Cardiovascular diseases (CVDs), particularly ischemic heart disease and stroke, remain the leading cause of mortality and a significant contributor to disability globally (1). According to the World Health Organization report from 2019, around 17.9 million deaths were attributed to CVDs, which corresponded to 31% of all the global mortality rate (2). Atherosclerosis, the primary etiology of CVD, poses a significant public health challenge owing to its asymptomatic nature over long terms, unfavorable prognosis, and reduced life expectancy (3, 4). Carotid atherosclerosis (CA), which encompasses increased carotid intima-media thickness (CIMT), plaque and stenosis, is widely recognized as a crucial indicator of generalized atherosclerosis and a predictor of cardiovascular disease events (5–8). Early detection of CA through regular non-invasive ultrasonography is advantageous in implementing proactive measures to prevent or manage CVD before its progression (9). Given the significant and escalating burden of CVD, it is imperative to promptly detect CA in the general population, identify promising biomarkers for early detection and implement preventive measures.

Recent accumulating evidence suggests that obesity, as measured by body mass index (BMI), is strongly associated with an increased risk of developing CA (10, 11). However, a significant number of individuals with normal weight but metabolic abnormalities also exhibit a cluster of metabolic risk factors as well as an elevated risk for carotid artery disease (12, 13). Furthermore, individuals with a normal weight, as determined by BMI, frequently perceive themselves as being in good health. Consequently, it is less probable that individuals with normal weight will undergo clinical screening and early intervention for CA compared to those who are obese. Therefore, timely detection of CA in individuals with normal body weight is imperative.

Insulin resistance (IR), defined as the attenuation of insulin responsiveness in tissues, not only expedites the advancement of atherosclerosis but also serves as a primary characteristic of metabolically obese normal-weight individuals (14). This implies that determining the degree of IR might be advantageous in predicting the likelihood of developing CA. The homeostatic model assessment of insulin resistance (HOMA-IR) has traditionally been employed to quantify insulin resistance; however, there has been no consensus regarding the association of HOMA-IR scores with the risk of CA (15–17). Furthermore, HOMA-IR is substantially limited in clinical practice by the need to measure insulin levels. The demand for a dependable and cost-effective indicator of IR has prompted the creation of innovative surrogate indexes, including triglyceride-glucose index (TyG) (18, 19), TyG-BMI (20), TyG-waist circumference (TyG-WC) (21), TyG-waist-to-height ratio (TyG-WHtR) (22), Chinese visceral adiposity index (CVAI) (23), visceral adiposity index (VAI) (24) and lipid accumulation product (LAP) (18), which are effective in assessing IR status. Reportedly, a few studies have evaluated the association between these partially surrogate IR indexes and atherosclerosis (25–29). However, many of these studies had small sample sizes and did not specifically target individuals with a normal weight, a population that has been understudied and often overlooked in early screening for CA. Furthermore, it remains ambiguous as to which indexes hold greater predictive value for CA.

Despite considerable research on the surrogate indexes of IR, no studies have investigated their relationship with CA prevalence in normal-weight individuals. Accordingly, the present large, cross-sectional study enrolled 26,795 community residents with normal weight to explore the relationship between surrogate IR indexes and CA. In addition, the study identified the index with the most predictive ability for IR in a normal-weight population.

Using data from CSHPSIP, this large-scale, cross-sectional study demonstrated a significant correlation between the prevalence of CA and TyG-WC, TyG-WHtR, CVAI, and LAP in normal-weight adults. However, no significant association was observed between the prevalence of CA and TyG, TyG-BMI or VAI. Furthermore, among all the indexes, CVAI exhibited superior predictive ability for determining CA, increased CIMT, plaque and stenosis.

