Association between brominated flame retardants (PBDEs and PBB153) exposure and hypertension in U.S. adults: results from NHANES 2005-2016

Abstract

Background: Brominated Flame Retardants (BFRs) have attracted widespread concern due to their environmental persistence and potential toxicity. This study aims to examine the association between BFRs exposure and hypertension.

Methods: We used data from the National Health and Nutrition Examination Survey (NHANES) spanning 2005 to 2016 for the cross-sectional analysis. To evaluate the individual and combined impacts of BFRs exposure on hypertension, we utilized multivariate models, including generalized additive models, weighted quantile sum (WQS) regression, and Bayesian kernel machine regression (BKMR) models.

Results: 9882 individuals (48% male) aged ≥ 20 were included in the final analysis, of whom 4114 had hypertension. After controlling for potential covariates, higher serum concentrations of PBDE100 (OR: 1.26; 95% CI: 1.01, 1.57) and PBDE153 (OR: 1.50; 95% CI: 1.18, 1.88) were significantly associated with hypertension. A nonlinear relationship between PBDE28 and hypertension was observed (P = 0.03). Moreover, BFRs mixture were positively associated with the prevalence of hypertension in both the WQS (β:1.09; 95% CI: 1.02, 1.17; P = 0.02) and BKMR models.

Conclusion: Our study suggested that BFRs exposure is positively associated with hypertension in the general population. To confirm this association and elucidate the mechanisms, further research is required.

Keywords: Bayesian kernel machine regression; Brominated flame retardants; Hypertension; NHANES; Weighted quantile sum regression.

Fig. 1

Flow chart of study participants. NHANES, National Health and Nutrition Examination Survey; BFRs, brominated flame retardants.

Fig. 2

Cubic splines for the associations of log-transformed serum BFRs with Hypertension. The horizontal dashed line represents the OR = 1.00. The red lines indicate multivariate-adjusted OR and the purple shaded area represents the 95%CI. All models were adjusted for age, gender, race, education level, PIR, serum cotinine, alcohol consumption, sleep disorders, depression, physical activity, dietary sodium intake, dietary potassium intake, BMI, eGFR, history of diabetes and NHANES cycles. lg, log-transformed; OR, odd ratio; CI, confidence interval

Fig. 3

Association between BFRs exposure and hypertension by WQS model. (A) The combined effects of mixed exposure to BFRs. (B) The weights of each BFR for hypertension in positive direction. The dashed red line represents the cutoff value (by default equal to the inverse of the number of elements in the mixture). The model was adjusted for age, gender, race, education level, PIR, serum cotinine, alcohol consumption, sleep disorders, depression, physical activity, dietary sodium intake, dietary potassium intake, BMI, eGFR, history of diabetes and NHANES cycles. WQS, weighted quantile sum; lg, log-transformed

Fig. 4

Association between combined BFRs exposure and hypertension analyzed by BKMR model. (A) Overall effects of BFRs mixture on hypertension at all concentrations ranged from the first quantile (10%) to the third quantile (90%) relative to the median (50%) level. (B) Posterior inclusion probabilities (PIPs) of each BFR for hypertension. Adjusted for age, gender, race, education level, PIR, serum cotinine, alcohol consumption, sleep disorders, depression, physical activity, dietary sodium intake, dietary potassium intake, BMI, eGFR, history of diabetes and NHANES cycles