Age-adjusted visceral adiposity index as a predictor of chronic kidney disease: insights from NHANES 2007–2018

Abstract

The age-adjusted visceral adiposity index (AVAI) is a novel marker reflecting visceral fat-related metabolic risk. However, its relationship with chronic kidney disease (CKD) remains unclear. This study aims to evaluate the association between AVAI and CKD prevalence in U.S. adults. We conducted a cross-sectional analysis using data from the National Health and Nutrition Examination Survey (NHANES) 2007–2018, including 7,760 participants aged 20 years and older. Logistic regression models were used to examine the association between AVAI and CKD, with adjustments for demographic, lifestyle, and clinical factors. Restricted cubic spline and threshold effect analyses were applied to explore the potential nonlinear patterns, and subgroup analyses assessed the potential effect modifiers. Receiver operating characteristic (ROC) curves compared the predictive performance of AVAI and the traditional visceral adiposity index (VAI). AVAI was significantly associated with higher CKD prevalence after adjustment, with each 1-unit increase corresponding to a 47% higher risk of CKD (OR = 1.47, 95% CI: 1.36–1.59). A nonlinear association was observed, with a threshold at AVAI = −6.5, beyond which CKD risk increased steeply. Subgroup analysis showed a stronger association in older adults. ROC analysis indicated that AVAI had better discriminatory ability for CKD (AUC = 0.7581) than VAI (AUC = 0.5685). These findings suggest that AVAI is a promising, practical tool for identifying individuals at high risk of CKD in the general population.

Graphical Abstract

1. Introduction

Chronic kidney disease (CKD) includes a range of conditions involving persistent kidney structural or functional abnormalities lasting more than 3 months, typically indicated by a low glomerular filtration rate (GFR) or elevated urine protein levels [1]. The prevalence of CKD is on the rise globally, especially among individuals with underlying conditions such as diabetes, hypertension, and obesity [2]. CKD leads to gradual worsening of renal function and is closely linked with the heightened risks of cardiovascular diseases and overall mortality. Given that early-stage CKD often presents with few or no symptoms, a substantial number of patients are only diagnosed at more advanced stages, underscoring the critical need for early detection and effective management of metabolic risk factors associated with CKD [3].

Visceral fat accumulation has emerged as a key focus of research on metabolic risk factors for CKD. Visceral fat is a metabolically active adipose tissue that exhibits high levels of inflammatory activity and endocrine functions. Obesity, particularly excessive visceral fat, is closely linked to diabetes, cardiovascular diseases, and metabolic syndrome, and is considered a significant risk factor for CKD [4].

Conventional obesity indicators, such as body mass index (BMI) and waist circumference (WC), lack the ability to distinguish visceral from subcutaneous fat, limiting their accuracy in assessing visceral fat-related metabolic risk[5]. To bridge this gap, Amato et al. proposed the Visceral Adiposity Index (VAI), an integrative metric designed to indirectly evaluate metabolic disturbances linked to visceral adiposity. While VAI has shown robust performance in predicting metabolic disease risk, particularly cardiovascular diseases and type 2 diabetes [6,7], its applicability across diverse age groups and ethnicities remains limited. The index does not account for age, a key factor influencing visceral fat distribution and metabolic function [8]. Consequently, VAI may underestimate the metabolic risks associated with visceral fat in older populations, leading to inaccurate assessments [9]. To overcome these limitations, the Age-Adjusted Visceral Adiposity Index (AVAI) was proposed, integrating age into the VAI framework. AVAI offers a more comprehensive evaluation of visceral fat metabolic abnormalities and demonstrates improved predictive accuracy across different age groups and sexes [10]. Studies suggest that AVAI outperforms VAI in predicting cardiovascular diseases, metabolic syndrome, and female infertility, highlighting its broader clinical utility [11,12].

Although AVAI has demonstrated robust associations with a variety of metabolic disorders, its relationship with CKD remains inadequately investigated. Moreover, its clinical applicability in the context of CKD screening and risk stratification has not been clearly delineated. Traditional anthropometric indices such as BMI and WC are limited in their ability to reflect the pathogenic impact of visceral adiposity on renal function. In contrast, AVAI incorporates age, lipid parameters, and anthropometric measurements, and can be readily derived from routine clinical data, offering a potentially scalable and cost-effective tool for identifying individuals at the elevated risk of CKD. Notably, our study suggests a nonlinear relationship between AVAI and CKD risk, with a possible threshold around −6.5 beyond which CKD prevalence increases sharply. Accordingly, this study aimed to systematically evaluate the cross-sectional association between AVAI and CKD using data from the NHANES, and to assess the utility of AVAI in enhancing early risk stratification in clinical practice.

