Association of metal and metalloid exposure with cardiovascular-kidney-metabolic syndrome: mediation by inflammation, oxidative stress, and aging Authors: Shuai Zhang, Hanhan Tang, Linqing Pan, Minglian Zhou Published: 2025 ABSTRACT BackgroundMetal and metalloid exposure has been linked to various health impairments; however, its association with Cardiovascular-Kidney‐Metabolic (CKM) syndrome has not been systematically investigated. This study aims to evaluate this relationship.MethodsThis cross-sectional study analyzed data from the National Health and Nutrition Examination Survey (NHANES) spanning 1999–2018. Urinary concentrations of nine metal and metalloid species (barium, cadmium, cobalt, cesium, molybdenum, lead, antimony, thallium, and tungsten) were examined. Weighted multinomial logistic regression was used to assess associations between individual metals and CKM syndrome stages. Additionally, mixture models including quantile g-computation (Qgcomp), weighted quantile sum (WQS), and Bayesian kernel machine regression (BKMR) were applied to evaluate combined effects of metals and metalloids. Mediation analyses explored whether inflammation, oxidative stress, and biological aging may partially mediate these associations. Subgroup analyses were conducted by sex, age, and race. Sensitivity analyses including urinary arsenic were also conducted (2003–2018 NHANES cycles).ResultsAfter applying exclusion criteria, a total of 6,650 participants were included (3,358 men and 3,292 women; mean age 47.3 years, SD = 16.5). CKM syndrome exhibited a high prevalence in the U.S. population. Except for lead, higher CKM stages were generally associated with higher urinary metal levels. Multiple statistical models consistently indicated significant positive associations between exposure to several metals and CKM syndrome. Barium, thallium, and antimony were significantly associated with the overall odds of CKM syndrome, as well as with the risks of Stage 1 and Stage 2 CKM, with barium showing the highest weight. For advanced CKM stages, antimony had the greatest contribution, followed by tungsten and cobalt. Significant variations were observed by sex, age, and race. Mediation analyses suggested that inflammation, oxidative stress, and biological aging may partially explain the observed associations between metal co-exposure and CKM syndrome. In the sensitivity analysis, inorganic arsenic showed the strongest positive association with CKM syndrome and contributed most prominently in the mixture models.ConclusionsExposure to metal and metalloid species is closely associated with the prevalence of CKM syndrome at different stages. Inflammation, oxidative stress, and biological aging may play partial mediating roles in these associations. As NHANES reflects exposure patterns in the U.S. general population, the findings are most applicable to community-level environmental exposures and should not be directly extrapolated to high-exposure occupational groups. Further validation in prospective cohort studies and mechanistic research is warranted.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12889-025-25894-0. Introduction Currently, cardiovascular disease (CVD), chronic kidney disease (CKD), and diabetes have emerged as the major epidemics of the 21 st century. It is estimated that by 2021, there were over 500 million individuals with diabetes worldwide, the vast majority of whom had type 2 diabetes (T2D). By 2045, the number of cases is projected to nearly double [1]. In the United States, approximately 2.5% of the population is affected by heart failure [2]. Globally, CKD caused 1.2 million deaths in 2017 alone, with nearly 700 million individuals diagnosed with the disease [3]. Growing evidence suggests that T2D, CVD, and CKD often act as significant risk factors for one another due to shared pathophysiological mechanisms. These conditions are closely interconnected and mutually influential. Given the close interconnection among the three, it is essential to adopt a unified term—cardio-metabolic-renal (CMR) diseases—for systematically describing their interdependence [4]. In 2023, the American Heart Association (AHA) introduced a new concept: cardiovascular-renal-metabolic (CKM) syndrome as a health condition emerging from the complex interplay between obesity, diabetes, CKD, and CVD [5]. This concept aims to elucidate the close relationships and interactions among metabolic diseases, CVD, and CKD. CKM syndrome manifests clinically as a result of interactions among various metabolic risk factors at the pathophysiological level [6]. Surveys indicate that the overall prevalence of CKM syndrome in the general U.S. population approaches 90%, with nearly 15% meeting advanced-stage criteria [7]. Given its strong association with elevated mortality risk and premature death [8], effectively preventing the onset and progression of CKM syndrome is crucial for reducing the significant disease burden it imposes. Chronic inflammation, low socioeconomic status, mental health disorders, sex-specific risk factors, and a family history of kidney failure or diabetes have all been identified as high-risk factors for CKM syndrome [5]. With advancing industrialization, human health has become increasingly affected by environmental factors. The growing global disease burden caused by environmental pollution has become a pressing concern, with toxic metals being of particular concern [9, 10]. Previous research has demonstrated significant associations between toxic metal exposure and the development of obesity [11], metabolic disorders [12], CVD [13], and CKD [14]. In addition to well-established toxicants such as lead and cadmium, emerging evidence suggests that several other metals—including barium (Ba), thallium (Tl), antimony (Sb), tungsten (W), and cobalt (Co)—may also contribute to cardiovascular, renal, and metabolic dysfunction [15–19]. Although these metals are less frequently emphasized in traditional priority frameworks, they are increasingly recognized as potential contributors to CKM-related conditions. Nevertheless, the association between exposure to these metals and the distinct stages of CKM syndrome, as well as the overall prevalence of CKM syndrome, remains uncertain, particularly in the general population under conditions of low exposure. CKM syndrome is a progressively developing condition. Under the influence of multiple factors, it evolves from early adipose tissue accumulation and metabolic disturbances to diabetes, CKD, and cardiovascular-related diseases [20]. Sympathetic overactivity, the renin-angiotensin-aldosterone system (RAAS), inflammation, and oxidative stress all play roles in the initiation and development of diseases linked to CKM syndrome [4, 21–24]. Extensive research has reported that various CKM syndrome factors, including hypertriglyceridemia, hypertension, metabolic syndrome (MetS), diabetes, and CKD, are intimately connected with aging [25, 26]. Studies conducted previously suggest that heavy metal exposure might enhance the expression of pro-inflammatory cytokines, including IL-1β and IL-6, thus playing a role in chronic inflammation [27, 28]. Additionally, it can disrupt reactive oxygen species (ROS) homeostasis via signaling pathways such as NF-κB, NRF2, JAK-STAT, JNK, and FOXO, leading to oxidative stress [29], which further accelerates organismal aging [30, 31]. Consequently, inflammation, oxidative stress, and aging might act as potential mediators in the connection between metal exposure and the development of CKM syndrome. Given the potential adverse health effects of metal exposure, it is essential to explore the potential associations between metal exposure and CKM syndrome. Therefore, we carried out a cross-sectional analysis leveraging data from the National Health and Nutrition Examination Survey (NHANES). This study explored the relationship between urinary concentrations of nine specific metals and metalloids—Ba, cadmium (Cd), Co, cesium (Cs), molybdenum (Mo), lead (Pb), Sb, Tl, and W—and the prevalence of CKM syndrome across various stages. Furthermore, three advanced models for mixture analysis were utilized—Quantile g-computation (Qgcomp), weighted quantile sum regression (WQS), and Bayesian kernel machine regression (BKMR)—to examine the relationships between metal mixtures and the stages of CKM syndrome. Key metals driving these associations across different stages were also identified. Furthermore, mediation analysis was utilized to assess the potential roles of inflammatory responses, oxidative stress, and biological aging as mediators in the connection between metal exposure and CKM syndrome. Methods Study population For this study, data were drawn from 10 consecutive NHANES cycles conducted from 1999 to 2018. The health and nutritional status of adults and children in the United States is assessed by NHANES through a complex, multistage probability sampling approach. The study sample represents the U.S. population across all age groups and is conducted by the National Center for Health Statistics (NCHS) under the Centers for Disease Control and Prevention (CDC). The study was authorized by the NCHS Ethics Review Board, and written informed consent was secured from all individual participants. The 10 NHANES cycles include a total of 101,316 participants. After screening, 55,081 people who were 20 years and above were selected. An additional 1,541 participants were excluded due to pregnancy. Subsequently, 31,865 participants were excluded due to missing fasting status laboratory data. Further, 539 participants with indeterminate CKM syndrome status and 