OBJECTIVE: To assess the risk of teratogenicity from maternal intake of the more widely used newer antiepileptic drugs, especially lamotrigine, levetiracetam and topiramate. MATERIALS AND METHODS: Use of confidence interval and regression methods to compare risks of foetal malformation in pregnancies in women exposed (n = 1572) and in women with epilepsy not exposed (n = 153) to antiepileptic drugs in the first trimester. RESULTS: Compared with the foetal malformation rate in women with epilepsy who were untreated in the first trimester (3.3%), the malformation rates for lamotrigine (4.6%), levetiracetam (2.4%) and topiramate (2.4%), all in monotherapy, were not statistically significantly different. However, the malformation rates for topiramate as part of polytherapy (14.1%) and for valproate in both monotherapy (13.8%) and polytherapy (10.2%) were statistically significantly higher. Regression analysis of combined monotherapy and polytherapy data showed no statistically significant increased risk of teratogenesis associated with lamotrigine or levetiracetam, but a statistically significant and dose-related risk for first trimester topiramate (P = 0.01) and valproate (P < 0.0001) exposure. CONCLUSIONS: Evidence from this and other studies suggests that lamotrigine and levetiracetam have low risk for teratogenesis, but that topiramate exposure early in pregnancy may be associated with dose-related anatomical teratogenesis, as valproate is already known to be.