Association and prognostic value of adiposity distribution indices for all-cause mortality in cardiovascular-kidney-metabolic syndrome stages 0

Abstract

Background:

Cardiovascular-kidney-metabolic syndrome (CKM) is a major public health challenge. While obesity is a central driver of CKM, body mass index (BMI) fails to adequately reflect body fat distribution and its associated prognostic heterogeneity.

Objectives:

To evaluate the prognostic value of emerging anthropometric indices reflecting fat distribution in predicting all-cause mortality risk among individuals in the early and intermediate stages of CKM (stages 0–3).

Design:

A population-based retrospective cohort study utilizing data from the National Health and Nutrition Examination Survey (NHANES) 1999–2006, with mortality linkage follow-up through 2019.

Methods:

We analyzed 5907 adults with CKM stages 0–3. Eleven adiposity indices were evaluated using four machine learning algorithms (LASSO-Cox, RSF-Boruta, RSF, survival XGBoost) to identify key mortality predictors. Statistical analyses incorporated multivariable Cox proportional hazards regression, restricted cubic splines, false discovery rate (FDR) correction for multiple testing, and a sensitivity analysis across the entire CKM continuum (stages 0–4).

Results:

Four indices reflecting peripheral-to-central adiposity predominance (TCWR, CWR, AWR, SWR) were consistently identified as key predictors. Higher levels independently correlated with lower all-cause mortality (all FDR-adjusted p < 0.05). TCWR and CWR showed linear inverse trends, whereas AWR and SWR exhibited nonlinear threshold effects. These associations persisted across BMI subgroups and remained robust in the full CKM continuum subanalysis. The combined index model yielded improved discriminative ability (AUC = 0.769) compared to single indices (TCWR AUC = 0.731) and conventional BMI-based models, providing higher net clinical benefit.

Conclusion:

Peripheral-to-central adiposity indices are strong predictors of mortality in CKM stages 0–3, providing prognostic information beyond BMI and serving as practical tools for early risk stratification. Future studies should evaluate their dynamic changes and validate their applicability across diverse populations.

Keywords

body fat distribution cardiovascular-kidney-metabolic syndrome mortality NHANES peripheral adiposity

Introduction

Cardiovascular-kidney-metabolic (CKM) syndrome is a major public health challenge.¹ CKM staging reflects the cumulative burden of multisystem comorbidities, providing a framework for risk prediction.² As cardiometabolic risks and mortality elevate even in early clinical stages (0–3),^3,4 accurate early risk stratification remains an urgent clinical requirement.

Obesity is considered a primary driver of CKM progression.⁵ However, traditional metrics like body mass index (BMI) fail to capture adipose tissue heterogeneity. Cardiometabolic pathophysiology relies primarily on anatomic fat distribution rather than absolute mass.⁶ Visceral adiposity promotes systemic inflammation and lipid dysregulation,⁷ secreting pro-inflammatory cytokines (e.g., TNF-α, IL-6) that drive insulin resistance and endothelial dysfunction. Conversely, peripheral subcutaneous fat acts as a protective metabolic sink, sequestering lipids and secreting beneficial adipokines.⁸ This functional divergence, modulated by environmental modifiers (e.g., diet, pollutants) and molecular mechanisms (e.g., epigenetics, adiponectin signaling),^9,10 synergistically accelerates vascular aging and myocardial remodeling. Consequently, adverse fat distribution exacerbates disease progression and mortality across the CKM continuum. While emerging anthropometric indices better distinguish central from peripheral adiposity and predict clinical outcomes,^11
–13 their prognostic value in CKM stages 0–3 remains unexplored.

Therefore, we analyzed National Health and Nutrition Examination Survey (NHANES) 1999–2006 data to evaluate the associations of 11 fat distribution indices with all-cause mortality in individuals with CKM stages 0–3. This study aims to clarify the prognostic value of specific fat distribution patterns and provide epidemiological evidence for individualized risk assessment tools beyond BMI.