2. Methods

2.1. Survey description and study population

The NHANES adopts a cross-sectional design, incorporating a stratified, multistage sampling technique aimed at assessing the health and dietary habits of adults across the United States. The survey employs a comprehensive and systematic sampling procedure to ensure a diverse and representative sample, thus strengthening the accuracy and reliability of the data collected. Ethical approval was secured for all aspects of the study, and informed consent was obtained from each participant prior to inclusion. This study utilized data from the NHANES cycles conducted from 2007 to 2018, spanning six continuous survey waves. Eligibility criteria included: (1) participants aged 20 years and older; (2) those with complete chronic kidney disease data; and (3) individuals with AVAI. Participants with missing values for key variables (CKD status or AVAI) were excluded, as illustrated in the flowchart (Figure 1). For other baseline covariates, missing data were minimal (all <3%) and not imputed. These were retained as “No record” categories in descriptive tables and regression models to preserve sample size and reduce potential bias. Comparisons between included and excluded participants are summarized in Supplemental Table S1, showing no substantial differences across demographic or clinical characteristics.

Figure 1.

Included participants in the process.

2.2. Calculation of VAI and AVAI

The Visceral Adiposity Index (VAI) and the Age-Adjusted Visceral Adiposity Index (AVAI) were calculated to evaluate the visceral fat-related metabolic risk. VAI is an established surrogate marker that integrates anthropometric and lipid parameters, while AVAI is a modified version incorporating age to enhance predictive performance. VAI was calculated based on the sex-specific formulas developed by Amato et al. [5], as follows:

Female:

$$\mathit{\text{VAI}}=[\mathit{\text{WC}}/(36.58+1.89\times \mathit{\text{BMI}})]\times (\mathit{\text{TG}}/0.81)\times (1.52/\mathit{\text{HDL}})$$

To ensure scientific transparency and reproducibility, we provide the detailed multivariable linear regression formula for AVAI proposed by Kuang et al. [5,8]:

$$\mathit{\text{AVAI}}=-16.186-1.369\times \mathit{\text{HDL}}+0.38\times \mathit{\text{WC}}+0.144\times \mathit{\text{Age}}-0.013\times \mathit{\text{BMI}}-0.151\times \mathit{\text{TG}}$$

The AVAI model was originally derived through multivariable linear regression using NHANES 2013–2018 data. Candidate variables included age, BMI, WC, HDL-C, and TG. Stepwise selection procedures and collinearity diagnostics were applied to determine the final model structure. The coefficients were estimated based on a training set, with model performance evaluated using internal validation (e.g., 10-fold cross-validation). The original model yielded an R² of 0.31, and the standard errors for the coefficients were: HDL (0.12), WC (0.04), Age (0.02), BMI (0.03), TG (0.06). The model’s predictive performance was validated with an AUC of 0.738 for infertility prediction. This regression equation was directly adopted in our study to calculate AVAI for each participant.

2.3. Definition of chronic kidney disease

In this study, chronic kidney disease (CKD) was diagnosed following the established clinical guidelines [13,14]. Renal function was assessed using the CKD-EPI equation, which estimates the glomerular filtration rate (eGFR) based on serum creatinine levels [15]. eGFR is a fundamental marker of kidney function and enables early detection of renal impairment. The albumin-to-creatinine ratio (ACR), calculated as urinary albumin (mg/dL) divided by creatinine (g/dL), was used as an additional diagnostic measure. CKD was defined by an eGFR of less than 60 mL/min/1.73 m² or an ACR above 30 mg/g, in line with the established diagnostic criteria. In line with the study objective, CKD was defined according to KDIGO guidelines without further staging differentiation.