14,396 participants missing urinary metal data were excluded. Finally, 90 participants with missing eGFR and BMI data were also excluded. As a result, 6,650 participants with complete data on CKM syndrome and metal exposure were included in the analysis. All the data employed in this research can be found publicly on the NHANES website: https://www.cdc.gov/nchs/nhanes/index.htm. A detailed overview of the data screening process is presented in Fig. 1. Fig. 1Participant screening flowchart Participant screening flowchart Metal and metalloid measurements Since 1999, NHANES has continuously monitored metal exposure within the general population of the U.S. Metal exposure data for participants are derived from urine sample analyses. Participants were asked to provide sufficient urine samples at mobile examination centers, and these samples underwent processing, storage, and shipment to the Division of Laboratory Sciences at the National Center for Environmental Health, CDC, Atlanta, GA, where inductively coupled plasma dynamic reaction cell mass spectrometry (ICP-DRC-MS) was used to analyze all metallic elements. In cases where metal concentrations were below the limit of detection (LOD), NHANES applies a substitution method by using the LOD divided by the √2. We limited our analysis to nine metals and metalloidls in urine (Ba, Cd, Co, Cs, Mo, Pb, Sb, Tl, W) with detection rates above 80% to minimize potential statistical bias from low detection rates. Detailed detection rates are provided in Supplementary Table 1. Urinary creatinine levels were assessed to adjust for differences in urine concentration. Definition and staging of CKM syndrome CKM syndrome is defined and staged according to the diagnostic and grading criteria outlined in the 2023 Presidential Advisory from the AHA [5]. We determined the staging of CKM syndrome based on participants’ responses to standard questionnaires, as well as their physical examination and laboratory test results, in order to differentiate severity levels. According to the current NHANES data, CKM syndrome is classified into the following stages: Stage 0: No identified risk factors for CKM syndrome. Stage 1: Overweight, dysfunctional obesity, or prediabetes. Stage 2: The presence of at least one condition, such as hypertriglyceridemia, hypertension, diabetes, any component of MetS, or moderate to high-risk CKD. Stage 3: CKD with a very high risk or a predicted 10-year CVD risk of 20% or more [32]. Stage 4: Self-reported clinical CVD, including coronary artery disease, angina, myocardial infarction, heart failure, or stroke. Detailed criteria and definitions for each stage of CKM syndrome are outlined in Supplementary Tables 2 and 3. The CKD-EPI equation based on serum creatinine, excluding adjustments for race or ethnicity was used to calculate the estimated glomerular filtration rate (eGFR) [33]. The 10-year CVD risk was calculated using the AHA Predicting Risk of CVD EVENTs (PREVENT) equations [32]. In the analysis, we combined CKM syndrome stages 3 and 4 to represent advanced CKM syndrome, which is associated with a high predicted risk for CVD [34]. Assessment of covariates Drawing on prior studies [35], the regression analysis incorporated the following covariates: continuous age, sex (male or female), race/ethnicity (including Non-Hispanic White, Non-Hispanic Black, Mexican American, Other Hispanic, and other racial groups), educational level (less than high school, high school graduate/GED, and above high school), smoking status (current smoker or non-smoker), poverty-to-income ratio (PIR: ≤1, 1–4, ≥ 4), marital status (married or unmarried), and NHANES survey cycles. Adjustment of urine Dilution Covariate-adjusted standardized creatinine was employed to account for the dilution of metals in urine [36–39]. This approach helps minimize potential biases from impaired kidney function. Because the application of standard methods for urine dilution adjustment, including the urine metal concentration divided by creatinine concentration, may lead to erroneous conclusions related to impaired renal function [39, 40]. As described in our previous study [41], the process involves two main steps. First, in the overall population, the natural logarithm of urinary creatinine is regressed against factors that may influence creatinine levels (age, sex, race/ethnicity, body mass index, and eGFR) to obtain predicted creatinine values. Next, the measured urinary metal values are divided by the ratio of measured creatinine to predicted creatinine to obtain the error-corrected actual exposure levels of urinary metals, attributed solely to hydration status. Statistical analysis The participants’ baseline characteristics are summarized as means (standard deviations, SD) for continuous variables and as percentages (%) for categorical variables. Due to the significant skewness of urinary metal and creatinine levels, a natural log-transformation was applied to reduce skewness and improve the distribution. Given the relatively low amount of missing covariate data (PIR: 8.1%, marital status: 0.6%, education level: 0.12%, smoking status: 0.09%), multiple imputation was applied to handle missing covariate data in order to maximize statistical power. After transforming the data with natural logarithms, Pearson correlation analysis was utilized to assess the relationships between urinary metals. After adjusting for all covariates, we calculated the geometric means and 95% confidence intervals of urinary metal concentrations across different CKM syndrome stages. The CKM syndrome Stage 0 group was used as the reference group to compare differences in urinary metal concentrations between participants in different CKM syndrome stages [42]. We used the CKM syndrome Stage 0 group as the reference group and employed weighted multinomial logistic regression to explore the association between individual urinary metals and different CKM syndrome stages [43]. To accommodate NHANES’ intricate sampling design, 2-year metal survey subsample weights were used in the weighted regression analysis [Https://Wwwn.Cdc.Gov/Nchs/Data/Nhanes/Analyticguidelines/99-10-Analytic-Guidelines.Pdf.]. The results are presented as odds ratios (ORs) with 95% confidence intervals. Three models were used for the analysis: Model 1 is the basic model; Model 2 accounts for variables such as age, sex, and race/ethnicity; Model 3 includes adjustments for all variables, including age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. The qgcomp model was utilized to examine the comprehensive connection between exposure to metal mixtures and CKM syndrome [44]. The qgcomp model allows for the analysis of outcome variables with multinomial structures and provides weighted estimates of the effects of exposure variables. In addition, the qgcomp model offers several advantages, including no assumptions of directional homogeneity, simplicity of interpretation, and computational ease. It also permits the modeling of both individual exposures and the overall mixture effects in a nonlinear and non-additive manner. The Stage 0 group of CKM syndrome was used as the reference group, and a WQS regression model with 1,000 iterations was employed to explore the relationship between metal mixtures and the stages of CKM syndrome [45]. In the WQS regression, data were randomly split into 40% training and 60% validation sets based on the assumption of directionality homogeneity. A WQS index was constructed, and the relationship between its quartiles and the outcome variable was examined to estimate the cumulative effect of metal mixtures on the different stages of CKM syndrome. Moreover, the model allows for evaluating the weights of each metal individually, thereby identifying key metal elements with significant effects. We used the CKM syndrome Stage 0 group as the reference group and applied the BKMR model with 30,000 iterations to investigate the dose-effect associations between urinary metals and CKM syndrome stages, while also estimating the overall effect of metal mixtures [41, 46]. Additionally, the BKMR model can identify metals with significant effects on CKM syndrome by calculating the posterior inclusion probability (PIP). The BKMR model also allows for the estimation of the risk associations between individual metal elements (at the 75th and 25th percentiles) and the outcome, while simultaneously fixing the other eight metals at the 25th, 50th, and 75th percentiles. Subgroup analysis and sensitivity analysis In the subgroup analysis, we conducted subgroup analyses based on weighted polynomial logistic regression and the qgcomp model for different genders, age groups (< 60 years and ≥ 60 years), and race groups (White, Black, and Other) to validate the robustness of the results across these subgroups. Considering arsenic’s well-established nephrotoxic and cardiometabolic effects and its top ranking on the ATSDR Substance Priority List, a sensitivity analysis was conducted to assess whether inclusion of urinary arsenic would influence the main results. Because arsenic speciation was unavailable in the 1999–2002 NHANES cycles, this analysis was restricted to cycles with complete arsenic data (2003–2018). Inorganic arsenic (iAs = arsenite [AsIII] + arsenate [AsV]) and total arsenic (ΣAs = iAs + MMA + DMA) were calculated following García-Villarino et al. [47]. Weighted logistic regression models were used to examine associations between urinary arsenic (iAs and ΣAs) and individual cardiometabolic and renal outcomes. In addition, the qgcomp model was re-fitted to include arsenic as part of the metal mixture to assess its influence on