RESUMEN
BACKGROUND: The triglyceride-glucose (TyG) index is regarded as a sophisticated surrogate biomarker for insulin resistance, offering a refined means for evaluating cardiovascular diseases (CVDs). However, prospective cohort studies have not simultaneously conducted baseline and multi-timepoint trajectory assessments of the TyG index in relation to CVDs and their subtypes in elderly participants. METHODS: After excluding data deficiencies and conditions that could influence the research outcomes, this study ultimately incorporated a cohort of 20,185 participants, with data chronicles extending from 2016 to 2022. The TyG index was calculated as Ln [fasting triglyceride (mg/dL) × fasting glucose (mg/dL)/2]. Latent Class Trajectory Model (LCTM) was used to assess the change trends of the TyG index over multiple time points. Utilizing the Cox proportional-hazards models, we assessed the relationship between the baseline quartiles of the TyG index and various trajectories with CVDs and subtypes. RESULTS: During the mean follow-up time of 4.25 years, 11,099 patients experienced new CVDs in the elderly population. After stratifying by baseline TyG quartiles, the higher TyG level was associated with an increased risk of CVDs; the aHR and 95% CI for the highest quartile group were 1.28 (1.19-1.39). Five trajectory patterns were identified by the LCTM model. The low gradual increase group as the reference, the medium stable group, and the high gradual increase group exhibited an elevated risk of CVDs onset, aHR and 95%CIs were 1.17 (1.10-1.25) and 1.25 (1.15-1.35). Similar results were observed between the trajectories of the TyG index with subtypes of CVDs. CONCLUSION: Participants with high levels of baseline TyG index and medium stable or high gradual increase trajectories were associated with an elevated risk of developing CVDs in elderly populations.