A drug or medication is any substance taken by a human with the aim of altering physiologic function.

The distinction between “recreational drugs” and medications is entirely cultural and in most cases counterproductive to meaningful discussions. Many of the antiseizure drugs (ASDs) are controlled substances by law in the United States and in many other countries. This is supposed to mean that they have some potential for dependence or “abuse.” Most strikingly, the drug cannabidiol (CBD), a prominent constituent of the plants Cannabis sativa and Cannabis indica is in one sense completely outlawed in the US (under the prohibition of said plants and their extracts containing cannabinoids) and in another sense explicitly indicated for the treatment of seizures in severe epilepsies and characterized by a potential for abuse leading to “moderate or low physical dependence or high psychological dependence” (under schedule IV). This is one of several examples of oddly chemistry-naïve double scheduling.



Valproate—also known as valproic acid, divalproex sodium, and sodium valproate—is a simple branched-chain carboxylic acid first manufactured in the C19 but whose antiseizure properties were discovered by accident in the 1970s by Pierre Eymard who used the lipophilic liquid solvent as a vehicle for several chemicals in experiments with khelline derivatives.1

It was subsequently approved for use by the US FDA in the late 1970s.

Mechanisms of Action

Valproate works in several ways. It increases GABA by induction of glutamic acid decarboxylase. It has Nav blocking actions at high doses although these may not be pharmacologically feasible. And it is thought to affect T-type Ca channels such as those involved in absence epilepsy.2

Adverse Effects

Common adverse effects are those of most other ASDs.

VPA often causes asymptomatic rises in ALT and sometimes ALP less than 3 times the upper limit of normal. In rare cases it can cause enduring hepatocellular damage and liver failure.

Pancreatitis is uncommon but has been attributed to VPA.

Of the blood cells, platelets are not uncommonly depressed. It is important when interpreting thrombocytopenia in VPA-taking patients to take into account that platelet function is also reduced by the medication. In some cases, severe thrombocytopenia will force discontinuation of the drug.

VPA can cause a polycystic ovarian syndrome–like syndrome in women: weight gain, androgenic hair growth, and oligomenorrhea.

Weight gain is a problem for all sexes, as is hair loss.

The teratogenicity of VPA is one of its biggest downsides. The North American AED Pregnancy Registry (NAAEDPR) estimates the rate of major congenital malformations in women taking VPA to be 9% or so—nearly five times the rate of women using levetiracetam. (Underlying rate without ASDs is around 1.5–2.0%.) Other papers have described dose-response curves, with higher rates of birth defects in women taking higher doses.

Deferring VPA therapy during the first and second trimesters, which can at least mitigate some of the worst birth defects (to wit, neural tube defects like spina bifida or the dreaded anencephaly) still has consequences. Use during the third trimester is still associated with an increased risk for autism spectrum disorders and developmental delay.

For these reasons, VPA is almost completely contraindicated in patients of child-bearing potential. The sole exceptions are for cases of PGE or LGS not responsive to other medications.


Valproate is gold-standard therapy for most types of primary generalized epilepsy (PGE) and many types of secondary generalized epilepsy, including Lennox-Gastaut Syndrome (LGS).

The AE profile generally retards its use in PGE unless patients have failed to respond to several other appropriate meds. The teratogenicity of valproate makes it among the only medications one would make every effort to extirpate during pregnancy, even in controlled patients.