Prognostic effect of triglyceride glucose-related parameters on all-cause and cardiovascular mortality in individuals with cardiovascular-kidney-metabolic syndrome: evidence from international multi-cohort studies

Abstract

Objectives

The emerging triglyceride–glucose (TyG) related index has attracted attention as a promising predictor of various cardiometabolic conditions. However, their prospective association with different stages of cardiovascular-renal metabolic (CKM) syndrome is still not fully established, and it remains unclear whether TyG related parameters have prognostic effects on mortality outcomes of CKM syndrome.

Methods

The data were derived from the China Health and Retirement Longitudinal Study (CHARLS), and which were determined by the use of a standardised questionnaire during follow-up. TyG and its related parameters (TyG-body mass index, TyG-waist circumference, TyG-waist to height ratio, and TyG-a body shape index (TyG-ABSI) were calculated. Multivariate Cox regression analysis was used to analyze hazard ratios (HRs) and 95% confidence intervals (CI), and Kaplan–Meier survival curve was used to analyze the associations of TyG-ABSI with all-cause mortality and cardiovascular mortality in patients with CKM syndrome. Additionally, the multivariate adjusted restricted cubic spine was employed to examine the dose-response relationship. Mediation analysis was conducted to assess whether white blood cell (WBC) and C-reactive protein (CRP) mediated the associations. Subgroup analyses and interaction tests were conducted to evaluate the risk within various demographics. The National Health and Nutrition Examination Survey (NHANES) was used as validation to improve the reliability of the study results.

Results

The study enrolled 11,235 participants with CKM syndrome from the CHARLS database, during the median follow-up of 5 years, a total of 747 (6.65%) all-cause mortality and 84 (0.75) cardiovascular mortality occurred. TyG-ABSI was associated with CKM syndrome (OR 1.55; 95% CI 1.35–1.79). Furthermore, among patients with CKM syndrome, TyG-ABSI was association with all-cause mortality (HR 1.14; 95% CI 1.04–1.35). In which continuous TyG-ABSI were converted to classified variable (tertile), compared to those with T1 group, the risk of advanced CKM syndrome was found to be 2.41-fold higher in those with T3 group (OR 2.41; 95% CI 1.18–3.20). Additionally, individuals in the T3 group had a 55% increased risk of all-cause mortality (HR 1.55; 95% CI 1.10–2.18). The mediation analysis results suggested that the relationship between TyG-ABSI and all-cause mortality risk is partially mediated by WBC, and CRP, the proportion of mediation were 15.16% and 11.83%. Additionally, analyses of 15,054 participants from the NHANES database indicated a significant positive association between TyG-ABSI and all-cause mortality and cardiovascular mortality among individuals diagnosed with CKM syndrome during the 10 years follow-up.

Conclusion

Higher TyG-ABSI is associated with an increased risk of advanced CKM syndrome and mortality. It further emphasizes the role of TyG-ABSI in the management of CKM syndrome stages and the risk of all-cause mortality and cardiovascular mortality.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-025-03057-0.

Introduction

The American Heart Association (AHA) recently proposed the concept of cardiovascular-kidney-metabolic (CKM). CKM is a multifaceted clinical condition defined by the complex interaction of metabolic disorders, chronic kidney disease (CKD), and cardiovascular disease (CVD) [1]. This interconnected pathophysiology leads to multiorgan damage and significantly increases the likelihood of adverse cardiovascular events [2]. The complex interactions and shared pathological mechanisms among these diseases further contribute to the disease burden and mortality risk [3]. CKM affected a large proportion (89.4%) of American adults. Among them, more than half were at stage 2 or higher, and during the period from 2011 to 2020, more than half of American adults remained at stage 2 or higher for CKM, with no improvement at all [4]. The CKM syndrome usually arises from either excess adipose tissue or its dysfunction, often a combination of the two. Dysfunctional visceral adipose tissue releases pro-inflammatory cytokines and oxidative stress mediators that adversely affect arterial, cardiac, and renal tissues. When these inflammatory substances are released into the blood, pro-oxidative and pro-inflammatory mediators may exacerbate atherosclerotic damage and myocardial damage [5]. Epidemiological data suggest that as individuals progress from CKM stage 0–3, the absolute risk of atherosclerotic CVD and heart failure is higher, but optimal strategies for risk assessment need to be refined [6]. Insulin resistance (IR) directly leads to abnormal lipid metabolism, enhanced inflammatory response and endothelial dysfunction [7]. Therefore, early identification of high-risk patients with IR in CKM population and early intervention to improve insulin sensitivity may help to improve the prognosis of patients.

IR is an important factor leading to adverse events of CVD, which interacts with other metabolic risk factors to promote various metabolic diseases and increase the risk of mortality [7]. Simple, noninvasive methods such as the homeostasis model assessment of IR (HOMA-IR) [8] and the triglycerid-glucose (TyG) index [9] are frequently used to assess insulin sensitivity. Among them, the TyG index is a simple indicator for assessing the risk of IR. It is calculated by fasting triglycerides (TG) and fasting blood glucose (FPG), and does not require insulin quantification. It is more widely used and has been proven to be related to cardiovascular diseases and prognosis [10–11]. In addition, body fat content and distribution, especially visceral fat, are also closely related to IR, CVD, and cardiovascular mortality [12–13]. A growing number of studies have combined TyG index with traditional obesity-related parameters such as body mass index (BMI) and waist-to-height ratio (WHTR) to determine whether it can improve risk stratification of cardiovascular outcomes. Compared with TyG index alone, the combination of TyG index and obesity index can provide better cardiovascular risk assessment [14], but other studies have come to the opposite conclusion [15–17]. Therefore, there is no consensus on using a combination of TyG index and traditional obesity-related parameters to predict cardiovascular risk.

Conventional anthropometric measures of abdominal obesity, including Waist circumference (WC) and WHTR, are closely related to BMI, limiting their ability to assess visceral fat content independently of BMI. A new anthropometric measure, a body shape index (ABSI), which uses BMI and height to normalize WC, focuses more on abdominal obesity by controlling for the confounding effects of weight and height. Thus, it shows a stronger association with increased cardiovascular risk and mortality [18–19]. The latest study constructed the simultaneous assessment of TyG index and ABSI and revealed a synergistic effect on cardiovascular mortality [20]. In addition, the association of TyG-ABSI with all-cause mortality and cardiovascular mortality was also found in individuals with hyperuricemia [21]. However, it is uncertain whether TyG-ABSI is better than TyG index and other TyG-derived indexes in predicting the risk of CKM syndrome, and whether TyG-ABSI can affect the risk of CVD and mortality in patients with CKM syndrome is still unknown.

Therefore, this study aims to explore the predictive effects of TyG-related indicators on the all-cause mortality and cardiovascular mortality of patients with CKM syndrome, and to further determine the association between TyG-ABSI and the stage of CKM syndrome, as well as to investigate the association between TyG-ABSI and the mortality risk of patients with CKM syndrome. It is helpful for clinicians to better evaluate the cardiovascular risk and prognosis of patients and formulate individualized treatment plans.

