Factors Associated with Urinary 1-Hydroxypyrene and Malondialdehyde among Adults near a Petrochemical Factory: Implications for Sex and Lifestyle Modification

Background: The association between the biomarkers of environmental exposure, oxidative stress, and health-related behaviors in community residents living in an endemic viral hepatitis area and near petrochemical industrial complexes remains unclear. From a health promotion perspective, healthcare providers must know what to do for residents concerned about their health and living environment, especially for individual-level and modifiable risk factors. Therefore, we aimed to explore the factors associated with urinary 1-hydroxypyrene (1-OHP) and malondialdehyde (MDA). Methods: A community-based, cross-sectional study was conducted between July 2018 and February 2019 in western coastal Yunlin County, Taiwan. All participants lived within a 10 km radius of a large petrochemical complex and did not work in the factory. This study was conducted with the local hospital through annual community health screening. Biological samples were collected and biomarkers determined and quantified in the central laboratory of the collaborating hospital. Results: A total of 6335 adult residents completed the study. The mean age was 47.7 (SD = 16) years. Out of the total population, 56.4% were female, 30.1% had metabolic syndrome (MetS), and 16.8% and 14.3% had hepatitis B virus antigen (HBsAg) and hepatitis C virus antibody (anti-HCV) positivity, respectively. The median 1-OHP and MDA level was 0.11 and 0.9 μg/g creatinine with an interquartile range of 0.07–0.18, and 0.4–1.5, respectively. The MDA levels correlated with specific diseases. The multivariable ordinal logistic regression model revealed that female sex, smoking, betel nut use, HBsAg, and anti-HCV positivity were associated with higher 1-OHP levels. In men, MetS was associated with higher 1-OHP levels and regular exercise with lower 1-OHP levels. High MDA levels were associated with smoking, betel nut users, HBsAg, and anti-HCV positivity. Conclusions: The findings highlight the importance of initiating individualized health promotion programs for residents near petrochemical factories, especially for adults with substance-use and cardiometabolic risk factors. Furthermore, it is crucial to provide further treatment to patients with viral hepatitis.


Introduction
Previous studies have indicated that petrochemical industrial complexes (PICs) have a negative effect on human health and environment, especially during the manufacturing hepatitis due to inflammation. Previous studies have shown that the coastal city of Yunlin, Taiwan has a high prevalence of residents who engage in alcohol consumption, betel nut chewing, cigarette smoking, inadequate dietary adoption, and low exercise, and live around an endemic viral hepatitis region [31][32][33]. This phenomenon was because, five decades ago, many rural residents of Taiwan were treated with inadequately disinfected medical equipment while sick. For instance, they were given injections by unqualified physicians, which caused many innocent people to be unknowingly infected by HCV [32,33]. Therefore, the aim of this study was to explore the factors associated with urinary 1-OHP and MDA levels among adult residents near large PICs in central southwestern Taiwan.

Study Design and Population
We conducted a community-based, cross-sectional study between July 2018 and February 2019 in Yunlin County. All participants lived within a 10 km radius of a large petrochemical factory for many years. Considering the independent age and ability to participate in this study, the inclusion criteria were (1) area of residence in Mailiao and Taihsi designated townships; (2) more than 20 years of age and not working inside the PICs; (3) able to communicate in Mandarin or Taiwanese; (4) able to walk or drive to the local hospital and community health activity center; and (5) agreeing to participate in this study and signing the informed consent form. The exclusion criteria were an inability to answer questions or incomplete data.

Procedure and Ethical Approval
This study was approved by the institutional review board of the research ethics committee (No: IRB 201800428B0). It was conducted in collaboration with a local hospital through annual community health screening. All procedures were performed in accordance with the Helsinki Declaration. The township heads sent messages (oral and poster) regarding free health check-ups and invited individuals to participate in this study. Six nursing students were recruited as research assistants and trained by the investigators. The research team described the study procedures (e.g., collecting blood and urine specimens and an interview to track health-related habits with a questionnaire). The questionnaire consisted of health-related lifestyle behaviors, which were designed by the research team and based on previous studies [31,32]. Blood samples were drawn 8 h after fast; blood and urine samples were stored and sent to the central laboratory of the cooperating hospital.

