Abstract
Objective: To determine the associations between triglyceride-glucose (TyG) index and mortality from all causes and cardiovascular causes in diabetic population. Methods: 3349 participants with diabetes mellitus (DM) from the 1999-2014 National Health and Nutrition Examination Surveys (NHANES), aged 18-85 years were included and grouped based on the TyG index in quintiles. Mortality was followed up through December 31(th), 2015. Cox proportional hazards models were used to assess the hazard ratios (HRs) and 95% confidence intervals (CIs). We clarified the shape of association between TyG index and mortality using restricted cubic splines and piecewise linear regression. Results: After a median follow-up period of 82 months, 800 (23.9%) deaths occurred, of which 190 (5.7%) were due to cardiovascular causes. Participants in the top quintile had higher risks of all-cause mortality (HR, 1.38; 95% CI, 1.04-1.48) and cardiovascular mortality (HR, 2.43; 95% CI, 1.32-4.45) than those in the lowest quintile. TyG index and all-cause mortality had a J-shaped relationship with a threshold value of 9.32, while TyG index and cardiovascular mortality had a reversed L-shaped relationship with a threshold value of 9.37. Higher TyG index was associated with increased risks of all-cause mortality (per SD increment, HR, 1.52; 95% CI, 1.27-1.82) and cardiovascular mortality (per SD increment, HR, 2.17; 95% CI, 1.54-3.04) when above the threshold values. The sensitivity analyses demonstrated similar findings. Conclusions: TyG index in diabetic patients was nonlinearly correlated with mortality risks, potentially predicting all-cause and cardiovascular mortality.