Proton Pump Inhibitor Use, Hypomagnesemia and Risk of Cardiovascular Diseases The Atherosclerosis Risk in Communities (ARIC) Study

• Authors: Mary R Rooney, Univ Minnesota; Elizabeth J Bell, Optum; Alvaro Alonso, Emory University; James S Pankow, University of Minnesota, Minneapolis; Ryan T Demmer, University of Minnesota, Minneapolis; Kyle D Rudser, University of Minnesota, Minneapolis; Lin Y Chen, Univ Minnesota Med Sch; Pamela L Lutsey, University of Minnesota, Minneapolis
• Journal Title: JOURNAL OF CLINICAL GASTROENTEROLOGY
• Volume: Volume 55, Number 8
• Publisher: LIPPINCOTT WILLIAMS & WILKINS | 2021-09-01, Pages 677-683
• Type of Work: Article | Post-print: After Peer Review
• Publisher DOI: http://doi.org/10.1097/MCG.0000000000001420
• Subjects: Health Sciences, Epidemiology; Biology, Biostatistics
• Keywords: Science & Technology; Life Sciences & Biomedicine; Gastroenterology & Hepatology; proton pump inhibitors; circulating magnesium; cardiovascular disease; prospective; ELECTROLYTE DISTURBANCES; ATRIAL-FIBRILLATION; GENERAL-POPULATION; SERUM MAGNESIUM; ISCHEMIC-STROKE; METAANALYSIS; DESIGN; ADULTS

Abstract

Goals: The goal of this study was to evaluate whether proton pump inhibitor (PPI) use is cross-sectionally associated with hypomagnesemia and whether hypomagnesemia mediates the prospective association between PPIs and cardiovascular disease (CVD) risk. Background: Use of PPIs has been associated with hypomagnesemia, primarily in case reports or within insurance databases. Both PPI use and low serum magnesium (Mg) have been associated with modestly higher CVD risk. Yet, the interrelation between PPI use and Mg in relation to CVD risk is unclear. Study: The 4436 Atherosclerosis Risk in Communities participants without prevalent CVD at visit 5 (baseline, 2011-2013) were included. Multivariable relative risk regression was used for cross-sectional analyses between PPI and hypomagnesemia prevalence (≤0.75 mmol/L). Incident CVD (defined by atrial fibrillation, coronary heart disease, CVD mortality, heart failure, stroke) was identified through 2017. Multivariable Cox regression was used to examine the PPI-CVD association. Results: Participants were mean±SD aged 75±5 years; 63% were women, 23% Black, and 24% were PPI users. PPI users had 1.24-fold (95% confidence interval: 1.08-1.44) higher prevalence of hypomagnesemia than nonusers. Over a median 5 years of follow-up, 684 incident CVD events occurred. PPI users had higher CVD risk [hazard ratio (95% confidence interval) 1.31 (1.10-1.57)] than nonusers. The effect estimate was largely unchanged when hypomagnesemia was added to the model as a potential mediator. Conclusions: In this elderly community-based study, PPI users had a higher prevalence of hypomagnesemia than in nonusers. PPI users also had higher CVD risk than nonusers; however, it appears unlikely that hypomagnesemia explains associations of PPIs with CVD risk.