2.4. Covariates

To ensure robust adjustment in the analysis, this study incorporated a range of covariates encompassing demographic attributes and health-related behaviors, including age, sex, race, poverty income ratio (PIR), educational level, body mass index (BMI), smoking, drinking, physical activity, diabetes, and hypertension. However, dietary intake and medication use were not included due to extensive missing data and potential recall bias in the NHANES questionnaires. The PIR was divided into three groups: less than 1, between 1 and 3, and greater than or equal to 3. BMI categories were defined as normal (<25 kg/m²), overweight (25–30 kg/m²), and obese (>30 kg/m²). Smoking status was self-reported. Participants who had smoked ≥100 cigarettes during their lifetime were classified as smokers. According to NHANES criteria, participants who reported consuming more than 12 alcoholic drinks in the past year were classified as alcohol drinkers. Physical activity levels were assessed using metabolic equivalent tasks (MET) derived from the physical activity questionnaire. Participants with MET values below 600 min per week were considered physically inactive. The calculation formula was: MET (min/week) = MET value × frequency (per week) × duration (per session). Health conditions such as diabetes and hypertension were identified through physician diagnoses and participant self-reports. Smoking status was determined via questionnaire, and individuals reporting more than 100 cigarettes smoked were classified as smokers. Health status, including diabetes and hypertension, was determined by physician diagnosis and participant self-report.

2.5. Statistical analysis

This study analyzed NHANES data from 2007 to 2018 using a cross-sectional approach. Participant characteristics by CKD status were summarized with descriptive statistics: means and standard deviations for continuous data, and percentages for categorical data. Associations between VAI, AVAI, and CKD prevalence were initially assessed using univariate logistic regression, followed by the multivariable adjustments for confounders. AVAI was divided into tertiles to examine CKD risk trends and validate findings. Restricted cubic spline analysis explored nonlinear relationships between AVAI and CKD, using threshold testing to detect key inflection points. Subgroup analyses evaluated the impact of demographic and clinical factors, such as age, sex, BMI, smoking, diabetes, and hypertension, on the AVAI-CKD relationship. ROC curve analysis compared the predictive accuracy of VAI and AVAI for CKD risk. Statistical analyses were conducted using R software (version 4.2.3; R Foundation for Statistical Computing, Vienna, Austria), with significance set at p < 0.05.

3. Results

3.1. Baseline characteristics

This study enrolled 7,760 participants, with details of the selection process depicted in Figure 1. The comparison of the characteristics of the subjects who were excluded due to missing variables and those who were finally included is shown in Supplementary Table 1. Although the excluded subjects were younger, their gender and race distributions were similar, suggesting a low risk of systematic selection bias. Table 1 provides a summary of the baseline characteristics of the participants, categorized by CKD status. Among the entire cohort, 6,345 individuals did not have CKD, whereas 1,415 were identified with the condition. The analysis revealed that individuals with CKD were generally older, had a higher likelihood of being overweight, or reported a history of smoking. Additionally, these participants frequently presented with chronic conditions, such as diabetes and hypertension. Notably, those diagnosed with CKD exhibited significantly elevated VAI and AVAI scores, indicating a possible association between visceral adiposity indicators and CKD.

Table 1.

Baseline characteristics of the study population.

