Environmental Exposure to Cooking Oil Fume and Fatty Liver Disease: a cross-sectional study.

Background The effect of cooking oil fume on development of fatty liver disease is limited. The present study aims to investigate the association between exposure to cooking oil fume and the risk of fatty liver disease. Method A total of 55959 participants aged between 40 and 75 years old participated in a community-based survey in Ningbo, China. Multiple logistic regression analyses were conducted to investigate the association between cooking oil fume exposure and fatty liver risk. Furthermore, ordered logistic regression was conducted to investigate the association between cooking oil fume exposure and the severity of fatty liver disease. Results Cooking oil fume exposure were significantly associated with fatty liver disease after adjusting for confounding factors compared with participants in the none oil fume exposure group. Moreover, interaction analyses indicated that females with heavy oil fume exposure had the highest odds ratios of fatty liver disease and severer disease extent. In the stratified analysis, compared to participants in the smokeless group, males and females in light, moderate and heavy cooking oil fume exposure groups all had significantly higher risk of fatty liver disease and severer disease extent, while participants with heavier cooking oil fume exposure tended to have higher risk of fatty liver disease and severer disease extent. Conclusion Our findings indicated that exposure to cooking oil fume potentiated the risk of fatty liver disease, and the associations might be dose-responsive. Furthermore, heavy oil fume exposure and female sex might have a synergistic effect on fatty liver disease.


Introduction
Fatty liver disease is characterized by fat deposits in liver cells and encompasses a broad array of liver pathology, ranging from asymptomatic steatosis to steatohepatitis, fibrosis, and cirrhosis [1]. It has placed a heavy economic burden on health care systems globally [1]. With changes of lifestyles, the prevalence of fatty liver in Asia area grows sharply in recent years, with the mean prevalence in China reaching 15% in 2009 [2]. Thus, it is important to seek related risk factors of fatty liver disease and effective strategies to prevent it.
Cooking is an essential part of daily life. Chinese-style cooking including stir-frying and deep-frying which need preheated about 25-100 ml of cooking oil to approximately 280 degrees Celsius and produce large quantities of cooking oil fume. [3]Cooking oil fume contains lots of carcinogens, such as polycyclic aromatic hydrocarbons (PAH), fine particulate matter (PM), etc. [4,5]It is reported that the annual emission rate of total PAH was 2038 kg/year in Chinese restaurants, which was approximately 8-fold higher than that at restaurants of western. [6]Another study reported that the proportion of PM2.5 emissions from cooking fuels was about 12% of total global PM2.5 emissions [7]. Several epidemiologic studies have related cooking oil fume to lung cancer and cardiovascular disease. [8,9] Recently, Jian etc. found that environmental exposure to cooking oil fume is also associated with diabetes.
[10] However, few studies has analyzed the association between cooking oil fume and fatty liver although Ames test and the SOS chromotest have found that cooking oil fume contains genotoxicity which related to fatty deposition [11,12]. Therefore, the present study is conducted to investigate whether exposure to cooking oil fume is related to incident and severity of fatty liver disease.

Participants
We performed a cross-sectional study base on Ningbo database of the Cancer Screening Program in Urban China (CanSPUC) [13]. CanSPUC is a ongoing national cancer screening program which was initiated in October 2012. Participants aged from 40 -75 years old and resided in Ningbo city, China for more than three years were recruited. Subjects with malignant tumors or other serious diseases under treatment were excepted. After obtaining signed written informed consent, all the eligible participants were interviewed face-to-face by trained staffs to collect information about their exposure to risk factors.

Ethics approval and consent to participate
This study was approved by the Ethics Committee of HwaMei Hospital, University of Chinese Academy of Sciences.

Declarations
We confirmed that all methods were performed in accordance with the relevant guidelines and regulations.

Questionnaires
Fatty liver disease, environmental exposure to cooking oil fume, gender, age, menopausal status, weight, height, waist, education, active and passive smoke, alcohol drinking, dietary fat intake, physical activity, and metabolic disease (hypertension, hyperlipemia and type 2 diabetes) were collected through face-to-face interview. Participants who had self-reported fatty liver disease were further grouped as light fatty liver, moderate fatty liver and heavy fatty liver by asking "how severer is the disease?". Cooking oil fume exposure was leveled as smokeless, light, moderate and heavy. Body mass index (BMI) is calculated by dividing body weight in kilograms by the square of height in meters.
Education was classified as illiteracy, primary or middle school, high school and college or above. Information on active smoking was obtained by asking current and lifetime smoking habits and classed as smoker (current or former smoker) or nonsmoker. Passive smoking was defined by asking "Are there smokers living in the participants' family or at their workplace". Alcohol drinking was classified as never, regular and quit. Dietary fat intake was leveled as high fat diet, moderate fat diet and low-fat diet. Regular exercise was defined according to whether participants do exercises at least 3 times a week and 30 minutes every time.

