Mediation of blood lipid levels on the relationship between BMI/WHR and liver function in normal, overweight, and obese individuals

Abstract

This study investigates the impact of Body Mass Index (BMI) and Waist-to-Hip Ratio (WHR) on Alanine Aminotransferase (ALT) and Aspartate Aminotransferase (AST) levels, as well as the mediating role of lipid parameters: total cholesterol, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol. Understanding these associations is crucial for exploring the interplay between obesity, fatty liver disease, liver function impairment, and dyslipidemia. Conducted in Nantong City, Jiangsu Province, from July to September 2024, this research used a convenience sampling method involving 1,816 participants. Generalized Linear Models (GLM) and Restricted Cubic Splines (RCS) analyzed BMI/WHR’s relationship with ALT/AST, while causal mediation analysis assessed lipid parameters’ mediating effects. The findings indicated a positive correlation between both BMI and WHR with AST and ALT levels. The lipid indicators significantly mediated the relationship between BMI/WHR and AST/ALT, suggesting that obesity correlates with elevated liver enzyme levels and that blood lipid abnormalities may play an important intermediary role. This study enhances the understanding of interactions among obesity, fatty liver disease, liver function impairment, and dyslipidemia. It underscores the importance of addressing lipid abnormalities in managing obesity-related liver disease and calls for future research to explore underlying mechanisms and develop targeted interventions.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-22211-y.

Introduction

In assessing liver function, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are critical indicators for diagnosing liver health. When hepatocytes experience damage or undergo necrosis, serum levels of these enzymes rise significantly, providing vital information that helps physicians evaluate the patient’s condition¹. In clinical practice, measuring AST and ALT has become a standard method for liver function assessment, enabling timely identification and management of liver-related issues. It is important to note that while AST is found in various tissues throughout the body, ALT is predominantly concentrated in the liver, which enhances its specificity in diagnosing liver diseases². Over the past few decades, the AST to ALT ratio (commonly referred to as the De Ritis Ratio, or DRR), initially proposed by Fernando De Ritis in 1957³, has been recognized as a significant predictor for liver diseases, including hepatitis and liver cancer^4–6. Moreover, recent studies have established connections between elevated AST and ALT levels and several vascular disorders, such as peripheral arterial occlusive disease⁷, acute myocardial infarction⁸, and acute ischemic stroke⁹. Therefore, gaining a comprehensive understanding of the role of AST and ALT in liver function assessment is essential for enhancing the diagnostic accuracy of liver diseases and related conditions.

It is essential to recognize the impact of Body Mass Index (BMI), waist-to-hip ratio (WHR), and dyslipidemias on the levels of AST and ALT. According to Slywitch’s study¹⁰, there is a clear correlation whereby increasing BMI is associated with elevated ALT and AST levels in the general population. Similarly, Villegas’s research¹¹, which examined 3,978 urban Chinese men aged 40 to 74 years, found direct associations between heightened BMI and WH and increased levels of ALT and AST. Additionally, dyslipidemias can lead to complications affecting the endocrine system, central nervous system, liver, and renal systems¹². The liver, closely linked to lipid metabolism, is particularly susceptible to the adverse effects of dyslipidemias. Consequently, abnormalities in lipid levels can disrupt liver metabolism and potentially cause damage to liver tissue¹³.

Studying the key biological mechanisms linking BMI and WH to AST and ALT is crucial for understanding the essential characteristics of BMI and WH. Dyslipidemias have been identified as both outcomes of BMI and WH and as risk factors for elevated AST and ALT levels^14,15. This suggests that dyslipidemias may serve as mediators in the biological mechanisms connecting BMI and WH to AST and ALT. Causal mediation analysis, a valuable tool in clinical epidemiological research^16–19, allows for the decomposition of total effects into direct and indirect effects. It is particularly useful in complex analytical scenarios that account for potential interactions between exposures and mediators²⁰. Therefore, employing causal mediation analysis is vital for examining the mediating effect of dyslipidemias in the relationship between BMI/WH and the risk of elevated AST/ALT, especially while controlling for confounding effects to ensure the robustness of the mediation analysis.

In this study, we conducted a cross-sectional survey within a hospital-based population to investigate the effects of BMI and WH on ALT and AST in individuals categorized as normal weight, overweight, and obese. We also assessed the potential mediating roles of four lipid parameters: total cholesterol, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol. To explore the relationships between BMI/WH and ALT/AST and to establish exposure-response (E-R) relationships, we employed Generalized Linear Models (GLM) and Restricted Cubic Splines (RCS). Furthermore, we conducted causal mediation analysis to evaluate the mediating effects of these four lipid parameters on the increase in AST/ALT induced by BMI/WH.

