First-Line Monotherapy for Focal Epilepsy: Levetiracetam or Lamotrigine
A single patient, newly diagnosed with focal epilepsy, choosing a first drug. Both options are genuinely guideline-endorsed — the disagreement is about which real-world constraint should decide it.
D.R., a 26-year-old man, delivers packages for a regional courier route he has driven for four years, work he genuinely likes and cannot easily replace with anything that pays as well without a license. Three weeks ago he had a witnessed focal-to-bilateral seizure in the courier depot's break room, preceded by what coworkers described as a blank stare and lip-smacking lasting under a minute before he stiffened and convulsed. He has a history of treated generalized anxiety and mild depression, stable for two years on sertraline 100mg daily with his primary care physician, and otherwise no significant medical history — no head injury, no febrile seizures in childhood, no family history of epilepsy that he knows of.
MRI showed a small area of left temporal encephalomalacia, likely old and incidental rather than causally clear, and EEG captured left temporal sharp waves during a brief episode of impaired awareness two days into monitoring — enough, together with the witnessed event, to establish a diagnosis of focal epilepsy rather than a single provoked seizure. State law requires a seizure-free interval, commonly six months, before his license can be reinstated, and he has already told the team directly that the clock on that interval is the only thing he is actually thinking about.
The real choice in front of him is between two drugs that, on paper, both belong on a short list of reasonable first agents for focal epilepsy. Levetiracetam requires no baseline hepatic dosing adjustment, has no clinically significant drug-drug interactions to reconcile against his sertraline, and can be titrated to a therapeutic dose within roughly a week. Lamotrigine works by a different mechanism — voltage-gated sodium channel blockade rather than SV2A binding — and its own titration schedule is deliberately slow, six to eight weeks to a typical maintenance dose, specifically to reduce the risk of a serious rash. SANAD-II (Marson et al., 2021), the largest trial comparing the two drugs head-to-head in focal epilepsy, was designed to test whether levetiracetam was non-inferior to lamotrigine on time to twelve-month remission. It failed to show that: the intention-to-treat hazard ratio of 1.18 crossed the non-inferiority margin without establishing lamotrigine's superiority outright, though the per-protocol analysis did favor lamotrigine (HR 1.32). On the secondary endpoint of time to treatment failure, levetiracetam was significantly inferior. The trial's own conclusion was that lamotrigine should remain first-line — a recommendation resting on a failed non-inferiority claim plus a tolerability gap, not on a clean primary-endpoint win.
In clinic, choosing the first drug
Lamotrigine first. SANAD-II (Marson et al., 2021) set out to show levetiracetam was non-inferior to lamotrigine in newly diagnosed focal epilepsy and could not do it — and on time to treatment failure levetiracetam was clearly inferior. I want to be precise about what that is and isn't: the intention-to-treat primary analysis didn't crown lamotrigine the winner, it failed to clear levetiracetam. But the trial's own authors concluded lamotrigine should stay first-line, and when the drug with the better tolerability record is also the one the challenger couldn't match, the titration schedule is the weaker argument to let override that.
The six-to-eight-week ramp is real, and I understand exactly why it matters to him right now. But a seizure during an inadequately controlled first month on either drug resets his eligibility clock the same way regardless of which drug caused the gap — so the actual question is which drug gets him to durable control fastest, not which drug reaches a full dose fastest.
I follow the trial data and I'm not disputing SANAD-II's result. But he has told us directly that the six-month clock is what he's optimizing for, and levetiracetam gets him to a therapeutic, seizure-suppressing dose in about a week rather than two months.
Calling the titration schedule the weaker argument only holds if a breakthrough seizure during an under-dosed lamotrigine ramp is actually less costly than one on a faster-acting drug — for a patient whose eligibility clock resets on any seizure regardless of which drug he was on, that's not obviously true.
Levetiracetam also has nothing to reconcile against his sertraline — no enzyme interaction, no dose adjustment either direction. That's not a small convenience for a patient already managing one chronic prescription he's stable on.
You're both right about different halves of this, and I don't think the answer is picking one drug and ignoring the other's real cost. Levetiracetam's psychiatric and behavioral adverse-effect profile — irritability, mood lability, and less commonly frank depressive symptoms — is well documented and dose-related, and it lands on a patient who is already vulnerable on exactly that axis. That's not a reason to rule it out reflexively, but it is a reason not to treat the two drugs as interchangeable once you set the trial data aside.
There's a version of this that doesn't force a choice between SANAD-II's evidence and his actual timeline. Start lamotrigine on the standard slow titration — the trial data are what they are — and see him back at two and four weeks instead of the usual six, specifically to catch any early breakthrough seizure or intolerance before it costs him a real month. If levetiracetam still ends up necessary later, we watch his mood and sertraline response deliberately rather than starting a drug with a real psychiatric signal on a patient with anxiety and depression in the record and hoping nothing shows up.
Agreed: lamotrigine started on the standard slow titration, with clinic follow-up moved to two and four weeks rather than the usual six specifically to shorten how long an early problem could go unnoticed. D.R. was told directly why the faster drug wasn't chosen, including the psychiatric-history reasoning, since it's his eligibility clock being weighed against his own mental health history and he asked to understand the trade explicitly.
Not agreed, and left open rather than resolved: whether levetiracetam becomes the fallback if lamotrigine's titration produces a breakthrough seizure in week four or five, or whether a different agent entirely would be the more honest next step at that point rather than reaching for the same drug the psychiatric concern was raised about in the first place.