Gender differences in the relationships between dietary phytosterols intake and prevalence of obesity in Chinese population

Abstract To investigate the associations between different phytosterols (PSs) intake and subtype of obesity in Chinese. Total 6073 adults aged ≥18 years was enrolled from China. General characteristics were completed by the validated dietary questionnaire. For total phytosterols intake, comparing Q4 with Q1 was inversely associated with the risks of overweight [odds ratio (OR) 95% confidence interval (CI), 0.82 (0.69, 0.96), p < .05]. The intake of stigmasterol, β‐sitosterol, β‐sitostanol and campestanol were associated with the lower risks of obesity, whereas no significant correlationss were found between campesterol intake and any subtype of obesity in the multivariable‐adjusted model. Interestingly, the stigmasterol intake was inversely related with the prevalence of central obesity in female, while the β‐sitostanol intake was found in male [OR 95% CI in Q3 of 0.78 (0.60–0.99) and 0.71 (0.56–0.91), respectively; p < .05]. The multiple linear regression models showed that fruits, vegetable‐oil, nuts and seeds may be important diet sources of PSs. The intake of total PSs, β‐sitosterol, stigmasterol, β‐sitostanol and campestanol were inversely associated with the prevalence of obesity. Moreover, the lower obesity risk for total PSs and PSs subgroups differed for the gender. The firm results deserve to be further verified in cohort studies.

pay attention to the associations between daily diet and NCDs, and raise the healthy and reasonable dietary recommendation.
The phytosterols (PSs, including plant sterols and plant stanols) are the essential non-nutritive but bioactive components in plantbased foods, the structures of which are 28-or 29-carbon alcohols with similar to cholesterol (Ghaedi et al., 2019). According to different food sources, it is mainly divided into β-sitosterol, stigmasterol and campesterol, with β-sitosterol as the dominant component, accounting for 60%-90% of the total PSs content .
Depending on the saturation state of the C-5 bond, PSs are divided into sterols and stanols. The β-sitostanol and campestanol are structures with saturated double bond of β-sitosterol and campesterol, respectively. Some studies reported that PS mixture intake has no effect on the susceptibility to obesity in mice model . However, the majority of evidences have suggested that PSs supplementation can significantly decrease triglyceride (TG), total cholesterol (TC) levels, fat absorption, and increase fecal lipid excretion as well thereby reducing body weight (Ghaedi et al., 2019). Recently, the meta-analysis of randomized controlled trials revealed that body mass index (BMI) significantly decreased following PSs supplementation (Ghaedi et al., 2019). Other possible anti-obesity effects of PSs could be by improving insulin sensitivity, increasing adiponectin levels and by modulating AMP-activated protein kinase (AMPK) activation and phospholipase A2 inhibition (Ghaedi et al., 2019;Xiong et al., 2018). The clinical trials found that 4 weeks PSs supplementation (1.6 g/day) could prevent obesityassociated inflammation by suppressing the expression of the C-C motif chemokine 2, monocyte chemotactic protein-1 (MCP-1), and interleukin-10-R and (Lambert et al., 2017) and decreasing circulating IL-6 and TNFα (Kurano et al., 2018). Particularly, β-sitosterol isolated from Moringa oleifera reduced the secretion of IL-6, IL-1β, IL-8 and TNFαin LPS-stimulated human keratinocytes and mouse macrophages (Liao et al., 2018). Another study reported the stigmasterol significantly reduced body weight comparing with mice fed high-fat diet at 17 weeks. However, -sitosterol was not as effective as stigmasterol . Stigmasterol decreases the intestinal bile acid levels that in turn decrease lipid absorption from diet, thereby attenuating weight gain (Feng, Gan, et al., 2018). Body weight, adipose expansion, and fatty liver changes are significantly suppressed with markedly elevated circulating sitosterol levels in mice (Kurano et al., 2018). In conclusion, although some studies have reported that PS mixture intake has no effect on obesity, some individual PS such as β-sitosterol, stigmasterol and sitosterol has impact on preventing obesity and related diseases. Studies on anti-obesity properties of PSs are controversial. Further researches are warranted to clarify whether individual PS may be protective against obesity.
Our previous study conducted in Chinese population have shown beneficial effect of PSs intake on obesity . However, limited evidence is available for large Chinese populations regarding the relationships between PSs intake and obesity. Moreover, there is few evidence revealed some details such as PSs type and differences between genders were not considered in the effects of PSs intake on obesity. Therefore, the aim of this study to examine the associations between the intake of different types of PSs and the prevalence of obesity in northern Chinese population. We focused on the five most commonly PSs types (stigmasterol, β-sitosterol, campesterol, β-sitostanol and campestanol) in order to investigate the associations and gender differences between these and overall and central obesity. In addition, gender difference was also explored.