Methods

This study was conducted in Nantong City, Jiangsu Province, from July to September 2024. We utilized a convenience sampling approach, specifically inviting individuals undergoing physical examinations at the Nantong People’s Hospital Examination Center to participate in our questionnaire survey. Participants were personally approached by our professional staff and thoroughly informed about the research objectives. Participation was entirely voluntary, and explicit informed consent was obtained from all participants prior to data collection. The questionnaire was designed to be concise, taking approximately 5 min to complete. As a token of appreciation for their valuable time, each participant who completed the questionnaire received a small gift—such as a notebook, water bottle, or snacks—valued between 10 and 15 yuan. The research protocol received formal approval from the Medical Ethics Committee of Nantong People’s Hospital on [number 2024KT370]. Ultimately, out of 2,272 invited participants, we successfully collected 2,067 valid surveys, achieving an impressive response rate of 91%.

The questionnaire content is comprehensive and detailed, encompassing not only basic demographic information such as sex, age, education level, monthly income, marital status, and residence but also meticulously recording key anthropometric measurements including weight, height, waist circumference, and hip circumference. Additionally, we assessed biochemical indicators such as blood lipid levels (including important markers like total cholesterol and triglycerides), AST, and ALT, providing a holistic view of the participants’ health status. We calculated the BMI and waist-to-hip ratio for each participant, enabling a more thorough assessment of their health. According to international health standards, a BMI between 18.5 and 23.9 kg/m² is classified as normal. A BMI below 18.5 kg/m² may indicate underweight, while a BMI exceeding 23.9 kg/m² suggests overweight. Notably, a BMI of 28 kg/m² or higher is generally defined as obesity. For the waist-to-hip ratio, the normal range for males is between 0.85 and 0.95, while for females, it falls between 0.67 and 0.80.

In this study, our research focused on normal, overweight, and obese populations, necessitating specific data screening. We excluded 159 individuals due to missing BMI information and an additional 92 individuals with a BMI below 18.5 kg/m². Following this screening process, the final sample size for the study was determined to be 1,816 participants.

Statistical analysis

For continuous variables, we summarized the data using the mean ± standard deviation (SD), while categorical variables were represented as frequency (percentage, %). To visually illustrate the distribution of ALT (alanine aminotransferase) and AST (aspartate aminotransferase) among participants, we fitted sample density curves, presenting them separately for BMI groups and waist-to-hip ratio groups. Additionally, we employed analysis of variance (ANOVA) to determine whether the differences in AST and ALT levels across the BMI groups and waist-to-hip ratio groups were statistically significant.

To investigate the impact of BMI and WH on AST and ALT, we employed multiple linear regression models. Initially, we constructed a crude model to examine the relationships among BMI, WH, AST, and ALT. Subsequently, we incorporated general demographic characteristics into the model. Finally, we controlled for blood lipid-related indicators, including Triglycerides (TG), High-Density Lipoprotein Cholesterol (HDLC), Low-Density Lipoprotein Cholesterol (LDLC), and Total Cholesterol (TC). Additionally, we utilized restricted cubic splines (RCS) with four knots to establish the exposure-response (E-R) relationships between these indicators (BMI, WH, AST, ALT, TG, HDLC, LDLC, TC).

To explore the mediating role of blood lipid indicators—specifically triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and total cholesterol—in the associations between BMI, WH, and AST and ALT, we conducted a causal mediation analysis. This allowed us to assess the proportion of the mediation effect attributed to these blood lipid indicators.

Additionally, we conducted sensitivity analyses based on gender and removing extreme values for our key variables (BMI, WHR, AST, and ALT).

Statistical analyses were conducted using R (Version 4.3.1). The linear model was implemented using the “rms” package, while restricted cubic splines (RCS) were visualized with the “plotRCS” package. The mediation effect model was executed utilizing the “mediation” package. A two-sided P-value of less than 0.05 was considered statistically significant.

Results

Descriptive statistics

Table 1 provides a detailed overview of the participants’ characteristics. A total of 1816 adults were recruited for this study, with an average age of 42.79 ± 13.94 years and an average BMI of 24.48 ± 3.44 kg/m². Among the participants, males were in the majority, totaling 1159 individuals and accounting for 63.8% of the sample. In terms of educational background, approximately half of the participants held a bachelor’s degree or above, with 877 individuals representing 48.3% of the total. Regarding monthly income, the largest proportion of participants, 400 individuals (22.0%), fell within the 6001–8000 RMB range. As for marital status, 1471 participants were married, accounting for 81.0% of the sample. In terms of residence, the majority of participants, 1342 individuals (73.9%), resided in Chongchuan District. Additionally, 653 participants had a normal waist-to-hip ratio (36.0%), while 984 had an increased WHR (54.2%); data for 179 participants were missing in this regard.

Table 1.

Socio-demographic characteristics of the study participants (N = 1816).

Characteristic	Number	Percent (%)
Age (years)	42.79 ± 13.94
Sex
Women	657	36.2
Men	1159	63.8
Education level
Primary school graduated or below	88	4.8
Junior high school graduated	281	15.5
High school or technical secondary school graduated	281	15.5
Junior college graduated	289	15.9
Bachelor’s degree or above	877	48.3
Monthly income (Yuan)
2000 and lower	165	9.1
2001–4000	227	12.5
4001–6000	378	20.8
6001–8000	400	22.0
8001–10,000	310	17.1
Over 10,000	336	18.5
Marital status
Unmarried	309	17.0
Married	1471	81.0
Others	36	2.0
Residence
Chongchuan District	1342	73.9
Tongzhou District	126	7.0
Haimen District	51	2.8
Others	287	16.3
BMI	24.48 ± 3.44
Normal	853	47.0
Overweight	703	38.7
Obesity	260	14.3
Waist-to-hip ratio	0.87 ± 0.07
Normal	653	36.0
Increased	984	54.2
Missing	179	9.9

Sample density curves revealed that AST and ALT levels were higher in obese and overweight individuals compared to the normal population (Fig. 1a). The results of the analysis of variance (ANOVA) were as follows: AST (F = 51.89, p < 0.001) and ALT (F = 128.11, p < 0.001). Figure 1b demonstrates that AST and ALT levels were higher in individuals with an increased WHR compared to those with a normal WHR. The ANOVA results were: AST (F = 34.00, p < 0.001) and ALT (F = 57.81, p < 0.001).

Fig. 1.

(a) Sample density curves of AST and ALT among normal, overweight, and obesity adults. (b) Sample density curves of AST and ALT among normal, elevated adults.

Association of BMI and waist-hip ratio with ALT and AST levels

Table 2 presents the results of the linear regression analysis examining the relationships between BMI and WH with levels of ALT and AST. Positive associations were observed across all models. In the adjusted Model 2, BMI demonstrated a significant association with AST (β = 0.606, P < 0.001) and ALT (β = 2.043, P < 0.001). Similarly, WH was associated with AST (β = 28.32, P < 0.001) and ALT (β = 88.57, P < 0.001). Notably, the changes in coefficient sizes from adjusted Model 1 to Model 2 indicate that lipid indices played a significant role in these associations.

Table 2.

Linear regression analysis examining the association of BMI and WHR with AST and ALT Levels.

Variables	AST		ALT
	β	P	β	P
BMI
Crude model	0.778	< 0.001	2.55	< 0.001
Adjust model 1	0.721	< 0.001	2.354	< 0.001
Adjust model 2	0.656	< 0.001	2.043	< 0.001
WH
Crude model	31.09	< 0.001	95.02	< 0.001
Adjust model 1	34.60	< 0.001	115.82	< 0.001
Adjust model 2	28.32	< 0.001	88.57	< 0.001

Crude model, without adjustment. Adjusted model 1, adjusted for sex, age, education level, monthly income, marital status, residence.; Adjusted model 2, adjusted for sex, age, education level, monthly income, marital status, residence, TG, HDLC, LDLC, TC. ALL variables (VIF < 10).

The exposure-response (E-R) relationships between BMI, WHR, and AST/ALT levels (Figs. 2 and 3)

Fig. 2.

(a) Association between BMI and AST using a restricted cubic spline regression model. Graphs show β for AST according to BMI adjusted for sex, age, education level, monthly income, marital status, residence. Data were fitted by a linear regression model, and the model was conducted with 4 knots at the 5th, 35th, 65th, 95th percentiles of BMI (reference is the 5th percentile). Solid lines indicate β, and shadow shape indicate 95% CIs. CI, confidence interval. (b) Association between BMI and ALT using a restricted cubic spline regression model. Graphs show β for ALT according to BMI adjusted for sex, age, education level, monthly income, marital status, residence. Data were fitted by a linear regression model, and the model was conducted with 4 knots at the 5th, 35th, 65th, 95th percentiles of BMI (reference is the 5th percentile). Solid lines indicate β, and shadow shape indicate 95% CIs. CI, confidence interval.

Fig. 3.

(a) Association between WH and AST using a restricted cubic spline regression model. Graphs show β for AST according to WH adjusted for sex, age, education level, monthly income, marital status, residence. Data were fitted by a linear regression model, and the model was conducted with 4 knots at the 5th, 35th, 65th, 95th percentiles of WH (reference is the 5th percentile). Solid lines indicate β, and shadow shape indicate 95% CIs. CI, confidence interval. (b) Association between WH and ALT using a restricted cubic spline regression model. Graphs show β for ALT according to WH adjusted for sex, age, education level, monthly income, marital status, residence. Data were fitted by a linear regression model, and the model was conducted with 4 knots at the 5th, 35th, 65th, 95th percentiles of WH (reference is the 5th percentile). Solid lines indicate β, and shadow shape indicate 95% CIs. CI, confidence interval.

Figure 2 presents the E-R function illustrating the relationships between BMI and AST/ALT levels, while Fig. 3 depicts the E-R function for WHR in relation to AST/ALT. In details, Fig. 2a and b elaborate on the E - R relationships between BMI and the results of ALT and AST. The results show that ALT and AST levels exhibit a linear increase with increasing BMI (P for overall < 0.05). In terms of the shape of the E - R functions, both the relationships between BMI - AST and BMI - ALT demonstrate nearly linear trends (P for non - linear > 0.05), indicating that the non - linear relationships are not statistically significant.

Figure 3a and b elaborate on the E - R relationships between WHR and the results of ALT and AST. The results reveal that ALT and AST levels also show a linear increase with increasing WHR (P for overall < 0.05). The shape of the E - R functions indicates that the relationship between WH and AST is nearly linear (P for non - linear > 0.05), meaning that the non - linear relationship is not statistically significant. However, the E - R function for WH and ALT levels appears to be non - linear (P for non - linear < 0.05) (as shown in Fig. 3b).

This study further explored the E - R associations between BMI, WH, and TG, HDLC, LDLC, and TC) (Figures S1- S8), as well as between TG, HDLC, LDLC, TC, and AST/ALT levels (Figures S9 – S16). The results showed that with increasing BMI and WH, TG, LDLC, and TC mostly exhibited linear increases, while HDLC mostly showed non - linear relationships. AST and ALT levels mostly increased linearly with rising TG, LDLC, and TC, and some had non - linear relationships with TG and LDLC.

The mediating effects of lipid profiles (TG, HDLC, LDLC, TC) in the relationships between BMI/WH and AST/ALT levels (Figs. 4, 5, 6 and 7)

Fig. 4.

Mediation effects of TG, HDLC, LDLC, TC in the associations of BMI with AST. ACME, average causal mediation effects (indirect effect); ADE, average direct effects. * P < 0.05, ** P < 0.01, and *** P < 0.001.

Fig. 5.

Mediation effects of TG, HDLC, LDLC, TC in the associations of BMI with ALT. ACME, average causal mediation effects (indirect effect); ADE, average direct effects. * P < 0.05, ** P < 0.01, and *** P < 0.001.

Fig. 6.

Mediation effects of TG, HDLC, LDLC, TC in the associations of WH with AST. ACME, average causal mediation effects (indirect effect); ADE, average direct effects. * P < 0.05, ** P < 0.01, and *** P < 0.001.

Fig. 7.

Mediation effects of TG, HDLC, LDLC, TC in the associations of WH with ALT. ACME, average causal mediation effects (indirect effect); ADE, average direct effects. * P < 0.05, ** P < 0.01, and *** P < 0.001.

In exploring the relationship between BMI and AST levels, we found that TG, LDLC, and TC played a mediating role, with mediation proportions of 9% (95% CI: 4%-15%), 5% (95% CI: 1%-10%), and 3% (95% CI: 1%-6%) respectively (Fig. 4). In the relationship between BMI and ALT levels, we also observed the mediating effects of TG, HDLC, LDLC, and TC, with mediation proportions of 8% (95% CI: 4%-12%), 5% (95% CI: 1%-9%), 6% (95% CI: 3%-10%), and 3% (95% CI: 1%-5%) respectively (Fig. 5). When studying the relationship between WH and AST levels (Figs. 6 and 7), we found that TG, LDLC, and TC exhibited mediating effects, with mediation proportions of 14% (95% CI: 7%-23%), 8% (95% CI: 1%-15%), and 4% (95% CI: 1%-8%) respectively. Similarly, in the relationship between WH and ALT levels, TG, HDLC, LDLC, and TC also demonstrated mediating effects, with mediation proportions of 14% (95% CI: 9%-20%), 10% (95% CI: 6%-16%), 10% (95% CI: 6%-16% ), and 4% (95% CI: 2%-6%) respectively.

We also conducted sensitivity analyses based on gender and removing extreme values for our key variables. The results showed that when gender was taken into account, the regression outcomes remained consistent. Specifically, BMI and WH continued to have a significant impact on ALT and AST (See Table 1s). These findings further validate the robustness of our research, indicating that within the context of our study, gender factors did not interfere with the intrinsic relationships between BMI, WHR, ALT, and AST.

Discussion

In our study population, categorized by normal BMI, overweight, and obesity, we observed a positive correlation between BMI, WHR, and increased risks of liver function indicators, specifically AST and ALT. We established significant E-R relationships between these anthropometric measures (BMI and WHR) and the risks associated with both AST and ALT levels. Furthermore, we found that four indices of blood lipids mediated the associations between BMI and AST, with mediation proportions ranging from 3% to 9%, as well as between BMI and ALT, which ranged from 3% to 8%. Similarly, these blood lipid indices mediated the relationships between WHR and AST, with mediation proportions from 4% to 14%, as well as between WHR and ALT, also ranging from 4% to 14%. This significant finding not only enhances our understanding of the intricate interactions among obesity, fatty liver disease, liver function impairment, and dyslipidemia but also provides a solid scientific foundation for potential intervention strategies in the prevention and treatment of related metabolic diseases.

Our study findings revealed a positive correlation between both BMI and WHR with AST and ALT levels. This observation aligns with established research, reinforcing the notion that excess adiposity is intricately connected to alterations in liver enzyme levels. In Leticia’s work²¹, the results indicated that overweight and obese children exhibited significantly higher concentrations of ALT, with a prevalence rate of 22%, compared to their normal-weight counterparts who showed a prevalence of 0%. Elevated ALT levels are a common manifestation of liver dysfunction in overweight and obese adolescents^22,23. AST and ALT serve as crucial biomarkers for assessing liver function; their elevation may indicate hepatocyte damage or the presence of inflammation. BMI and WH, widely employed metrics for obesity assessment, have frequently been linked to various health issues, including liver diseases¹¹. The findings of this study further substantiate the well-documented association between obesity and liver dysfunction. In obese individuals, the accumulation of adipose tissue can lead to insulin resistance, the development of fatty liver, and inflammatory responses, ultimately resulting in liver damage. Therefore, increased BMI and WHR may serve as potential risk factors for predicting liver dysfunction.

Our study successfully established significant E-R relationships between anthropometric measures (BMI and WHR) and the risks associated with elevated AST and ALT levels. The analysis revealed a corresponding linear increase in both ALT and AST levels with rising BMI and WHR, which was statistically significant (overall P < 0.05) (Figs. 2a and b and 3a and b). This finding underscores the direct association between obesity, quantified by BMI and WHR, and liver enzyme abnormalities, consistent with previous studies²⁴. Upon closer examination of the E-R functions, we observed approximately linear trends in the relationships between BMI and AST, BMI and ALT, as well as WHR and AST (non-linear P > 0.05) (Figs. 2a and b and 3a). These observations suggest that for these specific pairings, the risk of liver enzyme elevation increases relatively consistently with each unit increase in BMI or WHR. In contrast, the E-R function between WHR and ALT levels exhibited a non-linear pattern (non-linear P < 0.05) (Fig. 3b). This deviation from linearity indicates that the relationship between central obesity, represented by WHR, and ALT levels may be more complex. It suggests the possibility of threshold effects or other non-linear dynamics, where certain levels of WHR may disproportionately elevate the risk of ALT increase. The observed non-linear E-R function between WHR and ALT highlights the significance of considering the specific nature of the relationship between obesity measures and liver enzyme levels. WHR serves as an indicator of abdominal fat accumulation²⁵. Adipokines secreted by this active endocrine organ can induce liver inflammation through mechanisms such as NF-κB activation, leading to hepatocyte damage and subsequent ALT release^26,27. When WHR surpasses a certain threshold, adipokine secretion may increase exponentially, resulting in a significant elevation of ALT levels. It underscores the need for further research to elucidate the underlying mechanisms driving these associations and to identify potential thresholds or inflection points where the risk of liver dysfunction may escalate more rapidly.

