Abstract
OBJECTIVE: Topiramate has been linked to increased glaucoma risk, potentially through mechanisms involving ocular fluid shifts. However, comparative risks vs other antiseizure medications (ASMs) and variation by sex or indication remain uncertain. This study evaluates glaucoma incidence in topiramate initiators compared to valproate or lamotrigine users among patients with epilepsy or migraine. METHODS: We conducted a retrospective active-comparator, new-user cohort study using electronic health records from the IQVIA Medical Research Data among patients with epilepsy or migraine initiating topiramate, valproate, or lamotrigine. Patients with prior ASM use, limited washout period and follow-up, or pre-existing glaucoma were excluded. The outcome was incident glaucoma within 1 year, censored at glaucoma occurrence, death, discontinuation, switch, or September 30, 2023. Covariates included age, sex, race, lifestyle factors, comorbidities, and medication history. Propensity score-based inverse probability weighting balanced characteristics, and crude and weighted Cox models estimated hazard ratios (HRs) with 95% confidence intervals (CIs). Subgroup analyses were conducted by sex, age, and indication. RESULTS: The cohort included 688 topiramate, 4490 valproate, and 4179 lamotrigine initiators. After weighting, the 1-year absolute risk increase was approximately 2.4% when comparing topiramate to valproate, and about 2.0% compared to lamotrigine. Topiramate was associated with higher glaucoma risk vs valproate (adjusted HR 2.66, 95% CI 1.12-6.32) and lamotrigine (adjusted HR 3.57, 95% CI 1.76-7.26). Risks were elevated in female (vs valproate: HR 5.31, 95% CI 1.48-19.08; vs lamotrigine: HR 5.73, 95% CI 2.38-13.79) or epilepsy patients (vs valproate: HR 2.23, 95% CI 1.04-4.76; vs lamotrigine: HR 5.08, 95% CI 2.32-11.14), but not in male or migraine patients. SIGNIFICANCE: Topiramate use substantially increases glaucoma risk compared with valproate and lamotrigine, particularly among female or epilepsy patients. No significant association in male or migraine patients was observed. These findings may inform targeted ophthalmologic monitoring in high-risk groups and use of alternative ASMs.