Research suggests that atherosclerosis is characterized by an initial, extended asymptomatic phase, which can commence as early as adolescence or even childhood (33). Given the increasing burden of CVD, it is crucial to conduct early screening and timely interventions for atherosclerosis for reducing the incidence of CVD. Furthermore, it is less likely that the normal-weight population will undergo clinical screening and early intervention for CA compared with the obese population. However, the present study found that the prevalence of CA was as high as 50.1% in normal-weight individuals. This indicates that it is imperative to incorporate carotid ultrasound into routine annual physical examinations even for individuals with normal body weight.

IR is a robust predictor of atherosclerotic CVD, as it plays a pivotal role in the initiation and progression of atherosclerosis via complex pathophysiological processes such as endothelial injury, activation of the inflammatory response and oxidative stress (34). Many observational studies have shown that over 30% of individuals with a normal body weight exhibit metabolic abnormalities, such as abdominal adiposity and insulin resistance, similar to those typically seen in overweight or obese individuals (known as the “MONW” phenotype) (35, 36). Furthermore, Asian populations who are metabolically obese but have normal weight are at an elevated risk of developing CA compared with metabolically normal but obese or normal-weight individuals (12). However, there is a dearth of metabolic indicators to determine the onset of CA in individuals with normal body weight. To the best of our knowledge, the present study is the first to enroll 26,795 normal-weight individuals and to demonstrate that CVAI is the most effective surrogate index for predicting CA. Moreover, CVAI can be used as a reliable and easily quantifiable measure to identify high-risk populations and implement primary prevention strategies against atherosclerosis.

CVAI is indicative of the distribution of abdominal fat and dyslipidemia and is linked to insulin resistance, abnormal glucose metabolism and an elevated risk of CVD in adults (23, 37, 38). In a cohort study of 3,640 Chinese individuals from CHARLS, CVAI exhibited a higher predictive ability for metabolic syndrome in females than the other six surrogate indexes examined in our study (39). Another study enrolling 1,452 Chinese adults similarly demonstrated that CVAI had a superior predictive ability for identifying metabolic syndrome (40). In certain studies, the correlation between CVAI and CA has also been examined. A study based on 4,075 workers from a steel company has revealed that an increased CVAI is a predictive indicator of increased CIMT (29). However, it should be noted that steelworkers are more overweight and obese than the general population owing to occupational factors such as shift work, noise exposure and occupational stress (29). As such, the findings may not be generalizable to individuals within the normal-weight range. Furthermore, a retrospective study in Taiwan Province revealed a positive correlation between CVAI and the prevalence of CA (41). In addition, the study by Hu et al. observed a significant association between CVAI and the risk of carotid plaque (42). However, these studies did not include participants with normal weight; thus, the results may not be generalizable to the Chinese population with normal weight. Compared with previous studies, the present study specifically examined individuals with normal weight and identified a non-linear, positive association between CVAI and CA, thereby addressing gaps in our understanding of the relationship between CVAI and CA occurrence among normal-weight individuals. Notably, when CVAI exceeds 89.83, the prevalence of CA increases significantly. Therefore, for adults with normal weight, it is recommended to undergo further carotid ultrasound examination when their CVAI exceeds 89.83.

The LAP index, a measure of obesity that is based on WC and fasting TG level, is used for determining the burden of coronary atherosclerotic plaques and the risk of CVD (43, 44). To date, only one study has indicated a correlation between LAP and carotid plaque; however, its small sample size and exclusive focus on acromegalic populations limits the generalizability of the findings (27). The present study revealed a significant correlation between an increase of one SD in LAP and the development of atherosclerotic phenotype, particularly stenosis, among individuals with normal BMI. However, further research is necessary to validate the applicability of these findings to individuals who are either obese or underweight.

