Multiple tendon conditions · Review · 2026
Is radial or focused shockwave better for tendon problems?
In this review of 9 trials with 530 adults, neither type came out clearly ahead. The review reported better pain results with focused shockwave used on its own, and more wrist strength with radial shockwave. Discomfort during treatment and side effects did not differ clearly between the two.
- trials comparing radial and focused shockwave
- 9
- adults with tendon problems
- 530
- sessions, depending on the trial
- 3–5
Good to know: The trials covered several tendons, including the elbow, shoulder and kneecap.
Stania M, Pawłowski M, Benduch M, et al. Scientific Reports, 2026 · 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 the trials used
- Study-level dose variability
- Across trials, protocols ranged from 1,500 to 4,000 shocks (Król et al. RSWT: 2,000 + 2,000), 3.2–10 Hz, 0.09–0.3 mJ/mm² (Kaplan et al. and Şah & Delen print 0.02–0.60 mJ/mm² for FSWT) or approximately 1.2–4 bar, and 3–4 sessions once a week (Maher et al.: one session in alternate weeks, maximum 3–5; Tables 1–2 give no session count for Kaplan et al. or Şah & Delen); no pooled modality-level dose was specified.
- Representative focused regimen
- Calcific shoulder trial: 1,500 shocks at 4 Hz, 0.3 mJ/mm², 4 weekly sessions; sonographic marking of calcification.
- Representative radial regimen
- Calcific shoulder trial: 2,000 shocks at 10 Hz, 2.5 bar, 4 weekly sessions; sonographic marking of calcification.
- Co-interventions
- Four studies combined shockwave with exercise or conventional physiotherapy; these were analyzed separately from monotherapy.
Who took part
530 adults with tendon problems in the arm or leg, from 9 trials, each run at a single centre.
Side effects
Discomfort during treatment and side effects did not differ clearly between the two types. Trials reported brief pinpoint skin bleeding, mild swelling, redness, bruising, fainting and migraine.
All the numbers and the full appraisal
Every reported result
The results, in numbers.
Nine single-centre RCTs, 530 adults with upper- or lower-limb tendinopathy; diagnosis and interventions varied.
Pooled evidence from five trials in the review; trial-specific sample sizes are in Table 1 and not summed here because records may not contribute to every pooled endpoint.
Head-to-head pooled comparator; trial-specific samples vary by endpoint.
Four trials combined this modality with exercise or another conservative treatment; no common co-intervention across trials.
Matched modality comparator in the combined-care stratum; background treatment varied by trial.
Study-specific VAS or NRS for pain1 reported outcome
Pain intensity, monotherapy
Study-specific VAS or NRS for pain SMD
RSWT versus FSWT; reported narrative says the result favors FSWT but signed direction coding is not explicit
Statistically significant pooled difference (p=.006); source authors describe an FSWT advantage.
Analysis details and source
95% CI is not stated in accessible Results prose. I²=59.32% (p=.02), moderate heterogeneity. Mid-term SMD .73 (p=.008); short-term .32 (p=.17); subgroup difference p=.33. Direction of signed SMD not explicitly defined. Egger's test was significant (t=3.17; p=.02); the authors say publication bias cannot be ruled out. The number of trials pooled is not printed (the Fig. 4A forest plot is not in the XML); Table 1 lists a pain measure for all five monotherapy trials, and the Results confirm only that Król et al. contributed (sensitivity analysis).
Study-specific pain scale at rest1 reported outcome
Rest pain, combined conservative care
Study-specific pain scale at rest SMD
RSWT versus FSWT with another conservative treatment
2 studies
No statistically significant pooled difference (p=.06); certainty low.
Analysis details and source
95% CI not stated in accessible Results prose. I²=94.85% (p<.001), considerable heterogeneity. Sensitivity analysis excluding Akınoğlu et al. gave SMD −1.89 (p<.0001; I²=0%). Egger's test was significant (t=−5.38; p=.03); the authors say publication bias cannot be ruled out.
