Cardiovascular risk and metabolic characterization of metabolically healthy obesity classified by the newly proposed definition: a prospective study

Objective: The varying definition of metabolic health is responsible for the debate about whether metabolically healthy obesity (MHO) is associated with increased cardiovascular risk. An empirically derived definition of metabolic health has been newly proposed to identify individuals with obesity but without excess mortality risk. This study aimed to evaluate the association of MHO classified by the new definition with cardiovascular events and to reveal the metabolic characterization of MHO. Design and method: We classified 1331 Flemish people (mean age: 51.3 years; women: 50.3%) by their metabolic health status (systolic blood pressure < 130 mmHg, waist-to-hip ratio < 0.95 for women or 1.03 for men, and absence of antihypertensives and diabetes) and obesity status (body mass index > = 25 kg/m2). Plasma 44 metabolites were measured by nuclear magnetic resonance spectroscopy at baseline. Hazard ratios (HRs), odds ratios, and 95% confidence intervals (95% CI) were calculated using Cox proportional hazard regressions and logistic regressions, accounting for sex, age, sex, smoking, alcohol, and physical activity. Results: During 5.6 years of follow-up, 111 cardiovascular events and 33 diabetes incidents occurred. MHO was not associated with cardiovascular events (HR 1.61, 95% CI: 0.49–5.27) compared with metabolically healthy normal weight. The HRs were 3.19 (95% CI: 1.07–9.53) and 3.27 (95% CI: 1.15–9.32) for metabolically unhealthy normal weight and obesity, respectively. Individuals with metabolic health had a lower risk of diabetes (HR: 0.22, 95%CI: 0.06–0.77). Compared with metabolically unhealthy obesity, MHO was associated with eight metabolites, including six with positive association (2-aminobutyrate, 4-aminobutyrate, alanine, leucine, 4-hydroxybutyrate, and acetate) and two with negative association (valerate and lactate). Metabolites positively associated with MHO showed inverse associations with diabetes risk, while metabolites negatively associated with MHO showed positive associations with diabetes risk. The associations of these metabolites with MHO reversed their directions compared with metabolically healthy normal weight. None of the eight metabolites was associated with cardiovascular events. Conclusion: This new definition of metabolic health can be replicated to stratify cardiovascular risk for individuals with or without obesity. The metabolic characterization of MHO was not completely ‘healthy’, which might help to identify novel targets for intervention.