Plantar heel pain / plantar fasciopathy · Randomized trial · 2026
How does shockwave compare with a steroid injection and radiofrequency for heel pain?
In this trial, heel pain fell from about 8 to about 3.5 out of 10 a year after three shockwave sessions, one a week. People treated with radiofrequency, a needle-based heat treatment, did better, ending at about 1.6. Those who had a steroid injection ended at about 3.
Radiofrequency did better at 12 months- people (62 feet) with heel pain
- 49
- shockwave sessions, one a week
- 3
- months of follow-up
- 12
Good to know: Some people had both feet treated, and each foot was counted.
Tütüncü MN, Sekizkardeş M, Şahin E, et al. BMC Musculoskeletal Disorders, 2026 · Read the paper ↗
What happened
Heel pain
0–10 pain scale · start and after 12 months
StartAfter 12 months
Show the numbers
| Group | People | Start | After 12 months |
|---|---|---|---|
| Shockwave | — | 7.79 | 3.47 |
| Radiofrequency | — | 6.87 | 1.61 |
| Steroid injection | — | 8.1 | 3.05 |
The treatment
Shockwave
Shockwave, once a week for 3 weeks
- Sessions
- 3, one a week
- Pulses
- 2,000 per session
- Strength
- 4.0bar · 15 per second
- Numbing
- Not mentioned
- Device
- BTL-5000 SWT
- Everyone also got
- Anti-inflammatory tablets for five days, plus ice and stretching advice
Compared with: A steroid injection, or radiofrequency (a needle-based heat treatment)
Settings are what each study used, not a prescription.
Who took part
49 people (62 feet) with heel pain that had not settled with other care.
Side effects
The study does not report side effects.
All the numbers and the full appraisal
The results, in numbers.
Adults with chronic plantar heel pain refractory to conservative care; analyzed per foot. Narrative reports 20 CSI feet/18 patients, 19 ESWT feet/15 patients, and 23 radiofrequency feet/16 patients; Table 2 sex totals conflict.
The study analyzed feet, and some participants contributed both feet. Foot counts agree between the Results prose and Table 2 (20 corticosteroid, 19 ESWT, 23 radiofrequency), but Table 2's female/male figures sum to those foot counts rather than to the narrative patient counts (18, 15, 16); participant count is left null.
The study analyzed feet, and some participants contributed both feet. Foot counts agree between the Results prose and Table 2 (20 corticosteroid, 19 ESWT, 23 radiofrequency), but Table 2's female/male figures sum to those foot counts rather than to the narrative patient counts (18, 15, 16); participant count is left null.
The study analyzed feet, and some participants contributed both feet. Foot counts agree between the Results prose and Table 2 (20 corticosteroid, 19 ESWT, 23 radiofrequency), but Table 2's female/male figures sum to those foot counts rather than to the narrative patient counts (18, 15, 16); participant count is left null.
Visual Analogue Scale pain score1 reported outcome
Visual Analogue Scale pain score
Visual Analogue Scale (VAS), 10 cm VAS points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Corticosteroid injection | — | 8.1 (SD 1.373) | 3.05 (SD 1.432) |
| Extracorporeal shock-wave therapy | — | 7.79 (SD 1.084) | 3.47 (SD 2.01) |
| Radiofrequency thermal lesioning | — | 6.87 (SD 2.007) | 1.61 (SD 0.839) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radiofrequency had lower pain score than ESWT (reported difference 1.86 points; p<0.001).
Post-hoc comparison as reported by authors
Radiofrequency had lower pain score than ESWT (reported difference 1.86 points; p<0.001). Radiofrequency had lower pain score than corticosteroid injection (reported difference 1.44 points; p=0.006).
Analysis details and source
Group means and SDs are reported by foot; per-arm analyzed foot counts are not provided in the outcome table. Bilateral feet from some participants were treated as independent units; no clustering adjustment was reported.
