BackgroundPrevious studies has shown a significant relationship between baseline triglyceride-glucose (TyG) index and subsequent cardiovascular disease (CVD). However, the effect of longitudinal changes in TyG index on the risk of CVD remains uncertain. This study aimed to investigate the association between change in TyG index and the risk of CVD in the general population.MethodsThe current study included 62,443 Chinese population who were free of CVD. The TyG index was calculated as ln [fasting triglyceride (mg/dL) × fasting glucose (mg/dL)/2], and change in TyG index was defined as the difference between the TyG index in 2010 and that in 2006. Multivariable-adjusted Cox proportional hazard models and restricted cubic spline analysis were used to examine the association between change in TyG index and the risk of CVD.ResultsDuring a median follow-up of 7.01 years, 2530 (4.05%) incident CVD occurred, including 2018 (3.23%) incident stroke and 545 (0.87%) incident myocardial infarction (MI). The risk of developing CVD increased with the quartile of change in TyG index, after adjustment for multiple potential confounders, the hazard ratios for the Q4 group versus the Q1 group were 1.37 (95% confidence interval [CI], 1.21–1.54) for the overall CVD, 1.38 (95% CI, 1.19–1.60) for stroke, and 1.36 (95% CI, 1.05–1.76) for MI. Restricted cubic spline analysis also showed a cumulative increase in the risk of CVD with increases in the magnitude of change in TyG index. The addition of change in TyG index to a baseline risk model for CVD improved the C-statistics (P = 0.0097), integrated discrimination improvement value (P < 0.0001), and category-free net reclassification improvement value (P < 0.0001). Similar results were observed for stroke and MI.ConclusionsSubstantial changes in TyG index independently predict the risk of CVD in the general population. Monitoring long-term changes in TyG may assist with in the early identification of individuals at high risk of CVD.