Rooney, Mary2019-08-202019-08-202019-05https://hdl.handle.net/11299/206397University of Minnesota Ph.D. dissertation. May 2019. Major: Epidemiology. Advisors: Pamela Lutsey, James Pankow. 1 computer file (PDF); xiii, 145 pages.Low circulating magnesium (Mg) or hypomagnesemia is thought to be common, and is traditionally measured by circulating total Mg. Proton pump inhibitor (PPI) medication use is also common and has been linked with low circulating Mg. Both low circulating Mg and PPI use have been associated with elevated cardiovascular disease (CVD) risk. This dissertation further characterizes the complex relationship between circulating Mg and CVD among older adults. Using data from a double-blind pilot Mg supplementation randomized controlled trial, the first manuscript characterizes the interrelationship of different circulating Mg status biomarkers (ionized and total Mg) at baseline and in response to Mg supplementation. Baseline ionized and total Mg were modestly and positively associated. Mg supplementation versus placebo over 10 weeks resulted in increased concentrations of ionized and total Mg. In the second manuscript, we test cross-sectional associations of circulating total Mg with burden of atrial and ventricular arrhythmias as measured over 2 weeks on an ambulatory electrocardiographic monitoring patch in the Atherosclerosis Risk in Communities (ARIC) study. In this now elderly population, serum Mg was inversely associated with premature ventricular contraction burden. While effect estimates were in the hypothesized direction, we found little evidence of an association between circulating Mg and atrial arrhythmias. These findings were similar even among those without a history of CVD. The third manuscript explores cross-sectional associations of PPI use with circulating total Mg and prospective associations of PPI use, hypomagnesemia and CVD risk in the ARIC study. One in four participants had used a PPI within the last 2 weeks, and PPI users had a greater prevalence of hypomagnesemia than non-users. Additionally, PPI users had modestly elevated risk of CVD; however, presence of hypomagnesemia did not explain this elevated risk of CVD. Collectively, this dissertation helps refine our understanding of Mg homeostasis in relation to CVD.enarrhythmiascardiovascular diseasecirculating magnesiumepidemiologyproton pump inhibitorsThesis or Dissertation