In the current study, our findings underscore the crucial mediating role of four blood lipid indices in the complex relationships between BMI, WHR, and liver function markers, specifically AST and ALT. The results suggest that changes in blood lipid levels may serve as a significant intermediary pathway through which obesity, as assessed by BMI and WH, influences liver health, as evidenced by variations in AST and ALT levels. Piotrowska’s research indicates that overweight and obesity, particularly abdominal obesity, are significant risk factors for disorders in lipid metabolism²⁸. Additionally, previous studies have shown that the incidence of dyslipidemia and liver enzyme abnormalities is higher in individuals with diabetes and hypertension compared to healthy individuals. Approximately 61% of participants with dyslipidemia exhibited elevated levels of one or more liver enzymes. Obesity is often associated with an abnormal lipid profile, including elevated cholesterol (CHOL) and low-density lipoprotein (LDL) levels. These lipid abnormalities are reflective of elevated liver transaminases, which increase the risk of cirrhosis among obese individuals^29,30. Therefore, managing blood lipid levels could potentially mitigate some of the adverse effects of obesity on liver function. This emphasizes the importance of comprehensive lifestyle interventions that target both weight management and lipid control in promoting overall liver health.

Several limitations of this study should be acknowledged. Firstly, the primary limitation stems from the cross-sectional design, which restricts our ability to draw causal conclusions. The observed associations and mediating effects only suggest potential relationships that require further validation in longitudinal or experimental studies. Secondly, the absence of behavioral data—such as smoking habits, alcohol consumption, and exercise duration—represents another limitation. These factors may act as significant confounding variables, potentially introducing biases into the findings. However, due to practical research constraints, we were unable to include these factors in our study. To mitigate the impact of not considering these factors on the results, we conducted sensitivity analyses using gender as a surrogate variable. The results showed no substantial changes in the study findings, which to some extent ensures the reliability of the results. But we are fully aware of the limitations of this approach. In future research, we will make efforts to overcome the existing difficulties and incorporate these key lifestyle factors into the study to improve its comprehensiveness and accuracy, and also offer reference directions for subsequent related research. Lastly, this study exclusively recruited participants from the physical examination population at Nantong People’s Hospital, which may lead to selection bias. To increase the generalizability and representativeness of the findings, future research should aim for a larger, more diverse sample that includes individuals from various backgrounds and characteristics.

Conclusion

In conclusion, our study offers compelling evidence of a positive correlation between BMI, WH, and increased risks associated with liver function indicators, specifically AST and ALT levels. We successfully established significant E-R relationships between these anthropometric measures and liver enzyme levels, highlighting the direct association between obesity and liver dysfunction. Moreover, our findings emphasize the critical mediating role of blood lipid indices in these relationships, suggesting that alterations in blood lipids may serve as a significant intermediary pathway linking obesity to liver health. The observed non-linear E-R relationship between WH and ALT levels further underscores the complexity of these associations, indicating the necessity for additional research to elucidate underlying mechanisms and identify potential thresholds. Overall, this study enhances our understanding of the intricate interactions among obesity, fatty liver disease, liver function impairment, and dyslipidemia. It provides a solid scientific foundation for potential intervention strategies aimed at promoting overall liver health through comprehensive lifestyle interventions focused on both weight management and lipid control.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AST

Aspartate aminotransferase

ALT

Alanine aminotransferase

TG

Triglycerides

HDLC

High-density lipoprotein cholesterol

LDLC

Low-density lipoprotein cholesterol

TC

Total cholesterol

BMI

Body mass index

WH

Waist-to-hip ratio

Author contributions

CX Z, XF C analyzed the data and wrote manuscript; CX Z, ZZ D and XF C edited the manuscript. All authors read and approved the final manuscript.

Funding

This research was approved by: Jiangsu Provincial Hospital Association Hospital Management Innovation Research Project (JSYGY-2024-238); Nantong Science and Technology Plan Project (MSZ2024135); Nantong Municipal Health Commission Science and Technology Plan Project (MS2024031); High-level Talent Cultivation Fund of Nantong First People’s Hospital (YPYJJZD24006).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Participants provided written informed consent, and the Medical Ethics Committee of Nantong People’s Hospital on [number 2024KT370] approved this study. And all methods were performed in accordance with t Declaration of Helsinki.

Competing interests

The authors declare no competing interests.