Obesity- and lipid-related indices as a predictor of obesity metabolic syndrome in a national cohort study
Jiaofeng Gui,Yuqing Li,Haiyang Liu,Lei-lei Guo,Jinlong Li,Yunxiao Lei,Xiaoping Li,Lu Sun,Liu Yang,Ting Yuan,Congzhi Wang,Dongmei Zhang,Huanhuan Wei,Jing Li,Mingming Liu,Ying Hua,Lin Zhang
DOI: https://doi.org/10.3389/fpubh.2023.1073824
IF: 5.2
2023-02-14
Frontiers in Public Health
Abstract:Objective Metabolic syndrome is a common condition among middle-aged and elderly people. Recent studies have reported the association between obesity- and lipid-related indices and metabolic syndrome, but whether those conditions could predict metabolic syndrome is still inconsistent in a few longitudinal studies. In our study, we aimed to predict metabolic syndrome by obesity- and lipid-related indices in middle-aged and elderly Chinese adults. Method A national cohort study that consisted of 3,640 adults (≥45 years) was conducted. A total of 13 obesity- and lipid-related indices, including body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR), conicity index (CI), visceral adiposity index (VAI), Chinese visceral adiposity index (CVAI), lipid accumulation product (LAP), a body shape index (ABSI), body roundness index (BRI), and triglyceride glucose index (TyG-index) and its correlation index (TyG-BMI, TyG-WC, and TyG-WHtR), were recorded. Metabolic syndrome (MetS) was defined based on the criteria of the National Cholesterol Education Program Adult Treatment Panel III (2005). Participants were categorized into two groups according to the different sex. Binary logistic regression analyses were used to evaluate the associations between the 13 obesity- and lipid-related indices and MetS. Receiver operating characteristic (ROC) curve studies were used to identify the best predictor of MetS. Results A total of 13 obesity- and lipid-related indices were independently associated with MetS risk, even after adjustment for age, sex, educational status, marital status, current residence, history of drinking, history of smoking, taking activities, having regular exercises, and chronic diseases. The ROC analysis revealed that the 12 obesity- and lipid-related indices included in the study were able to discriminate MetS [area under the ROC curves (AUC > 0.6, P < 0.05)] and ABSI was not able to discriminate MetS [area under the ROC curves (AUC < 0.6, P > 0.05)]. The AUC of TyG-BMI was the highest in men, and that of CVAI was the highest in women. The cutoff values for men and women were 187.919 and 86.785, respectively. The AUCs of TyG-BMI, CVAI, TyG-WC, LAP, TyG-WHtR, BMI, WC, WHtR, BRI, VAI, TyG index, CI, and ABSI were 0.755, 0.752, 0.749, 0.745, 0.735, 0.732, 0.730, 0.710, 0.710, 0.674, 0.646, 0.622, and 0.537 for men, respectively. The AUCs of CVAI, LAP, TyG-WC, TyG-WHtR, TyG-BMI, WC, WHtR, BRI, BMI, VAI, TyG-index, CI, and ABSI were 0.687, 0.674, 0.674, 0.663, 0.656, 0.654, 0.645, 0.645, 0.638, 0.632, 0.607, 0.596, and 0.543 for women, respectively. The AUC value for WHtR was equal to that for BRI in predicting MetS. The AUC value for LAP was equal to that for TyG-WC in predicting MetS for women. Conclusion Among middle-aged and older adults, all obesity- and lipid-related indices, except ABSI, were able to predict MetS. In addition, in men, TyG-BMI is the best indicator to indicate MetS, and in women, CVAI is considered the best hand to indicate MetS. At the same time, TyG-BMI, TyG-WC, and TyG-WHtR performed better than BMI, WC, and WHtR in predicting MetS in both men and women. Therefore, the lipid-related index outperforms the obesity-related index in predicting MetS. In addition to CVAI, LAP showed a good predictive correlation, even more closely than lipid-related factors in predicting MetS in women. It is worth noting that ABSI performed poorly, was not statistically significant in either men or women, and was not predictive of MetS.
public, environmental & occupational health