Gestational diabetes mellitus (GDM) refers to varying degrees of glucose intolerance that occur or are identified for the first time during pregnancy, irrespective of pre-existing diabetes (1). Over the last few years, the incidence of GDM has gradually increased, ranging from 9.3% to 25.5% (2). GDM is often associated with preeclampsia, macrosomia, perinatal anomalies, and mortality, while being closely linked to the onset of metabolic syndrome and hyperglycemia in both mother and offspring (3). This condition significantly affects the well-being of pregnant women and fetuses and poses a concealed risk for future ailments (4, 5). Clinical diagnosis of GDM typically occurs during the 24-28 week gestational period using a 75g oral glucose tolerance test (OGTT) (6). However, empirical evidence suggests that by the time GDM is diagnosed at this stage, both the mother and fetus may have already been adversely affected to varying degrees, despite the potential benefits of symptom management (5, 7). Thus, early recognition of pregnancies at risk for GDM is essential in preventing negative outcomes for pregnancy and intergenerational transmission of metabolic disorders.

Early detection of insulin resistance (IR) in pregnant women has been shown to assist in predicting the onset of GDM before clinical diagnosis (8, 9). The TyG index, calculated from fasting plasma glucose (FPG) and serum triglycerides (TG), is considered a straightforward, economical, replicable, and reliable surrogate for IR (10, 11). While many studies have investigated the relationship between the TyG index and GDM, suggesting its potential as an early GDM risk indicator (12, 13), there may be variations between ethnic groups. For instance, Sánchez-García et al. (14) found no significant difference in TyG index values between Latin American pregnant women with and without gestational diabetes. A meta-analysis by Song et al. (15) indicated that a higher TyG index may predict GDM in Asian women but not in non-Asian women. Therefore, using data from the National Health and Nutrition Examination Survey (NHANES), we conducted a cross-sectional investigation with a population of pregnant women in the United States to assess the connection between the TyG index and GDM.

This study further explored the correlation between TyG index and GDM using the nationally representative NHANES database. According to the results, pregnant women in the highest quartile of the TyG index had a 2.92-fold increased risk of developing GDM compared to those in the lowest quartile. The association between gestational diabetes mellitus and TyG index remains significant even after adjusting for all possible confounders, while the AUC of TyG index for predicting GDM was 0.57. It is noteworthy that the present study is the first to report a linear positive association between TyG index and GDM after adjusting for confounding variables, utilizing a population of pregnant women from the NHANES database from 1999 to 2020. Furthermore, the results of the subgroup analyses demonstrate the robustness of our findings.

The TyG index is a composite biochemical indicator that reflects the integrated influence of blood lipids and glucose. It has been reported to demonstrate high sensitivity in identifying insulin resistance among healthy individuals (11). Li et al. (18) conducted a cohort study that demonstrated a distinct association of the TyG index with an elevated probability of incident diabetes, suggesting its potential as a predictive indicator for type 2 diabetes mellitus (T2DM). The TyG index has been shown to be valuable in various cardiovascular disease types, including stable coronary artery disease, acute coronary syndromes, in-stent restenosis, arterial stiffness, coronary artery calcification, and heart failure (19). Meanwhile, multiple research investigations have revealed a strong correlation between the various factors of the TyG index and the risk of cardiovascular disease occurrence in both normal-weight individuals (20) and those suffering from non-alcoholic fatty liver disease (21). In a study of pregnant women, Pan et al. (22) observed that the TyG index in the early trimester of pregnancy is closely associated with the development of gestational hypertension and adverse pregnancy outcomes. Moreover, recent cohort studies have demonstrated a statistically significant association between the TyG index and the risk of developing gestational diabetes mellitus (GDM), indicating its efficacy as a predictor for both GDM and large for gestational age newborns (12, 23, 24).

GDM’s pathophysiology involves beta cell dysfunction and tissue insulin resistance during pregnancy. Elevated glucose levels can trigger islet beta cells to create reactive oxygen species, leading to oxidative stress and cell dysfunction, culminating in insulin resistance (25). Pregnant women, especially those with GDM, often exhibit elevated serum triglyceride levels, potentially impairing pancreatic beta cell function (26, 27). Early identification of GDM risk using reliable insulin resistance indicators is crucial to prevent adverse consequences. The TyG index is considered a cost-effective and efficient indicator of insulin resistance compared to other methods like Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), as it eliminates the need for measuring insulin levels (28, 29). Its advantage lies in its ability to be derived from a single sample, making it highly suitable for various clinical applications (30).

