Other Investigations

Resources for Women with Epilepsy (all links will open a new page in an external website)


Retrospective NEAD Study (RetroNEAD information on ClinicalTrials.gov)

Children of women with epilepsy are at increased risk for cognitive impairment. However, it remains unclear if there are differential effects across AEDs (see detailed protocol for additional background). The recent retrospective study by Adab et al. (2001) raises serious concerns that valproate may impair cognitive development to a greater extent than other commonly used AEDs, but selection bias cannot be ruled out. If the same pattern of results was found in a completely separate cohort from a different country, it would provide striking evidence that the observation is genuine. The results of such a study would have direct impact on the management of women with epilepsy.

The specific aim of this study is to determine if in utero exposure produces differential antiepileptic drug (AED) effects on subsequent cognitive abilities and behavioral abnormalities in children.

The study will be conducted as a multicenter retrospective parallel group study. A total of 90 mother/child pairs and their maternal relatives and fathers will be assessed. Subjects will be assigned to a research group according to their specific AED monotherapy

Research Group


AED Monotherapy Type













 Subject groups will include 6 - 16 year/old children of women with epilepsy, who are in one of the above three groups in regards to in utero AED exposure. Based on data from the study by the Adab et al. (2001) study, 30 per group will be required. Thus, there will be a total of 90 children. In addition to the children, their mother, father and a primary relative of the mother will also participate in the study. Children cannot be enrolled if they were exposed to another known teratogen during their mother’s pregnancy other than the target AEDs in monotherapy; if the child has or has had a serious brain injury (e.g. severe head trauma) or serious medical illness (e.g. cancer), which is unrelated to AED exposure and may affect the child’s cognitive abilities. Other exclusion criteria include, maternal IQ <70; mother has a progressive brain lesion (e.g. tumor) or mother had serious medical illness or complication during this child’s pregnancy, which was unrelated to AED or epilepsy.

Children of the same mother may be enrolled if each child was exposed to a different AED monotherapy during their mother’s pregnancy.

The child should be off sympathomimetic medications (e.g. Ritalin) for 24 hours prior to testing if he/she is being treated for ADD or ADHD.

The primary outcome variable is the IQ (TONI-3) in the children. Additional outcome variables in the children include the Peabody Picture Vocabulary Test –3; WRAT spelling & reading subtests; Children’s Memory Scale: story memory subtest (immediate, delayed, & delayed recognition subscales); WISC-III coding subtest; Behavioral Assessment System for Children (parent and teacher scales) and the Conner’s Rating Scale – (parent and teacher rating scales).

Control variables include the IQ for the mother, father, and primary relative of the mother (if father and primary relative are available). The primary relative of the mother will be recruited according to the following criteria:

Other outcome measures will also include social and medical factors of the mother, specifically age, type of epilepsy, AED & estimated months exposed and average dose during pregnancy, estimated seizure frequency during pregnancy (by types including status), any other major medical illnesses during pregnancy, primary language, education level and socioeconomic status.

Factors assessed in the child include: age, gender, gestational age at birth (i.e. full or pre-term), birth order & # sibs, primary language at home, Birth defects, Hx ADD or ADHD; other major medical illnesses (include perinatal complications); special education requirements of the child; and remedial math or reading requirements of child.

Factors assessed in the father and primary relative include: age, relationship to mother (for primary relative), primary language, education level, socioeconomic status, any major medical illnesses (especially those which could affect IQ).


Congenital Malformations

The North American Pregnancy Registry has found that 6.5% of children exposed to phenobarbital and 10.7% of children exposed to valproate during pregnancy suffer major congenital malformations [1, 2]. These risks are statistically greater than the risk of 1.62% in the general population malformation. The risk for valproate was also statistically greater than the combined mean rate for all other AEDs.

The Australian Pregnancy Registry also reported a significantly greater risk of malformations for children exposed to valproate during pregnancy than children exposed to other AEDs or to no AED) [3]; 16.5% of the children exposed to valproate were born with congenital malformations. In addition, the effect of valproate appear to increase with greater valproate dosages.

