Blood lead and cardiovascular mortality in never and ever smokers: a 20-year follow-up of US adults in NHANES 1999–2018

Low-level lead exposure has been associated with cardiovascular mortality, but tobacco smoke is a major source of lead, and it remains debated whether the association reflects residual confounding by smoking, operates through blood pressure, and persists as exposure declines. We followed 41,629 adults aged ≥20 years from the National Health and Nutrition Examination Survey (NHANES) 1999–2018 for mortality through 2019 by linkage to the National Death Index. During 418,656 person-years (median 9.9 years), 2,115 cardiovascular deaths occurred. In survey-weighted Cox models adjusted for sociodemographic factors, body mass index, alcohol intake, serum cotinine and smoking status, each doubling of blood lead was associated with a hazard ratio (HR) of 1.18 (95% confidence interval 1.11, 1.26) for cardiovascular and 1.23 (1.14, 1.32) for coronary heart disease mortality, but not with cerebrovascular or cancer mortality. The association was similar in never smokers (1.18; 1.07, 1.31), never smokers with serum cotinine ≤1 ng/mL (1.15; 1.02, 1.30), former smokers (1.17; 1.06, 1.30) and current smokers (1.16; 0.98, 1.37; P for interaction 0.84). It was not attenuated by adjustment for systolic blood pressure (1.19; 1.12, 1.27), blood cadmium or kidney function, and it did not weaken in cohorts recruited in 2009–2018, when geometric mean blood lead was 35% lower than in 1999–2008 (1.23 vs 1.16; P for interaction 0.56). Associations were present below 3.5 µg/dL. Blood lead at the low concentrations typical of contemporary US adults is associated with cardiovascular, particularly coronary, mortality independently of smoking and largely independently of blood pressure.