ADA Scientific Sessions 2026 Abstract · Jun 5, 2026
Unraveling the Metabolic–Cardiovascular Disease Nexus: An Epidemiological Network Analysis Identifies Central Indicators
By CAI X
Introduction and Objective: The absence of widely accepted scoring systems to accurately gauge the comprehensive impacts of metabolic disorders on cardiovascular diseases (CVD) risk remains a significant challenge. This study developed a CVD assessment model based on metabolic clustering networks. Methods: The study (NCT07043166) included 4752 adults through three-stage stratified sampling in China, as a part of the STONE study. Adults aged 18-65 years who had resided in the selected community for more than 6 months and were free of severe diseases were included. 68 indices across 7 major dimensions of metabolic health were mapped using betweenness centrality and permutation testing, and logistic regression was combined with genetic algorithm, leading to the development of a novel CVD screening scoring system (CardioMet12). CardioMet12 was externally validated using NHANES population. Results: 1. Participants characteristics: A total of 4066 participants were enrolled and 922 met criteria for clinical or sub-clinical CVD (22.68 %).2. Key findings: 12 central indicators were identified because of statistical and clinical relevance to CVD and availability in routine examination to establish CardioMet12, including non-traditional factors such as total bilirubin and bone mineral density, whose betweeness centrality was markedly higher than in randomly permuted networks (p < 0.001).3. Model performance: CardioMet12 exhibited satisfactory screening performance on validation set, achieving an AUC of 0.782, with 0.731 sensitivity and 0.711 specificity. Individuals scoring ≥56 had a risk of CVD diagnosis >50% in the STONE and NHANES samples. Conclusion: Metabolic-related alterations influence CVD through multiple dimensions within systems biology. Based on analysis from Chinese and US epidemiological cohorts and network analysis, CardioMet12 emerges as a robust metric that considers the complex interplay among metabolic disorders, supporting a more proactive and comprehensive approach to CVD screening and evaluation. Disclosure: X. Cai: None. Funding: Shanghai Municipal Health Commission Grant (No.: GWIV-27.7 to YQ.S); Chongqing Overseas Support Program (to J.Z.); Chongqing Excellent Youth Project (No.: CSTB2024YCJH-KYXM0098 to J.Z.); General Project of National Natural Science Foundation of China (No.: 82170858, 82470935 to T.L.); “Dawn” Program of Shanghai Education Commission, China (No.: 23SG33, to T.L.).