Cardiometabolic index (CMI) and sarcopenia as predictors of all-cause and cardiovascular mortality in chronic kidney disease: a NHANES-based cohort study

Abstract

The Cardiometabolic Index (CMI) reflects visceral adiposity and lipid dysregulation, while sarcopenia indicates skeletal muscle depletion, both representing metabolic and functional decline in chronic kidney disease (CKD). This study aimed to investigate the impact of CMI and sarcopenia, as well as their integration into a metabolico-muscular composite risk model, on all-cause and cardiovascular mortality in CKD patients. Based on data from the National Health and Nutrition Examination Survey spanning 1999–2006 and 2011–2018, a total of 1,886 CKD patients were included. Kaplan–Meier survival analysis, multivariable Cox proportional hazards models, restricted cubic spline regression, and subgroup analyses were employed for assessment, while time-dependent receiver operating characteristic analysis was used to evaluate predictive performance. Results showed that patients with both high CMI and sarcopenia had the lowest survival rate. After multivariable adjustment, Cox regression demonstrated that patients in the highest CMI quartile had significantly increased risks of all-cause mortality (HR = 2.25, 95% CI: 1.35–3.73) and cardiovascular mortality (HR = 4.03, 95% CI: 1.52–10.70). Sarcopenia was also associated with increased risks of both mortality types (all-cause: HR = 1.27, 95% CI: 1.10–2.49; cardiovascular: HR = 1.12, 95% CI: 1.06–2.21). Further analysis identified nonlinear relationships between CMI and both all-cause and cardiovascular mortality, regardless of sarcopenia status. This longitudinal cohort study demonstrates that elevated CMI and sarcopenia are independently associated with increased mortality risk in CKD patients, with the highest risk observed when both conditions coexist. Therefore, this study positions CMI and sarcopenia as prognostic biomarkers for mortality risk stratification in CKD.

Graphical abstract

This figure was created with BioRender.com (https://www.biorender.com/)

1. Introduction

Chronic kidney disease (CKD), characterized by its distinctive “three highs and one low” epidemiological profile—high prevalence, high disability rate, high disease burden, and low awareness—has become a major global public health challenge [1–3]. With progressive renal function decline, many patients inevitably advance to end-stage kidney disease (ESKD), which is accompanied by systemic complications, particularly cardiovascular disease (CVD), the leading cause of death in this population [4–6]. Given its growing burden, the World Health Organization has officially recognized CKD as a priority non-communicable disease requiring global attention and intervention.

Dyslipidemia and obesity are key drivers in the progression of CKD. Existing evidence indicates that CKD-related dyslipidemia promotes the development of CVD, which is the leading cause of death in CKD patients. Studies to date have confirmed significant dysregulation of serum lipid profiles in CKD [7]. A study based on the Cardiovascular Health Study cohort found that plasma sphingolipids partially mediate the association between low estimated glomerular filtration rate (eGFR) and incident heart failure [8]. However, current indicators for assessing mortality risk in CKD patients based on lipid levels remain inadequate. The prevalence of obesity is higher in CKD patients than in the general population of the same age, and there is a recognized association between obesity and kidney failure [9]. Obesity can not only directly cause CKD but also indirectly impair renal function through mechanisms such as the secretion of various adipokines, modulation of inflammation and glucose metabolic homeostasis, and induction of functional lipid metabolism disorders [9]. The Cardiometabolic Index (CMI) is an indicator that combines obesity and dyslipidemia.

CMI integrates the waist-to-height ratio (WHtR) and the triglyceride (TG) to high-density lipoprotein cholesterol (HDL-C) ratio as a surrogate of cardiometabolic dysfunction (CMI = [TG/HDL-C] × WHtR) [10]. In the general population, elevated CMI has been consistently associated with cardiometabolic disorders and cardiovascular risk. However, how CMI influences its prognostic role in CKD remains poorly understood.

Sarcopenia, defined as a progressive decline in skeletal muscle mass and function, is increasingly recognized as a common comorbidity in CKD [11–14]. The uremic milieu creates a profound metabolic disturbance—characterized by protein-energy wasting, chronic inflammation, metabolic acidosis, and impaired protein synthesis—that accelerates muscle loss [15–17]. A recent meta-analysis involving more than 42,000 CKD patients reported that the prevalence of sarcopenia reached 16.7% in non-dialysis CKD, 28.8% in hemodialysis, 16.3% in peritoneal dialysis, and 18.4% in kidney transplant recipients [18]. Beyond impairing physical function, skeletal muscle is a crucial metabolic and endocrine organ, and its loss contributes to systemic inflammation, insulin resistance, and metabolic dysregulation, thereby worsening CKD prognosis [19,20].

Currently, metabolic abnormalities and sarcopenia are not only highly prevalent in CKD but also mechanistically intertwined. This suggests that their combined impact on adverse outcomes may be greater than the effect of either factor alone. In CKD management, metabolic health and muscle health are two sides of the same coin. Integrating metabolic and muscle indicators is a crucial step toward achieving precise risk stratification and personalized management for CKD patients. Therefore, this study utilizes data from the National Health and Nutrition Examination Survey (NHANES) to explore the independent and joint associations of the CMI and sarcopenia with the risks of all-cause and cardiovascular mortality in CKD patients. The aim is to provide new insights for risk stratification in CKD and to inform personalized prevention and intervention strategies for this high-risk population.

2. Methods

2.1. Data collection and study population

This study utilized data from the National Health and Nutrition Examination Survey (NHANES), a nationally representative cross-sectional survey administered by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC). NHANES employs a complex, multistage, stratified probability sampling design to obtain health-related data from the non-institutionalized civilian U.S. population via standardized interviews, physical examinations, and laboratory tests. The survey protocol was reviewed and approved by the NCHS Institutional Review Board, and all participants provided written informed consent prior to participation.

We extracted multidimensional data from the publicly available NHANES database (https://www.cdc.gov/nchs/nhanes/), including demographic characteristics, anthropometric measures, laboratory biomarkers, and questionnaire-based variables. The reporting of this study adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines to ensure methodological transparency and rigor. The NHANES 1999–2006 and 2011–2018 cycles were specifically selected because these periods provide consistent availability of dual-energy X-ray absorptiometry (DXA) data, which is essential for the assessment of sarcopenia-related variables. The gap between cycles (excluding 2007–2010) is due to the absence of DXA scans during those survey years, which would have introduced sample selection bias if included. Furthermore, these selected cycles are the only ones that concurrently offer both mortality linkage data and the complete set of variables required for sarcopenia evaluation. The initial combined cohort comprised 80,630 individuals.