Characteristic	Group	Overall	Non-CKD	CKD	P value
n		7,760	6,345	1,415
Age (%)	<50	3,912 (50.4)	3,606 (56.8)	306 (21.6)	<0.001
	>50	3,848 (49.6)	2,739 (43.2)	1,109 (78.4)
Sex (%)	Female	3,937 (50.7)	3,212 (50.6)	725 (51.2)	0.698
	Male	3,823 (49.3)	3,133 (49.4)	690 (48.8)
BMI (%)	Normal	2,225 (28.7)	1,881 (29.6)	344 (24.3)	<0.001
	Overweight	2,562 (33.0)	2,132 (33.6)	430 (30.4)
	Obese	2,973 (38.3)	2,332 (36.8)	641 (45.3)
Race (%)	Mexican American	1,168 (15.1)	977 (15.4)	191 (13.5)	<0.001
	Non-Hispanic black	1,591 (20.5)	1,236 (19.5)	355 (25.1)
	Non-Hispanic white	3,026 (39.0)	2,435 (38.4)	591 (41.8)
	Others	1,975 (25.5)	1,697 (26.7)	278 (19.6)
Education level (%)	Under high school	1,850 (23.8)	1,441 (22.7)	409 (28.9)	<0.001
	High school	1,781 (23.0)	1,410 (22.2)	371 (26.2)
	Above high school	4,123 (53.1)	3,490 (55.0)	633 (44.7)
	No record	6 (0.1)	4 (0.1)	2 (0.1)
PIR (%)	<1	1,485 (21.3)	1,183 (20.8)	302 (23.7)	<0.001
	1–3	3,019 (43.4)	2,418 (42.5)	601 (47.2)
	>3	2,453 (35.3)	2,084 (36.7)	369 (29.0)
Smoke (%)	No	4,165 (53.7)	3,479 (54.8)	686 (48.5)	<0.001
	Yes	3,589 (46.2)	2,861 (45.1)	728 (51.4)
	No record	6 (0.1)	5 (0.1)	1 (0.1)
Drink (%)	No	1,615 (22.3)	1,268 (21.5)	347 (25.9)	0.002
	Yes	5,626 (77.7)	4,633 (78.5)	993 (74.1)
	No record	2 (0.0)	2 (0.0)	0 (0.0)
Activity status (%)	Active	4,166 (53.7)	3,592 (56.6)	574 (40.6)	<0.001
	Inactive	3,594 (46.3)	2,753 (43.4)	841 (59.4)
Hypertension (%)	No	4,866 (62.7)	4,378 (69.0)	488 (34.5)	<0.001
	Yes	2,886 (37.2)	1,961 (30.9)	925 (65.4)
	No record	8 (0.1)	6 (0.1)	2 (0.1)
Diabetes (%)	No	6,461 (83.3)	5,565 (87.7)	896 (63.3)	<0.001
	Yes	1,084 (14.0)	615 (9.7)	469 (33.1)
	No record	215 (2.8)	165 (2.6)	50 (3.5)
WC (mean (SD)) (cm)		99.72 (16.39)	98.77 (16.21)	103.94 (16.56)	<0.001
SCR (mean (SD)) (mg/dL)		0.87 (0.32)	0.82 (0.18)	1.11 (0.59)	<0.001
UCR (mean (SD)) (mg/dL)		127.66 (78.45)	129.88 (80.21)	117.72 (69.16)	<0.001
eGFR (mean (SD))		92.38 (22.45)	97.17 (17.84)	70.87 (27.79)	<0.001
UACR (mean (SD))		46.51 (297.32)	8.34 (5.66)	217.66 (670.14)	<0.001
HDL (mean (SD)) (mmol/L)		1.39 (0.42)	1.40 (0.41)	1.38 (0.45)	0.104
TG (mean (SD)) (mmol/L)		1.38 (1.29)	1.34 (1.24)	1.55 (1.48)	<0.001
VAI (mean (SD))		1.68 (2.40)	1.61 (2.20)	2.02 (3.10)	<0.001
AVAI (mean (SD))		−7.70 (2.68)	−8.14 (2.53)	−5.74 (2.47)	<0.001

Mean (SD) for continuous variables, % for categorical variables. AVAI: age-adjusted visceral adiposity index; BMI: body mass index; CKD: chronic kidney disease; eGFR: estimated glomerular filtration rate; PIR: poverty index ratio; SCR: serum creatinine; UACR: urinary albumin-to-creatinine ratio; UCR: urinary creatinine; VAI: visceral adiposity index.

3.2. Association between VAI, AVAI, and CKD prevalence

Univariate logistic regression analyses were conducted to assess the impact of VAI and AVAI on the prevalence of CKD, with the results summarized in Table 2. In the unadjusted analysis (Model 1), both VAI and AVAI showed a statistically significant positive correlation with CKD prevalence. After adjustment for multiple covariates (Model 3), VAI was not statistically significantly associated with CKD prevalence, and each 1-unit increase in AVAI was associated with a 47% increase in CKD prevalence, suggesting that AVAI has a higher sensitivity in identifying CKD prevalence. For further analysis, AVAI was divided into tertiles in Model 1 to explore the association across different AVAI levels. When compared to the lowest tertile, the highest AVAI tertile exhibited a notably higher prevalence of CKD (OR = 7.49, 95% CI = 5.98–9.38). This positive association persisted even after controlling for all covariates, as indicated by a significant trend (P-trend < 0.001).