mixture weights and overall associations with CKM syndrome. Mediation analysis A mediation analysis was conducted to explore how inflammation, oxidative stress, and aging might mediate the link between metal mixtures and CKM syndrome [41, 48]. Alkaline phosphatase was chosen as a marker for inflammation, while albumin served as the marker for oxidative stress [49]. Using the BioAge R package, the Klemera-Doubal method (KDM) was employed to calculate biological age, which serves as a marker of biological aging. The biological age was then regressed against chronological age to obtain residuals, referred to as the KDM residual, with lower residual values indicating a younger biological age [50]. The final estimates and their 95% confidence intervals in the mediation analysis were derived from 1,000 simulations. R version 4.4.1 was used to conduct all analyses. The following R packages were used in the analysis: “qgcomp” (version 2.15.2), “gWQS” (version 3.0.5), “bkmr” (version 0.2.2), “survey” (version 4.4-2.4), “mice” (version 3.16.0), “preventr” (version 0.10.0), “svyVGAM” (version 1.2), “BioAge” (version 0.1.0), and “mediation” (version 4.5.0). Statistical significance was assigned to p-values under 0.05. Results General characteristics of the participants and the distribution of urinary metals As shown in Table 1, among the 6,650 participants, 814 (14.9%) met the criteria for stage 0 CKM syndrome. Participants with CKM syndrome at stage 1 or higher accounted for 85.1% of the total, including 1,122 (18.9%) participants in stage 1, 3,732 (55.6%) in stage 2, and 982 (10.6%) in the advanced stage. As the stage of CKM syndrome increased, the average age of participants significantly increased, along with a notable rise in the proportion of males and Non-Hispanic Black individuals. Higher educational levels and higher PIR were generally associated with lower CKM syndrome stages. Interestingly, the proportion of smokers among participants with advanced CKM syndrome was significantly lower than in the other groups, and serum cotinine levels showed a similar pattern. Regarding urinary creatinine levels, the average level showed a decreasing trend as CKM syndrome stage advanced. The distribution of CKM syndrome stages is presented in Table 1. Table 1Baseline characteristics of participants, (N = 6,650), NHANES, USAOverallCKM syndrome stageCharacteristicOverall, N = 6650 (100%)1Stage 0, N = 814 (14.9%)1Stage 1, N = 1122 (18.9%)1Stage 2, N = 3732 (55.6%)1Advanced CKM syndrome, N = 982 (10.6%)1p-value 2Age (years)47.3 (16.5)36.6 (13.1)40.5 (14.0)48.8 (15.0)66.4 (13.3)< 0.001 Sex < 0.001 female3292 (50%)455 (59%)651 (57%)1767 (47%)419 (46%) male3358 (50%)359 (41%)471 (43%)1965 (53%)563 (54%) Race/ethnicity < 0.001 Non-Hispanic White2941 (69%)401 (73%)437 (66%)1573 (68%)530 (74%) Non-Hispanic Black1320 (10%)125 (7.8%)227 (11%)755 (11%)213 (12%) Mexican American1176 (8.5%)127 (7.5%)226 (10%)714 (8.8%)109 (4.9%) Other/multiracial621 (6.3%)91 (6.4%)116 (5.7%)352 (6.8%)62 (4.6%) Other Hispanic592 (5.7%)70 (5.7%)116 (7.3%)338 (5.5%)68 (3.9%) Education attainment < 0.001 Above than high school3334 (58%)494 (68%)669 (64%)1789 (56%)382 (43%) High school Grad/GED1561 (25%)163 (19%)228 (23%)901 (26%)269 (30%) Less than high school1747 (17%)156 (12%)225 (13%)1036 (18%)330 (27%) PIR < 0.001 ≥ 41560 (35%)215 (37%)318 (40%)868 (35%)159 (24%) 1 ~ 43357 (52%)393 (49%)537 (49%)1873 (52%)554 (61%) ≤ 11195 (13%)145 (14%)183 (12%)677 (13%)190 (16%) Smoking status < 0.001 Non-smoker5302 (80%)616 (77%)931 (84%)2937 (78%)818 (85%) Smoker1342 (20%)197 (23%)189 (16%)792 (22%)164 (15%) Marital status < 0.001 Married3568 (57%)363 (46%)621 (61%)2050 (59%)534 (57%) Non-Married3041 (43%)447 (54%)499 (39%)1650 (41%)445 (43%)Serum cotinine (ng/mL)58 (127)60 (121)53 (130)60 (126)52 (132)< 0.001Urine creatinine (mg/dl)124.92 (76.47)118.03 (78.97)131.24 (74.44)127.45 (78.19)110.02 (63.84)< 0.001 CKM syndrome staging factors Overweight/obesity4739 (69%)0 (0%)1035 (92%)2938 (79%)766 (78%) Abdominal obesity3723 (55%)0 (0%)688 (62%)2380 (65%)655 (70%) Prediabetes926 (11%)0 (0%)134 (11%)628 (14%)164 (17%) Hypertriglyceridemia2223 (33%)0 (0%)0 (0%)1800 (51%)423 (45%) Hypertension3549 (49%)0 (0%)0 (0%)2681 (72%)868 (87%) Diabetes1163 (12%)0 (0%)0 (0%)742 (15%)421 (37%) MetS2246 (30%)0 (0%)0 (0%)1665 (43%)581 (59%) CKD risk low5536 (88%)814 (100%)1122 (100%)3096 (86%)504 (58%) middle981 (11%)0 (0%)0 (0%)636 (14%)345 (31%) high133 (1.1%)0 (0%)0 (0%)0 (0%)133 (11%)Clinical CVD707 (7.9%)0 (0%)0 (0%)0 (0%)707 (75%)10-y CVD risk score0.12 (0.09)0.00 (0.00)0.00 (0.00)0.08 (0.05)0.20 (0.10)1Mean (SD); n (unweighted) (%)2Design-based KruskalWallis test; Pearson’s X^2: Rao & Scott adjustment Table 1Baseline characteristics of participants, (N = 6,650), NHANES, USAOverallCKM syndrome stageCharacteristicOverall, N = 6650 (100%)1Stage 0, N = 814 (14.9%)1Stage 1, N = 1122 (18.9%)1Stage 2, N = 3732 (55.6%)1Advanced CKM syndrome, N = 982 (10.6%)1p-value 2Age (years)47.3 (16.5)36.6 (13.1)40.5 (14.0)48.8 (15.0)66.4 (13.3)< 0.001 Sex < 0.001 female3292 (50%)455 (59%)651 (57%)1767 (47%)419 (46%) male3358 (50%)359 (41%)471 (43%)1965 (53%)563 (54%) Race/ethnicity < 0.001 Non-Hispanic White2941 (69%)401 (73%)437 (66%)1573 (68%)530 (74%) Non-Hispanic Black1320 (10%)125 (7.8%)227 (11%)755 (11%)213 (12%) Mexican American1176 (8.5%)127 (7.5%)226 (10%)714 (8.8%)109 (4.9%) Other/multiracial621 (6.3%)91 (6.4%)116 (5.7%)352 (6.8%)62 (4.6%) Other Hispanic592 (5.7%)70 (5.7%)116 (7.3%)338 (5.5%)68 (3.9%) Education attainment < 0.001 Above than high school3334 (58%)494 (68%)669 (64%)1789 (56%)382 (43%) High school Grad/GED1561 (25%)163 (19%)228 (23%)901 (26%)269 (30%) Less than high school1747 (17%)156 (12%)225 (13%)1036 (18%)330 (27%) PIR < 0.001 ≥ 41560 (35%)215 (37%)318 (40%)868 (35%)159 (24%) 1 ~ 43357 (52%)393 (49%)537 (49%)1873 (52%)554 (61%) ≤ 11195 (13%)145 (14%)183 (12%)677 (13%)190 (16%) Smoking status < 0.001 Non-smoker5302 (80%)616 (77%)931 (84%)2937 (78%)818 (85%) Smoker1342 (20%)197 (23%)189 (16%)792 (22%)164 (15%) Marital status < 0.001 Married3568 (57%)363 (46%)621 (61%)2050 (59%)534 (57%) Non-Married3041 (43%)447 (54%)499 (39%)1650 (41%)445 (43%)Serum cotinine (ng/mL)58 (127)60 (121)53 (130)60 (126)52 (132)< 0.001Urine creatinine (mg/dl)124.92 (76.47)118.03 (78.97)131.24 (74.44)127.45 (78.19)110.02 (63.84)< 0.001 CKM syndrome staging factors Overweight/obesity4739 (69%)0 (0%)1035 (92%)2938 (79%)766 (78%) Abdominal obesity3723 (55%)0 (0%)688 (62%)2380 (65%)655 (70%) Prediabetes926 (11%)0 (0%)134 (11%)628 (14%)164 (17%) Hypertriglyceridemia2223 (33%)0 (0%)0 (0%)1800 (51%)423 (45%) Hypertension3549 (49%)0 (0%)0 (0%)2681 (72%)868 (87%) Diabetes1163 (12%)0 (0%)0 (0%)742 (15%)421 (37%) MetS2246 (30%)0 (0%)0 (0%)1665 (43%)581 (59%) CKD risk low5536 (88%)814 (100%)1122 (100%)3096 (86%)504 (58%) middle981 (11%)0 (0%)0 (0%)636 (14%)345 (31%) high133 (1.1%)0 (0%)0 (0%)0 (0%)133 (11%)Clinical CVD707 (7.9%)0 (0%)0 (0%)0 (0%)707 (75%)10-y CVD risk score0.12 (0.09)0.00 (0.00)0.00 (0.00)0.08 (0.05)0.20 (0.10)1Mean (SD); n (unweighted) (%)2Design-based KruskalWallis test; Pearson’s X^2: Rao & Scott adjustment Baseline characteristics of participants, (N = 6,650), NHANES, USA 1Mean (SD); n (unweighted) (%) 2Design-based KruskalWallis test; Pearson’s X^2: Rao & Scott adjustment The associations of metals in urine, after natural logarithmic transformation, are shown in Fig. 2. Pearson correlation analysis revealed significant positive correlations between urinary metals, with the strongest associations observed between Cs and Tl, as well as Cs and Mo. Fig. 2Correlation of nine urinary metal elements. Note: The color intensity of the red squares and the numerical values indicate the strength of the correlations Correlation of nine urinary metal elements. Note: The color intensity of the red squares and the numerical values indicate the strength of the correlations Urinary metal levels stratified by CKM syndrome stages Table 2 shows the differences in urinary metal levels stratified by CKM syndrome stages. Overall, compared to the CKM syndrome stage 0 group, participants in higher stages generally exhibited higher levels of metal exposure, with the exception of Pb. Additionally, no clear dose-dependent trend in urinary metal levels was observed across different CKM syndrome stages. For Ba, Cd, Cs, Pb, and Tl, metal levels in the CKM syndrome stage 1 group were higher than in the other groups, while for Co, Mo, Sb, and W, higher metal levels were found in participants with advanced CKM syndrome. As shown in Supplementary Fig. 1, the distributions of urinary metal concentrations across different stages of CKM syndrome in the overall population were visualized using violin plots. The distributions in different subgroups are presented in Supplementary Fig. 2. Table 2Differences in urinary metal levels stratified by CKM syndrome stagesUrine metals (ug/L)CKM syndrome stageGeometric mean (95%CI)p-valueBaStage 00.912(0.849,0.981)ReferenceStage 11.239(1.17,1.312)< 0.001Stage 21.213(1.166,1.262)< 0.001Advanced CKM syndrome1.08(0.986,1.184)0.017CdStage 00.265(0.248,0.284)ReferenceStage 10.291(0.278,0.304)0.028Stage 20.284(0.276,0.292)0.172Advanced CKM syndrome0.281(0.265,0.298)0.434CoStage 00.316(0.301,0.332)ReferenceStage 10.344(0.329,0.36)0.015Stage 20.341(0.333,0.35)0.012Advanced CKM syndrome0.411(0.384,0.441)< 0.001CsStage 04.167(3.987,4.354)ReferenceStage 14.673(4.518,4.832)< 0.001Stage 24.482(4.401,4.565)0.007Advanced CKM syndrome4.229(4.052,4.413)0.921MoStage 034.665(33.092,36.313)ReferenceStage 137.848(36.469,39.278)0.005Stage 238.493(37.533,39.477)< 0.001Advanced CKM syndrome41.911(39.867,44.059)< 