Measurements
This study collected the following information via individual interviews and questionnaires along with the physiological biomarkers: Demographic characteristics and health-related lifestyle behaviors, including sex, age, educational level (years), body height (cm), and body weight (kg). Participants were asked to report the frequency of seven health habits: (e.g., how often do you have . . . ?): regular alcohol consumption (never vs. current or former user), betel nut chewing (never vs. current or former user), cigarette smoking (never vs. current or former smoker), at least three servings of vegetables (1.5 bowls) per day, two servings of fruit (1 bowl) per day, at least 1500 mL of water per day, and at least 30 min of exercise (3 times per week). The research team determined the frequencies of eating proper servings of vegetables/fruit/water and of adopting exercise as low (never/seldom) or often (usually/always).
Urine 1-OHP and MDA: Spot urine samples were collected and stored below −80 • C until urine 1-OHP and MDA levels were quantified in the central laboratory of the collaborating hospital. Both 1-OHP and MDA were expressed as µg/g CRE. Urinary 1-OHP was analyzed using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Urinary MDA was quantified using the standard thiobarbituric acid reactive substances (TBARS) assay. Urinary creatinine concentration was used for urinary 1-OHP and MDA adjustments.

Statistical Analyses
Because there were no reference values for urine 1-OHP and MDA concentrations, we equally divided the participants into quartiles according to 1-OHP and MDA levels. The demographics and characteristics among the MDA or 1-OHP quartiles were compared using one-way analysis of variance for continuous variables and a chi-square test for categorical variables, with a Bonferroni multiple comparison when the overall test was significant. The linear trend of the demographics and characteristics across the MDA or 1-OHP quartiles was also tested for linear contrast in a general linear model for continuous variables and the Cochran-Armitage test for categorical variables. Factors associated with higher levels of 1-OHP or MDA were investigated using a multivariable odds proportional model (ordinal logistic regression). The analysis was performed on the entire cohort and stratified by sex. MetS and body mass index (BMI) components were not included in the multivariable model due to collinearity with MetS. Laboratory data for liver function (AST and ALT) were also not entered into the multivariable model due to the collinearity with HBsAg and anti-HCV. Finally, the relationship between 1-OHP and MDA levels was examined using the Spearman's rank correlation. All tests were two-tailed, and p < 0.05 was considered statistically significant. Data analyses were conducted using SPSS 25 (IBM SPSS Inc., Chicago, IL, USA).

Participant Demographic Characteristics
A total of 6335 participants aged ≥ 20 years who completed the examination were included in the study, of whom 3574 (56.4%) were female. The mean age was 47.7 years, and the average education level was 9.6 years. Approximately half of the subjects (55.5%) were overweight or obese and MetS was prevalent in 30.1% of the population (n = 1909). Of note, 16.8% (n = 1067) and 14.3% (n = 909) of the subjects were identified as HBsAg and anti-HCV positive, respectively. The median 1-OHP level was 0.11 µg/g creatinine, with an interquartile range (IQR) of 0.07 to 0.18. The median MDA level was 0.9 µg/g creatinine, with an IQR of 0.4 to 1.5 (Table 1). Compared to male subjects, females tended to have an older age, less education, lower prevalence of MetS, lower likelihood of substance use (smoking, alcohol drinking, and betel nut chewing), better dietary habits, and less exercise. Females also tended to have a lower prevalence of HBsAg and better liver function, but a higher potential for HCV infection. The results also showed that female subjects had slightly higher 1-OHP levels and slightly lower MDA levels than male subjects (Table 1). Additionally, the relationship between 1-OHP and MDA levels was examined ( Figure 1). The results showed that the correlation was significantly positive, but weak (correlation coefficient = 0.09).

