|My first photo... @ Wikipedia|
Anyhow, for those of you who that haven't read my previous discussions on the growing body of research suggesting that we should perhaps be particularly mindful about the use of certain antiepileptic medicines during pregnancy (see here and here), the crux of the 'association' being made is that there may be some elevated risk of offspring autism under certain circumstances.
An automatic response to hearing about this possible link (aside from asserting that 'correlation does not equal causation') might be to say something along the lines of 'ban them' during pregnancy, as per another medicine (see here) with a very unfortunate history. But here's the thing: anti-epileptics aren't just some routine medicine used as pain relief or to help with other grumbling health-related issues that can be readily substituted. They can be, and often are, lifesavers: be in no doubt that untreated epilepsy can kill. And whilst it sounds easy to say that we should be prohibiting their use during pregnancy, the question then arises about what alternative you might suggest to control maternal epilepsy? Even though I'm enthused about things like ketogenic diets (see here) for example, and their growing use in treating some types of epilepsy, such measures are not always a good alternative to medication nor potentially great news for the developing child either (see here**). So we have a quandary.
Going back to the Veiby paper, it followed a slightly different vein to the previous work which looked at 'diagnosed' childhood outcomes such as autism following maternal use of anti-epileptics. Instead inquiring about specific facets of behaviour and development of offspring at 18 months and 36 months of age based on maternal report.
Using data derived from the Norwegian Mother and Child Cohort Study (MoBa) which covered a considerable sized cohort of pregnant women (more details are reported in this paper by Per Magnus and colleagues*** open-access), several thousand children were followed and included for the current study. MoBa, by the way, was also the source of data for one of the big folic acid studies with autism in mind (see here) and other interesting work (see here).
Anyhow, based on the analysis of records, some 333 children were reported to be exposed to antiepileptics in-utero as a result of maternal epilepsy. These children, when compared with non-exposed children, showed an increased frequency of issues with things like motor skills at 18 months alongside greater scores based on maternal reports of autistic-like behaviours.
At 36 months of age, these issues were still present in the exposed group. The frequency percentages for autistic traits in exposed and non-exposed children at the two time points were interesting (3.5% vs. 0.9% at 18 months, and 6% vs. 1.5% at 36 months). Just in case you're thinking that there might have been an effect from mum's epilepsy on child development and not the medication, the authors also reported no similar increased risk of developmental issues in those children born to mothers with epilepsy but who were unmedicated during pregnancy. Likewise when dads were diagnosed with epilepsy.
It's hard not to say that the evidence is stacking up for some effect of such medication when it comes to autism as a diagnosis and as a collection of symptoms/traits bearing in mind that risk is risk not fact. The announcement that the US FDA has now issued a warning about using the antiepileptics in question during pregnancy (which includes valproate) for treating things like migraine is testament to how seriously this issues is being taken. For many people, this is probably going to be worrying news.
With my cold, objective science hat on I'd also reiterate the suggestions from my previous posts on this topic that we have an interesting model developing potentially illustrative of how autism, at least some autism, might come about. Yes, autism research already uses a valproate mouse model of autism (see here for example) illustrative of the potential teratogenicity of the drug. But further than that are the reports that valproate for example, might have some interesting epigenetic effects on the genome, as per it being "a powerful HDAC inhibitor"*** (HDAC = histone deacetylases); indeed what effects it might have on certain comorbidity too****. One has to wonder whether this might have some effect also.
To close, I will reiterate the sentiments of this blog about not providing medical or clinical advice. Anyone requiring further information about this association, is directed to their supervising medical practitioner for further advice. Don't mess with epilepsy.
Something musical to close.... The Smiths and Sheila Take A Bow and the news that Morrissey has pulled his autobiography (which I assume would have been quite an interesting read).
* Veiby G. et al. Exposure to antiepileptic drugs in utero and child development—A prospective population-based study. Epilepsia. 2013. doi: 10.1111/epi.12226
** Sussman D. et al. Effects of a ketogenic diet during pregnancy on embryonic growth in the mouse. BMC Pregnancy and Childbirth 2013, 13:109 doi:10.1186/1471-2393-13-109.
*** Haberland M. et al. The many roles of histone deacetylases in development and physiology: implications for disease and therapy. Nat Rev Genet. 2009 January; 10(1): 32–42.
**** Turgeon N. et al. HDAC1 and HDAC2 Restrain the Intestinal Inflammatory Response by Regulating Intestinal Epithelial Cell Differentiation. PLoS ONE. 2013. 8(9): e73785. doi:10.1371/journal.pone.0073785
Veiby G, Daltveit AK, Schjølberg S, Stoltenberg C, Oyen AS, Vollset SE, Engelsen BA, & Gilhus NE (2013). Exposure to antiepileptic drugs in utero and child development: A prospective population-based study. Epilepsia, 54 (8), 1462-72 PMID: 23865818