Study design and population
This was a cross-sectional study included 3,169 railway workers in southwest China from January to December 2021.
Inclusion criteria: All on-job railway drivers were included in this study. Exclusion criteria were as follows: The participants with incomplete clinical information (e.g., fasting blood glucose, blood lipid), severe systemic disease (such as cardiac, renal or liver failure, and malignant) were excluded.
Data Collection
The height and weight of participants were measured, and the body mass index (BMI) (kg/m2) was calculated. Subjects were divided into two BMI categories: BMI < 23 kg/m2, underweight or normal; BMI ≥ 23/m2, overweight or obesity. Blood pressure was presented in the seated position using a sphygmomanometer, and the average of the two readings was calculated as systolic blood pressure (SBP) and diastolic blood pressure (DBP). Elevated blood pressure was defined as a SBP ≥ 140 mmHg or DBP ≥ 90 mmHg.
Fasting blood samples were collected after an 8-h to 12-h overnight fast. Total cholesterol (TC), TG, high-density-lipoprotein cholesterol (HDL-C), low-density-lipoprotein cholesterol (LDL-C), FBG, and high-sensitivity C-reactive protein (hs-CRP) were uniformly analyzed by laboratory physician of Clinical Medical College & Affiliated Hospital of Chengdu University, Chengdu, Sichuan, China. The TyG index was calculated as Ln [(fasting TG (mg/dl) x FBG (mg/dl)/2][15].
The presence/absence of CAP was determined using a digital ultrasonic diagnostic system (EPIQ CX, Philips Ultrasound Inc., USA). The common carotid artery, the carotid artery bulb, and the near and far wall segments of the internal carotid artery were scanned bilaterally. Two physicians with more than five years of experience in vascular ultrasound imaging reviewed the images blindly. According to the Mannheim criteria[17], CAP was defined as a focal region encroaching into the arterial lumen by at least 0.5 mm, > 50% of surrounding intima-media thickness values, or thickness ≥ 1.5 mm above the distance of the interface between the lumen-intima and the media-adventitia.
Covariates
The TyG index and CAP confounders chosen by earlier studies[15, 16, 18, 19] were the potential factors in our analysis [15, 16, 18, 19]. The sociodemographic, personal medical history (diabetes, hypertension, hyperlipidemia, medication history), and lifestyle factors (smoking status, drinking status, dietary pattern, physical activity, etc.) were collected via a standard questionnaire by our trained doctors. Smoking and drinking status were defined as smoking 1-cigarette per day and consuming at least ≥ 50g alcoholic drink 1 time/week for more than 6 months. According to the total metabolic equivalent (MET) score, physical activity was divided into two groups (3000 MET min/w, 3000 MET min/w)[20]. Dietary patterns were assessed by intaking meat, vegetables, and fruits [21]. Dietary intake frequency categories were recoded as follows: "never" = 0, "less than once a week" = 0, "2–3 times a week" = 2, "4–6 times a week" = 5, and "once or more daily" = 7. Servings for meat were summed to create the frequency of unprocessed meat consumption. For fruit and vegetables, participants were asked how many heaping tablespoons of cooked/salad, raw, or pieces of fresh/dried fruit they ingested daily. Tablespoons of cooked/salad and raw vegetables were added to create the consumption of vegetables and pieces of fresh and dried fruit were added to create the consumption of fruit. We determined healthy diets using the healthy diet score (HDS), which was calculated using the following criteria: consumption of at least four tablespoons of vegetables daily, three pieces of fruit daily, two servings of fish weekly, two servings of unprocessed red meat weekly, and no more than two servings of processed meat weekly (median). The total diet score ranged from 0 to 5, and one point was awarded for each favorable dietary component. Participants were divided into three groups of poor dietary habits (score of 0 or 1), medium dietary habits (score of 2 or 3), and ideal dietary habits (score of 4 or 5). The participants' daily salt intake was self-reported and was divided into three groups: low intakes, moderate intakes, and high intakes.
In our investigation, metabolic dysfunction-associated fatty liver disease (MAFLD) was also considered a confounder. The proposed criteria for a positive diagnosis of MAFLD were based on hepatic steatosis detected by ultrasound image in addition to one of the following three criteria: overweight/obesity, presence of type 2 diabetes mellitus, or evidence of metabolic dysregulation [22]. Hypertension was defined as an elevated blood pressure, prior diagnosis or any use of antihypertensive agents. Diabetes was defined as FBG levels ≥ 7.0 mmol/L, prior diagnosis or any use of hypoglycemic agents. Dyslipidemia was defined as any self-reported history or use of lipid-lowering drugs, or TC levels ≥ 5.17 mmol/L or TG levels ≥ 1.69 mmol/L or LDL-C levels ≥ 3.62 mmol/L or HDL-C level ≤ 1.04 mmol/L.
Statistical analysis
According to the previous study, the prevalence of CAP in the general Chinese population was about 20.15%[23]. Assuming 80% power, a 2-sided α error of 0.05, and the allowable error was 3%; finally, we obtained a sample size of 687. Considering a dropout rate of 20%, we decided on a minimum total sample size of 825. The sample size of this study has reached this standard.
Participants were divided into 4 categories according to tertiles of the TyG index. Continuous variables were described as the median (interquartile ranges [IQRs]) and analyzed by a rank-sum test owing to the skewed distribution. Categorical variables were described using frequency (%) and analyzed by the Chi-square test. The logistic regression models were performed to estimate the association between the TyG index and CAP by calculating odds ratios (ORs) and 95% confidence intervals (CIs) after adjusting covariates step by step from models 1 to 3. When the TyG index was treated as a continuous variable, a restricted cubic spline with 3 knots at the 10th, 50th and 95th percentiles of the TyG index were performed to examine the detailed association between the TyG index and CAP. Subgroup analyses were stratified by age, MAFLD, smoking and drinking, physical activity, dietary pattern, salt intake, and elevated blood pressure. Besides, sensitivity analyses were performed to examine our results' robustness by excluding participants using hypoglycemic agents, antihypertensive agents, and lipid-lowering drugs.
All statistical analyses were conducted in R Studio (Version 4.0.5). A two-sided P < 0.05 was considered statistically significant.