The internal control group in a register of antiepileptic drug use in pregnancy.

The aim of this study was to determine how valid 68 first-trimester pregnancies of untreated epileptic women would prove as an internal control group for investigating foetal malformation rates in 709 simultaneously collected antiepileptic drug-exposed pregnancies in an Australian register of pregnancies in epileptic women. We carried out comparisons of values for parameters relating to personal details, obstetric aspects, and epilepsies prior to and during pregnancy in the drug-exposed and drug-unexposed pregnancies, with observations on subpopulations within the drug-unexposed group. Statistically significant (p<0.05) differences existed for only seven of more than 50 parameters compared. None of these seven parameters had a statistically significant influence on foetal malformation rates in the whole dataset. In 23 of the 65 epileptic pregnancies unexposed to antiepileptic drugs, therapy had been ceased shortly prior to pregnancy and was often resumed after the first trimester. In the remaining 42, therapy had been ceased earlier, often despite continuing seizures. Planned withdrawal of therapy did not appear to produce additional hazards for mothers and foetuses in the former subgroup. In the data collection studied, there did not appear to be evidence of statistically significant differences between untreated pregnancies and treated epileptic pregnancies that would be likely to invalidate the former group as an internal control for the latter, at least when assessing foetal malformation rates.