The following exclusion criteria were applied:

1. Individuals not meeting diagnostic criteria for CKD;

2. Missing data on estimated glomerular filtration rate (eGFR) or urinary albumin-to-creatinine ratio (UACR);

3. Age under 18 years;

4. Insufficient data for the calculation of sarcopenia or the cardiometabolic index (CMI);

5. Incomplete or ineligible follow-up for mortality outcomes.

After applying these exclusions, a final analytic sample of 1886 eligible participants was retained for analysis (Figure 1).

Figure 1.

Flow diagram illustrating participant selection from NHANES cycles (1999–2006 and 2011–2018) for inclusion in the chronic kidney disease (CKD) cohort with available cardiometabolic index (CMI), sarcopenia metrics, and mortality follow-up data.

2.2. Assessment of CKD

eGFR was estimated by Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) 2021 equation [21]. CKD was defined as eGFR < 60 mL/min/1.73 m² or UACR ≥ 30 mg/g. CKD stages are classified based on eGFR levels as I, II, III, IV and V.

2.3. Definitions of sarcopenia (FNIH criteria) and CMI (triglyceride/HDL-C × WHtR)

Body composition in the NHANES was evaluated by DXA. From these measurements, appendicular lean mass (ALM) was calculated as the total muscle mass of the four limbs. Sarcopenia was defined based on the Foundation for the National Institutes of Health (FNIH) criteria, which utilizes the ratio of ALM to body mass index (BMI) [22]. The cut-off values for sarcopenia index were < 0.789 kg/m² for men and < 0.512 kg/m² for women.

Based on anthropometric and laboratory data, the CMI was calculated using the formula established in previous studies. The initial calculation of the WHtR involved the use of waist circumference (WC) (cm) and body height (BH) (cm). Subsequently, the CMI was computed by integrating HDL-C and TG levels. The formula is as follows: CMI = TG/HDL-C × WHtR. WHtR = waist circumference/height [23]. For analytical purposes, study participants were categorized into four groups (Q1-Q4) based on the quartiles of the CMI value distribution (i.e., using the 25th, 50th, and 75th percentiles as cut points).

2.4. Assessment of covariates

To investigate the independent associations between sarcopenia and CMI with all-cause and cause-specific mortality in CKD, we adjusted for potential confounding variables that might influence these associations based on clinical relevance. The covariates included: age, sex, BMI (defined as weight (in kilograms) divided by the square of height (in meters), and categorized as follows: <18.5 (underweight), 18.5–24.9 (normal weight), or ≥25 (overweight or obese)), race (Mexican American, other Hispanic, non-Hispanic Black, non-Hispanic White, and other races), poverty-income ratio (PIR), education level (less than high school grade, more than high school grade); smoking (participants were divided into two groups according to current smoking status: yes or no), and drinking frequency (non-drinker, 1–5 drinks/month, 5–10 drinks/month, and >10 drinks/month); diabetes mellitus (participants were considered to have diabetes if they met at least one of the following criteria: a) diagnosis of diabetes; b) use of glucose-lowering medications; c) HbA1c ≥6.5%; or d) fasting glucose ≥126 mg/dL) [24] and hypertension (defined as systolic blood pressure [SBP] ≥ 130 mmHg or diastolic blood pressure [DBP] ≥ 80 mmHg, where SBP = [second SBP + third SBP]/2 and DBP = [second DBP + third DBP]/2) [25].

Medication use was also assessed and included as covariates, specifically the use of statins, renin-angiotensin system inhibitors (RASI), and steroids.

2.5. Determination of death

The primary outcomes were all-cause and cardiovascular mortality in the CKD population. The specific cause of death was identified using the 10th revision of the International Classification of Diseases (ICD-10). All-cause mortality refers to death from any cause, while cardiovascular mortality primarily includes ischemic heart disease and heart failure, among others. To ascertain vital status and specific causes of death during follow-up, we utilized the updated mortality data from NHANES, which are linked to the National Death Index [26].

2.6. Statistical analyses

In this study, eligible data were extracted from the NHANES database. We identified missing data in the chronic kidney disease cohort (with the exception of exposure and outcome variables). The distribution of missing values across the study variables is presented in Table S1 and Figure S1. Specifically, the proportions of missing data for variables included in the imputation model were as follows: LDL-C (17.5%), education (10.1%), smoking (9.4%), globulin (9.0%), uric acid (9.0%), potassium (9.0%), calcium (9.0%), ALT (9.0%), hypertension (7.1%), TC (6.7%), BMI (3.1%), PIR (2.9%), statins (0.1%), RASI (0.1%), and steroids (0.1%). Given that the missing data pattern was random (Missing at Random, MAR) and the proportion of missingness for any individual variable fell within an acceptable range, multiple imputation by chained equations (MICE) was ultimately employed to preserve the effective sample size and minimize selection bias [24].

Categorical variables are expressed as numbers and percentages (%), while continuous variables are expressed as medians and 25–75th percentiles. Kaplan-Meier survival analyses with log-rank tests were performed to compare event-free survival across CMI quartiles and sarcopenia categories. Associations of CMI and sarcopenia with all-cause and cardiovascular mortality were evaluated using multivariable Cox proportional hazards regression models via the “survival” package. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated in three models: Model 1 (unadjusted), Model 2 (adjusted for age, sex, race, poverty-income ratio, education, and BMI), and Model 3 (adjusted for all covariates). Restricted cubic spline (RCS) models, implemented with the “plotRCS” package, were used to explore potential non-linear relationships between exposures and outcomes. Subgroup analyses were conducted stratified by age, sex, race, education, smoking status, drinking frequency, diabetes, hypertension, CKD stage, and medication use. Time-dependent receiver operating characteristic (ROC) curves were generated to assess the predictive performance of CMI and sarcopenia for mortality at 36, 60, and 120 months. To rigorously evaluate the model’s robustness and mitigate potential overfitting, the discriminative performance was further validated using 5-fold cross-validation. All statistical tests were two-sided, and a p value < 0.05 was considered statistically significant. Analyses were performed using R software (version 4.4.2).