Table 2.

The relationship between AVAI and CKD.

		Model 1 OR (95%CI) P value	Model 2 OR (95%CI) P value	Model 3 OR (95%CI) P value
CKD	VAI	1.07 (1.03, 1.11) <0.001	1.07 (1.03, 1.12) <0.001	1.03 (1.00, 1.06) 0.077
	AVAI	1.47 (1.41, 1.54) <0.001	1.54 (1.44, 1.65) <0.001	1.47 (1.36, 1.59) <0.001
	Q1	[Reference]	[Reference]	[Reference]
	Q2	1.79 (1.38, 2.32) <0.001	1.63 (1.21, 2.20) 0.002	1.50 (1.05, 2.14) 0.028
	Q3	7.49 (5.98, 9.38) <0.001	6.24 (4.23, 9.22) <0.001	4.95 (3.06, 8.00) <0.001
	P for trend	<0.001	<0.001	<0.001

CI: confidence interval; CKD: chronic kidney disease; OR: odds ratio; Q: quartiles; AVAI: age-adjusted visceral adiposity index.

Model 1: no covariates adjusted; Model 2: adjusted for age, sex, and race; Model 3: adjusted for age, sex, race, BMI, educational level, smoke, drink, activity status, diabetes, hypertension.

3.3. Nonlinear relationship and saturation effect analysis

Restricted cubic spline (RCS) analysis revealed a significant nonlinear association between AVAI and CKD prevalence (P-nonlinear < 0.0001), as shown in Figure 2. As AVAI levels increased, the prevalence of CKD also rose. Threshold effect analysis identified a critical inflection point for AVAI at −6.5 (see Table 3). A histogram of AVAI distribution in the study population (Figure S1) further illustrates the proportion of participants above and below this threshold, supporting its relevance as a data-driven cutoff. When AVAI exceeded −6.5, further increases were significantly associated with higher CKD prevalence (OR = 1.83, 95% CI = 1.70–1.97). This threshold may serve as an epidemiologically meaningful cutoff, marking the point beyond which CKD prevalence rises sharply. Identifying individuals with AVAI values above −6.5 may help guide early clinical monitoring and metabolic interventions targeting visceral adiposity.

Figure 2.

RCS curve fits the Association of AVAI with CKD.

Adjusted for age, sex, race, BMI, educational level, smoking, drink, activity status, diabetes, and hypertension.

Table 3.

Two-stage logistic regression between AVAI and CKD.

	AVAI	OR (95%CI) P value
CKD	Standard linear model	1.50 (1.42, 1.58) <0.001
	AVAI < −6.5	1.17 (1.08, 1.27) <0.001
	AVAI > −6.5	1.82 (1.69, 1.96) <0.001
	Log-likelihood ratio test	<0.001

AVAI: age-adjusted visceral adiposity index; CKD: chronic kidney disease; CI: confidence interval; OR: odds ratio.

3.4. Subgroup analysis and sensitivity analysis

Stratified analysis was conducted based on various demographic and lifestyle factors (Figure 3; Table 4). The positive association between AVAI and CKD prevalence remained consistent across all subgroups. Interaction analysis revealed significant interaction effects for age, race, and BMI. Specifically, the positive association between AVAI and CKD was particularly significant among individuals aged over 50, while it was not statistically significant in those under 50. To further test the robustness of the study, sensitivity analysis was performed after excluding missing values of covariates. Finally, 7,027 participants were left for sensitivity analysis, including 5,738 non-CKD people and 1,289 CKD people. The results are shown in Table S2. The sensitivity analysis showed that it was consistent with the results of this study.

Figure 3.

Subgroup analysis of the association between AVAI and CKD prevalence.

Table 4.

Subgroup analysis between AVAI and CKD.

