Next Article in Journal
Toxicity of Per- and Polyfluoroalkyl Substances and Their Substitutes to Terrestrial and Aquatic Invertebrates—A Review
Next Article in Special Issue
The Effect of Selenium Against Cadmium-Induced Nephrotoxicity in Rats: The Role of the TRPM2 Channel
Previous Article in Journal
Short-Term Effect of Air Pollution on Daily Hospital Visits for Anxiety Disorders in Southern China with Low Pollution Concentrations
Previous Article in Special Issue
Per- and Polyfluoroalkyl Substances (PFAS) Affect Female Reproductive Health: Epidemiological Evidence and Underlying Mechanisms
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Toxics
Volume 13
Issue 1
10.3390/toxics13010046
toxics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Exposure to Volatile Organic Compounds in Relation to Visceral Adiposity Index and Lipid Accumulation Product Among U.S. Adults: NHANES 2011–2018

by
Ziyi Qian
1,
Chenxu Dai
1,
Siyan Chen
1,
Linjie Yang
1,2 and
Xia Huo
1,*
1
Laboratory of Environmental Medicine and Developmental Toxicology, Guangdong Key Laboratory of Environmental Pollution and Health, College of Environment and Climate, Jinan University, Guangzhou 511443, China
2
Department of Public Health and Preventive Medicine, School of Medicine, Jinan University, Guangzhou 510632, China
*
Author to whom correspondence should be addressed.
Toxics 2025, 13(1), 46; https://doi.org/10.3390/toxics13010046
Submission received: 3 November 2024 / Revised: 31 December 2024 / Accepted: 6 January 2025 / Published: 9 January 2025
(This article belongs to the Special Issue Traditional and Emerging Pollutants: Human Exposure Profiles and Health Effects)
Browse Figures
Versions Notes

Abstract

:
Volatile organic compounds (VOCs) are associated with obesity health risks, while the association of mixed VOCs with visceral adiposity indicators remains unclear. In this study, a total of 2015 adults from the National Health and Nutrition Examination Survey (NHANES) were included. Weighted generalized linear models, restricted cubic spline (RCS), weighted quantile sum (WQS), and Bayesian kernel machine regression (BKMR) were adopted to assess the association of VOC metabolites (mVOCs) with the visceral adiposity index (VAI) and lipid accumulation product (LAP). Multiple mVOCs were positively associated with the VAI and LAP in the single-exposure model, especially N-acetyl-S-(2-carboxyethyl)-L-cysteine (CEMA) and N-acetyl-S-(N-methylcarbamoyl)-L-cysteine (AMCC). The associations of mVOCs with VAI and LAP were more significant in <60-year-old and non-obese individuals, with interactions of CEMA with age and AMCC with obesity status. Nonlinear relationships between certain mVOCs and the VAI or the LAP were also observed. In the WQS model, co-exposure to mVOCs was positively correlated with the VAI [β (95%CI): 0.084 (0.022, 0.147)]; CEMA (25.24%) was the major contributor. The result of the BKMR revealed a positive trend of the association between mixed mVOCs and the VAI. Our findings suggest that VOC exposure is strongly associated with visceral obesity indicators. Further large prospective investigations are necessary to support our findings.

1. Introduction

Obesity is defined as an excessive buildup of body fat. According to the World Health Organization (WHO), more than 890 million adults worldwide were obese (16%) as of 2022 [1]. Obesity is associated with the prevalence of many diseases, notably metabolic diseases such as type 2 diabetes, cardiovascular disease, and hypertension [2,3,4]. It is important to note that individuals with a comparable body mass index (BMI) vary in their body fat distribution, metabolic profile, and the degree of related metabolic risk [5]. Obesity health risks are primarily affected by body fat distribution, with visceral fat (rather than subcutaneous fat) being a more crucial risk factor [5,6,7]. A previous review referred to metabolic syndrome as “visceral adiposity syndrome”. The expansion of visceral adipose tissue stimulates an inflammatory response and produces cytokines that directly interfere with insulin signaling, leading to insulin resistance, which is a key risk factor for metabolic disorders [8]. Genetics and lifestyle (e.g., diet and physical activity) have a remarkable effect on obesity health risk. Moreover, studies suggest that environmental pollutants are also related to insulin resistance, glucose and lipid homeostasis, and obesity [9,10].
Volatile organic compounds (VOCs) are ubiquitous pollutants, originating primarily from the emissions of numerous day-to-day products used in residential and commercial applications [11]. VOCs are predominantly found in the air, especially indoors, where multiple sources of exposure exist alongside inadequate ventilation [12]. This results in a wider and easier VOC exposure for the general population. Previous studies suggested that VOC exposure has adverse effects on human health and is correlated with the risk of several diseases, such as respiratory diseases, cancers, type 2 diabetes, cardiovascular diseases, and metabolic syndrome [11,13,14,15,16]. A recent cross-sectional study showed that concentrations of multiple VOC metabolites (mVOCs) were associated with obesity or abdominal obesity in adults [17]. However, conventional measurements of obesity, such as BMI and waist circumference (WC), only provide a crude assessment of obesity and cannot distinguish between subcutaneous and visceral fat. Epidemiological data on the association between VOC exposure (especially mixed exposure) and visceral adiposity are still sparse.
Various measurements of obesity have been developed and used in epidemiologic investigations. Some simple, inexpensive, and effective identifiers of visceral obesity have attracted attention, such as the visceral adiposity index (VAI) and lipid accumulation product (LAP). They involve the integration of anthropometric and lipid metabolic parameter information to effectively identify visceral obesity and associated cardiometabolic risk [5]. The VAI is useful for assessing visceral fat distribution and dysfunction [18]. The LAP serves to assess abdominal lipid accumulation status [19,20]. The identification of potential VOCs related to the VAI or LAP can contribute to a deeper understanding of the mechanisms by which VOCs affect obesity and its related health effects.
In this study, we aimed to systematically investigate the associations between VOC exposure and visceral adiposity indicators (i.e., the VAI and LAP) in general adults using data from the National Health and Nutrition Examination Survey (NHANES). Upon exposure, VOCs undergo metabolic transformation and are excreted as one or more mercapturic acids [21]. Considering that the physiological half-life of urinary mVOCs in urine is longer than that of parent VOCs in blood and considering the specificity of most mercapturic acid metabolites, urinary mVOCs were used as the indicator of VOC exposure in this study [21,22,23]. Our findings will provide new evidence to further elucidate the complex associations between chemical exposure and obesity.

2. Methods

2.1. Study Population

The NHANES is a large-scale, cross-sectional survey conducted in 2-year cycles to assess the nutritional and health status of the U.S. general population. Participants all provided informed consent to participate in the program. The population for this study was drawn from four NHANES cycles (2011–2012, 2013–2014, 2015–2016, and 2017–2018). Survey data were collected through household interviews, standardized body measures, and laboratory tests [24]. Our study focused on adults (≥20 years of age, n = 22617). Pregnant individuals (n = 247) were excluded. Figure S1 shows the flowchart of participant screening. Finally, this study included 2015 participants.

2.2. Measurement of Urinary VOC Metabolites

Urinary mVOCs were quantified using ultra-performance liquid chromatography–electrospray tandem mass spectrometry (UPLC-ESI/MSMS) as outlined by Alwis et al. [22]. Detailed information and experimental protocols are available in the laboratory method file (https://wwwn.cdc.gov/nchs/data/nhanes/public/2013/labmethods/UVOC_H_MET.pdf) (accessed on 26 December 2024).
The lower limits of detection (LLODs) for the metabolites remained consistent across all four survey cycles. Analyte concentrations below the LLOD were assigned a value of LLOD divided by the square root of 2. We excluded analytes with less than 75% detection and ultimately selected 16 mVOCs for analysis: N-acetyl-S-(2-carboxyethyl)-L-cysteine (CEMA), N-acetyl-S-(3-hydroxypropyl-1-methyl)-L-cysteine (HMPMA), N-acetyl-S-(4-hydroxy-2-butenyl)-L-cysteine (MHBMA3), N-acetyl-S-(2-cyanoethyl)-L-cysteine (CYMA), N-acetyl-S-(N-methylcarbamoyl)-L-cysteine (AMCC), N-acetyl-S-(2-hydroxypropyl)-L-cysteine (2HPMA), 2-methylhippuric acid (2MHA), N-acetyl-S-(3-hydroxypropyl)-L-cysteine (3HPMA), 3- and 4-methylhippuric acid (3-4MHA), N-acetyl-S-(2-carbamoylethyl)-L-cysteine (AAMA), 2-aminothiazoline-4-carboxylic acid (ATCA), N-acetyl-S-(n-propyl)-L-cysteine (BPMA), N-acetyl-S-(3,4-dihydroxybutyl)-L-cysteine (DHBMA), mandelic acid (MA), N-acetyl-S-(benzyl)-L-cysteine (BMA), and phenylglyoxylic acid (PGA). The parent compounds, LLODs, detection rates, and concentration distributions of these metabolites are presented in Table S1. Urine creatinine was introduced to correct mVOC levels to account for urine dilution. Specifically, mVOC levels were corrected by dividing them by the creatinine concentrations and ultimately expressed in ug/g creatinine (ug/g Cr).