Factors Associated with 1-OHP Level
According to the quartiles of 1-OHP, Table 2 shows that there were significant differences in age, sex, education, BMI, MetS and its components (except waist circumference), substance use, dietary habits, regular exercise, HBsAg, anti-HCV, and liver function. Based on the multivariable ordinal logistic regression model, Table 3 demonstrates that older age (odds ratio (OR) 0.97, 95% confidence interval (CI) 0.96-0.98) and a higher education level (OR 0.96, 95% CI 0.95-0.97) were associated with a lower 1-OHP level. In contrast, several factors were identified to be associated with a higher 1-OHP level, including female (OR 2.30, 95% CI 2.06-2.57), smoking (OR 6.94, 95% CI 5.96-8.08), betel nut chewing (OR 1.25, 95% CI 1.04-1.52), HBsAg (OR 1.27, 95% CI 1.12-1.43), and anti-HCV positivity (OR 1.43, 95% CI 1.12-1.43). After stratifying the analysis by sex, the results revealed that MetS was associated with a higher 1-OHP level in male subjects (OR 1.17, 95% CI 1.00-1.36) but not in female subjects. In addition, an association between frequent regular exercise and lower 1-OHP levels was observed only in male participants.

Factors Associated with 1-OHP Level
According to the quartiles of 1-OHP, Table 2 shows that there were significant differences in age, sex, education, BMI, MetS and its components (except waist circumference), substance use, dietary habits, regular exercise, HBsAg, anti-HCV, and liver function. Based on the multivariable ordinal logistic regression model, Table 3 demonstrates that older age (odds ratio (OR) 0.97, 95% confidence interval (CI) 0.96-0.98) and a higher education level (OR 0.96, 95% CI 0.95-0.97) were associated with a lower 1-OHP level. In contrast, several factors were identified to be associated with a higher 1-OHP level, including female (OR 2.30, 95% CI 2.06-2.57), smoking (OR 6.94, 95% CI 5.96-8.08), betel nut chewing (OR 1.25, 95% CI 1.04-1.52), HBsAg (OR 1.27, 95% CI 1.12-1.43), and anti-HCV positivity (OR 1.43, 95% CI 1.12-1.43). After stratifying the analysis by sex, the results revealed that MetS was associated with a higher 1-OHP level in male subjects (OR 1.17, 95% CI 1.00-1.36) but not in female subjects. In addition, an association between frequent regular exercise and lower 1-OHP levels was observed only in male participants.

Factors Associated with MDA Level
Based on the MDA level quartiles, Table 4 shows that there were significant differences in age, sex, education, BMI, MetS and its components, substance use, dietary habit of vegetable (but not fruit) consumption, both types of hepatitis, and liver function. Based on the multivariable ordinal logistic regression model, Table 5  , and anti-HCV positivity (OR 1.48, 95% CI 1.29-1.70) were associated with a higher MDA level. In contrast, a higher education level was associated with a lower MDA level (OR 0.98, 95% CI 0.97-0.99). After stratifying the analysis by sex, the results were generally consistent between male and female participants. The effect of smoking and betel nut chewing in male subjects was more apparent because the prevalence of betel nut chewing was higher in men.