Characteristic	Group	OR (95%CI) P value	P for interaction
Age	<50	1.07 (0.93, 1.23) 0.400	<0.001
	>50	1.71 (1.54, 1.90) <0.001
Sex	Male	1.43 (1.27, 1.61) <0.001	0.800
	Female	1.50 (1.33, 1.70) <0.001
Race	Mexican American	1.25 (1.02, 1.53) 0.034	0.036
	Non-Hispanic black	1.44 (1.27, 1.63) <0.001
	Non-Hispanic white	1.48 (1.33, 1.65) <0.001
	Others	1.61 (1.38, 1.89) <0.001
BMI	Normal	1.28 (1.09, 1.51) 0.003	0.004
	Overweight	1.64 (1.39, 1.94) <0.001
	Obese	1.54 (1.38, 1.73) <0.001
Smoke	No	1.45 (1.28, 1.65) <0.001	0.300
	Yes	1.49 (1.30, 1.71) <0.001
Hypertension	No	1.40 (1.23, 1.59) <0.001	0.400
	Yes	1.55 (1.36, 1.75) <0.001
Diabetes	No	1.44 (1.31, 1.59) <0.001	0.800
	Yes	1.57 (1.29, 1.91) <0.001

3.5. Comparison of VAI and AVAI in predicting CKD prevalence

A comparative analysis using ROC curves was conducted to evaluate the predictive performance of VAI and AVAI in estimating CKD prevalence (Figure 4; Table S3). The AUC for AVAI was 0.7581 (95% CI: 0.7436–0.7726), notably higher than that of VAI (AUC = 0.5685, 95% CI: 0.5520–0.5850), indicating superior discriminatory ability. Moreover, AVAI achieved a sensitivity of 66.9% and a specificity of 73.7%, representing a better balance between the two metrics compared to VAI. In addition, decision curve analysis (Figure S2) showed that AVAI provided a higher net clinical benefit than VAI across a range of threshold probabilities, further supporting its practical utility in CKD risk prediction.

Figure 4.

Diagnostic performance of obesity AVAI and VAI index on CKD prevalence.

4. Discussion

Based on a cross-sectional analysis of NHANES 2007–2018 data, this study identified a significant positive association between AVAI and CKD prevalence, independent of multiple covariates. Subgroup analysis revealed that this association was particularly pronounced among older adults. Compared with the conventional VAI, AVAI demonstrated superior discriminatory ability, suggesting its potential value as a noninvasive tool for early risk stratification and clinical management of CKD in high-risk populations.

The conclusions of this study align with prior research, further emphasizing the contribution of visceral fat to CKD pathogenesis. Qin et al. initially highlighted that individuals with the elevated VAI levels exhibited a markedly higher prevalence of CKD, suggesting that the accumulation of visceral fat could significantly impair kidney function [16]. However, the investigation by Wang et al. did not delve into the detailed mechanisms connecting VAI and CKD, nor did it account for more precise indicators of visceral fat distribution. Building on this, Fang et al. demonstrated a strong association between VAI and CKD risk, particularly in older or obese populations [17]. Nonetheless, its predictive performance was weaker in low-BMI or younger populations, highlighting the limitations of VAI. Zhu et al. noted that increased fat, particularly visceral fat, contributes to insulin resistance, inflammatory responses, and oxidative stress, all of which are key risk factors for kidney damage [18]. In contrast, VAI, as a traditional measure of visceral adiposity, may fail to fully capture the complex effects of visceral fat on metabolic and renal health [19]. Given that age is a well-established risk factor for CKD, the incorporation of age adjustment in AVAI allows for a more accurate representation of the relationship between visceral fat and CKD prevalence across different age groups, thereby enhancing its discriminative power in heterogeneous populations. In the present study, AVAI demonstrated significantly greater performance than VAI in identifying individuals with CKD, with an AUC of 0.7581. Recent studies based on NHANES data have reported that commonly used anthropometric and inflammatory indices show relatively the modest discriminative ability. For example, the AUCs for the A Body Shape Index (ABSI) and Conicity Index (C-index) in predicting CKD were 0.655 and 0.657, respectively, both of which outperformed BMI and waist circumference [20]. The Body Roundness Index (BRI) has also been studied, with an AUC of 0.625 (95% CI: 0.616–0.633) [21]. Additionally, the Aggregate Index of Systemic Inflammation (AISI), an inflammation-based biomarker, achieved an AUC of 0.563 (95% CI: 0.557–0.569) [22]. Compared to these indices, AVAI exhibited a superior AUC, suggesting that incorporating both anthropometric and metabolic parameters—along with age—may improve the ability to detect CKD in population-based settings (Table S4). However, further validation incorporating additional clinical, biochemical, and metabolic parameters is necessary to confirm its robustness and applicability in clinical settings.