2.3. Assessment of Outcomes

Serum samples for triglycerides (TGs) and high-density lipoprotein cholesterol (HDL-C) testing were collected after 8–12 h of fasting. Measurements of WC and BMI were performed by trained technicians in the Mobile Examination Center. The VAI and LAP were calculated according to the equations established in previous studies [18,20]. The equations are as follows:
M a l e s :   V A I = W C 39.68 + 1.88 × B M I × T G 1.03 × 1.31 H D L − C
F e m a l e s :   V A I = W C 36.58 + 1.89 × B M I × T G 0.81 × 1.52 H D L − C
M a l e s :   L A P = W C − 65 × T G
F e m a l e s :   L A P = W C − 58 × T G
where WC and BMI are in cm and kg/m2, respectively, and TGs and HDL-C are in mmol/L.

2.4. Assessment of Covariates

Covariates were calculated or categorized based on previous studies, including demographic variables, lifestyle, and chronic diseases [23,25,26,27,28,29]. Specifically included were sex, age, race, education level, poverty-to-income ratio (PIR), BMI, self-reported smoking status, serum cotinine, average daily alcohol consumption, physical activity, Healthy Eating Index-2015 (HEI-2015), diabetes, hypertension, and cardiovascular disease (CVD). The HEI-2015 was calculated using the R package “dietaryindex” [30], based on the data of total nutrient intakes from the first 24 h dietary recall interview (data file “DR1TOT”). Additional details on covariates are provided in Text S1 of the Supplementary Materials.

2.5. Statistical Analysis

Descriptive statistics were performed to analyze the baseline characteristics of the population. Categorical variables were reported as counts (percentages, %) and compared using chi-square tests. Continuous variables were reported as mean (standard deviation, SD) or medians [interquartile range, IQR] and were compared using parametric or nonparametric tests. Due to the right-skewed distribution, the VAI, LAP, and mVOC concentrations were natural log transformed (ln-transformed) before formal analysis. A Pearson correlation analysis was used to examine correlations among the mVOCs. Given the complex survey design of NHANES, for weighted analyses, we used the strata, cluster, and environmental subsample weight data.

2.5.1. Individual VOC Metabolite Exposure Analysis

Associations of individual mVOCs with the VAI or LAP were examined using survey-weighted generalized linear models (WGLMs). The regression coefficient (β) was interpreted as the average change in the ln-transformed VAI or LAP for each unit increase in each ln-transformed mVOC. Moreover, restricted cubic spline (RCS) models were fitted as sensitivity analyses. RCS models can capture both linear and nonlinear relationships between exposures and outcomes. The number of RCS knots for each mVOC was determined based on the minimum value of the Akaike information criterion to reduce the possibility of RCS under- and over-fitting. We also stratified the analyses by sex (males or females), age (<60 or ≥60 years of age), and obesity status (BMI < 30 kg/m2 is considered non-obesity, ≥30 kg/m2 is considered obesity). Interaction effects were analyzed by introducing interaction terms into the models.

2.5.2. Mixed VOC Metabolite Exposure Analysis

We used two mixed-exposure models to assess the relationship between mixed mVOCs and the VAI or LAP. The weighted quantile sum (WQS) regression analysis assesses the overall effect of mixed mVOCs by constructing weighted indices [31]. Parameter estimation and significance tests for the WQS index effect of the mixture were conducted to determine the association between the index and the outcome. The estimated weights of each mVOC reflect its contribution to the overall effect, with weights exceeding the threshold indicating greater contributions. The threshold is typically the inverse of the number of elements in the mixture. Given the limitations of the WQS on detecting effect direction, both positive and negative models were applied. Bayesian kernel machine regression (BKMR) models were further fitted to assess the exposure–response relationship between exposure and outcome as well as the overall effect of mixed exposures [32]. Specifically, single-exposure response function curves reflect the univariate relationship between each mVOC and outcome when all other mVOCs are fixed at the median. The overall effect of mVOCs was assessed by comparing mixtures of mVOCs fixed at specific percentiles (e.g., 25th and 75th) with those fixed at the median. Conditional posterior inclusion probability (PIP) was used to determine the importance of mVOCs on outcome, with the threshold set at 0.5 [33].
All analyses and plots were performed using R (version 4.3.2) and GraphPad Prism (version 10.1.0) software. Statistical significance was defined as a two-tailed p-value < 0.05. Weighted generalized linear models, RCS, WQS, and BKMR analyses were conducted using the R packages “survey”, “plotRCS”, “gWQS”, and “bkmr”, respectively.

3. Results

3.1. Baseline Characteristics of the Study Population

Table 1 demonstrates the survey-weighted baseline characteristics of the 2015 participants in this study. Participants were grouped by sex and included 1057 males and 958 females. The weighted mean (SD) age for all participants was 47.51 (16.72). The weighted medians [IQR] for the VAI and LAP were 1.33 [0.78, 2.31] and 42.22 [21.88, 72.13], respectively. Non-Hispanic whites (weighted percentage: 69.26%), above high school (62.85%), and BMI ≥ 30 kg/m2 (39.29%) accounted for a higher percentage of their subgroups. Comparative analyses revealed statistically significant differences between males and females in age, race, BMI, smoking, cotinine exposure, alcohol consumption, HEI-2015, physical activity, and CVD prevalence. We also examined demographic information stratified by VAI- or LAP-weighted tertiles (Tables S2 and S3). The results showed that the three VAI or LAP groups differed significantly in age, PIR, race, education, BMI, alcohol consumption, HEI-2015, and physical activity. The prevalence of hypertension, diabetes, and CVD increased with increasing tertiles of the VAI or the LAP.

3.2. Distribution and Correlation of Urinary VOC Metabolites

The distribution of creatinine-corrected urinary mVOC concentrations is shown in Table S4. The weighted median level of DHBMA was the highest among the 16 mVOCs, followed by 3-4MHA, PGA, and 3HPMA, and the lowest was CYMA. In addition, several mVOCs (3-4MHA, AMCC, ATCA, BMA, DHBMA, MA, and PGA) were significantly higher in females than in males, while 3HPMA was significantly higher in males. Levels of mVOCs were also characterized according to age and obesity status. Specific information is shown in Tables S5 and S6. Urinary concentrations of mVOCs were bivariate-correlated with the Pearson correlation coefficients from −0.02 to 0.89 (p < 0.001 for most correlations) (Figure 1). CYMA, 3HPMA, MHBMA3, and HMPMA were highly correlated with each other (r = 0.70–0.89). There were also high correlations between 2MHA and 3-4MHA (r = 0.86) and between CEMA and 3HPMA (r = 0.72). Other statistically significant correlations were relatively weak or moderate, with correlation coefficients ranging from 0.06–0.69. This suggests that these mVOCs probably have similar sources or relatively strong collinearity.

3.3. Association of Individual Urinary VOC Metabolites with VAI and LAP

Three weighted generalized linear models were fitted to explore univariate associations between individual mVOCs and the VAI and LAP. In all three models, AMCC, CEMA, CYMA, MHBMA3, and HMPMA were significantly positively associated with the VAI (Figure 2A); only AMCC and CEMA were significantly positively associated with LAP (Figure 2B). In crude model 1, increased exposure to AAMA, BPMA, and 2HPMA were all significantly associated with a decreased level of the LAP, and an increased level of 2HPMA was also significantly associated with a decreased level of the VAI. These negative associations were not significant after covariate adjustment. In the fully covariate-adjusted model 3, increased levels of 3-4MHA, AMCC, CEMA, CYMA, MHBMA3, and HMPMA were significantly associated with increased levels of the VAI and LAP, respectively. In addition, significant positive associations of 2MHA and 3HPMA with the VAI were observed in model 3.
Sensitivity analyses further employed RCS models to flexibly fit and visualize linear and nonlinear relationships between mVOCs and the VAI or the LAP. As shown in Figures S2 and S3, the RCS showed broadly similar results to the weighted generalized linear models. Consistently, we observed that AMCC, CEMA, CYMA, 3HPMA, MHBMA3, and HMPMA were positively correlated with the VAI (p for overall < 0.05 for all, p for nonlinear > 0.05 except for AMCC) and that AMCC, CYMA, and MHBMA3 were positively correlated with the LAP (p for overall < 0.05, p for nonlinear > 0.05). The results also revealed nonlinear correlations of AAMA, ATCA, 2HPMA, and PGA with the VAI or the LAP (p for nonlinear < 0.05), as well as a significant overall negative trend for 2HPMA and PGA (p for overall < 0.05).