Methods

Study population and design

This study utilized data from two nationally representative cohorts: the discovery set (CHARLS) in China (https://charls.pku.edu.cn) and the validation set (NHANES) in the United States (https://www.cdc.gov/nchs/nhanes/index.htm). Both surveys evaluated the health, nutrition, and socioeconomic status of adults in China or the United States, and both adhered to the STROBE observational study guidelines. CHARLS project aims to collect a set of high-quality microdata representing households and individuals aged 45 and above in China to analyse the aging of China’s population and promote interdisciplinary research on aging. CHARLS national baseline survey was conducted in 2011 using the multi-stage probability to proportional to size (PPS) sampling method. The samples covered 450 villages, 150 counties, and 28 provinces, involving more than 17,000 people from about 10,000 households. The CHARLS is an ongoing survey with exams performed every 2 to 3 years. The participants were interviewed face-to-face in their homes through computer-assisted personal interviewing technology. The survey included basic demographic information of the respondents and their families, transfer payments between family members, health status of the respondents, medical care and insurance, employment, income, expenditure and assets, and so on. Besides, CHARLS included 13 physical measurements and blood sample collection. The baseline survey was conducted from June 2011 to March 2012 (Wave 1), followed by four follow-up waves in 2013 (Wave 2), 2015 (Wave 3), 2018 (Wave 4), and 2020 (Wave 5). The CHARLS survey project was approved by the Biomedical Ethics Committee of Peking University, and all participants were required to sign informed consent. The NHANES is an ongoing survey of the national population of the United States (US) that employs a complex, multistage and probabilistic sampling technique to provide a plethora of information on nutrition and health of the US population. The NHANES is a major program of the National Center for Health Statistics (NCHS). It was designed to assess the health and nutritional status of adults and children in the US. All study procedures were authorized by the Ethical Review Board of the National Center for Health Statistics before data collection, and all participants gave their signed informed consent. According to the inclusion and exclusion criteria shown in Fig. 1, the final analysis included 11,235 participants from CHARLS and 15,054 participants from NHANES. The flowchart of inclusion and exclusion of the participants is presented in Fig. 1.

Fig. 1.

Flow chart of the screening of eligible participants

Definition of CKM syndrome state

CKM syndrome definition highlights the multifaceted interactions between metabolic disorders, CKD, and CVD [1, 6]. Metabolic disorders include overweight or obesity, abdominal obesity, prediabetes, diabetes mellitus, hypertension, dyslipidemia, and metabolic syndrome. Clinical CVD was defined as a history of chronic heart failure, coronary heart disease, myocardial infarction, or stroke. Subclinical CVD was defined as 10-year risk ≥ 20% or high risk for CKD. (Table S1) The AHA PREVENT equation was used to calculate the rate of subclinical CVD [22]. We used the 2021 Chronic Kidney Disease Epidemiology Collaboration creatinine equation to estimate the estimated glomerular filtration rate (eGFR), which does not account for race or ethnicity [23]. (Table S2) The classification of CKM syndrome stages, which range from 0 to 4, follows the criteria detailed in the AHA Presidential Advisory Statement on CKM Syndrome [24]. Stage 0 indicates no abnormalities, whereas stage 1 is characterized by obesity or prediabetes alone. Stage 2 included individuals with at least one additional metabolic disorder or CKD. Stage 3 was defined as the presence of subclinical CVD and metabolic disorders or CKD. Finally, stage 4 represents clinical CVD with metabolic disorders or CKD. The stages are defined as follows in Table S3. In this study, CKM stage was further classified into No advanced CKM (stage 0–2) and advanced CKM (stage 3–4) [25].

Exposure variable

Trained medical professionals performed anthropometric measurements and laboratory tests. Participants fasted overnight to obtain fasting plasma glucose (FPG), fasting triglycerides (TG), high-density lipoprotein cholesterol (HDL-C) and fasting insulin values. Standard method is used to measure the height, weight and WC. The calculation of IR-related indices is as follows.

Outcome variable

The outcome variables included all-cause mortality and cardiovascular mortality. The death data of CHARLS were obtained through follow-up in 2018 and 2020. Mortality data for the follow-up population were sourced from the NHANES Public-use link mortality files, with updates extending through December 31, 2019. For more information on accessing the restricted use linked mortality files, please refer to the official website (https://www.cdc.gov/nchs/data-linkage). The observation time was defined as the duration between baseline assessment (initial interview) and the subsequent occurrence of either mortality or study completion. All-cause mortality encompasses deaths from any cause, including heart disease, malignant neoplasms, unintentional injuries, cerebrovascular diseases, diabetes mellitus, and other causes. Cardiovascular mortality specifically refers to deaths attributed to heart disease and cerebrovascular diseases, specifically the codes for heart disease (ICD-10 codes I00-I09, I11, I13, and I20-I51) or cerebrovascular disease (ICD-10 codes I60-I69).

Covariates

Demographic data, physical examination, laboratory tests, lifestyle habits, and medical conditions of the participants were collected. Demographic data included age, gender, [race, education, family income, and poverty income ratio (NHANES only)]. Physical examination included BMI, WC, standing height, systolic blood pressure (SBP) and diastolic blood pressure (DBP). Laboratory test data included alanine aminotransferase (ALT), aspartate aminotransferase (AST), FPG, glycosylated hemoglobin (HbA1c), TG, total cholesterol (TC), HDL-C, low-density lipoprotein cholesterol (LDL-C), white blood cells (WBC), C-reactive protein (CRP) and eGFR. Lifestyle and medical conditions included smoking, drinking, physical activity, use of antihypertensive drugs, use of antidiabetic drugs, use of insulin, use of statins, hypertension, diabetes, and stroke. For CHARLS, smoking status was determined by the question “Do you smoke?” and drinking status by the question “Did you drink any alcoholic beverages last year?”. In NHANES, race was classified as Mexican Spanish-speaking, non-Spanish White, non-Spanish Black, or other. Education levels were categorized as below high school, high school or equivalent, or above university. Smoking status was recorded as never (defined as having smoked fewer than 100 cigarettes in one’s lifetime), previously (defined as having smoked 100 or more cigarettes in one’s lifetime but not currently smoking), or currently (defined as having smoked 100 or more cigarettes in one’s lifetime and currently smoking). Hyperlipidemia was defined as TC ≥ 6.2 mmol/L (240 mg/dL), TG ≥ 2.3 mmol/L (200 mg/dL), LDL-C ≥ 4.1 mmol/L (160 mg/dL), or HDL-C ≤ 1.0 mmol/L (40 mg/dL) [11]. SBP and DBP were averaged over four measurements by experienced technicians at the heart level using automatic blood pressure monitors. SBP ≥ 130 mmHg or DBP ≥ 80 mmHg was defined as hypertension [30]. Self-reported physician diagnosis of hypertension, or the use of antihypertensive medication, was described as hypertension. Diabetes [31] was defined as occurring in cases where any of the following conditions were met: (1) fasting glucose ≥ 126 mg/dL; (2) random plasma glucose ≥ 200 mg/dL; (3) 2-h oral glucose tolerance test glucose ≥ 200 mg/dL; (4) HbA1c ≥ 6.5%; (5) self-reported diabetes physician diagnosis; or (6) use of anti-diabetic medication. Stroke was defined by self-reported previous diagnosis by a physician during the face-to-face interview. Anyone who answered “yes” to the following question: “Have you ever been told by a physician or a health professional that you had a stroke?” was considered to have a stroke. To be noted, the use of self-reported measures is prone to recall bias, which may impact the interpretation of the data.