0.001PbStage 00.478(0.453,0.503)ReferenceStage 10.514(0.491,0.537)0.081Stage 20.486(0.473,0.5)0.848Advanced CKM syndrome0.453(0.43,0.478)0.405SbStage 00.057(0.053,0.06)ReferenceStage 10.06(0.058,0.063)0.093Stage 20.066(0.064,0.067)< 0.001Advanced CKM syndrome0.072(0.068,0.076)< 0.001TlStage 00.153(0.145,0.161)ReferenceStage 10.169(0.163,0.175)0.008Stage 20.168(0.165,0.172)0.001Advanced CKM syndrome0.158(0.149,0.167)0.729WStage 00.057(0.053,0.062)ReferenceStage 10.061(0.057,0.065)0.499Stage 20.066(0.063,0.068)0.002Advanced CKM syndrome0.083(0.076,0.09)< 0.001Note: Geometric means and 95% confidence intervalsModels were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Table 2Differences in urinary metal levels stratified by CKM syndrome stagesUrine metals (ug/L)CKM syndrome stageGeometric mean (95%CI)p-valueBaStage 00.912(0.849,0.981)ReferenceStage 11.239(1.17,1.312)< 0.001Stage 21.213(1.166,1.262)< 0.001Advanced CKM syndrome1.08(0.986,1.184)0.017CdStage 00.265(0.248,0.284)ReferenceStage 10.291(0.278,0.304)0.028Stage 20.284(0.276,0.292)0.172Advanced CKM syndrome0.281(0.265,0.298)0.434CoStage 00.316(0.301,0.332)ReferenceStage 10.344(0.329,0.36)0.015Stage 20.341(0.333,0.35)0.012Advanced CKM syndrome0.411(0.384,0.441)< 0.001CsStage 04.167(3.987,4.354)ReferenceStage 14.673(4.518,4.832)< 0.001Stage 24.482(4.401,4.565)0.007Advanced CKM syndrome4.229(4.052,4.413)0.921MoStage 034.665(33.092,36.313)ReferenceStage 137.848(36.469,39.278)0.005Stage 238.493(37.533,39.477)< 0.001Advanced CKM syndrome41.911(39.867,44.059)< 0.001PbStage 00.478(0.453,0.503)ReferenceStage 10.514(0.491,0.537)0.081Stage 20.486(0.473,0.5)0.848Advanced CKM syndrome0.453(0.43,0.478)0.405SbStage 00.057(0.053,0.06)ReferenceStage 10.06(0.058,0.063)0.093Stage 20.066(0.064,0.067)< 0.001Advanced CKM syndrome0.072(0.068,0.076)< 0.001TlStage 00.153(0.145,0.161)ReferenceStage 10.169(0.163,0.175)0.008Stage 20.168(0.165,0.172)0.001Advanced CKM syndrome0.158(0.149,0.167)0.729WStage 00.057(0.053,0.062)ReferenceStage 10.061(0.057,0.065)0.499Stage 20.066(0.063,0.068)0.002Advanced CKM syndrome0.083(0.076,0.09)< 0.001Note: Geometric means and 95% confidence intervalsModels were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Differences in urinary metal levels stratified by CKM syndrome stages Note: Geometric means and 95% confidence intervals Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Association between urinary metals and CKM syndrome As shown in Table 3, the results of the weighted multinomial logistic regression analysis indicate that, compared to the stage 0 group, higher levels of Ba, Co, and Mo were significantly positively associated with a higher odds of CKM syndrome at advanced stages. High levels of Cs and Tl were significantly positively correlated with the odds of CKM syndrome at stages 1 and 2, but not with advanced CKM syndrome. In addition, higher levels of Cd, Sb, and W were significantly associated with the odds of CKM syndrome at stage 2 and/or advanced stages, but not with stage 1. For Pb, no significant association was found between its levels and CKM syndrome at any stage. For overall CKM syndrome (stage 1 or higher), exposure to all metals was significantly positively associated with the odds of CKM syndrome. Although Cd and Pb also showed positive associations, these associations were not statistically significant. Table 3The association between individual urinary metals and the stages of CKM syndromeUrine metals (ug/L)CKM syndrome stageModel IModel IIModel IIIOR (95%CI)p-valueOR (95%CI)p-valueOR (95%CI)p-valueBaStage 0ReferenceReferenceReferenceStage 11.332(1.192,1.489)< 0.0011.471(1.3,1.663)< 0.0011.541(1.364,1.74)< 0.001Stage 21.284(1.16,1.421)< 0.0011.447(1.285,1.629)< 0.0011.475(1.313,1.659)< 0.001Advanced CKM syndrome0.956(0.827,1.105)0.5431.257(1.068,1.479)0.0061.297(1.098,1.532)0.002Stage all1.245(1.132,1.369)< 0.0011.446(1.294,1.615)< 0.0011.487(1.333,1.659)< 0.001CdStage 0ReferenceReferenceReferenceStage 11.279(1.075,1.522)0.0051.028(0.857,1.232)0.7681.197(0.973,1.473)0.089Stage 21.874(1.559,2.251)< 0.0011.124(0.92,1.374)0.2521.085(0.874,1.348)0.46Advanced CKM syndrome3.217(2.572,4.023)< 0.0011.44(1.137,1.823)0.0031.398(1.078,1.813)0.012Stage all1.813(1.531,2.148)< 0.0011.083(0.903,1.298)0.3931.112(0.909,1.36)0.305CoStage 0ReferenceReferenceReferenceStage 11.288(1.063,1.561)0.011.447(1.156,1.812)0.0011.343(1.072,1.683)0.01Stage 21.126(0.956,1.326)0.1551.43(1.164,1.757)< 0.0011.35(1.098,1.66)0.004Advanced CKM syndrome1.614(1.34,1.942)< 0.0012.076(1.631,2.642)< 0.0011.94(1.53,2.461)< 0.001Stage all1.219(1.049,1.417)0.0111.47(1.214,1.78)< 0.0011.377(1.137,1.667)0.001CsStage 0ReferenceReferenceReferenceStage 11.78(1.358,2.335)< 0.0011.681(1.264,2.236)< 0.0011.728(1.285,2.325)< 0.001Stage 21.694(1.344,2.134)< 0.0011.255(0.98,1.606)0.0711.326(1.03,1.707)0.029Advanced CKM syndrome1.327(1.001,1.759)0.0491.061(0.755,1.49)0.7351.245(0.876,1.769)0.222Stage all1.656(1.322,2.074)< 0.0011.333(1.049,1.694)0.021.401(1.096,1.792)0.008MoStage 0ReferenceReferenceReferenceStage 11.227(1.029,1.465)0.0231.317(1.089,1.593)0.0051.338(1.108,1.617)0.003Stage 21.246(1.073,1.446)0.0041.379(1.161,1.638)< 0.0011.411(1.185,1.679)< 0.001Advanced CKM syndrome1.279(1.063,1.539)0.0091.789(1.44,2.221)< 0.0011.914(1.537,2.385)< 0.001Stage all1.246(1.082,1.434)0.0031.373(1.166,1.616)< 0.0011.402(1.19,1.65)< 0.001PbStage 0ReferenceReferenceReferenceStage 11.068(0.887,1.286)0.4880.933(0.778,1.118)0.4491.208(0.989,1.477)0.065Stage 21.467(1.245,1.727)< 0.0010.98(0.839,1.146)0.8040.993(0.833,1.183)0.936Advanced CKM syndrome1.826(1.512,2.206)< 0.0010.914(0.741,1.128)0.4040.949(0.74,1.217)0.679Stage all1.404(1.198,1.645)< 0.0010.955(0.819,1.113)0.5551.043(0.88,1.236)0.625SbStage 0ReferenceReferenceReferenceStage 10.875(0.756,1.014)0.0750.93(0.787,1.098)0.3891.179(0.978,1.421)0.084Stage 21.084(0.942,1.248)0.2611.284(1.082,1.524)0.0041.481(1.224,1.791)< 0.001Advanced CKM syndrome0.948(0.79,1.137)0.5651.513(1.221,1.873)< 0.0011.969(1.535,2.525)< 0.001Stage all1.02(0.892,1.166)0.7741.165(0.994,1.365)0.0611.386(1.162,1.654)< 0.001TlStage 0ReferenceReferenceReferenceStage 11.362(1.037,1.789)0.0261.49(1.129,1.968)0.0051.402(1.052,1.869)0.021Stage 21.103(0.885,1.376)0.3831.287(1.024,1.617)0.0311.341(1.064,1.69)0.013Advanced CKM syndrome0.536(0.408,0.705)< 0.0011.114(0.809,1.533)0.5091.286(0.927,1.783)0.132Stage all1.045(0.848,1.289)0.6781.31(1.045,1.64)0.021.328(1.055,1.672)0.017WStage 0ReferenceReferenceReferenceStage 10.976(0.842,1.13)0.7411.046(0.89,1.23)0.5821.079(0.914,1.273)0.37Stage 21.062(0.947,1.192)0.3031.224(1.078,1.389)0.0021.22(1.069,1.393)0.003Advanced CKM syndrome1.168(1.012,1.348)0.0331.689(1.41,2.024)< 0.0011.741(1.441,2.104)< 0.001Stage all1.057(0.943,1.183)0.3441.18(1.038,1.342)0.0131.195(1.046,1.364)0.009Note: Model I is the crude model; Model II adjusts for covariates including age, sex, and race/ethnicity; Model III adjusts for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Table 3The association between individual urinary metals and the stages of CKM syndromeUrine metals (ug/L)CKM syndrome stageModel IModel IIModel IIIOR (95%CI)p-valueOR (95%CI)p-valueOR (95%CI)p-valueBaStage 0ReferenceReferenceReferenceStage 11.332(1.192,1.489)< 0.0011.471(1.3,1.663)< 0.0011.541(1.364,1.74)< 0.001Stage 21.284(1.16,1.421)< 0.0011.447(1.285,1.629)< 0.0011.475(1.313,1.659)< 0.001Advanced CKM syndrome0.956(0.827,1.105)0.5431.257(1.068,1.479)0.0061.297(1.098,1.532)0.002Stage all1.245(1.132,1.369)< 0.0011.446(1.294,1.615)< 0.0011.487(1.333,1.659)< 0.001CdStage 0ReferenceReferenceReferenceStage 11.279(1.075,1.522)0.0051.028(0.857,1.232)0.7681.197(0.973,1.473)0.089Stage 21.874(1.559,2.251)< 0.0011.124(0.92,1.374)0.2521.085(0.874,1.348)0.46Advanced CKM syndrome3.217(2.572,4.023)< 0.0011.44(1.137,1.823)0.0031.398(1.078,1.813)0.012Stage all1.813(1.531,2.148)< 0.0011.083(0.903,1.298)0.3931.112(0.909,1.36)0.305CoStage 0ReferenceReferenceReferenceStage 11.288(1.063,1.561)0.011.447(1.156,1.812)0.0011.343(1.072,1.683)0.01Stage 21.126(0.956,1.326)0.1551.43(1.164,1.757)< 0.0011.35(1.098,1.66)0.004Advanced CKM syndrome1.614(1.34,1.942)< 0.0012.076(1.631,2.642)< 0.0011.94(1.53,2.461)< 0.001Stage all1.219(1.049,1.417)0.0111.47(1.214,1.78)< 0.0011.377(1.137,1.667)0.001CsStage 0ReferenceReferenceReferenceStage 11.78(1.358,2.335)< 0.0011.681(1.264,2.236)< 0.0011.728(1.285,2.325)< 0.001Stage 21.694(1.344,2.134)< 0.0011.255(0.98,1.606)0.0711.326(1.03,1.707)0.029Advanced CKM syndrome1.327(1.001,1.759)0.0491.061(0.755,1.49)0.7351.245(0.876,1.769)0.222Stage all1.656(1.322,2.074)< 0.0011.333(1.049,1.694)0.021.401(1.096,1.792)0.008MoStage 0ReferenceReferenceReferenceStage 11.227(1.029,1.465)0.0231.317(1.089,1.593)0.0051.338(1.108,1.617)0.003Stage 21.246(1.073,1.446)0.0041.379(1.161,1.638)< 0.0011.411(1.185,1.679)< 0.001Advanced CKM syndrome1.279(1.063,1.539)0.0091.789(1.44,2.221)< 0.0011.914(1.537,2.385)< 0.001Stage all1.246(1.082,1.434)0.0031.373(1.166,1.616)< 0.0011.402(1.19,1.65)< 0.001PbStage 0ReferenceReferenceReferenceStage 11.068(0.887,1.286)0.4880.933(0.778,1.118)0.4491.208(0.989,1.477)0.065Stage 21.467(1.245,1.727)< 0.0010.98(0.839,1.146)0.8040.993(0.833,1.183)0.936Advanced CKM syndrome1.826(1.512,2.206)< 0.0010.914(0.741,1.128)0.4040.949(0.74,1.217)0.679Stage all1.404(1.198,1.645)< 0.0010.955(0.819,1.113)0.5551.043(0.88,1.236)0.625SbStage 0ReferenceReferenceReferenceStage 10.875(0.756,1.014)0.0750.93(0.787,1.098)0.3891.179(0.978,1.421)0.084Stage 21.084(0.942,1.248)0.2611.284(1.082,1.524)0.0041.481(1.224,1.791)< 0.001Advanced CKM syndrome0.948(0.79,1.137)0.5651.513(1.221,1.873)< 0.0011.969(1.535,2.525)< 0.001Stage all1.02(0.892,1.166)0.7741.165(0.994,1.365)0.0611.386(1.162,1.654)< 