|Home | About | Journals | Submit | Contact Us | Français|
From 1999–2014, obesity prevalence increased among adults and youth. Obese individuals may be uniquely susceptible to the proinflammatory effects of ozone because obese humans and animals have been shown to experience a greater decline in lung function than normal-weight subjects. Obesity is independently associated with limitations in lung mechanics with increased ozone dose. However, few epidemiologic studies have examined the interaction between excess weight and ozone exposure among adults. Using PubMed keyword searches and reference lists, we reviewed epidemiologic evidence to identify potential response-modifying factors and determine if obese or overweight adults are at increased risk of ozone-related health effects. We initially identified 170 studies, of which seven studies met the criteria of examining the interaction of excess weight and ozone exposure on cardiopulmonary outcomes in adults, including four short-term ozone exposure studies in controlled laboratory settings and three community epidemiologic studies. In the studies identified, obesity was associated with decreased lung function and increased inflammatory mediators. Results were inconclusive about the effect modification when data were stratified by sex. Obese and overweight populations should be considered as candidate at-risk groups for epidemiologic studies of cardiopulmonary health related to air pollution exposures. Air pollution is a modifiable risk factor that may decrease lung function among obese individuals with implications for environmental and occupational health policy.
Obesity prevalence has been on the rise among adults in the U.S. (Figure 1). Approximately two-thirds of the U.S. adult population were overweight or obese in 2010.1 The study of community air pollution and ozone exposures has not accounted for the rise in obesity prevalence. Scientific investigations of the respiratory effects of obesity spanning the past 40 years have reported that obesity and overweight conditions have a direct negative effect on respiratory well-being in addition to cardiovascular health, and these changes could have a direct effect on received dose and response to ozone (Figure 2).2–7 Understanding the effect of ozone exposures among adults with excess weight is important because the obesity problem is spreading around the globe with high calorie diets, sedentary lifestyles, and aging populations at a time when air pollution is increasingly adding to the global burden of disease.8
Most ozone exposure studies, however, do not account for how the increased prevalence of obesity or overweight status modifies the relationship between air pollution and cardiopulmonary health in the general population. Time series epidemiologic methods that examine death certificates or hospital admissions data do not provide direct information on patients’ weight and other anthropometric measures. Previous controlled ozone exposure laboratory, panel, and cohort studies do not reflect today’s increased prevalence of obesity. The National Institute of Health (NIH) defines obese a body mass index (BMI) of 30 kg/m2 or greater and overweight as adults with a BMI of 25–29.9 kg/m2.9 According to NHANES data, in 2013–2014, the age-adjusted prevalence of obesity in the U.S. was 37.7% (95% CI, 35.8%, 39.7%); among men, obesity prevalence was 35.0% (95% CI, 32.8%, 37.3%); and among women, it was 40.4% (95% CI, 37.6%, 43.3%).10
Ozone exposure has long been shown to reduce air flow and volume on a short-term basis (e.g., as measured by forced expiratory volume in one second (FEV1)) in mainly normal weight healthy adults.11,12 Epidemiologic studies have shown that chronic reduction in FEV1 is a powerful marker of future morbidity and cardiovascular mortality in the general population.13,14 Ozone exposures are associated with increased respiratory hospitalizations and mortality.15–19 Accordingly, the U.S. Environmental Protection Agency (EPA) has judged the relationships between short-term (e.g., hours, days, weeks) exposure to ozone and respiratory morbidity to be causal.11 Moreover, EPA has judged the relationship between long-term (e.g., months to years) ozone exposures and respiratory effects, cardiovascular disease, central nervous system effects, and total mortality as “likely causal.”11 In the EPA’s 2015 review of the science related to the national ambient air quality standards for ozone, EPA designated at-risk populations for these effects as shown in Table 1 (80 Federal Register 65291, October 26, 2015).
EPA’s 2015 review of at-risk populations considered ozone exposure studies from 2006 to July 2012 from the perspective of the Clean Air Act.20 Although EPA’s approach to at-risk populations characterized the available air pollution studies, the review did not consider other relevant biomedical evidence about the effects of obesity or overweight status from a population health perspective or the pulmonary health of overweight and obese groups, as summarized in Table 2. Moreover, evidence from animal models suggests that obese individuals may be susceptible to proinflammatory and oxidative stress injury of air pollution.21,22
Independently from air pollution exposure effects, pulmonary function is known to decline with increased abdominal adiposity or BMI, which may increase susceptibility to ozone-related effects.5,23–25 In the U.S., adults generally gain weight as they age, which may predispose them for worse outcomes from air pollution exposures than seen in previous epidemiologic studies.5 The increased mass of the chest wall in obese individuals reduces compliance and respiratory muscle endurance, which increases the work of breathing.26,27 Obese and overweight adults with increased abdominal fat mass may have less functional residual capacity than healthy-weight adults.5 Reduced lung volume contributes to closure of gas exchange units and ventilation-perfusion mismatching resulting in arterial hypoxemia and trapping of CO2.3,28,29 These abnormalities contribute to inflammation, the pathogenesis of asthma, obstructive sleep apnea, and other respiratory diseases.6,30–32 Although there are obese individuals with healthy lung function, the literature documenting the respiratory status of obese and overweight populations suggests they are compromised compared to normal-weight adults, and as a result, obese people may experience a chronic inflammatory state.
