The triglyceride glucose index was U-shape associated with all-cause mortality in population with cardiovascular diseases

The retrospective cohort study included individuals from the National Health and Nutrition Examination Survey (NHANES) between the periods of 1999–2014, a nationwide survey conducted by the National Center for Health Statistics (NCHS) in United States. The cardiovascular disease (CVD) was defined as self-reported congestive heart failure, coronary heart disease, angina pectoris, heart attack and stroke. The participant was recorded as having CVD if she/he answered “yes” to the following question: “Has a doctor or other health professional ever told you that you had congestive heart failure/coronary heart disease/angina pectoris/stroke?” in a validated questionnaire (https://​wwwn.​cdc.​gov/​Nchs/​Nhanes/​2013-2014/​MCQ_​H.​htm). Among 5,012 participants with CVD, we excluded 2,939 participants with missing data on fasting triglycerides and glucose, as well as one individual with missing mortality. In total, 2,072 participants were enrolled in our study. Figure 1 depicted the selection process. All participants provided written informed consent and the protocol was approved by NCHS Research Ethics Review Board (Protocol #98 − 12, Protocol #2005-06, and Protocol #2011-17).

TyG index was calculated as ln[fasting triglycerides (mg/dL) x fasting glucose (mg/dL)/2]. Both the concentrations of triglycerides and glucose were measured enzymatically using Roche Modular P chemistry analyzer. Serum triglyceride concentration was measured using the Roche Modular P and Roche Cobas 6000 chemistry analyzers. Fasting plasma glucose was measured by the hexokinase-mediated reaction using Roche/Hitachi Cobas C 501 chemistry analyzer. The primary outcome was all-cause mortality while the secondary outcome was cardiovascular mortality. Mortality status was obtained by linkage to the National Death Index by 31 December 2015. Cardiovascular disease was defined as ICD-10 codes I00-I09, I11, I13, I20-I51, I60-I69 and I70-78.

Information on age, sex, race, education level, smoking status, drinking status, physical activity, and comorbidities were collected by using standardized questionnaires. The height and weight of each participant were obtained from the physical examinations. Fasting serum low-density lipoprotein (LDL) and high-density lipoprotein (HDL) were measured enzymatically using Roche Modular P chemistry analyzer. Body mass index (BMI, kg/m2) was calculated as weight divided by height squared. Race/ethnicity was classified as non-Hispanic white, non-Hispanic black, Mexican American or other race. Education level was categorized as less than high school, high school or equivalent and college or above. Smoking status were defined as current, past and never. Physical activity was categorized as vigorous, moderate and inactive. Hypertension was defined as the self-report hypertension, or systolic blood pressure ≥ 140 mmHg, or diastolic blood pressure ≥ 90 mmHg, or taking antihypertensive drugs. Diabetes was defined as a history of diabetes or fasting glucose > 7 mmol/L or glycated hemoglobin A1c > 6.5% or use of hypoglycemic medication. Multiple imputation using predictive mean matching was performed for covariates with missing values. Predictor variables included all covariates from the main analysis, survey weight, and unique groupings of sampling strata and primary sampling unit. Ten imputed datasets were generated and pooled to obtain the overall result [12].

Data are presented as mean ± standard deviation or number (proportions). Differences among different TyG index groups were explored by one-way analysis of variance and chi-square test. Associations between TyG index and the risk of all-cause and cardiovascular mortality were estimated by multivariate Cox regression models with 95% confidence intervals. The selection of confounding variables was based on the significant difference among TyG quartiles and clinical relevance of all-cause and cardiovascular mortality. Model 1 was unadjusted. Model 2 was adjusted for age, and gender. Model 3 was adjusted for age, gender, race, education, BMI, smoking, drinking, physical activity, hypertension, diabetes, HDL and LDL. The specific associations between levels of TyG index and all-cause mortality was evaluated on a continuous scale with restricted cubic spline. All analysis were performed using the statistical package R version 3.6. All P values were two-sided with a significance level of < 0.05.