Erectile dysfunction · Single-group study · 2026
Which men with erection problems respond best to shockwave?
In this study of 219 men, about two in three (145) improved six months after shockwave twice a week for three weeks. Men under 65, without diabetes, with erection problems for under 36 months, or who had been helped by erection pills, were more likely to respond.
Two in three men improved- men with erection problems linked to blood flow
- 219
- weeks of treatment, twice a week
- 3
- months of follow-up
- 6
Good to know: No comparison group. The 7-point score for predicting response has not yet been tested in other clinics.
Anıl H, Yıldız A, Topuz AN, et al. Sexual Medicine, 2026 · Read the paper ↗
How this study worked
What happened
Men who responded at six months
Response: erection hardness up by 1 point or more, or erection score up by 5 or more
Show the numbers
| Group | Count | Share |
|---|---|---|
| Shockwave | 145 / 219 | 66.2% |
The treatment
Shockwave
Focused, twice a week for 3 weeks
- Pulses
- 3,000 per session · spread over six spots on the shaft; 18,000 in all
- Strength
- 0.09mJ/mm² · 3 per second
- Numbing
- Not mentioned
- Device
- Modus ESWT
Compared with: No comparison group
Settings are what each study used, not a prescription.
Who took part
219 men aged 18–80 with mild to moderate erection problems thought to come from blood flow, at two centres; none had had shockwave before.
Side effects
The available results do not report side effects for this group.
All the numbers and the full appraisal
The results, in numbers.
Men aged 18–80 with clinically diagnosed vasculogenic erectile dysfunction, baseline IIEF-5 EF 11–21, EHS 2–3, stable ED ≥6 months and no prior Li-ESWT; 219 analyzed. The responder/nonresponder groups are post-treatment strata defined by six-month response, not treatment arms. Penile Doppler was not routine.
Prospective, nonrandomized, single-arm intervention; responder/nonresponder groups defined after 6-month response.
Response: ≥1-point EHS increase or ≥5-point IIEF-5 EF increase1 reported outcome
Six-month Li-ESWT treatment response
Response: ≥1-point EHS increase or ≥5-point IIEF-5 EF increase responders / cohort
145/219 patients (66.2%) met the prespecified response definition at six months. This is a single-arm response rate, not a comparative treatment effect.
Analysis details and source
Responders n=145; nonresponders n=74. IIEF-5 EF and EHS are separate components of the OR-based response definition.
Seven-item 0–7 point score1 reported outcome
Clinical prediction score for six-month Li-ESWT response
Seven-item 0–7 point score points; response likelihood bands
One point each: cardiovascular risk factors present; age <65; FMD ≥5%; previous PDE5i benefit; cIMT <0.8 mm; no diabetes; ED duration <36 months. Reported response-rate bands: 0–2 points, 0–25%; 3–4, 34–60%; 5–7, 75–88.3%. AUC 0.819 (95% CI 0.762–0.877); cutoff 4.5, sensitivity 73%, specificity 77%, PPV 0.86, NPV 0.59, accuracy 74%.
Analysis details and source
The model is not externally validated; prediction applies to this cohort/device/protocol. ORs: no DM 2.67 (1.24–5.72); ED duration <36 months 2.23 (1.10–4.53); cIMT <0.8 2.04 (1.03–4.06); PDE5i benefit 2.47 (1.18–5.15); FMD ≥5% 2.57 (1.23–5.34); age <65 2.28 (1.07–4.83); CVD risk factors 2.23 (1.06–4.70). CAD was not independent (p=.118). Calibration: Hosmer–Lemeshow χ²=1.678, df=4, p=.795. Bootstrap internal validation (1,000 iterations) yielded AUC .82 (95% CI .76–.88); this is not external validation.
IIEF score2 reported outcomes
IIEF at baseline by eventual response status
IIEF score Source-reported mean±SD or median (IQR)
Nonresponders (n=74): 16.8±3.6; responders (n=145): 16.3±2.9; p=.266.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
IIEF at month 6 by response status
IIEF score Source-reported mean±SD or median (IQR)
Nonresponders (n=74): 18.2±3.9; responders (n=145): 21.7±3.5; p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
Erection Hardness Score (EHS), median (IQR)2 reported outcomes
EHS at baseline by eventual response status
Erection Hardness Score (EHS), median (IQR) Source-reported mean±SD or median (IQR)
Nonresponders (n=74): 2 (2–2); responders (n=145): 2 (2–3); p=.223.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
EHS at month 6 by response status
Erection Hardness Score (EHS), median (IQR) Source-reported mean±SD or median (IQR)
Nonresponders (n=74): 2 (2–2); responders (n=145): 3 (2–3); p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
Age (years)1 reported outcome
Age ROC
Age (years) AUC (95% CI), cut-off, sensitivity and specificity (%)
AUC .67 (95% CI .60–.74); cutoff ≤63.5 years; sensitivity 59.5%, specificity 82.1%; p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
ED duration, months1 reported outcome
ED duration ROC
ED duration, months AUC (95% CI), cut-off, sensitivity and specificity (%)
AUC .68 (95% CI .61–.76); cutoff ≤34.5 months; sensitivity 82.4%, specificity 41.4%; p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
Carotid intima-media thickness (cIMT), mm1 reported outcome
cIMT ROC
Carotid intima-media thickness (cIMT), mm AUC (95% CI), cut-off, sensitivity and specificity (%)
AUC .70 (95% CI .62–.78); cutoff <.765 mm; sensitivity 82.4%, specificity 36.6%; p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
Flow-mediated dilation (FMD), %1 reported outcome
FMD ROC
Flow-mediated dilation (FMD), % AUC (95% CI), cut-off, sensitivity and specificity (%)
AUC .74 (95% CI .66–.83); cutoff ≥4.8%; sensitivity 89.0%, specificity 60.8%; p<.001.
