Abstract P330: Comparative Effectiveness of Atenolol and Metoprolol for Prevention of CVD Events in Hypertensive Patients

Abstract

OBJECTIVE: Randomized controlled trials have demonstrated the efficacy of selected beta blockers for preventing cardiovascular (CV) events in patients following myocardial infarction (MI) or with heart failure (HF). However, the effectiveness of atenolol for preventing CV in patients with hypertension (HTN) has been questioned recently. METHODS: Using data from Cardiovascular Research Network HTN registry, we compared incident MI, HF, and stroke in 21,960 patients who were new beta blocker users between 2000-2009. Patients had no prior history of CV disease and had not previously filled a prescription for either atenolol or metoprolol tartrate. Exposure to either atenolol or metoprolol tartrate was ascertained from pharmacy data. In order to control for confounding by indication, patients were matched 1:1 based on the propensity to have metoprolol dispensed. The propensity score model included age, sex, race/ethnicity, year of beta blocker initiation, number of visits in the previous year, insurance payer, systolic blood pressure at baseline, and prevalence of lipid disorders, diabetes, chronic kidney disease, and whether the patient was on additional anti-HTN medication at baseline. Incident CV events, patient and provider characteristics were determined from electronic heath record , health plan enrollment and claims data. Cox proportional hazards regression was used to examine the associations of different beta blockers with incident CV events. RESULTS: The propensity-matched 10,980 new metoprolol tartrate users were similar to the 10,980 new atenolol users with regard to demographic and clinical characteristics. During follow-up (median 3.9 years), there were 743 incident MI, 806 incident HF, and 749 incident stroke events. Hazard rate ratios for MI, HF and stroke were 1.00 (95% confidence interval 0.96-1.02), 0.98 (95% CI 0.96-1.02), and 0.99 (95% CI 0.96-1.02), respectively. The results from multivariate models adjusted for the variables in the propensity model were similar. CONCLUSIONS: There were no statistically significant differences in incident MI, HF, stroke between atenolol and metoprolol tartrate users. These results do not support recent concerns that atenolol may not be as safe as other beta-blockers for preventing CV morbidity in patients with HTN. Large registries similar to the one used in this analysis may be useful for addressing comparative effectiveness questions that are unlikely to be resolved by randomized trials.