Foot Function Index total score1 reported outcome
Foot Function Index total score
Foot Function Index (FFI), total FFI points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Corticosteroid injection | — | 133.4 (SD 24.4053) | 50.25 (SD 23.5816) |
| Extracorporeal shock-wave therapy | — | 122.579 (SD 29.1725) | 56.737 (SD 33.7505) |
| Radiofrequency thermal lesioning | — | 97.783 (SD 38.3559) | 27.043 (SD 12.4261) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radiofrequency was better than ESWT (p=0.001) and corticosteroid injection (p=0.008), per the narrative post-hoc results.
Post-hoc comparison as reported by authors
Radiofrequency was better than ESWT (p=0.001) and corticosteroid injection (p=0.008), per the narrative post-hoc results.
Analysis details and source
Group means and SDs are reported by foot; per-arm analyzed foot counts are not provided in the outcome table. Bilateral feet from some participants were treated as independent units; no clustering adjustment was reported.
Roles and Maudsley score1 reported outcome
Roles and Maudsley score
Roles and Maudsley score Score points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Corticosteroid injection | — | 3.5 (SD 0.6882) | 1.7 (SD 0.5712) |
| Extracorporeal shock-wave therapy | — | 3.526 (SD 0.513) | 1.895 (SD 0.9366) |
| Radiofrequency thermal lesioning | — | 2.696 (SD 0.5588) | 1.174 (SD 0.3876) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Radiofrequency was reported superior to both ESWT and corticosteroid injection at 12 months (p=0.002); scale direction is not stated in the source, so no scale range or direction is assigned here.
Post-hoc comparison as reported by authors
Radiofrequency was reported superior to both ESWT and corticosteroid injection at 12 months (p=0.002); scale direction is not stated in the source, so no scale range or direction is assigned here.
Analysis details and source
Group means and SDs are reported by foot; per-arm analyzed foot counts are not provided in the outcome table. Bilateral feet from some participants were treated as independent units; no clustering adjustment was reported.
How treatment was delivered
BTL-5000 SWT; focused modality is not stated explicitly in the protocol; ultrasound-guided plantar heel application.
15 Hz, 2,000 pulses, 4.0 bar; three sessions, one week apart. No bar-to-energy conversion is made.
Ultrasound-guided medial injection: 1 mL betamethasone 40 mg/mL plus 2 mL bupivacaine 5 mg/mL.
Single session; NeuroTherm JK4A; 22G, 10-cm cannula; 50 Hz; 90°C for 90 seconds, repeated at up to three pain sites; local entry anesthetic 1 mL lidocaine 2%.
All groups received NSAIDs for five days and advice for ice and stretching; no activity/weight-bearing restriction recommended.
No structured adverse-event counts are reported. The Discussion names known corticosteroid risks, including fat-pad atrophy and plantar-fascia rupture, but does not give study-specific arm counts.
Authors report no placebo arm, no double blinding, and subjective outcomes.
Bilateral feet from some participants were analyzed as independent units without adjustment for intra-person clustering.
Radiofrequency group had longer symptom duration at baseline; authors note possible confounding.
Patient counts in Results prose and Table 2 are inconsistent; exact arm participant totals remain unresolved.
No outcome beyond 12 months was assessed.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Authors report no placebo arm, no double blinding, and subjective outcomes.
- Bilateral feet from some participants were analyzed as independent units without adjustment for intra-person clustering.
- Radiofrequency group had longer symptom duration at baseline; authors note possible confounding.
- Patient counts in Results prose and Table 2 are inconsistent; exact arm participant totals remain unresolved.
- No outcome beyond 12 months was assessed.
Source-linked critical reading
Randomized comparative clinical trial
Technical source record
All 14 article section records, three structured tables with footnotes, four figure captions and disclosures reviewed. Raw XML table-wrap count (3) matches parsed tables. Figure pixels unavailable locally. Supplementary-information headings are present but no supplementary contents were included.
All three treatments were followed by improvement in this study. Radiofrequency had better average pain and function scores at 12 months, but the analysis treated feet from the same person as independent and group counts are inconsistent. The results need confirmation before applying them broadly.
In adults with chronic plantar heel pain that has not improved after conservative care, how do ultrasound-guided corticosteroid injection, extracorporeal shock-wave therapy and radiofrequency thermal lesioning compare over 12 months?
