Predictive Value of the second-Trimester Triglyceride-Glucose Index and Its Derived Indices for Macrosomia in Gestational Diabetes Mellitus

Introduction

Gestational diabetes mellitus (GDM) is characterized by glucose intolerance initially identified during pregnancy,¹ affecting 9.3% to 25.5% of pregnancies worldwide.² The prevalence in China reaches approximately 14.8% and continues to rise each year. Macrosomia, defined as birth weight ≥4000 g.³ represents a frequent complication of GDM and correlates strongly with various adverse maternal and neonatal outcomes. Maternal complications often include obstetric trauma and postpartum hemorrhage, while neonates are predisposed to neonatal hypoglycemia and an elevated risk of developing metabolic disorders in adulthood.^4,5 Therefore, identifying effective predictive markers for early detection of macrosomia holds profound clinical significance in improving both short and long term adverse outcomes for mothers and infants. Although routine ultrasound examination at 24–28 weeks of gestation is essential for structural screening, its sensitivity for predicting macrosomia is limited. This limitation arises because macrosomia typically manifests as accelerated growth in the late third trimester, fetal biometric measurements often remain within the normal range at this earlier stage, and operator-dependent weight estimation formulas are inherently inaccurate. Consequently, the predictive value of a single second-trimester ultrasound is limited, highlighting the need for supplementary biochemical markers to improve early risk stratification.⁶

Insulin resistance (IR), a central pathological mechanism in glucose and lipid dysregulation, significantly contributes to the development of GDM.^7–9 Recent studies confirm that the triglycerides-glucose (TyG) index correlates strongly with both the hyperinsulinemic-euglycemic clamp (the gold-standard measure of insulin resistance) and the homeostasis model assessment of insulin resistance (HOMA-IR).^10,11 It is also closely related to diabetes,¹² cardiovascular disease,^13,14 hypertension and other diseases. These findings establish the TyG index as a practical and economical biomarker for assessing insulin resistance.^15–17 Although the TyG index is currently being used more and more in pregnant populations,^18–21 more research is needed to determine its specificity and sensitivity in GDM patients, who have noticeable IR characteristics.²² Body mass index (BMI) and gestational weight gain (GWG) are established independent risk factors for macrosomia. Anthropometric measures alone, however, may inadequately reflect underlying metabolic dysregulation, including insulin resistance. This limitation underscores the need for composite indices. Recent studies suggest that derivative indices like TyG-body mass index (TyG-BMI) and TyG-waist circumference (TyG-WC) could improve the predictive accuracy of the TyG index.^23–27 However, their ability to predict macrosomia in GDM pregnancies remains uncertain. This retrospective study aims to evaluate the predictive value of the TyG index and its derivatives for macrosomia in GDM pregnancies. The findings may provide a clinical basis for early identification of high-risk GDM individuals and mitigation of macrosomia risk.

Materials and Methods

Study Participants

Pregnant women with GDM who underwent regular prenatal examinations and delivered at Nanjing Women and Children’s Healthcare Hospital between January 2017 and June 2022 were enrolled. The inclusion criteria were as follows: (1) Singleton pregnancy with delivery between 37 and 42 weeks of gestation; (2) Age 18–45 years; (3) Diagnosis of GDM confirmed by a 75g oral glucose tolerance test (OGTT) at 24–28 weeks of gestation, in accordance with the diagnostic criteria;²⁸ a diagnosis of GDM was confirmed if one or more of the following plasma glucose thresholds were met or exceeded: fasting plasma glucose (FPG) ≥ 5.1 mmol/L, 1-hour plasma glucose ≥ 10.0 mmol/L, or 2-hour plasma glucose≥ 8.5 mmol/L.(4) Complete clinical data available. The exclusion criteria included: (1) Pre-existing type 1 or type 2 diabetes mellitus, diagnosis of diabetes mellitus before pregnancy, or lipid-lowering medication use prior to pregnancy; (2) Coexisting pregnancy complications such as gestational hypertension; (3) Presence of autoimmune diseases, cardiovascular diseases, severe hepatic or renal dysfunction or malignancy; (4) Requirement of pharmacotherapy for other chronic conditions; and (5) Fetal congenital anomalies, chromosomal abnormalities, or inherited metabolic disorders. A total of 5,502 GDM patients aged 18–45 years (mean ± SD: 29.93 ± 3.64) were included. This study was approved by the Ethics Committee of Nanjing Maternity and Child Health Hospital (Approval No.: 2022KY-068). (Figure 1) details the flow chart for the selection of participants.