Statistical analysis

Of the two databases, the CHARLS was used as the development cohort and the NHANES as the validation cohort. Impute the variables with a missing rate less than 15% using random forest in CHARLS. Statistical analyses followed NHANES guidelines that combined sample weights, stratification, and clustering to account for the complex survey design. TyG-ABSI was analyzed based on tertiles. Firstly, because of the non-normal distribution, continuous variables were described by median (1st Quartile, 3rd Quartile) according to the grouping of the study variables, and the Kruskal-Wallis test was used for group comparison. Categorical variables were described as counts and percentages, and group differences were assessed using Fisher’s exact test. Second, we used logistic regression to examine the relationship between IR-related indices (HOMA-IR, METS-IR, TyG index, TyG-BMI, TyG-WHtR, and TyG-ABSI) and CKM syndrome, Adjusting for age, gender, race and ethnicity, education level, family income, smoking, drinking, physical activity, and BMI. Third, Kaplan–Meier survival curves were used to illustrate the relationship between TyG-ABSI and all-cause and cardiovascular mortality in patients with CKM syndrome, and the log-rank test was used to test statistical differences. Fourth, to assess the association between TyG-ABSI and all-cause mortality and CVD mortality in patients with CKM syndrome, we utilized weighted univariate and multivariate Cox proportional hazards models with results expressed as hazard ratios (HR) and 95% confidence intervals (CI). Evaluation of multicollinearity using variance inflation factor (VIF) showed no significant multicollinearity problems. Fifth, based on Model 3, restricted cubic spline (RCS) regression was used to assess potential nonlinear associations. Furthermore, the “medflex” software package was used to analyze the mediating role of white blood cells (WBC) and C-reactive protein (CRP) in the association between TyG-ABSI and mortality among participants with CKM syndrome. Finally, to assess the robustness of the association between TyG-ABSI and CKM syndrome mortality, we conducted several sensitivity analyses to validate our main findings, including the association between TyG-ABSI and mortality outcomes after excluding patients with CKM syndrome who died during the first 2 years of follow-up in NHANES. Multivariate Logistics and Cox regression risk models were constructed using subgroup analysis. To estimate the effects of age, gender, smoking, drinking, physical activity, and different metabolic disease status (hypertension, hyperlipidemia, stroke, diabetes) on the association between TyG-ABSI and CKM syndrome stages and the risk of mortality in patients with different stages of CKM syndrome. All data processing and statistical calculations were performed with the use of R statistical software (Foundation for Statistical Computing, Vienna, Austria, version 4.2.3) and python (version 3.11). Two-sided P values < 0.05 were considered significant.

Results

Participants characteristics

The study included 11,235 participants with CKM syndrome from CHARLS and 15,054 participants with CKM syndrome from NHANES. For CHARLS, 9,800 (87.23%) with no advanced CKM and 1,435 (12.77%) with advanced CKM. Participants had a median age of 65 years. During a median follow-up of 5 years, 747 (6.65%) all-cause mortality and 84 (0.75%) cardiovascular mortality occurred. For NHANES, 12,676 (86.11%) with no advanced CKM and 2,378 (13.89%) with advanced CKM. Participants had a median age of 46 years. During a median follow-up of 10 years, 2,229 all-cause mortality (10.88%) and 611 cardiovascular mortality (2.86%) occurred. The baseline characteristics of the participants stratified by tertiles of the TyG-ABSI from CHARLS and NHANES were shown in Tables 1 and 2. The baseline characteristics of the participants stratified by CKM syndrome from CHARLS and NHANES were shown in Supplemental material (Table S4 and Table S5).

Table 1.

Baseline characteristics of study individuals according to the TyG-ABSI tertiles from CHARLS