Victorian Institute of Sport Assessment–Patella (VISA-P)1 reported outcome
VISA-P, combined conservative care
Victorian Institute of Sport Assessment–Patella (VISA-P) MD points
RSWT versus FSWT with another conservative treatment; the review's Discussion attributes the VISA-P improvement to FSWT
2 studies
The review reports a statistically significant pooled difference (p=.03) and states that FSWT was associated with a significant improvement in VISA-P scores. The evidence is low quality, and Table 2 summarises the two patellar tendinopathy trials as finding RSWT superior (Şah & Delen) and both modalities equally effective (van der Worp et al.).
Analysis details and source
95% CI not stated in the text. I²=80.72% (p<.001), substantial heterogeneity; low-certainty evidence. Subgroup difference by follow-up p=.97. Egger's test was significant (t=−8.03; p=.0005); the authors say publication bias cannot be ruled out. Table 2 gives the two trials' own conclusions: "RSWT showing superior efficacy" (Şah & Delen) and "equally effective" (van der Worp et al.).
Dynamometer assessment of wrist-extensor strength1 reported outcome
Wrist extensor strength, affected limb, monotherapy
Dynamometer assessment of wrist-extensor strength MD; source unit not stated
RSWT versus FSWT monotherapy
2 studies
Reported estimate and confidence interval.
Statistically significant pooled difference reported; the review interprets this as possible RSWT advantage, based on few studies.
Analysis details and source
The 95% CI (0.97 to 2.64) is given in the Abstract, not in the Results prose. I²=0% (p=.67). Subgroups: MD 1.48 (p=.003) short-term and 2.68 (p=.001) mid-term; subgroup difference p=.21. Symptom-duration groups MD 3.15 (p=.001) acute and 1.47 (p=.0021) chronic; subgroup p=.12. Repeated observations from two trials raise dependence concerns.
Dynamometer assessment of grip strength1 reported outcome
Grip strength, affected limb, monotherapy
Dynamometer assessment of grip strength MD; source unit not stated
RSWT versus FSWT monotherapy
2 studies
Reported estimate and confidence interval.
No statistically significant difference (p=.62); review rates certainty moderate.
Analysis details and source
The 95% CI (−1.68 to 2.82) is given in the Abstract, not in the Results prose. I²=0% (p=.88). Subgroup tests by follow-up p=.97 and symptom duration p=.18. Sensitivity estimate MD .16; p=.92. Repeated rows from two trials raise dependence concerns.
Study-reported pain during shockwave application1 reported outcome
Participants reporting pain during shockwave application
Study-reported pain during shockwave application RR
RSWT versus FSWT monotherapy
No statistically significant pooled difference (p=.18); low-certainty evidence.
Analysis details and source
95% CI not stated in accessible Results prose. I²=0% (p=.77). The number of trials pooled is not printed (the Fig. 5C forest plot is not in the XML); Table 1 lists no measure of pain during application for any monotherapy trial, and the Results confirm only that Król et al. contributed (sensitivity analysis).
Adverse effects reported in trial reports1 reported outcome
Participants experiencing adverse effects
Adverse effects reported in trial reports RR
RSWT versus FSWT monotherapy
No statistically significant pooled difference (p=.76); evidence certainty low.
Analysis details and source
95% CI not stated in accessible Results prose. Source gives I² as “%” with no numeric value in the main text. Harms listed include transient petechiae, mild edema, redness, hematoma, syncope, and migraine. The number of trials pooled is not printed (the Fig. 5D forest plot is not in the XML); the Discussion cites four trials for these adverse effects (references 9, 10, 23 and 24), and Table 1 lists side effects or adverse events as outcomes of Abo Al-Khair et al. and Li et al.
How treatment was delivered
Across trials, protocols ranged from 1,500 to 4,000 shocks (Król et al. RSWT: 2,000 + 2,000), 3.2–10 Hz, 0.09–0.3 mJ/mm² (Kaplan et al. and Şah & Delen print 0.02–0.60 mJ/mm² for FSWT) or approximately 1.2–4 bar, and 3–4 sessions once a week (Maher et al.: one session in alternate weeks, maximum 3–5; Tables 1–2 give no session count for Kaplan et al. or Şah & Delen); no pooled modality-level dose was specified.
