AUTHOR=Chen Tuotuo , He Haiqing , Tang Wei , Liu Ziyi , Zhang Hongliang TITLE=Association of blood trihalomethane concentrations with diabetes mellitus in older adults in the US: a cross-sectional study of NHANES 2013–2018 JOURNAL=Frontiers in Endocrinology VOLUME=15 YEAR=2024 URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1401131 DOI=10.3389/fendo.2024.1401131 ISSN=1664-2392 ABSTRACT=Background

Previous studies have demonstrated that there is a correlation between trihalomethanes and disease progression, such as allergic diseases. As we know, only few studies focused on the relationship between trihalomethanes and metabolic diseases, such as diabetes mellitus.

Objective

The aim of this study was to further explore the associations between blood trihalomethane concentrations and diabetes mellitus in older adults in the US.

Methods

Data were collected from the National Health and Nutrition Examination Study (NHANES) database in the survey cycle during 2013 to 2018, including 2,511 older adults in the US whose blood trihalomethane concentrations were measured, involving chloroform (TCM) and brominated trihalomethanes (Br-THMs). Br-THMs include bromodichloromethane (BDCM), dibromochloromethane (DBCM), and bromoform (TBM). Meanwhile, the concentration of total trihalomethanes (TTHMs) was also measured later. A multivariate logistic regression and restricted cubic spline were used to examine the relationship between blood THMs and diabetes mellitus. Meanwhile, we performed a subgroup analysis, which aims to explore the stability of this relationship in different subgroups. In order to further consider the impact of various disinfection by-products on diabetes, we also used weighted quantile sum (WQS). To explore the correlation in trihalomethanes, we plot a correlation heatmap.

Results

Adjusting for potential confounders, we found that there was a significant negative association between chloroform and diabetes mellitus [Model 1 (adjusted for covariates including age, sex, and race, OR = 0.71; 95% CI: 0.50–1.02; p = 0.068; p for trend = 0.094); Model 2 (adjusted for all covariates, OR = 0.68; 95% CI: 0.48–0.96; p = 0.029; p for trend = 0.061)]. In the bromodichloromethane, we reached a conclusion that is similar to TCM [Model 1 (adjusted for covariates including age, sex, and race, OR = 0.54; 95% CI: 0.35–0.82; p = 0.005; p for trend = 0.002); Model 2 (adjusted for all covariates, OR = 0.54; 95% CI: 0.35–0.82; p = 0.003; p for trend = 0.002)]. Meanwhile, the restricted cubic spline curve also further confirms this result (p overall = 0.0027; p overall< 0.001). Based on the analysis in the subgroups, we found that the value p for interaction in the majority of subgroups is higher than 0.1. Trihalomethanes and diabetes were inversely associated, and in the WQS, chloroform and bromodichloromethane were found to be the major contributors to this relationship. In the correlation analysis, we found that most trihalomethanes have a weak correlation, except for TBM and TCM with a strong correlation.

Conclusion

Our results in this study showed that blood chloroform, bromodichloromethane concentrations, and diabetes mellitus in older adults in the US are negatively correlated, suggesting that chloroform and bromodichloromethane can be protective factors for diabetes.