Methods

Data source

Since the critical anthropometric measurements required for calculating the core fat distribution indices were discontinued after the 2005–2006 cycle, data for this study were derived from four consecutive NHANES cycles (1999–2006). NHANES, conducted by the National Center for Health Statistics (NCHS), uses a multistage, stratified probability sampling design to collect demographic, lifestyle, examination, and laboratory data. All participants provided written informed consent, and survey protocols were approved by the NCHS Ethics Review Board in accordance with the Declaration of Helsinki. This study was reported in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.¹⁴

Study population

Inclusion criteria were: (1) age ⩾20 years; and (2) CKM stage 0–3 according to the 2023 American Heart Association (AHA) framework.¹⁵ Detailed clinical definitions and the stepwise classification algorithm for CKM stages are explicitly provided in Tables S1 and S2. Exclusion criteria were: (1) missing anthropometric measurements for body fat distribution; (2) lack of mortality follow-up data; and (3) missing key covariates. CKM stage 4 participants (characterized by established cardiovascular or end-stage kidney disease) were strictly excluded to prevent substantial confounding by disease severity, thereby clarifying prognostic associations specifically during preclinical and subclinical windows.

A final cohort of 5907 participants was established (Figure 1). Baseline characteristics were highly comparable between the included cohort and participants excluded solely due to missing data (n = 2017; Table S3), confirming that the complete-case analysis did not introduce substantial selection bias.

Figure 1.

Study flow diagram of participant inclusion and exclusion.

Definition of body fat distribution indices

Anthropometric parameters were measured in NHANES Mobile Examination Centers following standardized protocols. Based on previous literature,¹⁶ we calculated 11 adiposity indices conceptually classified into peripheral-to-central, global, and limb adiposity indicators, alongside conventional BMI. The specific definitions and mathematical calculation formulas for all evaluated indices are consolidated in Table S4.

Data collection

Demographic variables (age, gender, race, marriage, education, and poverty income ratio) and lifestyle factors (smoking and drinking) were collected at baseline. Comorbidities included cancer (self-reported), hypertension (blood pressure ⩾140/90 mmHg, self-reported diagnosis, or antihypertensive medication use), and diabetes (fasting glucose ⩾7.0 mmol/L, HbA1c ⩾6.5%, self-reported diagnosis, or glucose-lowering medication use).

Study outcome

The primary outcome was all-cause mortality, ascertained through linkage to the National Death Index (NDI). Follow-up time was calculated from the baseline survey date to death, loss to follow-up, or December 31, 2019.

Statistical analysis

Baseline characteristics and inter-index correlations were evaluated using Wilcoxon rank-sum tests, chi-square tests, and Spearman correlation analyses. Four machine learning algorithms (LASSO-Cox, RSF-Boruta, RSF, and survival XGBoost) identified key predictors, defined as consistently top-ranked indices across all methods. Model stability was validated via hyperparameter sensitivity analyses (Table S5). Survival differences were assessed using Kaplan–Meier curves and log-rank tests. Multivariable Cox proportional hazards models estimated hazard ratios (HRs) for indices as continuous (per 1-SD increment) and categorical (quartile) variables. A sensitivity subanalysis evaluated the entire CKM continuum (stages 0–4). The proportional hazards assumption and multicollinearity were verified using scaled Schoenfeld residuals (Figure S1) and variance inflation factors (VIF < 5.0; Table S6), respectively. Restricted cubic splines (four knots) explored nonlinear associations, and stratified subgroup analyses evaluated potential effect modifications. Predictive performance was assessed via time-dependent receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). Incremental predictive value beyond conventional risk factors and BMI was quantified using the C-statistic, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI). All analyses incorporated NHANES complex survey design elements using R software (version 4.3.2, R foundation for statistical computing, Vienna, Austria). The Benjamini–Hochberg procedure was applied to calculate false discovery rate (FDR)-adjusted p-values to control for multiple testing. Statistical significance was defined as a two-sided p < 0.05.

Results

Baseline characteristics and index correlations

A total of 5907 participants (1207 deaths) were followed for a median of 195 months. Non-survivors were significantly older, predominantly male, and had a higher prevalence of CKM stage 3, comorbidities, and adverse lifestyle factors (all p < 0.001; Table S7). Regarding fat distribution, non-survivors exhibited significantly lower SWR, AWR, CWR, TSFWR, TCWR, and WWR, but higher WHtR (all p < 0.001; Table 1). BMI showed no significant inter-group differences. Strong inter-index correlations were observed among indices from similar anatomical sites, while SWR exhibited weak correlations (|r| < 0.2; Figure S2).