Analysis details and source
Responder status is a post-treatment prognostic stratum, not a randomized comparator; these contrasts are not causal effects.
How treatment was delivered
Modus ESWT electrohydraulic generator with focusing probe, 0.09 mJ/mm², 3 Hz; 3,000 pulses/session twice weekly, 18,000 total over 3 weeks at six bilateral penile shaft sites. Active PDE5i was stopped with a one-month washout.
The accessible results do not report adverse events for the Li-ESWT cohort. This should not be read as evidence of no harms.
No sham/control group; response model developed and tested in the same 219-person cohort with bootstrap internal validation only. No external validation. Vasculogenic ED was diagnosed clinically without routine penile Doppler, which may include mixed vascular phenotypes. Two centers/operators, self-reported/clinical outcomes, device-specific 3-week protocol.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- No sham/control group; response model developed and tested in the same 219-person cohort with bootstrap internal validation only. No external validation. Vasculogenic ED was diagnosed clinically without routine penile Doppler, which may include mixed vascular phenotypes. Two centers/operators, self-reported/clinical outcomes, device-specific 3-week protocol.
Source-linked critical reading
Single-arm intervention and prediction-model study
Technical source record
Read full article, raw Tables 1–4, all figure captions, Li-ESWT protocol, exact six-month response definition, regression coefficients, score bands, discrimination/calibration and limitations.
The paper proposes a way to estimate response likelihood, but without a control and external validation it cannot show treatment benefit or predict an individual reliably.
Can clinical and vascular characteristics predict a six-month response to Li-ESWT in clinically diagnosed vasculogenic ED?
Prospective, nonrandomized, single-arm interventional study at two centers, planned January 2024–January 2025. A seven-factor prediction score was derived.
Men aged 18–80 with clinical vasculogenic ED, IIEF-5 EF 11–21, EHS 2–3, stable ED ≥6 months, and no Li-ESWT prior treatment. Penile Doppler was not routinely used.
Modus ESWT electrohydraulic focusing device; 0.09 mJ/mm², 3 Hz, 3,000 pulses twice weekly for three weeks (18,000 total) at six bilateral shaft sites.
No sham or untreated comparator.
Response at six months was ≥1-point EHS increase OR ≥5-point IIEF-5 EF increase. 145/219 (66.2%) responded. Table 1 also compares post-response strata (nonresponders vs responders): baseline IIEF 16.8±3.6 vs 16.3±2.9 (p=.266), month-6 IIEF 18.2±3.9 vs 21.7±3.5 (p<.001); baseline EHS median (IQR) 2 (2–2) vs 2 (2–3) (p=.223), month 6 2 (2–2) vs 3 (2–3) (p<.001). These are responder/nonresponder differences after treatment, not causal comparisons.
Score: one point each for CVD risk factors, age <65, FMD ≥5%, prior PDE5i benefit, cIMT <0.8 mm, no diabetes, ED <36 months. AUC 0.819 (95% CI 0.762–0.877); sensitivity 73%, specificity 77% at 4.5 cutoff. The four single-predictor ROC rows in Table 2 are separate from the final seven-item score: age AUC .67 (95% CI .60–.74), cutoff ≤63.5; ED duration .68 (.61–.76), ≤34.5 months; cIMT .70 (.62–.78), <.765 mm; and FMD .74 (.66–.83), ≥4.8%; each p<.001. The article reports Hosmer–Lemeshow χ²=1.678 (df=4), p=.795 and internal 1,000-iteration bootstrap AUC .82 (95% CI .76–.88), not external validation.
No adverse-event summary was found in accessible results.
No control group; score developed/tested in same cohort with internal bootstrap only, no external validation; clinical diagnosis without routine Doppler; device/protocol-specific, potential operator/site differences.
Results and score apply to this cohort and protocol; do not establish comparative efficacy or validated individual prediction.
External validation in independent multicenter cohorts and controlled trials are necessary before clinical use of the score.
Read the original article.
Development of a clinical scoring system to predict response to low-intensity shock wave therapy in clinically diagnosed vasculogenic erectile dysfunction: a prospective, non-randomized, single-arm, interventional study.
Anıl H, Yıldız A, Topuz AN, Kayra MV, Sözütok S, Altunkol A, Alma E, Topuz M, Bayazıt Y. · Sexual medicine · 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 →