Which antiseizure drug a person takes may alter outcomes on a DOAC

Design

Retrospective cohort emulating a target trial, TriNetX electronic health records, propensity matching

Population

9,529 adults with epilepsy starting a DOAC on antiseizure monotherapy; 57% on levetiracetam

Primary outcome

Thromboembolic composite, major bleeding and all-cause mortality

Effect

Levetiracetam vs lamotrigine/lacosamide: thromboembolism HR 1.98 (95% CI 1.37 to 2.87), mortality HR 1.60 (1.23 to 2.08)

A target trial emulation used the TriNetX network of 165 health organisations to study 9,529 adults with epilepsy who started a direct oral anticoagulant while on stable antiseizure monotherapy. Each drug group was compared with a pooled lamotrigine or lacosamide reference, after propensity matching. Participants with recent vitamin K antagonist use or strong interacting drugs were excluded.

Levetiracetam, taken by 57%, was associated with higher thromboembolic risk (HR 1.98, 95% CI 1.37 to 2.87) and mortality (HR 1.60, 1.23 to 2.08), with similar major bleeding. Strong enzyme-inducing drugs such as carbamazepine and phenytoin were linked to more thromboembolism (HR 1.55, 1.02 to 2.36) but less major bleeding (HR 0.62, 0.46 to 0.83). Valproate was associated with higher mortality (HR 1.49) and intracranial bleeding (HR 3.01, 1.45 to 6.26). About 28% of thromboembolic events were judged potentially preventable under the reference drugs. In people on vitamin K antagonists the thromboembolic risks were near null.

The cohort is observational. Patients given levetiracetam may be sicker or older, and the abstract does not say whether anticoagulant levels were measured. It supports a closer look at interactions, not switching a seizure-controlled patient's drug on this basis.

• Review the antiseizure drug whenever a DOAC is started or being considered in epilepsy
• Strong enzyme inducers such as carbamazepine and phenytoin can lower DOAC levels; check the product information
• Do not switch a seizure-controlled patient solely on this study; discuss with the epilepsy team
• Document the reasoning when combining valproate with a DOAC, given the intracranial bleeding signal
• Monitor for thromboembolic events and bleeding in the first months of combined therapy

Why it matters

It suggests that the seizure drug, usually chosen without regard to anticoagulation, may be a modifiable contributor to clot and bleeding outcomes.

Don't overread it

Observational findings cannot show that changing drug would lower thromboembolic risk, and seizure control comes first.

The statistics, in plain English

A hazard ratio of 1.98 for levetiracetam means roughly double the rate of thromboembolic events compared with the lamotrigine or lacosamide group over follow-up. Propensity matching balances recorded characteristics, but confounding by indication, where sicker patients are given certain drugs, is a persistent risk in this kind of study.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.