Rotator-cuff / subacromial pain: mixed or unspecified subtype · Single-group study · 2026
Does shoulder tendon pain improve after focused shockwave plus exercise and rehab?
In this study of 239 people, shoulder pain dropped from 6.5 to about 4 out of 10 four months after three sessions of focused shockwave, one a week, alongside home exercises and supervised rehab. Shoulder function scores rose by about 15 points. People with a higher body mass index tended to gain less function.
Improved over time- adults with shoulder tendon pain
- 239
- sessions, one a week
- 3
- months of follow-up
- 4
Good to know: Everyone also did daily home exercises and 16 weeks of supervised rehab; there was no comparison group. Followed for 4 months.
Nusca SM, Latini E, Santilli G, et al. Medical Sciences (Basel, Switzerland), 2026 · Read the paper ↗
How this study worked
What happened
Shoulder pain
0–10 pain scale · start and after four months
StartAfter 4 months
Show the numbers
| Group | People | Start | After 4 months |
|---|---|---|---|
| Shockwave + exercise and rehab | 239 | 6.5 | 3.96 |
The treatment
Shockwave + exercise and rehab
Focused, once a week for 3 weeks
- Sessions
- 3, one a week
- Pulses
- 2,400 per session
- Strength
- 0.15mJ/mm² · 4 per second
- Numbing
- None
- Device
- Modulith SLK
- Everyone also got
- Daily home exercises plus supervised rehabilitation three times a week for 16 weeks
Compared with: No comparison group
Settings are what each study used, not a prescription.
Who took part
Adults with long-lasting supraspinatus (rotator-cuff) tendon pain, with or without calcium deposits, confirmed on imaging. Of 312 people screened, 239 were analysed; 41 had incomplete follow-up.
Side effects
No serious side effects. Some people had brief local discomfort, and everyone analysed completed all three sessions.
All the numbers and the full appraisal
The results, in numbers.
Adults with chronic, imaging-confirmed supraspinatus tendinopathy (calcific and noncalcific), treated with focused ESWT; final cohort 239 of 312 screened; 73 were excluded, including 41 (13.1%) with incomplete follow-up.
Retrospective single-arm cohort with no untreated or active comparator. Of 312 patients screened, 73 were excluded (41 with incomplete follow-up, 18 rotator cuff tears, 8 recent injections, 6 other reasons), leaving 239 analysed; the paper prints no separate enrolled count. The printed loss to follow-up of 13.1% is 41 of the 312 screened.
10-point VAS; higher=worse1 reported outcome
Shoulder pain (VAS)
10-point VAS; higher=worse points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Focused ESWT plus home exercise and supervised rehabilitation | 239 | 6.5 (SD 1.35) | 3.96 (SD 2.09) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Within-cohort change, baseline to month 4
239 participants in this estimate
Baseline versus 4-month follow-up
Within-cohort pre/post comparison
Pain decreased by a mean 2.55 points (95% CI 2.29–2.81) at 4 months; within-cohort p<0.001.
Analysis details and source
Uncontrolled pre/post estimate in a retrospective cohort receiving concurrent home exercise and a 16-week supervised rehabilitation program; the change cannot be attributed to ESWT alone. The Methods describe a 100 mm VAS (0–100 mm), but Table 2 labels it 0–10 and reports values in that range; values are kept as printed.
Constant–Murley 0–100; higher=better1 reported outcome
Shoulder function (Constant–Murley)
Constant–Murley 0–100; higher=better points
| Group | n | Baseline | Follow-up |
|---|---|---|---|
| Focused ESWT plus home exercise and supervised rehabilitation | 239 | 60.38 (SD 14.53) | 75.88 (SD 15.52) |
SD: standard deviation. CI: confidence interval. IQR: interquartile range. —: not extracted.
Within-cohort change, baseline to month 4
239 participants in this estimate
Baseline versus 4-month follow-up
Within-cohort pre/post comparison
Mean function score increased 15.50 points (95% CI 13.67–17.34) by month 4; p<0.001.
Analysis details and source
Uncontrolled pre/post estimate in a retrospective cohort receiving concurrent home exercise and a 16-week supervised rehabilitation program; the change cannot be attributed to ESWT alone.
Multivariable linear regression for Delta VAS and Delta Constant–Murley1 reported outcome
Predictors of four-month treatment response
Multivariable linear regression for Delta VAS and Delta Constant–Murley Unstandardized regression coefficient β
Higher baseline VAS predicted larger absolute pain reduction (β=0.571, 95% CI 0.37–0.77; p<0.001). Higher baseline Constant–Murley also predicted pain reduction (β=0.025; 0.01–0.04; p=0.010). For functional change, higher baseline Constant–Murley (β=−0.504; −0.63 to −0.38) and higher BMI (β=−0.949 per kg/m²; −1.43 to −0.47), both p<0.001, predicted smaller improvement. Models explained adjusted R²=0.122 for VAS and 0.275 for function.
Analysis details and source
Exploratory backward stepwise models in one retrospective cohort; predictors require external validation. Diabetes, hypercholesterolemia and hypertension were not retained as independent predictors.