| Study population
We used data from the validated internet-based dietary questionnaire for Chinese (IDQC, www.yyjy3 65.org/diet;Feng et al., 2016). This cross-sectional study included 6994 adults aged ≥18 years, were enrolled for this study from March 2014 to December 2016 in northern Chinese area. The sample size (n) was calculated as n = t α 2 × p × q/e 2 , where p is the percentage of the study characteristic, q is calculated as 1 − p, e is the accepted margin of error (set at 0.05), and t α = 1.96 (95% CI) (Rodriguez Del Aguila & Gonzalez-Ramirez, 2014). The exclusive criteria were as follows: (1) incomplete information on the IDQC (n = 526); (2) extreme daily energy intake (<600 kcal/day for all or >4000 kcal/day for females and >4200 kcal/day for males) (n = 314); (3) pregnancy or menopause (n = 81). Finally, the sample of 6073 participants were available for the analysis.
This study was conducted with the approval by Ethics Committee of Harbin Medical University and signed informed consents from all participants, and in accordance with the ethical standards in the Declaration of Helsinki.  Han et al. (2007Han et al. ( , 2009). The database of PSs was established and then estimated PSs intake using the IDQC.

| Anthropometric measurements, the definition of overweight, obesity and central obesity
Bodyweight, height, and waist circumference (WC) of participants were measured by well-trained nurses. Participants were asked to wear light, thin clothing and no shoes. Bodyweight, height and WC were measured to the nearest 0.1 kg, 0.1 cm and 0.1 cm, respectively.
Body mass index (BMI) in kg/m 2 was calculated as body weight (kg) in kilograms divided by the square of the height (m). According to the 2006 Guidelines on Preservation and Control Overweight and Obesity in Chinese Adults classification , Overweight, obesity and central obesity were defined as 24 ≤ BMI < 28 kg/m 2 , BMI ≥ 28 kg/m 2 , WC ≥ 80 cm (for females) and WC ≥ 85 cm (for males), respectively.

| Other covariates
Other covariates involved age, gender, education, incomes, physical exercise and smoking status. These data were also gained by the IDQC. For instance, education was divided into college and postgraduate and above, senior high school, junior high school, primary school and below. Income was divided into <2000 Yuan, 2000-4000 Yuan, >4000 Yuan. Smoking was divided into >20 times/day, 10-20 times/day, 1-9 times/day, < once/day, quit smoking and never smoking. Quit smoking was defined abstinent for at least 30 days before the enrollment to the study.

| Statistical analysis
All continuous variables and categorical variables are presented as means ± SD and the percentage, separately. Chi-square test was used to compare the differences in frequency of categorical variables. One-way analysis of variance (one-way ANOVA) or Kruskal-Wallis test was used to compare the means of continuous variables. Multivariable logistic regression analyses were used to exploring relationships between dietary phytosterols intake and different types of obesity. Age, gender, education, incomes, physical exercise and smoking status were adjusted in models. All analyses were carried out with SPSS Statistics ver. 22.0 (IBM). The level of significance was set at p < .05.

| Participant characteristics and dietary PSs intake according to quartiles of total PSs intake
The characteristics and dietary PSs intake of participants were summarized in Table 1. Participants were classified into quartiles of total PSs intake. The lowest quartile of consumption (Q1) was considered as the reference category. The phytosterols intake was higher in females than male. Participants of Q4 tended to have higher levels of physical activity, education and income, but lower levels of alcohol intake and smoking versus the first quartile.
However, participants in Q4 were less likely to have obesity and central obesity (p < .05 for all).