3. Results

3.1. Baseline characteristics of the study population

A total of 1,886 participants with CKD were included, with a median age of 51.0 years (IQR: 41.0–61.0), of whom 26% were male. The majority were overweight (72%) and non-Hispanic White (44%). Baseline characteristics stratified by CMI quartiles, which were defined based on percentile cutoffs (≤25th, >25th to ≤50th, >50th to ≤75th, and >75th percentile) of the CMI distribution, are presented in Table 1. Compared with those in the lowest quartile (Q1), participants in the highest quartile (Q4) were more likely to be male, overweight, and non-Hispanic White; have a higher PIR; attain an education level beyond high school; report consuming ≥10 alcoholic drinks per month; and be current smokers. They also demonstrated a higher prevalence of diabetes mellitus, hypertension, and sarcopenia. Regarding medication use, the utilization rates of statins, renin-angiotensin system inhibitors (RASI), and steroids significantly increased across the CMI quartiles (Q1 to Q4: statins 7.6–23.0%; RASI 9.1–32.6%; steroids 6.1–9.4%). Furthermore, individuals in Q4 had elevated levels of uric acid, TG, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), alanine aminotransferase (ALT), globulin, and potassium, along with lower levels of HDL-C. All differences across CMI quartiles were statistically significant (p < 0.001).

Table 1.

Baseline demographic, metabolic, and clinical characteristics of chronic kidney disease (CKD) patients stratified by cardiometabolic index (CMI) quartiles in NHANES.

Characteristic	Total, N = 1886	CMI				P b
		Q1, N = 407 (25%)a	Q2, N = 475 (25%)a	Q3, N = 483 (25%)a	Q4, N = 521 (25%)a
Age, years	51.00 (41.00, 61.00)	47.00 (36.00, 54.00)	50.00 (40.00, 57.00)	54.00 (45.00, 61.00)	52.00 (42.00, 59.00)	<0.001
Sex						<0.001
Female	1400 (74.2%)	344 (84.5%)	386 (81.3%)	358 (74.1%)	312 (59.9%)
Male	486 (25.8%)	63 (15.5%)	89 (18.7%)	125 (25.9%)	209 (40.1%)
BMI, kg/m2						<0.001
Underweight	36 (1.9%)	28 (6.9%)	7 (1.5%)	1 (0.2%)	0 (0.0%)
Normal weight	499 (26.5%)	226 (55.5%)	133 (28.0%)	84 (17.4%)	56 (10.7%)
Overweight	1351 (71.6%)	153 (37.6%)	335 (70.5%)	398 (82.4%)	465 (89.3%)
Race						<0.001
Mexican American	275 (14.6%)	36 (8.8%)	45 (9.5%)	78 (16.1%)	116 (22.3%)
Non_Hispanic Black	532 (28.2%)	145 (35.6%)	171 (36.0%)	122 (25.3%)	94 (18.0%)
Non_Hispanic White	834 (44.2%)	180 (44.2%)	203 (42.7%)	213 (44.1%)	238 (45.7%)
Other Hispanic	109 (5.8%)	18 (4.4%)	23 (4.8%)	32 (6.6%)	36 (6.9%)
Other Race	136 (7.2%)	28 (6.9%)	33 (6.9%)	38 (7.9%)	37 (7.1%)
PIR	2.22 (1.15, 4.16)	4.10 (2.00, 5.00)	2.73 (1.36, 4.70)	2.72 (1.44, 4.86)	2.76 (1.33, 4.93)	<0.001
Education						<0.001
Less Than High School Grade	465 (24.7%)	63 (15.5%)	95 (20.0%)	127 (26.3%)	180 (34.5%)
More Than High School Grade	1421 (75.3%)	344 (84.5%)	380 (80.0%)	356 (73.7%)	341 (65.5%)
Drinking frequency						<0.001
Non-drinker	471 (25.0%)	106 (26.0%)	127 (26.7%)	117 (24.2%)	121 (23.2%)
1&5 drinks a month	533 (28.3%)	150 (36.9%)	134 (28.2%)	101 (20.9%)	148 (28.4%)
5&10 drinks a month	70 (3.7%)	12 (2.9%)	18 (3.8%)	16 (3.3%)	24 (5.0%)
10+ drinks a month	812 (43.1%)	139 (34.2%)	196 (41.3%)	249 (51.6%)	228 (43.8%)
Smoking						<0.001
Yes	816 (43.3%)	145 (35.6%)	192 (40.4%)	203 (42.0%)	276 (53.0%)
No	1070 (56.7%)	262 (64.4%)	283 (59.6%)	280 (58.0%)	245 (47.0%)
Diabetes mellitus						<0.001
Yes	1433 (76.0%)	381 (93.6%)	408 (85.9%)	359 (74.3%)	285 (54.7%)
No	453 (24.0%)	26 (6.4%)	67 (14.1%)	124 (25.7%)	236 (45.3%)
Hypertension						<0.001
Yes	1006 (53.3%)	284 (69.8%)	258 (54.3%)	254 (52.6%)	210 (40.3%)
No	880 (46.7%)	123 (30.2%)	217 (45.7%)	229 (47.4%)	311 (59.7%)
Sarcopenia						<0.001
Yes	155 (8.2%)	8 (2.0%)	22 (4.6%)	49 (10.1%)	76 (14.6%)
No	1731 (91.8%)	399 (98.0%)	453 (95.4%)	434 (89.9%)	445 (85.4%)
CKD stage						<0.001
I	242 (12.8%)	38 (9.3%)	44 (9.3%)	56 (11.6%)	104 (20.0%)
II	425 (22.5%)	87 (21.4%)	104 (21.9%)	101 (20.9%)	133 (25.2%)
III	1157 (61.3%)	278 (68.3%)	313 (65.9%)	309 (64.0%)	257 (49.3%)
IV	48 (2.5%)	3 (0.7%)	10 (2.1%)	15 (3.1%)	20 (3.8%)
Ⅴ	14 (0.7%)	1 (0.2%)	4 (0.8%)	2 (0.4%)	7 (1.3%)
Medication use						<0.001
Statins​	325 (17.2%)	31 (7.6%)	69 (14.5%)	105 (21.7%)	120 (23.0%)
RASI	441 (23.4%)	37 (9.1%)	110 (23.2%)	124 (25.7%)	170 (32.6%)
Steroids​	166 (8.8%)	25 (6.1%)	37 (7.8%)	55 (11.4%)	49 (9.4%)
eGFR, mL/min/1.73m²	56.85 (49.57, 74.01)	55.58 (50.63, 61.20)	56.25 (50.50, 62.97)	56.54 (49.15, 65.32)	58.07 (48.80, 84.57)	0.104
Uric acid, mg/dL	5.30 (4.40, 6.40)	4.30 (3.70, 5.10)	5.00 (4.30, 5.80)	5.40 (4.60, 6.30)	6.20 (5.30, 7.10)	<0.001
TG, mmol/L	1.25 (0.85, 1.92)	0.68 (0.53, 0.84)	1.01 (0.85, 1.17)	1.42 (1.22, 1.69)	2.46 (1.99, 3.11)	<0.001
TC, mg/dL	196.50 (173.00, 226.00)	193.00 (168.00, 217.00)	194.00 (175.00, 218.00)	200.00 (173.00, 226.00)	209.00 (181.00, 247.00)	<0.001
HDL-C, mg/dL	53.00 (44.00, 65.00)	75.00 (66.00, 87.00)	60.00 (52.00, 66.00)	51.00 (46.00, 57.00)	41.00 (36.00, 48.00)	<0.001
LDL-C, mg/dL	116.00 (94.00, 140.00)	104.00 (86.00, 124.00)	117.00 (97.00, 139.00)	121.00 (101.00, 144.00)	123.00 (98.00, 153.00)	<0.001
ALT, U/L	19.00 (15.00, 26.00)	17.00 (15.00, 22.00)	18.00 (15.00, 23.00)	20.00 (15.00, 26.00)	24.00 (18.00, 33.00)	<0.001
Globulin, g/dL	3.00 (2.70, 3.30)	2.80 (2.50, 3.10)	2.90 (2.60, 3.20)	2.90 (2.60, 3.20)	3.10 (2.80, 3.30)	<0.001
Potassium, mmol/L	4.00 (3.80, 4.20)	4.00 (3.80, 4.20)	4.00 (3.80, 4.20)	4.00 (3.78, 4.30)	4.10 (3.90, 4.35)	<0.001
Calcium, mg/dL	9.40 (9.20, 9.60)	9.40 (9.20, 9.60)	9.40 (9.20, 9.60)	9.40 (9.20, 9.60)	9.40 (9.20, 9.70)	0.574