Four-category ordinal patient-rated outcome; success defined by the source as score 1–21 reported outcome
Roles and Maudsley Score distribution
Four-category ordinal patient-rated outcome; success defined by the source as score 1–2 participants (%)
In this uncontrolled cohort (N=239), baseline to month 4 counts were: excellent 39 (16.3%) to 147 (61.5%); good 114 (47.7%) to 71 (29.7%); acceptable 75 (31.4%) to 20 (8.4%); poor 11 (4.6%) to 1 (0.4%). The source-defined success category (score 1–2) was 153 (64.0%) at baseline and 218 (91.2%) at month 4. These are within-cohort descriptive changes; the table reports no comparative p value.
Analysis details and source
Do not interpret this uncontrolled pre/post distribution as causal ESWT efficacy or as a between-group effect. The paper defines the score as a patient-rated treatment outcome (1 = no pain, full movement and activity), yet Table 2 gives a baseline distribution with 39 patients (16.3%) rated excellent, although all had shoulder pain for at least 3 months; counts are kept as printed.
How treatment was delivered
Modulith SLK electromagnetic generator with integrated inline ultrasound guidance, targeting supraspinatus insertion at greater tuberosity; no local anesthesia. EFD 0.15 mJ/mm², 2,400 pulses/session, 4 Hz, three treatments at weekly intervals. All received daily home pendulum exercises, passive and active-assisted range of motion, and progressive rotator cuff strengthening; no other physical modalities allowed.
Started with the first f-ESWT session and continued for 16 weeks: three sessions per week of about 45–60 min, supervised by a physical therapist specialized in shoulder rehabilitation (warm-up; progressive rotator cuff and scapular strengthening; posterior capsule and pectoralis minor stretching), progressed by tolerance with pain monitoring; adherence tracked by direct supervision and attendance records.
No serious adverse events related to focused ESWT were observed. Some patients reported transient local discomfort that did not require discontinuation; all 239 analyzed participants completed three sessions.
Retrospective single-center cohort with no comparator, concurrent daily home exercise and a 16-week supervised rehabilitation program, and 41 of 312 screened patients lost to follow-up (13.1%). Follow-up was four months. Backward stepwise models may be unstable and are hypothesis-generating; no external validation. Improvement cannot be assigned to ESWT alone.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Retrospective single-center cohort with no comparator, concurrent daily home exercise and a 16-week supervised rehabilitation program, and 41 of 312 screened patients lost to follow-up (13.1%). Follow-up was four months. Backward stepwise models may be unstable and are hypothesis-generating; no external validation. Improvement cannot be assigned to ESWT alone.
Source-linked critical reading
Retrospective treated-cohort prognostic study
Technical source record
Read full article, all raw tables/captions, cohort attrition, ESWT and exercise protocol, 4-month outcomes, regression models, safety and limitations.
People improved after ESWT plus exercise, but the design cannot show how much improvement was caused by shockwave therapy alone. Higher BMI was associated with smaller functional gains.
Among people treated with focused ESWT for supraspinatus tendinopathy, which baseline clinical/metabolic characteristics predict four-month pain and functional response?
Single-center retrospective cohort, consecutive patients treated June 2020–December 2025; 239 analyzed after 73 exclusions, 41 of them for incomplete follow-up.
Adults with chronic imaging-confirmed calcific or noncalcific supraspinatus tendinopathy, symptoms ≥3 months; full-thickness tears and recent injection/nerve block were excluded.
Three ultrasound-guided focused sessions at weekly intervals using Modulith SLK, EFD 0.15 mJ/mm², 2,400 pulses/session, 4 Hz, no local anesthesia. All received daily home exercise and 16 weeks of supervised rehabilitation three times a week; other modalities prohibited.
No untreated, exercise-only or alternative-treatment comparator.
At four months, VAS fell 2.55 points (95% CI 2.29–2.81); Constant–Murley rose 15.50 (13.67–17.34); both p<0.001. Models associated baseline severity and BMI with change. The Roles and Maudsley ordinal distribution is also reported at baseline and month 4: excellent 39/239 (16.3%) to 147/239 (61.5%), good 114 (47.7%) to 71 (29.7%), acceptable 75 (31.4%) to 20 (8.4%), and poor 11 (4.6%) to 1 (0.4%). Source-defined success (scores 1–2) was 153/239 (64.0%) to 218/239 (91.2%). This is uncontrolled cohort description; no comparative p value is supplied.
Higher BMI predicted smaller Constant–Murley improvement (β −0.949 per kg/m², 95% CI −1.43 to −0.47). Baseline scores predicted absolute change. Models explained 12.2% and 27.5% adjusted variance.
No serious ESWT adverse event; transient local discomfort occurred. All analyzed patients completed three sessions.
Retrospective, uncontrolled, single-center, four-month follow-up, 41 of 312 screened (13.1%) lost, concurrent exercise and stepwise model instability; prediction models not externally validated.
This cohort describes short-term outcomes after combined ESWT and exercise, not the independent treatment effect of ESWT.
Prospective external validation and controlled studies should assess prognostic variables and separate ESWT from exercise effects.
Read the original article.
Clinical and Metabolic Predictors of Response to Focused Extracorporeal Shockwave Therapy in Rotator Cuff Tendinopathy: A Retrospective Cohort Study.
Nusca SM, Latini E, Santilli G, Beccarini G, Bova V, Santoboni F, Santilli V, Felzani G, Perroni F, Vulpiani M, Sisti D, Vetrano M. · Medical sciences (Basel, Switzerland) · 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 →