Calcific rotator-cuff tendinopathy · Review · 2024
How does shockwave compare with needle lavage for calcium deposits in the shoulder?
In this review, three trials compared shockwave with ultrasound-guided needle lavage, which rinses out the calcium deposit. In the first six months, neither clearly eased pain more, and shoulder function showed no clear difference. At one year, pain was lower in the lavage groups.
Needle lavage did better for pain at one year- trials in the review, reported in 7 papers
- 6
- people with calcium deposits in a shoulder tendon
- 703
- trials compared lavage with shockwave
- 3
Good to know: This review was mainly about needle lavage; shockwave was the comparison in three trials.
Sconza C, Palloni V, Lorusso D, et al. European Journal of Physical and Rehabilitation Medicine, 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
- studies
- 3
- Compared with
- Ultrasound-guided needle lavage that rinses out the calcium deposit
Shoulder pain: lavage vs shockwave
Pooled trials · standardised difference (no units)
Show the numbers
| Difference (standardised difference (no units)) | likely range | |
|---|---|---|
| Pain at one year | −0.52 | −0.85 to −0.19 |
| Pain, under 12 weeks | −0.52 | −1.57 to 0.54 |
What the trials used
- Ultrasound-guided lavage (UGL)
- UGL was performed with needle lavage/aspiration of calcific deposits, commonly using saline and a single 18G needle; reported exceptions included 17G and 16G, and some trials used two needles. Subacromial corticosteroid/anesthetic was given before or after UGL depending on study. Techniques and timing varied.
- Comparators and follow-up
- UGL was compared with ESWT in three studies, steroid/anesthetic injection in two, sham lavage in one, and deposit needling without aspiration in one. Only the UGL-versus-ESWT studies were pooled; other comparisons were descriptive. Reported follow-up timepoints ranged from 6 weeks to 5 years.
Who took part
People with calcium deposits in a rotator-cuff tendon, mostly 3 mm or larger; about 6 in 10 were women.
Side effects
Across six trials, side effects were absent or mild and brief. They included a frozen shoulder that resolved, feeling faint, brief dizziness, nausea or headache, bruising where the needle went in, discomfort and bursa inflammation.
All the numbers and the full appraisal
Every reported result
The results, in numbers.
Results describes seven RCTs with 703 participants with calcific rotator cuff tendinopathy; the abstract instead says 709 subjects, a six-person discrepancy the article does not reconcile. Results reports 426 women (60.59%). One five-year report is follow-up of another included trial, not an independent cohort. Most included calcifications were ≥3 mm and Gartner types I–II.
Pain: UGL vs ESWT2 reported outcomes
Pain: UGL vs ESWT
VAS or NPRS standardized mean difference
UGL versus ESWT
3 studies · I²=93%
Reported estimate and confidence interval.
No significant pooled difference (SMD −0.52, 95% CI −1.57 to 0.54; p=0.34; I²=93%; 3 studies); certainty very low.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
Pain: UGL vs ESWT
VAS or NPRS standardized mean difference
UGL versus ESWT
3 studies · I²=38%
Reported estimate and confidence interval.
The 52-week pooled comparison was statistically significant and favored UGL as reported (SMD −0.52, 95% CI −0.85 to −0.19; p=0.002; I²=38%; 3 studies). Certainty was nevertheless very low.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
Pain: UGL vs ESWT1 reported outcome
Pain: UGL vs ESWT
VAS mean difference, scale points
UGL versus ESWT
2 studies · I²=82%
Reported estimate and confidence interval.
No significant pooled difference (MD −1.20, 95% CI −2.66 to 0.27; p=0.11; I²=82%; 2 studies); certainty very low.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
Function: UGL vs ESWT2 reported outcomes
Function: UGL vs ESWT
Constant–Murley Score (CMS) mean difference, score points
UGL versus ESWT
2 studies · I²=79%
Reported estimate and confidence interval.
No significant pooled difference (MD 3.34, 95% CI −11.45 to 18.12; p=0.66; I²=79%; 2 studies); certainty very low.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
CMS at 52 weeks: UGL+steroid vs steroid
Constant–Murley Score (CMS) mean difference, score points
UGL plus subacromial steroid injection versus steroid injection alone (single trial; non-pooled)
1 study
Reported estimate and confidence interval.
In one trial, UGL plus subacromial injection favored function at 52 weeks (MD 12.1, 95% CI 3.9 to 20.2; p=0.005). This is a single-trial comparison, not the UGL-versus-ESWT pooled estimate.
Analysis details and source
Single-trial, non-pooled function result described narratively by the review; the source-reported contrast is retained without recalculating direction or effect. The main text gives no certainty (GRADE) rating for this comparison.