Furthermore, experiments in animal models have shown that the pulmonary inflammatory response elicited by ozone exposure is enhanced in obese animals, which suggests that obese humans compared to nonobese individuals may be at risk of more adverse effects of air pollution.21,33–37 Enhanced pulmonary inflammation and injury have been shown with short-term ozone exposure in genetic and diet-induced obese mice.21,33–35,37–41 Higher levels of proinflammatory cytokines and chemokines (IL-6, CXCL1, MIP-2, MCP-1) have also been observed in obese animals following ozone exposure.21 Additional studies have investigated the role of diet and sedentary behavior on ozone exposure responses.42,43
A similar enhanced inflammatory response has been observed in obese human populations following ozone exposure. This response may augment bronchoconstriction, airway hyperreactivity, and mucus secretion, reducing the patency of conducting airways.44 Additionally, obese people may possess a greater number of peripheral blood leukocytes, which are known to contribute to pulmonary inflammation following exposure to ozone.45 Proinflammatory mediators may be produced locally in the lung or may accumulate in the lung with the leakage of plasma fluid following disruption of the alveolar epithelium.44 Higher levels of proinflammatory cytokines and adipokines have also been reported in the serum of obese subjects, which might contribute to greater airway inflammation.46,47
One of the underlying pathologies of obesity has been hypothesized to be linked to a chronic state of oxidative stress and impaired oxidant defense.48 Ozone mediates some of its adverse effects through oxidative stress; thus, antioxidant nutritional status may affect the risk of ozone-related health effects.49 In addition, individuals with reduced dietary intake of vitamins E and C are at increased risk for ozone-related health effects.20 Thus, poor diets associated with obesity might also be a factor.
Other air pollution literature about particulate matter exposures have highlighted enhanced responsiveness to cardiac end points with excess weight.50 Finally, emerging evidence suggests that improvements in particulate matter air pollution do not offer corresponding improvements for obese adults’ lung function.51
For all these reasons, we hypothesized that people with excess weight would be more reactive to the same ambient concentration of ozone. To test our hypothesis, we reviewed the epidemiological literature to identify potential response-modifying factors to determine if obese adults are at increased or decreased risk of health effects from ozone exposures. We also examined differences by sex in these associations.
To identify epidemiology studies addressing the interaction between air pollution and obesity/overweight status on respiratory system effects, keyword and reference lists searches using PubMed were conducted using keywords, medical headings, and medical subject headings from three groups connected with “AND”:
Conducted on December 1, 2015 and updated on August 29, 2016 and March 14, 2017, the search was restricted to studies published in English in the past 10 years and where the study subjects were adults. We searched for studies published since 2006 to be consistent with the 2015 EPA national ambient air quality standards review and integrated science assessment.
This search initially identified 170 studies; abstracts were reviewed for ozone exposure among adults and a cardiopulmonary outcome. Studies within the past 10 years were then examined for information about the effect of both weight and ozone exposure among adults regarding a cardiopulmonary function. We excluded studies which controlled for BMI or other measures of obesity as an independent predictor of lung function but did not evaluate an interaction between excess weight and ozone. Physiological and biomedical studies of obese adults and ozone exposure studies in humans and animal models were reviewed for factors related to obese and overweight status that may modify the relationship between both short-term and long-term ozone exposure and health effects.