Abbreviations: BMI, body mass index; PIR, poverty-income ratio; RASI, renin-angiotensin system inhibitor; eGFR, estimated glomerular filtration rate; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TC, total cholesterol; TG, triglycerides; ALT, alanine aminotransferase.

^aMedian (IQR) for continuous; n (%) for categorical.

^bKruskal-Wallis test for continuous variables; Chi-square test for categorical variables.

3.2. Association of CMI and sarcopenia with all-cause and cardiovascular mortality in patients with CKD

During a median follow-up period of 97 months, 400 out of 1,886 CKD patients (21.2%) died, among whom 144 (7.6%) died from cardiovascular diseases. The median survival time for all-cause mortality was 104.00 (IQR, 50.00–146.00) months, and for cardiovascular mortality, it was 98.50 (41.75–139.25) months. Kaplan-Meier (K-M) survival analysis revealed significant differences in the risks of all-cause and cardiovascular mortality among CKD patients based on CMI and sarcopenia status (Figure 2). Patients in the highest CMI quartile (Q4 (≥75th percentile of CMI distribution)) exhibited markedly lower survival probabilities for both all-cause mortality (Figure 2A, p < 0.0001) and cardiovascular mortality (Figure 2B, p < 0.0001), compared with those in the lower quartiles. Similarly, patients with sarcopenia demonstrated significantly reduced survival compared with non-sarcopenic individuals for both all-cause mortality (Figure 2C, p < 0.0001) and cardiovascular mortality (Figure 2D, p = 0.0045). Notably, when stratified by both CMI quartiles and sarcopenia status, the joint effect was evident: patients with both high CMI (Q4) and sarcopenia had the lowest survival probabilities, while those with low CMI (Q1) and without sarcopenia had the most favorable prognosis (Figures 2E and F, p < 0.0001). These findings suggest a synergistic impact of metabolic dysregulation and sarcopenia on mortality risk among CKD patients.

Figure 2.

Kaplan–Meier survival curves demonstrating associations between cardiometabolic index (CMI) quartiles, sarcopenia status, and risks of all-cause mortality (A, C, E) and cardiovascular mortality (B, D, F) in chronic kidney disease (CKD) patients from NHANES.

Table 2 shows the link between the CMI levels with all-cause and cardiovascular mortality in adults with CKD. Higher CMI levels were associated with increased risks of all-cause and cardiovascular mortality in the unadjusted model. After adjustment for all covariates, compared with those in the lowest quartile (Q1), participants in the highest quartile (Q4) exhibited significantly elevated risks of all-cause mortality (HR = 2.25, 95% CI 1.35–3.73, p = 0.002) and cardiovascular mortality (HR = 2.03, 95% CI 1.52–10.70, p = 0.005). Significant associations were also observed for the intermediate quartile Q2 with both all-cause mortality (HR = 1.43, 95% CI 1.19–3.45, p = 0.009) and cardiovascular mortality (HR = 1.75, 95% CI 1.61–10.70, p = 0.003). However, the associations were attenuated and not statistically significant in the Q3 group.

Table 2.

Association between cardiometabolic index (CMI) quartiles and risks of all-cause and cardiovascular mortality in adults with chronic kidney disease (CKD).

CMI group	All-cause mortality
	Model 1		Model 2		Model 3
	95% CI	p	95% CI	p	95% CI	P
Q1	ref		ref		ref
Q2	2.99 (1.87, 4.76)	<0.001	2.00 (1.19, 3.35)	0.009	1.43 (1.19, 3.45)	0.009
Q3	3.08 (1.98, 4.80)	<0.001	1.39 (0.82, 2.34)	0.218	1.15 (0.79, 2.20)	0.291
Q4	4.52 (2.97, 6.86)	<0.001	2.48 (1.50, 4.10)	<0.001	2.25 (1.35, 3.73)	0.002
CMI group	Cardiovascular mortality
	Model 1		Model 2		Model 3
	95% CI	P	95% CI	P	95% CI	P
Q1	ref		ref		ref
Q2	6.40 (2.72, 15.10)	<0.001	4.55 (1.74, 11.90)	0.002	1.75 (1.61, 10.70)	0.003
Q3	5.99 (2.36, 15.20)	<0.001	2.65 (0.99, 7.09)	0.052	1.22 (0.94, 5.72)	0.069
Q4	8.83 (3.49, 22.40)	<0.001	5.09 (1.88, 13.80)	0.001	2.03 (1.52, 10.70)	0.005

Model 1: no covariates were adjusted.

Model 2: Age, sex, race, PIR, education, BMI were adjusted.

Model 3: Age, sex, race, PIR, education, BMI, drinking frequency, smoking, diabetes mellitus, hypertension, medication use were adjusted.

Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.

In the unadjusted model, sarcopenia was significantly associated with an increased risk of both all-cause mortality (HR = 2.45, 95% CI: 1.64–3.65, p < 0.001) and cardiovascular mortality (HR = 2.47, 95% CI: 1.23–4.95, p = 0.001). After adjustment for age, sex, race, PIR, education, BMI, drinking frequency, smoking, diabetes mellitus, hypertension, and medication use, sarcopenia was independently associated with an increased risk of all-cause mortality (HR = 1.27, 95% CI: 1.10–2.49, p = 0.045) and cardiovascular mortality (HR = 1.12, 95% CI: 1.06–2.21, p = 0.048) (Table 3).

Table 3.

Independent association between sarcopenia status and all-cause and cardiovascular mortality among chronic kidney disease (CKD) patients.

Sarcopenia group	All-cause mortality
	Model 1		Model 2		Model 3
	95% CI	P	95% CI	p	95% CI	P
Non-sarcopenia	ref		ref		ref
Sarcopenia	2.45 (1.64, 3.65)	< 0.001	1.92 (1.21, 3.05)	0.006	1.27 (1.10, 2.49)	0.045
Group	Cardiovascular mortality
	Model 1		Model 2		Model 3
	95% CI	p	95% CI	p	95% CI	p
Non-sarcopenia	ref		ref		ref
Sarcopenia	2.47 (1.23, 4.95)	0.001	1.86 (1.13, 2.29)	0.024	1.12 (1.06, 2.21)	0.048

Model 1: no covariates were adjusted.

Model 2: Age, sex, race, PIR, education, BMI were adjusted.

Model 3: Age, sex, race, PIR, education, BMI, drinking frequency, smoking, diabetes mellitus, hypertension, medication use were adjusted.

Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.

Table 4 presents the association between the metabolic-muscle composite risk model (integrating CMI and sarcopenia) and mortality risk in patients with chronic kidney disease. This model stratified patients into different risk tiers based on CMI quartiles and sarcopenia status. For all-cause mortality, there was a trend of increasing mortality risk with ascending metabolic-muscle composite risk levels. In Model 3, compared with the lowest-risk tier (Q1 without sarcopenia), the intermediate-risk tiers (Q2 and Q3 with sarcopenia) showed significantly elevated risks (HR = 1.84, 95% CI: 1.41–2.18, p = 0.011; HR = 1.28, 95% CI: 1.09–2.19, p = 0.032), while the highest-risk tier (Q4 with sarcopenia) exhibited the peak risk (HR = 2.36, 95% CI: 1.22–3.81, p = 0.003).

Table 4.

Joint impact of cardiometabolic index (CMI) quartiles and sarcopenia on all-cause and cardiovascular mortality in chronic kidney disease (CKD).

CMI-Sarcopenia group	All-cause mortality
	Model 1		Model 2		Model 3
	95% CI	p	95% CI	p	95% CI	P
Q1 without Sarcopenia	ref		ref		ref
Q1 with Sarcopenia	2.73 (1.52, 14.40)	<0.001	1.78 (1.28, 3.35)	<0.001	1.38 (1.19, 3.29)	0.008
Q2 without Sarcopenia	2.87 (1.78, 4.65)	<0.001	2.05 (1.48, 2.84)	0.006	1.67 (1.26, 2.27)	0.035
Q2 with Sarcopenia	5.94 (2.33, 15.10)	<0.001	2.15 (1.98, 4.69)	<0.001	1.84 (1.41, 2.18)	0.011
Q3 without Sarcopenia	2.94 (1.87, 4.63)	<0.001	1.27 (0.96, 1.67)	0.089	1.12 (0.65, 1.39)	0.121
Q3 with Sarcopenia	5.23 (2.44, 11.20)	<0.001	1.91 (1.33, 2.75)	<0.001	1.28 (1.09, 2.19)	0.032
Q4 without Sarcopenia	4.19 (2.73, 6.44)	<0.001	2.39 (1.87, 3.06)	0.001	1.87 (1.34, 2.79)	0.021
Q4 with Sarcopenia	10.30 (4.85, 21.8)	<0.001	2.93 (1.53, 5.61)	<0.001	2.36 (1.22, 3.81)	0.003
CMI-Sarcopenia group	Cardiovascular mortality
	Model 1		Model 2		Model 3
	95% CI	p	95% CI	p	95% CI	p
Q1 without Sarcopenia	ref		ref		ref
Q1 with Sarcopenia	1.53 (1.23, 10.14)	<0.001	1.46 (1.21, 4.05)	<0.001	1.24 (1.11, 3.29)	0.005
Q2 without Sarcopenia	5.60 (2.39, 13.10)	<0.001	4.06 (1.57, 10.5)	0.004	2.01 (1.15, 3.51)	0.027
Q2 with Sarcopenia	18.1 (4.40, 74.6)	<0.001	3.74 (2.59, 23.6)	<0.001	2.10 (1.31, 5.08)	0.015
Q3 without Sarcopenia	5.11 (1.96, 13.40)	<0.001	1.86 (0.70, 4.96)	0.215	1.20 (0.70, 2.04)	0.510
Q3 with Sarcopenia	14.1 (4.11, 48.70)	<0.001	4.68 (1.54, 14.20)	0.006	1.77 (1.15, 3.69)	0.029
Q4 without Sarcopenia	8.70 (3.41, 22.20)	<0.001	4.22 (1.55, 11.50)	0.009	2.13 (1.26, 3.60)	0.038
Q4 with Sarcopenia	8.93 (2.19, 36.40)	<0.001	4.74 (1.34, 8.37)	<0.001	2.40 (1.14, 6.13)	0.002

Model 1: no covariates were adjusted.