Statistical analysis
The characteristics of the participants were summarized as mean ± standard deviation (SD) for continuous variables and as frequency (percentages) for the categorical variables.
Analyses of variance (ANOVA) and Bonferroni test were used to compare continuous variables, and chi-squared tests were used to compare categorical variables across different cooking oil fume exposure groups.
Multiple logistic regression analyses were conducted to investigate the association between cooking oil fume exposure and fatty liver risk after adjusting for gender, age, BMI, waist, education, active and passive smoke, alcohol drinking, dietary fat intake, regular exercise and metabolic disease.
Interactive analysis was used to evaluate the interaction between cooking oil fume exposure and gender. And then participants were divided into two groups as significant interaction was observed. Multiple logistic regression models were respectively rerun in males and females, and models in females were additionally adjusted for menopausal status. Furthermore, multiple ordinal logistic regression analyses were used to examine the association between cooking oil fume exposure and the severity of fatty liver disease with same confounders adjusted in the multiple logistic regression analyses (p value of proportional odds assumption<0.001). A value of p < 0.05 (two sided) was considered statistically significant unless otherwise indicated. Stata version 13 was used for data analyses (Stata Corporation, College Station, TX).

Results
The characteristics of participants are shown in Table 1. Participants who were exposed to heavier cooking oil fume were more likely to develop fatty liver disease (all p < 0.001).
Significant differences were found in waist, education, alcohol use, active and passive smoke, dietary fat intake, physical activity (all p< 0.001). Participants in the heavy cooking fume exposure group had the highest risk of metabolic disease (hypertension, dyslipidemia and type 2 diabetes, all p< 0.05). There were no significant differences in age, menopausal status and waist across cooking oil fume exposure groups (p > 0.05).  In the interactive analysis, we observed significant interaction between cooking oil Thus, models were rerun with participants being divided into two groups according to gender (Table 3). Compared to participants in the smokeless group, males and females in light, moderate and heavy cooking oil fume exposure groups all had significantly higher risk of fatty liver disease, while participants with heavier cooking oil fume exposure tended to have higher risk of fatty liver disease (all p<0.001).

Discussion
In the present study, we found a positive association between cooking oil fume exposure and fatty liver disease after adjusting for various potential confounders in Chinese adults.
Participants who exposure to heavier cooking oil fume tended to have higher risk of fatty liver disease and severer disease extent. In addition, we observed significant interaction between cooking oil fume exposure and gender on the risk of fatty liver disease. Female with heavy cooking oil fume exposure was associated with the highest observed odds ratio of fatty liver disease and severer disease extent.
Cooking fume exposure has been found to be related with many chronic diseases such as lung cancer, cardiovascular disease and diabetes.[8-10] Previous animal studies also found that carcinogens contained in cigarette smoke, which can also be found in cooking oil fume, might accelerate fat deposition in liver.
[14] However, clinical data on the association between environmental exposure to cooking oil fume and fatty liver disease is limited. In the present study, we firstly found that cooking oil fume is significantly associated with fatty liver disease and severer disease extent even after adjusting for possible confounders. In addition, participants with heavier cooking oil fume exposure tended to have higher odds ratios of fatty liver disease and severer disease extent, suggesting a dose-response relationship between cooking oil fume exposure and fatty liver. However, the mechanism underlying the association is still unclear. In the interaction analyses, we found a synergistic effect between heavy cooking oil fume exposure and female sex on fatty liver disease and severer disease extent. Female with heavy cooking oil fume exposure was associated with highest risk for fatty liver disease and severest disease extent, suggesting female were more likely to be affected by cooking The present study has several strengths. Our study firstly related cooking oil fume to fatty liver disease and found a dose-response relationship between cooking oil fume exposure and fatty liver. Secondly, the sample size was relatively large, making the statistical significance being robust. Thirdly, a series of important confounders, such as dietary fat intake, social economic factors and metabolic diseases, were taken into consideration in the analysis. Lastly, collapsing environmental exposure of cooking oil fume, severity of fatty liver into four categories separately enabled us to have a better understanding on the associations between cooking oil fume and fatty liver.
Conversely, limitations should also be noted. First, because of the cross-sectional design, a causal relationship between cooking oil fume and fatty liver may not be inferred.
Second, the measurement of cooking oil fume exposure and fatty liver were based on selfreport only, which may lead to recall bias and misclassification. Prospective and longitudinal studies should be made in the future with more accurate information on cooking oil fume exposure and fatty liver. Third, participants with nonalcoholic fatty liver disease cannot be distinguished from alcoholic fatty liver disease in our study. However, a sensitive analyze included only nondrinker observed similar results (table S1, table S2).

Conclusion
In conclusion, our study found that there may be a dose-dependent association between cooking oil fume exposure and fatty liver disease. In addition, heavy cooking oil fume exposure might have a synergistic effect with female sex on fatty liver disease. These findings add to the current understanding regarding fatty liver prevention. Future studies should be conducted to explore the underlying mechanism of this association and whether effective cooking oil fume control means like a fume extractor in home kitchen can reduce

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.