Predictive modeling the probability of suffering from metabolic syndrome using machine learning: A population-based study.

Background: There is an increasing trend of Metabolic syndrome (MetS) prevalence, which has been considered as an important contributor for cardiovascular disease (CVD), cancers and diabetes. However, there is often a long asymptomatic phase of MetS, resulting in not diagnosed and intervened so timely as needed. It would be very helpful to explore tools to predict the probability of suffering from MetS in daily life or routinely clinical practice. Objective: To develop models that predict individuals' probability of suffering from MetS timely with high efficacy in general population. Methods: The present study enrolled 8964 individuals aged 40-75 years without severe diseases, which was a part of the REACTION study from October 2011 to February 2012. We developed three prediction models for different scenarios in hospital (Model 1, 2) or at home (Model 3) based on LightGBM (LGBM) technique and corresponding logistic regression (LR) models were also constructed for comparison. Model 1 included variables of laboratory tests, lifestyles and anthropometric measurements while model 2 was built with components of MetS excluded based on model 1, and model 3 was constructed with blood biochemical indexes removed based on model 2. Additionally, we also investigated the strength of association between the predictive factors and MetS, as well as that between the predictors and each component of MetS. Results: In this study, 2714 (30.3%) participants suffer from MetS accordingly. The performances of the LGBM models in predicting the probability of suffering from MetS produced good results and were presented as follows: model 1 had an area under the curve (AUC) value of 0.993 while model 2 indicated an AUC value of 0.885. Model 3 had an AUC value of 0.859, which is close to that of model 2. The AUC values of LR model 1 and 2 for the scenario in hospital and model 3 at home were 0.938, 0.839 and 0.820 respectively, which seemed lower than that of their corresponding machine learning models, respectively. In both LGBM and logistic models, gender, height and resting pulse rate (RPR) were predictors for MetS. Women had higher risk of MetS than men (OR 8.84, CI: 6.70-11.66), and each 1-cm increase in height indicated 3.8% higher risk of suffering from MetS in people over 58 years, whereas each 1- Beat Per Minute (bpm) increase in RPR showed 1.0% higher risk in individuals younger than 62 years. Conclusion: The present study showed that the prediction models developed by machine learning demonstrated effective in evaluating the probability of suffering from MetS, and presented prominent predicting efficacies and accuracies. Additionally, we found that women showed a higher risk of MetS than men, and height in individuals over 58 years was important factor in predicting the probability of suffering from MetS while RPR was of vital importance in people aged 40-62 years.

Not performing an OGTT results in underdiagnosis, inadequate risk assessment and probable cost increases of (pre)diabetes in Han Chinese over 40 years: a population-based prospective cohort study.

OBJECTIVE: To explore the influence by not performing an oral glucose tolerance test (OGTT) in Han Chinese over 40 years. DESIGN: Overall, 6682 participants were included in the prospective cohort study and were followed up for 3 years. METHODS: Fasting plasma glucose (FPG), 2-h post-load plasma glucose (2h-PG), FPG and 2h-PG (OGTT), and HbA1c testing using World Health Organization (WHO) or American Diabetes Association (ADA) criteria were employed for strategy analysis. RESULTS: The prevalence of diabetes is 12.4% (95% CI: 11.6-13.3), while the prevalence of prediabetes is 34.1% (95% CI: 32.9-35.3) and 56.5% (95% CI: 55.2-57.8) using WHO and ADA criteria, respectively. 2h-PG determined more diabetes individuals than FPG and HbA1c. The testing cost per true positive case of OGTT is close to FPG and less than 2h-PG or HbA1c. FPG, 2h-PG and HbA1c strategies would increase costs from complications for false-positive (FP) or false-negative (FN) results compared with OGTT. Moreover, the least individuals identified as normal by OGTT at baseline developed (pre)diabetes, and the most prediabetes individuals identified by HbA1c or FPG using ADA criteria developed diabetes. CONCLUSIONS: The prevalence of isolated impaired glucose tolerance and isolated 2-h post-load diabetes were high, and the majority of individuals with (pre)diabetes were undetected in Chinese Han population. Not performing an OGTT results in underdiagnosis, inadequate developing risk assessment and probable cost increases of (pre)diabetes in Han Chinese over 40 years and great consideration should be given to OGTT in detecting (pre)diabetes in this population. Further population-based prospective cohort study of longer-term effects is necessary to investigate the risk assessment and cost of (pre)diabetes.

