For a recent evidence based paediatrics assignment we had to answer and present a clinical question. I’m sure you are well acquainted with the process; construct your question in standard PICO format, search your secondary and primary sources, critically appraise the evidence and draw your conclusions.

Having noted a trend towards starting lamotrigine rather than valproate in adolescent girls, because of the concerns of teratogenicity, and wondering if this is at the expense of good seizure control, my question was:“In adolescent girls with newly diagnosed generalised epilepsy (population) is lamotrigine (intervention) as effective as sodium valproate (comparison) at achieving seizure control (outcome)?”

Diligently I went through the process, starting with the NICE guidelines on the epilepsies. Their recommendation – valproate should be first line treatment for generalised tonic clonic seizures; but just bear in mind, and discuss with your female patients, the fact it is teratogenic. There is a note to say the guidelines are being currently being reviewed because MHRA have “strengthened its warnings” on prescribing valproate to women of childbearing potential – suggesting we’re a little more concerned about valproate than we used to be.

Noting that the evidence base for the NICE guidelines did not seem to contain anything comparing the two drugs in children or young people (how unusual!) I moved on. Cochrane very helpfully have a meta-analysis that will exactly answer my question – when it is completed next year. I then hunted through MEDLINE and EMBASE with as many keywords and MESH terms as I could think of. Having discarded a lot of irrelevant articles and gone through the remaining vaguely relevant ones, I’d decided it all pointed to valproate being better at controlling seizures but with more side effects.

Just for completeness (and because someone was bound to ask me), I thought the night before presenting I’d look up that MHRA alert for myself. Remember the NICE website said they’d strengthened their warning? In fact it didactically says to not prescribe valproate to any female infant, child or woman of childbearing potential until you’ve tried pretty much everything else.

Maybe you were already well aware of this. I unfortunately was not, and so frustratingly went through the whole process of trying to find an answer, only to find it had been made null & void by the regulatory authority.

So what have I learnt? Well clearly I’ll only be prescribing valproate to boys in the future. And I’m going to at least skim read all those clinical newsletters that appear in my inbox, which might tell me about important MHRA alerts. And if NICE (or anyone else) points me towards an alert, I’ll go and read it for myself, rather than relying on their interpretation…
– Rishwa Vithlani