CRP–triglyceride–glucose index (CTGI) as a predictor of preeclampsia: a population-based study of risk stratification Authors: Yuting Liang, Yanqiu Zhang, Yujing Li, Jun Cao, Bin Feng, Jieyu Jin, Sheng Zhang, Qingqin Tang, Longwei Qiao, Zhixing Jin Published: 2025 ABSTRACT AbstractBackgroundpreeclampsia (PE) remains a leading cause of maternal and perinatal morbidity and mortality worldwide. While metabolic and inflammatory factors are increasingly recognized in its pathogenesis, the clinical utility of composite biomarkers remains underexplored. This study aimed to investigate the association between the C-reactive protein-triglyceride-glucose (CRP-TG-glucose) index (CTGI), a novel marker of metabolic-inflammation stress, and the risk of preeclampsia.MethodsThis retrospective cohort study included 11,916 pregnant women, of whom 486 developed preeclampsia. Maternal baseline characteristics were compared between the PE and non-PE groups. Logistic regression analyses were conducted to identify factors associated with PE. The relationship between CTGI and PE risk was further explored using quartile stratification, restricted cubic spline regression, and threshold effect analyses. Subgroup analyses were also performed to assess interaction effects across maternal and obstetric variables.ResultsWomen with PE had significantly higher maternal age, body mass index (BMI), in vitro fertilization (IVF) conception, multifetal pregnancies, and elevated CTGI levels compared to non-PE counterparts (all P < .001). Multivariate logistic regression identified CTGI as an independent risk factor for PE (adjusted OR, 1.78; 95% CI, 1.51–2.09; P < .001), alongside BMI, maternal age, IVF, and multifetal gestation. A dose–response relationship was observed across CTGI quartiles, with the highest quartile showing a markedly increased PE risk (adjusted OR, 2.06; 95% CI, 1.52–2.81). Restricted cubic spline models and threshold analysis revealed a nonlinear association with a significant inflection point at CTGI = 2.244. Above this threshold, the risk of PE rose sharply (OR, 3.93; 95% CI, 2.09–7.39; P < .001). Subgroup analyses demonstrated consistent associations across maternal age, BMI, parity, plurality, and IVF status, without significant interaction.ConclusionsElevated CTGI in early pregnancy is independently and nonlinearly associated with an increased risk of preeclampsia, particularly above a critical threshold of 2.244. These findings underscore the potential clinical value of CTGI as an early risk stratification biomarker for PE, enabling timely intervention in high-risk pregnancies. Graphical Abstract Supplementary InformationThe online version contains supplementary material available at 10.1186/s12944-025-02829-7.Keywords: Pre-Eclampsia, Dose-response relationship, Risk assessment Background Preeclampsia (PE) is a leading cause of maternal and perinatal mortality worldwide, accounting for approximately 14% of maternal deaths and 10%–25% of perinatal deaths [1]. It is strongly associated with increased risks of intrauterine growth restriction, preterm birth, and perinatal mortality, including a fivefold higher risk of foetal death [2, 3]. Women who survive PE often face reduced life expectancy and elevated risks of stroke, cardiovascular disease, and diabetes [4, 5]. Similarly, infants born to mothers with PE are more likely to experience prematurity, perinatal death, neurodevelopmental disorders, and long-term cardiovascular and metabolic conditions [6]. Given its severe outcomes and enduring effects on both maternal and offspring health, PE represents a significant global public health burden. Early identification of at-risk individuals is therefore critical for timely intervention and improved prognosis. Various early prediction strategies for PE have been explored. Large-scale clinical trials have demonstrated that combining early pregnancy biomarkers (e.g., placental growth factor [PlGF]) with uterine artery Doppler screening can increase the detection rate of early-onset PE (< 37 weeks) and facilitate the prevention of some cases through low-dose aspirin administration [7, 8]. However, these strategies have shown limited effectiveness in predicting term PE [7]. Existing clinical prediction tools include antenatal blood pressure monitoring, serum PlGF measurement, and uterine artery Doppler velocimetry [9]. However, these methods require specialized equipment or costly reagents and have not achieved optimal sensitivity for overall PE prediction. For instance, although mid-gestational PlGF levels may indicate impending PE, the specificity and sensitivity vary considerably in routine screening. Similarly, uterine artery Doppler ultrasound alone typically detects fewer than half of PE cases. These limitations highlight the need for simple, accessible, and sensitive biomarkers to complement current screening methods. Inflammation is recognized as a fundamental mechanism in the pathogenesis of various diseases, including cardiovascular, renal, and autoimmune disorders and preeclampsia [10–12]. Given the associations between PE, inflammation, and metabolic dysregulation, the C-reactive protein–triglyceride–glucose index (CTGI, also referred to as the CTI) has been proposed as an integrated predictive indicator [13, 14]. This index incorporates three routinely available clinical markers—serum C-reactive protein (CRP), fasting plasma glucose (FPG), and serum triglycerides (TGs)—to quantify metabolic–inflammatory burden. The CTGI has been suggested as a theoretically suitable marker for early maternal risk assessment because of its ability to reflect both inflammation and insulin resistance. Emerging evidence has indicated that the CTGI has predictive value for cardiovascular and metabolic diseases [15, 16]. However, limited research has been conducted on the CTGI in pregnant individuals, and its value in early pregnancy for predicting PE remains unclear. To address this gap, the objectives of this study were to evaluate the association between CTGI levels in early pregnancy and subsequent PE development, assess the role of the CTGI as an independent predictive factor, investigate potential nonlinear threshold effects, and determine the stability of this index across different population subgroups. These findings can provide insight into the potential application of the CTGI in early PE risk screening and contribute to the improvement of maternal health care strategies. Methods Study population and design This retrospective cohort study included 11,916 pregnant women who underwent routine prenatal screening at the Affiliated Suzhou Hospital of Nanjing Medical University (Suzhou, China) between 2015 and 2024. Women with incomplete clinical or laboratory data, a history of chronic hypertension, foetal demise, termination of pregnancy, or spontaneous miscarriage were excluded. Participants with nonpreeclamptic pregnancies complicated by intrauterine growth restriction or preterm birth were also excluded. This exclusion was applied to reduce confounding results from obstetric complications that may independently alter maternal metabolic or inflammatory profiles. The final analytic sample comprised 486 individuals who developed preeclampsia and 11,430 who did not. Exposure measurement: the CRP–triglyceride–glucose index (CTGI) Fasting blood samples were collected before 14 weeks of gestation to obtain values for CRP, triglyceride, and glucose levels. The CTGI, a composite marker of systemic inflammation and metabolic stress, was calculated using the following formula: CTGI = ln [CRP (mg/L) × triglycerides (mmol/L) × glucose (mmol/L)/2]. The denominator of 2 was adopted by analogy to the calculation of the TyG index, where it serves as a scaling factor to stabilize the distribution of values and facilitate comparability across populations. This adjustment does not change the relative ranking of individuals but provides consistency with established metabolic indices. Participants were stratified into quartiles on the basis of the distribution of CTGI values. The primary exposure was the CTGI value, which was analysed both as a continuous variable (including in its ln-transformed form) and by quartiles. Outcome definition The primary outcome was the development of preeclampsia, defined according to the American College of Obstetricians and Gynecologists (ACOG) criteria as new-onset hypertension (systolic blood pressure ≥ 140 mm Hg or diastolic blood pressure ≥ 90 mm Hg) after 20 weeks of gestation, accompanied by proteinuria (≥ 1 + on a urine dipstick on two occasions, ≥ 300 mg per 24-hour urine collection, or a urine protein-to-creatinine ratio ≥ 30 mg/mmol). In the absence of proteinuria, preeclampsia was diagnosed if hypertension occurred with evidence of maternal organ dysfunction (e.g., thrombocytopenia, renal insufficiency, liver dysfunction, pulmonary oedema, or neurologic symptoms). Covariates Demographic and clinical covariates extracted from medical records included maternal age, body mass index (BMI), gravidity, parity, use of in vitro fertilization (IVF), and foetal plurality (singleton vs. multifoetal gestation). These were included as potential confounders in the regression models. Statistical analyses Descriptive statistics were used to summarize the baseline characteristics. Continuous variables are presented as medians with interquartile ranges (IQRs) and were