The observed positive relationship between AVAI and CKD likely underscores the pivotal role of visceral fat in CKD pathogenesis. Visceral fat, beyond its role as an energy reservoir, functions as an active endocrine tissue, releasing various pro-inflammatory and metabolic regulatory molecules essential for initiating and advancing CKD [19,23,24]. This chronic inflammation leads to metabolic disturbances and accelerates kidney injury and CKD progression by activating tubular epithelial cells and renal interstitial fibrosis signaling pathways [25,26]. Additionally, studies have shown that visceral fat accumulation is associated with increased urinary albumin excretion, indicating that chronic inflammation can compromise glomerular permeability, thereby elevating CKD risk [27]. The rise in free fatty acids (FFAs) within visceral fat induces lipotoxicity, causing direct damage to tubular cells and glomerular endothelial cells [28,29]. Lipotoxicity further induces insulin resistance, disrupts glucose and lipid metabolism, and contributes to the development of metabolic syndrome [30]. Insulin resistance is closely linked to a decline in eGFR and exacerbates the renal workload by altering renal hemodynamics [31]. Moreover, visceral fat accumulation is frequently accompanied by enhanced oxidative stress, with elevated reactive oxygen species (ROS) levels leading to direct damage to the glomeruli and renal tubules [32]. ROS can also worsen renal function decline by inducing glomerular matrix production and fibrosis [33]. Oxidative stress and metabolic disturbances, including insulin resistance and lipotoxicity, create a vicious cycle that accelerates CKD progression [34,35].

Restricted cubic spline regression analysis revealed a significant nonlinear relationship between AVAI and CKD prevalence, likely linked to the detrimental effects on kidney function once visceral fat accumulation surpasses a certain threshold. Studies indicate that excessive visceral fat significantly increases metabolic and inflammatory burdens, hastening renal function deterioration [36,37]. This threshold may represent a clinically actionable cutoff: patients exceeding this point exhibited significantly higher CKD risk, warranting enhanced surveillance—such as earlier eGFR and albuminuria testing—and targeted interventions including structured weight-loss programs, increased physical activity, and metabolic risk factor optimization. For instance, evidence shows that in CKD stages 1–4, intentional weight loss reduces proteinuria and glomerular hyperfiltration, improving renal outcomes (adjusted OR –0.6, p < 0.05) [38]. In contrast, individuals with AVAI below −8.0 (well below the threshold) may continue with standard monitoring without aggressive intervention. AVAI is derived from readily available measurements—WC, HDL, TG, BMI, and age—making it a low-cost surrogate for visceral adiposity. It correlates strongly with MRI-measured visceral fat (R ≈ 0.7, p < 0.01) while avoiding expensive imaging procedures like CT/MRI [5]. Previous economic analyses suggest that early CKD detection using simple risk indices can significantly curtail long-term dialysis-related costs, which exceeded $37 billion in the U.S. in 2019 alone [38]. Thus, AVAI integration into routine screening protocols may be cost-effective, facilitating timely intervention with minimal resource utilization. Nonetheless, AVAI’s accuracy diminishes in advanced CKD (stages 4–5) due to fluid retention, which inflates waist circumference and body weight, potentially distorting index values. Fluid overload is prevalent in late-stage CKD and can lead to misleading adiposity estimates using anthropometric methods[39]. Therefore, AVAI may be most appropriate for early CKD risk assessment (stages 1–3), while its application in advanced disease warrants cautious interpretation or supplemental methods when volume overload is present. In subgroup analyses, significant interactions between AVAI and both age and BMI were observed, indicating that the association between AVAI and CKD prevalence may be modified by individual metabolic status. In older adults, age-related physiological changes—such as increased oxidative stress, endothelial dysfunction, and mitochondrial decline—may enhance susceptibility to the deleterious effects of visceral adiposity on renal function [40,41]. Likewise, individuals with higher BMI are more likely to exhibit insulin resistance, chronic low-grade inflammation, and ectopic lipid deposition in renal tissues, all of which are implicated in glomerular and tubular injury [42,43]. These pathophysiological processes may partially account for the stronger associations observed in older and obese subgroups. More prospective studies are needed in the future to further explore the biological mechanism of AVAI and CKD prevalence. The variation observed across racial subgroups may similarly reflect population-level differences in fat distribution and metabolic risk profiles. Since AVAI components—including waist circumference, triglycerides, HDL-C, and BMI—are known to differ by ethnicity, the applicability of a single threshold across all populations may be limited. Although we applied a unified cutoff to enhance generalizability, further validation of population-specific thresholds may improve the precision of CKD risk stratification in diverse cohorts.