Associations of Urinary VOC Metabolites with VAI and LAP in Stratified Analyses

Stratified subgroup analyses of sex, age, and obesity status were conducted to identify populations sensitive to the effects of VOC exposure on the VAI or LAP. The results are summarized in Figure 3. For the VAI, 2MHA, 3-4MHA, CYMA, MHBMA3, and HMPMA were significantly and positively associated with the VAI regardless of sex subgroups. CEMA and AMCC were positively associated with the VAI in males and females, respectively. After stratification by age and obesity status, the majority of positive associations were observed in the <60 years of age and non-obesity subgroups. In the subgroup of <60 years of age, 2MHA, 3-4MHA, AMCC, CEMA, CYMA, 3HPMA, MHBMA3, and HMPMA were positively associated with the VAI, with significant interactions with age for CEMA (p for interaction = 0.019) and 3HPMA (p for interaction = 0.046). In the subgroup of non-obesity, 2MHA, 3-4MHA, AMCC, CEMA, CYMA, MHBMA3, and HMPMA were positively associated with the VAI, and AMCC had a significant interaction with BMI (p for interaction = 0.010). CEMA, CYMA, MHBMA3, and HMPMA were also associated with the VAI in the obesity subgroup. For the LAP, similar to the VAI, multiple mVOCs were significantly and positively associated with the LAP in both the <60 years of age and the non-obesity subgroups. In particular, for both the VAI and LAP, CEMA and 3HPMA showed significant interactions with age, as did AMCC with BMI. In addition, PGA was significantly negatively associated with the VAI and LAP in the subgroup ≥ 60 years of age, and the interaction with age was significant.

3.4. The Overall Effect of Mixed mVOCs on VAI and LAP

Since there are multiple highly correlated mVOCs, the use of traditional multivariate regression methods may result in collinearity and variance inflation problems or may have major limitations in the selection of relevant components. Therefore, we used the WQS and BKMR models to assess the association of mixed exposure to VOCs with the VAI and LAP and to identify the risk components among mVOCs. Due to the limitations of the WQS model on the direction of association, positive and negative modeling were performed, respectively. After adjusting for covariates, the WQS index for exposure to mixtures of mVOCs was significantly positively correlated with VAI levels [β (95% CI): 0.084 (0.022, 0.147), p = 0.008], while the positive correlation with the LAP was not significant (p > 0.05) (Table 2). No significant association of the WQS index with the VAI or the LAP was observed in the negative WQS regression model. Figure 4 illustrates the WQS weight distribution results. The threshold was set as 1/16. In the positive WQS regression model, for both the VAI and LAP, CEMA (25.24% for the VAI and 28.32% for the LAP) and AMCC (16.02% for VAI and 17.06% for LAP) made the greater contribution to the WQS index.
The BKMR models with a gaussian kernel function were further fitted to estimate the overall effect of 16 mVOCs. Figure 5A,B demonstrate the estimates of the VAI or the LAP when all mVOCs are fixed at different percentiles compared to when all are fixed at the 50th percentile. Similar to the results of the WQS model, the result suggested a positive trend of association of mixed mVOCs with the VAI and LAP, although not significant. According to the results of the univariate exposure–response function, CEMA and AMCC showed a positive trend with both the VAI and LAP when other mVOCs were fixed at the median (Figure 5C,D). The PIP for mVOCs calculated by the BKMR model was used to quantify the relative importance of mVOC exposure on outcomes, as shown in Table S7. AMCC had the highest PIP values for the VAI and LAP, indicating that AMCC had the highest contribution to the overall effect.

