Non-alcoholic fatty liver disease (NAFLD) is a significant global public health concern affecting over 25% of the world’s population (1). Histologically, NAFLD is classified into non-alcoholic fatty liver and non-alcoholic steatohepatitis (NASH) (2). NAFLD not only contributes to the development of type 2 diabetes and cardiovascular disease but may also progress to cirrhosis and hepatocellular carcinoma as the disease advances (3). Therefore, early detection and accurate prediction of NAFLD onset and progression are critical for reducing its health burden (4). Notably, recently, major liver societies such as the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL) have approved the naming of NAFLD as metabolic dysfunction associated fatty liver disease (MASLD) (5, 6). However, in this study, since our initial diagnosis and data collection were performed before the widespread adoption of MASLD criteria. To ensure that the previous terminology is clearly familiar to the reader, we have retained the term NAFLD throughout the text, while acknowledging the newer terminology and its clinical significance.

The triglyceride-glucose (TyG) index has been demonstrated to be a valuable marker for predicting and aiding in the diagnosis of metabolic-associated fatty liver disease (7). It is a reliable predictor of cardiovascular and all-cause mortality (8). As research advances, the TyG index is increasingly recognized for its simplicity and accessibility (9), and it has emerged as one of the most accurate predictors of NAFLD (10). Additionally, TyG-related indicators, such as TyG-waist circumference (TyG-WC) (11), TyG-body mass index (TyG-BMI) (12), and TyG-waist-height ratio (TyG-WHtR) (13), are strongly correlated with NAFLD and considered valuable for assessing and diagnosing various NAFLD stages (14). The gold standard for diagnosing and assessing NAFLD is costly, time-consuming, and invasive, limiting its practicality for widespread screening (15).

With the rapid aging of the global population, the proportion of individuals aged 60 years and older is increasing at an unprecedented rate. According to the World Health Organization, the number of people aged 60 and above is expected to rise from 1 billion in 2020 to 1.4 billion by 2030, and to 2.1 billion by 2050. This demographic shift poses significant public health challenges, particularly in the management of chronic metabolic diseases. Moreover, predicting the prevalence and risk factors of NAFLD in this population is challenging due to the complexity of age-related metabolic disorders (16). Although the Rotterdam Study suggested that NAFLD prevalence stabilizes with age (17), additional studies have consistently shown that NAFLD is more prevalent in older adults compared to younger individuals (18). This discrepancy is likely because most research includes broader age ranges, whereas research specifically targeting older populations remains limited. Additionally, invasive diagnostic techniques are unsuitable for elderly individuals. Previous meta-analyses have shown that non-invasive techniques such as vibration-controlled transient elastography (VCTE), point shear wave elastography (pSWE), 2-dimensional shear wave elastography (2DSWE), and magnetic resonance elastography (MRE) (19), as well as the FibroScan-AST (FAST) score (20), may achieve diagnostic accuracy comparable to liver biopsy in diagnosing NAFLD. However, their higher medical requirements and significant cost burdens make them difficult to widely promote in most regions of the world, especially for elderly populations who cannot undergo timely monitoring at qualified hospitals on a long-term basis. Moreover, the prevalence of NAFLD is growing at a rate far exceeding previous expectations, resulting in a heavy public health burden. Therefore, it is crucial to investigate noninvasive and easily accessible indicators for predicting and diagnosing NAFLD occurrence and progression. The TyG index, by integrating triglyceride and fasting glucose levels, reflects the mechanisms of insulin resistance. It can accurately assess the degree of insulin resistance and is closely related to metabolic disruptions in pathways such as fatty acid metabolism, glucose metabolism, and insulin signaling. This study aimed to explore the diagnostic value of the TyG index, TyG-WC, TyG-BMI, and TyG-WHtR in elderly individuals with NAFLD and to examine the correlation between changes in TyG values over 2 years and the progression of hepatic steatosis. Our findings may help establish practical, non-invasive predictors for risk stratification in an aging population.

In this study, we reached the following conclusions: (1) The TyG index and its related indices were highly correlated with NAFLD across different models. (2) The TyG index and its related indices demonstrated significant predictive value for both NAFLD and its varying degrees of severity. (3) Monitoring changes in the TyG index can provide insights into the progression or regression trends of NAFLD. The following sections present a detailed discussion of these findings.

Recent studies have demonstrated a strong correlation between the TyG index and related indices in NAFLD. Peng et al. suggested through their research that the TyG index is a crucial indicator of the risk of NAFLD (13). In a study involving 10,390 participants, the clinical value of the TyG index in predicting the risk of all-cause mortality in individuals with NAFLD was analyzed and confirmed. Furthermore, they highlighted its importance as a key indicator for monitoring the progression of NAFLD to liver fibrosis and underscored its role in follow-up and preventive strategies (27). A study involving 1,727 adults demonstrated that the TyG index and its related parameters can be effectively utilized for early screening and monitoring of disease progression in individuals with NAFLD (14). These studies collectively confirm the significance of the TyG index and related indices in identifying NAFLD pathology. Moreover, they demonstrated an independent positive association between TyG-BMI and the risk of NAFLD, offering new insights into the prevention and treatment of the disease (28). In this study, we aimed to explore the relationship between the TyG index and its related indices and the severity of NAFLD in an elderly population. Additionally, we sought to determine whether monitoring the changes in the TyG index could accurately assess NAFLD progression.

Fatty liver development is thought to result from the breakdown of triglycerides in adipose tissue into fatty acids, which are then transported to the liver through the bloodstream. This process is regulated by insulin on adipocytes. When insulin receptor signaling in adipose tissue is impaired, leading to insulin resistance, the balance of lipolysis is disrupted. Under these conditions, an excessive amount of fatty acids is transported to the liver (29). Insulin resistance (IR) is the primary cause of this imbalance (30). Since Simental-Mendia et al. (31) first introduced the concept of calculating the TyG index using fasting glucose and triglyceride levels, it has demonstrated significant efficacy in assessing insulin resistance, cardiovascular disease, all-cause mortality, and NAFLD (32). Emerging evidence suggests that NAFLD may be a poor prognostic factor for bladder cancer(BC), and the TyG index has shown high sensitivity in identifying individuals at risk of BC (33, 34). These findings highlight the broader clinical utility of the TyG index beyond liver-related outcomes. An increase in the TyG index often indicates impaired pancreatic β-cell function (35), reflecting not only the state of IR but also its close association with the onset and progression of various diseases (36). This, in turn, influences the development of NAFLD, potentially explaining the strong association observed between the TyG index and the occurrence and progression of NAFLD. For the elderly, this situation may be even more complex. Due to physiological decline and intricate immune responses, there may be additional factors influencing the development of fatty liver (37). For example, a significant decrease in mitochondrial oxidation and phosphorylation activity has led to enhanced insulin resistance in the elderly population, increasing their likelihood of developing NAFLD (38). Therefore, under such a complex set of risk factors, we aimed to investigate whether the correlation between the TyG index and NAFLD remains consistent with that reported in previous studies. Additionally, obesity indices, such as BMI, WC, and WHtR, are also important indicators for assessing NAFLD (11, 12). Therefore, combining TyG and obesity indices to derive a new parameter may enable a more accurate and sensitive assessment of NAFLD and its various stages. In our study, we simultaneously incorporated four parameters to summarize and compare the risk factors associated with NAFLD and predictive states of the disease.

Previous studies generally regarded middle-aged and elderly individuals as having a high prevalence of NAFLD (39). Based on these findings, age may influence the diagnostic value of the TyG index for fatty liver disease (9). However, research on the diagnostic value of TyG and its related indices in elderly populations remains limited. We investigated the role of TyG in NAFLD in 3,954 older adults. In this study, we established three models to confirm whether TyG levels and related indices are independent risk factors for NAFLD. Our findings indicate that even in the complex immunological landscape of the elderly, indices such as TyG can accurately reflect their correlation with NAFLD, providing evidence to support the clinical diagnosis of NAFLD in this age group. ROC curve analysis revealed that TyG and its related indices possess significant predictive value for different stages of NAFLD, particularly TyG-BMI, which demonstrated high predictive value across all three disease stages. This suggests that these indices may facilitate a more accurate and sensitive assessment of NAFLD status and severity in clinical practice. Although imaging techniques are widely used as the primary diagnostic approach for NAFLD worldwide (40), their accuracy can be compromised by confounding factors, such as chronic liver disease and liver fibrosis, which may interfere with the specific evaluation of disease status (41). The TyG index and related indices, when combined with imaging assessments, provide clinicians with valuable insights into the diagnosis and management of various diseases. Moreover, the predictive efficacy demonstrated by these four parameters can significantly enhance diagnostic capabilities. Numerous studies have confirmed that the TyG index can predict the prevalence of NAFLD; however, most of these studies were cross-sectional in nature, and the impact of changes in the TyG index on the varying outcomes of NAFLD is yet to be established. A meta-analysis of randomized controlled trials indicated that approximately 20% of patients with NAFLD may rapidly progress to advanced cirrhosis within a short period without exhibiting any clinical symptoms (42), However, the methods for identifying such rapid progression remain unclear. Therefore, we conducted a two-year follow-up study in this population to determine whether changes in the TyG index affect the progression of NAFLD. Over time, the progression of NAFLD results in various outcomes (43). These results indicate that as the TyG index increases, there is a progressive worsening trend in the severity of NAFLD, suggesting that monitoring changes in the TyG index can be a useful tool for assessing disease progression. This trend was particularly pronounced in groups with prior worsening and improvement of NAFLD, where an increase in the TyG index serves as a valuable risk indicator for predicting changes. In the future, combining TyG index with B-mode ultrasonography may help screen for high-risk populations with a potential progression of NAFLD, providing clinicians with more accurate non-invasive diagnostic tools, such as the FibroScan-AST (FAST) score and transient elastography. This approach could not only effectively reduce the waste of medical resources but also support early intervention, thereby improving patient outcomes.

With the increasing prevalence of NAFLD, it has become a significant cause of hepatocellular carcinoma (HCC) (44); however, there are currently no specific medications available for the treatment of NAFLD in clinical practice. Instead, management strategies primarily involve interventions such as dietary modifications and weight control to mitigate the effects of NAFLD (45). This treatment approach may be effective for early-stage NAFLD; however, as the progression of NAFLD advances, relying merely on dietary changes and weight management is insufficient to halt disease progression. There is a pressing need to implement additional therapeutic measures, including pharmacological interventions, aimed at controlling blood glucose levels and reducing blood lipids (46). The TyG index was calculated based on blood glucose and lipid levels. Monitoring changes in the TyG index not only allows for the prediction of the severity of NAFLD but also provides a dynamic assessment of disease progression. Moreover, compared with other indices, the TyG index and its related parameters demonstrate a significantly higher sensitivity in predictive value (13), which may effectively assist clinicians in implementing interventions such as blood glucose control and lipid-lowering medications when a sustained increase in the TyG index is observed, thereby preventing further progression of NAFLD. Although the effectiveness of the TyG index has been supported by numerous studies, factors such as regional differences in diet and lifestyle may influence its performance. Therefore, future research should focus on global validation and the establishment of standardized protocols to ensure the TyG index’s applicability and accuracy across different regions.

Our study has some limitations. Firstly, it diagnosed and classified NAFLD solely using B-mode ultrasound, which prevented the determination of whether patients would progress to NASH. Although simple steatosis is generally accepted as a diagnostic criterion for NAFLD, the limitations of this method must be considered. Secondly, owing to technical and financial constraints, we were unable to assess the degree of liver fibrosis using a liver biopsy. Several potential confounding factors may have influenced the observed association between changes in the TyG index and the progression of NAFLD/MASLD. Lifestyle modifications during the follow-up period, such as changes in dietary intake, physical activity, or body weight, could affect insulin resistance and lipid metabolism, thereby impacting TyG levels (47). In addition, comorbidities including diabetes, hypertension, and dyslipidemia (48), as well as the initiation or adjustment of relevant medications (e.g., statins, antihyperglycemic agents) (49), may have influenced both TyG dynamics and hepatic fat accumulation. Although our multivariable models adjusted for many of these covariates, residual confounding cannot be entirely ruled out due to the lack of detailed longitudinal data on lifestyle and pharmacologic interventions.

Despite these limitations, our findings underscore the clinical utility of TyG-based indices as practical and noninvasive tools for the screening and monitoring of NAFLD/MASLD. The TyG index, calculated from routine fasting glucose and triglyceride measurements, serves as a surrogate marker for insulin resistance and can be readily integrated into standard clinical workflows. Importantly, our results indicate that not only baseline TyG levels but also longitudinal increases in TyG are significantly associated with disease progression. This highlights the value of repeated TyG assessments over time to identify individuals at elevated risk, especially in older populations where scalable, cost-effective screening strategies are particularly needed.

In elderly populations, the TyG index and its related parameters hold significant value in the diagnosis of NAFLD and prediction of different stages of the disease. Additionally, monitoring changes in the TyG index can effectively indicate NAFLD progression. Given its simplicity and availability in routine laboratory testing, the TyG index may serve as a practical tool for early detection and longitudinal monitoring of NAFLD/MASLD in primary care settings. Incorporating TyG-based metrics into community health screening programs or clinical guidelines may facilitate timely intervention and help mitigate long-term hepatic and metabolic complications, particularly in aging populations.