Discussion
Our results provide important data for further study and evidence-based lifestyle modifications. This study had three important findings. First, smoking, betel nut chewing, HBsAg, and anti-HCV positivity were significantly associated with higher urinary concentrations of 1-OHP and MDA. Second, a high prevalence of cardiometabolic risk factors and chronic viral hepatitis was found in this population. Third, sex differences were found in adopting unhealthy lifestyles, and men tended to show substance use and inadequate diet, while females tended to take less regular exercise.
The present findings strongly indicate an association between smoking and urinary 1-OHP concentration, with an odds ratio of 6.94, which was statistically significant in women and men, and is in agreement with the findings of previous studies [12,13,35,36]. This is not surprising, because each cigarette contains 3.2-16.0 ng benzo(a)pyrene [37]. A substantial amount of evidence supports the suggestion that active or passive cigarette smoking-related oxidative stress causes inflammation, which in turn results in further generation of ROS and potentially increases oxidative damage to macromolecular targets that may lead to cancer initiation and/or progression [35,38]. This also explains why smokers tended to have higher urinary MDA levels, with an odds ratio of 1.32. The betel nut (also called areca) has been confirmed as a class I carcinogen and is a popular addictive substance in Asia [31]. Furthermore, some studies have confirmed that areca nut extract and other areca ingredients can induce ROS in oral epithelial cells, which may involve oxidative stress via redox imbalance [39]. Our study revealed that betel nut users had higher 1-OHP and MDA concentrations, with 1.25 and 1.22 odds ratios, respectively. This implies that the influence of betel nut-induced oxidative stress may not be confined to the oral cavity. Notably, previous studies found that cigarette smokers also tended to have betel nut-chewing and alcohol-drinking habits in this study area [31]. The present finding also indicates that education level was significantly associated with lower 1-OHP and MDA levels. A higher education level may enable individuals to adopt healthier lifestyles, including having a balanced diet, less substance use, and more regular exercise. Therefore, further studies are needed to initiate smoking and betel nut cessation programs, especially for community adults with lower socioeconomic status.
Theoretically, adopting adequate exercise and consuming fruits or vegetables rich in dietary antioxidants vitamin C, vitamin E, and carotenoids may lower oxidative stress and result in lower 1-OHP and MDA biomarker levels, as supported by the findings of many studies [12,28]. However, this benefit was not consistent with the findings of the present study. This could be because nutrients were quantified using a simple questionnaire, which might have caused a distortion of the real data or the dietary amount. Studies have shown that it is difficult to demonstrate the benefits of nutrient intervention in nutrient-deficient populations [20]. Therefore, it may not be feasible to manage low oxidative stress therapy using nutrient supplements. The present findings show that adopting regular exercise in male adults resulted in significant benefits in lowering 1-OHP concentration, with an odds ratio of 0.8, but did not influence the MDA biomarker. Hence, quantitation of specific types of exercise and diets should be clarified in further studies, to determine the most effective practices for relieving oxidative stress.
According to a report from the Nutrition and Health Survey in Taiwan (NAHSIT) 2005-2008, the prevalence of adult MetS was 25.5% and increased from 13.6% in NAH- SIT 1993SIT -1996. The present study demonstrated that 30.1% of the participants had MetS and 55.5% were overweight or obese, and these frequencies were significantly higher than those observed nationwide and in Asian regions [40]. Since ROS can initiate the inflammatory signaling pathway, in which cytokines induce endothelial cell dysfunction, the serial reactions would result in atheroma formation and other cardiometabolic diseases [22,27]. Hence, fundamental prevention strategies should focus on reducing MetS and cardiometabolic risks through lifestyle modification, while considering sex differences [41]. For instance, we should empower male adults to adopt a healthier diet, with sufficient servings of vegetables and fruit, and female adults to perform regular exercise.