Calcific shoulder trial: 1,500 shocks at 4 Hz, 0.3 mJ/mm², 4 weekly sessions; sonographic marking of calcification.
Calcific shoulder trial: 2,000 shocks at 10 Hz, 2.5 bar, 4 weekly sessions; sonographic marking of calcification.
Four studies combined shockwave with exercise or conventional physiotherapy; these were analyzed separately from monotherapy.
No significant pooled difference in application-related pain or adverse effects. Safety certainty was low; study reports included transient petechiae, mild edema, erythema, hematoma, syncope and migraine.
Pooled comparisons combine varied diagnoses, device parameters, outcome scales and follow-up times. The review does not state a dependence adjustment for repeated timepoint rows; few trials can contribute many rows. Exact CIs and visual direction could not be checked because forest plots and supplements were unavailable in retrieved XML. Signed pain SMD direction is not explicit in accessible narrative. The Discussion attributes the pooled VISA-P difference to FSWT, while Table 2 summarises the two patellar trials as finding RSWT superior (Şah & Delen) and both modalities equally effective (van der Worp et al.); that evidence is low quality and Egger's test was significant (p=.0005).
Most pooled findings have low or very low certainty; the source itself reports missing numeric I² for adverse effects (“I² = %”). Forest-plot images and supplementary tables were cited but not part of the retrieved source copy.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Pooled comparisons combine varied diagnoses, device parameters, outcome scales and follow-up times. The review does not state a dependence adjustment for repeated timepoint rows; few trials can contribute many rows. Exact CIs and visual direction could not be checked because forest plots and supplements were unavailable in retrieved XML. Signed pain SMD direction is not explicit in accessible narrative. The Discussion attributes the pooled VISA-P difference to FSWT, while Table 2 summarises the two patellar trials as finding RSWT superior (Şah & Delen) and both modalities equally effective (van der Worp et al.); that evidence is low quality and Egger's test was significant (p=.0005).
- Most pooled findings have low or very low certainty; the source itself reports missing numeric I² for adverse effects (“I² = %”). Forest-plot images and supplementary tables were cited but not part of the retrieved source copy.
Source-linked critical reading
Systematic review with meta-analysis
Technical source record
Full-text XML read, including all sections, Tables 1–3, all Results, Discussion, Limitations and Conclusion, plus five figure captions. Supplementary figures and GRADE table were referenced but not included in the retrieved source; the forest-plot images were not available for visual inspection. Follow-up attempts at official PMC image URLs and Europe PMC supplementaryFiles ZIP did not return inspectable assets (PMC page reCAPTCHA; image/PDF URLs inaccessible to reader; OA package 404; Europe PMC request timed out).
This review does not establish one shockwave modality as best across tendon conditions. It reports possible advantages for focused treatment in some pain/function comparisons and radial treatment for wrist-extensor strength, but pooled estimates are uncertain and affected by heterogeneity and reporting concerns. The result for a specific tendon problem should be judged from the relevant trial, not the overall pooled headline.
Among adults with upper- or lower-limb tendinopathy, how do radial shockwave therapy (RSWT) and focused shockwave therapy (FSWT) compare for pain, treatment success, function, strength, range of motion, perceived improvement and harms, as monotherapies or alongside conservative care?
PRISMA 2020 systematic review and random-effects meta-analysis of nine RCTs (530 participants), searched to March 2025; protocol CRD420251011209 registered 24 March 2025. Follow-up-duration and symptom-duration subgroup analyses were added beyond the registration. Two investigators independently screened and extracted data; three reviewers rated trial quality.
Nine single-centre RCTs covered 530 adults with heterogeneous tendon conditions: lateral epicondylitis (tennis elbow), calcific shoulder, non-calcific rotator cuff and patellar tendinopathies. Five compared modalities as monotherapies and four combined each modality with another conservative treatment. These are pooled across different trials and diagnoses, not one cohort.
