Abstracts

Rationale: Older-generation antiepileptic drugs (AEDs) with highly inducing activity of cytochrome P450 enzyme system may produce metabolic abnormalities and increase cardiovascular risk. However, the extent of increase in cardiovascular events among older AEDs users has rarely been explored and the results were still controversial. The objective of this study was to evaluate the risk of acute myocardial infarction and ischemic stroke in adult patients with epilepsy who newly initiate AEDs that highly induce cytochrome P450 activity (EI-AEDs) compared with those who initiate other AEDs. Methods: This was a retrospective cohort study of patients from the National Health Insurance Research Database in Taiwan who had been diagnosed as epilepsy and initiated AEDs between 2005 and 2010. A baseline period of 6 months was used to ensure no prior exposure of any AEDs and no hospitalization episode of acute myocardial infarction (AMI) or acute ischemic stroke (AIS). Patients were divided into cohorts of EI-AEDs (i.e. initiators of phenytoin, carbamazepine, phenobarbital) or other AEDs (i.e. valproic acid, gabapentin, oxcarbazepine, lamotrigine, levetiracetam, topiramte, tiagabine, and vigabatrin) according to the AEDs the initiated. The primary study outcome was combined cardiovascular (CV) events (i.e. AMI or AIS). Outcomes of AMI and AIS were also analyzed separately. We explored the incidence rate (per 1000 person-year) of combined CV events, AMI, and AIS in each cohort. A 1:1 propensity score (PS)-matched analysis was used to adjust potential confounding by indication. Cox-proportional hazard models were used to determine the odds ratios (ORs) and 95% confidence intervals (CIs) of EI AEDs for the increase of CV event risk. Results: The baseline demographics of patients were shown in Table 1. We identified 3,028 combined CV events in 732,180 unmatched patients and 2,102 combined CV events in 492,294 patients after PS-matching. For the unmatched and the 1:1 PS-matched patients, the incidences of combined CV events in cohorts of EI-AEDs and other AEDs were shown in Table 2. The OR of EI-AEDs for combined CV events was 1.95 (95% CI: 1.81-2.10) in the unmatched cohorts and 1.87 (95% CI: 1.71-2.04) in the PS-matched cohorts. For the PS-matched cohorts, the magnitude of increase for AIS among initiators of EI-AEDs (OR: 1.93; 95% CI: 1.76-2.12) was higher than that of increase for AMI (OR: 1.56; 95% CI: 1.25-1.94). Conclusions: In this exploratory analysis, there was a consistent and statistically significant effect of EI-AEDs on combined CV endpoints, as well as AMI and AIS as endpoints separately. The different magnitude of increase in AIS and AMI events may suggest the interaction of disease (epilepsy) itself and drug effects (metabolic derangement due to EI-AEDs).