Assessment of drug-related migraine in a real-world large-scale database

Abstract

Background: Drug-induced migraine represents a clinically significant yet under-investigated subtype of migraine. This study aims to evaluate the risk of drug-related migraine based on real-world data from the FDA Adverse Event Reporting System (FAERS).

Methods: A retrospective pharmacovigilance analysis was conducted using FAERS data from Q1 2004 to Q4 2024. Migraine cases were identified via standardized MedDRA (The Medical Dictionary for Regulatory Activities) terms. Only primary suspect drugs were included. Disproportionality analyses were performed using four algorithms: ROR, PRR, MGPS, and BCPNN. Drugs were classified by therapeutic indication and mechanism of action, and stratified by BCPNN values to assess risk levels.

Results: A total of 20,886 migraine-related adverse events were identified, predominantly among females (77.4%) with a mean age of 45.7 years. Sixty-six drugs yielded positive signals, and after exclusion criteria, 39 remained for further analysis. The highest-risk agents included lorcaserin (BCPNN = 3.33), tasimelteon (3.20), and botulinum toxin type A (3.06). High-risk therapeutic classes included immunosuppressants, estrogens/progestogens, and sedative-hypnotics.

Conclusion: This large-scale analysis identifies key drug categories and compounds associated with an elevated risk of migraine, providing actionable insights for clinicians. Especially lorcaserin, tasimelteon, and botulinum toxin as potential risk factors for migraine. Given the public health burden of migraine, pharmacovigilance efforts should incorporate such findings to mitigate iatrogenic risks. Further prospective studies are warranted to establish causal mechanisms and optimize therapeutic decision-making.

Keywords: BCPNN; FAERS; disproportionality analysis; drug-induced headache; migraine; pharmacovigilance.

FIGURE 1

The selection of migraine-related cases and the data cleaning process in the FAERS database.

FIGURE 2

Distribution of baseline data for patients reporting migraine-related adverse events in the FAERS database. (a) Age distribution pyramid by gender for patients reporting migraine-related adverse events. (b) Temporal distribution of migraine-related adverse event reports. (c) Distribution of routes of administration for migraine-related adverse events. (d) Histogram of outcomes for patients with migraine-related adverse events. (e) Distribution of drug use causing migraine-related adverse events.

FIGURE 3

Detection of positive signal drugs through disproportionality analysis and their categorization based on mechanism of action. Notes: This figure classifies these positive signal drugs according to their mechanisms of action, where higher signal values indicate a greater risk of drug-related macular degeneration. The Reporting Odds Ratio (ROR) is presented as a measure of relative risk. Abbreviations: CI, confidence interval.

FIGURE 4

Risk levels and reporting volumes for drug-related migraine adverse reactions, sorted by decreasing risk and reporting volume. (a) Risk levels of positive drugs assessed using the BCPNN algorithm. (b) Reporting volumes for drug-related migraine adverse reactions, decreasing from high to low.