Model 2: Age, sex, race, PIR, education, BMI were adjusted.

Model 3: Age, sex, race, PIR, education, BMI, drinking frequency, smoking, diabetes mellitus, hypertension, medication use were adjusted.

Abbreviations: 95% CI, 95% confidence interval; HR, hazard ratio.

In the analysis of cardiovascular mortality, significant associations were observed for the intermediate-risk tiers (Q2 and Q3 with sarcopenia) (HR = 2.10, 95% CI: 1.31–5.08, p = 0.015; HR = 1.77, 95% CI: 1.15–3.69, p = 0.029). Notably, patients in the highest-risk tier (Q4 with sarcopenia) demonstrated a further increased risk of cardiovascular mortality (HR = 2.40, 95% CI: 1.14–6.13, p = 0.002). These findings indicate that the metabolic-muscle composite risk model effectively identifies patient groups with differing mortality risks, and the composite risk state characterized by the coexistence of high CMI and sarcopenia is associated with the worst clinical prognosis.

In Figure 3, we used an RCS regression model to explore the association of CMI with all-cause and cardiovascular mortality, stratified by sarcopenia status. After adjustment for all covariates, a significant nonlinear relationship was observed between CMI and both outcomes (P for nonlinearity <0.001). For all-cause mortality (Figure 3A), the risk increased sharply with higher CMI, and the excess risk was more pronounced among individuals with sarcopenia, who exhibited consistently higher hazard ratios across the entire CMI range compared with those without sarcopenia. For cardiovascular mortality (Figure 3B), a similar pattern was observed: mortality risk rose with increasing CMI, and patients with sarcopenia showed a markedly higher risk, particularly at lower-to-moderate CMI levels. These findings highlight the synergistic impact of sarcopenia and elevated CMI on mortality risk in CKD patients.

Figure 3.

Restricted cubic spline (RCS) models showing nonlinear associations between cardiometabolic index (CMI), sarcopenia status, and adjusted hazard ratios for all-cause mortality (A) and cardiovascular mortality (B) in chronic kidney disease (CKD) patients. Models are adjusted for demographic, lifestyle, and clinical covariates.

To further assess the robustness of the observations, we performed a subgroup analysis. Tables S2–S5 present the associations of CMI levels and sarcopenia with all-cause and cardiovascular mortality across all subgroups. The positive association between CMI and all-cause mortality was generally consistent across most subgroups of CKD patients, with stronger effects observed among participants aged ≥ 45 years, females, non-Hispanic Whites, those with higher education, smokers, heavy drinkers, and individuals with diabetes or hypertension (all P for interaction <0.05). Notably, the risk of all-cause mortality increased markedly with CMI in patients with stage III CKD, while no significant associations were observed in early (stage I-II) or advanced (stage IV-V) stages. In addition, the coexistence of sarcopenia further amplified the impact of CMI, with significantly higher risks observed in sarcopenic patients compared with their non-sarcopenic counterparts. For cardiovascular mortality, similar patterns were found. The associations between CMI and cardiovascular death were particularly evident among participants ≥ 45 years, females, non-Hispanic Whites, non-Hispanic Blacks, those with higher education, smokers, heavy drinkers, and individuals with diabetes or hypertension (all p for trend < 0.05). Sarcopenia was independently associated with both all-cause and cardiovascular mortality, and its effect was more pronounced in participants with higher CMI levels. Similarly, the associations between sarcopenia and risks of all-cause and cardiovascular mortality were most pronounced in patients with stage III CKD.

3.3. The predictive ability of CMI, sarcopenia, and traditional indicators for all‑cause and cardiovascular mortality in patients with CKD

To holistically assess the metabolic and muscular health of patients with CKD, we developed a novel metabolic-muscular composite risk model through the integration of the CMI and sarcopenia. As shown in Figure 4, this composite model, along with its individual components (CMI alone and sarcopenia alone), demonstrated predictive ability for all-cause and cardiovascular mortality at 3-, 5-, and 10-year time points.

Figure 4.

Time-dependent receiver operating characteristic (ROC) curves evaluating predictive performance of cardiometabolic index (CMI), sarcopenia, and their combined model for all-cause mortality (A, C, E) and cardiovascular mortality (B, D, F) at 3-, 5-, and 10-year follow-up in chronic kidney disease (CKD) patients.

For all-cause mortality, the CMI model showed favorable discriminative ability, with area under the curve (AUC) values of 0.801, 0.805, and 0.829 at 3, 5, and 10 years, respectively (Figure 4A). The sarcopenia model also exhibited good predictive performance, with AUCs of 0.853, 0.818, and 0.803 (Figure 4C). Notably, the metabolic-muscular composite risk model further enhanced discriminative power, achieving AUCs of 0.873, 0.877, and 0.886 at the corresponding time points (Figure 4E), which indicates its superior overall predictive capability. For cardiovascular mortality, all three models demonstrated considerable predictive value. The CMI model yielded AUCs of 0.840, 0.800, and 0.816 at 3, 5, and 10 years (Figure 4B), while the sarcopenia model produced AUCs of 0.889, 0.824, and 0.807 (Figure 4D). Importantly, the metabolic-muscular composite model displayed the highest predictive accuracy, with AUCs reaching 0.895, 0.858, and 0.855 (Figure 4F), which further validates its advantage in assessing cardiovascular mortality risk.

To better contextualize the performance of our novel model, we further compared its predictive ability with conventional metabolic risk indicators such as BMI, waist circumference, and the TG/HDL-C ratio. As illustrated in Figure S2, the traditional indicators showed moderate predictive ability for all-cause mortality, with AUCs ranging between 0.60 and 0.85 across the three time points. In contrast, the metabolic-muscular composite model consistently maintained AUCs above 0.85, reflecting superior discriminative capacity. A consistent trend was observed for cardiovascular mortality (Figure S3), with the composite model again outperforming traditional indicators (all AUCs >0.85 vs. generally <0.85 for conventional measures). Moreover, when compared directly with CMI alone, the composite model showed improved AUCs across all time points for both all-cause and cardiovascular mortality, underscoring the additive predictive value gained by integrating sarcopenia.

These findings suggest that the metabolic-muscular composite risk model, by integrating multidimensional clinical information, provides a more accurate prediction of all-cause and cardiovascular mortality risk in CKD patients than traditional anthropometric and lipid-based indicators.