The current prevalence of HCV infection in Taiwan is approximately 4% [42]. A high prevalence of viral hepatitis was noted in the present study. This may be related to an incident which happened a few decades ago, where blood-borne virus transmission was caused by inadequate sterilization of medical equipment, or infection by unscreened blood products in Taiwan, specifically in the southwestern coastal area [42,43]. Previous studies have indicated that oxidative stress, viral hepatitis, and liver cancer may be interconnected [23,29]. HCV has evolved to manipulate pro-and antioxidant balance and generate sustainable oxidative stress that not only causes hepatic damage but also stimulates processes that reduce damage treatment. Lipid metabolism pathways present a clear danger to the accumulation of viral-induced ROS. These disorders tend to lead to cirrhosis and hepatocellular carcinoma (HCC), but there is little real-world data confirmation. Our large-scale data verified this theory and demonstrated that viral hepatitis carriers are prone to higher levels of 1-OHP and MDA, especially chronic hepatitis C carriers. Oxidative stress may trigger HCC in patients with chronic viral hepatitis [20,30]. Therefore, oxidative biomarker assessment may provide us with a better understanding of disease history and progress beyond traditional laboratory studies. Fortunately, the WHO has launched a global health strategy to eliminate hepatitis C by 2030 [30,43], and the Taiwanese government has implemented an elimination goal, with free antiviral treatment by National Health Insurance [42,44]. Recently, the emergence of direct-acting antivirals, such as asunaprevir/daclatasvir, has increased cure rates and treatment safety, with more than 95% efficiency, few adverse effects, and a short treatment time of 8-12 weeks [44,45]. Many adults near the PICs with chronic viral hepatitis do not receive the much-needed available information [32]. Therefore, it is crucial for us to recommend both HBV and HCV carriers for further antiviral treatment.
Compared with females, male sex was significantly associated with a higher prevalence of smoking, alcohol consumption, and betel nut chewing, and tended to also be associated with more regular exercise but improper diet habits, in agreement with published findings [31,33]. The present findings showed that the prevalence of current or former female smokers was only 4.5%, and this rate was lower than that of males (40.5%). However, the female sex was associated with significantly higher levels of urinary 1-OHP. This result is consistent with those of previous studies [8,11,12]. This could be due to the amount of time women spend in the kitchen, and cooking methods with frying and high temperatures can produce cooking oil fumes. In addition, poorly ventilated kitchens may increase fume inhalation [13,41]. The 1-OHP is a sensitive biomarker that reflects indoor air pollutant exposure [15]. Exposure to cooking oil fumes can lead to disturbed cardiovascular autonomic function and an increased risk of oxidative DNA injury [5]. Fur-thermore, numerous studies have reported an increase in the incidence of cardiovascular and respiratory diseases among women and children who lived around smokers [46,47]. As a cohort study in the United States, Gearhart-Serna et al. [36] indicated that smoking and secondhand smoke exposure accounted for the largest PAH intake, and that there were strong interactions between race/ethnicity and smoking or second-hand smoke, especially in vulnerable populations for high PAH exposure. From the perspectives of human and environmental health, it is necessary to conduct a tailored health promotion program, including strategies to reduce cooking oil fumes by improving ventilation devices and methods to avoid secondhand smoke. In addition, it is time to initiate more aggressive actions in the smoking cessation program and regulation for the vulnerable group near the PICs.
There are some limitations to this study. First, the limited geographical scope might limit the generalizability of the findings. Second, the health-related behavior assessments were mostly self-reported and were not measured more precisely, for example, water or vegetable intake, although we demonstrated servings with a standardized container. Third, due to the lack of information on the history of medications used for hypertension or diabetes, an underestimation of the prevalence of cardiometabolic risks might have occurred.

Conclusions
This study used large-scale data to explore the association between the biomarkers of environmental exposure, oxidative stress, and health-related behaviors in community residents living in an endemic area of viral hepatitis and nearby large PICs. A high prevalence of cardiometabolic risk factors and chronic viral hepatitis was found in this population. Furthermore, smoking, betel nut use, HBsAg, and anti-HCV positivity were associated with higher urinary 1-OHP and MDA levels. The findings highlight the importance of initiating individualized health promotion programs for adult residents, including cardiometabolic risk prevention, and smoking and betel-nut cessation, and regulations to avoid secondhand smoke. In addition, there is an urgent need to provide antiviral treatment to residents with chronic hepatitis. Institutional Review Board Statement: Ethical approval was obtained from the Institutional Review Board (IRB NO: 201800428B0) at the Chang Gung Memorial Hospital Foundation, Taiwan. All participants gave informed written consent to participate and were informed that they were free to withdraw from the study at any time.

Data Availability Statement:
The individual-level data used and/or analyzed for the current study are available from the corresponding author on request.