4. Discussion

To the best of our knowledge, a knowledge gap remains regarding the relationship between the exposure to VOCs and visceral obesity. This study explored the potential association between urinary mVOCs and the VAI and LAP. In the individual exposure analyses, adjusted for different covariates, concentrations of several mVOCs (including AMCC, CEMA, CYMA, MHBMA3, and HMPMA) were consistently associated with increased levels of the VAI or the LAP. Nonlinear relationships were found among some mVOCs and the VAI or the LAP. Significant interactions were observed between CEMA and age for the VAI and LAP and between AMCC and obesity status. In addition, there was also a positive trend of association between mixed mVOCs and the VAI.
Several studies suggested that the blood TG levels in the presence of large WC might be a simple and useful marker representing excess visceral adiposity [34,35]. Although studies directly investigating the link between VOC exposure and visceral adiposity are lacking, previous studies reported the association between VOC exposure and obesity and blood lipid profiles. An observational study found that higher benzene metabolite levels (t,t-MA) were associated with increased obesity and diabetes risk [16]. Similar to our results, Lei et al. reported positive associations of mVOCs (particularly CEMA) with obesity and abdominal obesity in the US general population [17]. Occupational exposure to VOCs in female nail technicians was correlated to elevated TG and reduced HDL-C levels. Meanwhile, plasma activities of AST and ALT were significantly increased in nail technicians compared to controls, suggesting possible adverse effects of VOC exposure on liver function, which plays a crucial role in lipid metabolism [36,37]. Another study observed significant correlations between certain mVOCs (especially CEMA) and elevated WC, TGs, fasting glucose, and reduced HDL-C [38]. These studies provided part of the support for our findings. Unlike traditional obesity measurement and lipid profiles, the VAI and LAP have higher accuracy in identifying visceral obesity and are more predictive of multiple metabolic diseases [39,40,41,42,43]. The present study was based on these findings and extended the investigation to visceral adiposity markers, filling an important research gap in this field.
The association between VOC exposure and the VAI and LAP may be related to insulin resistance. The expansion of visceral adipose tissue leads to inflammation and insulin resistance, which induces or exacerbates the dysregulation of glucose and lipid homeostasis, making individuals more susceptible to metabolic disorders [8,9,44,45]. Environmental pollution interferes with metabolic processes that regulate lipid accumulation and increase body adiposity [46]. Insulin resistance is more strongly related to regional adipose tissue distribution than to total fat mass [47]. Pollutant (e.g., VOCs) exposure can lead to oxidative stress and inflammatory responses and might result in impaired insulin signaling in insulin-sensitive target tissues (including muscle, liver, and adipose tissue), which can in turn trigger insulin resistance [48,49]. This probably interferes with lipid metabolism and further induces the deposition of visceral adiposity. Previous studies suggested that VOC exposure is associated with insulin resistance and impaired glucose and lipid homeostasis [37,50]. In addition, oxidative stress promotes the accumulation of white adipose tissue, stimulation of preadipocyte proliferation and differentiation, and enlargement of mature adipocytes [51]. Notably, VOCs represent a large class of pollutants, each with their unique toxicity and combinatorial toxicity, and possibly affect visceral adiposity through different mechanisms.
The parent compound of CEMA is acrolein, an unsaturated aldehyde found in tobacco smoke, in industrial emissions, and as a byproduct of lipid peroxidation [52]. Studies suggest that acrolein increases cardiovascular metabolic risk by promoting oxidative stress, inflammation, apoptosis, and systemic dyslipidemia [53,54]. Acrolein was also found to inhibit the insulin signaling pathway and reduce glucose uptake by muscle and adipose tissue cells [55,56]. This suggests that acrolein is a risk factor for abnormalities in glucose and lipid metabolism, which might explain the robust positive correlation of CEMA with the VAI and LAP. As another metabolite of acrolein, 3HPMA also showed a significant positive correlation with the VAI in the model adjusted for covariates. The corresponding parent compound of AMCC is N, N-dimethylformamide (DMF), and the liver is the most sensitive target organ for its toxic effects [57]. An animal study showed that DMF exposure in mice can cause heart and liver injuries, and oxidative stress was involved in the toxic effects [58]. Another in vitro study demonstrated that DMF can disrupt liver function by causing oxidative stress, mitochondrial dysfunction, and alterations in key metabolic pathways including N-glycan biosynthesis and bile acid metabolism, leading to lipid deregulation in hepatocytes [59,60]. Hepatic impairment disrupts lipid metabolism, which might lead to fat accumulation. For other mVOCs that showed positive correlations with the VAI or the LAP, exposure to acrylonitrile (CYMA) or crotonaldehyde (HMPMA) was reported to increase oxidative stress [61,62]. Additionally, 1,3-butadiene (MHBMA3) was found to cause disturbances in metabolism-related enzyme activities, such as the depletion of glutathione and saturation of oxidative metabolism [63]. This induces oxidative stress and the impairment of liver function, probably explaining the positive correlations between these mVOCs and the VAI or the LAP.
The positive associations of mVOCs with the VAI and LAP varied across age and obesity status subgroups. Specifically, the associations were more significant in populations <60 years or non-obese. Younger and non-obese individuals may be more sensitive to the effects of mVOCs since they have higher basal metabolic rates, stronger immune responses, and healthier patterns of fat distribution and inflammatory states. Sex-stratified analyses revealed significant associations of CEMA and AMCC with the VAI and LAP only in males and females, respectively, although no interaction was observed. Acrolein is abundant in cigarette smoke, and Table 1 showed a higher proportion of male smokers and cotinine-exposed individuals compared to females, possibly contributing to the complex exposure profile of males. DMF interferes with the interaction of estrogen receptors with estradiol, potentially making females more sensitive to AMCC [64,65]. Further studies are needed to explore the specific mechanisms by which VOC exposure affects different populations and consider developing targeted prevention and intervention strategies.
Mixed-pollutant exposure is a more realistic environmental health concern compared to single exposure. WQS analyses revealed that exposure to mixed mVOCs was significantly and positively correlated with the VAI, with CEMA and AMCC being the most weighted contributors. It was generally consistent with the individual exposure analyses. A trend of positive association of mixed exposure with the VAI and LAP was also observed in the BKMR model. Notably, the BKMR model showed negative associations of AAMA (acrylamide), ATCA (cyanide), 2HPMA (propylene oxide), and PGA (ethylbenzene and styrene) with the VAI or the LAP, and similar associations were also shown in the RCS model. Acrylamide and cyanide primarily affect the nervous system, potentially impairing central and peripheral nerve function [66,67]. This probably disrupts appetite regulation and stress response, indirectly affecting fat metabolism. High-dose cyanide exposure is lethal, while lower doses might cause weight loss [67]. This partly explains our results. In addition, several of the above VOCs were reported to have hematological effects or carcinogenicity [68]. For instance, occupational exposure to ethylbenzene can decrease hemoglobin levels, probably leading to anemia [69]. These disorders may be accompanied by malnutrition and increased energy expenditure, reducing body adiposity (including visceral adiposity) content. Further experimental validation is needed to elucidate the specific mechanisms.
This study filled a knowledge gap regarding the association between visceral adiposity and VOC exposure. Multiple mVOCs were found to be associated with an increased VAI and LAP, which provides new inspiration for obesity health risk prevention. The results of this study were derived from a large nationally representative sample. We adopted multiple statistical models to assess the individual and combined effects of mVOCs on the VAI and LAP to ensure the robustness of the results. This study also has several limitations. Firstly, a causal relationship between VOC exposure and visceral adiposity could not be determined due to the cross-sectional study design. On the one hand, some common VOCs (e.g., saturated and unsaturated hydrocarbons, aromatic hydrocarbons, and aldehydes) are lipophilic [70]. A high fat content might theoretically lead to a greater accumulation of VOCs in the body, which in turn could lead to higher levels of exposure in obese individuals. However, due to the short half-life of VOCs in the body [22], it is speculated that their accumulation in the body may not be obvious. On the other hand, the accumulation of visceral fat promotes reactive oxygen species (ROS) release and oxidative stress, and certain VOCs can be generated endogenously through the interaction of ROS with underlying cellular components [71]. Therefore, there may be reverse causality between certain VOCs and the VAI and LAP, and further studies are needed to clarify the corresponding endogenous VOC biomarkers in obese individuals and to explore the mechanisms of VOCs in fat accumulation. Secondly, although the effects of a wide range of covariates have been considered, there may still be residual confounders that affect the results. In addition, there is a lack of studies investigating the mechanisms by which VOC exposure affects fat distribution in animals or humans, and the specific mechanisms are still unclear.

5. Conclusions

Our findings revealed that multiple mVOCs are significantly positively associated with levels of visceral adiposity markers (the VAI and LAP), especially CEMA and AMCC. These associations were more significant in people <60 years old and in non-obese individuals. Exposure to mixed mVOCs also showed a positive association with increased VAI levels. In short, this study suggests an association between VOC exposure and visceral obesity. Although the exact mechanism is unclear, we hypothesize that insulin resistance may partially explain these relationships. Prospective studies and more experimental studies of relevant mechanisms are needed to further confirm the conclusions of this study.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/toxics13010046/s1, Text S1: Assessment of covariates; Table S1: Parent compounds, LLODs, detection rates, and descriptive statistics (n = 2015) for urinary mVOCs included in the analysis; Table S2: Weighted baseline characteristics of included participants divided by VAI-weighted tertiles; Table S3: Weighted baseline characteristics of included participants divided by LAP-weighted tertiles; Table S4: Weighted concentrations of mVOCs (μg/g Cr) in urine grouped by sex; Table S5: Weighted concentrations of mVOCs in urine grouped by age; Table S6: Weighted concentrations of mVOCs in urine grouped by obesity status; Table S7: Posterior inclusion probability (PIP) of VOC metabolites derived from BKMR modeling; Figure S1: Flowchart of the participant screening process for inclusion in the analysis; Figure S2: RCS results of the relationships between urinary mVOCs and the VAI; Figure S3: RCS results of the relationships between urinary mVOCs and the LAP.

Author Contributions

Conceptualization, X.H.; software, Z.Q.; formal analysis, Z.Q.; methodology, C.D. and Z.Q.; validation, S.C. and L.Y.; data curation, Z.Q. and L.Y.; writing—original draft preparation, Z.Q.; writing—review and editing, C.D., S.C., L.Y., and X.H.; visualization, Z.Q.; supervision, X.H.; funding acquisition, X.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (NSFC) (22376079).

Institutional Review Board Statement

This study was conducted following the Declaration of Helsinki and approved by the National Center for Health Statistics Research Ethics Review Board (https://www.cdc.gov/nchs/nhanes/about/erb.html, accessed on 7 January 2025). NCHS IRB/ERB Protocol Number or Description: Continuation of Protocol #2011-17 and Protocol #2018-01.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The original data presented in this study are openly available in the National Health and Nutrition Examination Survey at https://wwwn.cdc.gov/nchs/nhanes/Default.aspx (accessed on 7 January 2025).

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

WHO, World Health Organization; BMI, body mass index; VOCs, volatile organic compounds; mVOCs, volatile organic compound metabolites; WC, waist circumference; VAI, visceral adiposity index; LAP, lipid accumulation product; NHANES, National Health and Nutrition Examination Survey; UPLC-ESI/MSMS, ultra-performance liquid chromatography–electrospray tandem mass spectrometry; LLODs, lower limits of detection; CEMA, N-acetyl-S-(2-carboxyethyl)-L-cysteine; HMPMA, N-acetyl-S-(3-hydroxypropyl-1-methyl)-L-cysteine; MHBMA3, N-acetyl-S-(4-hydroxy-2-butenyl)-L-cysteine; CYMA, N-acetyl-S-(2-cyanoethyl)-L-cysteine; AMCC, N-acetyl-S-(N-methylcarbamoyl)-L-cysteine; 2HPMA, N-acetyl-S-(2-hydroxypropyl)-L-cysteine; 2MHA, 2-methylhippuric acid; 3HPMA, N-acetyl-S-(3-hydroxypropyl)-L-cysteine; 3-4MHA, 3- and 4-methylhippuric acid; AAMA, N-acetyl-S-(2-carbamoylethyl)-L-cysteine; ATCA, 2-aminothiazoline-4-carboxylic acid; BPMA, N-acetyl-S-(n-propyl)-L-cysteine; DHBMA, N-acetyl-S-(3,4-dihydroxybutyl)-L-cysteine; MA, mandelic acid; BMA, N-acetyl-S-(benzyl)-L-cysteine; PGA, phenylglyoxylic acid; TGs, triglycerides; HDL-C, high-density lipoprotein cholesterol; PIR, poverty-to-income ratio; HEI-2015, Healthy Eating Index-2015; CVD, cardiovascular disease; SD, standard deviation; IQR, interquartile range; WGLM, survey-weighted generalized linear models; RCS, restricted cubic spline; WQS, weighted quantile sum; BKMR, Bayesian kernel machine regression; ROS, reactive oxygen species.