Non-high-density lipoprotein cholesterol: High-density lipoprotein cholesterol ratio is an independent risk factor for diabetes mellitus: Results from a population-based cohort study.

BACKGROUND: Dyslipidemia predicts the development and progression of diabetes. A higher non-high-density lipoprotein cholesterol (HDL-C): HDL-C ratio is reportedly associated with metabolic syndrome and insulin resistance, but its relationship with glycemic levels and diabetes remains unclear. METHODS: In all, 4882 subjects aged ≥40 years without diabetes and not using lipid-lowering drugs were enrolled in the study. The non-HDL-C: HDL-C ratio was log10 transformed to achieve normal distribution. Multivariate logistic regression was used to investigate the association between the log10 -transformed non-HDL-C: HDL-C ratio and diabetes. Stratified analyses of the association by age, gender, and body mass index (BMI) were also performed. RESULTS: After 3 years of follow-up, 704 participants developed diabetes. After adjustment for age, gender, current smoking, current drinking, physical activity, BMI, systolic blood pressure, and family history of diabetes, each 1-SD increase in the log(non-HDL-C: HDL-C ratio) was associated with higher fasting blood glucose (FPG) levels (ß = 0.1; 95% confidence interval [CI] 0.1-0.1), 2-h postload plasma glucose levels (2-h glucose; ß = 0.2; 95% CI 0.1-0.2), and risk of diabetes (odds ratio [OR] 1.1; 95% CI 1.0-1.2). In a multivariate model, subjects in the top quartile of non-HDL-C: HDL-C ratio had higher FPG (ß = 0.2; 95% CI 0.2-0.3), 2-h glucose (ß = 0.5; 95% CI 0.3-0.7) and HbA1c (ß = 0.1; 95% CI 0.1-0.2) levels, and a 40% increased risk of diabetes (OR 1.4; 95% CI 1.1-1.8) than participants in the bottom quartile. CONCLUSIONS: The non-HDL-C: HDL-C ratio was found to be an independent risk factor for diabetes.

Genetic Regulation of the Thymic Stromal Lymphopoietin (TSLP)/TSLP Receptor (TSLPR) Gene Expression and Influence of Epistatic Interactions Between IL-33 and the TSLP/TSLPR Axis on Risk of Coronary Artery Disease.

The thymic stromal lymphopoietin (TSLP)/TSLP receptor (TSLPR) axis is involved in multiple inflammatory immune diseases, including coronary artery disease (CAD). To explore the causal relationship between this axis and CAD, we performed a three-stage case-control association analysis with 3,628 CAD cases and 3,776 controls using common variants in the genes TSLP, interleukin 7 receptor (IL7R), and TSLPR. Three common variants in the TSLP/TSLPR axis were significantly associated with CAD in a Chinese Han population [rs3806933T in TSLP, Padj = 4.35 × 10-5, odds ratio (OR) = 1.18; rs6897932T in IL7R, Padj = 1.13 × 10-7, OR = 1.31; g.19646A>GA in TSLPR, Padj = 2.04 × 10-6, OR = 1.20]. Reporter gene analysis demonstrated that rs3806933 and rs6897932 could influence TSLP and IL7R expression, respectively. Furthermore, the "T" allele of rs3806933 might increase plasma TSLP levels (R2 = 0.175, P < 0.01). In a stepwise procedure, the risk for CAD increased by nearly fivefold compared with the maximum effect of any single variant (Padj = 6.99 × 10-4, OR = 4.85). In addition, the epistatic interaction between TSLP and IL33 produced a nearly threefold increase in the risk of CAD in the combined model of rs3806933TT-rs7025417TT (Padj = 3.67 × 10-4, OR = 2.98). Our study illustrates that the TSLP/TSLPR axis might be involved in the pathogenesis of CAD through upregulation of mRNA or protein expression of the referenced genes and might have additive effects on the CAD risk when combined with IL-33 signaling.