| Odds ratios (OR) and 95% confidence intervals (CI) for different types of obesity according to total PSs intake
As shown in Table 2 Tables 3 and 4

| DISCUSS ION
We applied an internet-based dietary questionnaire to in a large northern Chinese population to investigate the associations between PSs intake and the prevalence of obesity incidence. Our findings indicated: (1) The intake of total PSs, β-sitosterol, stigmasterol, β-sitostanol, campestanol were associated with lower risks of obesity, whereas no significant associations were observed between TA B L E 1 General characteristics of participants according to quartiles of total PSs intake    (Vezza et al., 2020). However, no significant relationships were found between campesterol intake and obesity in the present study. Few data are available for the relationships between campesterol intake and obesity. Gaps in the evidence base a suggest that more research is urgently needed, especially relating to the associations between the intake of specific types of PSs and subtype of obesity.
There is a gender difference in the prevalence of obesity.
The global incidence of obesity is higher in female than in male in 141 countries in 2014 (Mauvais-Jarvis, 2017; NCD Risk Factor Collaboration (NCD-RisC), 2016). Interestingly, in recent decades, the prevalence of central obesity has increased more in female than in male in the USA (Mauvais-Jarvis, 2017). Therefore, the Multivariable logistic regression analyses were conducted to further investigate the gender differences of the relationships between PSs intake and the different subtypes of obesity. In gender-stratified analysis, the higher levels of campestanol were linked with the lower risk of overweight/obesity and overweight in female and male. However, the stigmasterol intake was associated with the prevalence of central obesity in female, while the β-sitostanol intake was found in male.
The recent majority of studies suggested that the prevalence of obesity might have influenced the positive effect of PSs intake on obesity. However, scarce data have investigated above associations differed between female and male. Further studies are warranted TA B L E 3 Odds ratios (ORs) and 95% confidence intervals (CIs) of obesity according to quartiles (Q1-Q4) of different PSs intake in female

TA B L E 5
The multiple linear regression in PSs to confirm these observational findings and explain the observed gender-specific differences.
The most important natural PSs dietary sources are nuts, vegetable oils, grains, and grain-derived products in human diet . Consistently, In the present study, our study showed the associations between fruits, vegetable-oil, nuts and seeds and plant sterols (R = .91, F = 4151.28, p < .001). The average daily intakes of total PSs usually range from 200 to 400 mg in general population and high daily intake of 500-1000 mg can be obtained by vegetarians (Ras & Trautwein, 2017). However, the bioavailability of PSs is very low, which only about 2%-6% can be absorbed through intestine and the left will be excluded into the feces. Most RCT studies give evidence that the health benefits are obtained at the high intake of PSs (around 2 g/day) , which cannot be achieved in habitual diets.
However, in most RCTs, the health benefits are observed at a high dose of phytosterols intake (about 2 g/day), which cannot be achieved in habitual diets. Further research is worth to investigate the safety and bioavailability issues of PSs and the derivatives in clinical trials.
To our knowledge, this study is the first to investigate the associations between different PSs intake and subtype of obesity.
Most importantly, we found total PSs and subgroups PSs intake were different for female and male in prevention of obesity risks.
Furthermore, the present study used multiple linear regression modeling for the dietary-related factors to investigate the potential mechanisms linking PSs intake and obesity. The results of the study should be interpreted with some limitations. First, the incidence of obesity was ascertained on the basis of self-reported information, therefore, information biases could be existed. In spite of recall bias is another limitation, clickable images were conducted for each food as a reference to assist in the estimation of food portions. Moreover, the cross-sectional nature of the data precludes conclusions about causality or its direction. Therefore, the results should be further drawn in further investigation.

| CON CLUS ION
In conclusion, the present study suggests that the intake of total PSs, stigmasterol, β-sitosterol, β-sitostanol and campestanol were inversely associated with the prevalence of obesity. In addition, lower obesity risk for total PSs and PSs subgroups differed for the gender.
These results deserve to be further verified in cohort studies.

ACK N OWLED G M ENTS
The work was supported by the grants from the National Natural Science Hospital Cultivation Fund of China (no: 04.03.19.153).

FU N D I N G I N FO R M ATI O N
Funding was supported by the National Natural Science Hospital Cultivation Fund of China (No. 04.03.19.153).

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data are available upon request from corresponding author.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all participants.