0.001TlStage 0ReferenceReferenceReferenceStage 11.362(1.037,1.789)0.0261.49(1.129,1.968)0.0051.402(1.052,1.869)0.021Stage 21.103(0.885,1.376)0.3831.287(1.024,1.617)0.0311.341(1.064,1.69)0.013Advanced CKM syndrome0.536(0.408,0.705)< 0.0011.114(0.809,1.533)0.5091.286(0.927,1.783)0.132Stage all1.045(0.848,1.289)0.6781.31(1.045,1.64)0.021.328(1.055,1.672)0.017WStage 0ReferenceReferenceReferenceStage 10.976(0.842,1.13)0.7411.046(0.89,1.23)0.5821.079(0.914,1.273)0.37Stage 21.062(0.947,1.192)0.3031.224(1.078,1.389)0.0021.22(1.069,1.393)0.003Advanced CKM syndrome1.168(1.012,1.348)0.0331.689(1.41,2.024)< 0.0011.741(1.441,2.104)< 0.001Stage all1.057(0.943,1.183)0.3441.18(1.038,1.342)0.0131.195(1.046,1.364)0.009Note: Model I is the crude model; Model II adjusts for covariates including age, sex, and race/ethnicity; Model III adjusts for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle The association between individual urinary metals and the stages of CKM syndrome Note: Model I is the crude model; Model II adjusts for covariates including age, sex, and race/ethnicity; Model III adjusts for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Association between urinary metal mixtures and CKM syndrome in Qgcomp regression The association between quartiles of metal mixtures and CKM syndrome in the qgcomp model is shown in Table 4. When compared to the Stage 0 group of CKM syndrome, exposure to metal mixtures was significantly positively associated with the prevalence of Stage 1 CKM syndrome (OR: 2.209, 95% CI: 1.78, 2.74). Likewise, higher urinary metal mixture levels were strongly associated with higher odds of CKM syndrome Stage 2 and more advanced stages (OR Stage 2: 2.242, 95% CI: 1.851, 2.715; OR advanced CKM syndrome: 3.401, 95% CI: 2.63, 4.399). In the overall association with CKM syndrome, for every one-quartile increase in the log-transformed urinary metal mixtures, participants in higher quartiles of urinary metal mixtures had 2.207-fold higher odds of CKM syndrome (95% CI: 1.835, 2.654) compared to the Stage 0 group. Table 4The association between urinary metal mixtures and different stages of CKM syndrome in Qgcomp and WQS regressionCKM syndrome stageQgcompWQSOR (95% CI)p-valuePositive OR (95% CI)p-valueNegative OR (95% CI)p-valueStage 0ReferenceReferenceReferenceStage 12.209(1.78,2.74)< 0.0012.412(2.078,2.801)< 0.001Not estimatedStage 22.242(1.851,2.715)< 0.0012.149(1.812,2.548)< 0.001Not estimatedAdvanced CKM syndrome3.401(2.63,4.399)< 0.0014.016(2.596,6.213)< 0.001Not estimatedStage all2.207(1.835,2.654)< 0.0012.201(1.914,2.532)< 0.001Not estimatedNote: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle Table 4The association between urinary metal mixtures and different stages of CKM syndrome in Qgcomp and WQS regressionCKM syndrome stageQgcompWQSOR (95% CI)p-valuePositive OR (95% CI)p-valueNegative OR (95% CI)p-valueStage 0ReferenceReferenceReferenceStage 12.209(1.78,2.74)< 0.0012.412(2.078,2.801)< 0.001Not estimatedStage 22.242(1.851,2.715)< 0.0012.149(1.812,2.548)< 0.001Not estimatedAdvanced CKM syndrome3.401(2.63,4.399)< 0.0014.016(2.596,6.213)< 0.001Not estimatedStage all2.207(1.835,2.654)< 0.0012.201(1.914,2.532)< 0.001Not estimatedNote: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle The association between urinary metal mixtures and different stages of CKM syndrome in Qgcomp and WQS regression Note: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle As shown in Fig. 3, in the weight analysis of the association between exposure to urinary metal mixtures and the odds of CKM syndrome Stage 1, Ba had the highest weight (25%), followed by Tl (20.6%), Sb (15.79%), and Cs (12.13%). For CKM syndrome Stage 2, Ba (25.55%) had the highest weight, followed by Sb (23.42%) and Tl (16.71%). In the odds of advanced CKM syndrome, Sb had the highest weight (28.1%), followed by W (22.16%) and Co (17.68%). In the weight analysis of the overall CKM syndrome (Stage 1 or higher), Ba (23.57%) had the highest weight, followed by Sb (22.09%) and Tl (16.97%) (see Supplementary Fig. 3). Fig. 3Weight analysis of the association of urinary metal mixtures with CKM syndrome in qgcomp regression. Note: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. The horizontal bars represent the relative weights of each metal in relation to CKM syndrome. Black bars indicate positive weights, and gray bars indicate negative weights. Longer bars correspond to greater contributions of the respective metals Weight analysis of the association of urinary metal mixtures with CKM syndrome in qgcomp regression. Note: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. The horizontal bars represent the relative weights of each metal in relation to CKM syndrome. Black bars indicate positive weights, and gray bars indicate negative weights. Longer bars correspond to greater contributions of the respective metals Association between urinary metal mixtures and CKM syndrome in WQS regression Pb showed an negative association with CKM syndrome in the qgcomp model (Fig. 3). However, considering its potential biological role and differences in model assumptions, Pb is retained as a positive contributing factor in the WQS analysis. The results of the WQS regression model are shown in Table 4. Compared to the stage 0 group, higher urinary metal mixture levels were significantly positively associated with the odds of CKM syndrome at all stages. Specifically, for each one-quartile increase in the natural logarithm-transformed urinary metal mixture, the odds of CKM syndrome at stage 1 were higher by 2.412 times (95% CI: 2.078, 2.801), at stage 2 by 2.149 times (95% CI: 1.812, 2.548), and at the advanced stage by 4.016 times (95% CI: 2.596, 6.213). For overall CKM syndrome (stage 1 or higher), exposure to the urinary metal mixture was associated with a 2.201-fold increase in the odds of CKM syndrome (95% CI: 1.914, 2.532) compared to the stage 0 group. As shown in Fig. 4, in the weight analysis of urinary metal mixtures’ exposure and the odds of CKM syndrome at stage 1, Ba had the highest weight (23.11%), followed by Tl (16.66%), Cs (14.07%), Sb (13.32%), and Cd (11.63%). In the weight analysis for CKM syndrome at stage 2, Ba (25.79%) had the highest weight, followed by Sb (22.59%), Tl (13.01%), and Co (11.54%). For the analysis of the odds of advanced CKM syndrome, Sb had the highest weight (32.99%), followed by W (17.63%) and Co (14.23%). For the overall CKM syndrome (stage 1 or higher), Ba (23.37%) had the highest weight, followed by Sb (17.85%), Co (12.59%), and Tl (12.39%). Fig. 4Weight analysis of the association of urinary metal mixtures with different stages of CKM syndrome in WQS regression. Note: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. The height of each colored bar represents the relative weight of a given metal, with taller bars indicating greater contributions. Different colors are used only to distinguish metals and do not imply directionality. The red dashed line denotes the threshold (1 divided by the number of metals) above which a metal is considered to contribute more than expected by chance Weight analysis of the association of urinary metal mixtures with different stages of CKM syndrome in WQS regression. Note: Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. The height of each colored bar represents the relative weight of a given metal, with taller bars indicating greater contributions. Different colors are used only to distinguish metals and do not imply directionality. The red dashed line denotes the threshold (1 divided by the number of metals) above which a metal is considered to contribute more than expected by chance Association between urinary metal mixtures and CKM syndrome in the BKMR model After adjusting for confounding factors, the BKMR model revealed a significant positive association between exposure to metal mixtures and higher odds of CKM syndrome at all stages (Supplementary Fig. 4). The results of the PIP analysis (see Supplementary Table 4) indicate the following: In the weight analysis of urinary metal mixtures’ exposure and the odds of CKM syndrome at stage 1, Tl had the highest weight, followed by Ba (23.75%), Cd (14.26%), and Sb (12.33%). For stage 2 CKM syndrome, Ba and Sb had the highest weights, followed by Tl and Mo. In the analysis of advanced CKM syndrome, Sb had the highest weight, followed by W and Co. For overall CKM syndrome (stage 1 or higher), the weights for Sb and Ba were the highest, followed by Tl. When the levels of other urinary metals were fixed at their median values, the estimated univariate exposure-response curves between each individual metal and the different stages of CKM syndrome are shown in Supplementary Fig. 5. Ba showed a positive linear association with CKM syndrome at stages 1, 2, and overall. Sb exhibited a nonlinear association with CKM syndrome at stages 1, 2, and overall, with a clear saturation effect; however, a positive linear trend was observed in relation to advanced CKM syndrome. The trend for Tl was similar to that of Sb, but there was no significant association with advanced CKM syndrome. Exposure to most other metals showed a positive correlation with CKM syndrome, whereas Pb exhibited a negative correlation with CKM syndrome at stages 2, advanced, and overall. When other metals are fixed at the 25th, 50th, and 75th percentiles, the association between individual metal elements and the odds of Stage 1, Stage 2, advanced, and overall CKM syndrome is shown in Fig. 5. Ba, Sb, and Tl are significantly positively correlated with the odds of Stage 1 CKM syndrome; Ba, Sb, Tl, and W are significantly