We included Chinese population who received primary health care in West China Hospital Health Management Center from January 2020 to Decemberr 2021. Subjects with serious cardiovascular and cerebrovascular diseases, severe liver and kidney disease, malignant tumors, and those receiving lipid-lowering drugs were excluded. Patients with a history of carotid artery disease were excluded. This study was approved by the Ethics Committee of West China Hospital, Sichuan University and was performed in accordance with the Declaration of Helsinki. All the participants signed informed consent forms.

Demographic data, smoking and drinking history, family history, and present disease history of subjects were collected. Participants were fasting for at least 8 h. Height, weight, waist circumference, hip circumference, waist-to-hip ratio, blood pressure, fasting plasma glucose (FPG), glycosylated hemoglobin A1c (HbA1c), blood uric acid, and blood lipids were measured. Blood lipids included TG, TC, LDL-C, HDL-C, and NHDL-C. Biochemical indexes were determined by an automatic biochemical analyzer (ROCHE COBASC702, Shanghai), and HbA1c was measured by high-pressure liquid chromatography (TOSOH HLC-723G8, Japan). Fatty liver was assessed by color Doppler ultrasound of the liver (Philips EPIQ7C, America). Body mass index (BMI) = weight/height² (kg/m²).

Bilateral carotid arteries were assessed by an experienced sonographer through a color Doppler ultrasound (Philips EPIQ7C, America). Carotid artery color Doppler ultrasound reports are issued by the operating doctor and reviewed in real time by senior physicians. The probe must be parallel to the vessel wall, and the beam should be perpendicular to the tube wall. The scanning mode of combined longitudinal and transverse sections was adopted. The vertical distance from the upper edge of the inner membrane to the upper edge of the outer membrane in the distal common carotid artery and/or bulb of the carotid artery was defined as CIMT (22). A CIMT less than 1 mm was normal. A 1.0 mm ≤ CIMT < 1.5 mm was defined as thickened CIMT (22). A CIMT ≥1.5 mm, protruding into vascular or localized thickness and 50% higher than peripheral CIMT, was defined as carotid plaque (22). It was discovered by chance during the first-time ultrasound screening.

SAS 9.4 software was used for statistical analysis. The central trend of measurement data is represented by the mean (X). Dispersive trends were expressed by the standard deviation (SD). Qualitative data were described by absolute numbers and rates. The measurement data of two independent samples were compared by t test. The comparison of rates was compared by the chi-square test of four lattice tables. Fisher's exact probability method was used if the test conditions were not satisfied. Multivariate linear regression analysis was used in the multivariate analysis if dependent variables were quantitative variables. Logistic regression was used for multivariate analysis when dependent variables were dichotomous variables. The test level α was 0.05.

Our study found that the prevalence of thickened CIMT and carotid plaque were 9.4% and 17.8%, respectively, in Chengdu residents. Another study, also from China, including 311 community residents, found that the prevalence of thickened CIMT and carotid plaque were 8.4% and 15.8%, respectively, which is similar to the results of our study (19). An epidemiological investigation from Jiangsu Province, China, showed a higher morbidity rate than our study (23). The prevalence of thickened CIMT was 13% in a Japanese study involving a 2012 population aged 34–88 years (24). Differences in prevalence may be related to the different gender compositions, ages, races, diseases, and other potential factors in the included participants. Participants in our study voluntarily underwent a physical check-up, and their literacy, incomes, and health requirements were relatively higher, which may be the main reason for the lower prevalence of thickened CIMT and carotid plaque. The subjects in our research were mainly Han Chinese. Race is an independent predictor of CIMT or carotid plaque in a large multiethnic cohort with a 10-year follow-up (25).

Liu et al. (26) included 3,214 healthy Chinese individuals undergoing physical examinations and found that serum HDL-C, NHDL-C, TC/HDL-C, and LDL/HDL-C were all associated with the incidence of carotid artery plaques. There are both similarities and differences with our research results. Although all the people we included were those undergoing health check-ups, the sample size of our research is larger than that of previous study.our criteria for inclusion and exclusion were different and we excluded patients with a previous history of carotid plaque. All of these might be the reasons for the inconsistent results.

To date, there has been a controversial correlation between TG or HDL-C and arteriosclerotic disease (17, 18, 20, 21, 27–29) Several studies have suggested that TG is independently associated with an increased risk of coronary artery disease CAD (17, 18). In contrast, a study reported that neither HDL-C nor TG can independently predict carotid plaque (27). A study has even found that TG is a protective factor against carotid plaque (28). The Guidelines of the American College of Cardiology and the European Society of Cardiology do not recommend HDL-C as a target for preventing ASCVD (29).

The controversy about the relationship between TG and arteriosclerosis may be attributed to the complexity of absorption and metabolism of TG and the role of different TG subtypes in the development of atherosclerosis (30). TG can trigger inflammation at the carotid artery intima-media and is then resolved, but cholesterol remains in it to promote foam cell formation and atherosclerosis (31). Assessing the benefits of triglyceride reduction alone is not easy, since many triglyceride-lowering drugs also lower cholesterol.

The concentration of high-density lipoprotein (HDL-P) rather than HDL-C particles is a better predictor of the function of HDL (32, 33). An observational study confirmed a stronger relationship between HDL-P and CAD than HDL-C (34, 35). There is evidence that HDL-P with cholesterol overload may be harmful, as experimental studies have observed that HDL-P with cholesterol overload affects the transport and clearance of cholesterol (36, 37). HDL-P with cholesterol overload is independently associated with the progression of carotid atherosclerosis, which may explain why increased HDL-C is not beneficial. HDL-P and HDL-C together determine antiatherosclerotic function, rather than a single parameter (38). The lipid parameter in our study was HDL-C rather than HDL-P; therefore, no correlation between HDL-C and IMT or carotid plaque was found.

