Cumulative visceral fat burden and progression to advanced cardiovascular kidney metabolic syndrome in Chinese middle-aged and older adults Authors: Zihan Fang, Anqi Xue, Qi Liu, Ran He, Danna Zheng, Juan Jin, Qiang He Published: 2026 ABSTRACT Visceral adiposity is increasingly recognized as a key contributor to cardiovascular–kidney–metabolic (CKM) syndrome, yet longitudinal evidence regarding both baseline visceral fat burden and cumulative metabolic exposure remains limited. The metabolic score for visceral fat (METS-VF) is a validated surrogate marker reflecting visceral adiposity and metabolic dysfunction. We investigated the associations of baseline METS-VF and cumulative METS-VF (cumMETS-VF) with progression to advanced CKM syndrome in a nationally representative cohort. We analyzed 3,322 middle-aged and older adults from the China Health and Retirement Longitudinal Study (CHARLS) who underwent repeated metabolic assessments in 2012 and 2015. Exposures included baseline METS-VF and cumMETS-VF derived from repeated measurements across the two waves. The outcome was incident advanced CKM syndrome (Stages 3–4). Multivariable logistic regression, restricted cubic spline analyses, threshold-effect analyses, receiver-operating-characteristic (ROC) analyses, subgroup analyses, and sensitivity analyses were performed. Threshold stability was further evaluated using Bootstrap resampling. During approximately four years of follow-up, 20.2% of participants progressed to advanced CKM syndrome. Higher baseline METS-VF and cumMETS-VF were independently associated with increased risk of advanced CKM syndrome after multivariable adjustment, with ORs of 1.13 (95% CI 1.06–1.22) and 1.06 (95% CI 1.04–1.09) per one-unit increment, respectively. Significant nonlinear associations were observed for both exposures (all P_non-linear < 0.001). Risk increased more markedly beyond identified thresholds of 6.71 for baseline METS-VF and 20.11 for cumMETS-VF. These threshold estimates were generally consistent across restricted cubic spline analyses, likelihood-based threshold models, ROC-derived optimal cutoffs, and Bootstrap resampling analyses. Compared with conventional adiposity indices, METS-VF and cumMETS-VF showed comparable or slightly better discrimination for advanced CKM syndrome progression (AUC approximately 0.60). Findings remained generally consistent across subgroup and sensitivity analyses. In this nationally representative cohort, both baseline and cumulative visceral adiposity burden assessed by METS-VF were independently associated with progression to advanced CKM syndrome. The observed nonlinear and threshold relationships suggest that sustained elevations in visceral adiposity burden may contribute to CKM syndrome progression. In comparative analyses, METS-VF demonstrated slightly better discriminatory performance than traditional adiposity indices. Together, these results suggest that METS-VF and cumulative METS-VF may contribute to improved risk stratification for identifying individuals at elevated risk.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-58351-y. Introduction CKM syndrome, formally defined by the American Heart Association (AHA) in 2023, describes a constellation of interconnected health disturbances arising from metabolic dysfunction, chronic kidney disease (CKD), and cardiovascular disease (CVD)1. Under this framework, CKM syndrome is classified into five stages (0–4); Stage 1 or higher denotes its presence, while Stages 3–4 indicate advanced CKM syndrome. The global burden of CKM syndrome continues to escalate, largely driven by the growing prevalence and frequent coexistence of its core components2,3. The CKM syndrome construct underscores the bidirectional, synergistic interactions across metabolic, renal, and cardiovascular systems—interactions that amplify disease progression and worsen clinical outcomes4. CKM syndrome has emerged as a major public health challenge: in the United States, nearly 90% of adults meet criteria for at least Stage 1 CKM syndrome, and 15% have progressed to advanced stages5. The burden is even heavier in China, where approximately 90% of middle-aged and older adults are affected, and the prevalence of advanced CKM syndrome reaches 26.5%6. Notably, individuals with advanced CKM syndrome experience substantially higher mortality compared with those in earlier stages7. Against this backdrop of rapidly rising prevalence, early and accurate identification of high-risk individuals—and timely intervention to prevent progression to advanced organ damage, heart failure, end-stage kidney disease, and cardiovascular mortality—has become a critical priority for clinical research and population health efforts8. Among the various proposed drivers of CKM syndrome, visceral adiposity has emerged as a factor of particular pathophysiological significance. It represents a pivotal pathological nexus linking metabolic dysfunction with cardiovascular injury and renal decline, and has therefore attracted growing scientific attention9–11. In contrast to general obesity, visceral fat is highly metabolically active and enriched with pro-inflammatory cytokines, adipokines, and free fatty acids. By promoting insulin resistance, endothelial dysfunction, sympathetic overactivation, and increased renal hemodynamic stress, visceral adiposity amplifies the multisystem cascade central to CKM syndrome progression12–15. Consequently, quantitative assessment of visceral fat burden has become essential for elucidating the biological underpinnings of CKM syndrome and enabling timely, targeted early intervention. Although conventional anthropometric measures—such as body mass index (BMI) and waist circumference (WC)—are widely used in clinical and epidemiologic settings to estimate overall adiposity, they fall short in capturing the true extent of visceral fat accumulation. Their major limitation is the inability to differentiate subcutaneous from visceral depots or to characterize the intra-abdominal distribution, metabolic activity, and inflammatory profile of adipose tissue16,17. To address these gaps, METS-VF derived from metabolic parameters including the Metabolic Score for Insulin Resistance (METS-IR), WHtR, age, and sex, has emerged as a low-cost, reproducible, and imaging-concordant surrogate marker of visceral adiposity18. Prior cohort studies have demonstrated its associations with elevated risks of hypertension, cardiovascular disease, and all-cause mortality19. Crucially, the cumulative