Prenatal Valproate Exposure Linked to Asymmetric Brain

Kate Johnson

June 27, 2013

MONTREAL, Quebec, Canada — Prenatal exposure to sodium valproate results in asymmetric brain development that can be seen at high-resolution T1-weighted MRI, researchers reported for the first time here.

The findings provide a structural explanation for previously described language and intellectual impairments in children whose mothers took the drug during pregnancy and also suggest a dose-response effect, said Amanda Wood, PhD, who is currently at the University of Birmingham, United Kingdom, but carried out the work at Monash University in Clayton, Australia.

"Certainly these data suggest a little caution is warranted in those women for whom valproate remains the only option," said Dr. Wood, who reported the findings at the 30th International Epilepsy Congress (IEC).

But the findings should be interpreted carefully, cautioned Graeme Jackson, MD, professor of neurology at Melbourne University's Florey Institute of Neuroscience and Mental Health in Australia when asked by Medscape Medical News for comment.

"We don't want to demonize valproate because it's a good drug and some women have to stay on it because they would probably terminate the pregnancy because of seizures otherwise," he said in an interview.

"In general the developmental abnormalities are at the higher dose of the drug, and that's confirmed in this study," he added. "These language deficits are a minor inefficiency — it doesn't mean someone goes from normal intelligence to intellectual disability. The study clearly says we should do a lot to minimize the use of valproate in women of childbearing age — and that message is largely already out there — but we don't want to demonize the drug for those women who really have to take it."

Relatively High Cognitive Function

The study included 16 children, mean age 7.2 years, who had been exposed to a "relatively moderate" mean dose of 885 mg of valproate per day across the entire pregnancy, said Dr. Wood.

"In terms of the cognitive characteristics of this group they were relatively high-functioning given what we know," she explained, reporting their mean Full Scale Intelligence Quotient score as 96.3, with 5 participants falling into the "low average" but not the intellectual disability range.

High-resolution T1-weighted MRI was used to look at participants' cortical structure compared with that of 16 age-matched controls.

"I was reasonably surprised when the analysis came through," Dr. Wood commented. "What we found was an increased cortical thickness in the left frontal operculum of valproate-exposed individuals compared to controls after correction for multiple comparisons (P < 0.05). We know in controls there is actually greater cortical thickness in the right rather than left frontal region. By contrast the valproate-exposed children did not show that pattern, so there seems to be an aberration of cortical thickness symmetry and that's where we end up with the increase in cortical thickness in the left front operculum associated with sodium valproate."

Comparing cortical thickness with language-specific impairments, Dr. Wood's study showed a nonsignificant trend (P = 0.12) toward correlation between increased left frontal cortical thickness in valproate-exposed children and poorer performance on the Verbal Comprehension Index on the Wechsler Intelligence Scale for Children.

"Although I will be the first to acknowledge this is not a significant result it certainly strongly suggests that there is an association of some sort between the cortical thickness in that left frontal region and which is abnormally increased and performance on one measure of verbal ability, suggesting that the greater the increase in cortical thickness the worse your performance is," she said.

Previous work by Dr. Wood's group has suggested a dose-dependent negative effect of prenatal valproate exposure on verbal intellectual abilities and working memory (J Int Neuropsychol Soc. 2011;17:133-142 and Neurology. 2011;76:719-726).

"I don't think this [current] study negates the earlier language processing impairments shown in our Neurology paper but the most severely affected children simply didn't participate in the imaging substudy," Dr. Wood told Medscape Medical News. "So I suspect the lack of statistical significance in the verbal/thickness association is because we are underpowered to detect that effect."

Looking at dose effects, the study did not find a dose-dependent impact on left cortical thickness; rather, there was a significant association between mean dose of valproate across the pregnancy and right cortical thickness, she said.

"The right frontal dose correlation is intriguing," she commented in the interview. "At this point it's hard to speculate on the basis for that association and we really will need to examine these issues in a group where we can more reliably separate out high from low dose. That is something we are doing in a new study but it is early days yet."

Pruning of Left Hemisphere

Commenting on the study, Dr. Jackson said, "it looks like in controls there is pruning of the left hemisphere and in valproate that pruning does not happen. And we know that left frontal thickness correlates with verbal comprehension."

Dr. Jackson has himself examined the effects of valproate on cortical structure and recently reported a significantly reduced parietal lobe cortical thickness in patients with epilepsy taking the drug (Neurology. 2013;80:1895-1900).

He said Dr. Wood's imaging findings strengthen his own. "We always need to understand the mechanism, because once we know that, bright people can figure out how to fix it. If you don't know why it happens it's hard to fix it."

While Dr. Wood agreed with Dr. Jackson that valproate should not be demonized, she emphasized that some children are severely affected, with life-long consequences for them and their families.

"My primary concern is that we understand the mechanisms underlying long-term outcomes so that we enable women and their treating doctors to make informed decisions. It is very, very important that this message is made in a measured fashion; the consequences of women not taking their medications can be dire and I hope that we will assist women, not complicate their decisions any further," she said.

Dr. Wood disclosed funding provided by GlaxoSmithKline, Janssen, Novartis, Pfizer, sanofi-aventis, and UCB provided to the Australian Pregnancy Register for Women with Epilepsy and Allied Disorder. Dr. Wood and several colleagues have been or continue to serve as members of the executive or scientific advisory boards for the register.

30th International Epilepsy Congress (IEC). Abstract 010. Presented June 24, 2013.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.