MetS is defined as a multifactorial and multicomponent clinical syndrome characterized by metabolically interconnected disorders, with core features including obesity, hyperglycemia, dyslipidemia (such as hypertriglyceridemia and low HDL-C), and hypertension (HTN). MetS not only directly promotes the development of atherosclerotic cardiovascular disease (ASCVD) but also increases the risk of T2DM. Consequently, individuals with MetS are considered a high-risk population for cardiovascular disease (CVD) (1). Given that IR is central to its pathogenesis, MetS shares key features with T2DM, including IR itself, HTN, and glucolipid metabolic disorders. Furthermore, the prevalence of both conditions has been rising steadily year by year (2). This contributes to a higher risk of CVD and patient mortality, thereby creating significant pressure on public health systems (3). In recent years, researchers have consequently dedicated efforts to discovering biomarkers for diagnosing and monitoring T2DM complicated with MetS. Such biomarkers hold promise for the early identification of high-risk individuals, assessment of disease progression, and formulation of personalized treatment strategies, which could facilitate early intervention, delay dis-ease progression, lower complication rates, and ultimately enhance patient quality of life and prognosis.

HOMA-IR serves as a simple surrogate marker for IR. However, due to limitations in health care resources and technical capabilities in some regions of China, the measurement of insulin has not yet been widely implemented. Given these limitations, using IR-sensitive biomarkers as clinical tools for evaluating metabolic health may offer practical advantages in research and clinical practice. Currently utilized to evaluate IR, the TyG index is a crucial indicator of HOMA-IR derived from TG and fasting plasma glucose (FPG) levels, which as a non-insulin-based marker, presents advantages over HOMA-IR in terms of lower cost and easier clinical implementation (4). Recent studies suggest that the TyG-BMI index may serve as a potential alternative measure of IR, which combines the TyG index and BMI. This composite indicator incorporates the influence of obesity on IR, thereby providing a more comprehensive assessment of IR severity (5). Moreover, the TG/HL-C ratio has been previously validated as a surrogate measure of IR (6).

Therefore, this study aims to evaluate the predictive value of the TyG index, the TyG-BMI index, and the TG/HDL-C ratio for T2DM complicated with MetS, with the goal of identifying the optimal parameter to serve as a practical tool for clinical diagnosis and management, thereby offering a simpler, more cost-effective, and efficient method.

A cohort of 2049 adults with T2DM was included. The overall MetS prevalence was 74.18% (males: 64.30%; females: 35.70%). The T2DM-MetS group demonstrated notably higher values in SBP, DBP, WL, WHR, BMI, TG, SUA, FPG, 2hPPG, FCP, and 2hCP compared to the T2DM-only group. Additionally, the TyG index, TyG-BMI index, and the TG/HDL-C ratio were significantly elevated in the T2DM-MetS group and as the clustering of MetS components increased, the levels of the TyG index, the TyG-BMI index, and the TG/HDL-C ratio also increased. Multivariate logistic regression con-firmed that these three indices were independent factors associated with MetS in T2DM. Notably, in contrast to the surrogate indices evaluated, HOMA-IR did not differ significantly between T2DM patients with and without MetS in our cohort. This finding may reflect the confounding influence of glucose-lowering therapies, particularly insulin and insulin secretagogues, which can elevate circulating insulin levels independent of underlying insulin resistance. This further supports the utility of non-insulin-based markers such as the TG/HDL-C ratio in clinical settings where insulin measurement is impractical or influenced by pharmacotherapy.

This is the first study to comprehensively evaluate the relationship and diagnostic capabilities of three IR indices (TyG, TyG-BMI, and the TG/HDL-C ratio) in relation to MetS within a cohort of 2409 T2DM patients. We discovered that among the IR indices, the strongest association with MetS was for the TG/HDL-C ratio. Multi-variate logistic regression provided additional evidence that the TG/HDL-C ratio is an independent determinant of MetS, maintaining a significant association even after adjusting for confounding variables such as age, sex, disease duration and the uric acid. In terms of diagnostic performance, ROC curve analysis revealed that the TG/HDL-C ratio demonstrated the greatest overall ability to distinguish between groups, high-lighting its potential as a non-invasive screening tool.