Figure 1 Study population selection and data analysis flowchart. Abbreviations: GDM, gestational diabetes mellitus; TyG, Triglycerides-glucose index; ROC, receiver operating characteristic; RCS, restricted cubic spline; NRI, net reclassification improvement; IDI, integrated discrimination improvement.

Research Method

Data Collection and Measurements

The relevant clinical data of the study subjects were collected through the electronic medical record information system, including age, height, pre-pregnancy body weight, gravidity, parity, family history of diabetes, present and past medical history, gestational weeks at delivery, pregnancy outcomes, and neonatal birth weight. Neonates with a birth weight ≥4000 g were defined as macrosomia.³ During the second-trimester (24–28 weeks of gestation), obstetric examinations were performed on the pregnant women. Standardized measurements were obtained for height, body weight, waist circumference (WC), diastolic blood pressure, and systolic blood pressure. Body mass index (BMI) and waist-to-height ratio (WHtR) were subsequently calculated.

Biochemical Index Measurement

Fasted for over 10 hours during the second-trimester (24–28 weeks of gestation), and 5 mL of venous blood was collected from the antecubital vein the following morning under fasting conditions. Fasting plasma glucose (FPG), total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and other biochemical parameters were measured using a fully automated biochemical analyzer (AU5800, Beckman Coulter). Glycated hemoglobin (HbA1c) was determined using a fully automated HbA1c analyzer (HA-8180, ARKRAY Inc). The TyG index and its derived indices were subsequently calculated.

Specific calculation formulas:

Grouping

Based on the tertile cutoff values of TyG, TyG-BMI, TyG-WC, and TyG-WHtR, participants were categorized into tertile 1 (T1), tertile 2 (T2), and tertile 3 (T3) groups, respectively. For trend analysis the first tertile served as the reference: TyG index (tertile 1, ≤ 8.849, tertile 2, 8.805–9.181; tertile 3, ≥ 9.181) TyG-BMI (tertile 1, ≤ 206.942, tertile 2, 206.942–236.037; tertile 3, ≥ 236.037) TyG-WC (tertile 1, ≤ 777.697, tertile 2, 777.697–845.598, tertile 3, ≥ 845.598) TyG-WHtR (tertile 1, ≤ 4.791, tertile 2, 4.791–5.221, tertile 3, ≥ 5.221).

Covariates

In the multivariate adjusted analysis, several demographic factors known to potentially influence the association between the TyG index and its derivatives and macrosomia were included as covariates, namely maternal age, gravidity (1, 2, or ≥ 3 times), parity (1 or ≥ 2 times), systolic blood pressure (SBP), and diastolic blood pressure (DBP). Additionally, adjustments were made for the following biochemical parameters as potential confounders: HbA1c, HDL-C, and LDL-C.

Statistical Analyses

Statistical analysis was performed using IBM SPSS 27.0. The normality of continuous variables was assessed by the Kolmogorov–Smirnov test. Normally distributed measurement data were expressed as mean ± standard deviation () and normally distributed continuous variables were analyzed using one-way ANOVA for intergroup comparisons, while categorical data were presented as n (%) and analyzed by χ² test. Patients were stratified into tertiles (T1-T3) based on TyG index and its derived indices. Continuous variables were compared using ANOVA or Kruskal–Wallis H-test, whereas categorical variables were analyzed by chi-square test. Spearman correlation analysis was conducted using the Correlation Plot module in Origin 2025 to examine the relationships between TyG and its derived indices with other parameters. Three multivariate-adjusted logistic regression models were developed to identify factors influencing macrosomia delivery in GDM patients. To further assess potential nonlinear associations between the TyG index and its derived indices with macrosomia, we performed restricted cubic spline (RCS) analysis, adjusting for all covariates included in Model 3. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to evaluate the diagnostic performance of TyG and its derived indices for predicting macrosomia in GDM pregnancies. DeLong’s test along with net reclassification improvement (NRI) and integrated discrimination improvement (IDI), was employed to compare the diagnostic efficacy among TyG, TyG-BMI, TyG-WC, and TyG-WHtR. Subgroup analyses were performed by stratifying participants based on age (<35 or ≥35 years), parity (1 or ≥2), to assess the influence of these variables. Interaction effects across subgroups were examined using likelihood ratio tests. Statistical analyses were conducted using R software (version 4.2.0). All tests were two-sided, with P <0.05 considered statistically significant.