positively correlated with the odds of Stage 2 CKM syndrome; Co, Mo, Sb, and W are significantly positively correlated with the odds of advanced CKM syndrome; and Ba, Sb, and Tl are significantly positively correlated with the odds of overall CKM syndrome. Subgroup analysis and sensitivity analysis In the gender-based subgroup analysis, the associations between metals and CKM syndrome showed some differences between males and females, as shown in Supplementary Table 5. For both sexes, the associations between CKM syndrome and Ba, Sb, and W were similar. However, compared to females, high levels of Cd were significantly positively associated with the odds of advanced CKM syndrome in males, while no such significant association was found in females. Co was significantly positively associated with the odds of CKM syndrome at all stages in males, whereas in females, it was only significantly associated with advanced CKM syndrome. For females, high levels of Cs were significantly positively associated with the odds of CKM syndrome at stage 2, while no significant association was observed in males. Mo was significantly positively associated with the odds of CKM syndrome at all stages in females, whereas in males, it was only significantly associated with advanced CKM syndrome. Tl was significantly positively associated with the odds of CKM syndrome at stage 2 in females, while in males, it was significantly associated with the odds at stage 1. In the subgroup analysis based on age, participants under 60 years of age exhibited stronger statistical associations between metal exposure and CKM syndrome. Among participants under 60, all metals, except for Pb, were significantly positively associated with CKM syndrome. In contrast, among participants aged 60 and above, only Ba was significantly positively associated with the odds of CKM syndrome Stage 1, while Mo and Sb were significantly positively associated with the odds of advanced CKM syndrome. As shown in Supplementary Table 6, further stratified analyses by race revealed significant differences in the associations between urinary metal exposure and CKM syndrome stages across different racial groups. In the White population, elevated levels of Ba, Cd, Co, Mo, Sb, and W were significantly associated with higher odds of advanced CKM stages. Among these, Ba showed a consistent increasing trend in odds across Stage 1, Stage 2, and advanced stages. Higher urinary levels of Mo, Sb, and W were also significantly associated with higher odds of Stage 2 CKM, while Cs was significantly related to increased risk at Stage 1. In the Black population, Co, Cs, Mo, Sb, Tl, and W exhibited the strongest associations. Notably, Cs was associated with a markedly higher odds (Stage 1: OR = 3.722, 95% CI: 1.738–7.974, P = 0.001; Stage 2: OR = 2.624, 95% CI: 1.261–5.458, P = 0.01), higher than in other racial groups. Co, Mo, Tl, and W demonstrated an increasing trend across CKM syndrome stages and were significantly associated with higher stage categories. In other racial groups, Ba, Cd, Co, Sb, Tl, and W were significantly associated with the odds of CKM syndrome. The results of the mixture analysis using the qgcomp model indicate that, compared with the Stage 0 group, exposure to urinary metal mixtures was significantly and positively associated with the odds of CKM syndrome across all stages in both male and female participants, as well as in those under 60 years of age. However, among participants aged 60 years and older, urinary metal mixture exposure was only significantly and positively associated with the odds of overall and advanced CKM syndrome compared with the Stage 0 group. In race-based analyses, urinary metal mixture exposure was significantly and positively associated with the odds of CKM syndrome at all stages across racial groups, compared with the Stage 0 group, as shown in Supplementary Table 7. The weight analysis results suggest that there are certain differences among male participants (Supplementary Figs. 6 and 7), female participants (Supplementary Figs. 8 and 9), participants under 60 years of age (Supplementary Figs. 10 and 11), those aged 60 and above (Supplementary Figs. 12 and 13), and participants of different racial groups (Supplementary Figs. 14–19). In the sensitivity analysis restricted to NHANES 2003–2018 cycles (n = 5,634) with available urinary arsenic data, higher concentrations of iAs were significantly associated with increased odds of CKM syndrome and its related conditions, including diabetes, hypertension, subclinical heart disease, metabolic syndrome, and CKD (Supplementary Table 8). When iAs was incorporated into the metal mixture using the qgcomp model, the overall association between the mixture and CKM syndrome stages remained significant (Supplementary Table 9). Arsenic showed the highest positive weight contribution, while the relative ranking and directions of other metals remained largely unchanged (Supplementary Fig. 20). Fig. 5Association between individual metal elements and the odds of Stage 1, Stage 2, advanced, and overall CKM syndrome in the BKMR model. Note: A-D: The association between individual metal elements and the odds of Stage 1, Stage 2, advanced, and overall CKM syndrome, with other metals fixed at the 25th, 50th, and 75th percentiles. Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. Each point represents the estimated effect of a given metal on CKM syndrome, while the horizontal lines indicate 95% confidence intervals. Colors denote the fixed percentile levels of other metals in the model: red = 25th percentile, green = 50th percentile, and blue = 75th percentile Association between individual metal elements and the odds of Stage 1, Stage 2, advanced, and overall CKM syndrome in the BKMR model. Note: A-D: The association between individual metal elements and the odds of Stage 1, Stage 2, advanced, and overall CKM syndrome, with other metals fixed at the 25th, 50th, and 75th percentiles. Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. Each point represents the estimated effect of a given metal on CKM syndrome, while the horizontal lines indicate 95% confidence intervals. Colors denote the fixed percentile levels of other metals in the model: red = 25th percentile, green = 50th percentile, and blue = 75th percentile Mediation analysis Mediation analysis suggested that serum albumin and biological aging may partially mediate the associations between metal mixture exposure and CKM syndrome at various stages, as well as overall CKM syndrome, while alkaline phosphatase partially mediated these associations, except for stage 1 CKM syndrome (Fig. 6). Among the mediators, biological aging, represented by KDM residuals, showed the largest estimated proportion of indirect association. Specifically, for the association between metal mixture exposure and CKM syndrome at stage 1, serum albumin mediated approximately 12.09% of the association, while KDM residuals mediated 9.15%. For stage 2 CKM syndrome, alkaline phosphatase mediated 8.87%, serum albumin mediated 9.14%, and KDM residuals mediated 35.79%. For advanced CKM syndrome, alkaline phosphatase mediated 6.99%, serum albumin mediated 8.25%, and KDM residuals mediated 30.35%. For overall CKM syndrome, alkaline phosphatase mediated 8.64%, serum albumin mediated 11.46%, and KDM residuals mediated 23.2%. Fig. 6The potential mediating role of inflammation, oxidative stress, and biological aging in the association between metal mixture exposure and the various stages of CKM syndrome. Note: A-D: The mediating role of alkaline phosphatase, albumin, and KDM residuals in the association between metal mixture exposure and Stage 1, Stage 2, advanced, and overall CKM syndrome. Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. ‘***’ <0.001, ‘**’ <0.01, ‘*’ <0.05. Klemera-Doubal biological age residual (KDM residual); Average direct effects (ADE); Average causal mediation effects (ACME). Proportion of mediation = ACME/Total effect The potential mediating role of inflammation, oxidative stress, and biological aging in the association between metal mixture exposure and the various stages of CKM syndrome. Note: A-D: The mediating role of alkaline phosphatase, albumin, and KDM residuals in the association between metal mixture exposure and Stage 1, Stage 2, advanced, and overall CKM syndrome. Models were adjusted for age, sex, race/ethnicity, education level, smoking status, marital status, PIR, and survey cycle. ‘***’ <0.001, ‘**’ <0.01, ‘*’ <0.05. Klemera-Doubal biological age residual (KDM residual); Average direct effects (ADE); Average causal mediation effects (ACME). Proportion of mediation = ACME/Total effect Discussion This cross-sectional analysis of U.S. adults revealed a notable prevalence of CKM syndrome within the general population, aligning with findings from earlier studies [7]. With the exception of Pb, participants with higher stages of CKM syndrome generally exhibited higher levels of metal exposure, although the levels of nine urinary metals did not show a consistent dose-response trend with increasing CKM syndrome stages. Weighted regression analysis revealed a significant positive association between exposure to specific metals and CKM syndrome in the U.S. general population. These findings suggest that metal exposure may be an important environmental correlate of CKM syndrome, with variations observed across sex, age, and racial/ethnic groups. Compared to older individuals, younger participants appeared to be more sensitive to metal exposure. Results from mixture models, including WQS and BKMR analyses, showed a significant positive