Greater trochanteric pain / gluteal tendinopathy · Randomized trial · 2023
How does shockwave compare with a steroid injection for outer hip pain?
In this trial of 60 people with outer hip pain, three sessions of radial shockwave, one a week, lowered pain from about 7 to about 1 out of 10 by three months. A single steroid injection gave a similar result, from about 7 to under 1.
Both helped- people with outer hip pain
- 60
- sessions, one a week
- 3
- months of follow-up
- 3
Good to know: There was no exercise programme during the study.
Çağlar Yağcı H, Yağcı İ, Bağcıer F Turkish Journal of Physical Medicine and Rehabilitation, 2023 · Read the paper ↗
What happened
Outer hip pain
0–10 pain scale · start and after three months
StartAfter 3 months
Show the numbers
| Group | People | Start | After 3 months |
|---|---|---|---|
| Radial shockwave | 32 | 6.84 | 1.03 |
| Steroid injection | 28 | 6.96 | 0.78 |
The treatment
Radial shockwave
Radial, once a week for 3 weeks
- Sessions
- 3, one a week
- Pulses
- 2,000 per session
- Strength
- 2bar · 12 per second
- Numbing
- Not mentioned
- Device
- Modus
- Everyone also got
- Paracetamol allowed if needed; no exercise programme
Compared with: One steroid injection (40 mg triamcinolone with a local anaesthetic)
Settings are what each study used, not a prescription.
Who took part
60 adults aged 34 to 65 with outer hip pain (greater trochanteric pain syndrome) diagnosed by examination.
Side effects
Some people had short-lived pain after a session that usually settled within 24 hours; 12 people in the shockwave group took paracetamol. Nobody stopped shockwave early.
All the numbers and the full appraisal
The results, in numbers.
Adults with clinically diagnosed GTPS (ages 34–65, although Methods screened patients aged 40–70). Of 96 patients screened, 64 were eligible; groups are reported as SWT n=32 and corticosteroid injection n=28 (60 in total). Four injection-group patients are then said to have been excluded (Methods: missed follow-up visits; Results: started NSAIDs for other diseases), yet Tables 1–2 keep n=28 and the completed sample is 60; the randomized, excluded and analyzed counts do not reconcile.
Table 2 uses n=32 and n=28. Methods report 64 eligible patients assigned as 32 and 28 (60 in total), then four injection-group exclusions (missed follow-up in Methods; NSAID use in Results) while still reporting a completed sample of 60; the source does not reconcile these counts, so the randomized allocation is not recoverable.
Table 2 uses n=32 and n=28. Methods report 64 eligible patients assigned as 32 and 28 (60 in total), then four injection-group exclusions (missed follow-up in Methods; NSAID use in Results) while still reporting a completed sample of 60; the source does not reconcile these counts, so the randomized allocation is not recoverable.
Overall lateral hip pain2 reported outcomes
Overall lateral hip pain
VAS, 0–10 cm cm
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 6.84 (SD 0.95) | 1.03 (SD 1.03) |
| Triamcinolone plus prilocaine injection | 28 | 6.96 (SD 0.99) | 0.78 (SD 0.83) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=0.789). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test was not significant (p=0.378); both groups improved over time (p=0.001). This single test covers all three visits, so the same p value appears with the week-3 outcome; the source gives no between-group difference or confidence interval for any single visit.
Analysis details and source
The paper reports repeated-measures tests, not an adjusted mean difference or confidence interval. Nonsignificance does not establish equivalence. Table 2 prints the same shockwave means (6.84 → 1.03) for pain as for tenderness (the SDs differ); transcribed as printed.
Greater-trochanter tenderness
VAS, 0–10 cm cm
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 6.84 (SD 1.01) | 1.03 (SD 0.99) |
| Triamcinolone plus prilocaine injection | 28 | 7 (SD 1.05) | 0.78 (SD 0.73) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=1.711). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test was not significant (p=0.185); both groups improved over time (p=0.001). This single test covers all three visits, so the same p value appears with the week-3 outcome; the source gives no between-group difference or confidence interval for any single visit.
