Abstract P010: Dose-Response Meta-Analysis of Arsenic Exposure and Coronary Heart Disease Risk
Introduction: Inorganic arsenic is an established toxicant and carcinogen found in drinking water and in some foods. Exposure to high levels of arsenic in drinking water (>100 μg/L) has consistently been associated with cardiovascular disease risk. In recent years, evidence of an association between cardiovascular disease and drinking water arsenic levels that are most relevant to the U.S. general population (<100 μg/L) has been building. Several high-quality prospective cohort studies are now available. In 2013, the National Academy of Science recommended that the U.S. Environmental Protection Agency include coronary heart disease as a critical endpoint in its ongoing reevaluation of the arsenic risk assessment. One critical aspect of this assessment is the shape of the dose-response relationship across low (<50 μg/L) and moderate (50-150 μg/L) levels of arsenic.
Hypothesis: We examined the hypothesis that the positive relationship between arsenic exposure and coronary heart disease risk is linear when modeling arsenic concentrations as log2-transformed.
Methods: We conducted a systematic review to identify all epidemiologic studies of arsenic and coronary heart disease conducted in general, non-occupationally exposed populations. We excluded studies with fewer than three exposure categories and studies without arsenic measurements in urine or water. Urine and water measurements were converted into estimates of internal daily dose (μg/kg/day) as a common exposure metric. Arsenic internal daily dose was log2-transformed. We calculated pooled relative risks using a two-stage random-effects meta-analysis based on the mean arsenic dose within each category and the categorical relative risks. The shape of the dose-response was examined using restricted quadratic splines. Studies for incident (fatal and non-fatal) and fatal coronary heart disease were examined in separate models.
Results: We identified ten studies meeting our inclusion criteria, including four prospective, two cross-sectional, and four ecological studies. Across the two prospective studies of coronary heart disease mortality, the pooled relative risk of fatal coronary heart disease was 1.23 (95% CI: 1.10, 1.38) per doubling of arsenic dose (μg/kg/day). Combining the three prospective studies of incident coronary heart disease, the pooled relative risk of incident coronary heart disease was 1.12 (95% CI: 1.05, 1.17) per doubling of arsenic dose (μg/kg/day). We found no evidence of non-linearity.
Conclusions: In conclusion, this analysis strengthens the evidence of a positive association between arsenic exposure at low-moderate levels and the risk of coronary heart disease. Further, it provides timely evidence for the current U.S. EPA risk assessment regarding the shape of the dose-response relationship at levels in drinking water relevant to the U.S. general population.
Author Disclosures: K.A. Moon: None. S. Oberoi: None. A. Barchowsky: None. E. Guallar: None. K.E. Nachman: None. Y. Chen: None. K.A. James: None. A. Navas-Acien: None.
- © 2015 by American Heart Association, Inc.