RSWT and FSWT differed by device, anatomical target, pressure/energy, impulse count, frequency and sessions. Table 1 includes four weekly sessions for calcific shoulder (FSWT 1,500 shocks, 0.3 mJ/mm²; RSWT 2,000 shocks, 2.5 bar) and noncalcific rotator cuff (both 3,000 shocks, four weekly sessions, with modality-specific frequency/energy). Tennis-elbow and patellar protocols also varied. In four studies each modality accompanied exercise or physiotherapy. Exact details are study-specific, not one review dose.
The pooled contrasts were RSWT versus FSWT within monotherapy or matched combined-care strata. Individual trials also included sham or conservative-care arms, but those were not the pooled head-to-head contrast.
Primary outcomes included pain, treatment success, patient-reported function/disability, strength, range of motion and perceived improvement; secondary outcomes were treatment-application pain and adverse effects. Continuous effects were MD or SMD, dichotomous effects RR, generally with 95% CIs in the forest plots; the retrieved XML narrative gives several pooled estimates and p values but omits their CIs, except the Abstract's 95% CIs for wrist-extensor and grip strength. Main forest-plot pixels and supplements were not available for visual verification.
The review reports an FSWT-favoring pain difference for monotherapy (SMD 0.49, p=.006; I²=59.32%), but accessible narrative does not define the scale-direction coding for the signed SMD. Combined-care rest pain was not statistically different (SMD −1.27, p=.06; I²=94.85%). Combined-care VISA-P differed by MD 7.67 (p=.03; I²=80.72%), which the Discussion attributes to FSWT; this evidence is low quality, Egger's test was significant (p=.0005), and Table 2 summarises the two patellar trials as finding RSWT superior (Şah & Delen) and both modalities equally effective (van der Worp et al.). RSWT favored wrist-extensor strength (MD 1.81, p<.001; I²=0%), while grip strength was not different (MD 0.57, p=.62; I²=0%). No significant differences in treatment pain (RR 1.17, p=.18) or adverse events (RR .87, p=.76). Apart from the Abstract's 95% CIs for wrist-extensor strength (0.97 to 2.64) and grip strength (−1.68 to 2.82), CIs for these pooled estimates are not in the narrative.
Pooled treatment-application pain was not significantly different (RR 1.17, p=.18); pooled adverse effects were also not different (RR .87, p=.76). The review lists transient petechiae, minor oedema, redness, hematoma, syncope and migraine in primary reports. Certainty for safety outcomes was low.
Few and small direct trials, heterogeneous diagnoses, protocols, scales and timepoints; most outcomes had low/very low certainty. The analysis states random-effects pooling but does not describe adjustment for repeated timepoints from the same participants, although its follow-up and symptom-duration subgroups imply several rows per trial. Forest plot images and referenced supplementary figures/tables were absent from the retrieved source, so exact CIs, row-level sample sizes and visual direction could not be checked. The positive signed SMD for monotherapy pain lacks explicit direction coding in narrative; the VISA-P advantage the Discussion assigns to FSWT rests on low-quality evidence with a significant Egger's test (p=.0005), and Table 2 summarises the two patellar trials as finding RSWT superior and both modalities equally effective. The source reports I² for harms as “%” without a numeric value.
These findings address head-to-head radial-versus-focused treatment, but pool distinct tendons, regimens and co-interventions. They cannot determine a best device or dose for an individual diagnosis. The review excluded languages other than English, Polish and Slovak, and certainty is mostly low.
Use adequately powered, diagnosis-specific head-to-head RCTs with prespecified clinical timepoints, one independent contribution per cohort or a justified dependence model, transparent dose parameters and full reporting of group data and CIs. Report the trial-level VISA-P data behind the pooled result and make data/code and supplements available.
Read the original article.
Efficacy of radial and focused shockwave therapy for tendinopathy: a systematic review and meta-analysis.
Stania M, Pawłowski M, Benduch M, Dudon A, Hirjaková Z, Bzdúšková D, Kimijanová J. · Scientific reports · 2026
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →