Systemic immune-inflammation index mediates the association between abdominal obesity and serum klotho levels

Abstract

The weight-adjusted waist index (WWI) has emerged as a reliable indicator of abdominal obesity. α-Klotho, a transmembrane protein, functions as a suppressor of aging. However, the relationship between these two factors remains underexplored. This study aims to investigate the association between WWI and serum α-Klotho levels in middle-aged and elderly Americans, with a focus on exploring the potential mediating role of the systemic immune inflammation index (SII). A cross-sectional study was conducted using data from 6997 middle-aged and elderly Americans participating in the National Health and Nutrition Examination Surveys (NHANES) between 2011 and 2016. Multiple linear regression analysis was employed to assess the relationship between WWI and serum α-Klotho concentrations. Additionally, mediation analysis was performed to investigate the mediating effect of SII on the relationships. Our analysis revealed a significant negative correlation between WWI and serum α-Klotho levels in the survey-weighted multiple linear regression models (adjusted percent change: -7.79; 95% CI: -10.15, -5.37). Mediation analysis demonstrated that the association between WWI and α-Klotho levels was partially mediated by SII (adjusted percent change: -0.88; 95% CI: -1.24, -0.45), with the proportion of mediation amounting to 11.6%. Further age-stratified results showed that the mediating role of SII was more pronounced among individuals aged ≥ 60 years, exhibiting a mediating effect of 26.3%, in contrast to 4.2% for those < 60 years. The findings suggest that WWI is inversely associated with serum α-Klotho concentrations and that this association is partially mediated by SII, especially in older people.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-88015-2.

Introduction

The membrane protein α-Klotho, encoded by the Klotho gene discovered by Kuro-o et al.¹, plays a pivotal role in suppressing aging through various biological processes². Animal studies have demonstrated that Klotho-deficient mice exhibit premature death¹, whereas mice overexpressing the Klotho gene enjoy a lifespan extended by 20–30%³. Furthermore, α-Klotho levels decline with age⁴ and are associated with age-related diseases such as chronic kidney disease⁵, diabetes⁶, cardiovascular disease⁷, Alzheimer’s disease⁸ and cancer⁹ in humans studies. The membrane protein α-Klotho is an antiaging marker.

Recently, accumulating evidence suggests that obesity could exacerbate aging due to the significantly increased risk of age-related comorbidities^10–13. Some studies have reported lower α-Klotho concentrations in serum among individuals with obesity compared to control participants^14–16. However, cross-sectional studies have yielded inconsistent results, indicating a positive association between obesity and α-Klotho concentrations^17,18. Therefore, the relationship between obesity and serum concentrations of α-Klotho remains to be further evaluated.

Obesity was previously assessed using the body mass index (BMI), a widely employed method for measuring obesity. However, recent studies have challenged the accuracy of BMI, pointing out its limitations in distinguishing lean body mass from fat mass, as well as its inability to differentiate between peripheral and visceral adipose tissues¹⁹. Specifically, BMI provides a general measure of obesity without considering the distribution of body fat. In contrast, some studies have demonstrated that waist circumference (WC) is a superior metric as it better distinguishes between peripheral and visceral adipose tissues^20,21. More recently, to control the influence of weight on WC, Park et al.²² introduced a novel obesity index called WWI, which represents weight-independent central obesity. Numerous studies have shown that WWI offers a more accurate assessment of obesity compared to BMI^22–24.

A prevailing theory suggests that inflammation serves as a “common soil” for the development of numerous chronic diseases^25,26, and it is also recognized as one of the hallmarks of aging²⁶. Furthermore, a mounting body of research has demonstrated that inflammation can modulate Klotho levels^27–30. In fact, animal studies indicate that inflammatory cytokines regulate α-Klotho expression, and a decrease in α-Klotho expression can be alleviated by the use of inflammatory cytokines antibodies^29,30. Epidemiologic studies also reveal a negative association between inflammatory cytokines and α-Klotho levels^31–33. Moreover, an increasing number of studies suggest that inflammation serves as a critical mechanism through which obesity promotes aging and aging-related diseases, such as cancer and diabetes^34–36. Consequently, it is plausible that inflammation may mediate the relationship between obesity and α-Klotho levels, an aging-related protein.

A recently developed inflammatory measure, known as SII, has demonstrated high effectivity and stability in indicating systemic inflammation, relying on neutrophil, platelet, and lymphocyte counts^37,38. Recent investigations have revealed a favorable correlation between SII and obesity, as well as components related to obesity^39–41. Despite these findings, the role of SII as a potential mediator between obesity and α-Klotho concentrations remains unexplored. Future studies are warranted to assess this relationship and further elucidate the complex interactions between inflammation, obesity, and α-Klotho levels.

Accordingly, leveraging the distinct advantages of WWI and SII in measuring obesity and inflammation, respectively, the current study aimed to evaluate the associations between WWI and α-Klotho levels. Furthermore, it sought to investigate whether SII serves as a potential intermediary in the relationship between WWI and α-Klotho levels.

Methods

Study population

The present study utilized serial cross-sectional data derived from the National Health and Nutrition Examination Survey (NHANES) conducted between 2011 and 2016⁴². NHANES is a program designed to assess the health and nutritional status of adults and children in the United States. Initially, the study included 25,152 participants who had available data on weight and waist circumference. However, to ensure the accuracy and reliability of the research findings, several exclusion criteria were applied. Firstly, individuals with missing data on SII or serum a-Klotho levels were excluded from the study, resulting in the removal of 17,684 participants. Secondly, pregnant women were excluded as pregnancy can significantly alter hormonal and physiological states, potentially confounding the results. This exclusion criterion led to the removal of an additional 6 participants. Lastly, participants with missing data on other relevant covariates were also excluded to maintain the consistency and completeness of the dataset. This exclusion criterion accounted for the removal of 465 participants. After applying all exclusion criteria, a total of 6,997 participants were included in the final study population (Fig. 1).

Fig. 1.

The flow chart of participants selection. NHANES National Health and Nutrition Examination Survey.

Assessment of weight-adjusted-waist index

Obesity was assessed using the weight-adjusted waist index (WWI), as outlined by Park et al.²². An individual exhibiting higher levels of obesity would correspondingly possess a greater WWI score. The WWI (cm/√kg) is calculated by dividing WC (cm) by the square root of weight (kg). The waist circumference and weight of each participant were meticulously measured, rounded to the nearest 0.1 centimeter and 0.1 kg, respectively. A precise digital floor scale was employed for weight measurement, while a tape measure was used to determine waist circumference at the topmost lateral border of the ilium. These measurements were collected by trained health technicians in mobile examination centers, ensuring the accuracy and reliability of the data. In the present research, WWI serves as an exposure variable, providing a quantitative metric for analyzing and understanding the impact of obesity on various health outcomes.

Assessment of serum α-Klotho

In this study, serum samples were gathered from middle-aged and elderly Americans over a five-year period, spanning from 2011 to 2016. These specimens were then preserved at a temperature of -80 ℃ and subsequently transported to the Northwest Lipid Metabolism and Diabetes Research Laboratory at the University of Washington. The objective was to assess the concentrations of α-klotho during the period of 2019 to 2020. To accurately measure the levels of α-Klotho in the serum, a sandwich enzyme-linked immunosorbent assay (ELISA) test was employed, sourced from IBL International in Japan. This method, as described by Yamazaki et al.⁴³, was chosen for its reliability and precision in determining α-Klotho concentrations. To guarantee the integrity and accuracy of the results, duplicate measurements were conducted for each sample. The final value reported for each sample was calculated as the average of these two measurements. Additionally, the performance of all samples was rigorously evaluated to ensure compliance with the laboratory’s standardized criteria, thus ensuring the highest quality of data.

Evaluation of systemic immune-inflammation index

Our study employed SII to evaluate the overall inflammatory status in participants. SII, originally proposed by Hu et al.³⁷, was initially designed to evaluate the prognostic significance in hepatocellular carcinoma. Recent studies have demonstrated that individuals with other cancer and cardiovascular diseases also had higher SII than their control counterparts^44,45. When compared to other inflammation markers, such as the neutrophil to lymphocyte ratio and platelet to lymphocyte ratio, the SII has emerged as a more objective indicator that reflects the intricate balance between the host’s inflammatory and immune response status³⁷. Systemic immune-inflammation index (SII) was cal culated as total peripheral platelets count (P) × neutro phil-to-lymphocyte ratio (N/L) (SII = P × N/L ratio)³⁷.

Other covariates

The statistical analysis covariates were carefully chosen based on prior research exploring the relationship between serum α-Klotho concentrations and obesity^15,16. After thorough consideration, we settled on 12 covariates, encompassing diverse aspects such as basic demographic characteristics, lifestyle habits, health status, and survey cycles from NHANES. The demographic characteristics encompassed age, gender, race/ethnicity, family PIR, and educational level. The lifestyle habits included serum cotinine status, alcohol intake, physical activity, and dietary quality. Additionally, we took into account the diseased status of the participants, specifically whether they had hypertension or diabetes. Hypertension was defined as having an average systolic blood pressure of 140 mm Hg or higher, a diastolic blood pressure of 90 mm Hg or higher, current use of anti-hypertensive medication, or self-reported hypertension. Diabetes was defined as having an HbA1c level of 6.5% or higher, self-reported diabetes, current use of hypoglycemic medications, or self-reported insulin use.

Statistical analysis

Descriptive statistics were employed to characterize the fundamental features of the study participants and variables. For continuous variables exhibiting normal distribution, they were represented by means and accompanying standard deviations (Mean ± SD). Conversely, those deviating from normal distribution were presented through geometric means and percentiles or via median values along with an interquartile range. Categorical variables were delineated by their frequency (proportion).

Linear regression

The relationships between WWI and serum concentrations of α-Klotho were investigated using linear regression. In this analysis, WWI was designated as a continuous exposure variable, whereas serum α-Klotho levels served as a variable representing continuous outcomes. The modified regression coefficient, expressed as the percentage change in α-Klotho levels for each 0.25-fold increase in WWI, was derived using the formula: 100% of (e ^{(ln 1.25× β)} − 1)⁴⁶. To further assess the correlation between serum α-Klotho levels and WWI within the context of linear regression models, WWI was stratified into quartiles. By treating the median of each WWI quartile as a continuous variable, the P trend was determined. Additionally, the percentage change in α-Klotho levels associated with the quartiles of WWI was estimated using the formula: (e ^β − 1) × 100%⁴⁷. To ensure the accuracy of the findings, various adjusted variables were taken into account, including age, sex, race, education, serum cotinine level, drinking habits, physical activity, diet quality, survey cycle, family income to poverty ratio, hypertension, and diabetes. Notably, the associations between WWI and SII, as well as between SII and serum α-Klotho concentrations, were similar to the examination of the relationships between α-Klotho levels and WWI described above.

Mediation analysis

The mediation analysis, conducted using the R package “mediation” aimed to investigate the mediating role of SII in the relationship between WWI and serum α-Klotho levels. Additionally, it sought to quantify the magnitude of this mediation effect. The total effect (TE) was disentangled into a direct effect (DE) and an indirect effect (IE)⁴⁸. In this study, DE represented the impact of WWI on serum α-Klotho levels without the involvement of SII, whereas IE encompassed the effects of WWI on α-Klotho levels mediated through SII. To determine TEs, DE, and IE in the mediation analysis, multiple linear regression analysis was employed, accounting for potential confounders. The proportion of mediation was calculated by dividing IE by TE. To compute standard errors and confidence intervals for effect estimates, 10,000 bootstrap resampling iterations were utilized. Notably, the adjustment covariates employed in all mediation analyses were consistent with those used in the primary analysis of WWI and α-Klotho levels.

Sensitivity analysis

Sensitivity analysis was performed to explore the robustness of the results by excluding individuals with serum cotinine levels ≥ 10 ng/mL and by excluding participants with any liver condition, weak/failing kidneys, or cardiovascular disease, respectively.

All data analyses were conducted using R (version V.4.2.1) and SPSS (version 25). Statistical significance was set at P < 0.05.

Results

Baseline characteristics of participants

The characteristics of the 6,997 individuals participating in the present study are presented in Table 1. Of these individuals, 46.1% were aged over 60, 51.7% were female, and 39.4% were Non-Hispanic White. Additionally, more than half of the participants reported not smoking (59.8%), drinking alcohol (70.6%), possessing a high level of education (54.1%), and engaging in physical activity (56.7%). Furthermore, 35.9% of the participants had a family PIR falling within the range of 1.3 to 3.4.

Table 1.

Characteristics of included participants from NHANES 2011–2016 (N = 6,997).

Characteristics	N (%)
Age (years)
40–59	3774 (53.9)
≥ 60	3223 (46.1)
Sex
Male	3381 (48.3)
Female	3616 (51.7)
Race/ethnicity
Mexican American	1013 (14.5)
Other Hispanic	884 (12.1)
Non-Hispanic White	2756 (39.4)
Non-Hispanic Black	1465 (20.9)
Other race	915 (13.1)
Family PIR
< 1.3	2123 (30.3)
1.3–3.4	2512 (35.9)
≥ 3.5	2362 (33.8)
Education
Under high school	1696 (24.2)
High school or equivalent	1516 (21.7)
Above high school	3785 (54.1)
Serum cotinine (ng/mL)
< 0.15	4183 (59.8)
0.15–9.9	1246 (17.8)
≥ 10	1568 (22.4)
Alcohol assumption (drinks/year)
< 12	2055 (29.4)
≥ 12	4942 (70.6)
Physical activity (MET-minutes/week)
< 600	3034 (43.4)
600–7999	3210 (45.9)
≥ 8000	753 (10.8)
Diet quality
Excellent/very good	2126 (30.4)
Good	2918 (41.7)
Fair/poor	1953 (27.9)
Hypertension
Yes	3866 (55.3)
No	3131 (44.7)
Diabetes
Yes	1611 (23.0)
No	5386 (77.0)
Survey cycle
2011–2012	2160 (30.9)
2013–2014	2483 (35.5)
2015–2016	2354 (33.6)

NHANES National Health and Nutrition Examination Survey, Family PIR family income to poverty ratio, MET metabolic equivalent.

Distribution of WWI, SII, and α-Klotho

The distribution of weight-adjusted waist index and inflammatory markers, serum concentrations of α-Klotho among the study participants have been compiled in Table 2. The median values were 11.274 cm/√kg (interquartile range (IQR): 10.774, 11.775 cm/√kg), 444.706 cells/L (IQR: 316.875, 627.000 cells/L) and 807.40 pg/ml (IQR: 657.600, 994.90 pg/ml) for WWI, SII and serumα-Klotho levels, respectively.

Table 2.

Distribution of WWI, SII and serumα-Klotho levels among participants included in the study.

	GM	Percentile
		5th	25th	50th	75th	95th
WWI (cm/√kg)	11.262	10.093	10.774	11.274	11.775	12.565
SII (cells/L)	444.091	188.177	316.875	444.706	627.000	1050.257
α-Klotho (pg/ml)	809.932	472.940	657.600	807.40	994.90	1395.940

GM geometric mean, WWI weight-adjusted waist index, SII systemic immune inflammation index.

Association between WWI and serum α-Klotho levels

Table 3 presents the impact of WWI on serum α-Klotho concentrations. In a crude model, WWI were linked with serum α-Klotho concentrations (percent change: -7.79; 95% CI: -10.15, -5.37). A statistically significant correlation was also identified between WWI and serum α-Klotho levels after controlling for covariates (percent change: -7.61; 95% CI: -10.35, -4.80). To detect possible non-linear trends, the study investigated the relationship between serum α-Klotho concentrations and quartiles of WWI exposure. The findings revealed that the highest exposure levels (Q4) were associated with the most significant decrease in α-Klotho concentrations (percent change: -5.20; 95% CI: -7.51, -2.83). Lower exposure quartiles (Q3 and Q2) also showed decreases in α-Klotho levels, although to a lesser extent. Additionally, when evaluating WWI on a linear scale, a clear trend emerged, indicating that higher levels of WWI were consistently associated with lower concentrations of α-Klotho (P trend < 0.001).

Table 3.

The association of weight-adjusted waist index with serum α-Klotho levels.

Exposure	Percent changea (95% CI)	P value	Percent changeb (95% CI)	P value
WWI
Continuous	− 7.79 (− 10.15, − 5.37)	< 0.001	− 7.61 (− 10.35, − 4.80)	< 0.001
Q1 (< 10.774)	Reference		Reference
Q2 (10.774–11.274)	− 3.59 (− 5.68, − 1.46)	0.001	− 3.30 (− 5.41, − 1.16)	0.003
Q3 (11.274–11.775)	− 4.87 (− 6.93, − 2.77)	< 0.001	− 4.28 (− 6.43, − 2.07)	< 0.001
Q4 (≥ 11.775)	− 5.59 (− 7.63, − 3.50)	< 0.001	− 5.20 (− 7.51, − 2.83)	< 0.001
P for trend	< 0.001		< 0.001

CI confidence interval, WWI weight-adjusted waist index. ^aUnadjusted confounder. ^bAdjusted for age, sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

To further understanding the findings, a stratified analysis was conducted based on age categories. Among both younger (< 60 years) and older (≥ 60 years) participants, similar associations were observed between WWI and serum α-Klotho levels (Table 4).

Table 4.

The association of weight-adjusted waist index with serum α-Klotho levels stratified by age.

Exposure	< 60 years old	P value	≥ 60 years old	P value
	Percent change (95% CI)		Percent change (95% CI)
WWI
Continuous	− 8.02 (− 11.63, − 4.26)	< 0.001	− 6.67 (− 10.80, − 2.34)	0.003
Q1	Reference		Reference
Q2	− 2.13 (− 4.76, 0.57)	0.121	− 5.35 (− 8.86, − 1.70)	0.004
Q3	− 3.69 (− 6.53, − 0.76)	0.014	− 5.44 (− 8.85, − 1.89)	0.003
Q4	− 5.24 (− 8.43, − 1.94)	0.002	− 5.82 (− 9.34, − 2.18)	0.002
P for trend	< 0.001		0.011

CI confidence interval, WWI weight-adjusted waist index. The models were adjusted for age, sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

Association between WWI and SII

The effects of WWI on SII are presented in Fig. 2. In a crude model, WWI were connected to SII (percent change: 23.04; 95% CI: 17.96, 28.33). After adjusting for confounders, a statistically significant correlation was also found between WWI and SII (percent change: 14.09; 95% CI: 8.71, 19.74).

Fig. 2.

Association of weight-adjusted-waist index with systemic immune inflammation index. The models were adjusted for age, sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

In addition, the highest (Q4) WWI exposure levels were associated with higher SII (percent change: 10.23; 95% CI: 5.94, 14.71) in the adjusted analyses, followed by the third quartile (Q3) (percent change: 3.60; 95% CI: -0.13, 7.47), and the second quartile (Q2) (percent change: 0.83; 95% CI: -2.66, 4.46). Similarly, when WWI was evaluated on linear trends, higher WWI was associated with elevated SII (P trend < 0.001).

In the age stratified analysis, we found that the association of WWI with SII levels was large among individuals aged ≥ 60 years compared with individuals aged < 60 years. For example, when WWI as continuous variable, the percentage increase in SII levels for each 1-fold rise in WWI was 21.68 (95% CI: 12.70, 31.38) among participants aged ≥ 60 years, compared with individuals aged < 60 years (percent change: 8.35; 95% CI: 1.84, 15.26). Similar results were also found when WWI as categorical variable (Fig. 3).

Fig. 3.