TC and LDL-C are well-recognized risk factors for atherosclerosis. Studies have shown that TC and LDL-C have the greatest predictive value for thickened CIMT or carotid plaque (19, 39). However, it has been reported that LDL-C and TC are not associated with carotid plaques in populations with a high risk of stroke (21). The researchers explain that the controversy may be due to the high-risk participants included in the study (21). In high-risk populations, the role of lipids may be attenuated (40). The participants in this study were from the general population rather than the high-risk population.

Another interesting finding was that there was no correlation between TC or LDLD-C and carotid artery before adjusting for other confounding factors, but after adjusting for confounding factors, there was a significant correlation between them. Therefore, when conducting multi-factor analysis, it is essential to identify potential confounding factors and make adjustments to prevent the true relationships from being concealed.

Most of the studies indicate that the NHDL-C level is a better predictor of IMT and carotid plaque than TC/HDL-C (41–43).

However, Tamada M's study reported that non-HDL-C is not an independent predictor of carotid atherosclerosis (44). The negative results may be attributed to a large number of low-risk subjects in the study, with lower levels of NHDL-C in low-risk subjects than patients with CAD at baseline (44). NHDL-C, which reflects the total particles leading to atherosclerosis, includes very low density lipoprotein cholesterol (VLDL-C), intermedium density lipoprotein cholesterol (IDL-C), and LDL-C and is equal to the TC concentration minus the HDL-C concentration (45). NHDL-C is a good predictive indicator for atherosclerosis.

At present, most studies believe that TC/HDL or LDL-C/HDL-C ratios are better predictors of atherosclerosis than TC, LDL-C, or HDL-C alone (15, 16). In contrast, some studies hold controversial points (46). The study has shown that LDL-C/HDL-C is positively associated with CIMT and carotid plaque in male patients with T2DM; however, no correlation has been observed in female patients (45), which may be due to estrogen (47). Our study found that LDL-C/HDL-C was an independent risk factor for thickened CIMT and was positively correlated with CIMT. There was no correlation between TC/HDL-C and the prevalencerate of thickened CIMT or CIMT. There was no correlation between LDL-C/HDL-C or TC/HDL-C and carotid plaque. Possible reasons are as follows. First, LDL-C is more predictive of atherosclerosis than TC (39), so LDL-C/HDL-C is more predictive than TC/HDL-C. Second, HDL-C was not a predictor of atherosclerosis in this study; therefore, the predictive values of LDL-C/HDL-C and TC/HDL-C were reduced. Third, there are differences in risk factors between CIMT and carotid plaque (48). Thickened CIMT and carotid plaque are different stages of the development of atherosclerosis, and their risk factors may be discrepant. Kocaman SA et al. (48) discovered LDL cholesterol, which appear to have a role in later stages of atherosclerosis but not in early stages.

Our study found NHDL-C/HDL-C significant correlation between carotid intima-media thickness or carotid plaquein, which was agreement with previous studies. Wang et al. (49) included 27,436 urban workers and among them, 7,161（26.1%） participants were diagnosed with carotid artery plaques. The results show that NHDL-C/HDL-C was related to carotid artery plaques. Liu et al. (50) included 839 subjects at high risk of stroke and found that the non-HDLc/HDLc was positively correlated with the incidence of carotid artery plaques. This study included the high-risk population of cerebral stroke, mainly focusing on non-HDLc/HDLc and carotid plaques. Although the populations included in the three studies were different, the conclusions reached by the three studies were the same. Our research further confirms the previous conclusions.

We included a population that was actively engaged in primary health care, a population that has rarely been reported before. This study included nearly ten thousand participants, while the sample size of previous studies ranged from several hundred to several thousand cases. A larger sample size may produce more reliable results. In this study, patients with newly diagnosed thickened CIMT or carotid plaque screened by physical check-up may benefit from early secondary prevention to prevent ASCVD and stroke. Because our study excluded patients with serious cardiovascular and cerebrovascular diseases, those receiving lipid-lowering drugs and those with a history of carotid artery disease, the conclusion is relatively reliable. In addition, this study analyzed the correlation of various lipid parameters with CIMT, thickened CIMT and carotid plaque, which is relatively comprehensive.

The participants in this study were Chinese residents in Chengdu who voluntarily underwent a physical check-up in a single center, so the results are difficult to generalize to other ethnicities or other regions. Moreover, this study is a cross-sectional study. Although some confounding factors were adjusted, there may still be some potential confounding factors affecting the results. The cross-sectional study could only prove the correlation between blood lipids and CIMT or carotid plaque but not the temporal relationship between lipid parameters and the progression of CIMT or plaque development. So the conclusion we draw is correlation rather than causation. This study did not analyze the relationship between blood lipids and plaque quantity, type, area, thickness, and so on. Carotid artery ultrasound during physical examination is a preliminary screening and does not describe the number and extent of plaques. Carotid ultrasound examinations were performed by experienced physicians and reviewed by senior operators; however, this does not substitute for formal assessment of measurement reproducibility, which might be one of the potential factors influencing the outcome. Furthermore, non-fasting samples, absence of remnant lipoprotein quantification may affect association between TG and HDL-C. Therefore, prospective multicenter studies with higher quality are needed to further confirm causality.