association between mixed metal exposure and the incidence of CKM syndrome at various stages. Mediation analyses further suggested that inflammation, oxidative stress, and biological aging may partially mediate or help explain the observed associations between metal co-exposure and CKM syndrome. As far as we know, no study to date has been published that comprehensively examines the combined effects of metal exposure on CVD, CKD, and metabolic-related diseases. Most existing research focuses on the association between metal exposure and specific diseases. In our study, multiple statistical models consistently indicated that Ba, Tl, and Sb were significantly associated with the odds of overall CKM syndrome, as well as with CKM syndrome at stages 1 and 2, with Ba having the highest weight. In the analysis of risk factors for advanced CKM syndrome, Sb had the highest weight, followed by W and Co. A previous meta-analysis revealed that Ba exposure elevates the risk of obesity, whereas exposure to Cd and Pb appears to decrease this risk [15], aligning with our findings. Furthermore, multiple cross-sectional studies have reported comparable results [51–53]. Beyond obesity, other studies have demonstrated that elevated Ba exposure is closely linked to a higher risk of conditions such as T2D, MetS, dyslipidemia, and kidney dysfunction [54–57]. Ba is a commonly encountered metal in daily life, widely used in industrial pigments, medical diagnostics, and other fields. As a toxic metal harmful to human health, Ba exposure has been demonstrated in toxicological studies to diminish the functions of superoxide dismutase, catalase, and acetylcholinesterase in rats subjected to BaCl2 concentrations of 150, 300, and 600 mg/L for 7 days. This reduction results in heightened oxidative stress and elevated IL-6 levels [58, 59]. Furthermore, Ba exposure can damage kidney function, increasing creatinine, urea, and uric acid levels in both blood and urine [60]. Some studies have indicated that exposure to Ba elevates circulating lipid levels [61]. With rising BaCl2 concentrations, plasma low-density lipoprotein cholesterol, triglycerides, and total cholesterol levels increase significantly, whereas high-density lipoprotein cholesterol levels decrease markedly [58]. Thallium is an uncommon but extremely poisonous heavy metal extensively utilized in the semiconductor and electronics sectors, with toxicity greater than that of mercury, Cd, Pb, copper, or zinc [62]. Earlier research has identified a link between plasma Tl levels and a heightened risk of T2D [16], as well as a significant correlation with the risk of gestational diabetes [63]. Furthermore, Tl exposure has been closely linked to obesity and MetS [18, 64]. The potential mechanisms underlying Tl toxicity include its ability to replace K + in Na+/K + ATPase, inhibiting ATPase activity, and suppressing cellular respiration [62]. Tl can also react with thiol groups in proteins and interact with mitochondrial membranes, disrupting cellular metabolism and mitochondrial oxidative phosphorylation [62, 65]. Additionally, Tl can interfere with calcium homeostasis and interact with riboflavin and riboflavin-based cofactors to exert toxicity [66]. It is also known to increase ROS while inhibiting glutathione (GSH) activity, thereby enhancing oxidative stress responses [66, 67]. Furthermore, Tl can disturb the balance of Bad, Bcl-2, and Bax proteins, leading to the activation of caspase-3, caspase-8, and caspase-9 in mitochondria, thereby increasing apoptosis [66]. Sb is a carcinogen, with exposure primarily occurring through ingestion via the gastrointestinal tract and inhalation through the respiratory tract [68]. Multiple studies have identified a strong link between Sb exposure and the onset of diabetes [17, 69]. Exposure to Sb may lead to both cellular and functional harm to the heart, a relationship confirmed by numerous studies. Exposure to Sb has been reported to induce electrocardiogram abnormalities and hypertension [70–72] and is strongly linked to a higher risk of CVD, such as acute myocardial infarction (AMI) and heart failure [73]. Mechanistic studies suggest that the toxic effects of Sb are closely linked to disturbances in redox homeostasis [74].Sb causes oxidative stress, leading to a decrease in cell viability depending on the dose. Exposure to Sb in cells causes a notable rise in mitochondrial ROS (mtROS) levels, leading to mitochondrial dysfunction and cytotoxicity, which are important factors in Sb-induced cell toxicity [75]. Additionally, Sb exposure can induce acute or chronic pancreatic damage [76]. Sb is also a potent metal estrogen, capable of binding to estrogen receptors (ER) [77]. Estrogen receptors are crucial regulators of glucose homeostasis. Overstimulation of ERα can cause insulin secretion imbalance, and persistent disruption may lead to insulin resistance and pancreatic β-cell failure, eventually resulting in diabetes [78]. Moreover, research has shown that exposure to Sb enhances the expression of pro-inflammatory genes, such as TNF-α and COX-2. In cells treated with Sb, TNF-α protein secretion was notably elevated in a dose-dependent fashion [79]. Several cross-sectional studies have reported a link between W exposure and a heightened risk of CKD [19], diabetes [80], elevated blood pressure [81], peripheral artery disease (PAD) [82], stroke [83], and self-reported CVD [84]. Co exposure has been linked to a higher incidence of MetS [18], myocardial dysfunction [85], and CVD [86]. In summary, our study revealed stage-specific dominant metals: Ba, Tl, and Sb played more prominent roles in the early stages of CKM syndrome, whereas Sb, W, and Co emerged as key contributors in the advanced stages. This stage-specific pattern is biologically plausible. In the early stages, low-grade inflammation, oxidative stress, and metabolic dysregulation are central mechanisms of disease initiation. Ba promotes oxidative stress and inflammation [58, 59, 61], Tl disrupts mitochondrial function and energy metabolism [62, 65, 66], and Sb, as an endocrine disruptor, interferes with glucose metabolism, thereby contributing to early metabolic disturbances [77, 78]. By contrast, advanced CKM stages are characterized by overt kidney dysfunction and cardiovascular complications. W has been strongly linked to nephrotoxicity and vascular injury, providing a mechanistic explanation for its stronger association with advanced CKM syndrome [19, 80–84]. Co exposure has been associated with myocardial dysfunction, MetS, and oxidative damage, which are hallmarks of advanced CKM pathology [85, 86]. Notably, Sb remains influential at this stage, likely due to its multifaceted toxic effects—including oxidative stress, inflammation, and endocrine disruption—that not only drive disease onset but also accelerate progression. Overall, these stage-specific associations reinforce the biological plausibility of our findings and underscore the need for longitudinal studies to clarify causal pathways. Mediation analysis suggested that multiple biological pathways—including inflammation, oxidative stress, and aging—may partially mediate or help explain the observed associations between metal exposure and CKM syndrome. Numerous studies have highlighted the close relationship between toxic metal exposure and the activation of inflammation, oxidative stress, and aging pathways [27, 29, 87]. In the pathogenesis of CKM syndrome, excessive and dysfunctional adipose tissue functions as an endocrine organ, secreting adipokines that are thought to contribute to elevated oxidative and inflammatory mediator levels [88]. Obesity-related inflammation and oxidative stress intensify vascular inflammation and promote the pathological processes underlying atherosclerosis [89], thereby raising the risk of hypertension and CVD. Additionally, oxidative stress, inflammation, and hemodynamic changes contribute to kidney damage and glomerulosclerosis [90, 91]. Chronic inflammation in adipose tissue significantly increases the risk of diabetes, with obesity-induced inflammation reducing insulin sensitivity, leading to impaired glucose tolerance and diabetes [92]. In hyperglycemic states, an abundance of intracellular glucose results in the production of mitochondrial superoxide, worsening oxidative stress. This process results in diabetic complications through several mechanisms, such as increased polyol pathway activity, higher advanced glycation end (AGE) production, enhanced expression of AGE receptors and their ligands, activation of protein kinase C isoforms, and excessive activation of the hexosamine pathway [93, 94]. In CKD, sodium and water retention, as well as activation of the RAAS, exacerbate hypertension and increase both the preload and afterload on the heart. Furthermore, cytokine release and chronic inflammation associated with CKD further worsen coronary atherosclerosis and ventricular remodeling [95]. Insulin resistance and endothelial dysfunction have a significant two-way relationship [96]. Heart failure (HF) is a pro-inflammatory disease, with inflammatory biomarkers in HF being closely linked to the development of new-onset T2D. The association indicates that immune-inflammatory processes could be involved in the development of diabetes linked to heart failure [97]. Aging is closely linked to the development of multiple diseases, with diabetes, CVD, and CKD being particularly prominent. Our study found that biological aging, represented by KDM residuals, mediated the strongest association between metal exposure and CKM syndrome, followed by oxidative stress and inflammation. This