References

  1. World Health Organization. Obesity and Overweight—Fact Sheets. Available online: https://www.who.int/en/news-room/fact-sheets/detail/obesity-and-overweight (accessed on 10 October 2024).
  2. Klein, S.; Gastaldelli, A.; Yki-Järvinen, H.; Scherer, P.E. Why does obesity cause diabetes? Cell Metab. 2022, 34, 11–20. [Google Scholar] [CrossRef] [PubMed]
  3. Powell-Wiley, T.M.; Poirier, P.; Burke, L.E.; Després, J.P.; Gordon-Larsen, P.; Lavie, C.J.; Lear, S.A.; Ndumele, C.E.; Neeland, I.J.; Sanders, P.; et al. Obesity and Cardiovascular Disease: A Scientific Statement from the American Heart Association. Circulation 2021, 143, e984–e1010. [Google Scholar] [CrossRef]
  4. Kotsis, V.; Stabouli, S.; Papakatsika, S.; Rizos, Z.; Parati, G. Mechanisms of obesity-induced hypertension. Hypertens. Res. 2010, 33, 386–393. [Google Scholar] [CrossRef]
  5. Piché, M.E.; Tchernof, A.; Després, J.P. Obesity Phenotypes, Diabetes, and Cardiovascular Diseases. Circ. Res. 2020, 126, 1477–1500. [Google Scholar] [CrossRef]
  6. Ibrahim, M.M. Subcutaneous and visceral adipose tissue: Structural and functional differences. Obes. Rev. 2010, 11, 11–18. [Google Scholar] [CrossRef]
  7. Ahima, R.S.; Lazar, M.A. The health risk of obesity—Better metrics imperative. Science 2013, 341, 856–858. [Google Scholar] [CrossRef]
  8. Lopes, H.F.; Corrêa-Giannella, M.L.; Consolim-Colombo, F.M.; Egan, B.M. Visceral adiposity syndrome. Diabetol. Metab. Syndr. 2016, 8, 40. [Google Scholar] [CrossRef]
  9. Khalil, W.J.; Akeblersane, M.; Khan, A.S.; Moin, A.S.M.; Butler, A.E. Environmental pollution and the risk of developing metabolic disorders: Obesity and diabetes. Int. J. Mol. Sci. 2023, 24, 8870. [Google Scholar] [CrossRef]
  10. Wang, C.; Meng, X.-c.; Huang, C.; Wang, J.; Liao, Y.-h.; Huang, Y.; Liu, R. Association between ambient air pollutants and lipid profile: A systematic review and meta-analysis. Ecotoxicol. Environ. Saf. 2023, 262, 115140. [Google Scholar] [CrossRef]
  11. Zhou, X.; Zhou, X.; Wang, C.; Zhou, H. Environmental and human health impacts of volatile organic compounds: A perspective review. Chemosphere 2023, 313, 137489. [Google Scholar] [CrossRef]
  12. Liu, N.; Bu, Z.; Liu, W.; Kan, H.; Zhao, Z.; Deng, F.; Huang, C.; Zhao, B.; Zeng, X.; Sun, Y.; et al. Indoor exposure levels and risk assessment of volatile organic compounds in residences, schools, and offices in China from 2000 to 2021: A systematic review. Indoor Air 2022, 32, e13091. [Google Scholar] [CrossRef] [PubMed]
  13. Kwon, J.W.; Park, H.W.; Kim, W.J.; Kim, M.G.; Lee, S.J. Exposure to volatile organic compounds and airway inflammation. Environ. Health 2018, 17, 65. [Google Scholar] [CrossRef] [PubMed]
  14. Lin, L.Y.; Chuang, H.C.; Liu, I.J.; Chen, H.W.; Chuang, K.J. Reducing indoor air pollution by air conditioning is associated with improvements in cardiovascular health among the general population. Sci. Total Environ. 2013, 463–464, 176–181. [Google Scholar] [CrossRef] [PubMed]
  15. Dong, R.; Chang, D.; Shen, C.; Shen, Y.; Shen, Z.; Tian, T.; Wang, J. Association of volatile organic compound exposure with metabolic syndrome and its components: A nationwide cross-sectional study. BMC Public Health 2024, 24, 671. [Google Scholar] [CrossRef] [PubMed]
  16. Lee, I.; Park, H.; Kim, M.J.; Kim, S.; Choi, S.; Park, J.; Cho, Y.H.; Hong, S.; Yoo, J.; Cheon, G.J. Exposure to polycyclic aromatic hydrocarbons and volatile organic compounds is associated with a risk of obesity and diabetes mellitus among Korean adults: Korean National Environmental Health Survey (KoNEHS) 2015–2017. Int. J. Hyg. Environ. Health 2022, 240, 113886. [Google Scholar] [CrossRef]
  17. Lei, T.; Qian, H.; Yang, J.; Hu, Y. The association analysis between exposure to volatile organic chemicals and obesity in the general USA population: A cross-sectional study from NHANES program. Chemosphere 2023, 315, 137738. [Google Scholar] [CrossRef]
  18. Amato, M.C.; Giordano, C.; Galia, M.; Criscimanna, A.; Vitabile, S.; Midiri, M.; Galluzzo, A. Visceral Adiposity Index: A reliable indicator of visceral fat function associated with cardiometabolic risk. Diabetes Care 2010, 33, 920–922. [Google Scholar] [CrossRef]
  19. Zhou, T.; Chen, S.; Mao, J.; Zhu, P.; Yu, X.; Lin, R. Association between obstructive sleep apnea and visceral adiposity index and lipid accumulation product: NHANES 2015–2018. Lipids Health Dis. 2024, 23, 100. [Google Scholar] [CrossRef]
  20. Kahn, H.S. The ”lipid accumulation product” performs better than the body mass index for recognizing cardiovascular risk: A population-based comparison. BMC Cardiovasc. Disord. 2005, 5, 26. [Google Scholar] [CrossRef]
  21. Ding, Y.S.; Blount, B.C.; Valentin-Blasini, L.; Applewhite, H.S.; Xia, Y.; Watson, C.H.; Ashley, D.L. Simultaneous determination of six mercapturic acid metabolites of volatile organic compounds in human urine. Chem. Res. Toxicol. 2009, 22, 1018–1025. [Google Scholar] [CrossRef]
  22. Alwis, K.U.; Blount, B.C.; Britt, A.S.; Patel, D.; Ashley, D.L. Simultaneous analysis of 28 urinary VOC metabolites using ultra high performance liquid chromatography coupled with electrospray ionization tandem mass spectrometry (UPLC-ESI/MSMS). Anal. Chim. Acta 2012, 750, 152–160. [Google Scholar] [CrossRef] [PubMed]
  23. Wang, X.; Chen, Z.; Cheng, D.; Cao, Y.; Xie, X.; Zhou, J.; Wu, Y.; Li, X.; Yu, J.; Yang, B. Association between urinary metabolites of volatile organic compounds and cardiovascular disease in the general population from NHANES 2011–2018. Ecotoxicol. Environ. Saf. 2023, 264, 115412. [Google Scholar] [CrossRef]
  24. Paulose-Ram, R.; Burt, V.; Broitman, L.; Ahluwalia, N. Overview of Asian American Data Collection, Release, and Analysis: National Health and Nutrition Examination Survey 2011–2018. Am. J. Public Health 2017, 107, 916–921. [Google Scholar] [CrossRef]
  25. Tan, Y.; Fu, Y.; Yao, H.; Li, H.; Wu, X.; Guo, Z.; Liang, X.; Kuang, M.; Tan, L.; Jing, C. The relationship of organophosphate flame retardants with hyperuricemia and gout via the inflammatory response: An integrated approach. Sci. Total Environ. 2024, 908, 168169. [Google Scholar] [CrossRef]
  26. Tian, X.; Xue, B.; Wang, B.; Lei, R.; Shan, X.; Niu, J.; Luo, B. Physical activity reduces the role of blood cadmium on depression: A cross-sectional analysis with NHANES data. Environ. Pollut. 2022, 304, 119211. [Google Scholar] [CrossRef]
  27. Krebs-Smith, S.M.; Pannucci, T.E.; Subar, A.F.; Kirkpatrick, S.I.; Lerman, J.L.; Tooze, J.A.; Wilson, M.M.; Reedy, J. Update of the Healthy Eating Index: HEI-2015. J. Acad. Nutr. Diet. 2018, 118, 1591–1602. [Google Scholar] [CrossRef]
  28. Wang, X.; He, W.; Wu, X.; Song, X.; Yang, X.; Zhang, G.; Niu, P.; Chen, T. Exposure to volatile organic compounds is a risk factor for diabetes: A cross-sectional study. Chemosphere 2023, 338, 139424. [Google Scholar] [CrossRef]
  29. Lu, L.; Ni, R. Association between polycyclic aromatic hydrocarbon exposure and hypertension among the U.S. adults in the NHANES 2003-2016: A cross-sectional study. Environ. Res. 2023, 217, 114907. [Google Scholar] [CrossRef]
  30. Zhan, J.J.; Hodge, R.A.; Dunlop, A.L.; Lee, M.M.; Bui, L.; Liang, D.; Ferranti, E.P. Dietaryindex: A user-friendly and versatile R package for standardizing dietary pattern analysis in epidemiological and clinical studies. Am. J. Clin. Nutr. 2024, 120, 1165–1174. [Google Scholar] [CrossRef]
  31. Carrico, C.; Gennings, C.; Wheeler, D.C.; Factor-Litvak, P. Characterization of Weighted Quantile Sum Regression for Highly Correlated Data in a Risk Analysis Setting. J. Agric. Biol. Environ. Stat. 2015, 20, 100–120. [Google Scholar] [CrossRef]
  32. Bobb, J.F.; Valeri, L.; Claus Henn, B.; Christiani, D.C.; Wright, R.O.; Mazumdar, M.; Godleski, J.J.; Coull, B.A. Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures. Biostatistics 2015, 16, 493–508. [Google Scholar] [CrossRef] [PubMed]
  33. Zhang, Z.; Zhong, Q.; Qian, Z.; Zeng, X.; Zhang, J.; Xu, X.; Hylkema, M.N.; Nolte, I.M.; Snieder, H.; Huo, X. Alterations of gut microbiota and its metabolomics in children with 6PPDQ, PBDE, PCB, and metal(loid) exposure. J. Hazard. Mater. 2024, 475, 134862. [Google Scholar] [CrossRef] [PubMed]
  34. Arsenault, B.J.; Lemieux, I.; Després, J.-P.; Wareham, N.J.; Kastelein, J.J.; Khaw, K.-T.; Boekholdt, S.M. The hypertriglyceridemic-waist phenotype and the risk of coronary artery disease: Results from the EPIC-Norfolk prospective population study. Can. Med. Assoc. J. 2010, 182, 1427–1432. [Google Scholar] [CrossRef] [PubMed]
  35. LeBlanc, S.; Coulombe, F.; Bertrand, O.F.; Bibeau, K.; Pibarot, P.; Marette, A.; Alméras, N.; Lemieux, I.; Després, J.P.; Larose, E. Hypertriglyceridemic waist: A simple marker of high-risk atherosclerosis features associated with excess visceral adiposity/ectopic fat. J. Am. Heart Assoc. 2018, 7, e008139. [Google Scholar] [CrossRef]
  36. Grešner, P.; Król, M.B.; Świercz, R.; Gromadzińska, J. Blood plasma levels of biomarkers of liver status and lipid profile among nail technicians occupationally exposed to low-level mixture of volatile organic compounds. Int. Arch. Occup. Environ. Health 2021, 94, 487–494. [Google Scholar] [CrossRef]
  37. Chen, W.-Y.; Fu, Y.-P.; Tu, H.; Zhong, W.; Zhou, L. The association between exposure to volatile organic compounds and serum lipids in the US adult population. Lipids Health Dis. 2023, 22, 129. [Google Scholar] [CrossRef]
  38. Tan, L.; Liu, Y.; Liu, J.; Liu, Z.; Shi, R. Associations of individual and mixture exposure to volatile organic compounds with metabolic syndrome and its components among US adults. Chemosphere 2024, 347, 140683. [Google Scholar] [CrossRef]
  39. Wang, J.; Jin, X.; Chen, K.; Yan, W.; Wang, A.; Zhu, B.; Wang, W.; Gao, Z.; Tang, X.; Yan, L. Visceral adiposity index is closely associated with urinary albumin-creatinine ratio in the Chinese population with prediabetes. Diabetes Metab. Res. Rev. 2021, 37, e3424. [Google Scholar] [CrossRef]
  40. Ioachimescu, A.G.; Brennan, D.M.; Hoar, B.M.; Hoogwerf, B.J. The lipid accumulation product and all-cause mortality in patients at high cardiovascular risk: A PreCIS database study. Obesity 2010, 18, 1836–1844. [Google Scholar] [CrossRef]
  41. Kouli, G.M.; Panagiotakos, D.B.; Kyrou, I.; Georgousopoulou, E.N.; Chrysohoou, C.; Tsigos, C.; Tousoulis, D.; Pitsavos, C. Visceral adiposity index and 10-year cardiovascular disease incidence: The ATTICA study. Nutr. Metab. Cardiovasc. Dis. 2017, 27, 881–889. [Google Scholar] [CrossRef]
  42. Kahn, H.S. The lipid accumulation product is better than BMI for identifying diabetes: A population-based comparison. Diabetes Care 2006, 29, 151–153. [Google Scholar] [CrossRef] [PubMed]
  43. Chen, Z.-y.; Liu, L.; Zhuang, X.-x.; Zhang, Y.-c.; Ma, Y.-n.; Liu, Y.; Wen, D.-l. Lipid accumulation product is a better predictor of metabolic syndrome in Chinese adolescents: A cross-sectional study. Front. Endocrinol. 2023, 14, 1179990. [Google Scholar] [CrossRef] [PubMed]
  44. Lin, X.; Li, H. Obesity: Epidemiology, pathophysiology, and therapeutics. Front. Endocrinol. 2021, 12, 706978. [Google Scholar] [CrossRef]
  45. Rana, M.N.; Neeland, I.J. Adipose tissue inflammation and cardiovascular disease: An update. Curr. Diabetes Rep. 2022, 22, 27–37. [Google Scholar] [CrossRef] [PubMed]
  46. Martínez-Esquivel, A.; Trujillo-Silva, D.J.; Cilia-López, V.G. Impact of environmental pollution on the obesogenic environment. Nutr. Rev. 2022, 80, 1787–1799. [Google Scholar] [CrossRef]
  47. Park, S.E.; Park, C.-Y.; Choi, J.M.; Chang, E.; Rhee, E.-J.; Lee, W.-Y.; Oh, K.W.; Park, S.W.; Kang, E.S.; Lee, H.C. Depot-specific changes in fat metabolism with aging in a type 2 diabetic animal model. PLoS ONE 2016, 11, e0148141. [Google Scholar] [CrossRef]
  48. Shen, Q.; Liu, Y.; Li, G.; An, T. A review of disrupted biological response associated with volatile organic compound exposure: Insight into identification of biomarkers. Sci. Total Environ. 2024, 948, 174924. [Google Scholar] [CrossRef]
  49. Keane, K.N.; Cruzat, V.F.; Carlessi, R.; de Bittencourt, P.I.H., Jr.; Newsholme, P. Molecular events linking oxidative stress and inflammation to insulin resistance and β-cell dysfunction. Oxid. Med. Cell Longev. 2015, 2015, 181643. [Google Scholar] [CrossRef]
  50. Duan, X.; Chen, Z.; Xia, C.; Zhong, R.; Liu, L.; Long, L. Increased levels of urine volatile organic compounds are associated with diabetes risk and impaired glucose homeostasis. J. Clin. Endocrinol. Metab. 2024, 109, e531–e542. [Google Scholar] [CrossRef]
  51. Yan, S.; Chen, S.; Liu, Y.; Liang, H.; Zhang, X.; Zhang, Q.; Xiu, J. Associations of serum carotenoids with visceral adiposity index and lipid accumulation product: A cross-sectional study based on NHANES 2001–2006. Lipids Health Dis 2023, 22, 209. [Google Scholar] [CrossRef]
  52. Henning, R.J.; Johnson, G.T.; Coyle, J.P.; Harbison, R.D. Acrolein can cause cardiovascular disease: A review. Cardiovasc. Toxicol. 2017, 17, 227–236. [Google Scholar] [CrossRef] [PubMed]
