Migraine is a common chronic neurological disorder that poses a significant burden to global health. Investigating the risk factors for migraine is crucial for its prevention and management. Air pollution, one of the greatest environmental risks to public health today, has not yet been conclusively linked to migraine, and the quality of existing studies varies widely. To avoid potential confounders in observational studies, this research utilizes the 2-sample Mendelian randomization (MR) approach for the first time to reassess the causal relationship between air pollution indicators such as PM2.5, PM10, PM2.5 to 10, nitrogen dioxide (NO2), and nitrogen oxides (NOx), and migraine. This study employs a 2-sample MR design, utilizing genome-wide association summary data on air pollution and migraine from the IEU and FinnGen databases. The primary method used is inverse-variance weighting, combined with 4 other MR analysis techniques, to thoroughly investigate the data. Concurrently, a series of sensitivity analyses are conducted using Cochran Q test, MR-Egger intercept regression, and MR-PRESSO. The combined MR and sensitivity analysis of various air pollutants and migraine showed a high degree of robustness across all data except for the PM10 results. After applying the Bonferroni correction, all MR analysis results did not reach the significance threshold for p-values, indicating no statistically significant causal relationship between air pollution and migraine. This study did not find a significant causal relationship between air pollution and migraine. Combining previous studies, we speculated on the potential reasons behind our analysis results. Considering the limitations of this study, we recommend further research to validate our analysis results and to explore this topic more deeply.
Zhu et al. (Fri,) studied this question.