Table 1.

Baseline characteristics of participants stratified by survival status.

Variables	Total	Survivor	Non-survivor	p
Unweighted number	5907	4700	1207
Weighted number	34,383,182	28,428,973	5,954,209
BMI	26.03 (23.21, 29.16)	25.98 (23.19, 29.09)	26.29 (23.36, 29.48)	0.066
SWR	0.20 (0.16, 0.25)	0.21 (0.16, 0.26)	0.18 (0.14, 0.24)	<0.001
AWR	0.34 (0.32, 0.36)	0.34 (0.33, 0.37)	0.32 (0.30, 0.34)	<0.001
CWR	0.41 (0.38, 0.44)	0.41 (0.38, 0.44)	0.38 (0.35, 0.41)	<0.001
TSFWR	0.18 (0.12, 0.25)	0.19 (0.13, 0.26)	0.16 (0.12, 0.22)	<0.001
TCWR	0.56 (0.51, 0.61)	0.57 (0.53, 0.61)	0.51 (0.47, 0.55)	<0.001
WWR	0.80 (0.74, 0.87)	0.81 (0.75, 0.88)	0.76 (0.70, 0.82)	<0.001
WHtR	0.55 (0.50, 0.60)	0.54 (0.49, 0.60)	0.58 (0.53, 0.63)	<0.001
STR	1.12 (0.86, 1.49)	1.12 (0.87, 1.48)	1.13 (0.85, 1.52)	0.884
TSFCCR	0.46 (0.31, 0.62)	0.46 (0.31, 0.63)	0.44 (0.31, 0.60)	0.052
TSFTCR	0.33 (0.23, 0.45)	0.33 (0.23, 0.46)	0.33 (0.23, 0.44)	0.557

AWR, arm circumference-to-waist circumference ratio; BMI, body mass index; CKM, cardiovascular-kidney-metabolic syndrome; CWR, maximal calf circumference-to-waist circumference ratio; STR, subscapular-to-triceps skinfold thickness ratio; SWR, subscapular skinfold-to-waist circumference ratio; TCWR, thigh circumference-to-waist circumference ratio; TSFCCR, triceps skinfold-to-maximal calf circumference ratio; TSFTCR, triceps skinfold-to-thigh circumference ratio; TSFWR, triceps skinfold-to-waist circumference ratio; WHtR, waist circumference-to-height ratio; WWR, weight-to-waist circumference ratio.

Feature selection and algorithmic consistency

Across four complementary machine learning approaches (LASSO-Cox, RSF-Boruta, standard RSF, and survival XGBoost), TCWR, CWR, AWR, and SWR consistently emerged as the top-ranked and overlapping consensus features for mortality risk prediction (Figures 2 and S3). Consequently, these four peripheral-to-central adiposity indices were selected for subsequent analyses.

Figure 2.

Feature selection results from RSF and survival XGBoost models, and consistency analysis across methods. (a) Top variables ranked by RSF; (b) Top variables ranked by survival XGBoost; (c) Overlapping key features consistently identified by LASSO-Cox, RSF-Boruta, RSF, and survival XGBoost.

Survival analysis and multivariable Cox proportional hazards

Higher levels of TCWR, CWR, AWR, and SWR were significantly associated with lower all-cause mortality rates (log-rank p < 0.001; Figure 3). In the fully adjusted Cox regression (Model 3), participants in the highest quartiles (Q4) of TCWR, CWR, AWR, and SWR exhibited significantly lower risks of all-cause mortality compared with those in the lowest quartiles (Table S8). Modeled continuously (per 1-SD increment), each index consistently conferred an independent lower mortality risk. All associations maintained statistical significance after FDR correction (FDR-adjusted p < 0.05). In the sensitivity subanalysis incorporating CKM stage 4, these indices maintained significant, albeit slightly attenuated, protective associations (FDR-adjusted p < 0.001; Table S9).

Figure 3.

Kaplan–Meier survival curves for all-cause mortality stratified by quartiles of body fat distribution indices. (a) TCWR, (b) CWR, (c) AWR, and (d) SWR.