Association of weight-adjusted-waist index with systemic immune inflammation index stratified by age. The models were adjusted for sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

Association between SII and serum α-Klotho levels

Table 5 displays the impact of SII on serum α-Klotho concentration. In a crude model, higher SII and lower serum α-Klotho concentrations were related, with a percent change of -1.61 (95% CI: -1.92, -1.30). This correlation persisted even after adjusting for various covariates, with a percent change of -1.51 (95% CI: -1.83, -1.19). Individuals with the highest SII levels (Q4) exhibited the most significant decrease in α-Klotho concentrations, with a percentage change of -8.38 (95% CI: -10.37, -6.34). This decrease was followed by those in the third (Q3) and second (Q2) quartiles, with percentage changes of -5.78 (95% CI: -7.81, -3.71) and − 4.51 (95% CI: -6.55, -2.42), respectively. Consistent with these findings, a linear trend analysis confirmed that higher SII levels were associated with lower α-Klotho concentrations, with a statistically significant trend (P trend < 0.001).

Table 5.

The association of systemic immune inflammation index with serum α-Klotho levels.

Exposure	Percent changea (95% CI)	P value	Percent changeb (95% CI)	P value
SII
Continuous	− 1.61 (− 1.92, − 1.30)	< 0.001	− 1.51 (− 1.83, − 1.19)	< 0.001
Q1	Reference		Reference
Q2	− 4.67 (− 6.72, − 2.57)	< 0.001	− 4.51 (− 6.55, − 2.42)	< 0.001
Q3	− 6.08 (− 8.11, − 4.01)	< 0.001	− 5.78 (− 7.81, − 3.71)	< 0.001
Q4	− 9.07 (− 11.03, − 7.07)	< 0.001	− 8.38 (− 10.37, − 6.34)	< 0.001
P for trend	< 0.001		< 0.001

CI confidence interval, SII systemic immune inflammation index. ^aUnadjusted confounder. ^bAdjusted for age, sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

An age-stratified analysis further revealed that the correlation between SII and α-Klotho levels was stronger among individuals over 60 years old compared to those under 60. For instance, when considering SII as a continuous variable, a 0.25-fold increase in SII was associated with a percentage change in α-Klotho levels of -1.96 (95% CI: -2.41, -1.52) among those over 60, while the corresponding change among those under 60 was − 1.08 (95% CI: -1.54, -0.62). Similar outcomes were observed when SII was analyzed as a categorical variable. (Table 6).

Table 6.

The association of systemic immune inflammation index with serum α-Klotho levels stratified by age.

Exposure	< 60 years old	P value	≥ 60 years old	P value
	Percent change (95% CI)		Percent change (95% CI)
SII
Continuous	− 1.08 (− 1.54, − 0.62)	< 0.001	− 1.96 (− 2.41, − 1.52)	< 0.001
Q1	Reference		Reference
Q2	− 3.75 (− 6.56, − 0.86)	0.012	− 5.43 (− 8.37, − 2.38)	< 0.001
Q3	− 4.98 (− 7.77, − 2.11)	< 0.001	− 6.89 (− 9.81, − 3.87)	< 0.001
Q4	− 6.15 (− 8.95, − 3.27)	< 0.001	− 11.10 (− 13.91, − 8.19)	< 0.001
P for trend	< 0.001		< 0.001

CI confidence interval, SII systemic immune inflammation index. The models were adjusted for sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes.

SII as potential mediators of the effect of WWI on α-Klotho levels

As previously mentioned, the statistically significant associations were found between WWI and SII, as well as between SII and α-Klotho concentrations. These findings suggest that the impact of WWI on α-Klotho concentrations may be partially explained by changes in SII. To further explore these associations, we conducted covariate-adjusted causal mediation analyses. Our results indicate that an increase in WWI leads to a reduction in α-Klotho levels, with a total effect value of -7.40% (95% CI: -10.09%, -4.58%). This effect comprises a direct effect of -6.57% (95% CI: -9.35%, -3.72%) and significant indirect “mediation” effects through SII of -0.88% (95% CI: -1.24%, -0.45%). Notably, SII accounts for approximately 11.6% of the total impact of WWI on α-Klotho concentrations. Figure 4 shows a visual representation of the causal mediation analysis.

Fig. 4.

Mediation analysis of systemic immune inflammation index in the association between weight-adjusted waist index and serum α-Klotho levels. The models were adjusted for age, sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes. IE indirect effect, DE direct effect, TE total effect; Proportion of mediation = IE/(DE + IE).

