Abstract
BackgroundPostpartum cardiovascular disease (CVD) following hypertensive disorders of pregnancy (HDP) involves multifactorial mechanisms linking metabolic susceptibility to vascular damage. Identifying predictive biomarkers and modifiable risk factors is crucial for early stratification in the context of predictive, preventive, and personalized medicine (PPPM). We aimed to investigate the association of triglyceride-glucose (TyG) indices with incident CVD in women with prior HDP and to assess the causal relationship using Bayesian weighted Mendelian randomization (BWMR) analysis, with a focus on the implications for PPPM strategies in postpartum management.MethodsIn this longitudinal cohort study, 1,642 women with prior HDP from the UK Biobank were followed for a median of 14.05 years postpartum. TyG and modified TyG indices were assessed at enrollment. Associations with incident CVD were evaluated using Cox proportional hazards models and restricted cubic spline regression, while predictive performance was assessed with Harrell's C-index. BWMR was further used to validate causal relationships.ResultsDuring follow-up, 466 (28.4%) women developed postpartum CVD (mean age 51.05 ± 8.28 years). Elevated baseline TyG indices were associated with an increased risk of CVD. The highest TyG tertile had an adjusted hazard ratio (HR) of 1.35 for CVD (95% CI, 1.04-1.74; P = 0.023), with similar associations for modified TyG indices (adjusted HRs for the highest tertile ranging from 1.31 to 1.39 [P < 0.05]). C-indices were around 0.6. The stratified analysis showed more pronounced associations in younger, normotensive, and non-obese subgroups. The BWMR analysis supported a significant causal effect of genetically predicted TyG levels on CVD risk (P < 0.05), confirming the observational findings.ConclusionsElevated TyG indices are significantly associated with increased postpartum CVD risk in women with prior HDP, particularly in subgroups without overt clinical risk factors. Integration of observational and genetic evidence supports a causal role of insulin resistance in this multifactorial pathology. TyG indices may serve as promising predictive biomarkers for early risk stratification and potential targets for personalized metabolic interventions, in line with PPPM strategies. Our findings highlight the importance of monitoring metabolic trajectories for effective PPPM-guided postpartum CVD prevention and management.</p>