Exposure to Acrylamide and the Risk of Cardiovascular Diseases in the National Health and Nutrition Examination Survey 2003–2006

Yu Zhang,Mengmeng Huang,Pan Zhuang,Jingjing Jiao,Xinyu Chen,Jun Wang,Yongning Wu
DOI: https://doi.org/10.1016/j.envint.2018.04.047
IF: 11.8
2018-01-01
Environment International
Abstract:Background: Long-term exposure to acrylamide (AA) from diet sources may induce oxidative stress and chronic inflammation. However, the association between AA exposure and the prevalence of cardiovascular diseases (CVD) remains unclear. Objectives: We aimed to examine the association between blood exposure levels of AA biomarkers and the prevalence of main types of CVD in a general population of US adults. Methods: We analyzed the associations between AA hemoglobin biomarkers [ hemoglobin adducts of acrylamide (HbAA) and glycidamide (HbGA), sum of HbAA and HbGA (HbAA+HbGA), and ratio of HbGA to HbAA (HbGA: HbAA)] and self-reported diagnosis of CVD in 8290 adults (>= 20 years of age) from the National Health and Nutrition Examination Survey (NHANES) 2003-2006. Multivariable logistic regression models were employed for estimating the associations in three groups classified by the combination of smoking status and serum cotinine levels. Results: In people exposed to environmental tobacco smoke (n = 4670), HbGA, HbAA+HbGA, and HbGA: HbAA were significantly and inversely associated with the prevalence of total CVD (p < 0.0001, p = 0.0155, and p = 0.0014 for trend, respectively) after adjusting for various covariates. The odd ratios (ORs) for total CVD in the highest quartiles of HbGA, HbAA+ HbGA, and HbGA: HbAA were 0.311 [ 95% confidence interval (CI): 0.193-0.500], 0.664 (95% CI: 0.485-0.911), and 0.495 (95% CI: 0.326-0.752) when compared with the individual lowest quartiles. In active smokers (n = 2432), HbAA was positively associated with CVD risk (p = 0.0088 for trend), while HbGA: HbAA was inversely related to total CVD (p = 0.0137 for trend). However, no significant associations of any AA hemoglobin biomarker with total and individual CVD prevalence were observed in the nonsmoking group (n = 1188). Conclusions: AA hemoglobin biomarkers are significantly associated with CVD in the active smoking group and the group exposed to environmental tobacco smoke but not in the nonsmoking group. Further prospective studies should clarify the causal relationship between HbAA and HbGA and the prevalence of CVD.
What problem does this paper attempt to address?