Variables	Overall	T1	T2	T3	P value
	(N = 11235)	(n = 3745)	(n = 3745)	(n = 3745)
Gender, n (%)					< 0.001
Male	5253 (46.76)	1999 (53.38)	1725 (46.06)	1529 (40.83)
Female	5982 (53.24)	1746 (46.62)	2020 (53.94)	2216 (59.17)
Age, years, median (IQR)	65.00 (57.00,72.00)	63.00 (55.00,70.00)	65.00 (57.00,72.00)	67.00 (59.00,73.00)	< 0.001
Death, n (%)	747 (6.65)	212 (5.67)	227 (6.06)	308 (8.22)	< 0.001
CVD Death, n (%)	84 (0.75)	25 (0.67)	24 (0.64)	35 (0.93)	0.264
CKM syndrome stage, n (%)					< 0.001
No advance CKM	9800 (87.23)	3474 (92.76)	3273 (87.40)	3053 (81.52)
Advance CKM	1435 (12.77)	271 (7.24)	472 (12.60)	692 (18.48)
Smoking, n (%)					0.024
NO	10470 (93.19)	3458 (92.34)	3517 (93.91)	3495 (93.32)
YES	765 (6.81)	287 (7.66)	228 (6.09)	250 (6.68)
Disabetes, n (%)					< 0.001
NO	9622 (85.64)	3541 (94.55)	3319 (88.63)	2762 (73.75)
YES	1613 (14.36)	204 (5.45)	426 (11.38)	983 (26.25)
Hypertension, n (%)					< 0.001
NO	4667 (41.54)	1937 (51.72)	1564 (41.76)	1166 (31.14)
YES	6568 (58.46)	1808 (48.28)	2181 (58.24)	2579 (68.87)
Hyperlipidemia, n (%)					< 0.001
NO	7291 (64.90)	3170 (84.65)	2706 (72.26)	1415 (37.78)
YES	3944 (35.11)	575 (15.35)	1039 (27.74)	2330 (62.22)
Stroke, n (%)					0.001
NO	11081 (98.63)	3708 (98.75)	3692 (98.59)	3691 (98.56)
YES	154 (1.37)	47 (1.26)	53 (1.42)	54 (1.44)
Drink, n (%)					< 0.001
NO	7247 (64.50)	2285 (61.02)	2449 (65.39)	2513 (67.10)
YES	3988 (35.45)	1460 (38.99)	1296 (34.61)	1232 (32.90)
Physical activity, n (%)					0.050
NO	1090 (9.70)	340 (9.08)	351 (9.37)	399 (10.65)
YES	10145 (90.30)	3405 (90.92)	3394 (90.63)	3346 (89.35)
CKM syndrome stage, n (%)					< 0.001
0	1455 (12.95)	890 (23.77)	410 (10.95)	155 (4.14)
1	3317 (29.52)	1374 (36.69)	1053 (28.12)	890 (23.77)
2	5028 (44.75)	1210 (32.31)	1810 (48.33)	2008 (53.62)
3	656 (5.89)	115 (3.07)	217 (5.79)	324 (8.65)
4	779 (6.93)	156 (4.17)	255 (6.81)	368 (9.83)
TyG, median (IQR)	8.62 (8.25,9.09)	8.17 (7.96,8.41)	8.61 (8.37,8.86)	9.25 (8.92,9.67)	< 0.001
METS-IR, median (IQR)	35.02 (30.36,40.34)	31.31 (27.97,35.35)	35.14 (30.97,39.54)	39.38 (34.49,44.62)	< 0.001
TyG-BMI, median (IQR)	205.19 (179.40,234.60)	183.14 (164.60,204.71)	206.06 (183.03,229.00)	231.64 (204.96,258.61)	< 0.001
TyG-WHtR, median (IQR)	4.71 (4.16,5.29)	4.03 (3.70,4.38)	4.73 (4.37,5.07)	5.45 (5.04,5.86)	< 0.001
ABSI, median (IQR)	0.83 (0.80,0.86)	0.79 (0.76,0.81)	0.83 (0.81,0.85)	0.86 (0.84,0.89)	< 0.001
TyG-ABSI, median (IQR)	7.17 (6.68,7.70)	6.49(6.20,6.68)	7.17 (7.01,7.34)	7.94 (7.70,8.32)	< 0.001
TyG-WC, median (IQR)	743.14 (659.55,834.30)	640.03 (586.79,693.62)	748.25 (692.58,804.21)	857.57 (789.13,928.73)	< 0.001
WC, cm, median (IQR)	86.00 (78.80,93.00)	78.50 (72.50,84.50)	86.50 (80.80,92.50)	92.10 (86.20,98.50)	< 0.001
Weight, kg, median (IQR)	59.00 (52.00,67.10)	56.40 (50.30,63.50)	59.80 (52.60,67.60)	61.20 (53.30,70.30)	< 0.001
Height, cm, median (IQR)	157.80 (152.10,164.20)	158.50 (153.00,164.30)	157.90 (152.30,164.40)	156.90 (151.20,163.80)	< 0.001
FPG, mg/dl, median (IQR)	95.50 (88.29,106.31)	90.09 (84.69,97.30)	95.50 (88.29,102.70)	102.70 (93.69,124.32)	< 0.001
TC, mg/dl, median (IQR)	180.70 (158.69,205.41)	171.04 (151.35,193.82)	181.08 (159.85,204.25)	191.89 (167.95,217.00)	< 0.001
TG, mg/dl, median (IQR)	114.16 (82.30,169.91)	78.76 (64.60,98.23)	113.27 (90.27,145.13)	188.50 (140.71,268.14)	< 0.001
HDL-C, mg/dl, median (IQR)	49.81 (43.24,57.53)	53.67 (47.10,62.16)	49.81 (43.63,57.14)	46.33 (40.54,53.28)	< 0.001
BMI, Kg/m², median (IQR)	23.64 (21.30,26.23)	22.35 (20.34,24.69)	23.82 (21.49,26.30)	24.86 (22.39,27.28)	< 0.001
Crea, mg/dl, median (IQR)	0.76 (0.66,0.90)	0.77 (0.672,0.896)	0.76 (0.66,0.90)	0.75 (0.64,0.89)	< 0.001
eGFR, median (IQR)	90.50 (79.97,98.17)	91.44 (82.44,98.89)	90.47 (79.70,97.65)	89.37 (77.72,97.79)	< 0.001
WBC, 103/dL, median (IQR)	5.70 (4.77,6.88)	5.37 (4.49, 6.46)	5.79 (4.80, 6.80)	6.00 (5.09, 7.20)	< 0.001
CRP, mg/l, median (IQR)	1.40 (0.70, 2.60)	0.90 (0.50, 1.80)	1.30 (0.70, 2.30)	2.10 (1.20, 3.70)	< 0.001
LDL-C, mg/dl, median (IQR)	100.00 (82.24, 119.31)	95.37 (79.537, 113.13)	103.86 (86.49, 122.01)	101.54 (82.24, 122.39)	< 0.001
SBP, mmHg, median (IQR)	128.00 (116.00, 144.00)	124.00 (112.00, 139.00)	128.00 (116.00, 143.00)	133.00 (119.00, 148.00)	< 0.001
DBP, mmHg, median (IQR)	76.00 (68.00, 84.00)	74.00 (67.00, 82.00)	76.00 (69.00, 84.00)	77.00 (70.00, 86.00)	< 0.001
HbA1c,%, median (IQR)	5.80 (5.50, 6.10)	5.70 (5.40, 5.90)	5.80 (5.50, 6.10)	5.90 (5.60, 6.40)	< 0.001
Survivaltime, median (IQR)	4.92 (4.92, 5.00)	4.92 (4.92, 5.00)	4.92 (4.92, 5.00)	4.92 (4.92, 5.00)	0.087

Data are presented as Median (1st Quartile, 3rd Quartile) or number (proportion, %). T1: TyG-ABSI < 6.85; T2:6.85 ≤ TyG-ABSI < 7.509; T3: TyG-ABSI ≥ 7.51

SBP: systolic blood pressure; DBP: diastolic blood pressure; TC: total cholesterol; TG: triglycerides; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; HbA1c: glycated hemoglobin; WC: waist circumference; CKM, cardiovascular-kidney-metabolic; BMI, body mass index; EGFR, estimated glomerular filtration rate; TyG, triglyceride–glucose index; WHTR, waist-to-height ratio; ABSI: a body shape index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; WBC, white blood cell; METS-IR, metabolic score for insulin resistance; HOMA-IR, homeostasis model assessment of insulin resistance; T, tertiles; M: median; Q₁: 1st Quartile; Q₃: 3st Quartile

Table 2.

Baseline characteristics of study individuals according to the TyG-ABSI tertiles from NHANES