Earlier research has demonstrated a significant association between a higher TG/HDL-C ratio and unfavorable metabolic traits, such as dyslipidemia, obesity, diabetes, and Metabolic Dysfunction-Associated Fatty Liver Disease (MAFLD) (10–12). Recently there has been increasing attention on the association between this ratio and MetS. A cross-sectional study, involving over 5000 Iranian participants compared various lipid ratios for detection of MetS in the Iranian general population. The researchers concluded that the TG/HDL-C was the best indicator for identifying MetS compared to the other ratios (13). Another study in Iran demonstrated that the high TG/HDL-C ratio was associated with a 2.12 times increased risk of developing MetS, using a cut-off point of 4.03 for males and 2.86 for females (14). In 2021, a cross-sectional study was conducted in the elderly Chinese population, which included a total of 1267 participants ≥ 65 years of age, to investigate a correlation between TG/HDL-C ratio and MetS. They determined that TG/HDL-C ratio values exceeding the cut-off values of 1.437 for men and 1.196 for women predicted a higher risk of developing MetS (15). In line with these, data from our study on the T2DM population showed that the TG/HDL-C ratio conveyed a significantly higher risk for the diagnosis of MetS (adjusted OR = 14.795, 95% CI: 11.283- 19.401; P = 0.000), using an optimal cut-off point of 1.562. Undoubtedly, the TG/HDL-C ratio was a very satisfactory predictor for MetS in T2DM patients. Nonetheless, taking into consideration the different cut-off values of multiple trials, based on ethnicity, genetics and lifestyle, the afore mentioned ratio cannot be considered an absolute parameter without calibration.

Studies have demonstrated that the TyG index, which integrates FPG and TG, is a convenient tool for detecting IR, one of the main components of MetS (16). A me-ta-analysis and systematic review of 13 studies (N = 49,325) on the diagnostic accuracy of the TyG index for MetS in adults showed the summary ROC analysis yielded an AUC of 0.90 (79% specificity, 82% sensitivity) in males and 0.87 (85% specificity, 81% sensitivity) in females, supporting its high diagnostic accuracy for MetS (17). A large, community-based, prospective cohort over 12 years of follow-up conducted by Da-Hye Son et al. from South Korea, evaluated the comparative predictive utility of the TyG index and HOMA-IR for the prevalence and incidence of MetS (18). It was demonstrated that the TyG index showed higher predictive power for prevalent MetS than HOMA-IR, with optimal cutoffs of 8.718 for prevalence and 8.518 for incidence. Further evidence from a large-scale population-based study conducted in Wuhu, China, involving 298,652 participants, demonstrated that the prevalence of MetS showed a corresponding increase across ascending quartiles of the TyG index which also identified the optimal TyG index cutoff for diagnosing MetS to be 8.85, with a sensitivity of 81% and a specificity of 91% (19). In line with these, data from our study on the T2DM population showed that the TyG index conveyed a significantly higher risk for the diagnosis of MetS (adjusted OR = 6.049, 95% CI: 4.8877- 7.489; P = 0.000), using an optimal cut-off point of 9.171.

Notably, the TyG-BMI index, which is modified by BMI, has shown superior performance in evaluating the severity of IR. This index, calculated as Ln [TG (mg/dl) × FPG (mg/dl)/2] × BMI, was first proposed in 2016 by Er et al. (20). It was also noted that, with a range of 16.6%, the TyG-BMI index had a strong correlation with HOMA-IR among the visceral obesity indices and TyG-related values. A Study by Lim et al. not only confirmed the above findings, but further suggested that the TyG-BMI index was a more accurate predictor of IR than the TyG index and the TyG-WC index alone (21). Gui et al. revealed that among middle-aged and elderly adults, various obesity- and lipid-related indices could predict MetS, with the TyG-BMI index being the strongest predictor in males (22). Additionally, Tamini et al. demonstrated that the TyG-BMI index served as a promising non-invasive instrument to assess MetS risk in obese pediatric and adolescent populations (23). The studies reviewed above demonstrate that the TyG-BMI index could serve as an effective, non-invasive clinical biomarker for the early recognition of MetS. In line with previous researches, our findings also established the TyG-BMI index as an independent determinant of MetS within the T2DM population (adjusted OR = 1.030, 95% CI: 1.026- 1.034; P = 0.000), using an optimal cut-off point of 230.780.