  53. Aydın, B.; Şekeroğlu, Z.A.; Şekeroğlu, V. Effects of whey protein and conjugated linoleic acid on acrolein-induced cardiac oxidative stress, mitochondrial dysfunction and dyslipidemia in rats. Biomed. Pharmacother. 2018, 107, 901–907. [Google Scholar] [CrossRef]
  54. Conklin, D.J.; Barski, O.A.; Lesgards, J.-F.; Juvan, P.; Rezen, T.; Rozman, D.; Prough, R.A.; Vladykovskaya, E.; Liu, S.; Srivastava, S. Acrolein consumption induces systemic dyslipidemia and lipoprotein modification. Toxicol. Appl. Pharmacol. 2010, 243, 1–12. [Google Scholar] [CrossRef] [PubMed]
  55. Jhuo, J.-Y.; Tong, Z.-J.; Ku, P.-H.; Cheng, H.-W.; Wang, H.-T. Acrolein induces mitochondrial dysfunction and insulin resistance in muscle and adipose tissues in vitro and in vivo. Environ. Pollut. 2023, 336, 122380. [Google Scholar] [CrossRef] [PubMed]
  56. Medina-Navarro, R.; Guzmán-Grenfell, A.M.; Díaz-Flores, M.; Duran-Reyes, G.; Ortega-Camarillo, C.; Olivares-Corichi, I.M.; Hicks, J.J. Formation of an adduct between insulin and the toxic lipoperoxidation product acrolein decreases both the hypoglycemic effect of the hormone in rat and glucose uptake in 3T3 adipocytes. Chem. Res. Toxicol. 2007, 20, 1477–1481. [Google Scholar] [CrossRef] [PubMed]
  57. Wang, C.; Yang, J.; Lu, D.; Fan, Y.; Zhao, M.; Li, Z. Oxidative stress-related DNA damage and homologous recombination repairing induced by N, N-dimethylformamide. J. Appl. Toxicol. 2016, 36, 936–945. [Google Scholar] [CrossRef]
  58. Rui, D.; Daojun, C.; Yongjian, Y. Liver and heart toxicity due to 90-day oral exposure of ICR mice to N, N-dimethylformamide. Environ. Toxicol. Pharmacol. 2011, 31, 357–363. [Google Scholar] [CrossRef]
  59. Xu, L.; Zhao, Q.; Luo, J.; Ma, W.; Jin, Y.; Li, C.; Hou, Y.; Feng, M.; Wang, Y.; Chen, J. Integration of proteomics, lipidomics, and metabolomics reveals novel metabolic mechanisms underlying N, N-dimethylformamide induced hepatotoxicity. Ecotoxicol. Environ. Saf. 2020, 205, 111166. [Google Scholar] [CrossRef]
  60. Li, M.-J.; Zeng, T. The deleterious effects of N, N-dimethylformamide on liver: A mini-review. Chem. Biol. Interact. 2019, 298, 129–136. [Google Scholar] [CrossRef]
  61. Lin, C.-Y.; Lee, H.-L.; Sung, F.-C.; Su, T.-C. Investigating the association between urinary levels of acrylonitrile metabolite N-acetyl-S-(2-cyanoethyl)-L-cysteine and the oxidative stress product 8-hydroxydeoxyguanosine in adolescents and young adults. Environ. Pollut. 2018, 239, 493–498. [Google Scholar] [CrossRef]
  62. Zhang, B.; Li, S.; Men, J.; Peng, C.; Shao, H.; Zhang, Z. Long-term exposure to crotonaldehyde causes heart and kidney dysfunction through induction of inflammatory and oxidative damage in male Wistar rats. Toxicol. Mech. Methods 2019, 29, 263–275. [Google Scholar] [CrossRef] [PubMed]
  63. Kirman, C.R.; Albertini, R.J.; Sweeney, L.M.; Gargas, M.L. 1, 3-Butadiene: I. Review of metabolism and the implications to human health risk assessment. Crit. Rev. Toxicol. 2010, 40, 1–11. [Google Scholar] [CrossRef] [PubMed]
  64. Lian, X.; Guo, J.; Wang, Y.; Wang, S.; Li, J. Association between Volatile Organic Compound Exposure and Sex Hormones in Adolescents: The Mediating Role of Serum Albumin. Toxics 2024, 12, 438. [Google Scholar] [CrossRef] [PubMed]
  65. Sasson, S.; Notides, A.C. The effects of dimethylformamide on the interaction of the estrogen receptor with estradiol. J. Steroid Biochem. 1988, 29, 491–495. [Google Scholar] [CrossRef]
  66. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological Profile for Acrylamide. Available online: https://www.atsdr.cdc.gov/toxprofiles/tp203.pdf (accessed on 14 October 2024).
  67. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological Profile for Cyanide. Available online: https://www.atsdr.cdc.gov/toxprofiles/tp8.pdf (accessed on 14 October 2024).
  68. Li, A.J.; Pal, V.K.; Kannan, K. A review of environmental occurrence, toxicity, biotransformation and biomonitoring of volatile organic compounds. Environ. Chem. Ecotoxicol. 2021, 3, 91–116. [Google Scholar] [CrossRef]
  69. Agency for Toxic Substances and Disease Registry (ATSDR). Toxicological Profile for Ethylbenzene. Available online: https://www.atsdr.cdc.gov/ToxProfiles/tp110.pdf (accessed on 14 October 2024).
  70. Rowan, D.D. Volatile metabolites. Metabolites 2011, 1, 41–63. [Google Scholar] [CrossRef]
  71. Oyerinde, A.S.; Selvaraju, V.; Babu, J.R.; Geetha, T. Potential Role of Oxidative Stress in the Production of Volatile Organic Compounds in Obesity. Antioxidants 2023, 12, 129. [Google Scholar] [CrossRef]
Toxics 13 00046 g001
Figure 1. Correlation matrix for urinary VOC metabolites (ln-transformed, μg/g Cr). Notes: CEMA, N-acetyl-S-(2-carboxyethyl)-L-cysteine; HMPMA, N-acetyl-S-(3-hydroxypropyl-1-methyl)-L-cysteine; MHBMA3, N-acetyl-S-(4-hydroxy-2-butenyl)-L-cysteine; CYMA, N-acetyl-S-(2-cyanoethyl)-L-cysteine; AMCC, N-acetyl-S-(N-methylcarbamoyl)-L-cysteine; 2HPMA, N-acetyl-S-(2-hydroxypropyl)-L-cysteine; 2MHA, 2-methylhippuric acid; 3HPMA, N-acetyl-S-(3-hydroxypropyl)-L-cysteine; 3-4MHA, 3- and 4-methylhippuric acid; AAMA, N-acetyl-S-(2-carbamoylethyl)-L-cysteine; ATCA, 2-aminothiazoline-4-carboxylic acid; BPMA, N-acetyl-S-(n-propyl)-L-cysteine; DHBMA, N-acetyl-S-(3,4-dihydroxybutyl)-L-cysteine; MA, mandelic acid; BMA, N-acetyl-S-(benzyl)-L-cysteine; and PGA, phenylglyoxylic acid.
Figure 1. Correlation matrix for urinary VOC metabolites (ln-transformed, μg/g Cr). Notes: CEMA, N-acetyl-S-(2-carboxyethyl)-L-cysteine; HMPMA, N-acetyl-S-(3-hydroxypropyl-1-methyl)-L-cysteine; MHBMA3, N-acetyl-S-(4-hydroxy-2-butenyl)-L-cysteine; CYMA, N-acetyl-S-(2-cyanoethyl)-L-cysteine; AMCC, N-acetyl-S-(N-methylcarbamoyl)-L-cysteine; 2HPMA, N-acetyl-S-(2-hydroxypropyl)-L-cysteine; 2MHA, 2-methylhippuric acid; 3HPMA, N-acetyl-S-(3-hydroxypropyl)-L-cysteine; 3-4MHA, 3- and 4-methylhippuric acid; AAMA, N-acetyl-S-(2-carbamoylethyl)-L-cysteine; ATCA, 2-aminothiazoline-4-carboxylic acid; BPMA, N-acetyl-S-(n-propyl)-L-cysteine; DHBMA, N-acetyl-S-(3,4-dihydroxybutyl)-L-cysteine; MA, mandelic acid; BMA, N-acetyl-S-(benzyl)-L-cysteine; and PGA, phenylglyoxylic acid.
Toxics 13 00046 g001
Toxics 13 00046 g002
Figure 2. The association analysis between mVOCs and the VAI or the LAP: (A) for the VAI and (B) for the LAP. Model 1 is a crude model unadjusted for any covariates. Model 2 was adjusted for sex, age, race, education, PIR, and BMI. Model 3 was further adjusted for serum cotinine, physical activity, HEI-2015, self-reported smoking status, average daily alcohol consumption, diabetes, hypertension, and CVD based on model 2.