Of the 170 studies identified, four studies were excluded because the subjects were children,52–56 and 159 were excluded because they did not examine the interaction of obesity and air pollution. Many excluded studies examined BMI as a covariate but did not analyze the interaction of BMI and ozone exposure. As shown in Table 3, seven studies met the criteria of examining the interaction of excess weight and ozone exposure on health outcomes in adults.57–63 They can be divided into controlled ozone exposure human studies in laboratories and community studies. Two of the community cross-sectional studies examined cardiovascular end points in the same Chinese study population.61,63
All studies used BMI as a quantitative measure of obesity, and one included waist circumference as a secondary selection criterion, and measured percent body fat by bioelectric impedance analysis although these measures were not analyzed further.58 The controlled human studies generally reported a positive interaction between obesity and ozone exposure with increasing FEV1 responsiveness, with the exception of one retrospective analysis of 40 adults reporting no interaction.64 The one longitudinal cohort study among older men reported a greater decline in lung function among the obese group.57
Most studies reported p-values for effect modification across strata, with one study reporting a three-way interaction trend test among ozone, obesity, and airway hyper-responsiveness.57 The community epidemiologic studies used a variety of ozone exposure metrics including short-term and long-term exposure metrics. In the Boston study, a mean of ozone measurements 48 h prior to the exam, averaged across four local monitors using EPA protocols was selected after an assessment of averages of measurements on 1–5 days prior to exam.57 In the long-term exposure studies in China, researchers used the 3-year annual average ozone (removing outliers) from a central monitor within 1 km of participants’ homes in 33 northeastern Chinese cities. Compared to U.S. ambient levels, the Chinese study assessed significantly higher ozone concentrations: the Chinese 3-year mean was 1.2589 ppm (SD 0.35856) compared to Boston’s 48 h average of 0.0244 ppm (SD 0.011). As a point of reference, the U.S. EPA set the 2015 national ambient air quality standards for ozone at 0.070 ppm maximum 8 h average (80 Federal Register 65291, October 26, 2015).
Weight gain has been shown to reduce lung function in healthy overweight and obese adults.28 In a cohort study of healthy subjects, the quartile who gained the most weight over 10 years compared to the lowest quartile had the largest decrease in forced vital capacity (FVC) and FEV1.5 Thus, increased obesity prevalence contributes to the poor respiratory health of the general adult population. The altered physiological and proinflammatory states typical of obese populations may place an additional burden on their cardiac and respiratory systems from air pollution exposure. Experimental evidence from animal models also supports that obesity modifies the association between ozone exposure and cardiorespiratory health via inflammation34 and oxidative stress.22
Evaluating pulmonary susceptibility is complicated, however, because obesity is a complex metabolic condition that influences many systems and results in a variety of comorbidities (e.g., hypertension, asthma) that may also affect respiratory health. In the seven human subject studies identified, increased BMI was associated with ozone-related decreased lung function and increased inflammatory mediators. Controlled ozone exposure studies in human subjects provided the strongest evidence that the ozone-induced decrement in lung function is greater in the obese group compared to normal weight participants.58,60 Results from a longitudinal cohort study supported this finding; however, the subjects were mainly older white males.57 In the larger Chinese cross-sectional studies, greater positive associations with cardiovascular outcomes were reported for obese than normal-weight participants for long-term relatively higher ozone concentrations (mean 3-year average ozone level of 1.26 ppm).
There is evidence that obese populations receive an increased dose of ozone for the same ambient concentration (Table 2). In controlled air pollution exposure studies among human subjects, researchers can control the dose. In the four studies identified in this review, researchers assigned exposures to known doses of ozone in laboratory-controlled settings among generally healthy nonsmoking adult volunteers. Researchers administered known concentrations of ozone via chamber or facemask exposures and used exercise and body surface area to calculate and control target ventilation. This controlled exposure study design allowed for direct causal inference, control over confounding by copollutants, and assignment of the received dose of ozone to construct dose–response curves. The investigators collected detailed anthropomorphic information, such as body weight and percent body fat measurements, controlled for age, asthma status, smoking, and sex, and used a relatively large sample size. In general, for the ozone-induced effect, subjects served as their own control; investigators compared outcomes with the ozone dose to outcomes with a filtered air exposure. A further advantage was that the lung function measurements were well documented using standardized, reproducible techniques.
While significant insights were gained from controlled human exposure studies, three of the studies were not originally designed to evaluate the potential effect modification by obesity. Thus, few obese subjects were included, and the analysis of effect modification by BMI was typically conducted as a secondary analysis. As a result, BMI may more appropriately be interpreted as differences in body size among normal individuals rather than the effect of obesity on the relationship between ozone dose and lung function. A notable exception was the Bennett et al. 2016 study, which was designed to examine the effect of obesity on ozone exposure effects among healthy nonsmoking women with 0.4 ppm short-term ozone exposures. Future research should consider the detection of effect modification by obesity and overweight status of the ozone dose–response relationship over a wider range of concentrations, averaging times, and types of subjects.
While there are significant advantages to controlling ozone exposure in a laboratory setting, the pattern may deviate significantly from ambient exposure patterns. Another limitation is that chamber study subjects differ from the general population and are generally healthier, younger, and nonsmoking. Thus, insights from community epidemiologic studies augment these inherent limitations. In community studies, fuller ozone exposure patterns are evaluated among larger populations including a broader spectrum of health, lung function, and disease status. Community studies are especially important in studying ozone exposures because of the heterogeneous responses to ozone including presence of weak responders and smaller response among those older than 35 years.65 Strengths of the Normative Aging Study included its study design, long follow-up period, well-characterized spirometry outcomes, control of known confounders, and the quality of the short-term ozone measurements.57 Strengths of the Chinese studies included the large sample size with significant numbers of overweight (n = 8 764) and obese (n = 1 435) participants with a broad age and air pollution range.54,63 Limitations of the Chinese studies included lack of temporality between the exposure and outcome; ozone exposure misclassification; selection bias (e.g., healthy subjects were more likely to participate); and information bias (e.g., recall bias and self-reported end point, the use of prevalence rather than incidence of hypertension). Because of a moderate correlation between ambient ozone and particulate matter concentrations in the study districts, there may also be confounding by particulate matter, which is more strongly associated with cardiovascular outcomes.50
In general, few studies examined the interaction of excess weight and ozone exposure. The studies generally reported greater lung function decrements associated with ozone at increased weight categories. However, the results were inconclusive about effect modification when data were stratified by sex; these results are consistent with primary evidence (not examining the interaction with excess weight) about potential differences by sex in ozone exposure epidemiologic studies of respiratory hospital admissions.66,67 In general, after multiple exposures over a period of days, individuals have lower responses to ozone.68–70 However, young women lose ozone-responsiveness with multiple exposures to ozone three-times faster than young men, although in middle age, men and women lose responsiveness at the same rates.65 In addition, Vancza et al. reported small strain-dependent differences in effects by sex in adult mice with respect to pulmonary inflammation and injury after ozone exposure, with adult females generally more at risk. Lactating female mice incurred the greatest lung injury and inflammation among several strains of mice.71 However, some toxicological studies found some strains exhibiting greater risk in males. Thus, although experimental studies have provided potential biological plausibility for potential sex differences for the effect of ozone exposure on lung function, the results of these limited number of studies are inconclusive.
While the absence of a formal meta-analysis of the epidemiologic studies may be viewed as a limitation of this review, too many differing study designs and too few studies were identified to facilitate meaningful pooling of effect. However, the McDonnell et al. study provides a set of equations for researchers to test new controlled human subject data as they become available.60
The pulmonary inflammatory response elicited by ozone is enhanced in obese animals, which suggests that obese humans may be at increased risk of adverse effects of air pollution. Because obese and overweight populations exhibit limitations in pulmonary function and receive a greater dose of ambient pollution, the extent to which exposure to ozone induces adverse effects should be directly evaluated in additional studies. However, the current evidence generally supports a positive effect of excess weight on the relationship between ozone and lung function. There is a suggestion of a positive interaction for cardiac end points; however, the results by sex are inconclusive in cross-sectional studies. If confirmed, an interaction between excess weight and enhanced ozone response may provide public health advocates and clinicians with additional reason to promote the maintenance of a healthy body weight and diet.
The World Health Organization estimated that approximately 2.3 billion adults worldwide in 2015 were overweight.32 As obesity prevalence increases, there is increasing uncertainty as to how well the principal studies used to calculate end points in risk assessment, economic benefit assessment, or burden of disease calculations represent the impacts with respect to obese or overweight adult and elderly populations. In addition, while further evidence is required, recognition of this susceptibility could help regulators to designate obese populations as at-risk populations under the Clean Air Act for consideration in standard setting and public health warnings for air pollution.
P.D.K. was supported by the University of Michigan, Center for Occupational Health and Safety Engineering with Grant No. T42 OH008455, funded by the Centers for Disease Control and Prevention and the National Institute for Occupational Safety (NIOSH) and with Grant No. P30ES017885 from the National Institute of Environmental Health Sciences, National Institutes of Health. P.M. was supported by grants from the Flight Attendants Medical Research Institute (FAMRI) CIA-103071 and NIH ES017885. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, Centers for Disease Control and Prevention or the National Institute for Occupational Safety (NIOSH).
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemrestox.7b00077.
Brief description of each ozone and obesity study in the review (PDF)
Patricia D. Koman: 0000-0002-8445-6333
Author ContributionsThe manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. P.D.K. wrote most of the first draft of the manuscript with portions written by P.M. Both of the authors contributed to the analysis and interpretation of data and editing of the final draft.
The authors declare no competing financial interest.