It works pretty well for some people

The exact mechanism of action of VPA remains unclear. Some data suggest that the drug may be a gamma-aminobutyric acid (GABA) agonist, acting directly or indirectly by enhancing GABA receptor activity and/or inhibiting GABA metabolism. The drug may also facilitate chloride influx at the neuronal level.

If you want a more technical description of how it (may) work:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2656326/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2757443/

Thanks. I've taken interest in pharmacodynamics since Martin Shkreli's arrest.

Yes, Lamictal does that.

Sodium channels are responsible for causing nerve impulses. During a seizure, these nerve impulses occur in a sporadic and uncontrolled fashion which leads to a seizure. By either blocking or restricting sodium channels, you are essentially controlling the flow of impulses being sent to the neurons.

Obviously, my explanation is a gross oversimplification. But that's how it was explained to me in layman's terms.

Afaik, no one really knows the full answer.

CT's are very good at finding masses, or at least in finding ones that are symptomatic. In a general sense, if you have a brain mass large enough to cause either headaches or seizures, it'll show up easily on CT. You'd only need an MRI if you either had a positive finding on CT that they want to look at more closely, or in the case of seizures repeated ones without an explanation.

In passing, I think it's interesting (and somewhat hospital/system dependent ) as to workups. We'll generally not start any anti-seizure meds in a patient with a normal CT and even a vague reason (not enough sleep, vagal episode, alcohol). They get sent for a follow up EEG and no driving until cleared by neurologist, but no meds. Apparently in other places, this isn't so.

MM MD