Variables	Overall (N = 15054)	T1 (n = 4606)	T2 (n = 4800)	T3 (n = 5648)	P value
Gender, n (%)					< 0.001
Male	7747 (51.44)	2193 (49.07)	2637 (54.79)	2917 (50.36)
Female	7307 (48.56)	2413 (50.93)	2163 (45.21)	2731 (49.64)
Age, years, M (Q1, Q3)	46.00 (33.00, 58.00)	39.00 (29.00, 50.00)	46.00 (33.00, 58.00)	53.00 (41.00, 66.00)	< 0.001
Age, years, n(%)					< 0.001
20–44	6208 (47.14)	2722 (62.72)	2045 (47.27)	1441 (31.44)
45–59	3928 (30.03)	1140 (26.34)	1334 (31.71)	1454 (31.99)
≥ 60	4918 (22.82)	744 (10.94)	1421 (21.01)	2753 (36.56)
Race, n (%)					< 0.001
Mexican American	2846 (8.50)	576 (6.76)	969 (9.17)	1301 (9.54)
Other Hispanic	1356 (5.59)	348 (5.09)	448 (5.38)	560 (6.31)
Non-Hispanic White	6525 (67.48)	1741 (62.42)	2151 (70.47)	2633 (69.47)
Non-Hispanic Black	2962 (11.40)	1649 (20.66)	797 (8.56)	516 (5.08)
Other Race - Including Multi-Racial	1365 (7.03)	292 (5.07)	435 (6.42)	638 (9.60)
Marital, n (%)					< 0.001
Married	8039 (56.89)	2175 (52.63)	2718 (59.92)	3146 (58.04)
Widowed	1171 (5.15)	176 (2.37)	292 (3.88)	703 (9.23)
Divorced	1510 (9.77)	446 (9.13)	439 (9.08)	625 (11.14)
Separated	502 (2.45)	175 (2.75)	141 (1.97)	186 (2.64)
Never married	2539 (17.68)	1141 (23.43)	791 (17.36)	607 (12.25)
Living with partner	1154 (8.06)	459 (9.70)	370 (7.79)	325 (6.70)
PIR, M (Q1, Q3)	2.92 (1.45, 4.93)	3.01 (1.51, 4.94)	3.11 (1.57, 5.00)	2.62 (1.31, 4.67)	< 0.001
Education, n (%)					< 0.001
Less than high school	1959 (6.70)	283 (3.94)	536 (5.73)	1140 (10.45)
High school	5759 (35.90)	1689 (33.12)	1866 (35.33)	2204 (39.26)
More than high school	7320 (57.34)	2634 (62.94)	1309 (53.81)	2293 (50.22)
Smoking, n (%)					< 0.001
No	8061 (52.98)	2831 (61.47)	2541 (52.24)	2689 (45.25)
Yes	6977 (47.02)	1768 (38.53)	2255 (47.76)	2954 (54.75)
Drinking, n (%)					< 0.001
No	4064 (24.29)	1227 (24.53)	1173 (21.48)	1664 (26.92)
Yes	10081 (75.71)	3061 (75.47)	3352 (78.52)	3668 (73.08)
Physical activity, n (%)					< 0.001
Low physical activity	4698 (33.71)	1332 (29.78)	1439 (32.04)	1927 (39.90)
High physical activity	7994 (66.29)	2756 (70.22)	2690 (67.96)	2548 (60.10)
BMI, kg/m2	27.83 (23.90, 32.30)	30.89 (25.90, 36.90)	27.80 (23.85, 31.60)	26.06 (22.81, 29.13)	< 0.001
EGFR	104.18 (87.20, 119.42)	107.56 (91.52, 121.92)	104.65 (87.79, 119.88)	99.92 (81.73, 115.96)	< 0.001
Diabetes, n (%)	2860 (14.59)	520 (9.00)	713 (11.37)	1627 (23.52)	< 0.001
Hyperlipidemia, n (%)	11261 (73.75)	2692 (58.12)	3614 (75.56)	4955 (87.51)	< 0.001
Hypertension, n (%)	6639 (38.98)	1828 (35.12)	1948 (36.10)	2863 (45.82)	< 0.001
Stroke, n (%)	492 (2.59)	94 (1.75)	137 (2.16)	261 (3.87)	< 0.001
CKM syndrome stage, n(%)					< 0.001
0	1267 (10.30)	548 (12.98)	467 (12.13)	252 (5.74)
1	2184 (15.75)	1051 (24.20)	770 (16.25)	363 (6.79)
2	9225 (60.06)	2526 (53.84)	2846 (57.97)	3853 (68.43)
3	902 (6.05)	202 (4.40)	292 (6.58)	408 (7.17)
4	1476 (7.83)	279 (4.58)	425 (7.06)	772 (11.87)
CKM syndrome stage					< 0.001
No advance CKM	12676 (86.11)	4125 (91.02)	4083 (86.35)	4468 (80.96)
Advance CKM	2378 (13.89)	481 (8.98)	717 (13.65)	1180 (19.04)
Tyg-ABSI	1.86 (1.73, 2.00)	1.67 (1.59, 1.73)	1.86 (1.82, 1.90)	2.07 (2.00, 2.17)	< 0.001
TyG, M (Q1, Q3)	8.65 (8.22, 9.07)	8.25 (7.90, 8.65)	8.64 (8.27, 8.99)	9.05 (8.67, 9.47)	< 0.001
TyG-BMI, M (Q1, Q3)	243.07 (202.17, 288.29)	256.45 (206.53, 314.84)	240.31 (199.14, 282.06)	236.29 (201.62, 274.70)	< 0.001
WHTR, M (Q1, Q3)	0.58 (0.51, 0.64)	0.59 (0.51, 0.68)	0.57 (0.51, 0.63)	0.57 (0.52, 0.63)	< 0.001
TYG-WHTR, M (Q1, Q3)	5.03 (4.31, 5.74)	4.90 (4.09, 5.76)	4.93 (4.26, 5.61)	5.21 (4.61, 5.86)	< 0.001
METS-IR, M (Q1, Q3)	41.60 (33.84, 50.37)	44.22 (35.46, 55.07)	41.03 (33.17, 49.16)	40.30 (33.38, 47.88)	< 0.001
HOMA-IR, M (Q1, Q3)	2.40 (1.44, 4.08)	2.34 (1.36, 4.12)	2.34 (1.42, 3.95)	2.51 (1.55, 4.20)	< 0.001
Glucose, mg/dL, M (Q1, Q3)	99.00 (92.00, 107.60)	97.00 (90.00, 104.00)	99.00 (92.10, 106.80)	102.00 (94.40, 113.20)	< 0.001
HbAc1,%, mean (SD)	5.40 (5.20, 5.70)	5.30 (5.10, 5.60)	5.40 (5.10, 5.70)	5.50 (5.20, 5.90)	< 0.001
SBP, mmol/L (mean (SD)	120.00 (112.00, 132.00)	120.00 (110.00, 130.00)	120.00 (110.00, 130.00)	124.00 (114.00, 138.00)	< 0.001
DBP, mmol/Hg, M (Q1, Q3)	72.00 (64.00, 78.00)	72.00 (64.00, 78.00)	72.00 (64.00, 78.00)	70.00 (64.00, 78.00)	< 0.001
TC, mg/dL, M (Q1, Q3)	194.00 (168.00, 222.00)	183.00 (160.00, 207.00)	195.00 (170.00, 221.00)	206.00 (178.00, 236.00)	< 0.001
TG, mg/dL, M (Q1, Q3)	112.00 (76.00, 166.00)	79.00 (57.00, 113.00)	111.00 (80.00, 156.00)	161.00 (112.00, 233.00)	< 0.001
LDL-C, mg/dL, M (Q1, Q3)	114.00 (92.00, 139.00)	110.00 (88.00, 132.00)	115.00 (94.00, 140.00)	118.00 (95.00, 144.00)	< 0.001
HDL-C, mg/dL, M (Q1, Q3)	50.00 (42.00, 62.00)	53.00 (44.00, 63.00)	50.00 (42.00, 62.00)	47.00 (39.00, 59.00)	< 0.001
ALT, U/L, M (Q1, Q3)	22.00 (17.00, 30.00)	21.00 (17.00, 29.00)	22.00 (17.00, 30.00)	22.00 (17.00, 30.00)	< 0.001
AST, U/L, M (Q1, Q3)	23.00 (19.00, 27.00)	22.00 (19.00, 27.00)	23.00 (19.00, 27.00)	23.00 (20.00, 28.00)	< 0.001
CRP, mg/dL, M (Q1, Q3)	0.20 (0.08, 0.45)	0.21 (0.08, 0.53)	0.18 (0.07, 0.41)	0.20 (0.09, 0.43)	< 0.001
WBC,103/uL, M (Q1, Q3)	6.50 (5.50, 7.80)	6.30 (5.30, 7.60)	6.40 (5.40, 7.70)	6.80 (5.70, 8.10)	< 0.001
All-cause mortality, n (%)	2229 (10.88)	293 (5.01)	527 (8.25)	1409 (19.45)	< 0.001
Cardiovascular mortality, n (%)	611 (2.86)	86 (1.38)	149 (2.23)	376 (4.97)	< 0.001

Data are presented as Median (1st Quartile, 3rd Quartile) or number (proportion, %). T1: TyG-ABSI < 1.78; T2:1.78 ≤ TyG-ABSI < 1.95; T3: TyG-ABSI ≥ 1.95

