. Stiffness is an imaging measure, not a symptom.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect. Figure 2E bar heights do not match Table 2 (for example, about 23 kPa in both groups at 1 month in the figure versus 28.59 and 27.16 kPa in the table); values here follow Table 2. Figure 1 instead places this assessment after the third treatment; the text and limitations indicate it followed the first session.
Trigger-point stiffness (shear modulus)
Shear wave elastography (Aixplorer) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 46.87 (SD 6.79) | 28.59 (SD 3.88) |
| Dry needling | — | 44.83 (SD 6.37) | 27.16 (SD 3.26) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Shear modulus fell in both groups. The group×time interaction was significant (F=16.61, P<0.01), driven by the larger immediate drop after the first ESWT session; the text reports a between-group difference at that visit (P<0.01), but Table 2 prints no between-group marker. There was no significant between-group difference at 1 and 3 months. Stiffness is an imaging measure, not a symptom.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect. Figure 2E bar heights do not match Table 2 (for example, about 23 kPa in both groups at 1 month in the figure versus 28.59 and 27.16 kPa in the table); values here follow Table 2.
Trigger-point stiffness (shear modulus)
Shear wave elastography (Aixplorer) kPa
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial shockwaves | — | 46.87 (SD 6.79) | 30.98 (SD 5.21) |
| Dry needling | — | 44.83 (SD 6.37) | 31.01 (SD 5.65) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus dry needling across all four assessments (F=1.68)
Repeated-measures ANOVA, group main effect
Shear modulus fell in both groups. The group×time interaction was significant (F=16.61, P<0.01), driven by the larger immediate drop after the first ESWT session; the text reports a between-group difference at that visit (P<0.01), but Table 2 prints no between-group marker. There was no significant between-group difference at 1 and 3 months. Stiffness is an imaging measure, not a symptom.
Analysis details and source
Mean ± SD as printed in Table 2. Table 2 does not print n per cell; Results (p. 680) say 30 ESWT and 32 dry-needling participants completed and the analysis followed exclusion of the three dropouts. Within-group change from baseline is not a between-group treatment effect. Figure 2E bar heights do not match Table 2 (for example, about 23 kPa in both groups at 1 month in the figure versus 28.59 and 27.16 kPa in the table); values here follow Table 2.
How treatment was delivered
Swiss DolorClast (Electro Medical Systems, Nyon), radial shockwaves, applied by a designated physiotherapist with coupling gel.
2,000 shocks per session (1,500 on the trigger point, 500 around the taut band); energy flux density printed as 0.10 mJ/mm². Pressure in bar and frequency were not reported.
Three sessions at 1-week intervals (6,000 shocks in total). Local anesthesia was not mentioned.
One active upper trapezius trigger point, confirmed by palpation (Travell and Simons criteria) and by the twitch response to the probe.
Dry needling once weekly for three weeks: 0.30 × 50 mm solid needle inserted 30–35 mm into the trigger point with the patient seated, eliciting a twitch response, moved in and out 10 times. Participants were instructed not to take analgesics or receive physiotherapy during treatment and follow-up; three who received other treatment were excluded.
Opaque sealed envelopes from random.org numbers; examiners and sonographers blinded; participants not blinded.
No complications or reported soreness after either treatment during the 3-month follow-up, and no withdrawals for adverse effects. Events were collected by participants reporting intolerable soreness or other feelings to the research nurse by telephone; no structured adverse-event schedule is described.
Author-reported: participants were not blinded; there was no inactive or healthy control group; only patients with a single trigger point were enrolled; ESWT dose was not varied; follow-up ended at 3 months; the authors ask for results to be read as a noninferiority design with 80% power.
Extractor: no noninferiority margin or equivalence test was prespecified; the sample size targeted a repeated-measures ANOVA on VAS. A nonsignificant group effect therefore does not establish that the two treatments are equivalent, despite the authors' conclusion that the treatments are equally effective and safe.
Extractor: 65 were randomized (32/33) and 62 completed (30/32); the analysis excluded the three who received other treatments, without an intention-to-treat analysis. Table 2 does not print cell-level n.
Internal inconsistencies: the Results text reports a between-group shear-modulus difference immediately after the first session (P<0.01), but Table 2 shows no between-group marker; Figure 2E bar heights differ from Table 2 means; NDI time-effect F values in the text (176.54 and 146.88) differ from Table 2 (318.92); the limitations say outcomes were not assessed "after the third therapy", although 1- and 3-month assessments followed the third session. Figure 1 places the "immediate" assessment after the third treatment, whereas the Abstract, Outcome Measures and limitations indicate it followed the first session.
Extractor: baseline pain was mild (mean VAS 3.57 and 3.78 of 10 cm), leaving limited room for change; the authors themselves call the VAS reductions "relatively minimal". The paper states registration ChiCTR-INR-17014029 but not its date; recruitment ran through 2016.
Funding: National Natural Science Foundation of China, Natural Science Foundation of Guangdong, Guangzhou Science and Technology Plan and a Sun Yat-Sen clinical research programme. No conflicts of interest reported. CONSORT and STROBE checklists are online supplements not supplied here.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Author-reported: participants were not blinded; there was no inactive or healthy control group; only patients with a single trigger point were enrolled; ESWT dose was not varied; follow-up ended at 3 months; the authors ask for results to be read as a noninferiority design with 80% power.
- Extractor: no noninferiority margin or equivalence test was prespecified; the sample size targeted a repeated-measures ANOVA on VAS. A nonsignificant group effect therefore does not establish that the two treatments are equivalent, despite the authors' conclusion that the treatments are equally effective and safe.