5. Strengths and limitations

One major strength of this study lies in its use of the extensive, representative NHANES dataset, paired with a complex stratified sampling strategy, which bolsters the generalizability of the results. Furthermore, the novel AVAI metric demonstrated superior predictive capability for assessing CKD risk. However, this study has several limitations that should be considered. First, its cross-sectional design precludes determination of the temporal relationship between AVAI and CKD, limiting causal inference. Second, the absence of longitudinal follow-up data prevents the assessment of the long-term predictive value of AVAI for CKD onset or progression. Therefore, any application of AVAI in clinical settings must be approached with caution. Prospective studies involving diverse populations and longitudinal validation are needed to confirm the generalizability and reliability of AVAI-based risk stratification before its implementation in clinical practice. Third, although AVAI showed good predictive performance in this study population, it has not been externally validated against imaging-based gold standards such as computed tomography (CT) or magnetic resonance imaging (MRI), and its accuracy in reflecting visceral adiposity across different demographic groups remains uncertain. Fourth, despite adjusting for multiple confounders, important variables such as dietary intake and medication use—both of which may influence kidney function—were not included, raising the possibility of residual confounding. Lastly, as NHANES is a cross-sectional survey, survivorship bias may exist, as individuals with advanced CKD or multiple comorbidities could have been underrepresented due to early mortality or exclusion from the survey. This survivorship bias may lead to an underestimation of the true prevalence and severity of CKD in the general population, particularly among individuals with more advanced disease who may be less likely to participate in household-based surveys like NHANES. As a result, the association between AVAI and CKD identified in this study may reflect a healthier surviving cohort, potentially attenuating the strength of observed associations. Future prospective studies with longitudinal follow-up are warranted to confirm these findings and account for possible survivor effects.

6. Conclusion

In this nationally representative US population, AVAI was significantly associated with CKD prevalence and exhibited better discriminative performance than the traditional VAI. These findings suggest that AVAI may be a useful marker for identifying individuals at the elevated risk of CKD. However, given the cross-sectional nature of this study, causal inferences cannot be drawn, and the clinical applicability of AVAI requires further investigation. Prospective longitudinal studies and external validation across diverse populations are needed to determine whether AVAI can be integrated into routine CKD screening and risk stratification strategies. If validated, AVAI may contribute to more targeted preventive interventions in high-risk groups, potentially improving early CKD detection and management.

Glossary

Abbreviation List

ACR

albumin-to-creatinine ratio

AST

aspartate aminotransferase

AUC

area under the curve

AVAI

Age-Adjusted Visceral Adiposity Index

ALT

alanine aminotransferase

BMI

Body Mass Index

CKD

chronic kidney disease

CRP

C-reactive protein

eGFR

estimated glomerular filtration rate

FFAs

free fatty acids

GFR

glomerular filtration rate

HDL-C

high-density lipoprotein cholesterol

IL-6

interleukin-6

NHANES

National Health and Nutrition Examination Survey

OR

odds ratio

PIR

poverty income ratio

RCS

restricted cubic spline

ROC

receiver operating characteristic

ROS

reactive oxygen species

TG

triglycerides

TNF-α

tumor necrosis factor alpha

VAI

Visceral Adiposity Index

Funding Statement

The authors received no financial support for this article.

Ethics approval and consent to participate

NHANES is conducted by the Centers for Disease Control and Prevention and the National Center for Health Statistics. The National Center for Health Statistics Research Ethics Review Committee reviewed and approved the NHANES study protocol. All participants signed written informed consent.

Competing interests

The authors declare no conflict of interest.

Consent for publication

Not applicable.

Availability of data and materials

The National Health and Nutrition Examination Survey dataset is publicly available at the National Center for Health Statistics of the Center for Disease Control and Prevention (https://wwwn.cdc.gov/nchs/nhanes/default.aspx).