The International Lamotrigine Pregnancy Registry, a pharmaceutical company-based registry, found that the children of women on AED polytherapy that included valproate had a higher risk (12.5%) of major malformations than children whose mothers were on AED polytherapy without valproate (2.7%) [4].

In a preliminary report, the United Kingdom Pregnancy Registry has noted a risk of 6.0% for valproate, which is significantly greater than carbamazepine [5]. As discussed above, the preliminary results for our NEAD Study also implicates a greater risk for valproate.

Cognitive Effects of AED Exposure During Pregnancy

A retrospective study from Denmark found that prenatal exposure to phenobarbital significantly lowered verbal IQ scores (approximately 7 IQ points) in two groups of men exposed during their mothers’ pregnancies [6].

A retrospective study from Scotland reported 24% risk of developmental delay in children exposed to AEDs compared to 11% in non-exposed siblings [7]. T he risk was elevated for carbamazepine, phenytoin and valproate monotherapy. AED polytherapy appeared to have the highest risk, and phenobarbital monotherapy had the lowest relative risk.

A prospective study from Finland reported a greater risk for impaired cognition in children exposed to valproate [8]. Verbal IQ was 82 for children exposed to valproate monotherapy, which was significantly worse than children exposed to carbamazepine monotherapy (verbal IQ = 96) or children not exposed to drugs (verbal IQ = 95).

In a series of retrospective studies our colleagues from Liverpool and Manchester, UK have reported greater risk for cognitive problems in children exposed to valproate during pregnancy [9-11]. They reported that 30% of the children exposed in utero to monotherapy valproate required special education compared to 3% to 6% for other AED monotherapy groups [9]. Further, verbal IQ in the children at 6 to 16 years old was 8 to 15 points lower than children exposed to other AEDs. Valproate also produced developmental delay in a second group of children less than 6 years old compared to children exposed to other AEDs [10].

The above studies provide information on cognitive effects of AED exposure during pregancy, but the results are not conclusive. Additional studies are needed to confirm these findings and to clarify the risks across all AEDs.


Thus, investigations of 3 separate patient groups for phenobarbital and 8 separate patient groups for valproate have implicated increased risk for malformations or behavioral problems in children exposed to these 2 AEDs during their mother’s pregnancy.


Table 1. Anatomical Teratogenecity.* Percent major congenital malformations (%CM) with 95% confidence intervals (95% CI) from six recent prospective observational studies in women with epilepsy. CBZ = carbamazepine, LTG = lamotrigine, PB = phenobarbital, VPA = valproate.




95% CI

Significant Findings

Published Studies

N. American Registry 2004 2



2.1 -14.5

Greater than general population (1.62%)

Australian Registry 2004 4



2.25 -20.3

Greater no AED (3.12%) & other AEDs (3.00%)

*prospective data only

N. American Registry 2005 3



6.3 -16.9

Greater than gen pop (1.62%) & other AEDs (2.9%)

International Lamotrigine Registry 2005 5









Polytherapy without VPA

Polytherapy with VPA

Preliminary Reports

UK Registry 2005 6




Greater than CBZ (2.3%)

NEAD Study 2005 7



Greater than other AEDs


Table 2. Behavioral Teratogenecity.* Findings in selected studies of developmental delay in children of women with epilepsy, who were exposed in utero to AEDs.


Type of Study



Denmark 1995 9



Verbal IQ reduced 7 points

Liverpool/Manchester, UK

2001 10, 2004 11, 2005 12

Retrospective (2 cohorts)


1. Increased need for special

education in children exposed

to VPA (30% vs. 3-11%)

2. Verbal IQ reduced 8-15 points

3. Increase memory deficits

4. Delayed developmental in

2 nd cohort < 6 years old

Finland 2004 13



Verbal IQ reduced 11-12 points for VPA, but no reduction for CBZ.

*Tables 1 and 2 reproduced with permission from Epilepsy Currents.