Function: UGL vs ESWT1 reported outcome
Function: UGL vs ESWT
CMS or Oxford Shoulder Score (OSS) standardized mean difference
UGL versus ESWT
2 studies · I²=30%
Reported estimate and confidence interval.
No significant pooled difference (SMD 0.10, 95% CI −0.40 to 0.60; p=0.69; I²=30%; 2 studies); certainty very low.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
Calcific resorption: UGL vs ESWT1 reported outcome
Calcific resorption: UGL vs ESWT
Calcification resorption/disappearance risk ratio
UGL versus ESWT (as stated in Results; the abstract instead says UGL plus subacromial corticosteroid injection versus injection alone)
2 studies · I²=0%
Reported estimate and confidence interval. Logarithmic ratio scale.
In Results, the pooled resorption rate favored UGL over ESWT (RR 1.63, 95% CI 1.34 to 1.98; p<0.00001; I²=0%; 2 studies); evidence certainty was very low. The abstract attributes the same RR to UGL plus subacromial corticosteroid injection versus injection alone; the article does not reconcile the comparators, and this extraction follows the Results.
Analysis details and source
Review-level pooled estimate. The source-reported contrast is retained without recalculating direction or effect. Certainty was very low; nonsignificant differences do not establish equivalence.
Study-specific calcific resorption rates1 reported outcome
Study-specific calcific resorption rates
Radiographic/ultrasound deposit disappearance/resorption percent and count as reported
De Witte 2013: UGL+steroid 13/23 (56.5%) vs steroid 6/25 (24%) at 52 weeks. De Witte 2017 follow-up: steroid 16/25 (73%) vs UGL 13/23 (62%) at 5 years. Moosmayer: UGL+steroid 33/72 (45.83%) vs sham UGL+steroid 6/70 (8.57%) at 4 months. The five-year report is related follow-up, not an independent trial.
Analysis details and source
These are study-specific denominators and differing comparators; they are not pooled or directly comparable across studies. Percentages are as printed; for the five-year follow-up, 16/25 is 64% and 13/23 is 56.5%, not 73% and 62%.
Pain outcomes in non-pooled comparisons1 reported outcome
Pain outcomes in non-pooled comparisons
VAS at rest/motion/night/activity VAS scores and MDs as reported
De Witte: UGL+steroid vs steroid, post-intervention VAS rest 22.1±20.8 vs 19.6±24.2 (MD 2.5, 95% CI −11.3 to 16.3); motion 23.6±22.0 vs 25.0±23.9 (MD −1.4, −15.3 to 12.6). Moosmayer at 4 months: night MD −8.43 (−17.03 to 0.18), rest −4.09 (−11.53 to 3.35), activity −3.24 (−12.38 to 5.90); none significant.
Analysis details and source
Separate trials and outcome definitions; reported means are not assigned to a pooled estimate.
WORC: UGL plus steroid vs steroid2 reported outcomes
WORC: UGL plus steroid vs steroid
Western Ontario Rotator Cuff Index (WORC) mean difference, score points
UGL plus subacromial corticosteroid injection versus injection alone
1 study
Reported estimate and confidence interval.
At 52 weeks, the single-trial WORC estimate was not statistically significant (MD 14.0; 95% CI −1.7 to 29.7; p=0.08).
Analysis details and source
Single-trial, non-pooled quality-of-life result; the confidence interval includes the null. Do not interpret the nonsignificant result as equivalence.
WORC: UGL vs steroid at five years
Western Ontario Rotator Cuff Index (WORC) mean difference, score points
UGL plus subacromial corticosteroid injection versus injection alone
1 study
Reported estimate and confidence interval.
At five years, the related De Witte follow-up reported no statistically significant WORC difference (MD 4.2; 95% CI −11.0 to 19.4; p=0.58). This is follow-up, not an independent cohort.
Analysis details and source
Single-trial, non-pooled quality-of-life result; the confidence interval includes the null. Do not interpret the nonsignificant result as equivalence.
EQ-5D-5L: UGL vs sham1 reported outcome
EQ-5D-5L: UGL vs sham
EQ-5D-5L mean difference, index points
UGL plus subacromial corticosteroid injection versus sham UGL plus injection
1 study
Reported estimate and confidence interval.
At four months, the single-trial EQ-5D-5L difference was not statistically significant (MD −0.04; 95% CI −0.08 to 0.01); the Results section does not report a p value.
Analysis details and source
Single-trial, non-pooled quality-of-life result; the confidence interval includes the null. Do not interpret the nonsignificant result as equivalence.