Results

Baseline Characteristics

All pregnant women with GDM were stratified into tertiles based on their TyG index values (Table 1), T1 group (TyG ≤ 8.848, n=1838), T2 group (TyG 8.849–9.180, n=1830), and T3 group (TyG ≥ 9.180, n=1834).A progressive increase was observed across T1, T2, and T3 groups in maternal age, waist circumference (WC), body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), FPG, HbA1c, TG, infant weight and the proportion of macrosomia (P < 0.001). Conversely, gestational age at delivery, HDL-C, ALT, and AST levels demonstrated a sequential decline (P < 0.001). No significant differences were detected among TyG index tertiles for gestational weight gain (GWG) or infant sex. Similar trends were stratified by TyG-BMI,TyG-WC, and TyG-WHtR tertiles.

Table 1 Baseline Characteristics of Participants According to Tertiles of the TyG Index [, n(%)]

Spearman Correlation Analysis of the TyG Index and Related Parameters with Other Variables

As shown in (Figure 2), Spearman correlation analysis revealed that the TyG index was positively correlated with LDL-C, FPG, HbA1c, TG, TC, and infant weight (P < 0.05), while exhibiting a negative correlation with HDL-C (P < 0.001).The TyG-derived indices (TyG-BMI, TyG-WC, TyG-WHtR) demonstrated significant positive correlations with GWG, FPG, HbA1c, TG, and infant weight (P < 0.001), whereas they were negatively correlated with HDL-C and TC (P < 0.05).

Figure 2 Spearman correlation analysis of the correlation between TyG index related indices and other indicators. Abbreviations: GWG, Gestational weight gain; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; TG, triglycerides; TC, total cholesterol; BMI, Body mass index; WC, Waist circumference; WHtR, Waist-to-height ratio; TyG, Triglycerides-glucose index. Notes: *P≤0.05, **P≤0.01, ***P≤0.001.

Comparative Predictive Value of the TyG Index and Its Derived Indices for Macrosomia in Pregnant Women with GDM

Using macrosomia delivery by GDM mothers as the dependent variable, the TyG index and its derived indices were incorporated into the model according to their respective tertiles (T1, T2, T3), with T1 as the reference group (Figure 3). Logistic regression analysis was performed, and the results demonstrated that after adjusting for relevant confounding factors (age, gravidity, parity, SBP, DBP, HbA1c, HDL-C, LDL-C), the OR (95% CI) for macrosomia delivery in the T3 groups of TyG index, TyG-BMI, TyG-WC, and TyG-WHtR were 1.769 (1.278–2.450), 3.458 (2.427–4.926), 3.718 (2.603–5.312), and 2.279 (1.641–3.164) times higher (all P < 0.01), respectively, compared to the T1 group.

Figure 3 Forest plots show the Logistic regression analysis of factors influencing macrosomia in GDM pregnancies. (a) TyG index; (b) TyG-BMI; (c) TyG-WC; (d) TyG-WHtR. Model 1: Unadjusted; Model 2: Adjusted age, gravidity, parity, SBP, DBP; Model 3: Adjusted age, gravidity, parity, SBP, DBP, HbA1C, HDL-C, LDL-C. Abbreviations: T, tertiles; Ref, Reference group; SBP, systolic blood pressure; DBP, diastolic blood pressure, HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein; HbA1c, glycated hemoglobin. BMI, Body mass index; WC, Waist circumference; WHtR, Waist-to-height ratio; TyG, Triglycerides-glucose index; GDM, gestational diabetes mellitus.

Restricted cubic spline analysis showed a dose–response relationship between the TyG index and its derived indices and macrosomia after adjusting for multiple covariates (all P for overall < 0.001; Figure 4). The incidence of macrosomia increased linearly with rising TyG-BMI and TyG-WHtR (all P for Nonlinear > 0.05). The association between TyG and macrosomia risk followed an approximate inverted “U”-shaped curve (P for Nonlinear < 0.05). TyG < 9.0, the risk of macrosomia rose with increasing TyG levels; above this threshold, the risk plateaued and declined slightly. In contrast, TyG-WC exhibited a “J”-shaped relationship with macrosomia risk (P for Nonlinear < 0.05). Once TyG-WC exceeded 900, further increases were associated with a sharp rise in macrosomia incidence.

