Proton Pump Inhibitor Use and Incident Epilepsy in the UK Biobank: Evidence From Pharmacovigilance and Multi‐Omics Analyses
Minghao Li, Yiqian Li, Yuxi Zhang, Balkissou Seebo Saibou, Yifei Chen, Jiayi Xing, Chao Jin, Yuchen Xu, Jiahe Lin, Xinshi WangABSTRACT
Aims
Proton pump inhibitors (PPIs) are widely used, but their association with incident epilepsy remains uncertain. We examined this association in the UK Biobank and integrated evidence from pharmacovigilance and multi‐omics.
Methods
We included 492,858 UK Biobank participants without epilepsy at baseline. Cox models, age‐ and sex‐stratified analyses, active‐comparator analyses against H2RA users, propensity score matching, competing‐risk models, and sensitivity analyses were performed. FAERS disproportionality, Olink proteomics, and NMR metabolomics provided supportive evidence.
Results
Baseline PPI use was associated with incident epilepsy after full adjustment (HR 1.43, 95% CI 1.29–1.59) and in the PPI‐H2RA propensity score‐matched analysis (HR 1.41, 95% CI 1.03–1.94). In the GP‐linked subset, requiring an ASM prescription within 365 days after diagnosis yielded HR 1.47 (95% CI 1.07–2.02). FAERS showed a PPI‐epilepsy reporting signal (ROR 1.70, 95% CI 1.61–1.79). Proteomic and metabolomic analyses identified immune‐inflammatory and metabolic associations.
Conclusion
Baseline PPI use was associated with incident epilepsy in the UK Biobank. Active‐comparator, outcome‐definition, and prescription‐supported analyses supported the primary association. Pharmacovigilance and molecular findings provided complementary evidence but did not establish causality.