SBP: systolic blood pressure; DBP: diastolic blood pressure; TC: total cholesterol; TG: triglycerides; HDL-C: High-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; HbA1c: glycated hemoglobin; PIR: income to poverty ratio; WC: waist circumference; CKM, cardiovascular-kidney-metabolic; BMI, body mass index; EGFR, estimated glomerular filtration rate; TyG, triglyceride–glucose index; WHTR, waist-to-height ratio; ABSI: a body shape index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; WBC, white blood cell; METS-IR, metabolic score for insulin resistance; HOMA-IR, homeostasis model assessment of insulin resistance; T, tertiles; M: median, Q₁: 1st Quartile, Q₃: 3st Quartile

Association of TyG-ABSI with CKM syndrome stage

Compared with no advanced CKM, after multivariable adjusted, logistic regression analysis to examine the association between IR-related indices (HOMA-IR, METS-IR, TyG index, TyG-BMI, TyG-WHtR, and TyG-ABSI) and CKM syndrome. Logistic regression analysis showed that TyG-ABSI had the highest risk associated with CKM syndrome (OR 4.50; 95% CI 2.77–7.32), and the ROC curve results showed that the area under the ROC curve of TyG-ABSI was the largest (Figure S1A, Figure S2, Table S6). As shown in Table 3, In CHARLS, When TyG-ABSI was analyzed as a continuous variable, compared with no advanced CKM syndrome, TyG-ABSI was independently associated with advanced CKM syndrome after multivariable adjusted models (OR 1.55; 95% CI 1.35–1.79). On further analysis, in which continuous TyG-ABSI were converted to classified variable (tertile), compared to those with T1 group, the risk of advanced CKM syndrome was found to be 2.41-fold higher in those with T3 group (OR 2.41; 95% CI 1.82–3.20). And the P for trend of TyG-ABSI with CKM syndrome stage in the unadjusted or adjusted model was consistent with results obtained when the TyG-ABSI served as a continuous variable (P-trend < 0.001). Odds ratio (95% CI) of TyG-ABSI and advanced CKM syndrome stage from NHANES was similarly association.

Table 3.

Odds ratio (95% CI) of TyG-ABSI and advanced CKM syndrome stage

	Model 1		Model 2
	OR (95% CI)	P value	OR (95% CI)	P value
Variable (CHARLS)
TyG-ABSI	1.68 (1.54–1.84)	< 0.001	1.55 (1.35–1.79)	< 0.001
T1	1.00 (Reference)		1.00 (Reference)
T2	1.90 (1.58–2.29)	< 0.001	1.78 (1.34–2.37)	< 0.001
T3	2.94 (2.47–3.51)	< 0.001	2.41 (1.82–3.20)	< 0.001
P-trend		< 0.001		< 0.001
Variable (NHANES)
TyG-ABSI	5.18 (4.11–6.53)	< 0.001	4.34 (2.98–6.30)	< 0.001
T1	1.00 (Reference)		1.00 (Reference)
T2	1.60 (1.38–1.86)	< 0.001	1.50 (1.20–1.87)	< 0.001
T3	2.38 (2.07–2.75)	< 0.001	2.03 (1.61–2.55)	< 0.001
P-trend		< 0.001		< 0.001

Compared with no advanced CKM stage. OR, odds ratio; CI, confidence interval; T, tertiles; CKM, cardiovascular-kidney-metabolic; TyG, triglyceride–glucose index; ABSI: a body shape index. In CHARES, Model 1: unadjusted. Model 2: adjusted for gender, age, BMI, SBP, DBP, drink, smoking, physical activity, Crea, WBC, CRP, HDL, and eGFR. In NHANES, Model 1: unadjusted. Model 2: adjusted for age, gender, race, marital, education, family income, smoking, drinking, physical activity, and BMI

Association of TyG-ABSI with all-cause mortality and cardiovascular mortality among patients with CKM syndrome

Similarly, after multivariable adjusted, Cox proportional analysis showed that TyG-ABSI had the highest risk in association with all-cause mortality (HR 2.77; 95% CI 1.97–3.11) and cardiovascular mortality (HR 2.77; 95% CI 1.71–4.48) among patients with CKM syndrome (Figure S1B, Table S6). In CHARLS, When TyG-ABSI was analyzed as a continuous variable, TyG-ABSI was independently associated with all-cause mortality (HR 1.14; 95% CI 1.04–1.35) in the patients with CKM syndrome after multivariable adjusted. When analyzed as categorical variables, compared with T1 individuals, T3 individuals had a 55% increased risk of all-cause mortality (HR 1.55; 95% CI 1.10–2.18) (Table 4). In NHANES, when TyG-ABSI was analyzed as a continuous variable, TyG-ABSI was independently associated with all-cause mortality (HR 2.77; 95% CI 1.97–3.91) and cardiovascular mortality (HR 2.77; 95% CI 1.71–4.48) in the patients with CKM syndrome after multivariable adjusted. When analyzed as categorical variables, compared with T1 individuals, T3 individuals had a 43% and 56% increased risk of all-cause mortality (HR 1.43; 95% CI 1.12–1.83) and cardiovascular mortality (HR 1.56; 95% CI 1.07–2.26), respectively (Table 5). In addition, TyG-ABSI was independently associated with all-cause mortality and cardiovascular mortality in patients with different CKM syndrome. (Table S8)

Table 4.

Hazard ratio (95% CI) of TyG-ABSI and all-cause mortality, cardiovascular mortality among participants with CKM syndrome from CHARLS

Variable (CHARLS)	Model 1		Model 2		Model 3
	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
All-cause mortality
TyG-ABSI	1.21 (1.08–1.37)	0.001	1.16 (1.07–1.25)	< 0.001	1.14 (1.04–1.35)	0.004
T1	1.00 (Reference)		1.00 (Reference)		1.00 (Reference)
T2	1.00 (0.80–1.27)	0.431	1.13 (0.81–1.53)	0.663	1.12 (0.81–1.54)	0.665
T3	1.48 (1.21–1.82)	< 0.001	1.52 (1.09–2.12)	0.005	1.55 (1.10–2.18)	0.008
P-trend		< 0.001		< 0.001		< 0.001
Cardiovascular mortality
TyG-ABSI	1.14 (0.86–1.50)	0.091	1.32 (0.93–1.87)	0.157	1.31 (0.89–1.93)	0.137
T1	1.00 (Reference)		1.00 (Reference)		1.00 (Reference)
T2	0.66 (0.35–1.29)	0.906	2.35 (0.82–6.74)	0.160	2.50 (0.85–7.36)	0.167
T3	1.06 (0.59–2.93)	0.203	2.12 (0.68–6.68)	0.130	2.20 (0.61–7.87)	0.121
P-trend		0.386		0.136		0.156

Model 1: unadjusted

Model 2: adjusted for gender, age, BMI, SBP, DBP, physical activity, drink, and smoking

Model 3: adjusted for Model 2 and Crea, WBC, CRP, HDL, and eGFR

HR, hazard ratio; CI, confidence interval; T, tertiles; CKM, cardiovascular-kidney-metabolic; TyG, triglyceride–glucose index; ABSI: a body shape index

Table 5.

Hazard ratio (95% CI) of TyG-ABSI and all-cause mortality, cardiovascular mortality among participants with CKM syndrome from NHANES

Variable (NHANES)	Model 1		Model 2		Model 3
	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
All-cause mortality
TyG-ABSI	12.61 (9.01–17.64)	< 0.001	3.23 (2.30–4.54)	< 0.001	2.77 (1.97–3.91)	< 0.001
T1	1.00 (Reference)		1.00 (Reference)		1.00 (Reference)
T2	1.54 (1.28–1.86)	< 0.001	1.04 (0.84–1.28)	0.708	1.02 (0.83–1.27)	0.831
T3	3.73 (3.08–4.50)	< 0.001	1.53 (1.22–1.93)	< 0.001	1.43 (1.12–1.83)	0.004
P-trend		< 0.001		0.002		0.020
Cardiovascular mortality
TyG-ABSI	12.26 (8.60–17.47)	< 0.001	3.77 (2.37–5.99)	< 0.001	2.77 (1.71–4.48)	< 0.001
T1	1.00 (Reference)		1.00 (Reference)		1.00 (Reference)
T2	1.61 (1.21–2.14)	< 0.001	1.17 (0.82–1.69)	0.384	1.12 (0.78–1.62)	0.528
T3	3.92 (2.98–5.16)	< 0.001	1.77 (1.23–2.54)	0.002	1.56 (1.07–2.26)	0.021
P-trend		< 0.001		< 0.011		0.084

Model 1: unadjusted

Model 2: adjusted for age, gender, race, marital, education, family income, smoking, drinking, physical activity, and BMI

Model 3: adjusted for Model 2, diabetes, hyperlipidemia, hypertension, stroke, and eGFR

HR, hazard ratio; CI, confidence interval; T, tertiles; CKM, cardiovascular-kidney-metabolic; TyG, triglyceride–glucose index; ABSI: a body shape index

The Kaplan–Meier survival analysis curves for the primary outcome rates among the groups stratified by the tertiles of the TyG-ABSI index are presented in Figs. 2 and 3. The survival probability exhibited an decrease as TyG-ABSI index increased in the patients with overall CKM syndrome and no advanced CKM (Log rank P < 0.001). In addition, the same trend was observed for all-cause mortality and cardiovascular mortality in the patients with no advanced CKM and advanced CKM syndrome from NHANES (Log rank P < 0.001). However, there was no statistical difference among the patients from the CHARLS study (Log rank P>0.05).

Fig. 2.

Kaplan–Meier survival curves between the TyG-ABSI and all-cause mortality, cardiovascular mortality among participants with CKM syndrome. All-cause mortality and cardiovascular mortality among overall CKM syndrome from CHARLS (A, B) and NHANES (C, D)

Fig. 3.

Kaplan–Meier survival curves between the TyG-ABSI and all-cause mortality, cardiovascular mortality among participants with different CKM syndrome. All-cause mortality and cardiovascular mortality among no advanced CKM syndrome from CHARLS (A, B) and NHANES (C, D). All-cause mortality and cardiovascular mortality among advanced CKM syndrome from CHARLS (E, F) and NHANES (G, H)

Dose-response relationship between TyG-ABSI and all-cause mortality and cardiovascular mortality

According to RCS analysis, the association between TyG-ABSI and all-cause mortality and cardiovascular mortality in patients with CKM syndrome stage was J-shaped after adjusting for gender, age, BMI, SBP, DBP, drink, smoking, physical activity, Crea, WBC, CRP, HDL, and eGFR. And there was no nonlinear association between the risk of all-cause mortality and TyG-ABSI among CKM syndrome stage (Fig. 4 ).

Fig. 4.

Dose-response relationship between TyG-ABSI and all-cause mortality, cardiovascular mortality among CKM syndrome stage according to multivariable-adjusted restricted cubic spline analysis. All-cause mortality and cardiovascular mortality among overall CKM syndrome from CHARLS (A, B) and NHANES (G, H). All-cause mortality and cardiovascular mortality among no advanced CKM syndrome from CHARLS (C, D) and NHANES (I, J). All-cause mortality and cardiovascular mortality among advanced CKM syndrome from CHARLS (E, F) and NHANES (K, L). Multivariable-adjusted for gender, age, BMI, SBP, DBP, drink, smoking, physical activity, Crea, WBC, CRP, HDL, and eGFR in CHARLS. Multivariable-adjusted for age, gender, race, marital, education, family income, smoking, drinking, physical activity, BMI, hypertension, diabetes, hyperlipidemia, stroke, and eGFR in NHANES

Mediation analysis

The mediation analysis results suggested that the relationship between TyG-ABSI and all-cause mortality risk is partially mediated by WBC and CRP. the proportion of mediation were 15.16%, and 11.83%. (Fig. 5)

Fig. 5.

Mediation analysis of mortality risk factors in CHARLS

Subgroup analysis

To further explore the association between TyG-ABSI index and mortality among patients with CKM syndrome, we performed subgroup and interaction analyses stratified by age, gender, smoking, drinking, physical activity status and various comorbidities, including hypertension, diabetes, hyperlipidemia, stroke, and different stages of CKM syndrome. In the subgroup analysis between the TyG-ABSI index and patients with CKM syndrome, significant interactions were found only among patients with hypertension (P-interaction < 0.001) but not in other subgroups (P-interaction > 0.05) (Supplementary Figure S3-S5). No significant interaction was detected in the subgroup analysis of TyG-ABSI index and mortality of patients with CKM syndrome (P -interaction > 0.05). The association between TyG-ABSI index and all-cause mortality in patients with CKM syndrome was relatively stable across subgroups, but for cardiovascular mortality, the relationship was significantly different in patients aged 45–59 years.

Sensitivity analysis

To determine the robustness of the association of TyG-ABSI with all-cause mortality and cardiovascular mortality in patients with CKM syndrome state, in sensitivity analyses, the association between TyG-ABSI and mortality outcomes remained significant after excluding patients with CKM syndrome who died during the first 2 years of follow-up in NHANES (Tables S9 and S10).

Discussion

This is the first study to investigate different IR-related indices with the risk of CKM syndrome stage and mortality, and TyG-ABSI was found to have the highest association. This study evaluated the association between the TyG-ABSI and all-cause and cardiovascular mortality using data from two nationally representative cohorts, CHARLS and NHANES. In this study, the relationship between TyG-ABSI and CKM syndrome, and the association between TyG-ABSI and all-cause mortality and cardiovascular mortality in patients with CKM syndrome were further studied. The present study found that a higher TyG-ABSI was associated with a higher risk of CKM syndrome.

Several studies have shown that TyG index and its modified indices (TyG-WHTR, TyG-WC, TyG-BMI) are significantly associated with the risk of all-cause mortality in patients with CKM stage 0–3. Moreover, the association with the risk of all-cause mortality is stronger in CKM stage 1–3 [32]. In addition, TyG related indices can be used as independent predictors of the incidence of CVD in patients with CKM stage 0–3, and the modified model of TyG combined with WC or WHTR shows better predictive efficiency [33–34]. This study compared the IR-related indices (HOMA-IR, METS-IR, TyG index, TyG-BMI, TyG-WHtR, and TyG-ABSI) with the risk of CKM syndrome stage and mortality. The combination of TyG-ABSI was found to be a better predictor of advanced CKM syndrome. Consistent with previous studies, TyG index may be a useful tool to assess the risk of advanced CKM syndrome [35], and TyG-BMI can be used as an important indicator for risk stratification and prognosis prediction in patients with CKM stage 4 [36]. The integration of TyG index associated with atherosclerosis risk and ABSI independently predicts CVD, which may more fully capture the heterogeneity of metabolic syndrome. Each person varies in the levels of different biomarkers, and joint index analysis can be performed according to the specific individual. Moreover, ABSI is not affected by the “obesity paradox” caused by IR [18]. Mechanistically, the synergistic effect of TyG-ABSI may drive the progression of CKM syndrome through multiple pathophysiological pathways. On the one hand, IR directly leads to abnormal lipid metabolism, enhanced inflammatory response and endothelial dysfunction. Hyperinsulinemia induced by IR accelerates kidney injury by promoting renal cell proliferation, activating angiotensin II receptor and releasing growth factors. In addition, IR can reduce the production of nitric oxide, leading to endothelial dysfunction and metabolic abnormalities [37–38]. On the other hand, obesity further exacerbates metabolic imbalance by releasing free fatty acids and pro-inflammatory cytokines through ectopic deposits in adipose tissue. Obesity-related pro-inflammatory factors (such as TNF-α and IL-6) and free fatty acids produced by lipolysis further induce local inflammation, oxidative stress and β-cell dysfunction, which aggravate cardiovascular and metabolic disorders. Decreased levels of lipocalin weaken anti-inflammatory and anti-atherosclerotic ability, forming a vicious circle [39–40]. In conclusion, the TyG-ABSI combination may better predict advanced CKM syndrome and its all-cause mortality and cardiovascular mortality, and metabolic syndrome should be evaluated in multiple domains.

In subgroup and sensitivity analyses, higher TyG-ABSI index was independently associated with an increased risk of advanced CKM syndrome. Females and unhealthy lifestyles may have a higher risk of advanced CKM syndrome. Compared with males, females with metabolic syndrome are more likely to suffer from CVD [41]. Among participants with impaired glucose tolerance, females had a higher risk of coronary heart disease relative to males. These findings partially shed light on the higher prevalence of CVD risk factors in females, which is consistent with our current observations [29]. A gradual decline in metabolic capacity in people may result in weight gain, adipose tissue accumulation, and muscle mass loss. This process is associated with an increased risk of developing chronic diseases, thereby exacerbating the progression of the CKM syndrome. However, there was an interaction in people with or without hypertension, which may be because TyG-ABSI mainly reflects metabolic risk in people without hypertension, while its predictive value in the hypertensive subgroup was significantly enhanced by superimposed hemodynamic impairment rather than simple fat accumulation. In sensitivity analyses, we excluded patients who presented with death in the first 2 years, indicating the generalizability and stability of our findings across all populations in the context of cardiovascular mortality analysis. In conclusion, these results highlight the need to emphasize the association of TyG-ABSI with the risk of advanced CKM syndrome and mortality in individuals with different gender, age, lifestyle and metabolic state.

The clinical significance of this study lies in that the TyG-related parameters provide a readily accessible and cost-effective risk stratification tool for patients with CKM syndrome. The relevant parameters can be calculated through simple blood tests and physical examinations. In clinical practice, middle-aged patients with hypertension and mild obesity but with normal blood lipids may have a moderate overall risk score. However, if the TyG-ABSI calculation reveals a significant increase, it may indicate severe IR, which suggests that their actual cardiovascular metabolic risk has been underestimated. As a result, this patient can benefit from more active lifestyle interventions as well as earlier drug interventions. Similarly, for patients with CKM syndrome who already have cardiovascular diseases or diabetes, continuous monitoring of the TyG index can serve as a dynamic biomarker for evaluating the state of IR and the effectiveness of lifestyle interventions. TyG-ABSI can more comprehensively reflect the metabolic risks caused by excessive fat deposition, and is helpful in identifying special high-risk groups such as normal-weight metabolic obesity”. In order to realize the application of the TyG index in clinical practice, future research will integrate the TyG index into existing cardiovascular risk prediction models (such as the PCE score) to verify whether it can enhance the discriminative ability of the model. Conduct a prospective intervention study to explore whether risk stratification based on elevated TyG index can guide more effective individualized treatment and ultimately improve the endpoint outcome. Although its wide application still faces challenges in standardization and clinical recognition, this study provides strong multi-cohort evidence for the promotion of this inexpensive and informative biomarker.

This is a prospective, large-sample cohort study, which is the first to estimate the risk of TyG-ABSI and the association between TyG-ABSI and mortality outcomes in patients with CKM syndrome, expanding the clinical significance of TyG-ABSI. However, this study also has some limitations. First, the TyG-ABSI was assessed at baseline rather than longitudinally throughout the study. Second, the NHANES database does not provide more specific details on cardiovascular mortality (e.g., fatal myocardial infarction or fatal stroke). Therefore, they could not be analyzed in our study. Finally, there may be other unmeasured residual confounding factors, such as genetic predisposition.

Conclusion

TyG-ABSI is a reliable surrogate marker for IR, there is a significant positive association between the risk of TyG-ABSI and advanced CKM syndrome, and TyG-ABSI is a strong predictor of all-cause mortality and cardiovascular mortality in individuals with CKM syndrome. TyG-ABSI included in the basic risk model can significantly delay the staging of CKM and improve the prognosis of patients, and promote the clinical application of precise management of chronic diseases. In addition, future prospective studies should further validate these findings in larger and more diverse cohorts to explore potential mechanistic pathways underlying the pathophysiology between TyG-ABSI and CKM syndrome.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ABSI

A body shape index

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

BMI

Body mass index

CHARLS

China Health and Retirement Longitudinal Study

CI

Confidence intervals

CKM

Cardiovascular-kidney-metabolic

CRP

C-reactive protein

CVD

Cardiovascular disease

DBP

Diastolic blood pressure

eGFR

Estimated glomerular filtration rate

FPG

Fasting plasma glucose

HbA1c

Glycated haemoglobin

HDL-C

High-density lipoprotein cholesterol

HOMA-IR

Homeostasis model assessment of insulin resistance

HR

Hazard ratios

IR

Insulin resistance

LDL-C

Low-density lipoprotein cholesterol

METS-IR

Metabolic score for insulin resistance

NHANES

National Health and Nutrition Examination Survey

OR

Odds ratio

PIR

Income to poverty ratio

RCS

Restricted cubic spine

SBP

Systolic blood pressure

TC

Total cholesterol

TG

Triglycerides

TyG

Triglyceride–glucose index

T2DM

Type 2 diabetes

WBC

White blood cells

WC

Waist circumference

WHTR

Waist-to-height ratio

Author contributions

Zhu Li participated in the study design, statistical analysis, analyzed the data together and drafted the manuscript; Xiang Fan, Zhu Li, and Yige Wu participated in the revision of the manuscript; Yunqing Yang, Lei Wang, Shihao Dou, and Fangming Sun participated in data verification. All authors have reviewed and approved the final manuscript.

Funding

This work was supported by National Natural Science Foundation of China (82405046), Zhejiang medicine and health science and technology project (2025KY949), Natural Science Foundation of Zhejiang Province (LQH25H270015) and the Talent Special Project of Zhejiang Chinese Medical University (2023RCZXZK01) .

Data availability

The datasets generated and analyzed during the current study are available in the CHARLS and NHANES website, available in http://charls.pku.edu.cn/en and https://www.cdc.gov/nchs/nhanes/index.htm, respectively.

Declarations

Ethics approval and consent to participate

The NHANES is approved by the National Center for Health Statistics Research Ethics Review Board, and all participants provide informed consent. The CHARLS is approved by the Biomedical Ethics Review Committee of Peking University, and all participants provide informed consent.

Competing interests

The authors declare no competing interests.