Age, race, obesity, inheritance, smoking, and various risk factors collectively contribute to insulin resistance or GDM occurrence (31, 32). In this investigation, we comprehensively considered age, race, BMI, history of hypertension, hypercholesterolemia, alcohol use, smoking status, and gestational age as stratifying variables. Subgroup analysis and interaction tests consistently demonstrated correlations across different groups. Interestingly, our results from the subgroup analysis revealed that individuals aged <35 years, Mexican Americans, and those with a BMI ≥25 kg/m² exhibited a higher risk of GDM. BMI and age, two important characteristics related with insulin resistance, have a major influence on the development of GDM. It has been observed that maternal pre-pregnancy obesity substantially increases the risk of developing GDM, with an odds ratio (OR) of 1.63 (95% CI: 1.320-2.019), whereas the adjusted OR for older mothers (≥35 years) was 1.45 (95% CI: 1.184-1.776). This suggests that while age is indeed a risk factor for GDM, BMI exerts a greater influence on its development (33). Moreover, BMI serves as a crucial indicator for GDM in pregnant women with polycystic ovary syndrome (PCOS). Previous reports have highlighted that pregnant women affected by PCOS face an elevated risk of developing GDM if their BMI exceeds 28 kg/m² (34). However, the results of our analysis showed that a higher maternal TyG index during the first trimester of pregnancy did not significantly associate with GDM, raising concerns about using the TyG index as an early predictor of GDM. This finding aligns with a recent meta-analysis, suggesting that triglyceride variations during pregnancy and individual differences may contribute to these results (35). The limitations of the TyG index, derived from static measurements of fasting glucose and triglycerides, might not fully capture the intricate interactions of insulin sensitivity, insulin secretion, and glucose metabolism as assessed through oral or intravenous glucose tolerance tests (22, 36). Moreover, various factors affecting triglyceride metabolism, such as nutrition, alcohol use, antioxidants, medications, liver function, and genetic variations, could also impact the TyG index (19, 37, 38).

The predictive ability of the TyG index for GDM has been extensively investigated by multiple researchers. Liu et al. (24) reported an AUC of 0.686 (95% CI: 0.615-0.756) for the TyG index in predicting the risk of GDM. Similarly, Li et al. (12) demonstrated the potential of the TyG index in detecting the risk of GDM, with an AUC of 0.637 (95% CI: 0.626~0.649). Another study by Khan et al. (39) found that the TyG index exhibited the highest AUC, surpassing HbA1c and other biomarkers, with a predictive AUC of 0.712 (95% CI: 0.631-0.793) for GDM. In our study, we observed a limited effectiveness of the TyG index in predicting GDM, with an AUC of 0.57 (95% CI: 0.50-0.62). Differences in the reported ROC curves for the diagnosis of GDM using the TyG index may arise from various factors, including disparities in study populations, sample sizes, calculation methods, and reference standards utilized to diagnose GDM. Notably, in our study, the use of fasting blood glucose rather than the oral glucose tolerance test (OGTT) as the reference standard for GDM may have contributed to the observed discrepancy. While the TyG index as a composite indicator seems to reflect insulin resistance in pregnant women, further research is indispensable to fully explore its potential for early prediction of gestational diabetes.

Acknowledging the limitations of our research is essential. Firstly, the cross-sectional design using data from the NHANES database prevented us from establishing a direct causal association between the TyG index and the risk of developing GDM during pregnancy. Secondly, GDM recognition in this study relied solely on fasting plasma glucose, rather than using the more comprehensive OGTT. Several studies have indicated that our GDM categorization based only on fasting glucose may have an approximate 26% misclassification rate (40), potentially leading to an underestimation of GDM instances and influencing result interpretation. Thirdly, due to the cross-sectional design of the NHANES, all variable measurements were taken at a single point during a woman’s pregnancy, and trimester verification relied on pregnant women’s self-reports. Additionally, to avoid biased results, we did not compare the TyG index and HOMA-IR as distinct risk parameters for GDM, as the data for insulin levels in pregnant women had a substantial number of missing values. Future research is necessary to better predict GDM and its complications by incorporating the TyG index along with other clinical and biochemical parameters. Investigating potential mechanisms underlying the association between the TyG index and GDM, including the roles of adipokines, oxidative stress, inflammation, and insulin resistance, is also crucial. Despite these limitations, our study demonstrates a correlation between the gestational TyG index and GDM.

The present research demonstrated a positive correlation between the TyG index and gestational diabetes in pregnant women in the US. However, the diagnostic validity of TyG index for GDM is limited. Further investigation is needed to fully explore the potential of the TyG index as a predictor of GDM risk in early pregnancy.

Publicly available datasets were analyzed in this study. This data can be found here: https://wwwn.cdc.gov/nchs/nhanes.

YZ: conceptualization, statistical analysis, manuscript writing and editing. LY: methodology, reviewing and editing. XY: statistical analysis, reviewing. DZ: conceptualization, methodology, reviewing and editing. All authors contributed to the article and approved the submitted version.