Antiepileptic drugs and neurocognitive development.

In a prospective observational study published in the current issue of Neurology®, Shallcross et al. compared cognitive and language skills of children born to women with epilepsy (WWE) exposed in utero to levetiracetam (LEV, n 5 53) or sodium valproate (VPA, n5 44), and control children born to mothers without epilepsy or medication exposure during pregnancy (n5 133). WWE were recruited from the UK Epilepsy Pregnancy Registry. Children were evaluated at the age of 36–54 months, building on previous findings at 24 months of age. The authors found no difference between LEV-exposed and control children. VPA-exposed children had lower scores on tests of gross motor skills, comprehension, and expressive language abilities than LEV-exposed children. The authors conclude that in utero exposure to VPA, but not LEV, is associated with neurodevelopmental deficits of motor and language abilities. The present study adds information regarding potential effects of antiepileptic drugs (AEDs) used by WWE during pregnancy on the child’s cognitive development. In the Neurodevelopmental Effects of Antiepileptic Drugs (NEAD) study, children born to WWE using VPA monotherapy during pregnancy had lower IQs (7–10 points) and verbal and memory abilities at 6 years of age than children born to WWE treated with lamotrigine, carbamazepine, and phenytoin. Both studies show an adverse effect of VPA exposure on cognitive development, carrying important implications for clinical practice. The present study also suggests that LEV, like lamotrigine, carbamazepine, and phenytoin, may not adversely affect cognitive development. One of the strengths of the study is inclusion of a control group, absent in the NEAD study. However, the study has several shortcomings that may limit generalization of its findings. The number of exposed subjects was relatively small. The LEV and VPA groups differed markedly in their response rate to the invitation to participate in the study (82% of LEV-contacted mothers responded vs 33% VPA-contacted mothers), raising the possibility of a bias, and in socioeconomic status (SES; more professional, fewer manual/unemployed mothers in the LEV vs VPA groups, more smokers in the VPA group). The difference in SES could indicate a difference in maternal biology and nurturing environment between the groups in favor of LEV that could affect cognitive development of the offspring independently of AED treatment. There is a paucity of information regarding seizure frequency, type, and severity. Seizures during pregnancy were associated with poorer developmental outcome. However, it is not known whether seizure type, severity, or frequency was important in determining that association, and whether these variables differed between the VPA and LEV groups. For example, occurrence of generalized tonic-clonic (GTC) seizures during pregnancy is associated with shorter gestational age and reduced birthweight, which could affect subsequent development. In clinical practice, refractory primary GTC seizures are preferentially treated with VPA, so it would have been helpful to know whether the VPA group had more GTC seizures. There is no information about AED doses and levels, nor about use of folate, which has recently been shown to positively affect IQ development. Although the authors state that “no dose effect was detected,” they do not provide supporting data; in the NEAD study, dose effects were present. Accumulating reports of VPA’s teratogenicity, and now its potential to impair neurodevelopmental outcome, have resulted in a growing reluctance by neurologists to prescribe VPA to women of reproductive age. American Academy of Neurology and American Epilepsy Society guidelines suggest that if seizures can be controlled without substantial side effects by one of the AEDs for which pregnancy outcome and neurodevelopmental data appear superior to VPA, i.e., lamotrigine, carbamazepine, and levetiracetam, those AEDs should be used in preference to VPA. The decision is more difficult for those women whose seizures fail to respond to these other AEDs and may respond to VPA. In this relatively small but important group of women with refractory primary generalized epilepsy or symptomatic epilepsy with primary generalized seizures, the practitioner is challenged with the common dilemma of whether the potential benefit of treatment outweighs the potential harm of the medication. For refractory generalized seizures, the data are lacking to make an informed, rational choice. Adding to the choice