Analysis details and source
The paper reports repeated-measures tests, not an adjusted mean difference or confidence interval. Nonsignificance does not establish equivalence. The shockwave means (6.84 → 1.03) match the shockwave pain row (the SDs differ), and the injection week-3 and month-3 values are both 0.78 ± 0.73; transcribed as printed.
WOMAC total score1 reported outcome
WOMAC total score
WOMAC 0–96; higher worse WOMAC points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 47 (SD 8.8) | 6.73 (SD 6.54) |
| Triamcinolone plus prilocaine injection | 28 | 46.78 (SD 9.56) | 4.27 (SD 4.35) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=1.153). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test was not significant (p=0.287); both groups improved over time (p=0.001). This single test covers all three visits, so the same p value appears with the week-3 outcome; the source gives no between-group difference or confidence interval for any single visit.
Analysis details and source
The paper reports repeated-measures tests, not an adjusted mean difference or confidence interval. Nonsignificance does not establish equivalence.
VAS pain at week 31 reported outcome
VAS pain at week 3
Visual Analog Scale (VAS) centimeters
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | — | 1.09 (SD 0.96) |
| Triamcinolone plus prilocaine injection | 28 | — | 0.78 (SD 0.73) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=0.789). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
Week-3 means from Table 2. The between-group p value (0.378, F=0.789) is the same single repeated-measures test reported with the baseline-to-month-3 pain outcome, not a separate week-3 comparison; both groups improved over time (p=0.001).
Analysis details and source
Table 2 lists F=0.789 and p=0.378 for the between-group test (**), and p=0.001 for change over time (*). Do not mistake the F statistic 0.789 for the between-group p value. The injection value 0.78 ± 0.73 also appears in the injection tenderness rows at week 3 and month 3; transcribed as printed.
Greater trochanter tenderness at week 31 reported outcome
Greater trochanter tenderness at week 3
Trochanteric tenderness VAS centimeters
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | — | 1.18 (SD 1.06) |
| Triamcinolone plus prilocaine injection | 28 | — | 0.78 (SD 0.73) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=1.711). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
Week-3 means from Table 2. The between-group p value (0.185, F=1.711) is the same single repeated-measures test reported with the baseline-to-month-3 tenderness outcome, not a separate week-3 comparison; both groups improved over time (p=0.001).
Analysis details and source
Table 2 footnote marks the within-time test with * (p=0.001) and the between-group test with ** (p=0.185); do not conflate them. Values are means±SD. The injection value is identical at week 3 and month 3 (0.78 ± 0.73); transcribed as printed.
WOMAC at week 31 reported outcome
WOMAC at week 3
Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | — | 7.59 (SD 7.59) |
| Triamcinolone plus prilocaine injection | 28 | — | 5.1 (SD 3.83) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: one between-group test from the repeated-measures model covering baseline, week 3 and month 3 (Table 2, **; F=1.153). Table 2 prints it in the week-3 row; it is not a test at a single visit.
General linear model for repeated measures: between-group effect (Table 2, **)
Week-3 means from Table 2. The between-group p value (0.287, F=1.153) is the same single repeated-measures test reported with the baseline-to-month-3 WOMAC outcome, not a separate week-3 comparison; both groups improved over time (p=0.001).
Analysis details and source
Table 2 footnote marks the within-time test with * (p=0.001) and the between-group test with ** (p=0.287); do not conflate them. Values are means±SD. The shockwave SD equals its mean (7.59 ± 7.59); transcribed as printed.
Emotional role limitation1 reported outcome
Emotional role limitation
SF-36 emotional role limitation SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 58.5 (SD 19.67) | 59.21 (SD 18.6) |
| Triamcinolone plus prilocaine injection | 28 | 59.22 (SD 19.83) | 60.92 (SD 18.24) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.805). Table 2 reports p=0.069 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.069 with * and between-group p=0.805 with **.
Energy/vitality1 reported outcome
Energy/vitality
SF-36 energy SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 59.06 (SD 15.78) | 61.25 (SD 13.55) |
| Triamcinolone plus prilocaine injection | 28 | 54.82 (SD 12.58) | 57.32 (SD 10.49) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.859). Table 2 reports p=0.226 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.226 with * and between-group p=0.859 with **.
