Greater trochanteric pain / gluteal tendinopathy · Review · 2024
How does shockwave compare with exercise for outer hip pain?
In this review, two trials with 184 people compared shockwave with an exercise programme for outer hip pain. At least 28 weeks later, hip pain was similar in the two groups.
Similar to exercise- trials in the review (733 people)
- 6
- trials compared shockwave with exercise
- 2
- weeks of follow-up
- 28+
Good to know: The two trials used different kinds of shockwave (radial in one, focused in the other) and different schedules.
Kjeldsen T, Hvidt KJ, Bohn MB, et al. Physiotherapy, 2024 · 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
- trials
- 2
- people
- 184
- Compared with
- An exercise programme
Long-term hip pain: shockwave or exercise
Difference on a 0–10 pain scale · 28 weeks or more after treatment
Show the numbers
| Difference (points on a 0–10 pain scale) | likely range | |
|---|---|---|
| Hip pain, 28+ weeks later | +0.32 | −0.47 to 1.12 |
What the trials used
- Rompe 2009 radial shockwave protocol
- Handheld device; area of maximal tenderness treated circumferentially; 3 sessions/week; session duration, impulses, energy and total treatment duration not reported.
- Notarnicola 2023 focal shockwave protocol
- Focal ESWT at greater trochanter and gluteal tendon enthesis; once weekly for 3 weeks; impulses, energy, frequency and session duration not reported. Crossover design; review counts crossovers as dropouts.
- Shockwave-related comparator and co-interventions
- Rompe: exercise vs radial shockwave; co-intervention guidance included avoidance of co-interventions. Notarnicola: exercise vs focal shockwave; common background care not specified in Table 2.
Who took part
Adults with long-lasting outer hip pain in six trials; two of them compared exercise with shockwave.
Side effects
No serious side effects were reported. One trial noted skin irritation with shockwave; the other did not report side effects.
All the numbers and the full appraisal
Every reported result
The results, in numbers.
Six RCTs (733 participants) evaluated exercise for clinically diagnosed GTPS, gluteal tendinopathy or lateral hip pain. Two trials compared exercise with shockwave: Rompe 2009 and Notarnicola 2023.
Hip pain1 reported outcome
Exercise vs sham/wait-and-see: hip pain
Hip pain SMD
Exercise versus sham or wait-and-see; positive favors exercise by review direction
3 studies · 356 participants in this estimate
Reported estimate and confidence interval.
Small long-term improvement; low certainty.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed.
Patient-reported physical function1 reported outcome
Exercise vs sham/wait-and-see: self-reported physical function
Patient-reported physical function SMD
Exercise versus sham or wait-and-see; positive favors exercise by review direction
3 studies · 356 participants in this estimate
Reported estimate and confidence interval.
Slight long-term improvement; moderate certainty.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed.
Exercise vs sham/wait-and-see: disease severity2 reported outcomes
Exercise vs sham/wait-and-see: disease severity
VISA-G MD
Exercise versus sham or wait-and-see; source direction favors exercise
3 studies · 356 participants in this estimate
Reported estimate and confidence interval.
Slight reduction in disease severity; moderate certainty.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed.
Isotonic vs isometric exercise: VISA-G
VISA-G score
Both groups seemed to improve slightly; no between-group difference was found.
Analysis details and source
One trial, n=30; not a shockwave comparison.
Exercise vs sham/wait-and-see: global response2 reported outcomes
Exercise vs sham/wait-and-see: global response
Global Rating of Change RR
Exercise versus sham or wait-and-see
3 studies · 333 participants in this estimate
Reported estimate and confidence interval. Logarithmic ratio scale.
May increase response, but CI includes no difference; low certainty.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed.
Exercise vs corticosteroid injection: global response
Global Rating of Change RR
Exercise versus a single corticosteroid injection
2 studies · 186 participants in this estimate
Reported estimate and confidence interval. Logarithmic ratio scale.
May increase meaningful response; low certainty.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed.
Hip pain (NRS 0–10)1 reported outcome
Exercise vs shockwave: long-term hip pain
Hip pain (NRS 0–10) MD
Exercise versus shockwave; source says positive direction favors shockwave
2 studies · 184 participants in this estimate
Reported estimate and confidence interval.
Little-to-no difference; very-low certainty, downgraded for very high risk of bias and imprecision.
Analysis details and source
Meta-analysis from the review; no trial-arm means reconstructed. The review gives I² 82.8% as the reason only long-term hip pain was pooled for this comparison (Results, Exercise compared to shockwave therapy). Its Methods exclude comparisons with I² above 50% from meta-analysis, so that value refers to the exercise–shockwave comparison(s) left unpooled; the text does not report the I² of this pooled estimate (Fig. 5 is not in the snapshot), and Table 3 does not downgrade it for inconsistency.
How treatment was delivered
Handheld device; area of maximal tenderness treated circumferentially; 3 sessions/week; session duration, impulses, energy and total treatment duration not reported.
Focal ESWT at greater trochanter and gluteal tendon enthesis; once weekly for 3 weeks; impulses, energy, frequency and session duration not reported. Crossover design; review counts crossovers as dropouts.
Rompe: exercise vs radial shockwave; co-intervention guidance included avoidance of co-interventions. Notarnicola: exercise vs focal shockwave; common background care not specified in Table 2.