The TyG index serves as a practical surrogate for IR and exhibits superior performance compared with the traditional HOMA-IR (45). However, there has been no consensus regarding the correlation between elevated TyG and an increased prevalence of CA. In respective studies, Irace et al. and Li et al. demonstrated a significant correlation between the TyG index and CA (15, 46). However, Zhao et al. failed to establish any association between the TyG index and increased CIMT or carotid plaque (47). Irrespective, detailed analyses of different types of CA or subgroups of participants with normal weight were not conducted in these studies. Li et al. revealed that overweight individuals in the fourth quartile of TyG were at a higher risk for developing CA, increased CIMT, plaque formation and stenosis than those in the first quartile (28). Nevertheless, there were no notable correlations between either carotid plaques or stenosis with regard to individuals with normal BMI in the second to fourth quarters of the TyG index (28). The current investigation additionally revealed no significant correlation of TyG with CA, increased CIMT, plaque, and stenosis. These findings suggest that TyG may not be a dependable indicator for evaluating CA, particularly in individuals with normal BMI.

TyG index-related parameters are composite indicators that incorporate the TyG index with BMI, WC, and WHtR, as initially proposed by Er et al. (20). As mounting evidence suggests a close association between visceral adiposity and IR, the combination of visceral adiposity and TyG may offer a greater potential for identifying IR than TyG alone (20–22). Similarly, although the current study revealed no significant correlation between TyG and increased prevalence of CA among normal-weight participants, a noteworthy association was observed when combined with a composite measure of visceral adiposity (including WC and WHtR).

The association between VAI and atherosclerosis has yielded inconsistent findings in prior research studies. Several studies, despite their relatively small sample sizes, demonstrated a significant correlation between VAI and CA in the general population (25, 26). Xu et al. enrolled 3,363 older adult Chinese individuals and demonstrated that increased VAI was not linked with CA risk (48). Similarly, a cross-sectional study involving 788 Spanish patients did not reveal any correlation between VAI and CA risk (49).The sample size in this study was higher than that reported previously, leading to more robust findings. However, there was no significant correlation between an increase in per VAI SD and an elevated prevalence of CA among individuals with normal weight. This suggests that VAI may not be an effective predictor of CA.

The subgroup analysis conducted in this study demonstrated that the impact of CVAI, TyG-WC, TyG-WHtR, and LAP on the prevalence of CA was notably more significant among females and individuals aged between 40 and 49 years. This finding may be attributed to the heightened risk of atherosclerotic complications during the menopausal transition period for females, as previously suggested in literature (50). The decrease in estrogen secretion during peri-menopause and menopause is known to result in the accumulation of central adiposity and insulin resistance (51). In the context of insulin resistance, the absence of estrogen’s safeguarding impact on endothelial function increases the vulnerability to atherosclerosis. To substantiate this conjecture, additional external validation in more representative populations is imperative.

While benefiting from the well-established cohort and its relatively large size, this study is subject to certain limitations. First, certain confounding factors such as dietary habits and postmenopausal status were not considered in this investigation, which may have affected study results. Second, the study could not determine causality between surrogate IR indexes and CA due to its cross-sectional design. Finally, the present findings should be cautiously generalized to other populations, as the study only included Chinese middle-aged and older adult participants with normal weight.

Our study has provided evidence suggesting that CVAI, TyG-WC, TyG-WHtR, and LAP are potential predictors of CA in middle-aged and older adult individuals who have normal weight. Specifically, CVAI may be the most appropriate index for predicting CA in a normal-weight population.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were reviewed and approved by Xiangya Hospital Ethics Committee. The protocol and informed consent for the study of the China Stroke High-risk Population Screening and Intervention Program were reviewed and approved by the Institutional Review Board at the Capital Medical University Xuanwu Hospital early. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

ZL and JX: conceptualization. JF: data curation, resources, and writing – review and editing. ZL: formal analysis and writing – original draft. RT: investigation. ZL, BD, and QH: methodology. JX: project administration. BD, QH, and FY: software. RT, FY, and JF: validation. All authors contributed to the article and approved the submitted version.

This research was supported by the National Key Research and Development Projects (2022YFC3602400, 2022YFC3602401), the National Natural Science Foundation of China (82271369), the National Natural Science Foundation Youth Fund (82001392), the Natural Science Foundation of Hunan Province (Grant No. 2021JJ31109), and the Fundamental Research Funds for the Central Universities of Central South University (2022ZZTS0821).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.