Additionally, our stratified analysis by age categories revealed varying sizes of mediation effects of SII on the relationship between WWI and α-Klotho levels. Specifically, among participants aged 60 years and older, SII played a more significant mediating role in the relationship between WWI and α-Klotho concentrations, accounting for 26.3% of the effect, compared to 4.2% for participants younger than 60 years (Fig. 5).

Fig. 5.

Mediation analysis of systemic immune inflammation index in the association between weight-adjusted waist index and serum α-Klotho levels among participants < 60 years of age (A) and participants ≥ 60 years of age (B). The models were adjusted for sex, race, family income to poverty ratio, education, serum cotinine level, drinking, physical activity, diet quality, survey cycle, hypertension and diabetes. IE indirect effect, DE direct effect, TE total effect; Proportion of mediation = IE/(DE + IE).

Sensitivity analyses

After excluding individuals with serum cotinine levels ≥ 10 ng/mL (n = 1568), the sensitivity analysis yielded consistent results. For instance, statistically significant decrease in serum α-Klotho levels was observed as both WWI (percent change: -7.32; 95% CI: -10.43, -4.10) and SII (percent change: -1.51; 95% CI: -1.88, -1.15) increased following the exclusion of serum cotinine levels (Table S1 and Table S2). Furthermore, there were significant indirect “mediation” effects of WWI on α-Klotho levels mediated through SII (percent change: -1.18; 95% CI: -1.64, -0.67), with SII accounting for approximately 15.7% of the total effect (Figure S1).

Similarly, the association of WWI with serum α-Klotho levels (percent change: -7.71; 95% CI: -10.69, -4.64), WWI with SII (percent change: 13.08; 95% CI: 7.29, 19.17), and SII with serum α-Klotho levels (percent change: -1.26; 95% CI: -1.62, -0.90) was found in adjusted models by excluding participants with any liver condition, weak/failing kidneys, or cardiovascular disease (n = 1372) (Table S3, Table S4, and Table S5). SII played a small but significant mediating role in the relationship between WWI and α-Klotho concentrations (percent change: -0.03; 95% CI: -0.05, -0.02), accounting for 8.35% of the total effect (Figure S2).

Discussion

In the present study, we examined the relationships between WWI and α-Klotho levels, further exploring the potential mediating role of SII in the linkage between obesity and serum α-Klotho levels within the NHANES survey cycle of 2011–2016. Our findings revealed a significant association between elevated WWI and reduced serum α-Klotho concentrations, with SII emerging as a crucial mediator in this association, particularly among older individuals. As far as we know, this is the first study to investigate the relationship between WWI and serum α-Klotho levels.

We found that the correlation between elevated WWI and lower serum α-Klotho concentrations. Previously, inconsistent results have reported in the association of the obesity with α-Klotho levels. We hypothesize that these inconsistencies may stem from variations in the methods used to assess obesity. To date, various metrics have been employed to measure obesity, including BMI^{14,17,18,49,50}, WC^15,50,51 and the visceral adiposity index¹⁶. Interestingly, our research was consistent with the majority of studies that have reported a negative association between obesity and α-Klotho concentrations ^{14–16,49−51}. Notably, the exceptions to this trend were studies that utilized BMI as the measure of obesity^17,18. Some experts suggest that BMI, while commonly used, may not provide an accurate assessment of obesity due to its inability to account for differences in lean body mass and fat mass¹⁹. In fact, Amaro-Gahete et al.¹⁷ demonstrated that the apparent positive association between BMI and α-Klotho concentrations disappeared after adjusting for lean mass index, a more precise measure of lean body mass. Therefore, given its comprehensive consideration of abdominal fat accumulation, WWI, along with WC and VAI, may be recognized as a superior indicator of obesity in its association with serum α-Klotho levels.

