We aimed to confirm or reject previous reports on the association of prenatal antibiotic exposure and development of epilepsy in offspring by accounting for known and unidentified confounding factors. In a retrospective cohort investigation, we enrolled children aged 3-18 years born between 1998 and 2012 at a single regional hospital and their mothers. A computerized medication database was linked with hospital records. The exposed group included children whose mothers purchased 1 or more antibiotic medications for use during pregnancy. Epilepsy was defined by epilepsy diagnosis and/or by chronic dispensing of antiepileptic drugs. We analyzed maternal exposure to antibiotics 2 years after delivery (but not during pregnancy and/or the 2 years following delivery) as part of the specificity analysis. We enrolled 88 899 children and their 74 416 mothers. The group exposed prenatally to antibiotics comprised 36 622 children (41.2%). Of them, 326 (0.9%) developed epilepsy compared with 370 of 52 277 (0.7%) in the unexposed group (relative risk [RR], 1.24; 95% confidence interval [CI], 1.07-1.44: P = .004). Exposure during the first, second, and third trimesters was characterized by incidence of epilepsy in 0.8% (P = .943), 0.9% (P = .266), and 0.9% (P = .073) of children, respectively, compared with the unexposed group, with an RR of 1.01 (95%CI, 0.83-1.23), 1.12 (95%CI, 0.92-1.36), and 1.19 (95%CI, 0.98-1.45), respectively. Similarly, prenatal exposure by antibiotic class was associated with epilepsy. Nevertheless, the specificity analysis strongly suggested the possibility of confounding by indication. Our findings indicated that pregnant women should receive the indicated antibiotic treatment with no fear of the development of epilepsy in their children.
Keywords: antibiotics; childhood epilepsy; pregnancy; prenatal; specificity analysis.
© 2020, The American College of Clinical Pharmacology.