Figure 4 Dose-responsive relationship of the TyG index and its derived indices with the risk of macrosomia. (a) TyG; (b) TyG-BMI; (c) TyG-WC; (d) TyG-WHtR; Models were adjusted for Model 3. Abbreviations: OR, odds ratio; CI, confidence interval; BMI, Body mass index; WC, Waist circumference; WHtR, Waist-to-height ratio; TyG, Triglycerides-glucose index; GDM, gestational diabetes mellitus.

Model Prediction Performance

Compare the discriminative ability of the TyG index and its derived indices for predicting macrosomia in pregnant women with GDM (Table 2 and Figure 5). The AUC (95% CI) values for TyG, TyG-BMI, TyG-WC, and TyG-WHtR were 0.596 (0.572–0.621), 0.682 (0.658–0.706), 0.681 (0.658–0.705), and 0.651 (0.626–0.675), respectively. The corresponding sensitivities were 86.2%, 61.0%, 63.3%, and 61.2%, while the specificities were 28.2%, 66.3%, 63.0%, and 61.0%. The optimal cutoff values were 8.79, 233.05, 832.03, and 5.11, respectively. DeLong’s test revealed that the AUC of TyG-BMI and TyG-WC were significantly higher than that of TyG (P < 0.05), whereas no statistically significant difference was observed between TyG-BMI and TyG-WC (P = 0.881).

Table 2 Comparative Analysis of TyG and Its Derived Indices in Predicting Macrosomia in GDM

Figure 5 ROC curve analysis of four indexes for macrosomia delivery in GDM Pregnancies. Abbreviations: BMI, Body mass index; WC, Waist circumference; WHtR, Waist-to-height ratio; TyG, Triglycerides-glucose index; GDM, gestational diabetes mellitus.

Regarding the NRI and IDI, TyG-BMI and TyG-WC exhibited slightly superior performance compared to the other indices. The IDI values were 0.030 and 0.029, while the NRI values were 0.443 and 0.450, (P < 0.01). These findings indicate that TyG-BMI and TyG-WC serve as more effective predictive factors with slightly better discriminative ability than TyG and TyG-WHtR. However, no statistically significant difference in predictive ability was observed between TyG-BMI and TyG-WC (P > 0.05).

We further evaluated whether the composite indices TyG-BMI and TyG-WC possessed higher predictive value than the individual indices BMI and WC. The AUC (95% CI) for BMI and WC alone were 0.677 (0.652–0.701) and 0.674 (0.650–0.698), respectively. Although these values were slightly lower than those of TyG-BMI and TyG-WC, the differences did not reach statistical significance (BMI vs TyG-BMI, P = 0.103; WC vs TyG-WC, P = 0.203). However, comparisons using the NRI and IDI demonstrated that TyG-BMI and TyG-WC, which incorporate the TyG index, outperformed BMI and WC alone. The NRI values were 0.219 and 0.190 (P < 0.01), and the IDI values were 0.014 and 0.016 (P < 0.01).

Subgroup Analysis

To further investigate the association of TyG, TyG-BMI, TyG-WC, and TyG-WHtR with macrosomia in women with GDM, subgroup analyses were performed based on maternal age and parity (Figure 6 and Table 3). After adjusting for confounding factors in Model 3, no significant interaction was observed between subgroup variables (age, parity) and the association of TyG and its derived indices with the risk of macrosomia in GDM pregnancies (P for interaction >0.05). All four indices exhibited a positive correlation with macrosomia risk. Notably, TyG-BMI and TyG-WC demonstrated significant correlations in both age and parity subgroups, whereas TyG and TyG-WHtR showed no significant association in the parity ≥1 subgroup and the age ≥35 subgroup, respectively.

Table 3 Subgroup Analysis for the Association of the TyG and Its Derived Indices Tertiles for Macrosomia in GDM Pregnancies

Figure 6 Subgroup analysis of the association between TyG and its derived indices for macrosomia in GDM pregnancies. (a) TyG; (b) TyG-BMI; (c) TyG-WC; (d) TyG-WHtR, Models were adjusted for Model 3. Abbreviations: BMI, Body mass index; WC, Waist circumference; WHtR, Waist-to-height ratio; TyG, Triglycerides-glucose index; GDM, gestational diabetes mellitus.

