Preterm birth and risk of epilepsy in Swedish adults

Objective: To determine whether preterm birth is associated with epilepsy in a national cohort of adults aged 25-37 years. Methods: We conducted a national cohort study of 630,090 infants born in Sweden from 1973 through 1979, including 27,953 born preterm (+37 weeks), followed from 2005 to 2009 for 1) hospitalization for epilepsy and 2) outpatient and inpatient prescription of antiepileptic drugs. Epilepsy diagnoses and medication data were obtained from all hospitals and pharmacies throughout Sweden. Results: We found a strong association between preterm birth and epilepsy that increased by earlier gestational age. After adjusting for fetal growth and potential confounders, odds ratios for hospitalization for epilepsy were 4.98 (95% confidence interval [CI] 2.87-8.62) for those born at 23-31 weeks, 1.98 (95% CI 1.26-3.13) for those born at 32-34 weeks, and 1.76 (95% CI 1.30-2.38) for those born at 35-36 weeks, relative to those born full-term (37-42 weeks). A similar but slightly weaker trend was observed for the association between preterm birth and antiepileptic drug prescription. These associations persisted after excluding individuals with cerebral palsy, inflammatory diseases of the CNS, cerebrovascular disease, and brain tumors. Conclusions: These findings suggest that preterm birth, including late preterm birth, is strongly associated with epilepsy in Swedish adults aged 25-37 years. This association was independent of fetal growth and was not mediated by cerebral palsy or other comorbidities.