Mediation Analysis Of Insulin Resistance Between Refined Carbohydrate Intake And The Prevalence Of Non-alcoholic Fatty Liver Disease

Objective The prevalence of Non-alcoholic Fatty Liver Disease（NAFLD）is increasing in China.Cereals have been consumed as the main staple food in China since ancient times,and with the continuous economic development,fruit juices,cookies,cakes and milk tea have gradually become popular in China,and the intake of refined carbohydrates has been increasing.However,the relationship between refined carbohydrate intake and NAFLD is not very clear.Based on the assessment of the distribution characteristics of NAFLD and refined carbohydrate intake status,we analyzed the association between refined carbohydrate intake and NAFLD and explored the mediating role of insulin resistance（IR）in the association to provide guidance for the prevention of NAFLD.Methods A cohort was established in collaboration with a community hospital in Yaohai District,Anhui Province for participating workers between November 2019 and January2020 to conduct questionnaires,physical examinations,and laboratory tests.The diagnosis of NAFLD was determined by a combination of liver ultrasound and alcohol consumption;diet-related information was collected primarily through a validated food frequency questionnaire.Refined carbohydrate intake was assessed using a validated food frequency questionnaire（FFQ）.Univariate and multifactorial logistic regression models were used to analyze the relationship between refined carbohydrate intake and the risk of NAFLD prevalence,with a stratified analysis by gender;restricted cubic splines were used to fit the dose-response relationship between refined carbohydrate intake and the risk of NAFLD prevalence,and a sequential test was used to analyze the role of IR in refined carbohydrate intake and NAFLD prevalence The mediating effect of IR in the relationship between refined carbohydrate intake and the risk of NAFLD was analyzed using a restricted cubic strip.Results A total of 673 individuals were included in this study,of whom 456（67.8%）were men and 217（32.2%）were women;the mean age was 48.47±5.02 years.There were 371 study subjects with NAFLD,accounting for 55.1%of the total population.Univariate logistic regression analysis showed that intake and GL of total carbohydrates,refined carbohydrates,refined grains,and white rice were associated with an increased risk of NAFLD,and intake of whole grains was associated with a decreased risk of NAFLD（P<0.05）.Compared with the lowest quartile,the OR（95%CI）for NAFLD in the highest quartile of total carbohydrate,refined carbohydrate,refined grains,and white rice intake and GL values were 1.94（1.26-2.99）,1.88（2.94-4.59）,2.92（1.87-4.57）,1.58（1.00-2.48）,and 1.90（1.23-2.93）,with an OR（95%CI）of 0.43（0.28-0.67）for whole grains,all with a P_trend<0.05.After multifactorial adjustment,high refined carbohydrate intake(OR=1.09,95%CI:0.49-2.43,P_trend=0.032),high refined grain intake(OR=2.06,95%CI:0.98-4.31,P_trend=0.037)elevated the risk of NAFLD,while whole grains(OR=0.41,95%CI:0.23-0.72,P_trend=0.008)reduced the risk of NAFLD.After performing gender stratification,among men,the highest quartile group of whole grain intake after multifactorial adjustment was associated with a reduced risk of NAFLD(OR=0.48,95%CI:0.24-0.94,P_trend=0.047).In contrast,no statistical association was found between different mass carbohydrate intake and NAFLD prevalence in women.Dose-response relationship analysis showed that increased intake of refined carbohydrates and refined grains were both associated with an increased risk of NAFLD prevalence(P_non-liner=0.730and 0.523,respectively,P_overall<0.001),and increased intake of whole grains was negatively associated with the risk of NAFLD prevalence(P_non-liner=0.617,P_overall<0.001).Analysis of intermediate effects showed that the direct effect of refined carbohydrate intake was 0.046（95%CI:0.012,0.08 P<0.001）and the indirect effect of IR was-0.001（95%CI:-0.009,0.01 P=0.7）for the risk of NAFLD,total effect=0.045（95%CI:0.008,0.08 P<0.001）,with a mediating percentage effect of-0.036,P=0.7.There was no mediating effect of IR between refined carbohydrate intake and risk of NAFLD prevalence.Conclusions Refined carbohydrate intake is associated with NAFLD prevalence in middle-aged and older adults in Yaohai District,Hefei,and that this association occurs mainly in men.In addition,refined grain intake was associated with an increased risk of NAFLD,and whole grain intake was associated with a decreased risk of NAFLD.the mediating role of IR in the association between refined carbohydrate intake and the risk of NAFLD needs to be further investigated.