Prevalence and Risk Factors of Metabolic-Associated Fatty Liver Disease in School-Age Children and Adolescents in Shenyang, China

Background: Metabolic-associated fatty liver disease (MAFLD) refers to abnormal accumulation of fat in the liver for metabolic dysfunction. With modern socioeconomic conditions and changes in dietary patterns, the prevalence of MAFLD is increasing, to the detrimental of children’s learning and quality of life. We investigated the prevalence and inuencing factors of metabolic-associated fatty liver disease (MAFLD) in school-aged children and adolescents in Shenyang, China. Methods: In 2019, we collected demographic, anthropometric and liver health assessments from a random sample of Shenyang’ s school-aged children (7-12 years old) and adolescents (13-18 years old). Experienced hepatologists used transient elastography to diagnose the presence of fatty liver in the students. A random subsample was selected to complete a questionnaire to explore the impact of lifestyle habits on fatty liver disease. Results: The overall prevalence of MAFLD in these students was 23.83%, with a non-signicant difference between children (22.73%) and adolescents (24.43%). The prevalence of MAFLD was signicantly higher among boys than among girls. Compared with non-overweight students, a signicantly higher proportion of the overweight group had fatty liver. Moreover, questionnaire responses on exercise habits, normal diet, and parental factors were associated with fatty liver. Conclusions: MAFLD is very prevalent in children and adolescents in Shenyang city. Due to the close relationship between MAFLD and obesity, lifestyle plays a major role in the occurrence of MAFLD. -

for 1.58% of Shenyang's total student population. The study was approved by the First A liated Hospital of China Medical University ([2020]2020-258-2).

Anthropometric measurement
We recorded grade, education, gender, and age from school record systems. Height, weight, waist circumference, and hip circumference were measured in the morning; students had been requested to fast the previous day until after the measurements. Body mass index (BMI) was calculated by weight (kg) divided by the square of height (m 2 ). We de ned normal weight as a BMI less than 24.0, overweight as a BMI of 24.0 to 27.9, and obesity as a BMI of 28.0 or higher. Waist-to-hip ratio (WHR) was calculated using waist circumference (cm) divided by hip circumference (cm). A WHR ≥0.9 (boys) or ≥0.85 (girls) was de ned as abdominal obesity.
Fatty liver examination Fatty liver examination was performed by two experienced hepatologists, using transient elastography (Fibroscan). Fibroscan is newly noninvasive device for liver examination. The detection is fast and accurate, especially in the detection of fatty liver. It can not only provide quantitative results of fat but also display information on liver stiffness. Based on manufacturer' s recommendations, the diagnosis of fatty liver was made according to the value of the resulting according to the controlled attenuation parameter (CAP) value [10] . CAP values were categorized as non-MAFLD (CAP value <238 dB/m), mild fatty liver (238 dB/m ≤ CAP value <259 dB/m), moderate fatty liver (259 dB/m ≤ CAP value <292 dB /m), and severe fatty liver (CAP value ≥292 dB/m).

Questionnaire
We randomly selected three classes from one elementary school, one middle school, and one high school from among the study schools; their parents completed lifestyle questionnaires created by the First A liated Hospital of China Medical University, with speci c versions for students and parents. The questionnaire included basic information on the respondent' s date of birth, physical activity levels, entertainment practices, diet, and learning condition. Parents were also asked about their opinions about MAFLD. The participants were divided into the fatty liver group and the non-fatty liver group according to the FAP results. The relationship between fatty liver disease and living habits was compared in children (7-12 years) and adolescents (13-18 years).

Statistical analysis
Students' results were analyzed by age, gender, grade level, overweight, obesity, and MAFLD. Age was generally used as a dichotomous variablechildren (7-12 years) and adolescents (13-18 years). Data for continuous variables are reported as the mean ± standard deviation and compared between groups using Student' s t-test. Data for categorical variables are expressed as percentages (%) or proportions and compared between groups using chi-square tests. The level of signi cance was set at p<0.05. According to the difference of univariate analysis, it will be included in multivariate logistic regressionAll analyses were performed using SPSS software (IBM SPSS Statistics for Windows, Version 23.0. IBM Corp., Armonk, NY, USA).

Basic information
The ow of the subjects was seen in Table 1. The average age of participants was 13.31 ± 2.98 years old (Table 2). A total of 310 students were diagnosed with MAFLD, for a prevalence of 23.83%. Looking separately by age group, 22.7% of children and 24.43% of the adolescents were diagnosed with MAFLD.

Risk factors for MAFLD
Results of the bivariate analyses showed no signi cant differences between the MAFLD groups in terms of age. However, in both age groups and the combined sample, anthropometric measures were signi cantly higher in the MAFLD group than in the non-MAFLD group (p<0.001 for all comparisons). Logistic regression analyses revealed height, weight, BMI, waist circumference, and hip circumference as signi cant risk factors for MAFLD in children and adolescents; results for age and learning stage were not signi cant. However, liver stiffness in the MAFLD group was signi cantly higher (p<0.001) than that in the non-MAFLD group; this result was found for adolescents and the combined sample but was not found for children (Tables 3 and 4).

BMI and Fatty liver
A total of 345 students of the 1,301 study subjects were overweight (BMI ≥24); 224 of the overweight students had varying degrees of fatty liver.
The prevalence of fatty liver among obese students was quite high-90.54% (134 of the 148 students, 95% CI: 0.858-0.953). Among them, the proportions of boys and girls with fatty liver were 93.33% (95% CI: 0.881-0.986) and 86.21% (95% CI: 0.771-0.954), respectively. These prevalence was not signi cantly different (p>0.05) ( Table 5). The prevalence of overweight, obesity, and abdominal obesity was signi cantly higher in the MAFLD group (p<0.001 for all comparisons) ( Table 3). After strati cation by sex and age, the prevalence of MAFLD in both boys and girls increased with BMI; this pattern was seen for both age groups ( Figure 1).

Grade level and MAFLD
There were 395 elementary school students, 427 middle school students, and 479 high school students in the study, among whom 91, 107, and 112, respectively, were diagnosed with fatty liver. The overall prevalence rate was 23.8%, with grade-level results of 23.04% (95% CI: 0.189-0.272), 25.06% (95% CI: 0.209-0.292), and 23.38% (95% CI: 0.196-0.272), respectively. These differences were not statistically signi cant (p>0.05). Most of the elementary school students with fatty liver had a mild level, accounting for 15.19% of their cases. The proportion of junior high school students with moderate fatty liver was the highest. Nonetheless, there was no signi cant difference in the prevalence of moderate fatty liver among the three grades (p=0.023). The prevalence of severe fatty liver increased signi cantly with educational level (p<0.001). There were no signi cant differences between junior and senior high school students in FAP or liver stiffness. However, both groups were signi cantly higher on these measures than were the elementary school students (p<0.001). In addition, the value of other anthropometric factors and clinical measurements increased signi cantly with improvement of academic level (p<0.001 for all comparisons) ( Table 6).

Sex and MAFLD
The 1,301 study subjects included 625 boys and 676 girls. Among the boys, 182 were diagnosed with fatty liver (29.12%; 95% CI: 0.255-0.327); the 128 girls with fatty liver represented a prevalence of 18.9% (95% CI: 0.160-0.219). The overall prevalence for boys was signi cantly higher than that of girls (p<0.001), which mainly re ected the differing prevalence of severe fatty liver. There was no signi cant difference in age between the two groups. However, the values of height, weight, BMI, waist circumference, hip circumference, WHR, FAP (p<0.001 for all comparisons), and liver stiffness (p<0.01) were signi cantly higher in boys ( Table 7). The overall prevalence of fatty liver increased with age; for instance, the prevalence of fatty liver in students younger than 8 years old was 2.38%, while the prevalence in students older than 17 years old was 24.76%. After strati cation by sex and age, only the 13-14 age group showed a signi cantly higher prevalence among boys than among girls (p<0.001). Similarly, after strati cation by gender and education, the prevalence of fatty liver in the elementary and middle schools, but not high schools, was signi cantly higher in boys than in girls (p<0.05) ( Table 8).

Questionnaire
The questionnaire was valid for 117 of the 123 students. Tables 9 and 10 present the comparisons of the questionnaire factors between students with and without fatty liver. The basic information of children is consistent with that shown in the epidemiological survey. Fathers of children in the MAFLD group were more likely to have an educational level of high school or lower (p<0.05). Regarding exercise habits, the proportion of students in the MAFLD group who did not exercise or who had poor grade was signi cantly higher. In terms of eating habits, the students in the MAFLD group ate breakfast less frequently than those in the non-MAFLD group (p<0.05) but had a higher frequency of consuming fast food and snacks (p<0.05). Furthermore, the ratio of eating carefully and slowly was higher in the non-MAFLD group. Parents of adolescents with MAFLD also had higher BMI. Similar to children, more students ate breakfast every day in non-MAFLD group (p<0.05). Parents of adolescents with MAFLD had less knowledge of nutrition and health. Finally, no signi cant differences were found between the two groups in basic conditions of birth, recreation, and learning (p>0.05).

Discussion
1. Many studies have investigated the prevalence of MAFLD. Jeffery conducted a retrospective survey of 742 children aged 2-19 years from 1993 to 2003, reporting an average prevalence of fatty liver of 9.6% [11] . A 15-year meta-analysis showed an average prevalence of 7.6% for fatty liver in children and adolescents (ages 1-19 years) [7] . The prevalence of fatty liver appears to differ, however, across global populations, including Asia (10.2%) [9] and Europe (2.5%) [12] . Jain (2019) also demonstrated this point in a recent dissertation [13] , and different countries also have different prevalence rates; for example, in Haryana, India, 22.4% of children aged 5-10 years had fatty liver [14] . The average prevalence of fatty liver in Chinese children has been estimated at 9.03% [15] . Our study was designed to produce results that are representative for children and adolescents in the city of Shenyang; the prevalence of 23.83% is high. This phenomenon maybe related to diet and lifestyle, with many children and adolescents in the area preferring sweets and diets heavy in meat. In addition, the weather in the area is generally cold, causing children and adolescents to spend more time indoors with limited opportunity for exercise.
2. Among overweight children, our estimate of 64.9% of is signi cantly higher than the prevalence in non-overweight children. A study conducted in Beijing, China, found that 174 of 387 obese children and adolescents had non-alcohol fatty liver (NAFLD), with a prevalence of 45.0% [16] . An earlier Chinese study also reported a prevalence of 65.9% among 308 obese children aged 9-14 [17] . These results are supported by those of our study, with the prevalence of fatty liver closely related to overweight, obesity, and abdominal obesity. It has been reported that fatty acids are mainly catabolized in the liver [18] . Hepatocytes convert triglycerides formed by fatty acid esteri cation into very low-density lipoprotein and release them. Obese children and adolescents have increased levels of fatty acids entering the liver, which, at some point, will exceed the liver's ability to metabolize them. This can cause the deposition of a large number of fat droplets in the liver, creating the likelihood of liver dysfunction and promoting the formation of fatty liver. In addition, free fatty acids also increase the release of in ammatory mediators and insulin resistance [19] , increasing the production of intrahepatic fat [20] . Among obese people, central obesity (abdominal obesity) mainly re ects the amount of visceral fat, which has more serious consequences for the liver than peripheral obesity has. In our research, non-overweight children and adolescents still had a 9.0% prevalence of fatty liver. Fatty liver in non-overweight people increases the risk of cardiac metabolic disease and accompanying serious consequences. On one hand, fatty liver may be related to genetic factors, and many genetic variations have been found. Polymorphisms in the PNPLA3 gene can increase liver fat content and the risk of MAFLD and are possibly related to the severity level of MAFLD [21] . The TM6SF2, MBOAT7, and GCKR genes all appear to affect the occurrence and development of MAFLD in children and adolescents [22] . On the other hand, fatty liver may be related to irregular lifestyles; excessive intake of fructose promotes the formation of MAFLD, although fructose does not necessarily cause obesity [23] . Studies in adults have found that elevated serum uric acid can increase the risk of MAFLD in lean people [24] . However, further exploration is needed to determine whether this conclusion is equally applicable to children and adolescents.
3. We found that boys had a higher prevalence of MAFLD than girls; this conclusion is the same as that drawn in most studies. Estrogen plays a major role in preventing the occurrence and progression of fatty liver [25] , and through a variety of mechanisms, it can prevent the accumulation of triglycerides in the liver and thus liver brosis. In our survey, boys had a higher prevalence in all age groups, with the difference signi cant for students aged 13 to 14. In China, this age coincides with the average time of menarche, during which the ovaries begin to produce estrogen. Moreover, according to our questionnaire, boys had higher BMI values than girls (Table 11). The proportion of obese boys, especially those with abdominal obesity, was relatively large compared to that of girls. In terms of diet, boys' questionnaire answers revealed a preference for sweet, high-fat, and high-calorie foods, all of which increase body fat content and the metabolic burden on the liver. Surprisingly, we found a signi cant difference between MAFLD groups on whether breast milk was exclusively consumed in the rst four months after birth. It is not clear whether breast milk consumption is related to the high prevalence among boys. However, the earlier addition of complementary foods increases the burden on the digestive system because of imperfect digestive function in newborns. 4. Some studies have reported that MAFLD can aggregate within families. When exploring the relationship between the prevalence of MAFLD and family living habits, we found several links between parental characteristics and fatty liver in their children. In particular, a high parental BMI was associated with an increased risk of fatty liver in children. Part of the reason may be an inherited tendency for obesity. However, parents' high BMI is related to living and eating habits which will largely be shared by their children. Second, students' MAFLD status was related to parental education levels. Parents with a higher level of education may have a relatively deeper understanding or awareness of health and diseases and may also have increased access to health care systems for the prevention and control of their children' s fatty liver and other diseases.
5. For children and adolescents, increasing exercise should be the rst prevention target as physical exercise is the most effective method to prevent and treat MAFLD. Physical exercise can reduce the formation of free fatty acids, inhibit insulin resistance, reduce the risk factors for MAFLD, and has even been shown to reverse the condition [26] . Our study also found that children with fatty liver appeared to engage in less exercise, which is consistent with the results of the study of adults conducted by Mansour-Ghanaei et al. However, we did not nd a relationship between exercise intensity and fatty liver disease [27] . It has been reported that aerobic exercise and resistance exercise can reduce steatosis of MAFLD; people with poor cardiopulmonary function are more suited to resistance exercise [28] .
6. Moreover, we should pay attention to eating habits and ensure a healthy diet. It is crucial to ingest appropriate types and amounts of fruits and vegetables. Vegetables and fruits contain a large number of vitamins, which have antioxidant properties and can reduce the risk of liver steatosis [29] . Children and adolescents must control the frequency of desserts and sweet beverages, as these foods contain a large amount of fructose, which can induce insulin resistance and promote fat deposition in the liver [30] . We found that snacks and fried foods are frequently consumed by children with fatty liver. These are high-calorie foods that are di cult to digest and absorb. They will, eventually, be converted into fat stored in various parts of the body, which will increase the burden on the liver. In addition to the points mentioned above, a regular breakfast has a major effect on the prevention of fatty liver. We have found that eating breakfast less frequently increases the risk of fatty liver in children and adolescents. One study have found that skipping breakfast can cause overweight and obesity [31] . This may be the reason why these students are prone to fatty liver. Furthermore, chewing as slowly as possible and eating carefully during meals can prevent excessive dietary intake and fat accumulation. At present, the relationship between sleep and MAFLD has not been determined. Some studies suggest that a short sleep time increases the risk of fatty liver, but we did not nd a correlation between sleep time and MAFLD. Finally, our research suggests that sharing of information publicly in communities and other places is effective in reducing the probability of fatty liver.
Widely available information can increase parents' general health knowledge and, speci cally, the harms of obesity. Strengthening the popularization of science is likely related to the occurrence of MAFLD in school-aged children and adolescents.

Conclusions
The prevalence of MAFLD among school-aged children and adolescents aged 7 to 18 in Shenyang, China is higher than the national average.

Consent for publication: Not applicable
Availability of data and materials: The datasets generated and/or analysed during the current study are not publicly available because the public data has not obtained the consent of the participants' guardian, but the data acquisition and analysis have obtained the consent of the participants' guardian. But they are available from the corresponding author on reasonable request.

Competing interests: The authors declare that they have no competing interests
Funding: This research did not receive any speci c grant from funding agencies in the public, commercial, or not-for-pro t sectors.
Authors' contributions: GL and XZ carried out drafting and revising the article, HT and XJ carried out acquisition of data, DL and NW carried out designing the experiment, JS and XW carried out analysis of data, ZZ carried out interpretation of data, YL carried out nal approval of the article.All authors read and approved the nal manuscript P value: comparison between non-fatty liver group and fatty liver group.  P value: comparison among primary school students, middle school students and high school studentsnon.