Abstract P500: Proton Pump Inhibitor Use and Incident Cardiovascular Disease in Postmenopausal Women, Results From the Women’s Health Initiative

Introduction: Evidence from in-vitro and animal studies suggests that proton pump inhibitor (PPI) use is associated with incident cardiovascular disease (CVD). Observational studies in humans have shown conflicting results and didn’t account for relevant confounders such as diet and physical activity or confounding by indication. Methods: This study included 148,452 postmenopausal women without history of CVD at enrollment into the Women’s Health Initiative (WHI) observational study and clinical trials in 1993-1998. The primary exposure was PPI use (no, yes) determined by medication inventory conducted at baseline. The main outcome was incident primary CVD defined as a composite endpoint of physician-adjudicated myocardial infarction, coronary revascularization, stroke, and CVD mortality. Follow-up time was to September 2010. Cox proportional hazards regression was used to estimate hazards ratios (HR) and 95% confidence intervals (CI) for the association of PPI use and CVD. All models were adjusted for baseline age and stratified according to WHI study component. Additional models were further adjusted for demographics (education, income), lifestyle (smoking, alcohol, diet quality, physical activity, sleep duration), and major CVD risk factors (body mass index, family history of CVD, treated hypercholesterolemia, treated diabetes, treated hypertension). Results: Baseline prevalence of PPI use was 1.91%. After a mean follow-up of 11.3 years, there were 10,944 CVD events. PPI use was associated with a 28% higher risk of CVD compared to non-use in the age adjusted model (HR: 1.28, 95%CI: 1.13-1.45). The association was attenuated but remained significant after adjusting for potential confounders and was materially unchanged when using propensity score adjustment to better address confounding by indication (Table). Conclusions: PPI use was associated with higher risk of incident CVD in postmenopausal women. Further investigations are warranted to confirm this finding and identify potential mechanisms.