Figure 2. The association analysis between mVOCs and the VAI or the LAP: (A) for the VAI and (B) for the LAP. Model 1 is a crude model unadjusted for any covariates. Model 2 was adjusted for sex, age, race, education, PIR, and BMI. Model 3 was further adjusted for serum cotinine, physical activity, HEI-2015, self-reported smoking status, average daily alcohol consumption, diabetes, hypertension, and CVD based on model 2.
Toxics 13 00046 g002
Toxics 13 00046 g003
Figure 3. Estimated effects of urinary mVOCs on VAI and LAP grouped by sex (A for VAI, D for LAP), age (B for VAI, E for LAP), and obesity status (C for VAI, F for LAP) based on interaction models. Notes: *, p for interaction < 0.05; obesity status was categorized as non-obesity (BMI < 30 kg/m2) and obesity (BMI ≥ 30 kg/m2).
Figure 3. Estimated effects of urinary mVOCs on VAI and LAP grouped by sex (A for VAI, D for LAP), age (B for VAI, E for LAP), and obesity status (C for VAI, F for LAP) based on interaction models. Notes: *, p for interaction < 0.05; obesity status was categorized as non-obesity (BMI < 30 kg/m2) and obesity (BMI ≥ 30 kg/m2).
Toxics 13 00046 g003
Toxics 13 00046 g004
Figure 4. Weights for urinary mVOCs associated with the VAI and LAP from the weighted quantile sum (WQS) regression model: (A) positive weights for the VAI, (B) negative weights of the VAI, (C) positive weights for the LAP, (D) negative weights for the LAP. The red dashed line is the weight threshold for determining the importance of an element, which was set as the inverse of the number of elements in the mixture (i.e., 1/16).
Figure 4. Weights for urinary mVOCs associated with the VAI and LAP from the weighted quantile sum (WQS) regression model: (A) positive weights for the VAI, (B) negative weights of the VAI, (C) positive weights for the LAP, (D) negative weights for the LAP. The red dashed line is the weight threshold for determining the importance of an element, which was set as the inverse of the number of elements in the mixture (i.e., 1/16).
Toxics 13 00046 g004
Toxics 13 00046 g005
Figure 5. Overall relationship between the mixed mVOCs and the (A) VAI and (B) LAP estimated by the Bayesian kernel machine regression (BKMR) model. Univariate exposure–response functions between mVOCs and the (C) VAI and (D) LAP estimated by the BKMR model.
Figure 5. Overall relationship between the mixed mVOCs and the (A) VAI and (B) LAP estimated by the Bayesian kernel machine regression (BKMR) model. Univariate exposure–response functions between mVOCs and the (C) VAI and (D) LAP estimated by the BKMR model.
Toxics 13 00046 g005
Table 1. Weighted baseline characteristics of the study population grouped by sex.
Table 1. Weighted baseline characteristics of the study population grouped by sex.
OverallMaleFemalep-Value
N20151057958
Age (years, mean (SD))47.57 (16.72)46.09 (16.44)49.17 (16.88)0.007
PIR (median [IQR])2.95 [1.49, 4.92]3.06 [1.59, 4.92]2.89 [1.32, 4.92]0.121
Race, n (%) 0.011
   Hispanic461 (13.39)242 (14.83)219 (11.82)
   Non-Hispanic White817 (69.26)432 (70.06)385 (68.39)
   Non-Hispanic Black443 (10.12)229 (8.76)214 (11.58)
   Other races294 (7.24)154 (6.35)140 (8.21)
Education, n (%) 0.322
   High school graduate and below867 (37.15)489 (38.87)378 (35.28)
   Above high school1148 (62.85)568 (61.13)580 (64.72)
BMI 1 0.013
   <25 kg/m2576 (28.21)298 (26.11)278 (30.49)
   25–30 kg/m2649 (32.50)390 (36.92)259 (27.71)
   ≥30 kg/m2790 (39.29)369 (36.97)421 (41.80)
Smoker, n (%) <0.001
   No1108 (55.50)492 (48.68)616 (62.90)
   Yes907 (45.50)565 (51.32)342 (37.10)
Cotinine, n (%) 2 0.014
   <0.015 ng/mL643 (35.01)296 (30.60)347 (39.78)
   ≥0.015 ng/mL1372 (64.99)761 (69.40)611 (60.22)
Alcohol consumption per day (drinks, median [IQR])0.08 [0.00, 0.57]0.20 [0.02, 1.00]0.03 [0.00, 0.28]<0.001
HEI-2015 (mean (SD))50.25 (13.86)49.06 (13.54)51.55 (14.10)0.014
Physical activity, n (%) <0.001
   No479 (19.40)215 (15.39)264 (23.73)
   Moderate896 (45.60)422 (39.83)474 (51.85)
   Vigorous640 (35.01)420 (44.78)220 (24.42)
Hypertension, n (%) 0.190
   No1137 (61.32)570 (59.53)567 (63.26)
   Yes878 (38.68)487 (40.47)391 (36.74)
Diabetes, n (%) 0.397
   No1568 (83.11)804 (82.12)764 (84.18)
   Yes447 (16.89)253 (17.88)194 (15.82)
CVD, n (%) 0.008
   No1804 (92.09)921 (90.31)883 (94.01)
   Yes211 (7.91)136 (9.69)75 (5.99)
VAI (median [IQR])1.33 [0.78, 2.31]1.34 [0.76, 2.36]1.30 [0.81, 2.25]0.932
LAP (median [IQR])42.22 [21.88, 72.13]42.96 [23.07, 72.93]40.95 [21.45, 69.83]0.171
1 Body mass index (BMI) was categorized into normal weight (<25 kg/m2), overweight (25–30 kg/m2), and obesity (≥30 kg/m2); 2 serum cotinine concentration was used to determine whether an individual is tobacco-free (<0.015 ng/mL) or tobacco-exposed (≥0.015 ng/mL). Notes: PIR, poverty-to-income ratio; HEI-2015, Healthy Eating Index-2015; CVD, cardiovascular disease; SD, standard deviation; IQR, interquartile range; VAI, visceral adiposity index; and LAP, lipid accumulation product.
Table 2. β (95% CI) of the VAI and LAP associated with exposure to mixed mVOCs by positive and negative weighted quantile sum (WQS) regression analyses.
Table 2. β (95% CI) of the VAI and LAP associated with exposure to mixed mVOCs by positive and negative weighted quantile sum (WQS) regression analyses.
Positive WQS ModelNegative WQS Model
β (95% CI)p-valueβ (95% CI)p-value
VAI0.084 (0.022, 0.147)0.0080.003 (−0.056, 0.061)0.933
LAP0.046 (−0.012, 0.105)0.120−0.029 (−0.090, 0.031)0.342
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Qian, Z.; Dai, C.; Chen, S.; Yang, L.; Huo, X. Exposure to Volatile Organic Compounds in Relation to Visceral Adiposity Index and Lipid Accumulation Product Among U.S. Adults: NHANES 2011–2018. Toxics 2025, 13, 46. https://doi.org/10.3390/toxics13010046

AMA Style

Qian Z, Dai C, Chen S, Yang L, Huo X. Exposure to Volatile Organic Compounds in Relation to Visceral Adiposity Index and Lipid Accumulation Product Among U.S. Adults: NHANES 2011–2018. Toxics. 2025; 13(1):46. https://doi.org/10.3390/toxics13010046

Chicago/Turabian Style

Qian, Ziyi, Chenxu Dai, Siyan Chen, Linjie Yang, and Xia Huo. 2025. "Exposure to Volatile Organic Compounds in Relation to Visceral Adiposity Index and Lipid Accumulation Product Among U.S. Adults: NHANES 2011–2018" Toxics 13, no. 1: 46. https://doi.org/10.3390/toxics13010046

APA Style

Qian, Z., Dai, C., Chen, S., Yang, L., & Huo, X. (2025). Exposure to Volatile Organic Compounds in Relation to Visceral Adiposity Index and Lipid Accumulation Product Among U.S. Adults: NHANES 2011–2018. Toxics, 13(1), 46. https://doi.org/10.3390/toxics13010046

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop