糖尿病发病时间趋势及其与中国内脏脂肪指数的关系：前瞻性队列研究

BackgroundDiabetes mellitus is a global public health issue. Cross sectional studies have found that visceral fat is closely related to the prevalence of diabetes melliteswhile prospective cohort studies on the trend of onset time of diabetes mellitus and its correlation with Chinese visceral adiposity indexCVAIare scant. ObjectiveTo analyze the trend of onset time of diabetes mellitus and its correlation with CVAI in Pudong New Area residents by the prospective cohort study thus providing evidence for the scientific prevention and treatment of diabetes mellitus. MethodsThis was a prospective cohort study involving 5 236 residents from 12 townships and 35 village committees who participated in the chronic disease risk factor monitoring project in Pudong New Area in 2013. Baseline data were collectedincluding CVAIvisceral adiposity indexVAI body mass indexBMIwaist circumferenceWCwaist-to-hip ratioWHRwaist-to-height ratioWHtR body shape indexABSIand body adiposity indexBAI. Follow-up was conducted in 2016 and 2019. By the end of follow-up in October 2019the incidence of new onset of diabetes mellitus in this cohort was calculated through questionnaire surveylaboratory testingmedical system visit information and vital statistics information system. According to the baseline quartilethe CVAIVAIBMIWCWHRWHtRABSIand BAI of the included population were divided into Q1 to Q4 quartiles. The number of cases in CVAI Q1-Q4 groups was 1 3061 3071 307and 1 307respectively. The number of cases in VAI Q1-Q4 groups was 1 30013161 306and 1 306respectively. The number of cases in BMI Q1-Q4 groups was 1 3051 3021 312and 1 311respectively. The number of cases in WC Q1-Q4 groups was 1 3021 2731 287 and 1 367respectively. The number of cases in WHR Q1-Q4 groups was 1 1801 2031 332and 1 514respectively. The number of cases in WHtR Q1-Q4 groups was 1 1991 3931 400and 1 237respectively. The number of cases in ABSI Q1- Q4 groups was 13161 3021 302and 1 308 respectively. The number of cases in BAI Q1-Q4 groups was 1 3101 304 1 308and 1 307respectively. The multivariable COX regression was used to analyze the correlation of CVAI and other obesity indicators with the onset of diabetes mellitus. The predictive potential of CVAI and other obesity indicators in diabetes mellitus was assessed using Receiver Operator CharacteristicROCcurves. ResultsThe incidence density of diabetes mellitus in Pudong New Area was 33.55/1 000 person-years from 2013 to 2016and 23.25/1 000 person-years from 2016 to 2019. With agingthe total incidence density of diabetes mellitus showed an increasing trend2013-20162 =28.503Ptrend value < 0.0012016- 20192 =25.600Ptrend value < 0.001. By 2016the baseline CVAI quartile was positively correlated with the cumulative incidence of diabetes mellitusCVAI2 =131.865Ptrend value <0.001and the incidence densityCVAI2 =100.105 Ptrend value < 0.001. Cox regression analysis after adjusting for relevant confounders showed that compared with CVAI in Q1 the risk of diabetes mellitus in men with CVAI in Q4 increased by 79.4%HR=1.79495%CI=1.044-3.083P<0.05. Women had a 371.2% increased risk of diabetes mellitusHR=4.71295%CI=2.601-8.538P<0.05. ROC curve results showed that in predicting the incidence of male diabetesthe area under the ROC curveAUCof CVAI for male diabetes was 0.600 95%CI=0.561-0.640with the Youden index of 0.181and the cutoff value of 104.118. DeLong test showed that CVAI had the highest accuracy in predicting female diabetes mellitusAUC=0.699with the Youden index of 0.317and the optimal cutoff value of 104.609. ConclusionFrom 2013 to 2019the incidence density of diabetes mellitus increased with the increased age in Pudong New Area. Compared with other obesity indicatorsCVAI can be used as an indicator to predict the risk of diabetes mellitus.