burden of visceral adiposity—reflecting the duration and magnitude of exposure—may better represent chronic metabolic stress and its lasting impact on cardiac, renal, and metabolic systems than single-time measurements20. Accordingly, the development of the cumMETS-VF provides an important new tool for investigating the causal links between long-term visceral fat exposure and CKM syndrome progression. Most existing research relies on single time-point measures of obesity or metabolic status, which fail to reflect the dynamic nature of metabolic load and its true impact on organ-level injury. Moreover, robust, cohort-based evidence quantifying the associations between METS-VF, its cumulative exposure, and the risk of advanced CKM syndrome remains scarce among middle-aged and older adults. Against this backdrop, our study systematically evaluated the predictive value of METS-VF and cumMETS-VF for advanced CKM syndrome risk. These findings not only support more precise identification of individuals at highest risk but also offer essential insights for designing intervention strategies that consider the temporal dimension of metabolic exposure. Study population and data source The China Health and Retirement Longitudinal Study (CHARLS) is a nationally representative cohort designed to track long-term trends in health status, socioeconomic conditions, and demographic characteristics among Chinese adults. Using a multistage, stratified probability sampling strategy, the study recruited 17,708 participants from 10,257 households across 450 communities and 150 counties/districts in 28 provinces, ensuring broad geographic coverage and strong national representativeness. The baseline survey was conducted between June 2011 and March 2012 (Wave 1), with subsequent follow-up waves completed in 2013, 2015, 2018, and 2020. The study protocol was approved by the Institutional Review Board of Peking University (IRB00001052-11015), and all participants provided written informed consent prior to data collection. Study sample selection We initially considered all 17,708 participants from Wave 1 as the baseline cohort. After excluding individuals younger than 45 years, those with missing information required for CKM syndrome staging, and those lacking data necessary to compute METS-VF, 8,945 participants remained for descriptive analyses. Because blood samples were available only in 2011 and 2015, longitudinal analyses were restricted to these two waves. Additional exclusions were made for participants who had advanced CKM syndrome at baseline and for those in 2015 with missing CKM syndrome staging or incomplete data for METS-VF calculation, resulting in a final analytic sample of 3,322 participants for the prospective analyses. Exposure definition The exposures of interest were: The baseline METS-VF measured in 2012.The cumulative METS-VF (cumMETS-VF) from 2012 to 2015. The baseline METS-VF measured in 2012. The baseline METS-VF measured in 2012. The cumulative METS-VF (cumMETS-VF) from 2012 to 2015. The cumulative METS-VF (cumMETS-VF) from 2012 to 2015. Cumulative exposure was derived by integrating METS-VF measurements across the two waves, thereby capturing the time-accumulated burden of visceral adiposity. The computation formula is presented below: WHtR=WC/Height\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:METS-IR=\left[ln\right((2\times\:FBG)+TG)\times\:BMI]/\left[ln\right(HDL-C\left)\right]$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:METS-VF=4.466+0.011\times\:\left[{\left(ln\right(METS-IR\left)\right)}^{3}\right]+3.239\times\:\left[{\left(ln\right(WHtR\left)\right)}^{3}\right]+0.319\times\:sex+0.594\times\:\left[ln\right(age\left)\right]$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:CumMETS-VF=\left(\right({METS-VF}_{2012}+{METS-VF}_{2015})/2)\times\:time(2015-2012)$$\end{document} Specifically, fasting blood samples were collected by trained nurses after at least 8 h of fasting. Samples were processed within 2 h using centrifugation at local Centers for Disease Control or township health centers, and subsequently analyzed at the Youanmen Center for Clinical Laboratory of Capital Medical University. Fasting blood glucose (FBG), triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C) were measured using enzymatic colorimetric methods. The coefficients of variation were 0.90%, 1.50%, and 1.00%, respectively, indicating good measurement reliability. In addition, sex was defined as a binary variable (female = 0, male = 1). All laboratory measurements were conducted following standardized protocols with internal quality control procedures. Definition of CKM syndrome and study outcomes CKM syndrome was classified according to the American Heart Association’s Presidential Advisory statement1, which delineates five stages of disease progression: Stage 0: absence of metabolic, renal, or cardiovascular risk factors; Stage 1: overweight, obesity, or abnormal fat distribution without other metabolic abnormalities or CKD; Stage 2: presence of metabolic derangements—such as hypertension, hyperglycemia, or hypertriglyceridemia—or moderate to high CKD risk; Stage 3: substantial CVD risk or very high-risk CKD; Stage 4: established CVD, including coronary artery disease, heart failure, or stroke. Given that our study population comprises Chinese adults, very high-risk CKD (G4–G5) and a high 10-year CVD risk estimated using the Framingham Risk Score were applied as risk equivalents for subclinical CVD21. Estimated GFR was derived using the China-modified MDRD (C-MDRD) equation22, and CKD stages were assigned according to KDIGO guidelines. The primary outcome was progression from baseline to advanced CKM syndrome, a stage characterized by substantially elevated risks of CKD and CVD, including clinically manifest cardiovascular disease5. The comprehensive staging criteria are detailed in the Supplementary Methods. Data collection At baseline, trained interviewers conducted standardized questionnaires to collect information on demographic characteristics (including age, sex, education level, geographic region, and marital status), health-related lifestyle behaviors (such as sleep duration, physical activity, smoking, and alcohol consumption), and self-reported medical history (including diabetes, hypertension, heart disease, stroke, and the use of glucose-lowering, lipid-lowering, and antihypertensive medications). Anthropometric measurements, including height, weight, and waist circumference, were obtained by trained healthcare personnel using standardized protocols. In addition, fasting venous blood samples were collected from all participants after at least 8 h of fasting for laboratory assessments. Biomarkers included fasting blood glucose, glycated hemoglobin (HbA1c), lipid profiles (HDL-C, LDL-C, and TC), creatinine (Cr), blood urea nitrogen (BUN), C-reactive protein (CRP), uric acid (UA), white blood cell count (WBC), and platelet count (PLT). Detailed definitions of covariates are provided in the Supplementary Methods. Statistical analysis All analyses were conducted using R version 4.4.3. Missing covariates were handled through multiple imputation with the “mice” package.Baseline characteristics were summarized across CKM syndrome stages, with continuous variables reported as medians (IQRs) and compared using the Kruskal–Wallis or Wilcoxon rank-sum tests, and categorical variables summarized as counts (percentages) and compared using chi-square tests. Baseline and cumulative METS-VF values were presented as means ± SDs and quartile distributions.Because CKM status could only be assessed at two discrete time points, the exact timing of disease progression could not be determined; therefore, logistic regression was used as the primary analytical approach.For longitudinal analyses, multivariable logistic regression models were constructed to examine the associations of baseline and cumulative METS-VF with the risk of developing advanced CKM syndrome, yielding ORs and 95% CIs. Exposure variables were evaluated both continuously (per IQR increment) and categorically (quartiles Q1–Q4), with linear trends formally tested.Three models were specified: Model 1: unadjusted; Model 2: adjusted for sex, age, marital status, educational attainment, and urban/rural residence; Model 3: additionally adjusted for smoking, alcohol consumption, length of sleep, physical activity, SUA, BUN, WBC count, CRP, and PLT. All covariates included in the multivariable models were obtained from baseline assessments (2011). Multicollinearity was examined using variance inflation factors (VIF; Tables S1–S4).To assess potential linear and nonlinear exposure–response patterns, restricted cubic spline–based multivariable models were fitted, with knots at the 5th, 35th, 65th, and 95th percentiles of the METS-VF distribution. Threshold effects were assessed using likelihood ratio tests. Inflection points were identified based on maximum likelihood estimation within fully adjusted models. Piecewise regression models were then applied to estimate the associations below and above the identified thresholds, with covariates adjusted as specified in Model 3. To evaluate the stability of the estimated thresholds, bootstrap resampling (200 iterations) was performed, and the distribution of threshold estimates was summarized. Subgroup analyses were performed to assess the consistency of associations across key demographic and lifestyle strata, including sex, age, educational attainment, marital status, urban–rural residence, smoking status, alcohol consumption, diabetes status and hypertension status. Interaction terms between METS-VF (or cumMETS-VF) and subgroup variables were included in fully adjusted models, and statistical significance of effect modification was evaluated using likelihood ratio tests. Sensitivity analyses were performed to assess the robustness of the findings. Sensitivity analyses were conducted to evaluate the robustness of the findings. First, participants receiving glucose-lowering, lipid-lowering, or antihypertensive medications at baseline were excluded. In addition, Cox proportional hazards models were applied as an alternative analytical approach. Given the absence of precise event timing, a uniform follow-up duration was assumed. The consistency of results across these analyses was assessed. Receiver operating characteristic (ROC) analyses were performed to compare the predictive performance of METS-VF and cumMETS-VF for advanced CKM syndrome, benchmarked against conventional adiposity metrics (WWI, VAI, ABSI, BRI, WC, and BMI). A two-sided p value < 0.05 was considered statistically significant. Participant characteristics Figure 1 depicts the baseline distribution of CKM syndrome stages and highlights marked sex-related differences. Overall, 10.1%, 8.0%, 43.8%, 24.0%, and 14.0% of participants were categorized into CKM syndrome Stage 0 through Stage 4, respectively. Men were disproportionately represented in Stage 3, a stage defined by substantially elevated cardiovascular risk (39.7%), while they were less frequently classified in the metabolically driven Stage 2 (28.7%). Conversely, women were more likely to fall into Stage 2 (57.0%), indicating distinct sex-specific patterns across the CKM syndrome risk continuum (Fig.2). Fig. 1Study flowchart. Study flowchart. Baseline characteristics for the 3,322 participants included in the longitudinal analysis are presented in Table 1. Over the 4-year follow-up period, 671 individuals (20.2%) progressed to advanced CKM syndrome (Stage 3 or Stage 4). Compared with participants who did not progress, progressors were older and exhibited markedly higher prevalence of hypertension, metabolic syndrome, and use of related medications (all P < 0.001). Although diabetes prevalence was comparable between groups, progressors had higher fasting glucose and HbA1c levels. They also showed elevated serum creatinine and uric acid concentrations, together with increased C-reactive protein levels, consistent with declining renal function and heightened systemic inflammation. Importantly, both baseline and cumulative levels of METS-VF were significantly higher among progressors (all P < 0.001), underscoring the potential role of visceral adiposity burden in driving CKM syndrome progression. A total of 2,412 participants (42%) were excluded due to loss to follow-up or missing outcome data. Baseline characteristics of participants lost to follow-up were compared with those included in the final analysis (Table S5). No substantial differences were observed in most baseline variables, except for hypertension status.Table 1Baseline characteristics of participants in the longitudinal analysis.CharacterOverall(n = 3322)Advanced CKM syndrome (Stages 3 or 4)P-valueNo(n = 2651)Yes(n = 671)Age(years), mean (SD)56.25±(7.58)55.36±(7.29)59.73±(7.67)< 0.001Sex, n (%)< 0.001 Female2114(64%)1808(68%)306(46%) Male1208(36%)814(32%)365(54%)Residence, n (%)0.575 Rural2248(68%)1800(68%)448(67%) Urban1074(32%)851(32%)223(33%)Education_status, n (%)0.209 No formal education1665(50%)1345(51%)320(48%) Primary school678(20%)525(20%)153(23%) Middle school686(21%)541(20%)145(22%) High school or above293(8.8%)240(9.1%)53(7.9%)Marital_status, n (%)0.075 Married2909(88%)2335(88%)574(86%) Non-Married413(12%)316(12%)97(14%)Alcohol, n (%)< 0.001 Current976(29%)723(27%)253(38%) Former206(6.2%)143(5.4%)63(9.4%) Never2140(64%)1785(67%)355(53%)Smoking, n (%)< 0.001 Current675(20%)458(17%)217(32%) Former244(7.3%)158(6.0%)86(13%) Never2403(72%)2035(77%)368(55%)Length of sleep, n (%)0.023 Short sleep501(15%)383(14%)118(18%) Suitable sleep2273(68%)1843(70%)430(64%) Long sleep548(16%)425(16%)123(18%)Physical activity, n (%)0.564 High-intensity378(11%)306(12%)72(11%) Moderate-intensity324(9.8%)252(9.5%)72(11%) Low-intensity2620(79%)2093(79%)527(79%)Hypertension, n (%)< 0.001 Yes1418(57%)1040(39%)378(56%) No1904(43%)1611(61%)293(44%)Diabetes, n (%)0.583 Yes198(6.0%)155(5.8%)43(6.4%) No3124(94%)2496(94%)455(94%)MetS, n (%)< 0.001 Yes1746(53%)1332(50%)414(62%) No1576(47%)1319(50%)257(38%)Antidiabetic agents, n (%)0.120 Yes40(1.2%)28(1.1%)12(1.8%) No3282(99%)2623(99%)659(98%)Antihypertension agents, n (%)< 0.001 Yes394(12%)242(9.1%)152(23%) No2928(88%)2409(91%)519(77%)Antihyperlipidemic agents, n (%)< 0.001 Yes108(3.3%)70(2.6%)38(5.7%) No3214(97%)2581(97%)633(94%)BMI(kg/m2), mean (SD)24.50±(14.43)23.79±(15.98)27.33±(14.74)0.138TC(mg/dL), mean (SD)189.04±(34.81)188.71±(34.46)190.36±(36.17)0.185HDL-C(mg/dL), mean (SD)52.84±(14.69)52.98±(14.48)52.28±(15.47)0.160LDL-C(mg/dL), mean (SD)113.67±(31.73)113.35±(31.18)114.92±(33.80)0.186TG (mg/dL), mean (SD)136.24±(84.45)132.63±(81.60)150.50±(93.59)0.173CRP(mg/dL), mean (SD)2.30±(7.28)2.18±(7.00)2.77±(8.26)< 0.001SUA(mg/dL), mean (SD)4.20±(1.12)4.13±(1.09)4.47±(1.19)< 0.001SCr(mg/dL), mean (SD)0.74±(0.16)0.73±(0.15)0.77±(0.16)< 0.001BUN(mg/dL), mean (SD)15.36±(4.22)15.35±(4.21)15.41±(4.26)0.875FBG(mg/dL), mean (SD)103.07±(23.66)102.32±(20.09)106.01±(34.19)< 0.001HbA1c(%), mean (SD)5.17±(0.61)5.15±(0.58)5.24±(0.74)< 0.001WBC(×109),mean (SD)6.10±(1.82)6.07±(1.84)6.22±(1.74)0.007PLT(×109),mean (SD)212.06±(69.83)211.57±(70.31)214.00±(67.95)0.362METS-VF2012, mean (SD)6.28±(2.44)6.20±(2.56)6.59±(1.89)< 0.001METS-VF2015, mean (SD)6.30±(2.68)6.23±(2.82)6.59±(2.01)< 0.001Cumulative METS-VF, mean (SD)18.87±(5.50)18.65±(5.75)19.76±(4.25)< 0.001MetS metabolic syndrome, BMI body mass index, TC total cholesterol, HDL-C high-density lipoprotein cholestero, LDL-C low-density lipoprotein cholestero, TG triglyceride, CRP C-reactive protein, SCr serum creatinine, SUA serum uric acid, BUN blood urea nitrogen, FBG fasting blood glucose, HbA1c glycated Hemoglobin A1c, WBC white blood cell, PLT platelet, METS-VF metabolic score for visceral fat. Table 1Baseline characteristics of participants in the longitudinal analysis.CharacterOverall(n = 3322)Advanced CKM syndrome (Stages 3 or 4)P-valueNo(n = 2651)Yes(n = 671)Age(years), mean (SD)56.25±(7.58)55.36±(7.29)59.73±(7.67)< 0.001Sex, n (%)< 0.001 Female2114(64%)1808(68%)306(46%) Male1208(36%)814(32%)365(54%)Residence, n (%)0.575 Rural2248(68%)1800(68%)448(67%) Urban1074(32%)851(32%)223(33%)Education_status, n (%)0.209 No formal education1665(50%)1345(51%)320(48%) Primary school678(20%)525(20%)153(23%) Middle school686(21%)541(20%)145(22%) High school or above293(8.8%)240(9.1%)53(7.9%)Marital_status, n (%)0.075 Married2909(88%)2335(88%)574(86%) Non-Married413(12%)316(12%)97(14%)Alcohol, n (%)< 0.001 Current976(29%)723(27%)253(38%) Former206(6.2%)143(5.4%)63(9.4%) Never2140(64%)1785(67%)355(53%)Smoking, n (%)< 0.001 Current675(20%)458(17%)217(32%) Former244(7.3%)158(6.0%)86(13%) Never2403(72%)2035(77%)368(55%)Length of sleep, n (%)0.023 Short sleep501(15%)383(14%)118(18%) Suitable sleep2273(68%)1843(70%)430(64%) Long sleep548(16%)425(16%)123(18%)Physical activity, n (%)0.564 High-intensity378(11%)306(12%)72(11%) Moderate-intensity324(9.8%)252(9.5%)72(11%) Low-intensity2620(79%)2093(79%)527(79%)Hypertension, n (%)< 0.001 Yes1418(57%)1040(39%)378(56%) No1904(43%)1611(61%)293(44%)Diabetes, n (%)0.583 Yes198(6.0%)155(5.8%)43(6.4%) No3124(94%)2496(94%)455(94%)MetS, n (%)< 0.001 Yes1746(53%)1332(50%)414(62%) No1576(47%)1319(50%)257(38%)Antidiabetic agents, n (%)0.120 Yes40(1.2%)28(1.1%)12(1.8%) No3282(99%)2623(99%)659(98%)Antihypertension agents, n (%)< 0.001 Yes394(12%)242(9.1%)152(23%) No2928(88%)2409(91%)519(77%)Antihyperlipidemic agents, n (%)< 0.001 Yes108(3.3%)70(2.6%)38(5.7%) No3214(97%)2581(97%)633(94%)BMI(kg/m2), mean (SD)24.50±(14.43)23.79±(15.98)27.33±(14.74)0.138TC(mg/dL), mean (SD)189.04±(34.81)188.71±(34.46)190.36±(36.17)0.185HDL-C(mg/dL), mean (SD)52.84±(14.69)52.98±(14.48)52.28±(15.47)0.160LDL-C(mg/dL), mean (SD)113.67±(31.73)113.35±(31.18)114.92±(33.80)0.186TG (mg/dL), mean (SD)136.24±(84.45)132.63±(81.60)150.50±(93.59)0.173CRP(mg/dL), mean (SD)2.30±(7.28)2.18±(7.00)2.77±(8.26)< 0.001SUA(mg/dL), mean (SD)4.20±(1.12)4.13±(1.09)4.47±(1.19)< 0.001SCr(mg/dL), mean (SD)0.74±(0.16)0.73±(0.15)0.77±(0.16)< 0.001BUN(mg/dL), mean (SD)15.36±(4.22)15.35±(4.21)15.41±(4.26)0.875FBG(mg/dL), mean (SD)103.07±(23.66)102.32±(20.09)106.01±(34.19)< 0.001HbA1c(%), mean (SD)5.17±(0.61)5.15±(0.58)5.24±(0.74)< 0.001WBC(×109),mean (SD)6.10±(1.82)6.07±(1.84)6.22±(1.74)0.007PLT(×109),mean (SD)212.06±(69.83)211.57±(70.31)214.00±(67.95)0.362METS-VF2012, mean (SD)6.28±(2.44)6.20±(2.56)6.59±(1.89)< 0.001METS-VF2015, mean (SD)6.30±(2.68)6.23±(2.82)6.59±(2.01)< 0.001Cumulative METS-VF, mean (SD)18.87±(5.50)18.65±(5.75)19.76±(4.25)< 0.001MetS metabolic syndrome, BMI body mass index, TC total cholesterol, HDL-C high-density lipoprotein cholestero, LDL-C low-density lipoprotein cholestero, TG triglyceride, CRP C-reactive protein, SCr serum creatinine, SUA serum uric acid, BUN blood urea nitrogen, FBG fasting blood glucose, HbA1c glycated Hemoglobin A1c, WBC white blood cell, PLT platelet, METS-VF metabolic score for visceral fat. Baseline characteristics of participants in the longitudinal analysis. MetS metabolic syndrome, BMI body mass index, TC total cholesterol, HDL-C high-density lipoprotein cholestero, LDL-C low-density lipoprotein cholestero, TG triglyceride, CRP C-reactive protein, SCr serum creatinine, SUA serum uric acid, BUN blood urea nitrogen, FBG fasting blood glucose, HbA1c glycated Hemoglobin A1c, WBC white blood cell, PLT platelet, METS-VF metabolic score for visceral fat. Association of METS-VF and cumMETS-VF with the risk of progression to advanced CKM syndrome The associations between visceral adiposity burden and progression to advanced CKM syndrome were generally consistent across models. As shown in Table 2, both baseline METS-VF and cumulative METS-VF were independently associated with the risk of advanced CKM syndrome. In the fully adjusted model (Model 3), each unit increase in baseline METS-VF was associated with a 13% higher risk (OR 1.13, 95% CI 1.06–1.22; p = 0.001), while each unit increase in cumulative METS-VF was associated with a 6% higher risk (OR 1.06, 95% CI 1.04–1.09; p < 0.001).When analyzed by quartiles, a graded association was observed. Compared with the lowest quartile, participants in the highest quartile had a 2.41-fold higher risk for baseline METS-VF and a 2.66-fold higher risk for cumulative METS-VF (both p for trend < 0.001). Intermediate quartiles showed moderate increases in risk, particularly in the third quartile.Table 2Association of baseline METS-VF, cumMETS-VF and the risk of advanced CKM syndrome.Model 1Model 2Model 3OR (95%CI)P valueOR (95%CI)P valueOR (95%CI)P valuecumMETS-VF1.06(1.03,1.09)< 0.0011.06(1.04,1.09)< 0.0011.06(1.04,1.09)< 0.001cumMETS-VF (Quartile) Quartile 1ReferenceReferenceReference Quartile 21.06(0.82,1.39)0.6271.13(0.85,1.49)0.4071.14(0.85,1.51)0.381 Quartile 31.42(1.10,1.84)0.0071.57(1.20,2.06)0.0011.65(1.25,2.17)< 0.001 Quartile 42.39(1.88,3.08)< 0.0012.44(1.89,3.16)< 0.0012.66(2.03,3.48)< 0.001 P for trend< 0.001< 0.001< 0.001 METS-VF1.13(1.06,1.24)0.0011.12(1.05,1.21)0.0011.13(1.06,1.22)0.001METS-VF (Quartile) Quartile 1ReferenceReferenceReference Quartile 21.17(0.90,1.53)0.2271.13(0.86,1.49)0.3741.14(0.86,1.51)0.359 Quartile 31.37(1.06,1.77)0.0161.38(1.05,1.81)0.0201.49(1.13,1.97)0.005 Quartile 42.25(1.77,2.87)< 0.0012.21(1.71,2.86)< 0.0012.41(1.85,3.17)< 0.001 P for trend< 0.001< 0.001< 0.001Model 1: unadjusted. Model 2: adjusted for sex, age, marital status, education status and residence. Model 3: adjusted for sex, age, marital status, education status, residence, smoking, alcohol, length of sleep, physical_activity, SUA, BUN, WBC, CRP and PLT. Significant values are in bold. Table 2Association of baseline METS-VF, cumMETS-VF and the risk of advanced CKM syndrome.Model 1Model 2Model 3OR (95%CI)P valueOR (95%CI)P valueOR (95%CI)P valuecumMETS-VF1.06(1.03,1.09)< 0.0011.06(1.04,1.09)< 0.0011.06(1.04,1.09)< 0.001cumMETS-VF (Quartile) Quartile 1ReferenceReferenceReference Quartile 21.06(0.82,1.39)0.6271.13(0.85,1.49)0.4071.14(0.85,1.51)0.381 Quartile 31.42(1.10,1.84)0.0071.57(1.20,2.06)0.0011.65(1.25,2.17)< 0.001 Quartile 42.39(1.88,3.08)< 0.0012.44(1.89,3.16)< 0.0012.66(2.03,3.48)< 0.001 P for trend< 0.001< 0.001< 0.001 METS-VF1.13(1.06,1.24)0.0011.12(1.05,1.21)0.0011.13(1.06,1.22)0.001METS-VF (Quartile) Quartile 1ReferenceReferenceReference Quartile 21.17(0.90,1.53)0.2271.13(0.86,1.49)0.3741.14(0.86,1.51)0.359 Quartile 31.37(1.06,1.77)0.0161.38(1.05,1.81)0.0201.49(1.13,1.97)0.005 Quartile 42.25(1.77,2.87)< 0.0012.21(1.71,2.86)< 0.0012.41(1.85,3.17)< 0.001 P for trend< 0.001< 0.001< 0.001Model 1: unadjusted. Model 2: adjusted for sex, age, marital status, education status and residence. Model 3: adjusted for sex, age, marital status, education status, residence, smoking, alcohol, length of sleep, physical_activity, SUA, BUN, WBC, CRP and PLT. Significant values are in bold. Association of baseline METS-VF, cumMETS-VF and the risk of advanced CKM syndrome. Model 1: unadjusted. Model 2: adjusted for sex, age, marital status, education status and residence. Model 3: adjusted for sex, age, marital status, education status, residence, smoking, alcohol, length of sleep, physical_activity, SUA, BUN, WBC, CRP and PLT. Significant values are in bold. Fig. 2Prevalence of CKM syndrome stages by gender, CHARLS 2011–2012. Prevalence of CKM syndrome stages by gender, CHARLS 2011–2012. Restricted cubic spline analyses (Fig. 3) indicated significant non-linear associations for both baseline METS-VF and cumulative METS-VF (all P_non-linear < 0.05). For baseline METS-VF, the exposure–response curve showed a relatively gradual increase across lower levels, followed by a more pronounced upward trend beyond approximately 6.7. In contrast, cumulative METS-VF exhibited a relatively stable association across lower ranges, with risk increasing more steeply after approximately 20. Notably, the confidence intervals were wider at the extreme ends of the exposure distribution, indicating greater uncertainty in these regions, whereas the estimates were more stable within the central range. In threshold-effect analyses (Table 3), a cutoff value of 20.11 was identified for cumulative METS-VF. Below this threshold, a modest but statistically significant association was observed (OR 1.03, 95% CI 1.01–1.06; p = 0.012), while above the threshold, the association became more pronounced (OR 1.34, 95% CI 1.18–1.53; p < 0.001). Similarly, for baseline METS-VF, a threshold of 6.71 was identified. Below this level, a modest association was present (OR 1.06, 95% CI 1.01–1.14; p = 0.040), whereas above the threshold, the risk increased substantially (OR 2.63, 95% CI 1.74–4.00; p < 0.001). Likelihood ratio tests supported the presence of threshold effects for both exposures (both p < 0.001). Bootstrap resampling further demonstrated that the estimated thresholds were stable, with mean cutoff values of 20.35 (95% CI 19.22–21.32) for cumulative METS-VF and 6.78 (95% CI 6.37–7.13) for baseline METS-VF (Table S6).Fig. 3The RCS analysis between the baseline METS-VF control levels, cumMETS-VF and the risk of advance CKM syndrome. (A) unadjusted; (B) adjusted for sex, age, marital status, education status, residence, smoking, alcohol, SUA, BUN, WBC, CRP and PLT; (C) unadjusted; (D) adjusted for sex, age, marital status, education status, residence, smoking, alcohol, SUA, BUN, WBC, CRP and PLT.Table 3Segmented association of METS-VF and cumMETS-VF with advanced CKM syndrome.CharacterOR (95%CI)P ValuecumMETS-VF Threshold20.11 Below threshold (< 20.11)1.03 (1.01,1.06)0.012 Above threshold (≧ 20.11)1.34 (1.18,1.53)< 0.001 Likelihood ratio test< 0.001METS-VF Threshold6.71 Below threshold (< 6.71)1.06 (1.01,1.14)0.040 Above threshold (≧ 6.71)2.63 (1.74,4.00)< 0.001 Likelihood ratio test< 0.001 The RCS analysis between the baseline METS-VF control levels, cumMETS-VF and the risk of advance CKM syndrome. (A) unadjusted; (B) adjusted for sex, age, marital status, education status, residence, smoking, alcohol, SUA, BUN, WBC, CRP and PLT; (C) unadjusted; (D) adjusted for sex, age, marital status, education status, residence, smoking, alcohol, SUA, BUN, WBC, CRP and PLT. Table 3Segmented association of METS-VF and cumMETS-VF with advanced CKM syndrome.CharacterOR (95%CI)P ValuecumMETS-VF Threshold20.11 Below threshold (< 20.11)1.03 (1.01,1.06)0.012 Above threshold (≧ 20.11)1.34 (1.18,1.53)< 0.001 Likelihood ratio test< 0.001METS-VF Threshold6.71 Below threshold (< 6.71)1.06 (1.01,1.14)0.040 Above threshold (≧ 6.71)2.63 (1.74,4.00)< 0.001 Likelihood ratio test< 0.001 Segmented association of METS-VF and cumMETS-VF with advanced CKM syndrome. Comparative predictive performance of METS-VF and conventional adiposity indices ROC analyses (Fig. 4) indicated modest variation in discrimination across adiposity indices. Among all measures, cumulative METS-VF showed the highest AUC (0.60, 95% CI 0.58–0.63), followed by baseline METS-VF (AUC 0.59, 95% CI 0.57–0.62). Other cumulative indices, including cumWC (AUC 0.57) and cumABSI (AUC 0.56), demonstrated comparable but slightly lower discrimination, whereas conventional indices such as BMI, BRI, and VAI showed relatively lower AUC values (approximately 0.51–0.53) (Table 4). At their respective optimal cutoff points, cumulative METS-VF yielded a sensitivity of 0.52 and specificity of 0.65, reflecting a relatively balanced performance compared with other indices. In contrast, indices such as ABSI and BRI were characterized by higher specificity but lower sensitivity, whereas others (e.g., cumWC and cumWWI) exhibited higher sensitivity at the expense of lower specificity.Overall, METS-VF, particularly in its cumulative form, demonstrated comparatively better discrimination than traditional adiposity measures, although the overall predictive performance remained moderate. Fig. 4ROC curves of baseline ABSI control levels, baseline BMI control levels, baseline BRI control levels, baseline METS-VF control levels, baseline VAI control levels, baseline WC control levels, baseline WWI control levels, cumABSI, cumBMI, cumBRI, cumMETS-VF, cumVAI, cumWC, cumWWI in predicting the advance CKM syndrome. ROC curves of baseline ABSI control levels, baseline BMI control levels, baseline BRI control levels, baseline METS-VF control levels, baseline VAI control levels, baseline WC control levels, baseline WWI control levels, cumABSI, cumBMI, cumBRI, cumMETS-VF, cumVAI, cumWC, cumWWI in predicting the advance CKM syndrome. Table 4ROC analysis results.IndexAUC95%CICut-off pointSensitivitySpecificitycumABSI0.560.54,0.582.530.410.69cumBMI0.520.50,0.5571.340.480.58cumBRI0.530.51,0.5613.330.440.63cumMETS-VF0.600.58,0.6320.390.520.65cumVAI0.510.49,0.545.420.450.59cumWC0.570.54,0.59249.780.630.48cumWWI0.530.50,0.5533.200.560.49ABSI0.550.52,0.570.860.250.83BMI0.520.49,0.5423.760.460.59BRI0.520.50,0.555.400.210.84METS-VF0.590.57,0.626.720.530.61VAI0.510.49,0.531.130.410.61WC0.560.54,0.5984.150.530.55WWI0.520.49,0.5411.880.210.84 Table 4ROC analysis results.IndexAUC95%CICut-off pointSensitivitySpecificitycumABSI0.560.54,0.582.530.410.69cumBMI0.520.50,0.5571.340.480.58cumBRI0.530.51,0.5613.330.440.63cumMETS-VF0.600.58,0.6320.390.520.65cumVAI0.510.49,0.545.420.450.59cumWC0.570.54,0.59249.780.630.48cumWWI0.530.50,0.5533.200.560.49ABSI0.550.52,0.570.860.250.83BMI0.520.49,0.5423.760.460.59BRI0.520.50,0.555.400.210.84METS-VF0.590.57,0.626.720.530.61VAI0.510.49,0.531.130.410.61WC0.560.54,0.5984.150.530.55WWI0.520.49,0.5411.880.210.84 ROC analysis results. Subgroup and sensitivity analyses As shown in Fig. 5, the positive associations of both baseline METS-VF and cumulative METS-VF with advanced CKM syndrome were generally consistent across most demographic, lifestyle, and clinical subgroups. No significant effect modification was observed by sex, age, educational attainment, marital status, or urban–rural residence (all P_interaction > 0.05). However, significant interactions were identified for baseline METS-VF across smoking status, alcohol consumption, and residential status. In addition, diabetes status significantly modified the associations of both baseline METS-VF and cumulative METS-VF with advanced CKM sydrome. No significant interaction was observed for hypertension status. Fig. 5Subgroup analysis of the association between the baseline METS-VF control levels, cumMETS-VF and the risk of advance CKM syndrome. (A) baseline METS-VF control levels; (B) cumMETS-VF. Subgroup analysis of the association between the baseline METS-VF control levels, cumMETS-VF and the risk of advance CKM syndrome. (A) baseline METS-VF control levels; (B) cumMETS-VF. Sensitivity analyses further supported the robustness of our findings. Excluding individuals receiving glucose-lowering, lipid-lowering, or antihypertensive medications at baseline did not materially alter the results, with both METS-VF and cumulative METS-VF remaining significantly associated with the risk of progression to advanced CKM syndrome (Tables S7–S10). In addition, Cox proportional hazards models were applied as an alternative analytical approach. Despite the assumption of a uniform follow-up duration due to the lack of precise event timing, the results were consistent with those from the primary logistic regression analyses (Table S11). Discussion Using data from a nationally representative prospective cohort, this study comprehensively investigated the associations of visceral adiposity burden, assessed by METS-VF and cumulative cumMETS-VF, with progression to advanced CKM syndrome. Both baseline and cumulative measures demonstrated independent and consistent positive associations with CKM syndrome progression across a range of statistical analyses, underscoring the robustness of the findings and reinforcing the central role of visceral adiposity in multisystem impairment along the CKM syndrome continuum. Notably, the convergence of findings from RCS analyses, maximum likelihood–based threshold-effect modeling, and ROC-derived optimal Youden cutoffs revealed highly concordant risk inflection points. Collectively, these results suggest that METS-VF and cumMETS-VF may represent practical, inexpensive, and scalable metabolic markers for early risk stratification of CKM syndrome progression. Accumulating evidence has established a close link between visceral adiposity and a broad spectrum of cardiovascular and metabolic disorders. Recently, METS-VF, a composite index integrating metabolic parameters with adiposity distribution characteristics23, has emerged as a promising marker for cardiometabolic risk assessment. Several large-scale cohort studies have demonstrated its superior predictive performance, particularly among individuals with impaired glucose metabolism, in stratifying the risks of cardiovascular events and all-cause mortality24–26. In the Kailuan prospective cohort, cumulative METS-VF showed a clear dose–response association with cardiovascular disease and all-cause mortality, outperforming conventional obesity measures such as BMI and waist circumference20. Likewise, a multistate analysis involving more than 420,000 participants from the UK Biobank revealed that elevated METS-VF levels substantially increased the likelihood of progression from a healthy state to cardiometabolic disease and subsequent multimorbidity27. Together, these findings suggest that markers capturing the metabolic burden of visceral adiposity may provide a more accurate reflection of long-term cardiovascular–renal–metabolic risk than traditional anthropometric measures alone. Importantly, many previous investigations have relied on single time-point measurements, potentially overlooking the cumulative impact of prolonged metabolic burden. As a key approach for analyzing repeated-measurement data, cumulative exposure captures both the intensity and duration of exposure over time28. Compared with cross-sectional assessments, cumulative metrics are more capable of reflecting the sustained organ damage and biological consequences associated with chronic metabolic stress29,30. Accordingly, by jointly evaluating baseline METS-VF and cumulative exposure (cumMETS-VF), the present study provides a more comprehensive characterization of long-term metabolic burden and its contribution to CKM progression from the complementary perspectives of exposure magnitude and exposure duration. Our findings are broadly consistent with previous prospective evidence. In a cohort study involving 4,069 participants, cumulative METS-VF was significantly associated with incident stroke, with individuals in the highest quartile exhibiting more than a twofold increased risk compared with those in the lowest quartile (HR = 2.12). Notably, stroke risk increased sharply once METS-VF exceeded a specific threshold value31. Similarly, another study focusing on individuals with stages 0–3 CKM syndrome demonstrated that baseline METS-VF, cumulative METS-VF, and long-term trajectory patterns were all independently associated with cardiovascular disease risk, with cumulative exposure showing the strongest association (highest quartile HR = 1.82)32. Building upon these observations, our study further expands the current evidence base by suggesting that visceral adiposity metabolic burden may influence not only isolated cardiovascular outcomes, but also the broader multisystem progression of CKM syndrome. Nevertheless, most previous studies have primarily focused on isolated clinical outcomes, such as cardiovascular disease, stroke, or chronic kidney disease, with relatively limited attention to the integrated multisystem progression characteristic of the CKM continuum. A key strength and innovation of the present study is that, rather than examining a single endpoint in isolation, we investigated the transition from early-stage to advanced CKM syndrome as a dynamic and systemic process. This approach enables a more comprehensive assessment of the complex interactions among metabolic, renal, and cardiovascular dysfunction. Compared with conventional single-disease frameworks, our study design more closely reflects the underlying pathophysiological nature of CKM syndrome as a progressive multi-organ continuum. From a mechanistic perspective, visceral adipose tissue functions as a highly metabolically active endocrine organ and may drive CKM progression through several interconnected biological pathways. Excess visceral fat releases substantial amounts of free fatty acids into the portal circulation, contributing to hepatic lipotoxicity, insulin resistance, and dysregulated lipid metabolism10,33,34. At the same time, visceral adiposity promotes the secretion of pro-inflammatory cytokines and adverse adipokines—including TNF-α, IL-6, MCP-1, leptin, and resistin—thereby sustaining chronic low-grade inflammation13,14,35,36, impairing endothelial function, and accelerating atherosclerotic processes37–39. Moreover, visceral fat accumulation may activate both the sympathetic nervous system and the renin–angiotensin–aldosterone system (RAAS), increasing blood pressure and renal hemodynamic stress40–42, which in turn contributes to glomerular hyperfiltration, proteinuria, and renal interstitial fibrosis43,44. Prolonged exposure to excessive visceral adiposity may therefore create a self-perpetuating cycle of metabolic, cardiovascular, and renal injury, ultimately accelerating multisystem deterioration along the CKM syndrome continuum. Importantly, we observed pronounced nonlinear associations between both METS-VF and cumMETS-VF and the progression of CKM syndrome, with risk increasing sharply beyond specific threshold levels. In the lower exposure range, risk accumulation appeared relatively modest; however, once these thresholds were surpassed, the risk escalated substantially. These findings suggest that the detrimental effects of visceral adiposity may not follow a purely linear pattern of accumulation. Instead, once a critical metabolic burden is exceeded, compensatory physiological mechanisms may progressively fail, leading to an accelerated transition toward multisystem dysfunction and advanced CKM progression. Importantly, the risk inflection points identified across restricted cubic spline analyses, maximum likelihood–based threshold-effect models, and ROC-derived optimal Youden indices were highly concordant, collectively reinforcing the robustness and reliability of the observed threshold effects. We further assessed threshold stability using Bootstrap resampling, which demonstrated only minimal variability in the estimated cut-off values, indicating favorable statistical stability. Nevertheless, these thresholds were derived internally within the present cohort. Although they may carry potential clinical relevance, their generalizability across different populations, ethnic backgrounds, and age groups remains uncertain and warrants further validation in independent external cohorts. Notably, the threshold values identified in our study closely align with findings from previous investigations. A CT-based study using visceral fat area (VFA) as the reference standard reported that METS-VF outperformed most conventional adiposity indices in detecting visceral fat accumulation, with optimal ROC cut-off values of approximately 6.4 for men and 6.5 for women23, remarkably consistent with the thresholds observed in our cohort. Similarly, a cross-sectional analysis from NHANES demonstrated a significant positive association between METS-VF and CKD, identifying risk thresholds around 6.10 and 7.5545. Prior studies investigating coronary artery calcification (CAC) also reported an optimal METS-VF cut-off value close to 6.446. Collectively, these findings across diverse populations and cardiometabolic outcomes provide external support for the robustness and potential clinical relevance of the threshold effects observed in the present study. From a clinical and public health perspective, although the discriminatory performance of METS-VF (AUC approximately 0.60) remains modest and may not be sufficient for use as a standalone predictive marker, the consistently observed associations indicate that it could serve as a valuable adjunct to existing risk stratification frameworks rather than a replacement for established clinical models. Importantly, in primary care settings and resource-constrained environments where imaging-based assessments such as CT or MRI are not routinely feasible, METS-VF can be readily derived from standard anthropometric measurements and commonly available laboratory parameters. This practical accessibility, combined with its low cost and scalability, highlights its potential utility for large-scale population screening and early identification of individuals at elevated CKM syndrome risk. Furthermore, the risk thresholds identified in the present study may offer clinically relevant guidance for early preventive strategies. Individuals approaching or exceeding these cut-off values may benefit from prioritized interventions targeting weight control, regular physical activity, dietary modification, and sleep optimization, together with intensified cardiovascular and renal surveillance, with the aim of slowing or preventing further CKM syndrome progression. Importantly, METS-VF may also facilitate the identification of individuals with “hidden” or metabolically unhealthy obesity—those with a normal BMI but elevated visceral fat–related metabolic burden—who are frequently overlooked by conventional BMI-based screening approaches. Early recognition of this high-risk subgroup may help avoid missed opportunities for timely intervention and risk reduction. Several notable strengths of this study should also be acknowledged. First, the analysis was conducted within the nationally representative CHARLS prospective cohort, enabling a more comprehensive characterization of the CKM syndrome risk profile among middle-aged and older Chinese adults and enhancing the generalizability of the findings. Second, by jointly examining both baseline and cumulative exposure, our study extends beyond the limitations of conventional single time-point assessments and provides a more realistic representation of the long-term metabolic burden associated with visceral adiposity. Third, we employed a comprehensive analytical framework—including multivariable-adjusted models, restricted cubic spline analyses, threshold-effect modeling, Bootstrap-based internal validation, subgroup analyses, and multiple sensitivity analyses—to rigorously evaluate the consistency and robustness of the observed associations across different methodological approaches. Finally, by systematically comparing METS-VF with several conventional and emerging adiposity indices, this study further clarifies its potential utility in CKM syndrome risk stratification and population-level risk assessment. Several limitations should be acknowledged. First, a proportion of participants were lost to follow-up or lacked outcome data, which may introduce potential selection bias. However, baseline comparisons showed largely similar characteristics between included and excluded participants, suggesting that the impact on the overall findings is likely limited. Second, cumulative exposure was defined using two time points, which may not fully capture long-term exposure trajectories. This was constrained by the availability of fasting blood measurements in the CHARLS cohort, which were conducted only in 2011 and 2015. Nevertheless, prior epidemiological studies have demonstrated that repeated measurements at two time points can still provide meaningful insights into cumulative exposure, and consistent associations observed for both baseline and cumulative METS-VF in our analyses further support the robustness of the findings. Third, METS-VF serves as a metabolic surrogate for visceral adiposity rather than a direct imaging-based measure, and CKM syndrome staging was based on proxy indicators derived from available data rather than clinical adjudication. These factors may introduce some degree of misclassification. However, the definitions used were based on established clinical criteria and are commonly applied in large-scale epidemiological studies, suggesting that any misclassification is likely to be non-differential.Finally, although a wide range of covariates, including lifestyle factors, were adjusted for, residual confounding cannot be entirely excluded. In addition, as an observational study, causal relationships cannot be established. Future research priorities include validating thresholds and cumulative exposure in independent, multi-ethnic cohorts; conducting calibration studies combining imaging modalities (CT/MRI) with metabolic scores; long-term follow-up for causal inference; and designing randomized controlled trials targeting visceral fat reduction (via lifestyle or pharmacological interventions) to evaluate causal effects on CKM syndrome outcomes. Conclusion In this nationally representative cohort, both baseline and cumulative visceral adiposity burden, assessed by METS-VF, were independently associated with progression to advanced CKM syndrome. Compared with traditional adiposity indices, METS-VF showed comparable or slightly better discrimination. These findings underscore the role of visceral adiposity in multisystem involvement along the CKM continuum and suggest that METS-VF and cumulative METS-VF may provide incremental value for risk stratification when considered alongside established clinical factors. Further studies are warranted to validate these observations and clarify their applicability in broader populations. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1