Antiseizure Medication Choice Alters Stroke and Bleeding Risk in Epilepsy Patients on DOACs
核心洞察
A target trial emulation of 9,529 adults with epilepsy (搜索) found that antiseizure medication choice significantly influences thromboembolic, bleeding, and mortality outcomes during direct oral anticoagulant (DOAC) therapy.
Levetiracetam was linked to a 1.98 thromboembolic hazard ratio and a 60% increase in all-cause mortality, while strong enzyme inducers raised thromboembolic risk by 55% but cut major bleeding by 38%.
Valproate showed a threefold increase in major intracranial hemorrhage and a 49% rise in all-cause mortality, with risks near null among patients using vitamin K antagonists.
Antiseizure medications combined with direct oral anticoagulants (搜索) (DOACs) produce markedly different vascular and mortality outcomes in adults with epilepsy (搜索), according to a target trial emulation published in JAMA Neurology. The study, led by Schubert KM et al., evaluated 9,529 treatment initiators and found that individual antiseizure drug classes diverged substantially in thromboembolic events, major hemorrhage, and all-cause mortality compared with a low-interaction reference group receiving lamotrigine or lacosamide.
Using data from an international electronic health record network, researchers emulated a target trial among adults with epilepsy (搜索) receiving concurrent DOAC therapy. From 2.29 million adults with epilepsy, 40,932 were eligible for DOAC treatment, and 9,529 treatment initiators were included in the final analysis. The mean age was 63 years, and 51% were female.
Divergent Risk Profiles Across Drug Classes
Levetiracetam, despite being widely regarded as a low-interaction agent in routine practice, was associated with an elevated thromboembolic hazard ratio of 1.98 and a 60% increase in all-cause mortality, while displaying major bleeding rates comparable to the reference cohort.
Strong enzyme-inducing antiseizure medications, such as carbamazepine and phenytoin, exhibited a 55% increase in thromboembolic hazard alongside a 38% reduction in major bleeding risk. This inverse pattern aligns with enhanced cytochrome P450 and P-glycoprotein induction, which accelerates DOAC clearance, impairing antithrombotic efficacy while reducing systemic bleeding.
In contrast, valproate therapy demonstrated a threefold increase in major intracranial hemorrhage and a 49% rise in all-cause mortality, accompanied by a heightened risk of noncerebral embolic events. Moderate enzyme inducers showed no statistically significant differences in primary vascular or bleeding endpoints.
Preventable Events and Pharmacokinetic Mechanism
Counterfactual population modeling indicated that approximately 28% of observed thromboembolic events were potentially preventable if all patients had initiated the lamotrigine or lacosamide reference strategy. Notably, levetiracetam accounted for 71% of these excess ischemic events, while strong enzyme inducers contributed 16%.
In a sensitivity analysis of patients receiving vitamin K antagonists, thromboembolic and mortality hazards were near null across all drug groups, reinforcing that the observed risks represent drug-specific pharmacokinetic interactions with DOACs rather than intrinsic properties of the antiseizure medications themselves.
Clinical Implications for Patient Safety
These observational findings underscore that co-prescribing antiseizure medications and DOACs requires careful clinical risk assessment. Clinicians should consider prioritizing agents with low interaction potential, including lamotrigine and lacosamide, for patients requiring oral anticoagulation.
When clinical necessity dictates the use of higher-risk anticonvulsants, switching from DOACs to vitamin K antagonists with regular international normalized ratio monitoring may safeguard clinical outcomes. The authors concluded that antiseizure medication choice represents a potentially modifiable factor influencing outcomes in adults with epilepsy (搜索) receiving DOACs, highlighting the importance of treatment selection when balancing thromboembolic and bleeding risks.