Prospective three-arm randomized controlled trial. A computer-generated 1:1:1 sequence was prepared by an independent researcher, with group assignment concealed until intervention. The study planned 60 feet and reports final outcomes per foot.
Adults with heel pain for at least six months, symptoms despite at least two conservative treatments, and clinical plantar-fascia findings. The final results report 20 feet from 18 patients in the corticosteroid group, 19 feet from 15 patients in the ESWT group and 23 feet from 16 patients in the radiofrequency group. However, Table 2's female/male figures sum to 20, 19 and 23 respectively, matching the foot counts rather than the patient counts; counts are inconsistent. Some patients contributed both feet, analyzed independently.
Corticosteroid group: ultrasound-guided medial injection of 1 mL betamethasone 40 mg/mL plus 2 mL bupivacaine 5 mg/mL. ESWT: BTL-5000 SWT, 15 Hz, 2,000 pulses, 4.0 bar, three weekly sessions, ultrasound guided. Radiofrequency thermal lesioning: single session under ultrasound guidance; NeuroTherm JK4A, 22G 10-cm cannula, 50 Hz, 90°C for 90 seconds at up to three pain sites; 1 mL 2% lidocaine at entry. All groups received five days of NSAIDs, ice advice and stretching; no weight-bearing restriction was advised.
The three active treatments were compared directly. All groups received the same short course of NSAIDs and advice on ice and stretching; there was no placebo or untreated group.
Primary outcome was pain on a 10-cm VAS, anchored 0 (no pain) to 10 (worst imaginable). Other outcomes were Foot Function Index (FFI; total and pain, disability, activity-limitation components) and Roles and Maudsley score. Measurements occurred at baseline and months 1, 6 and 12. Table 3 reports group means, SDs and CIs; narrative reports post-hoc pairwise results.
All groups improved from baseline. At 12 months, mean VAS was 1.61 (SD 0.839) radiofrequency, 3.47 (2.010) ESWT and 3.05 (1.432) corticosteroid. The authors report radiofrequency better than ESWT (p<0.001) and corticosteroid (p=0.006); the reported differences were 1.86 and 1.44 VAS points, respectively, exceeding the authors’ 1.2-point MCID. FFI total scores at 12 months were 27.043 (SD 12.4261), 56.737 (33.7505) and 50.250 (23.5816), respectively; radiofrequency was reported better than ESWT (p=0.001) and corticosteroid (p=0.008). Roles and Maudsley score favored radiofrequency at 12 months (p=0.002). No group difference was found for FFI pain component. These subjective outcomes require caution given lack of double blinding.
The article does not provide a structured adverse-event table or arm-level adverse-event counts. It discusses known corticosteroid risks such as fat-pad atrophy and plantar fascia rupture, but does not report study-specific event counts for the three arms.
Authors note no placebo arm, lack of double blinding and subjective outcomes. Both feet from some participants were analyzed independently without adjustment for within-person clustering. Symptom duration was longer in radiofrequency group; authors note this may confound the apparent long-term advantage. The sample counts conflict between Results narrative and Table 2, so exact participant-level arm sizes are unresolved. Twelve-month follow-up does not establish longer-term outcome.
Results suggest radiofrequency may offer greater improvement in selected outcomes at 12 months in this refractory cohort; they do not establish it as first-line therapy. The per-foot analysis and baseline symptom-duration imbalance may inflate apparent precision or affect comparison. ESWT was delivered at a stated 4.0 bar; no conversion to energy flux density is justified.
Authors call for larger double-blind placebo-controlled trials and longer follow-up. Future work should account for bilateral feet clustered within participants, report participant and foot counts consistently, and assess objective pain sensitivity alongside function.
Read the original article.
Comparative efficacy of corticosteroid injection, extracorporeal shock wave therapy, and radiofrequency ablation for chronic plantar fasciitis: a prospective randomized controlled trial.
Tütüncü MN, Sekizkardeş M, Şahin E, Karaismailoğlu B. · BMC musculoskeletal disorders · 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 →