Nonlinear associations and subgroup stability

Restricted cubic splines (RCS) demonstrated significant overall associations for all four indices with mortality (Figure 4). TCWR and CWR exhibited linear inverse relationships (p-overall <0.001, p-nonlinear >0.05). In contrast, AWR and SWR demonstrated nonlinear threshold effects (p-nonlinear <0.001), where mortality risk declined steeply at lower ranges but plateaued at moderate-to-high levels, further underscoring their value in prognostic risk stratification.

Figure 4.

RCS analyses of the associations between body fat distribution indicators and all-cause mortality. (a) TCWR, (b) CWR, (c) AWR, and (d) SWR. Subgroup analyses: consistent protective effects across different populations.

Stratified analyses confirmed that the protective effects of these four indices were robust across age, sex, lifestyle, and comorbidity subgroups (all p for interaction > 0.05; Tables 2 and S10), maintaining significance after FDR correction. Notably, higher TCWR maintained a significant inverse association with mortality even among individuals with a normal BMI (<25 kg/m²; FDR-adjusted p < 0.05). The absence of significant interactions across BMI strata demonstrates that peripheral-to-central fat distribution provides prognostic value beyond overall adiposity.

Table 2.

Subgroup analyses of the association between TCWR and all-cause mortality.

Variables	Events/no.	HR (95% CI per 1 SD)	p	FDR-adjusted p	p for interaction
All participants	1207/5907	0.81 (0.75–0.88)	<0.001	<0.001
Age					0.071
<65	422/4705	0.56 (0.49–0.64)	<0.001	<0.001
⩾65	785/1202	0.71 (0.64–0.79)	<0.001	<0.001
Gender					0.242
Male	658/2929	0.77 (0.66–0.89)	<0.001	<0.001
Female	549/2978	0.80 (0.72–0.89)	<0.001	<0.001
PIR					0.814
<1.3	322/1414	0.73 (0.62–0.86)	<0.001	<0.001
1.3–3.5	602/2483	0.79 (0.70–0.89)	<0.001	<0.001
⩾3.5	283/2010	0.90 (0.75–1.08)	0.258	0.264
BMI					0.728
< 25	466/2425	0.84 (0.74–0.95)	0.007	0.009
25–30	478/2325	0.79 (0.69–0.91)	0.001	0.002
⩾30	263/1157	0.73 (0.62–0.87)	<0.001	<0.001
Smoking					0.104
Yes	698/2818	0.77 (0.69–0.86)	<0.001	<0.001
No	509/3089	0.81 (0.72–0.92)	<0.001	0.001
Drinking					0.665
Yes	763/3991	0.80 (0.71–0.89)	<0.001	<0.001
No	444/1916	0.79 (0.69–0.89)	<0.001	<0.001
Hypertension					0.993
Yes	508/1403	0.77 (0.68–0.88)	<0.001	<0.001
No	699/4504	0.80 (0.71–0.89)	<0.001	<0.001
Diabetes					0.628
Yes	231/480	0.78 (0.63–0.96)	0.017	0.019
No	976/5427	0.79 (0.72–0.87)	<0.001	<0.001
Cancer					0.097
Yes	237/451	0.92 (0.74–1.15)	0.485	0.485
No	970/5456	0.77 (0.71–0.85)	<0.001	<0.001

FDR-adjusted p-values were calculated using the Benjamini–Hochberg procedure to correct for multiple testing across the primary models.

BMI, body mass index; FDR, false discovery rate; SD, standard deviation; TCWR, thigh circumference-to-waist circumference ratio.

Predictive performance and incremental value

Among single indices, TCWR exhibited the highest discriminative ability (AUC = 0.731). Integration of the four key indices outperformed any single indicator (overall AUC = 0.769), maintaining stable time-dependent AUCs of 0.759, 0.781, and 0.785 at 3, 5, and 10 years, respectively (Figure 5). Compared with a conventional clinical model including established risk factors and BMI (C-statistic = 0.772), adding the four novel indices significantly improved discrimination (C-statistic = 0.884, p < 0.001) (Table S11). This enhanced model also provided significant improvements in risk reclassification (continuous NRI = 0.976; IDI = 0.226, both p < 0.001) and demonstrated superior calibration and net clinical benefit.

Figure 5.

Predictive performance of the selected body fat distribution indices. (a) ROC curves of TCWR, CWR, AWR, and SWR for all-cause mortality prediction. (b) Time-dependent ROC curves of individual indices during follow-up. (c) ROC curves comparing the joint model with individual indices. (d) Time-dependent ROC curves of the joint model at 3, 5, and 10 years. (e) Calibration curve of the joint model. (f) DCA of the joint model versus individual indices.

Discussion

This study evaluated the associations of 11 adiposity distribution indices with all-cause mortality in individuals at CKM stages 0–3. Using four machine learning approaches, we identified TCWR, CWR, AWR, and SWR as consistent protective predictors. Restricted cubic spline analysis revealed linear inverse associations for TCWR and CWR, and nonlinear threshold effects for AWR and SWR. Subgroup and predictive analyses confirmed their independent prognostic value in early CKM.

While fat distribution reflects metabolic risk better than overall obesity, the prognostic value of specific indices in early CKM was previously undefined.^17,18 Our findings indicate that indices reflecting a predominance of peripheral over central adiposity (TCWR, CWR, AWR, SWR) are highly informative for long-term outcomes. Anthropometrically, peripheral circumferences proxy both subcutaneous adipose tissue (SAT) and skeletal muscle mass. Biologically, these tissues exert protective metabolic effects by acting as efficient “glucose sinks” and secreting beneficial adipokines, improving glucose homeostasis.^19,20 Conversely, waist circumference represents visceral adipose tissue, which secretes pro-inflammatory cytokines that drive chronic inflammation and endothelial dysfunction.^21,22 Therefore, peripheral-to-central indices capture the net physiological balance between protective peripheral metabolic reserves and detrimental central visceral adiposity. The observed inverse associations reflect the survival benefit of maintaining this favorable body composition profile in early CKM. Regarding predictive performance, TCWR exhibited the highest discriminative ability among single indices, followed by CWR and AWR. The relatively lower contribution of SWR may stem from measurement variability inherent to skinfold thickness assessments.^23,24 Integrating all four indices improved overall predictive accuracy, time-dependent discrimination, and net clinical benefit. The application of ensemble learning in our study aligns with recent evidence confirming the utility of advanced computational approaches in predicting cardiovascular outcomes.^25,26 Combining multiple indices overcomes the limitations of single measures, facilitating robust risk identification in clinical settings.

Subgroup analyses demonstrated that these inverse associations remained significant even among normal-weight individuals (BMI <25 kg/m²), identifying high-risk patients potentially overlooked by BMI. Furthermore, these findings help clarify the cardiovascular “obesity paradox.” The apparent survival benefit associated with higher BMI is likely confounded by its inability to differentiate between protective peripheral mass and detrimental visceral adiposity. Our results suggest this paradox is primarily driven by the preservation of peripheral skeletal muscle and subcutaneous fat, rather than absolute weight. Clinically, the need for accessible risk stratification is increasingly urgent, as the global burden and cardiovascular complications of metabolic syndromes have been exacerbated following the COVID-19 pandemic.^27,28 Recent studies highlight the prognostic utility of laboratory-derived parameters like novel biomarker profiles and platelet indices.²⁹ While offering high pathophysiological specificity, these require laboratory infrastructure. Anthropometric indices provide a complementary, inexpensive, and non-invasive alternative easily incorporated into routine primary care, improving early risk identification. These results underscore that optimizing body composition through maintaining peripheral muscle and adiposity is a practical prognostic strategy.

Several limitations exist. First, complete-case analysis may introduce residual selection bias despite comparable baseline characteristics. Second, low peripheral circumferences may partly reflect unmeasured sarcopenia or aging-related frailty. Although machine learning diagnostic approaches for sarcopenia are advancing,³⁰ the lack of comprehensive baseline frailty indices in NHANES means reverse causation cannot be completely ruled out. Third, adiposity indices were assessed solely at baseline, preventing the evaluation of dynamic changes. Finally, excluding CKM stage 4 restricts the external validity of the primary findings to preclinical populations. Although sensitivity analyses across the entire CKM continuum (stages 0–4) supported the robustness of these indices, recall bias from self-reported covariates, unmeasured confounders, and restriction to a U.S. cohort limit broader generalizability.