We discovered that the SII, an indicator of overall inflammatory status, serves as a mediator in the relationship between obesity and α-Klotho concentrations in present study. Although the precise mechanism underlying the link between obesity and decreased klotho levels remains elusive, we speculate that inflammation might play a pivotal role in this association. A cross-sectional study conducted by Lewitt et al.⁵² revealed a correlation between the accumulation of abdominal fat and a low-grade rise of inflammatory markers. Furthermore, Hamdy et al.⁵³ reviewed literature indicating that adipose tissue, especially visceral fat, produces pro-inflammatory cytokines such as interleukin-6 and tumor necrosis factor-alpha (TNF-α). Recent research suggests that both local and systemic inflammation can suppress klotho expression in the kidneys. This downregulation can occur through the induction of specific cytokines like TNF-α or via the epigenetic silencing of Klotho transcription triggered by inflammatory cytokines⁵⁴. Additionally, Ma et al.⁵⁵ demonstrated that chronic inflammation, which can be assessed through a pro-inflammatory dietary pattern, can reduce plasma α-Klotho levels. Consequently, as a feature of the state of obesity, low-grade elevation of inflammatory, primarily from fat tissue, may contribute to decreased serum α-Klotho levels.

Furthermore, our mediation analysis revealed age-stratified results indicating that the mediating effect of SII on the relationship between WWI and α-Klotho levels is significantly larger in older individuals compared to middle-aged participants. Two potential mechanisms may explain this observed mediation. Firstly, the impact of obesity on systemic inflammation may become more pronounced with age. This hypothesis is supported by Frasca et al., who suggested that the combination of obesity and aging significantly elevates low-grade chronic inflammation through immune cells infiltrating adipose tissue⁵⁶. Consistent with this, our study also demonstrated a significantly greater effect of obesity on inflammation in individuals aged 60 years and above compared to those younger than 60 (Fig. 3). Secondly, aging may exacerbate the effects of systemic inflammation on the aging marker α-Klotho levels. Recently, numerous shared molecular mechanisms linking systemic inflammation and aging have been discovered⁵⁷. Similarly, our findings revealed a stronger association between SII and α-Klotho levels in participants over 60 years old compared to those under 60 (Table 6). Finally, the increase in systemic inflammation with age may be attributed to immune system dysfunction and a higher predisposition to mild chronic inflammation in older individuals.

Our research holds several distinct advantages. Firstly, to our knowledge, this marked the inaugural study to demonstrate that the linkage between WWI and human serum levels of α-Klotho was mediated by systemic inflammation. This revelation offers insights into the mechanisms underlying resistance to aging or its postponement, encompassing the maintenance of a healthy weight and the mitigation of chronic inflammation. Secondly, we capitalized on a relatively extensive sample size of 6997 individuals sourced from the NHANES, allowing for more robust and reliable estimates of our findings.

However, it is imperative to acknowledge that our research is not without limitations. A primary constraint lies in the inability to establish a causal relationship between WWI and serum concentrations of α-Klotho due to the inherent limitations of a cross-sectional study design. Specifically, while a decrease in α-Klotho levels may contribute to obesity, it is conceivable that visceral adiposity may conversely trigger a reduction in α-Klotho levels.

Conclusion

Utilizing a nationally representative sample, this study has revealed a negative correlation between WWI and serum α-Klotho levels among adult Americans. Furthermore, our findings suggest that SII serves as a mediating factor in the association between WWI and α-Klotho levels, particularly among older individuals. These outcomes hint at potential anti-aging and health benefits derived from the meticulous management of obesity and systemic inflammation through the elevation of serum α-Klotho concentrations, especially in the elderly population. In the future, large-scale, meticulously planned prospective studies was needed to confirm the causal link and the mediating role of systemic inflammation.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Chenchun Chen: Conceptualization, Methodology, Writing - Original draft preparation, Writing - Review and Editing, Visualization; Peng Tang: Conceptualization, Visualization, Formal analysis, Investigation, Methodology, Writing – Review & Editing; Wei Zhu: Supervision, Conceptualization, Methodology, Writing - Review and Editing.

Funding

This study was supported by Basic Research Priorities Program of Guangzhou (2024A03J0558).

Data availability

The data used in this study are publicly available online (https://wwwn.cdc.gov/nchs/nhanes/).

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The portions of this study involving human participants, human materials, or human data were conducted in accordance with the Declaration of Helsinki and were approved by the NCHS Ethics Review Board. The patients/participants provided their written informed consent to participate in this study.