- Extractor: 65 were randomized (32/33) and 62 completed (30/32); the analysis excluded the three who received other treatments, without an intention-to-treat analysis. Table 2 does not print cell-level n.
- Internal inconsistencies: the Results text reports a between-group shear-modulus difference immediately after the first session (P<0.01), but Table 2 shows no between-group marker; Figure 2E bar heights differ from Table 2 means; NDI time-effect F values in the text (176.54 and 146.88) differ from Table 2 (318.92); the limitations say outcomes were not assessed "after the third therapy", although 1- and 3-month assessments followed the third session. Figure 1 places the "immediate" assessment after the third treatment, whereas the Abstract, Outcome Measures and limitations indicate it followed the first session.
- Extractor: baseline pain was mild (mean VAS 3.57 and 3.78 of 10 cm), leaving limited room for change; the authors themselves call the VAS reductions "relatively minimal". The paper states registration ChiCTR-INR-17014029 but not its date; recruitment ran through 2016.
- Funding: National Natural Science Foundation of China, Natural Science Foundation of Guangdong, Guangzhou Science and Technology Plan and a Sun Yat-Sen clinical research programme. No conflicts of interest reported. CONSORT and STROBE checklists are online supplements not supplied here.
Source-linked critical reading
randomized trial
Technical source record
Complete local PDF read (8 pages), including all body sections, Tables 1–2, Figures 1–2, funding, conflict statement and references. Online CONSORT/STROBE checklists were not supplied or read.
In 65 people aged 16–60 with one active trapezius trigger point, three weekly sessions of radial shockwaves or of dry needling were followed by similar improvements in pain, pressure sensitivity and neck disability up to 3 months. The trial was not built to prove the two equivalent and had no sham group, so it shows similar course, not how much each treatment itself helped.
Does radial ESWT relieve pain, disability and trigger-point stiffness as well as dry needling in people with an active upper trapezius trigger point?
Single-centre randomized trial: 65 participants allocated by sealed opaque envelopes (32 ESWT, 33 dry needling); assessors and sonographers blinded, participants not. Sample size was powered for a repeated-measures ANOVA on VAS, the primary outcome; no noninferiority margin was set. Registered as ChiCTR-INR-17014029.
Outpatients aged 16–60 at a Guangzhou hospital rehabilitation department (recruited January–December 2016) with a single active trigger point in one upper trapezius and symptoms for at least 6 months. Completers: 30 ESWT (22 women, mean age 32.47) and 32 dry needling (21 women, mean age 33.09); mean baseline VAS 3.57 and 3.78 of 10 cm.
Radial ESWT with the Swiss DolorClast (EMS): 2,000 shocks per session at an energy flux density printed as 0.10 mJ/mm² (1,500 on the confirmed trigger point, 500 around the taut band), three sessions one week apart (6,000 shocks). Frequency was not reported; trigger points were confirmed by the twitch response to the probe.
Dry needling once weekly for three weeks: 0.30 × 50 mm solid needle inserted 30–35 mm into the trigger point with the patient seated, eliciting a twitch response, moved in and out 10 times. Participants were instructed not to take analgesics or receive physiotherapy during treatment and follow-up; three who received other treatment were excluded.
VAS (primary), pressure pain threshold by algometer (kPa), Neck Disability Index and shear modulus by elastography, assessed at baseline, 15–30 minutes after the first session (Figure 1 instead places this visit after the third session), and 1 and 3 months after the third session (NDI only at baseline, 1 and 3 months).
Pain (primary outcome), pressure pain threshold and NDI improved in both groups, with no statistically significant between-group difference: 3-month VAS 1.50±0.82 vs 1.69±1.03 cm (ESWT/dry needling), group P=0.60; NDI 9.07±1.70 vs 9.47±1.87, P=0.90. Shear modulus fell more right after the first ESWT session (33.35 vs 41.70 kPa; interaction P<0.01) but was similar at 1 and 3 months. The Results text also reports a between-group shear-modulus difference right after the first session, which Table 2 does not mark.
No complications or reported soreness after either treatment during the 3-month follow-up, and no withdrawals for adverse effects. Events were collected by participants reporting intolerable soreness or other feelings to the research nurse by telephone; no structured adverse-event schedule is described.
Sealed-envelope randomization with blinded assessors, but participants were unblinded, the analysis excluded the three dropouts, and there was no sham or untreated arm. The sample size was powered for an ANOVA, not a noninferiority margin, so the authors' equivalence conclusion overstates what a nonsignificant contrast shows. Text and Table 2 disagree on a between-group shear-modulus difference, and Figure 2E does not match Table 2. Funded by Chinese public grants; no conflicts declared. Baseline pain was mild.
Findings concern people aged 16–60 with a single trapezius trigger point, low baseline pain and no cervical disease, treated with this radial protocol. They describe similar short-term courses after two active treatments; they do not show equivalence, effects beyond 3 months, or results for multiple trigger points.
A trial with a prespecified noninferiority margin, a sham or usual-care arm, intention-to-treat analysis, structured harms recording and follow-up beyond 3 months, including people with several trigger points.
Read the original article.
Randomized Trial on Comparison of the Efficacy of Extracorporeal Shock Wave Therapy and Dry Needling in Myofascial Trigger Points.
Luan S, Zhu ZM, Ruan JL, Lin CN, Ke SJ, Xin WJ, Liu CC, Wu SL, Ma C. · Am J Phys Med Rehabil · 2019
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →