Non-alcoholic fatty liver disease (NAFLD) is defined as the excessive accumulation of fat in the liver in the absence of significant alcohol consumption. It is often regarded as the hepatic manifestation of metabolic syndrome and is commonly associated with metabolic disorders such as obesity, type 2 diabetes, and hyperlipidemia (1). In recent years, the incidence of NAFLD has increased, surpassing that of viral hepatitis to become the predominant chronic liver disease globally (2). The pathogenesis of NAFLD is complex and ranges from simple fatty liver, characterized by excess fat in the liver without significant inflammation or fibrosis, to non-alcoholic steatohepatitis (NASH), which not only involves fat accumulation but also accompanies liver cell inflammation and damage, ultimately leading to liver fibrosis (LF) (3). LF is a key pathological feature in the progression of NAFLD and a major risk factor for the development of cirrhosis and hepatocellular carcinoma. In recent years, an increasing number of studies have focused on the epidemiology of LF caused by NAFLD, with results indicating that the prevalence of LF significantly increases with the severity of NAFLD (4, 5). Therefore, it is crucial to promptly identify the risk factors for LF in patients with NAFLD.

Insulin resistance (IR) is a well-recognized factor in the pathogenesis of NAFLD and plays a critical role in its progression. IR leads to an imbalance in lipid metabolism, promoting hepatic fat accumulation and contributing to liver inflammation and fibrosis (6). Given the close relationship between IR and NAFLD, indicators of IR, such as fasting blood glucose (FBG), fasting insulin, and the homeostasis model assessment of insulin resistance (HOMA-IR), have been widely used as biomarkers to assess metabolic dysfunction in patients with NAFLD (7, 8). In addition to traditional markers of IR, the triglyceride-glucose index (TyG) has drawn increasing attention in recent years. The TyG index is a calculated measure based on fasting triglycerides and FBG. Due to its simplicity, ease of availability, and strong correlation with IR, it has been widely utilized for assessing IR and cardiovascular disease risk (9, 10). Moreover, the indicators combining TyG with body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHtR) further enhance the assessment of an individual’s metabolic status and have been shown to be closely associated with the presence and severity of NAFLD (11). Although the association between NAFLD and IR is well-established, the exact relationship between various IR indicators (including the TyG index) and the degree of LF in NAFLD remains unclear. Most previous studies have primarily focused on the presence of NAFLD and its progression to NASH, with comparatively less attention given to the specific correlation between these IR indicators and the degree of hepatic fibrosis in NAFLD patients (12, 13). A deeper exploration of the association between the TyG index and HOMA-IR with LF in NAFLD patients will enhance our understanding of the metabolic mechanisms underlying the disease and provide new insights for early risk assessment.

Histopathological examination of liver biopsy specimens has long been considered the gold standard for diagnosing NAFLD and LF. Nonetheless, this method presents several limitations, including its invasive nature, low acceptability, and high cost (14). In recent years, liver ultrasound transient elastography (LUTE) has emerged as an accurate and non-invasive technique for assessing the degree of steatosis and fibrosis in patients with NAFLD (15). A meta-analysis found that LUTE exhibits good sensitivity and specificity for LF, with sensitivity and specificity values of 0.79 and 0.78, respectively (16). Previous research has focused on developing non-invasive diagnostic methods for LF. Several studies have developed serological models based on biochemical markers and clinical information to predict LF, including the fibrosis-4 index, aspartate aminotransferase (AST) to platelet ratio, AST to alanine aminotransferase (ALT) ratio, Forns index, and BARD score (17–19). However, when these scoring systems are used to predict LF in NAFLD patients, they do not include metabolic indicators such as the TyG index. The lack of these key metabolic markers may reduce the accuracy of the models, failing to fully reflect the fibrosis risk caused by NAFLD.

In our study, we aim to utilize data from the National Health and Nutrition Examination Survey (NHANES) database to assess NAFLD and LF using LUTE. We will then explore the correlation between LF and various IR indicators in NAFLD patients. Additionally, we will attempt to develop a predictive model for NAFLD-related LF based on these metabolic indicators. This study will provide valuable insights into the potential role of IR in the progression of LF and help identify key markers for early risk stratification and management, thereby enabling a more accurate prediction of fibrosis risk in NAFLD patients.

This study systematically investigated the relationship between various IR indices and LF among patients with NAFLD. Our results demonstrated that indices such as TyG-WHtR, TyG-BMI, and HOMA-IR were significantly associated with the occurrence of LF in NAFLD patients, with TyG-WHtR emerging as the most prominent predictor. Even after adjusting for a range of covariates, TyG-WHtR maintained a strong correlation, suggesting its potential utility as an independent predictor of LF in this patient population. Additionally, we developed a predictive model for LF in NAFLD patients, which highlights the potential of these indices to be incorporated into routine clinical practice for risk assessment and early intervention.

The findings of this study have significant implications for clinical practice, particularly in the early identification and management of LF in patients with NAFLD. The strong association between TyG-WHtR and LF underscores its potential as a simple, non-invasive tool for risk stratification in routine clinical settings. The TyG-WHtR is a non-invasive, simple, low-cost index that only tests TG, FPG, WC, and height to produce results. Compared to liver puncture biopsy, CT, and MRI, TyG-WHtR offers a superior cost-benefit ratio. By incorporating TyG-WHtR and other IR indices into standard metabolic assessments, clinicians can more effectively identify high-risk patients who may benefit from closer monitoring or early intervention. The predictive nomogram developed in this study, which integrates AST, TyG, TyG-BMI, and diabetes, provides a practical and accessible tool for individualized risk assessment. This approach is particularly valuable in resource-limited settings where advanced diagnostic tools may not be readily available. Clinicians can use this nomogram to estimate fibrosis risk using readily available clinical data, enabling targeted therapies such as lifestyle modifications, weight management, and insulin-sensitizing treatments to slow or prevent fibrosis progression (33, 34). These findings advocate for the integration of IR indices into routine clinical practice to enhance early detection, risk assessment, and personalized management of NAFLD-related fibrosis.

Our findings emphasize that obesity, as reflected by higher BMI and WC, and other metabolic factors, such as increased ALT, AST, GGT, and HbA1c levels, were more prevalent in NAFLD patients with LF compared to those without. The liver, an essential organ for metabolic processes, regulates the metabolism of both lipids and glucose. Chiang et al. reported that increased obesity and IR significantly contribute to the progression from NASH to fibrosis through the development of a profibrotic environment in the liver (35). Additionally, Koppe et al. reported that IR leads to widespread metabolic disturbances, resulting in a net effect of TG accumulation in the liver. Some patients may develop hepatocellular injury and LF, which can progress to cirrhosis (36). In comparison to previous research, Khamseh et al. identified TyG-WC, TyG-BMI, and TyG-WHtR as the best predictors of metabolic-associated fatty liver disease (37). Although their study did not establish a clear relationship between these indicators and LF, our research further confirms this link. We found that several IR markers, particularly TyG-WHtR, TyG-BMI, and HOMA-IR, were significantly elevated in LF patients, indicating that metabolic dysfunction plays a central role in the pathogenesis of the disease.

IR indices are not only widely applied in metabolic diseases such as type 2 diabetes and obesity but are also used in other conditions, including cardiovascular diseases, chronic kidney disease, and polycystic ovary syndrome, where they have also been shown to predict adverse outcomes (38–41). Several studies have demonstrated that IR triggers lipotoxic pathways in the liver, leading to an accumulation of toxic lipid species such as ceramides and diacylglycerol, which further exacerbate liver injury and fibrogenesis (42–44). TyG-WHtR showed a significant correlation with LF in this study. High levels of TyG-WHtR indicate severe visceral fat accumulation, which is a key factor in the progression of LF (45). This relationship is particularly significant in the context of NAFLD, where visceral fat plays a crucial role in metabolic dysfunction. The accumulation of visceral fat is linked to various metabolic complications, including increased liver fat content, which can exacerbate liver inflammation and fibrosis (46). Therefore, TyG-WHtR can serve as an independent predictor of LF risk in NAFLD patients, with potential clinical utility.

In recent years, the TyG index has been increasingly applied in liver diseases, especially in predicting the progression of NAFLD, showing a significant association with NAFLD and LF (37, 47). To better evaluate the combined effects of IR and obesity, TyG-BMI integrates BMI, which reflects overall body weight status, making it more advantageous in assessing metabolic risk (22, 48). Our study found that TyG-BMI levels were significantly elevated in LF patients, highlighting the core role of the synergy between obesity and IR in LF progression. Compared to single IR indicators, TyG-BMI provides a more comprehensive assessment of metabolic risk, offering important insights for early identification and intervention of LF. Additionally, in our study, we found that TyG-GGT is not an independent risk factor for LF in NAFLD patients. In contrast, Lei Jin et al. suggested that TyG-GGT has strong predictive accuracy for advanced LF in overweight or obese patients (25). However, their study was limited by a small sample size, a retrospective design that did not fully control for confounding factors, and a lack of strong statistical significance. Additionally, both METS-IR and HOMA-IR are strongly associated with LF, reflecting the relationship between IR and metabolic syndrome. Consistent with previous studies, our findings show that these indicators have strong predictive power for LF in NAFLD patients (49–51). Additionally, HOMA-IR is considered an independent predictor of advanced LF in non-diabetic NAFLD patients (50). It accelerates fibrosis progression by promoting liver fat accumulation, inflammatory responses, and hepatic stellate cell activation (52). Therefore, METS-IR and HOMA-IR can serve as effective tools in clinical practice for assessing the risk of fibrosis in NAFLD patients. In addition to our findings, several studies from China have also reported a strong association between IR and the progression of NAFLD, as well as its correlation with fibrosis staging (47, 53). By integrating these findings, our study contributes to a growing body of evidence that underscores the clinical utility of IR indices in predicting LF risk in NAFLD patients.

This study has several notable strengths. First, it focuses on a specific population of NAFLD patients with LF, making the results more targeted and clinically relevant, thereby providing important insights for the management of this high-risk group. Second, we systematically employed various analytical methods, including logistic regression and RCS, to comprehensively evaluate the relationship between multiple IR indices and LF, clarifying the predictive value of these indices. Finally, we developed a LF prediction model based on multivariable logistic regression and constructed a nomogram, providing a scientific and effective tool for early risk identification and individualized intervention in clinical practice, with high practical value.

Despite providing further evidence of the close relationship between IR and LF in NAFLD patients, our study has several limitations that warrant discussion. First, as this study is based on cross-sectional data, we cannot establish a causal relationship between IR and LF. Longitudinal studies are therefore needed to verify the causal role of IR in the progression of NAFLD. Second, the relatively small sample size and the fact that our cohort was limited to the U.S. population may limit the external validity of the findings, particularly across different ethnicities and regions. Future research should include larger cohorts from diverse populations to validate the applicability and predictive value of these IR indices in a broader context. Additionally, our study mainly focused on epidemiological associations and lacked an in-depth exploration of the underlying molecular mechanisms. Thus, future basic research should aim to elucidate how IR promotes LF through specific cellular signaling pathways, providing theoretical support for targeted interventions.

In summary, this study identifies TyG-WHtR, TyG-BMI, and other IR indices as independent predictors of LF in NAFLD patients, highlighting their clinical utility in early risk stratification. These findings underscore the importance of integrating metabolic indicators into routine clinical practice to enhance early detection and intervention. The predictive nomogram developed in this study offers a practical, non-invasive tool for clinicians to identify high-risk patients. By focusing on metabolic risk factors, clinicians can implement targeted therapies—such as lifestyle modifications and insulin-sensitizing treatments—to slow or prevent fibrosis progression, ultimately improving long-term patient outcomes.