Conclusion

In summary, indices reflecting peripheral-to-central adiposity predominance (TCWR, CWR, AWR, SWR) are independent predictors of lower all-cause mortality in CKM stages 0–3. Combining these indices further enhances predictive accuracy. These findings highlight the prognostic importance of body fat distribution, providing accessible, non-invasive tools for individualized early risk stratification. Future research should assess longitudinal changes and validate these indices across diverse populations.

Supplemental Material

sj-docx-1-tae-10.1177_20420188261449986 – Supplemental material for Association and prognostic value of adiposity distribution indices for all-cause mortality in cardiovascular-kidney-metabolic syndrome stages 0–3: a nationwide study

Supplemental material, sj-docx-1-tae-10.1177_20420188261449986 for Association and prognostic value of adiposity distribution indices for all-cause mortality in cardiovascular-kidney-metabolic syndrome stages 0–3: a nationwide study by Nan Tang, Shurun Zuo, Ya’nan Hu, Shizhong Cheng, Xuejin Chen, Haoran Li, Qingdui Zhang, Ji Hao and Chunmei Qi in Therapeutic Advances in Endocrinology and Metabolism

Supplemental Material

sj-docx-2-tae-10.1177_20420188261449986 – Supplemental material for Association and prognostic value of adiposity distribution indices for all-cause mortality in cardiovascular-kidney-metabolic syndrome stages 0–3: a nationwide study

Supplemental material, sj-docx-2-tae-10.1177_20420188261449986 for Association and prognostic value of adiposity distribution indices for all-cause mortality in cardiovascular-kidney-metabolic syndrome stages 0–3: a nationwide study by Nan Tang, Shurun Zuo, Ya’nan Hu, Shizhong Cheng, Xuejin Chen, Haoran Li, Qingdui Zhang, Ji Hao and Chunmei Qi in Therapeutic Advances in Endocrinology and Metabolism

Footnotes

Acknowledgements

Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of our research. The authors gratefully acknowledge the National Health and Nutrition.

Declarations

ORCID iD

Chunmei Qi

Supplemental material

Supplemental material for this article is available online.

References

Kittelson

Junior

Fillmore

, et al. Cardiovascular-kidney-metabolic syndrome–an integrative review. Prog Cardiovasc Dis 2024; 87: 26–36.

Xie

Cui

, et al. Global burden of the key components of cardiovascular-kidney-metabolic syndrome. J Am Soc Nephrol 2025; 36(8): 1572–1584.

Aggarwal

Ostrominski

Vaduganathan

Prevalence of cardiovascular-kidney-metabolic syndrome stages in US adults, 2011–2020. JAMA 2024; 331(21): 1858–1860.

Zhang

Zeng

, et al. Association between triglyceride-glucose related indices and all-cause and cardiovascular mortality among the population with cardiovascular-kidney-metabolic syndrome stage 0–3: a cohort study. Cardiovasc Diabetol 2025; 24(1): 92.

Khan

Coresh

Pencina

, et al. Novel prediction equations for absolute risk assessment of total cardiovascular disease incorporating cardiovascular-kidney-metabolic health: a scientific statement from the American heart association. Circulation 2023; 148(24): 1982–2004.

Chen

Liang

, et al. Association between body roundness index and advanced cardiovascular-kidney-metabolic syndrome. Front Nutr 2025; 12: 1623766.

Shen

Kong

, et al. Association between the triglyceride glucose-body mass index and future cardiovascular disease risk in a population with cardiovascular-kidney-metabolic syndrome stage 0–3: a nationwide prospective cohort study. Cardiovasc Diabetol 2024; 23(1): 292.

Ayesh

Nasser

Ferdinand

, et al. Gender-specific factors influencing obesity in women: bridging the gap between science and clinical practice. Circ Res 2025; 136(6): 594–605.

Ajayi

Ogunjobi

TT.

Environmental exposures and cancer risk: a comprehensive review. MEDIN 2025; 2(2): 80–92.

10.

Feng

Cao

Tang

, et al. Microcystins exposure associated with blood lipid profiles and dyslipidemia: a cross-sectional study in Hunan province, China. Toxins (Basel) 2023; 15(4): 293.

11.

