ABSTRACT A total of 222 pregnancies in 56 mothers were studied, 48 pregnancies occurred in 17 mothers before the onset of epilepsy (Group 1). The same group went on to develop epilepsy, receive anticonvulsants and further conceive 43 pregnancies (Group la). In the other 39 mothers 131 pregnancies occurred after the onset of epilepsy and while receiving anticonvulsants (Group 2). The incidence of major congenital malformations among live births was significantly higher in Group 2 (14.4%) when compared with all pregnancies in Groups 1 and la (1.2%), P<0.025 and in Group 1 alone (2.2%) P<0.05. There was no significant difference between Groups 1 and Group la. Age at first pregnancy in Group 1 was 18.9 years ± 3.7 SD, which was significantly younger than Group 2, 21.7 years ± 5.1 SD, P<0.05. The age of onset of epilepsy was significantly younger in Group 2 than in Group 1, 15.1 years ± 6 SD vs 23.3 years ± 4.6 SD, P<0.001. The duration of epilepsy was significantly longer in Group 2 compared with Group 1, 13.8 years ± 8.7 SD vs 6.1 years ± 2.6 SD, P<0.001. Patients on polytherapy had a significantly higher risk of major malformations compared to those on monotherapy, 20.2% vs 1.3%, P<0.001. Our data suggest that epileptic mothers with the highest risk of having offspring with major malformations were those with a long duration of epilepsy combined with exposure to multiple anticonvulsant drugs.