finding highlights the central role of aging-related mechanisms in CKM pathogenesis and progression. From a translational perspective, interventions targeting biological aging may provide promising opportunities to mitigate metal-induced CKM progression. In recent years, several geroprotective agents such as metformin and rapamycin have been tested in clinical trials [98, 99], while lifestyle interventions including caloric restriction and regular physical activity have been shown to attenuate aging-related pathologies [100, 101]. Moreover, emerging anti-aging therapies are being extensively researched and demonstrate significant potential for clinical translation [102, 103]. Our subgroup analyses revealed that the relationship between metal exposure and CKM syndrome was different for each gender, a trend noted in other research as well. For instance, a study examining the relationship between cadmium exposure and peripheral artery disease (PAD) indicated that, compared to women, the association between cadmium exposure and PAD in men showed a significant dose-response relationship, which was not influenced by smoking status [104]. Co has been reported to have a significant positive correlation with diabetes-related indicators, while this association was not significant in women [105]. Variations in hormone levels between genders [106], redox homeostasis [107], sex-specific gene expression patterns [108], and along with different dietary and behavioral patterns [109], might explain the observed gender differences in the link between metal exposure and CKM syndrome. We also found that the association between metal exposure and CKM syndrome was more pronounced in the younger population compared to older adults. We hypothesize that several mechanisms may explain this phenomenon. First, younger individuals generally have stronger metabolic and detoxification abilities. As a result, after exposure to toxic metals, their bodies may experience more pronounced oxidative stress and inflammatory responses, which are thought to be closely associated with the development of CKM syndrome. Second, older adults may have experienced long-term, low-level exposure to toxic metals, with the effects accumulating over time. These cumulative effects may manifest as chronic damage to cardiovascular and renal function, making CKM syndrome a more subtle, long-term outcome. In contrast, younger individuals exposed to short-term or medium-term metal exposure may be more susceptible to the initiation and progression of CKM syndrome. Third, older individuals were likely exposed to pollution from earlier industrialization or traditional lifestyles (e.g., occupational exposure, past water contamination), whereas younger individuals are more exposed to toxic metals in modern environments (e.g., from electronic products, and fast food packaging). These differences in exposure patterns could lead to varied health outcomes. Additionally, younger individuals could be more prone to the direct metabolic effects of metals, whereas the disease status in older adults may mask or diminish the visibility of such associations. The observed racial differences in the associations between metal and metalloid exposure and CKM syndrome may be explained by variations in environmental exposures, socioeconomic factors, and genetic susceptibility across populations. The additional sensitivity analysis incorporating urinary arsenic provides valuable context for interpreting the robustness of our findings. Arsenic is one of the most extensively studied environmental toxicants, with consistent epidemiologic and mechanistic evidence linking inorganic arsenic exposure to chronic kidney disease, diabetes, and cardiovascular dysfunction [110, 111]. In our arsenic-restricted analysis (2003–2018 NHANES cycles), urinary iAs was significantly associated with higher risks of CKM syndrome and its related components, supporting arsenic’s established role in cardiometabolic and renal toxicity. Importantly, integrating iAs into the metal mixture slightly strengthened the overall associations without materially changing the contribution pattern of other metals, suggesting that the observed mixture effects are not solely driven by arsenic exposure. This consistency underscores the robustness of our primary results while indicating that excluding arsenic in the pooled analysis likely resulted in a conservative estimate of the total environmental metal burden. Pb showed an inverse trend in qgcomp but was retained as a positive contributor in WQS analysis. This divergence likely reflects differences in model assumptions (bidirectional vs. unidirectional weights) and biological factors such as reverse causality—where reduced renal excretion in CKM may lower urinary Pb levels despite higher systemic burden [112]. Our analysis has several key strengths. First, the data comes from the nationally representative NHANES survey, which provides a large sample size for robust analysis. Second, we also applied several analytical techniques, such as weighted multinomial logistic regression, qgcomp for multinomial outcomes. The use of multiple approaches allows for cross-validation, enhancing the stability and reliability of our findings. Third, we applied a novel creatinine adjustment method, which helped avoid biases associated with abnormal metal levels due to impaired kidney function. Fourth, in addition to examining the association between individual and combined metal exposures and CKM syndrome, we also identified key metals associated with different stages of CKM syndrome, offering valuable insights for targeted prevention strategies. Fifth, we conducted a mediation analysis, which helps explain the underlying mechanisms by which metal exposure affects CKM syndrome, thus providing a foundation for more targeted prevention and treatment measures. However, this research is not without its limitations. The most important is its cross-sectional design, which inherently precludes any inference of causality. Thus, the observed associations should be interpreted with caution, as reverse causality cannot be ruled out. Specifically, CKM-related kidney dysfunction may itself alter urinary excretion patterns of metals, thereby complicating the directionality of the observed relationships. Longitudinal studies are warranted to confirm the temporal sequence and further clarify causal pathways. Second, the metal exposure levels were based on spot urine samples, which may introduce some bias. Third, the metal mixtures analysis models do not account for weighting, which may limit their applicability in some scenarios. Fourth, Self-reports from participants were used to diagnose CKM syndrome, particularly clinical CVD, potentially causing recall bias. Fifth, the study’s focus was on the general U.S. population, so the results must be confirmed with additional datasets to validate their general applicability. In addition, only nine urinary metals were included in the main analyses. Although NHANES has measured several additional metals (e.g., manganese, tin, uranium, beryllium, platinum), they were not considered in this study primarily because of incomplete availability across survey cycles or very low detection rates. Metals with inconsistent measurements across cycles (e.g., uranium, beryllium, platinum) are difficult to analyze in pooled datasets, while those with low detection frequencies may reduce statistical power and potentially introduce bias into regression models [113]. Future studies with larger samples or more comprehensive measurements are warranted to further evaluate the roles of these metals. Finally, urinary metal levels primarily reflect excretory capacity and may underestimate systemic exposure in individuals with impaired renal function or compromised nutritional status. More broadly, urinary metal concentrations are influenced by absorption, biotransformation, and renal clearance, and therefore may not fully reflect overall exposure levels. They primarily represent renal excretion capacity rather than total body burden. Therefore, they may underestimate systemic exposure in individuals with impaired kidney function or nutritional deficiencies, where excretion efficiency is reduced despite ongoing environmental intake. This issue is particularly relevant in populations with CKM syndrome, where impaired renal and metabolic function may reduce excretion despite ongoing exposure. Similar concerns have been discussed in our recent work [114]. Conclusions In this study, we found that metal and metalloid exposure is closely associated with the odds of developing various stages of CKM syndrome. Multiple statistical models indicated significant associations between Ba, Tl, and Sb with the overall odds of CKM syndrome, as well as with the odds of Stage 1 and Stage 2 CKM syndrome. Among participants with advanced CKM syndrome, Sb showed the highest contribution, followed by W and Co. Mediation analysis revealed that inflammation, oxidative stress, and biological aging may partially mediate the associations between metal co-exposure and CKM syndrome. These results improve our understanding of CKM syndrome by suggesting that environmental metal exposure may be associated with CKM stages. However, as NHANES reflects exposure patterns in the U.S. general population, the results are most applicable to community-level environmental exposures and should not be directly extrapolated to high-exposure occupational groups. Further validation through prospective cohort studies and mechanistic research is warranted. Supplementary Information Supplementary Material 1. Supplementary Material 1.