Mental health1 reported outcome
Mental health
SF-36 mental health SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 63.84 (SD 13.72) | 63.96 (SD 14.18) |
| Triamcinolone plus prilocaine injection | 28 | 64.28 (SD 11.77) | 64.35 (SD 11.93) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.902). Table 2 reports p=0.687 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.687 with * and between-group p=0.902 with **.
Social function1 reported outcome
Social function
SF-36 social function SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 53.64 (SD 15.8) | 59.89 (SD 13.09) |
| Triamcinolone plus prilocaine injection | 28 | 56.62 (SD 12.64) | 60.91 (SD 10.24) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.546). Table 2 reports p=0.279 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.279 with * and between-group p=0.546 with **.
General health1 reported outcome
General health
SF-36 general health SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 55.93 (SD 15.26) | 56.28 (SD 14.76) |
| Triamcinolone plus prilocaine injection | 28 | 56.07 (SD 13.28) | 56.03 (SD 12.5) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.287). Table 2 reports p=0.783 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.783 with * and between-group p=0.287 with **.
Physical functioning1 reported outcome
Physical functioning
SF-36 physical functioning SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 58.9 (SD 13.42) | 70.9 (SD 13.42) |
| Triamcinolone plus prilocaine injection | 28 | 58.75 (SD 11.98) | 70.75 (SD 11.98) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.962). Table 2 reports p=0.001 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.001 with * and between-group p=0.962 with **. Each group's SD is identical before and after treatment (13.42; 11.98); transcribed as printed.
Role limitations due to physical health1 reported outcome
Role limitations due to physical health
SF-36 role limitations due to physical health SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 36.71 (SD 16.78) | 43.81 (SD 16.9) |
| Triamcinolone plus prilocaine injection | 28 | 36.6 (SD 14.4) | 45.42 (SD 14.67) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.853). Table 2 reports p=0.001 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.001 with * and between-group p=0.853 with **.
Bodily pain1 reported outcome
Bodily pain
SF-36 bodily pain SF-36 score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Radial ESWT | 32 | 46.18 (SD 11.06) | 53.34 (SD 10.77) |
| Triamcinolone plus prilocaine injection | 28 | 45.75 (SD 12.4) | 49.64 (SD 9.4) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radial ESWT versus corticosteroid injection: between-group test from the repeated-measures model covering baseline and month 3 (Table 2, **), printed in the month-3 row.
General linear model for repeated measures: between-group effect (Table 2, **)
The repeated-measures between-group test across baseline and month 3 was not statistically significant (p=0.432). Table 2 reports p=0.001 for change over time (*).
Analysis details and source
Baseline and month-3 means±SD are transcribed by group. Table 2 marks within-time p=0.001 with * and between-group p=0.432 with **.
How treatment was delivered
Radial extracorporeal shockwave therapy using Modus device
2,000 beats, 20-mm applicator, 12 Hz, 2 bar
One session weekly for three weeks, greater trochanter region, lateral decubitus
One injection: 40 mg triamcinolone acetonide plus 2 mL 2% prilocaine
Paracetamol 1,000 mg/day permitted; no exercise program during study
Twelve ESWT participants used paracetamol; transient post-procedure pain generally resolved within 24 hours. No ESWT discontinuations. Four injection participants were excluded after starting NSAIDs for other illnesses; this is not described as an injection adverse event.
Only three months of follow-up; long-term efficacy and recurrence are unknown.
Four injection-group participants were reported excluded after randomization, with different reasons in Methods (missed follow-up visits) and Results (NSAID use), yet the reported group sizes of 32 and 28 and the 64 eligible patients do not add up; the randomized denominators are not reported consistently, and no intention-to-treat analysis is reported.
No imaging confirmation separated tendinopathy from bursitis; no confidence intervals for between-group comparisons were reported.
