Valproate in Pregnancy Linked to Reduced IQ in Children

January 23, 2013

Further evidence that use of valproate in pregnancy leads to a reduction in cognitive abilities in offspring has been reported.

The results, from the well-known Neurodevelopmental Effects of Antiepileptic Drugs (NEAD) study, are reported by Kimford J. Meador, MD, Emory University, Atlanta, Georgia, and colleagues and are published online January 23 in The Lancet Neurology.

"Valproate is a very poor first choice for a woman of childbearing potential," Dr. Meador commented to Medscape Medical News. "It has a known malformation rate of about 10%, which started to become evident in 2004. On top of this is the reduced cognition effect, which was first seen in 3 year old children earlier in this study. The current results at 6 years drum this home."

Their latest results show that valproate was associated with an IQ reduction of around 7 to 10 points at age 6. "That is pretty dramatic," he said. "It would affect a person's ability to work and earn a good living. You wouldn't want that for your child and most women would not choose to take valproate if they knew the risks."

Dr. Meador said that the drug may have to be used in a subset of women with primary generalized epilepsy, but he would always try other drugs first. "And if valproate is the only one that works, I would try and keep the dose as low as possible." He added, "I would say at this point lamotrigine, carbamazepine and phenytoin look fairly safe in pregnancy."

Dr. Kimford J. Meador

The findings were first released at the American Academy of Neurology meeting in 2012 and reported by Medscape Medical News at that time.

NEAD Study

The NEAD study enrolled 305 pregnant women with epilepsy taking 1 antiepileptic drug (carbamazepine, lamotrigine, phenytoin, or valproate) between 1999 and 2004 at 25 epilepsy centers in the United Kingdom and United States. The primary outcome was IQ in the children at 6 years adjusted for maternal IQ, antiepileptic drug type, standardized dose, gestational birth age, and use of periconceptional folate.

Of the 311 resulting children, 224 completed 6 years of follow-up. Results showed that IQ at age 6 was lower after exposure to valproate than after exposure to the other drugs.

Table 1. Mean IQ at Age 6 in Children With In Utero Exposure to Antiepileptic Drugs

Antiepileptic Drug Mean IQ (95% Confidence Interval) P Value (vs Valproate)
Valproate 97 (94 - 101)
Carbamazepine 105 (102 - 108) .0015
Lamotrigine 108 (105 - 110) .0003
Phenytoin 108 (104 - 112) .0006


Children exposed to valproate also did poorly on measures of verbal and memory abilities compared with those exposed to the other antiepileptic drugs and on nonverbal and executive functions compared with lamotrigine (but not carbamazepine or phenytoin).

The negative effects of valproate were dose related, which is what would be expected for a teratogenic drug, Dr. Meador said.

Maternal IQ was related to child IQ for all the drugs except valproate, suggesting that valproate exposure disrupts this otherwise strong association, the authors report.

They also found that both valproate and lamotrigine were associated with a lower frequency of right-handedness than that in a sample of normal children used a reference, which they say may suggest that fetal exposure to some antiepileptic drugs might affect normal cerebral lateralization.

Dr. Meador noted that there is already a warning in the labeling for valproate about malformations and reduced IQ, and current guidelines from the American Association of Neurology also recommend against use in pregnancy.

[W]ith all the evidence we have that if women of child bearing potential are prescribed this drug without warning of this effect on the baby then doctors are leaving themselves open to a lawsuit. Dr. Kimford J. Meador

"But there are still a lot of women on valproate," he said. "Its use has decreased in tertiary epilepsy centers but it has not dramatically altered in general yet. It is still used extensively by primary care doctors and psychiatrists. There needs to be a lot more education to clinicians and the general public on this."

He pointed out that cognitive loss may occur at lower doses than those associated with malformations. "I have had patients who thought they were home and dry as the baby had no malformations, but then the child is discovered to have severe language delay."

"I do believe that now with all the evidence we have that if women of child-bearing potential are prescribed this drug without warning of this effect on the baby then doctors are leaving themselves open to a lawsuit," he declared.

Increase in IQ With Periconceptional Folate

Another interesting observation in the study was that patients who took folic acid around the time of conception had children with higher IQ rates than those who didn't.

Table 2. Mean IQs at Age 6 in Children Exposed to Periconceptional Folate vs Those Not Exposed

Group Mean IQ (95% Confidence Interval)
Exposed to folate 108 (106 - 111)
Not exposed to folate 101 (98 - 104)

P for difference = .0009.

"We saw about a 7-point increase in IQ with periconceptional folate, which appeared to be dose dependent," Dr. Meador said. "That is a substantial effect."

Noting that other studies have shown similar results in nonepileptic populations, he added, "We already recommend folate to reduce malformation risk but now it appears that it increases cognition as well. This is an additional reason to take folic acid if thinking about becoming pregnant."

The positive effect on IQ of folate was limited to periconceptional use. This effect wasn't seen if folate was started later in pregnancy, he noted.

"Not Evidence Based"

In an accompanying commentary, however, Dick Lindhout, MD, PhD, from the Department of Medical Genetics at the University Medical Center Utrecht, the Netherlands, warns that the folate finding regarding IQ should be regarded as preliminary.

He notes that although low-dose periconceptional folate (0.1 to 0.4 mg per day) is commonly recommended to the general population, periconceptional high-dose folic acid supplementation (4 to 5 mg per day) "as an antidote for teratogenic risks of antiepileptic drug is not evidence-based and potentially hazardous."

He calls for randomized trials comparing low-dose with high-dose folic acid in women receiving antiepileptic drugs who want to have children.

"Population-based prospective pregnancy registries and preconception outpatient clinics with focus on women on antiepileptic drugs might provide the best opportunity to do such studies including long-term neurodevelopmental outcome," Dr. Lindhout concludes.

The study was supported by the National Institutes of Health and the Epilepsy Research Foundation. Dr. Meador has received research support from GlaxoSmithKline, Eisai Medical Research, Myriad Pharmaceuticals, Marinus Pharmaceuticals, NeuroPace, Pfizer, SAM Technology, Schwartz Biosciences, and UCB Pharma; received salary support to Emory University from the Epilepsy Consortium for consultant work for NeuroPace, Novartis, Upsher-Smith, and Vivus; served as a consultant for Eisai, GlaxoSmithKline, Johnson & Johnson (Ortho McNeil), Medtronics Spherics, and UCB Pharma (but the monies went to charity); and received travel support from sanofi-aventis. Dr. Lindhout has disclosed no relevant financial relationships.

Lancet Neurol. Published online January 23, 2013. Abstract Commentary