Abstracts

Rationale: Despite the availability of a wide range of antiseizure medications with diverse mechanisms of action, none have been proven to prevent the occurrence of epilepsy in humans. Animal studies have, however, demonstrated that the angiotensin receptor blocker (ARB), losartan, can inhibit the development of spontaneous seizures in epilepsy models, an effect which may be due to its ability to block brain transformimg growth factor beta (TGF-β) signaling and normalize aquaporin-4 (AQP4) expression. Hypertension is a common cause of vascular lesions and increases the risk of epilepsy. A recent large population study in Germany reported a lower incidence of epilepsy with the use of ARBs compared to other classes of antihypertensive medications. Nevertheless, observational studies are susceptible to bias from residual unmeasured confounding, necessitating replication to confirm reported results. This study aimed to compare the risk of epilepsy between ARB and other antihypertensive classes in adult patients with a diagnosis of hypertension.

Methods: The present retrospective cohort study utilized data from a national health administrative database with a total 2,261,964 eligible beneficiaries aged ≥18 years diagnosed with primary hypertension (ICD-9 code 401 and ICD-10 code I10) and dispensed at least one outpatient prescription of ARB, angiotensin receptor enzyme inhibitor (ACEI), β-blocker (BB), or calcium channel blocker (CCB) from 2010 to 2017. Patients with a diagnosis of epilepsy, or dispensed antiepileptic drugs 12 months prior to or within 90 days after initiating the study drugs were excluded. Propensity scores generated from prespecified covariates were used to match patients who received ARBs with those who received ACEIs, BBs, CCBs, or a combination of these other antihypertensive classes. Cox regression analyses with robust sandwich covariance estimation were used to evaluate adjusted hazard ratios in matched cohorts. These analyses were repeated for the various sub-types of ARBs and patients with or without pre-existing stroke or cardiovascular diseases.

Results: The PS matched cohort groups had the following number of patients: ARB-ACEI = 619,858, ARB-BB = 619,828, and ARB-CCB = 601,002 patients. ARBs were associated with lower epilepsy incidence vs. ACEIs (HR: 0.75; 95%CI: 0.58-0.96), BBs (HR: 0.70; 95%CI: 0.54-0.90), CCBs (HR: 0.80; 95%CI: 0.61-1.04), and a combination of these antihypertensive classes (HR: 0.72; 95%CI: 0.56-0.95). The effect was most robust for losartan and among patients with no preexisting stroke or cardiovascular conditions. Note sample sizes for other ARBs and patients with preexisting stroke or cardiovascular conditions were small.

Conclusions: ARB use was associated with a lower incidence of epilepsy compared to other antihypertensive agents in adult hypertensive patients, particularly in those on losartan or with no preexisting stroke or cardiovascular disease. Future pharmacological studies including randomized clinical trials are needed to establish antiepileptogenic properties of angiotensin receptor blockers.

Funding: none