Discussion

This study analyzed clinical data from women with singleton pregnancies who received routine prenatal care and delivered at the Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital. It examined the association between the TyG index and its derived indices (TyG-BMI, TyG-WC, TyG-WHtR) measured during the second trimester and the risk of macrosomia in women with GDM. These findings may support earlier identification and intervention in high-risk populations. The results showed that the probability of delivering a macrosomic infant increased with higher TyG index values. Spearman correlation analysis revealed that TyG and its derived indices were positively correlated with TG, TC, and LDL-C but negatively correlated with HDL-C (P < 0.01). After adjusting for potential confounding factors, the highest tertiles of TyG, TyG-BMI, TyG-WC, and TyG-WHtR were associated with 1.769-fold, 3.458-fold, 3.718-fold, and 2.279-fold increased risks of macrosomia, respectively, compared with the lowest tertiles (all P < 0.05). It indicates that TyG and its derived indicators are independent risk factors for macrosomia in women with GDM.

The TyG index, a metabolic status evaluation metric based on glycemic and lipid parameters, was initially introduced in 2008 as a novel surrogate marker for assessing IR.²⁹ With advancing understanding of GDM, studies have demonstrated that IR induced by GDM plays a significant role in macrosomia.³⁰ Research has revealed that the TyG index exhibits diagnostic performance comparable to the OGTT for GDM between 24–28 weeks of gestation.³¹ Ning Ma et al³² confirmed the predictive value of first-trimester TyG index for GDM and its significant correlation with macrosomia. Retrospective analyses have identified pre-pregnancy TyG index as an independent risk factor for macrosomia in GDM patients, demonstrating its combined predictive value with GWG for macrosomia.³³ The underlying mechanism may involve elevated TG and blood glucose during the second-trimester activating serine-threonine kinases that inhibit insulin signaling pathways, thereby exacerbating IR.³⁴ IR can induce fetal β-cell hyperplasia and accelerate the maturation of the stimulus-secretion coupling mechanism in β-cells. Consequently, the fetus produces excessive insulin in response to hyperglycemia caused by increased placental glucose transfer. As insulin acts as a growth hormone, this leads to enhanced fetal growth and adiposity. The combined effects of hyperglycemia and hyperinsulinemia promote increased fetal fat and protein storage, ultimately resulting in macrosomia.³⁴

The results of this study demonstrated that the TyG index during the second-trimester of pregnancy predicted macrosomia in women with GDM with an AUC of 0.596. The optimal cut-off value was 8.79, yielding a sensitivity of 86.3% and specificity of 28.2%. A retrospective cohort study conducted at two tertiary hospitals reported a similar AUC of 0.599 for TyG in predicting macrosomia,³⁵ which is consistent with our findings. The RCS analysis confirmed these results. Although TyG and its related indices were correlated (all P for overall <0.001), the RCS curve for TyG demonstrated a nonlinear, approximately inverted U-shaped relationship with macrosomia. This phenomenon may result from the combined effects of pathophysiological mechanisms and clinical interventions. The severe glucolipotoxicity associated with an extremely high TyG index may lead to vascular endothelial injury and placental insufficiency,³⁶ thereby limiting nutrient transport. Furthermore, competitive risks such as the saturation of placental transporters or preterm birth may also restrict further fetal weight gain under extreme metabolic conditions.^37,38 Clinically, a high TyG index often indicates severe hyperglycemia, prompting earlier and more intensive clinical management. However, when glycemic targets are not met, the initiation of insulin therapy effectively curbs excessive fetal growth. This finding suggests that for such high-risk pregnant women, relying solely on routine gestational weight management is often insufficient, and the TyG index can serve as a key tool to guide the timely initiation of intensive metabolic management. The high sensitivity but low specificity of the TyG index suggests that using a single parameter to predict macrosomia in GDM patients has certain limitations. This may be attributed to the multifactorial nature of GDM,³⁹ with numerous factors influencing fetal macrosomia, including maternal age, gravidity, parity, and excessive GWG.³⁵ Therefore, relying solely on a single indicator to predict macrosomia in GDM patients has inherent limitations.