No serious adverse events were reported across trials. Reporting was incomplete: Rompe lists skin irritation with shockwave; adverse events were not reported for Notarnicola. No serious events does not establish zero risk.
Only two trials compared exercise with shockwave, and both were judged high risk of bias. Because of substantial heterogeneity (I²=82.8%), the review pooled only long-term hip pain, and that pooled result is very-low certainty.
The shockwave protocols differ in modality (radial vs focal), and the review does not report energy, pulse count or frequency. The preserved text snapshot does not include figure pixels for visual cross-check of forest-plot geometry.
No serious adverse events were reported, but not all studies fully reported harms; one trial did not report events and another reported only serious events.
The review included only English, Swedish, Norwegian or Danish reports and few RCTs; exercise fidelity was not reported.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Only two trials compared exercise with shockwave, and both were judged high risk of bias. Because of substantial heterogeneity (I²=82.8%), the review pooled only long-term hip pain, and that pooled result is very-low certainty.
- The shockwave protocols differ in modality (radial vs focal), and the review does not report energy, pulse count or frequency. The preserved text snapshot does not include figure pixels for visual cross-check of forest-plot geometry.
- No serious adverse events were reported, but not all studies fully reported harms; one trial did not report events and another reported only serious events.
- The review included only English, Swedish, Norwegian or Danish reports and few RCTs; exercise fidelity was not reported.
Source-linked critical reading
Systematic review with meta-analysis
Technical source record
Complete article text snapshot read across 12 pages/797 lines, including all Methods, Results, Discussion, Tables 1–3 and figure captions. The publisher PDF pixels were not available for visual inspection; forest plots therefore could not be visually cross-checked.
This review found very uncertain evidence that long-term hip pain differs between exercise and shockwave therapy. It cannot tell whether the treatments are equivalent or which shockwave protocol is preferable.
In clinically diagnosed GTPS, how does exercise compare with control, corticosteroid injection, shockwave therapy, or another exercise program?
Systematic review and meta-analysis of RCTs, registered prospectively in PROSPERO (CRD42021261380). Six RCTs (733 participants); five contributed to meta-analyses. Searches updated September 2023.
Trials enrolled people with clinically diagnosed GTPS, gluteal tendinopathy, or lateral hip pain; age means 46–62.5 years and 69–100% were women. Criteria required palpation pain and pain within 30 seconds of single-leg stance.
The shockwave comparators were heterogeneous: Rompe 2009 used radial shockwave over maximal tenderness, three times weekly, but did not report duration/session dose; Notarnicola 2023 used focal ESWT over the greater trochanter and gluteal tendon enthesis, once weekly for three weeks. Energy, impulses and frequency were not reported in the review table.
Exercise regimens included home exercise plus physiotherapy sessions; the two ESWT trials compared exercise with shockwave, while the review pooled them only for long-term hip pain. No common background care is established across the trials.
For exercise versus control at ≥52 weeks, Table 3 reports hip pain SMD 0.24 (95% CI 0.01–0.46), physical function SMD 0.30 (0.09–0.51), VISA-G disease severity MD 5.67 (2.03–9.30), and global response RR 1.23 (0.94–1.59). For exercise versus corticosteroid, global response RR 1.51 (1.24–1.83). For exercise versus shockwave, long-term hip-pain MD 0.32 (−0.47 to 1.12), favoring shockwave by sign convention.
Exercise versus shockwave showed little or no long-term difference in hip pain; the estimate is imprecise and very-low certainty. Both contributing trials had high risk of bias, and substantial heterogeneity (I² 82.8%) meant that only the long-term pain comparison was synthesized. Exercise versus a different exercise program (one 30-person trial) showed no between-group difference in VISA-G.
Across all modalities, no serious adverse events were found; one trial did not report adverse events and another reported only serious events. For shockwave, the review lists skin irritation in Rompe 2009; adverse events were not reported for Notarnicola 2023. These omissions do not establish zero risk.
The shockwave evidence is only two trials, both high risk of bias; certainty is very low. Protocols differ (radial vs focal), and key dose settings are missing. Only long-term hip pain was pooled, because of substantial heterogeneity (I² 82.8%). The review includes few trials, restricts languages, and reports no fidelity data for exercise. Figure/forest-plot images were not available for visual inspection in the preserved text snapshot.
The exercise-versus-shockwave estimate applies to trial populations with chronic, clinically diagnosed GTPS/lateral hip pain and long-term follow-up (≥28 weeks). It does not establish equivalence, preferred modality, or a device/dose protocol.
Trials should use clearly specified radial/focused devices and dose, consistent diagnostic criteria, adequate allocation concealment and blinding where feasible, prespecified outcomes and timing, intention-to-treat analyses, complete harms and adherence reporting, and enough participants to resolve the heterogeneous long-term comparison.
Read the original article.
Exercise compared to a control condition or other conservative treatment options in patients with Greater Trochanteric Pain Syndrome: a systematic review and meta-analysis of randomized controlled trials.
Kjeldsen T, Hvidt KJ, Bohn MB, Mygind-Klavsen B, Lind M, Semciw AI, Mechlenburg I. · Physiotherapy · 2024
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →