Abstracts

Rationale: Studies suggest that enzyme-inducing antiepileptic drugs (EIAEDs) elevate serum lipids and other vascular risk markers. This would be expected to increase the risk for vascular disease. But since lipids are often measured routinely, hyperlipidemia might be unwittingly discovered and treated without recognition that the drug was responsible. We assessed the incidence of both new vascular disease and new prescription for HMG-CoA reductase inhibitors ( statins ) among patients who were newly prescribed antiepileptic drugs (AEDs) for seizures, particularly among those whose lipids were checked.Methods: We queried the MarketScan databases between July 2009 to January 2013, which included 66 million enrollees with commercial or Medicare Supplemental insurance. We identified individuals who had a new diagnosis of epilepsy or convulsions on 2 distinct claims, prescription drug coverage, continuous enrollment in the database from 6 months prior to 24 months after the epilepsy diagnosis, no utilization of an AED prior to that diagnosis, and at least 1 new AED prescription. Those prescribed an EIAED (phenytoin, carbamazepine, barbiturates) for 30 days were counted as inducer-exposed patients, while those prescribed other AEDs (NEIAED) were considered non-inducer exposed patients. Rates of vascular disease (cardio-, cerebro-, and peripheral) between 6 and 24 months after AED prescription were assessed in the two groups, as were the rates of new statin prescriptions beginning 30 days after AED treatment. These rates were also examined in the subgroup of patients in whom a lipid panel was obtained. Univariate and multivariate analyses (accounting for age and gender) were performed.Results: There were 1959 patients meeting the criteria who were EIAED-exposed and 9934 NEIAED-treated patients. The EIAED group was older (mean 49 vs. 38 years). Among patients with no prior vascular diagnosis, new vascular disease was observed in 65 of 1129 EIAED patients (5.8%), and in 345 of 9772 NEIAED patients (3.6%; p=0.0002). However, this difference was no longer significant after controlling for age and gender. Among patients who had a lipid panel ordered, a statin was newly prescribed in 125 of 816 (15.3%) EIAED-treated patients, and in 351 of 3559 (9.9%) NEIAED patients (p < 0.0001). This difference remained highly significant after accounting for age and gender (p=0.0008). A patient starting an EIAED was 47% more likely to be subsequently started on a statin than an NEIAED-treated patient (95% CI 1.17 - 1.83).Conclusions: Statins were more often newly-prescribed to those started on EIAEDs than to those started on NEIAEDs. This provides further evidence that EIAEDs elevate lipids in a clinically meaningful manner. The absence of significantly higher rates of vascular disease in EIAED patients might be due to routine lipid surveillance and statin treatment, masking the drugs atherogenic effects. The limited time horizon also may have reduced our ability to detect long-term vascular effects. EIAEDs appear to be preferentially prescribed to older patients in the U.S.