Network meta-analysis of 189 trials: shockwave therapy eases short-term frozen shoulder pain

A systematic review and network meta-analysis of randomised controlled trials, available via dx.doi.org, compared therapeutic interventions for adhesive capsulitis (frozen shoulder) in adults. It included 189 trials with 12,988 participants and 59 interventions. Outcomes were pain, function, range of motion and adverse events. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

The authors searched Medline, the Cochrane Library, Embase, Web of Science and Scopus from inception to 1 May 2025, with an update in November 2025. The protocol is registered in PROSPERO (CRD420251070468). Most trials contributing to the main outcomes were at high risk of bias: 71% for pain, 72% for function and 71% for range of motion. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

Pain and function: extracorporeal shockwave therapy

At short-term follow-up, moderate-certainty evidence suggested that extracorporeal shockwave therapy probably reduced pain compared with usual care. The mean difference was -1.97 on a 0-10 scale (95% CI -2.97 to -0.97). dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

Low-certainty evidence suggested that the therapy may improve function (mean difference -11.42 on a 0-100 scale, 95% CI -16.68 to -6.15). Direct and indirect estimates for this outcome were limited by important incoherence. At mid-term follow-up, the possible improvement in function was also of low certainty, and pain reduction was supported only by very low certainty evidence. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

Range of motion: manipulation under anaesthesia

For short-term range of motion, moderate-certainty evidence suggested that manipulation under anaesthesia combined with intra-articular steroid injections probably improved flexion (mean difference 33.29°, 95% CI 11.19° to 55.39°). Manipulation under anaesthesia alone showed a similar result (26.16°, 8.19° to 44.13°). dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

Shockwave therapy may also improve flexion (22.42°, 12.99° to 31.84°) and abduction (18.62°, 8.09° to 29.14°). Both estimates are low-certainty and limited by important incoherence. Low-certainty evidence also pointed to a possible mid-term flexion gain with passive mobilisation. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

Limits of the evidence

Most other comparative estimates rested on low or very low certainty evidence. Comparative safety remains highly uncertain because adverse events were reported incompletely and inconsistently. The authors state that the absence of reported adverse events should not be read as absence of harm. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗

The available evidence is mainly short-term and specific to individual outcomes. Longer-term comparative effectiveness, overall benefits and harms, and applicability beyond predominantly primary adhesive capsulitis remain uncertain. dx.doi.orgdx.doi.orgdx.doi.orgOpen source ↗