ComparativeEffectivenessofBotulinumToxinInjectionforChronicShoulderPain:AMeta-AnalysisofRandomizedControlledTrials

Chronic shoulder pain is a widespread problem that affects about 16% of the general population, with lifetime prevalence estimates as high as 67%. The condition significantly reduces work efficiency and quality of life while creating substantial economic and medical burdens on society. While shoulder pain often stems from joint and tendon problems like rotator cuff disorders, myofascial pain characterized by trigger points is another common cause. Various injection therapies are used to treat these conditions, including corticosteroids, local anesthetics, and more recently, botulinum toxin (BoNT).

Although BoNT is well-known for treating conditions like post-stroke spasticity and cervical dystonia, its effectiveness for chronic shoulder pain—both joint-related and myofascial—has remained unclear. Previous reviews have reached conflicting conclusions, with some suggesting BoNT was effective while others found it no better than local anesthetic injections. Importantly, earlier reviews often mixed different patient populations together, including stroke patients with hemiplegic shoulders, making it difficult to draw clear conclusions about BoNT's effectiveness for chronic shoulder pain in the general population. This meta-analysis was designed to provide clearer, more specific evidence by focusing only on randomized controlled trials in non-stroke patients and by separately analyzing joint pain and myofascial pain conditions.

The researchers conducted a comprehensive systematic review and meta-analysis of randomized controlled trials (RCTs) published up to March 2020. They searched PubMed, Embase, the Cochrane Central Register of Controlled Trials, and other databases using specific search terms related to botulinum toxin, shoulder pain, and myofascial pain. Two reviewers independently screened articles and extracted data using predetermined forms. They included only RCTs that enrolled patients with shoulder pain (including tendon and joint disorders like adhesive capsulitis, subacromial impingement, and osteoarthritis, as well as myofascial pain syndrome) and that measured pain quantitatively before and after treatment.

Studies of stroke patients with hemiplegic shoulders were excluded, as were animal studies, case reports, and quasi-experimental designs.

The primary outcome was pain reduction measured by the Visual Analogue Scale (VAS), with data extracted at baseline, within one month, and between one to three months after intervention. The researchers used standardized mean differences (SMD) to compare BoNT injections against various reference treatments including saline, corticosteroids, dry needling, and lidocaine. They applied random-effects models for statistical pooling and assessed study quality using the Cochrane Risk of Bias Tool. Publication bias was evaluated using funnel plots and Egger's test.

Nine RCTs comprising 666 patients were included in the final analysis. The studies were divided into two groups: 182 patients with shoulder joint pain and 484 patients with shoulder myofascial pain. For shoulder joint pain, four studies were analyzed. At one month post-intervention, BoNT injections showed similar efficacy to reference treatments: compared to saline, the SMD was −0.605 (95% CI: −1.242 to 0.032), and compared to corticosteroids, the pooled SMD was −0.180 (95% CI: −0.514 to 0.153).

Neither comparison reached statistical significance. However, between one and three months, BoNT injections were significantly superior to corticosteroids, with a pooled SMD of −0.648 (95% CI: −1.071 to −0.225).

For shoulder myofascial pain, five studies were analyzed. At one month post-intervention, BoNT injections did not differ significantly from reference treatments: compared to saline, the pooled SMD was −0.212 (95% CI: −0.551 to 0.127; I² = 68.8%); compared to dry needling, the SMD was 0.665 (95% CI: −0.260 to 1.590); and compared to lidocaine, the SMD was 1.093 (95% CI: 0.128 to 2.058). Between one and three months, BoNT injections were significantly superior to saline, with a pooled SMD of −0.314 (95% CI: −0.516 to −0.111; I² = 3.3%).

Regarding safety, eight of nine trials clearly reported adverse events. Most documented adverse events were temporary and resolved spontaneously, with mild to moderate severity. Common adverse effects included injection site discomfort, flu-like symptoms, and muscle weakness. Only one study reported serious adverse events (chest pain and scheduled cataract surgery), but these were attributed to underlying diseases rather than BoNT injections.

Publication bias was detected for myofascial pain at one month (Egger test p = 0.03 for joint pain, p = 0.047 for myofascial pain) but not for the one to three month time period.

This meta-analysis provides important evidence that BoNT injections can be a safe and effective alternative for patients with chronic shoulder pain, with particular advantages for longer-term pain relief. For patients with shoulder joint pain conditions like osteoarthritis, subacromial bursitis, and adhesive capsulitis, BoNT offers comparable short-term pain relief to corticosteroid injections but appears to provide more durable benefits between one and three months. This is clinically significant because corticosteroid injections, while commonly used, often lead to recurrent pain, and their effects typically last only about eight weeks compared to approximately three months for BoNT.

For patients with myofascial pain and trigger points, BoNT injections provide similar immediate pain relief as other interventions like dry needling, lidocaine injections, or saline injections. The similar short-term effects across all these treatments may be partly due to the "needling effect" itself—simply inserting a needle into a trigger point can provide temporary analgesia and increase muscle circulation. However, BoNT's pharmacological mechanism appears to provide longer-lasting benefits beyond this initial needling effect, as evidenced by its superiority over saline at one to three months.

Clinically, these findings suggest that BoNT injections may be particularly valuable when longer-lasting pain relief is desired, or when patients have not responded well to corticosteroid injections. The safety profile appears favorable, with mostly mild and temporary adverse effects. However, clinicians should be aware that the total evidence base is still relatively small (only nine RCTs), follow-up periods were limited to three months, and the optimal dosing and injection techniques remain to be fully established. Future research with longer follow-up periods and functional outcome measures would help strengthen recommendations for clinical practice.