Pan

, et al. Suppression effect of folate on poly- and perfluoroalkyl substance-induced alterations in lipids and the atherogenic index of plasma in adolescents. Lipids Health Dis 2025; 24(1): 213.

12.

Zhong

Tan

Zhu

, et al. Normal-weight central obesity and risk of cardiovascular and microvascular events in adults with prediabetes or diabetes: Chinese and British cohorts. Diabetes Metab Res Rev 2023; 39(8): e3707.

13.

Wang

Chen

Wei

, et al. Association between body fat distribution and asthma in adults: results from the cross-sectional and bidirectional Mendelian randomization study. Front Nutr 2024; 11: 1432973.

14.

von Elm

Altman

Egger

, et al. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147(8): 573–577.

15.

Ndumele

Rangaswami

Chow

, et al. Cardiovascular-kidney-metabolic health: a presidential advisory from the American heart association. Circulation 2023; 148(20): 1606–1635.

16.

Yang

Zhu

, et al. The combination of fat distribution and BMI redefines obesity: result from NHANES. J Cachexia Sarcopenia Muscle 2025; 16(4): e70013.

17.

Zheng

Michaëlsson

Fall

, et al. The metabolomic profiling of total fat and fat distribution in a multi-cohort study of women and men. Sci Rep 2023; 13(1): 11129.

18.

Xing

. Exploring the association between regional fat distribution and atrial fibrillation risks: a comprehensive cohort study. Front Endocrinol (Lausanne) 2024; 15: 1367653.

19.

Huang

Gan

Gao

, et al. Discrepancies between general and central obesity in arterial stiffness: observational studies and Mendelian randomization study. BMC Med 2024; 22(1): 325.

20.

Emamat

Jamshidi

Farhadi

, et al. The association between the visceral to subcutaneous abdominal fat ratio and the risk of cardiovascular diseases: a systematic review. BMC Public Health 2024; 24(1): 1827.

21.

Bai

Zhang

, et al. Normal weight central obesity and its impact on type 2 diabetes mellitus. Curr Diab Rep 2024; 25(1): 3.

22.

Tian

Chen

Wang

, et al. Insulin resistance mediates obesity-related risk of cardiovascular disease: a prospective cohort study. Cardiovasc Diabetol 2022; 21(1): 289.

23.

Abreo

Bailey

Abreo

. Associations between calf, thigh, and arm circumference and cardiovascular and all-cause mortality in NHANES 1999–2004. Nutr Metab Cardiovasc Dis 2021; 31(5): 1410–1415.

24.

Gomes

TLN

Ivo

JFM

Mainardi

, et al. Positive association of the thigh muscle thickness to intracellular-water ratio with handgrip strength, but not with the risk of mortality in hospitalized cancer patients: a longitudinal study. Eur J Clin Nutr 2025; 79(7): 611–615.

25.

Sahrawat

Muskan Sharma

, et al. Interactome-based computational approach to identify association between cardiomyopathy and cardiovascular disease. MEDIN 2025; 2(2): 99–106.

26.

Zaman

MAU

Adepoju

OG.

A solution approach with ensemble-based learning technology for predicting early readmission among patients with heart failure (HF) diagnosis using electronic health records (EHR). MEDIN 2025; 2(3): 145–153.

27.

GBD 2023 Disease and Injury and Risk Factor Collaborators. Burden of 375 diseases and injuries, risk-attributable burden of 88 risk factors, and healthy life expectancy in 204 countries and territories, including 660 subnational locations, 1990–2023: a systematic analysis for the global burden of disease study 2023. Lancet 2025; 406(10513): 1873–1922.

28.

Xie

Bowe

, et al. Long-term cardiovascular outcomes of COVID-19.Nat Med 2022; 28(3): 583–590.

29.

Abd Elmoniem

Thabet

Abd El-Hameed

, et al. Could higher platelet indices on admission predict re-infarction & adverse outcomes in STEMI patients undergoing primary percutaneous coronary intervention (PPCI)? A single-center experience. Eurasian J Med Oncol 2024; 8(3): 311–321.

30.

, et al. Exploration of a machine learning approach for diagnosing sarcopenia among Chinese community-dwelling older adults using sEMG-based data. J Neuroeng Rehabil 2024; 21(1): 69.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

3.58 MB

0.03 MB

0.00 MB