Association of triglyceride glucose-related parameters with all-cause mortality and cardiovascular disease in NAFLD patients: NHANES 1999–2018

Abstract Background The relationship between the triglyceride-glucose (TyG) index and its derived index, the triglyceride glucose-waist height ratio (TyG-WHtR), with mortality and cardiovascular diseases (CVDs) in patients with non-alcoholic fatty liver disease (NAFLD) remains unclear. Methods This study enrolled 6627 adults aged 18 and above diagnosed NAFLD from the National Health and Nutrition Examination Survey (NHANES, 1999–2018). Binary weighted logistic regression analyses, cox proportional hazards model and restricted cubic spline (RCS) were used to analyze the relationship between TyG and TyG-WHtR with all-cause mortality, CVD mortality and CVDs. Mediation analysis explored the mediating role of glycohemoglobin, insulin and hypertension in the above relationships. Meanwhile, the incremental predictive value of the TyG index and TyG-WHtR was further assessed. Results Except for no significant association between the TyG index and both all-cause mortality and chronic heart failure (CHF), both TyG and TyG-WHtR exhibited significant positive correlations or trends of positive correlation with all-cause mortality, CVD mortality, total-CVD, CHF, coronary heart disease (CHD) and angina pectoris. For all-cause mortality, CVD mortality and CHF, TyG-WHtR was a better predictor than TyG (TyG-WHtR: HR 1.31, 95%CI 1.03–1.66; HR 2.22, 95%CI 1.42–3.47; OR 3.99, 95%CI 1.79–8.93). In contrast, TyG index demonstrated a stronger association with total-CVD, CHD and angina pectoris (TyG index: OR 2.00, 95%CI 1.26–3.18; OR 1.85, 95%CI 1.19–2.91; OR 2.93, 95%CI 1.23-7.00). RCS analysis showed that after adjusting for covariates, most of the aforementioned relationships were linear(P overall < 0.0001, P-nonlinear > 0.05), while the associations of the TyG index and TyG-WHtR with all-cause mortality and CHF were non-linear(P overall < 0.0001, P nonlinear < 0.05). The addition of the TyG index and TyG-WHtR to the basic model for outcomes improved the C-statistics, net reclassification improvement value, and integrated discrimination improvement value. Conclusions The predictive value of TyG or TyG-WHtR for all-cause mortality and cardiovascular risk in NAFLD patients was significant. The TyG index and TyG-WHtR might be valid predictors of cardiovascular outcomes of patients with NAFLD.