Several Table 2 cells repeat exactly: shockwave pain and tenderness have the same means (6.84 → 1.03); the injection value 0.78 ± 0.73 appears for tenderness at week 3 and month 3 and for pain at week 3; shockwave WOMAC at week 3 is 7.59 ± 7.59; SF-36 physical-function SDs are identical before and after treatment, and both groups rise by exactly 12.00 points; and the SF-36 general-health between-group F and p (1.153; 0.287) repeat WOMAC's. They are transcribed as printed; the pattern suggests possible copying errors in the published table.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Only three months of follow-up; long-term efficacy and recurrence are unknown.
- Four injection-group participants were reported excluded after randomization, with different reasons in Methods (missed follow-up visits) and Results (NSAID use), yet the reported group sizes of 32 and 28 and the 64 eligible patients do not add up; the randomized denominators are not reported consistently, and no intention-to-treat analysis is reported.
- No imaging confirmation separated tendinopathy from bursitis; no confidence intervals for between-group comparisons were reported.
- Several Table 2 cells repeat exactly: shockwave pain and tenderness have the same means (6.84 → 1.03); the injection value 0.78 ± 0.73 appears for tenderness at week 3 and month 3 and for pain at week 3; shockwave WOMAC at week 3 is 7.59 ± 7.59; SF-36 physical-function SDs are identical before and after treatment, and both groups rise by exactly 12.00 points; and the SF-36 general-health between-group F and p (1.153; 0.287) repeat WOMAC's. They are transcribed as printed; the pattern suggests possible copying errors in the published table.
Source-linked critical reading
Randomized comparative clinical trial
Technical source record
Complete article XML read, including Methods, Results, Tables 1–2, Discussion, Conclusion and disclosures.
Both options were followed by improvement in this small active-comparator study, with no detected between-group difference through three months. The evidence does not prove equal benefit; discuss the distinct procedures, preferences and uncertainty with a clinician.
In adults with clinically diagnosed greater trochanteric pain syndrome, how did three weekly radial shockwave sessions compare with one corticosteroid/local-anesthetic injection through three months?
Single-center, single-blind randomized active-comparator study. Sixty participants completed follow-up: 32 in ESWT and 28 in injection; four injection participants were excluded after starting NSAIDs. Initial randomized denominators are not explicitly stated in the reported results.
Adults aged 40–70 with at least six weeks of lateral hip pain and clinical provocation/tenderness criteria. Imaging was not used to distinguish bursal from tendon pathology.
Radial Modus device; lateral decubitus; 2,000 beats using a 20-mm applicator, 12 Hz, 2 bar, once weekly for three weeks. Paracetamol 1,000 mg/day was permitted.
Single injection of 40 mg triamcinolone acetonide plus 2 mL of 2% prilocaine at the most tender point. No exercise program was provided during the study.
Pain, trochanteric tenderness and WOMAC were assessed at baseline, week 3 and month 3, and SF-36 at baseline and month 3. Primary outcomes improved within both groups; ESWT arm mean pain was 6.84 cm at baseline and 1.03 at month 3; injection was 6.96 and 0.78.
The reported between-group tests were nonsignificant for VAS (p=.378), tenderness (p=.185), WOMAC (p=.287), and SF-36 subscales (all p>.05). These tests do not establish equivalence or noninferiority.
Twelve ESWT participants reported paracetamol use; treatment-related pain generally resolved within 24 hours and was tolerated. Four injection participants were excluded after using NSAIDs for other conditions. No patient discontinued ESWT.
Follow-up ended at three months; four injection participants were excluded, and imaging confirmation was absent. The study did not report between-group effect estimates with confidence intervals.
This trial informs a short-term choice between one injection and a three-session radial protocol for clinically diagnosed GTPS. It does not show that either treatment is better than no treatment, nor that they are interchangeable.
A larger registered trial should retain randomized participants, use intention-to-treat analysis, prespecify a clinically justified noninferiority/equivalence margin if intended, report confidence intervals and track recurrence, medication use and adverse events beyond three months.
Read the original article.
Comparison of shock wave therapy and corticosteroid injection in the treatment of greater trochanteric pain syndrome: A single-blind, randomized study.
Çağlar Yağcı H, Yağcı İ, Bağcıer F. · Turkish journal of physical medicine and rehabilitation · 2023
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →