Comparative Efficacy of Acupuncture, Pharmacological Interventions, Sham Acupuncture, and Placebo for Migraine Prevention: A Bayesian Network Meta-Analysis.

OBJECTIVE

This study aimed to compare the preventive effects of acupuncture (ACU), pharmacological interventions (PIs), sham ACU, and placebo for migraine through a Bayesian network meta-analysis and to evaluate the relative efficacy of different treatment approaches.

METHODS

By reviewing PubMed, Embase, MEDLINE, and Web of Science databases, 32 randomized controlled trials (RCTs) were included in this study. Bayesian network analyses were performed using Stata and R software to compare the effects of ACU, PI, sham ACU, and placebo on reducing headache frequency, reducing monthly migraine days, alleviating headache intensity, and decreasing the use of analgesics. Treatment effects were estimated using the Markov Chain Monte Carlo method, and model convergence and heterogeneity were evaluated. The results were presented as pairwise comparisons and forest plots.

RESULTS

A total of 32 RCTs were included. Network meta-analysis evaluated four intervention categories: ACU, PI, sham ACU, and placebo. For monthly migraine attack frequency, ACU was significantly more effective than placebo (MD = -2.30, 95% CrI: -3.70 to -1.00), and PI also showed superiority over placebo (MD = -1.70, 95% CrI: -2.50 to -0.82), while no significant differences were observed between ACU and PI or sham ACU. For monthly migraine days, both ACU (MD = -1.78, 95% CrI: -2.89 to -0.70) and PI (MD = -1.72, 95% CrI: -2.29 to -1.16) were superior to placebo, with no significant difference between ACU and PI (MD = -0.07, 95% CrI: -1.00 to 0.87). ACU demonstrated greater effects than PI in reducing headache intensity (MD = -0.99, 95% CrI: -1.94 to -0.02) and analgesic use frequency (MD = -1.12, 95% CrI: -1.98 to -0.08). However, ACU did not significantly differ from sham ACU across evaluated outcomes. Ranking analysis consistently indicated that ACU had the highest probability of being the most effective intervention, whereas placebo ranked lowest. These findings suggest that ACU may provide clinically meaningful benefits for migraine prevention, while the measurable effects observed in sham ACU groups highlight the methodological challenges of using sham ACU as an inert placebo control.

CONCLUSION

This Bayesian network meta-analysis suggests that ACU is effective for migraine prevention, showing significant benefits compared with placebo and comparable efficacy to PI. However, the lack of significant differences between ACU and sham ACU highlights the methodological challenges of sham ACU as a control intervention. Future studies should optimize sham ACU design to better evaluate the specific effects of ACU.