ORLANDO — Combining an enzyme-inducing antiseizure medication with a direct-acting oral anticoagulant (DOAC) does not significantly increase the risk of thromboembolic events in patients with epilepsy, preliminary results of a new study show.
These new data are important, “particularly when we’re talking about a more global perspective, given the vital role of enzyme-inducing antiseizure medications in epilepsy care across many middle- and low-income countries where they may be the only readily available treatment options,” said study investigator Emily K. Acton, PhD candidate in epidemiology and a medical student, University of Pennsylvania Perelman School of Medicine, Philadelphia, and University of Illinois College of Medicine, Chicago.
The findings also suggest that use of enzyme-inducing antiseizure medication with DOACs may be associated with a reduction in major bleeding events, although Ms. Acton stressed this requires more research.
The findings were presented at the American Epilepsy Society annual meeting.
Important Implications
Enzyme-inducing antiseizure medications may induce key drug metabolizing enzymes that result in wide-ranging interactions, Ms. Acton told this news organization. “But, in many cases, the clinical significance of these pharmacokinetic interactions is not completely understood.”
This has important implications for managing anticoagulation, said Ms. Acton. “The ease of DOAC use, and growing evidence of the drugs’ safety and efficacy compared to vitamin K antagonists, has led to widespread shifts in clinical practice towards DOACs.”
Due to the relative novelty of DOACs, their interaction profiles have been less than complete, she explained. Evidence that enzyme-inducing antiseizure medications may reduce absorption and accelerate metabolism of DOACs, potentially lowering DOAC levels and elevating thromboembolism risk, comes mainly from in vitro and animal studies.
“Research in humans is lacking and complicated in interpretation by inconsistent findings and methodological limitations,” she said.
The investigators wanted to address the “clinical uncertainty” surrounding the real-world relevance of enzyme-inducing antiseizure medications and DOAC interactions but conducting a randomized trial “would be neither feasible nor ethical,” said Ms. Acton.
Using healthcare claims data from October 2010 to September 2021, the researchers conducted an active comparator, new-user cohort study among a nationally representative sample of adults with epilepsy who had been co-prescribed these drugs.
They compared thromboembolic and major bleeding event rates between exposure to DOACs with enzyme-inducing antiseizure medications vs exposure to DOACs with non-enzyme inducing antiseizure medications.
Enzyme-inducing antiseizure medications included in the study were carbamazepine, oxcarbazepine, phenobarbital, phenytoin, primidone, and topiramate. Non-enzyme-inducing antiseizure medications included gabapentin, lacosamide, lamotrigine, levetiracetam, and pregabalin.
The researchers used data-adaptive high-dimensional propensity score matching to control for “hundreds and hundreds” of observed confounders, and proxies for unobserved confounders, said Ms. Acton. They identified outcomes based on validated diagnostic coding algorithms for thromboembolic and major bleeding events and estimated adjusted hazard ratios (aHRs) using Cox proportional hazard models with robust variance estimators to account for clustering within matched pairs.
Reduced Risk of Major Bleeding
Outcomes were analyzed in three separate cohorts. These included patients on DOACs for any indication (indication-agnostic); those on DOACs for atrial fibrillation (AF); and those taking DOACs for deep vein thrombus/pulmonary embolism (DVT/PE).
In the indication-agnostic analysis, the investigators examined thromboembolic events among 5989 episodes in patients taking both DOACs and enzyme-inducing antiseizure medications, compared witha reference group of 14,671 episodes in patients taking DOACs and non-enzyme-inducing antiseizure medications.
The reference group was generally older and had a greater prevalence of a number of major comorbidities compared with the exposed group, noted Ms. Acton.
For the indication-agnostic analysis, the aHR was 1.11 (95% CI 0.89-1.39). Results were similar for the AF indication (aHR 1.10; 95% CI 0.82-1.46) and for the DVT/PE indication (aHR 1.11; 95% CI 0.81-1.51).
“This research provides large-scale, real-world evidence enzyme-inducing antiseizure medication use alongside DOACs does not significantly elevate risk of thromboembolic events among a nationally representative epilepsy population,” said Ms. Acton.
However, “it’s always important to consider risk factors for thromboembolic and bleeding events at the level of the individual patient,” she added.
With respect to major bleeding events, there was a slightly reduced risk in the exposed group, specifically in the analysis of subjects with atrial fibrillation, where the aHR was 0.63 (95% CI 0.44-0.89).
“A potential explanation may be pharmacokinetic interaction with enzyme-inducing antiseizure medications occurring to a degree that lowers DOAC levels without necessarily negating therapeutic effects,” said Ms. Acton.
However, she cautioned that more research is needed.
As for the differential potency among the various enzyme-inducing antiseizure medications studied, Ms. Acton said results from a secondary analysis in the atrial fibrillation assessment that removed the potentially less potent enzyme inducers, oxcarbazepine and topiramate, didn’t significantly change the study results.