4. Discussion

Based on a nationally representative sample of U.S. adults with CKD, this study found an independent association between the CMI and the risk of all-cause mortality in CKD patients. Notably, the risk of mortality was further exacerbated when elevated CMI coexisted with sarcopenia, suggesting that sarcopenia mediates metabolic risk in the progression of CKD. More importantly, the combined assessment of CMI and sarcopenia can effectively identify CKD patients with high-risk clinical phenotypes. Given the potential synergistic effect between metabolic disorders and impaired muscle health, this study constructed an integrated “metabolic-muscle composite risk model,” treating CMI and sarcopenia as an interdependent and co-acting overall risk phenotype. These findings emphasize the important prognostic value of jointly assessing metabolic indicators and muscle status in the risk stratification of CKD patients, providing a simple and effective clinical tool for the early identification of high-risk individuals and the development of targeted management strategies.

4.1. CMI And sarcopenia are independently and synergistically associated with an increased risk of mortality in patients with CKD

In recent years, multiple studies have focused on the relationship between metabolism-related indicators and clinical outcomes in CKD patients. For example, elevated serum TG levels are significantly associated with a decline in eGFR, while higher levels of HDL-C are positively correlated with better renal function [27–29]. Furthermore, the TG/HDL-C ratio, a lipid metabolism-related indicator, has been confirmed as an independent determinant of CKD risk [30]. The plausible mechanisms involve lipid deposition in renal tissue triggering local inflammatory responses, oxidative stress, and dysregulated autophagy, thereby accelerating glomerulosclerosis and tubulointerstitial damage [31]. Concurrently, central obesity can induce hemodynamic abnormalities such as glomerular hyperfiltration, further aggravating renal injury [28,32,33]. As an emerging metabolic risk assessment tool, CMI has been proven to be closely associated with various metabolic abnormalities. Its association with adverse clinical outcomes provides a new perspective for the prevention and treatment of metabolic diseases. Yang et al. indicated a significant correlation between CMI and albuminuria in hypertensive patients, with CMI performing better than other obesity-related indicators in diagnosing albuminuria [34]; Miao et al. reported that CMI is an independent risk factor for microalbuminuria, particularly showing a stronger association in diabetic patients [35]. Tang et al. further confirmed that elevated baseline CMI in the elderly is positively associated with both the risk of CKD development and rapid renal function decline, suggesting its potential value in predicting renal injury [36]. Xu’s study found that elevated CMI levels in the elderly are associated with an increased risk of all-cause and cardiovascular mortality [37]. Although the aforementioned studies primarily focused on the correlation between CMI and renal function indicators or mortality risk, this study further clarifies that CMI is independently associated with increased mortality risk in CKD patients. However, after adjusting for multiple confounders, the mortality risk was significantly higher in the Q2 and Q4 groups compared to the Q1 group, while the risk in the Q3 group did not reach statistical significance. This result may stem from a non-linear relationship between them, where the CMI level corresponding to the Q3 group might fall precisely within a “plateau” of risk. Additionally, unmeasured effect modifiers or the relative insufficiency of statistical power within this subgroup could also contribute to the non-significant result in the Q3 group.

This study also found that sarcopenia is independently associated with increased mortality risk in CKD patients. Existing literature supports this conclusion; for instance, lower derived paraspinal muscle density is independently associated with an increased risk of all-cause mortality in hemodialysis patients [38]. A meta-analysis further indicated that sarcopenia is highly prevalent in end-stage renal disease patients requiring dialysis and is an important predictor of cardiovascular events and mortality [39]. The underlying mechanisms involve multiple factors: CKD patients often experience malnutrition, insufficient energy reserves, chronic inflammation, and metabolic disorders, which collectively promote skeletal muscle protein breakdown, inhibit synthesis, and accelerate muscle loss [40,41]. The uremic environment-specific persistent low-grade inflammation (e.g., elevated interleukin-6 (IL-6), tumor necrosis factor-alpha), metabolic acidosis, and oxidative stress also exacerbate muscle cell damage and atrophy. Functionally, reduced muscle mass leads to decreased mobility and limited physical activity, thereby increasing the risk of falls, infections, and complications, while also worsening insulin resistance and energy metabolism disorders [42]. Furthermore, sarcopenia promotes endothelial dysfunction and atherosclerosis through inflammatory and metabolic pathways, further increasing cardiovascular risk [43,44]. These findings are consistent with the results obtained in this study, suggesting that early identification and interventions aimed at maintaining or increasing muscle mass may help reduce mortality risk in CKD patients. This study further extends existing knowledge by finding that the “metabolic-muscle composite risk model,” integrating CMI and sarcopenia, can identify high-risk phenotypes among CKD patients, indicating a potential synergistic amplification of mortality risk.

4.2. Pathophysiologic interplay between metabolic dysregulation and muscle loss

As one of the largest metabolic organs in the human body, skeletal muscle plays a key role in maintaining systemic metabolic homeostasis. Recent research increasingly highlights the role of sarcopenia in the development and progression of metabolic disorders; its association with abdominal obesity, hypertriglyceridemia, and an increased risk of metabolic syndrome has been confirmed [45]. Plasma metabolomic and lipidomic analyses show significant differences in metabolite profiles between sarcopenic patients and healthy controls, with some metabolites correlating with muscle mass indicators [46], suggesting a close link between sarcopenia and metabolic disorders. The results of this study further indicate that in the progression of CKD, the metabolic abnormalities represented by CMI and sarcopenia may have a synergistic effect, jointly exacerbating patient mortality risk. The mechanism may stem from the reciprocal interaction between metabolic dysregulation and muscle wasting: reduced skeletal muscle mass worsens metabolic dysfunction by decreasing glucose uptake capacity and altering myokine secretion profiles, while insulin resistance, chronic inflammation, and lipotoxicity further promote the progression of sarcopenia via protein degradation pathways [47,48]. Specifically, skeletal muscle plays a central role in glucose homeostasis through mechanisms such as glucose uptake, regulation of muscle-derived cytokines (e.g., feimin [49], IL-6, interleukin-8, interleukin-15, fibroblast growth factor 21, irisin, myostatin [50], among others), maintenance of muscle mass, and exercise adaptation. Loss of muscle mass or myokine dysfunction may lead to dysregulated glucose and lipid metabolism [49,50]. Additionally, glycerophosphocholine phosphodiesterase 1 (Gpcpd1), an enzyme abundant in muscle, hydrolyzes glycerophosphocholine (GPC), and Gpcpd1 deficiency leads to GPC accumulation and impaired insulin signaling, severely disrupting glucose metabolism [51]. Normal metabolic status is also crucial for maintaining muscle health: lipid metabolism disorders often accompany chronic low-grade inflammation [52], which can activate p38 MAPK and NF-κB pathways, upregulate muscle-specific E3 ubiquitin ligases, and synergistically activate protein degradation pathways like the ubiquitin-proteasome system and autophagy, leading to excessive protein breakdown [53,54]. Furthermore, ectopic deposition of excess lipids (e.g., ceramides, diacylglycerols) in muscle cells directly interferes with insulin signaling, blocks the PI3K/Akt pathway, and hinders muscle protein synthesis [53,55,56]. Other studies have found that deficiency in phosphatidylethanolamine cytidyltransferase, a key enzyme for synthesizing phosphatidylethanolamine (PE), reduces PE synthesis, impairing the stability of the muscle fiber membrane lipid bilayer and mitochondrial function, leading to progressive muscle weakness and atrophy [57]. These metabolic imbalances collectively disrupt muscle protein homeostasis, promoting the onset and progression of sarcopenia.