Previous studies have confirmed that maternal body mass index (BMI) is an independent risk factor for macrosomia. Maternal obesity during pregnancy is not only closely associated with hyperinsulinemia and dyslipidemia, but also correlated with impaired endothelial function and increased inflammation.^40,41 A meta-analysis comprising 23 studies (including a total of 17,711 women) demonstrated a significant association between maternal HbA1c levels and macrosomia.⁴² A prospective cohort study further revealed that elevated maternal serum lipid levels remained an independent risk factor for macrosomia even after adjusting for maternal blood glucose levels and pre-pregnancy BMI.

Therefore, further analysis was conducted to evaluate the predictive capabilities of TyG-derived indices incorporating obesity-related parameters (TyG-BMI, TyG-WC, and TyG-WHtR) for macrosomia in women with GDM, with their predictive performances compared using DeLong’s test. The results demonstrated that TyG-BMI, TyG-WC, and TyG-WHtR yielded AUC values of 0.682, 0.681, and 0.651, respectively, for predicting macrosomia in GDM patients, all significantly superior to the TyG index alone (P < 0.05). Notably, TyG-BMI and TyG-WC showed comparable discriminative power, suggesting their equivalent utility in GDM screening (P = 0.881). However, compared to BMI alone (AUC = 0.677) or WC alone (AUC = 0.674), the improvement in AUC was modest and not statistically significant (P > 0.05). Among these composite indices, BMI and WC appear to be the primary drivers of predictive performance. NRI and IDI analyses further confirmed that TyG-BMI and TyG-WC significantly improved risk reclassification and discriminative ability (all P < 0.01). These results suggest that the incorporation of obesity parameters (BMI or WC) provides a more comprehensive assessment of the synergistic effect between metabolic abnormalities and adiposity accumulation, which is consistent with previous studies demonstrating that TyG combined with other risk factors enhances predictive performance.⁴³

The underlying mechanism may involve excessive free fatty acids, reactive oxygen species (ROS), and proinflammatory cytokines released by adipose tissue in overweight or obese individuals, which exacerbate insulin resistance, stimulate fetal hyperinsulinemia, promote excessive fetal growth, and consequently increase macrosomia risk. As composite indices derived from TG, FPG, and anthropometric parameters, TyG-BMI and TyG-WC integrate glycolipid metabolic risk factors into clinically actionable metrics. Rather than replacing traditional BMI assessment, these indices serve as complementary tools that enable early screening and lifestyle modification to reduce macrosomia risk in GDM pregnancies, thereby providing valuable guidance for perinatal care.

Furthermore, our subgroup analysis demonstrated that the associations of TyG-BMI and TyG-WC with macrosomia in GDM pregnancies remained consistent across different age and parity groups. However, TyG and TyG-WHtR showed no significant correlation in the age ≥35 years group, and TyG exhibited no significant association in the parity ≥1 group. This may be attributed to the decline in metabolic function among advanced-age pregnant women, who are exposed to additional risk factors beyond IR or obesity. Additionally, altered insulin sensitivity in multiparous women may render single indicators (eg. TyG) less reliable. In contrast, TyG-BMI and TyG-WC, which incorporate body size parameters, exhibit more stable predictive performance.

The profound significance of this study lies in its value for optimizing risk stratification using readily accessible and low-cost indicators. Current clinical practice primarily focuses on metrics such as BMI, blood glucose, and GWG. Incorporating composite indicators like the TyG index and its derivatives for synergistic assessment can capture underlying metabolic risks. These indicators can be calculated during the OGTT at 24–28 weeks of gestation, thereby aiding in the early collaborative identification of high-risk pregnant women and facilitating early, stratified, and precise management.

This study has several limitations. First, dietary influences on blood glucose and lipid levels were not evaluated. Second, the analysis relied solely on second-trimester single-point data, omitting longitudinal tracking throughout pregnancy. Third, for women with GDM, we could not determine whether insulin therapy weakened the observed associations between the TyG index, its derived indices, and macrosomia incidence.

Future studies should integrate nutritional assessments and monitor key indicators longitudinally. Incorporating additional clinical examination data would enable a more comprehensive assessment of TyG and its derived indices in predicting macrosomia risk among GDM patients.

Conclusion

In summary, elevated TyG index and its derived indices during the second-trimester are associated with an increased risk of macrosomia in women with GDM, all demonstrating predictive value. Notably, compared to the TyG index alone, the TyG-BMI and TyG-WC demonstrated better predictive performance. These composite indices can serve as effective screening adjuncts to better identify high-risk pregnancies, particularly by capturing the synergistic risk of obesity and insulin resistance.