Dextrose prolotherapy for chronic symptomatic rotator cuff tendinopathy and subacromial bursitis: A systematic review...

Dextrose prolotherapy is used for musculoskeletal injuries, but evidence of its effectiveness is limited. We assess the evidence base on effects of dextrose prolotherapy for chronic symptomatic rotator cuff disorders and/or bursitis.

Ovid/Medline, Ovid/Embase, and Scopus from inception to March 2026. Content experts provided additional citations.

We included randomized controlled trials (RCTs) or observational studies evaluating dextrose prolotherapy for shoulder pain. Outcomes of interest included pain-related functioning, physical performance, pain severity, quality of life, adverse events, costs, and treatment burden. Study risk of bias (RoB) used Cochrane RoB-2 or RoB In Non-Randomized Studies of Interventions (ROBINS-I). Certainty of evidence (COE) used Grading of Recommendations Assessment, Development, and Evaluation. Meta-analyses were conducted if more than three studies evaluated similar interventions, comparators, and outcomes.

We identified 14 studies (10 high RoB), including 13 RCTs and 1 observational study. These included participants with mixed rotator cuff tendinopathy and/or subacromial bursitis (k = 9) or only supraspinatus tendinopathy (k = 5). Studies enrolled individuals with pain/dysfunction >3 months. Dextrose concentrations varied (12.5%-25%); eight studies used a single injection. Studies were short- to medium-term (k = 10 < 6 months) and mostly small (n = 12-129). Comparators included normal saline (k = 5), corticosteroid injections (k = 4), physical therapy (PT) (k = 4), platelet-rich plasma (PRP) (k = 2) and atelocollagen (k = 1). For mixed bursitis and rotator cuff tendinopathy, the evidence is very uncertain regarding the effect of dextrose prolotherapy compared with normal saline on short- and medium-term pain-related functioning (standardized mean difference [SMD]_medium-term: -0.3 [-1.0; 0.4] (k = 3; n = 149); very low COE) and may have little to no benefit at long-term follow-up (low COE). Compared with corticosteroid injections, the evidence is very uncertain regarding the effect of dextrose prolotherapy on short- and medium-term pain-related functioning (SMD_medium-term: 0.6 [-2.2; 3.4] (k = 3; n = 147); moderate COE) and little to no difference in long-term pain-related functioning (low COE). Evidence comparing dextrose prolotherapy with PT or autologous blood products, as well as for other efficacy outcomes and adverse events, was sparse (mostly very low COE). Each study evaluating prolotherapy for supraspinatus tendinopathy used a different comparator (saline, corticosteroid, PT, PRP, or atelocollagen), limiting synthesis of their findings. Cost, health-related quality of life, and treatment burden were not reported.

For chronic symptomatic mixed rotator cuff tendinopathy and/or subacromial bursitis, dextrose prolotherapy may have little to no benefit compared with normal saline. Compared with corticosteroid injections, dextrose prolotherapy probably results in worse physical performance. Limited evidence exists for supraspinatus tendinopathy. High-quality RCTs are needed to better understand benefits and harms.