How treatment was delivered
UGL was performed with needle lavage/aspiration of calcific deposits, commonly using saline and a single 18G needle; reported exceptions included 17G and 16G, and some trials used two needles. Subacromial corticosteroid/anesthetic was given before or after UGL depending on study. Techniques and timing varied.
UGL was compared with ESWT in three studies, steroid/anesthetic injection in two, sham lavage in one, and deposit needling without aspiration in one. Only the UGL-versus-ESWT studies were pooled; other comparisons were descriptive. Reported follow-up timepoints ranged from 6 weeks to 5 years.
Six studies reported adverse events; overall they were absent or minimal/transient. Events included frozen shoulder (resolved), vasovagal reaction, temporary dizziness/nausea/headache, injection-site hematoma, discomfort and subacromial bursitis. No pooled event rate is reported.
Only three RCTs contributed to the UGL-versus-ESWT pooled comparisons, and at least two of six assessed studies were at high overall risk of bias.
Treatment technique, injection timing, calcification morphology, eligibility and follow-up varied; most long-term evidence was limited and some patients received alternative treatments during follow-up.
Evidence certainty for pooled UGL-versus-ESWT results was very low; other comparisons were descriptive and study-specific.
ROM and time to recovery were not investigated in any included study; quality-of-life outcomes were incompletely reported.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- Only three RCTs contributed to the UGL-versus-ESWT pooled comparisons, and at least two of six assessed studies were at high overall risk of bias.
- Treatment technique, injection timing, calcification morphology, eligibility and follow-up varied; most long-term evidence was limited and some patients received alternative treatments during follow-up.
- Evidence certainty for pooled UGL-versus-ESWT results was very low; other comparisons were descriptive and study-specific.
- ROM and time to recovery were not investigated in any included study; quality-of-life outcomes were incompletely reported.
Source-linked critical reading
Systematic review with meta-analysis
Technical source record
Complete acquired XML reviewed end-to-end. All seven RCT descriptions, related follow-up overlap, pooled and non-pooled pain/function/resorption results, adverse events, QoL, missing ROM/recovery data, limitations and conclusion checked. Supplement tables not available.
Lavage may remove calcifications and can help some outcomes, but this review found uncertain evidence and no clear functional advantage over shockwave therapy. The best option depends on the clinical and imaging context.
Does ultrasound-guided lavage (UGL) improve pain, function and calcific resorption in rotator-cuff calcific tendinopathy compared with ESWT and other treatment approaches?
Systematic review/meta-analysis of seven RCTs. Results reports 703 participants, whereas the abstract reports 709; the six-person discrepancy is unresolved. One five-year publication is follow-up from another trial, not an independent cohort. UGL-versus-ESWT pooled estimates came from three trials.
Calcific rotator-cuff disease; most deposits were >3 mm and Gartner I–II. Calcification classification and tendon location varied. The article reports 703 participants in Results versus 709 in the abstract.
UGL used needle lavage/aspiration, generally saline and an 18G needle; some studies used different gauges or two needles. Steroid/anesthetic timing differed.
Comparisons included ESWT, steroid/anesthetic injection, sham lavage and needling without aspiration. Only ESWT comparisons were pooled; others are descriptive.
Pain versus ESWT was not statistically different at under 12 weeks or at 26 weeks but favored UGL at 52 weeks (SMD −0.52, 95% CI −0.85 to −0.19; p=0.002). Function did not differ clearly at 6 or 52 weeks. Results attributes the pooled resorption RR 1.63 (95% CI 1.34–1.98) to UGL versus ESWT; the abstract attributes that estimate to UGL plus subacromial corticosteroid injection versus injection alone. This comparator discrepancy is unresolved. Evidence certainty was very low.
UGL showed selected advantages for pain at 52 weeks and calcific resorption, not clear functional superiority. Comparisons with injection or sham were mixed, and source certainty remains very low.
Six studies reported no or minimal transient events; no pooled safety rate is available.
Few pooled trials, high risk of bias, heterogeneous protocols/morphology/comparators, limited long-term evidence, unavailable supplements, and no ROM or recovery-time outcomes.
Interpret by symptom, deposit morphology and comparison treatment. UGL-versus-ESWT evidence has very low certainty; the review cannot establish a gold-standard technique.
Run adequately powered, sham-controlled trials with standardized lavage, steroid timing, calcification criteria, function, pain, resorption, ROM and adverse-event outcomes, plus long follow-up.
Read the original article.
Ultrasound-guided percutaneous lavage for the treatment of rotator cuff calcific tendinopathy: a systematic review with meta-analysis of randomized controlled trials.
Sconza C, Palloni V, Lorusso D, Guido F, Farì G, Tognolo L, Lanza E, Brindisino F. · European journal of physical and rehabilitation medicine · 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 →