Insulin Resistance-Adiposity Burden and Incident Cardiovascular Disease Across Glycaemic Transitions in Prediabetes: Risk Stratification After Reversion to Normoglycaemia.

AIMS: To examine whether insulin resistance (IR), assessed by cumulative triglyceride-glucose (TyG)-derived indices and change patterns, stratifies incident cardiovascular disease (CVD) risk after glycaemic transition in adults with prediabetes. MATERIALS AND METHODS: Data were obtained from 1468 participants with prediabetes in the China Health and Retirement Longitudinal Study. Glycaemic transition and TyG-derived indices were assessed from Wave 1 to Wave 3. Indices included TyG, TyG-ABSI (a body shape index), TyG-BRI (body roundness index), TyG-CVAI (Chinese visceral adiposity index) and TyG-WWI (weight-adjusted waist index). Cumulative indices summarised measurements across Waves 1 and 3 and exploratory k-means clustering identified broad two-wave exposure patterns. Incident CVD was assessed from Waves 3 to 5. Cox models estimated hazard ratios (HRs) and 95% confidence intervals (CIs). RESULTS: During a median 5.0-year follow-up, 294 CVD events occurred. Compared with persistent prediabetes, reversion to normoglycaemia was not associated with lower CVD risk (HR 0.97, 95% CI 0.72-1.31), whereas progression to diabetes was associated with higher risk (HR 1.53, 95% CI 1.14-2.07). In joint analyses, reversion with low cumulative indices was not associated with increased CVD risk, whereas reversion with high indices was significantly associated with elevated risk for most indices (HRs 1.55-2.05). Among participants who reverted to normoglycaemia, high cumulative TyG-derived indices were significantly associated with higher CVD risk (adjusted HRs 1.89-2.52; all p < 0.05). Change-pattern analyses were concordant. CONCLUSIONS: Glycaemic transition status did not fully stratify subsequent CVD risk. Cumulative TyG-related measures further stratified CVD risk across glycaemic transition groups, including among participants who reverted to normoglycaemia.