Comparative efficacy of vitamins and nutraceuticals for migraine prevention: A systematic review and network meta‐analysis of randomized controlled trials
Muhammad Ahmed, Ume Aiman, Niraj Kukreja, Syed Muhammad Rayyan, Umer Bin Shahzad, Huzaifa Sabir Nawaz, Muhammad Talha, Talha Ilyas Khan, Bareera Asad, Aamna Raheel Sheikh, Muhammad Faisal Iqbal, Hamad SaifAbstract
Objectives/Background
Migraine is one of the leading causes of neurological disability globally, and conventional prophylactic pharmacotherapy may be limited by incomplete efficacy, poor tolerability, contraindications, and concerns about long‐term safety. Vitamins and nutraceuticals including riboflavin, coenzyme Q10 (CoQ10), magnesium, and vitamin D3 are increasingly used as adjunctive or alternative preventive strategies, yet their comparative efficacy within a single evidence synthesis remains uncertain. Our objective was to evaluate and rank the comparative efficacy and safety of vitamins and nutraceutical interventions for migraine prevention using a network meta‐analysis (NMA) of randomized controlled trials (RCTs).
Methods
We conducted a Preferred Reporting Items for Systematic Reviews and Meta‐Analyses‐compliant systematic review and NMA International Prospective Register of Systematic Reviews (PROSPERO: CRD420261389392). PubMed, Embase, Scopus, Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, and Google Scholar were searched from inception to March 2026 for double‐blind RCTs evaluating vitamins or nutraceuticals for migraine prevention in adults or children. The primary outcome was monthly attack frequency; achievement of ≥50% responder status was a secondary outcome. Secondary outcomes included attack duration, pain severity, and migraine days. The primary network was restricted to adults; pediatric trials were summarized separately. Shared‐unit outcomes (attack frequency, migraine days, attack duration) were reported as mean differences (MDs) and pain severity as standardized MDs; responder status was reported as odds ratios, all with 95% confidence intervals (CIs). Certainty of evidence was assessed using the Confidence in Network Meta‐Analysis (CINeMA) (Grading of Recommendations Assessment, Development and Evaluation‐NMA) framework.
Results
Twenty unique double‐blind RCTs (reported in 21 publications; n = 1829 randomized; 1600 analyzed) were included; 10 adult trials informed the primary attack‐frequency network. For attack frequency, riboflavin 400 mg showed the most reliable reduction in adults (MD = −1.95 attacks per month, 95% CI = −2.42 to −1.49; moderate certainty); larger point estimates were observed for vitamin B12 500 μg (MD = −2.50, 95% CI = −4.35 to −0.65), vitamin B6 80 mg (MD = −2.33, 95% CI = −4.19 to −0.48), and vitamin D3 (MD = −2.29, 95% CI = −3.03 to −1.54). The vitamin B12 and B6 estimates were derived from single small trials and the vitamin D3 estimate from two small trials, all of low to very low certainty. Riboflavin 100 mg and 200 mg contributed no adult attack‐frequency data (all estimates versus placebo). For ≥50% responder status, CoQ10 300 mg was the only intervention with a statistically significant advantage in adults (odds ratio = 4.83, 95% CI = 1.17–19.89; very low certainty). The previously reported riboflavin 200 mg signal was derived from a pediatric trial. CoQ10 400 mg significantly reduced attack duration (MD = −3.30 h, 95% CI = −5.72 to −0.88; low certainty, downgraded for imprecision) and pain severity (standardized MD = −0.70, 95% CI = −1.12 to −0.27; low certainty). However, responder‐status findings were limited by network sparsity and imprecision and should be interpreted cautiously. Magnesium showed no detected benefit across outcomes; adverse‐event proportions were highest for magnesium but are reported descriptively only and are not comparative evidence.
Conclusion
In this systematic review and NMA, in adults, riboflavin 400 mg showed the most consistent, moderate‐certainty reduction in migraine attack frequency, whereas CoQ10 400 mg showed potential benefit for reducing attack duration and pain severity. Vitamin D3 and selected B vitamins also demonstrated frequency‐lowering effects, although estimates were limited by sparse evidence and imprecision. Magnesium showed no detected benefit; adverse‐event proportions were descriptive only and not comparative evidence. Because several treatment nodes were informed by few studies and responder estimates were imprecise, these findings should be interpreted cautiously. Larger, standardized, and adequately powered trials are needed to confirm comparative efficacy and safety.