Comment on "Environmental Cadmium and Mortality from Influenza and Pneumonia in U.S. Adults".

Park et al. (2020) conducted a prospective study on the association between cadmium (Cd) exposure and subsequent mortality associated with influenza and pneumonia. The adjusted hazard ratio (HR) of participants with creatinine-corrected urinary Cd (U-Cd) in the 80th vs. 20th percentiles for influenza- or pneumonia-associated mortality was 1.15 (95% confidence interval: 1.05, 1.26), and the significance was maintained among never-smokers. In contrast, no significant association was reported between blood Cd levels and influenza- or pneumonia-associated mortality, which raises some questions about their study.

First, the level of Cd exposure was low, presenting a mean value of of creatinine in the urine, suggesting that participants lived in non-Cd–polluted areas (Chaumont et al. 2013). The background level of Cd exposure is important for considering the mechanism of association between Cd and influenza- or pneumonia-associated mortality. The effects of Cd on the kidney cannot be observed in this population, and the association with influenza is less likely to be related to Cd nephrotoxicity.

Second, I speculate that the discrepancy between the associations of U-Cd and blood Cd with influenza- or pneumonia-associated mortality is important. Huang et al. (2013) reported that mean U-Cd in men was significantly higher in current and former smokers than in never-smokers. Cd has a shorter half-life in blood than in urine, and blood Cd more likely reflects recent exposure, including smoking (Lauwerys et al. 1994; Adams and Newcomb 2014). Therefore, middle- to long-term Cd exposure can be monitored based on urine samples. I suggest that future studies of Cd and lung tissue should pay more attention to U-Cd than to blood Cd.

Third, Park et al. (2020) used Cox regression analysis to predict influenza- or pneumonia-associated mortality by Cd exposure. U-Cd concentrations varied widely by age, and there are different results of the adjusted HRs for mortality in each age class. Adjustment of sociodemographic and socioeconomic data, such as race, ethnicity, income, education, and occupational position, could provide insight into whether the association of Cd exposure with influenza and pneumonia might vary by age.

Although a statistically significant association was observed in this study using a sufficiently large sample, sensitivity analysis showed that the significance was observed in participants ages 65–84 y. Further studies are needed to elucidate age-related factors affecting the relationship between Cd exposure and influenza- or pneumonia-associated mortality.