Postmarketing pharmacovigilance study of levothyroxine based on FDA adverse event reporting system (FAERS) and Mendelian randomization.

This study utilizes adverse event records for levothyroxine from the FAERS database to explore and evaluate its post-marketing adverse events. Using levothyroxine adverse event records from the FAERS database, four methods-report ratio odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multivariable gamma-Poisson shrinkage model (MGPS)-were applied to assess pharmacovigilance signals for levothyroxine adverse reactions. Causality between levothyroxine and adverse events was assessed via two-sample Mendelian randomization (MR) analysis. This study identified multiple adverse events not listed in levothyroxine prescribing information, including dizziness, depression, myalgia, migraine, celiac disease, lactose intolerance, autoimmune thyroiditis, and nail disorders. Mendelian randomization analysis indicated a causal relationship between levothyroxine and myalgia, migraine, celiac disease, lactose intolerance, and autoimmune thyroiditis. These findings provide valuable reference information for reducing clinical risks associated with levothyroxine use. Clinicians should be vigilant for these adverse reactions when prescribing levothyroxine.