4.3. The modifying effect of CKD stage on the risk association: a focus on stage 3 patients

Subgroup analysis showed that the impact of CMI and sarcopenia on all-cause and cardiovascular mortality was most significant in stage 3 CKD patients, while the association was weaker or non-significant in early-stage (1–2) or late-stage (4–5) patients. The stronger association in stage 3 CKD may be because this stage often marks the transition from the compensatory phase to the decompensated phase of the disease. Previous studies have shown that patients with moderate CKD are more likely to die from cardiovascular disease than to progress to end-stage renal disease [58]; the cardiovascular disease mortality rate is highest in stage 3 CKD patients compared to stages 2, 4, and 5 [59], suggesting that stage 3 may be a prognostic turning point [60,61]. In early stages (1–2), the effects of metabolic abnormalities and sarcopenia might be partially masked by compensatory mechanisms, whereas in late stages (4–5), severe complications such as anemia, electrolyte imbalances, heart failure, and infections dominate the mortality risk, diminishing the marginal effects of CMI and sarcopenia. Stage 3 patients have significantly reduced renal function but have not reached end-stage failure, making the hazards of metabolic disorders and muscle loss more prominent; the high prevalence of cardiovascular comorbidities at this stage may further amplify their risk. Data from the China Kidney Disease Network indicate that up to 71.6% of stage 3 CKD patients were undiagnosed at baseline, consistent with undiagnosed rates of 56.0%–95.5% in international cohorts [62]. These findings suggest that strengthening the monitoring of CMI and sarcopenia in stage 3 CKD patients is crucial for improving prognosis.

4.4. Clinical implications and translational potential

The results of this study have important clinical implications. The coexistence of elevated CMI and sarcopenia can significantly amplify the mortality risk in CKD patients, suggesting that clinical management should simultaneously focus on metabolic health and muscle status. In CKD diagnosis and treatment, combining CMI screening with muscle mass assessment holds promise as a feasible risk stratification tool. Specifically, routine monitoring of CMI and sarcopenia may facilitate early risk stratification, and targeted metabolic and muscle-preserving interventions should be prioritized, especially in stage 3 CKD. Such a dual-focus strategy helps in the early identification of high-risk patients and provides a basis for developing targeted interventions to delay disease progression and reduce mortality.

4.5. Strengths and limitations

The main strengths of this study include the use of a nationally representative sample and the adjustment for multiple covariates, including socioeconomic variables like the PIR and education level, enhancing the generalizability of the results and providing a reference for future public health strategies aimed at improving muscle health and cardiometabolic indicators in CKD patients. Nevertheless, several limitations must be acknowledged: First, the NHANES database only partially includes grip strength data, limiting the comprehensive assessment of sarcopenia and potentially affecting the generalizability of the results. Second, the covariate adjustment did not include laboratory indicators such as albumin, C-reactive protein, cholesterol subfractions, uric acid, and IL-6, leaving room for potential residual confounding. Third, the analytical sample size was relatively small due to the limited number of participants with muscle function assessments and complete CMI calculation variables, which may affect statistical power and result robustness. Fourth, the observational design inherently limits causal inference, and reverse causality cannot be fully excluded, as worsening CKD may drive sarcopenia, which then exacerbates outcomes. Future prospective cohort studies are needed to further validate these associations.

5. Conclusion

In summary, this study confirms that elevated CMI and sarcopenia are independent predictors of all-cause and cardiovascular mortality in CKD patients, and their coexistence synergistically significantly increases mortality risk, highlighting the important prognostic impact of the synergistic effect between metabolic disorders and muscle loss. Incorporating CMI and sarcopenia into routine risk assessment helps identify high-risk CKD patients early and provides a basis for targeted interventions. Future prospective studies and intervention trials are needed to validate these findings and to further explore whether strategies improving metabolic control and maintaining muscle health can reduce mortality risk and improve long-term outcomes in this population.

Funding Statement

Construction of Demonstration Pilot Project of Traditional Chinese Medicine Inheritance and Innovation Development in Pudong New Area, Shanghai (Construction of High-level Research Hospital of Traditional Chinese Medicine No. YC-2023-0901); Famous Traditional Chinese Medicine Practitioners of Pudong New Area (PDZY-2025-0703); Plucked Project of the Oriental Talent Program in 2023; Tertiary Management of Renal Disease in Shanghai Three-Year Action Plan (1-2-1); Shanghai Science and Technology Commission Science and Technology Innovation Action Plan (23Y11921400); National Natural Science Foundation of China (2022–82274451).

Disclosure statement

The authors declare no conflicts of interest.

Ethics statement

This research followed the STROBE reporting guideline. The study involving human participants was reviewed and approved by the NCHS Research Ethics Review Board (ERB). The approval number for this study is not applicable as the analysis was conducted on de-identified, publicly available data from the National Health and Nutrition Examination Survey (NHANES).

Data availability statement

The data that support the findings of this study were derived from the following resource available in the public domain: National Health and Nutrition Examination Survey (NHANES), available at: https://www.cdc.gov/nchs/nhanes/index.html.