Based on the optimal predictive indicator for MetS, the TG/HDL-C ratio, combined with other key indicators (sex, WHR, and FCP), a prediction model for the development of MetS in patients with T2DM was further developed in this study. The ROC curve analysis yielded an AUC of 0.922 (95% CI: 0.909, 0.936), indicating good predictive performance. The DCA analysis showed that within a risk threshold range of 0.00 to 1.00, the model achieved a maximum net benefit of 0.742, supporting its favorable clinical utility. Similarly, the calibration curve indicated a close alignment of predictions with actual observations, suggesting good model calibration. Thus, the model has significant utility for evaluating MetS risk among individuals with T2DM, guiding both clinical risk prediction and preventive strategies.

In addition to the biochemical and anthropometric indicators evaluated in this study, it is worth emphasizing that early and accurate anthropometric assessment remains a cornerstone in identifying individuals at high risk for MetS. While traditional measures such as BMI and WHR are valuable, emerging evidence highlights the importance of body fat distribution in metabolic risk stratification. For instance, recent studies suggest that echocardiographic imaging modalities- particularly speckle tracking echocardiography- can provide non-invasive insights into cardiometabolic risk. Myocardial and atrial strain parameters have been shown to correlate with MetS probability, particularly in individuals with android (central) obesity, whereas those with gynoid fat distribution exhibit lower metabolic risk (24). Integrating such imaging-based assessments with anthropometrics and biochemical indices like the TG/HDL-C ratio could facilitate a more comprehensive, multimodality approach to early risk detection and personalized prevention strategies in T2DM patients.

In conclusion, the surrogate indices for IR (the TyG index, the TyG-BMI index, and the TG/HDL-C ratio) were risk factors for T2DM complicated with MetS. This is the first large-sample study (n = 2,409) to systematically compare three IR surrogate indices (TyG, TyG-BMI, TG/HDL-C) specifically in a T2DM population for predicting MetS. While previous studies have investigated these indices individually or in general populations, no study has evaluated their comparative diagnostic performance in T2DM patients using a comprehensive analytic framework including ROC, DeLong’s test, and nomogram construction. We further demonstrated that TG/HDL-C ratio outperforms TyG and TyG-BMI in this specific population, and constructed a clinically applicable nomogram integrating TG/HDL-C, sex, WHR, and FCP, achieving an AUC of 0.922. This provides a cost-effective, non-insulin-dependent tool for early MetS risk stratification in T2DM patients, particularly in resource-limited settings. An important methodological strength of this study lies in the standardized fasting protocol. Unlike outpatient or community-based retrospective studies, in which fasting duration is often documented as “at least 8 hours” with considerable inter-individual variability, all participants in our cohort were hospitalized under a controlled dietary regimen. Dinner was served at a fixed time (5:30 PM), and no caloric intake was permitted thereafter until blood collection the following morning. This ensured a consistent fasting duration of approximately 12–13 hours for all subjects, thereby minimizing confounding effects of variable fasting intervals on triglyceride and glucose measurements. This design enhances the internal validity of our findings and supports the reliability of IR surrogate indices such as the TG/HDL-C ratio. However, our study has some limitations. First, this is a single center retrospective study. Future large-scale, multi-center trials are needed to establish the causal relationship between surrogate IR indices for T2DM complicated with MetS, which would provide further validation for our conclusions. Second, all participants were recruited from Beijing, China, and the MetS diagnostic criteria used were tailored to abdominal obesity features characteristic of the Chinese population, which may limit the generalizability of our findings to other regions or ethnic groups, particularly those with different body composition patterns or diagnostic thresholds. Third, this study did not systematically adjust for the confounding effects of glucose-lowering medications (such as insulin, insulin secretagogues, metformin, etc.). These drugs may directly influence insulin levels and metabolic parameters, thereby affecting the interpretation of the study’s indicators. Future research should systematically collect and analyze detailed medication data to validate the robustness and generalizability of indicators such as the TG/HDL-C ratio in treated T2DM populations.