compared using the Mann–Whitney U test. Categorical variables are expressed as counts and percentages and were compared using the chi-square test. Univariate logistic regression was first used to evaluate the association between individual covariates and the odds of developing preeclampsia. Multivariable logistic regression models were then constructed to estimate adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for the CTGI, adjusting for maternal age, BMI, parity, IVF use, and multifoetal gestation. To examine the dose–response relationship, the CTGI was analysed across quartiles in three models: Model 1: unadjusted; Model 2: adjusted for maternal age, BMI, and parity; and Model 3: adjusted for IVF and multifoetal pregnancy. To assess the trend across CTGI quartiles, quartiles were treated as an ordinal variable by assigning a numeric value (using the median value of the CTGI within each quartile as the score for that group) and included as a continuous variable in the logistic regression models. The P value from this analysis represents the test for linear trends (P for trend) across quartiles. Restricted cubic spline (RCS) regression was applied to assess nonlinear associations between the CTGI and preeclampsia risk. The likelihood ratio test was used to compare models with and without spline terms. Two-piecewise linear regression was performed to detect potential threshold effects, with the inflection point determined by a recursive algorithm. Subgroup analyses were conducted to evaluate the robustness of the CTGI–preeclampsia association across strata defined by maternal age (< 35 vs. ≥35 years), obesity status (BMI < 25 vs. ≥25 kg/m²), parity (nulliparous vs. multiparous), plurality (singleton vs. multifoetal gestation), and conception methods (natural vs. IVF). To test for effect modification, an interaction term between the CTGI and each stratification variable was included in the multivariable logistic regression models. The resulting P value represents the test for interaction (P for interaction). Interaction terms were considered statistically significant at a P for interaction < 0.05. A total of 6524 pregnant women were included. Maternal blood samples were obtained during the first trimester, and PlGF levels and mean arterial pressure (MAP) were measured. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminative performance of the CTGI, PlGF, and MAP for the early prediction of PE. The area under the curve (AUC) with 95% CIs was calculated for each marker. Comparisons of AUCs were conducted using the DeLong test. All analyses were performed using R software, and a two-sided P <.05 was considered to indicate statistical significance. Results Baseline characteristics of the participants A total of 11,916 pregnant women were included in the analysis, comprising 11,430 without preeclampsia and 486 with preeclampsia. The median gestational age at delivery was significantly lower for women with preeclampsia (37.29 weeks [IQR: 35.29–39.00]) than for those without preeclampsia (39.57 weeks [IQR: 39.00–40.29]; Z = − 24.16, P <.001). Maternal age was slightly but significantly greater in the preeclampsia group (32.00 years [IQR: 29.00–35.00]) than in the nonpreeclampsia group (31.00 years [IQR: 28.00–34.00]; Z = − 4.14, P <.001) (Table 1). Table 1Baseline maternal and pregnancy characteristics in women with and without preeclampsiaVariablesTotal (n = 11916)0 (n = 11430)1 (n = 486)Statistic P Gestational Age at Delivery, M (Q₁, Q₃)39.43 (38.86, 40.14)39.57 (39.00, 40.29)37.29 (35.29, 39.00)Z=−24.16 <.001 Age, M (Q₁, Q₃)31.00 (28.00, 34.00)31.00 (28.00, 34.00)32.00 (29.00, 35.00)Z=−4.14 <.001 BMI, M (Q₁, Q₃)21.86 (20.20, 23.86)21.79 (20.20, 23.73)24.03 (21.48, 26.56)Z=−12.83 <.001 Gravidity, M (Q₁, Q₃)2.00 (1.00, 2.00)2.00 (1.00, 2.00)2.00 (1.00, 2.00)Z=−0.760.450CTGI, M (Q₁, Q₃)1.36 (0.93, 1.79)1.35 (0.92, 1.77)1.74 (1.26, 2.24)Z=−11.79 <.001 Parity, M (Q₁, Q₃)0.00 (0.00, 1.00)0.00 (0.00, 1.00)0.00 (0.00, 1.00)Z=−3.56 <.001 Multifetal pregnancyχ²=289.67 <.001 No11,737 (98.50)11,303 (98.89)434 (89.30) Yes179 (1.50)127 (1.11)52 (10.70)IVF pregnanciesχ²=155.21 <.001 No10,826 (90.85)10,462 (91.53)364 (74.90) Yes1090 (9.15)968 (8.47)122 (25.10)Z: Mann-Whitney test, χ²: Chi-square testM: Median, Q₁: 1 st Quartile, Q₃: 3 st Quartile, Bold indicates P value < 0.05 Baseline maternal and pregnancy characteristics in women with and without preeclampsia M: Median, Q₁: 1 st Quartile, Q₃: 3 st Quartile, Bold indicates P value < 0.05 BMI values were significantly greater among women who developed preeclampsia (24.03 [IQR: 21.48–26.56]) than among those who did not (21.79 [IQR: 20.20–23.73]; Z = − 12.83, P <.001), indicating a greater prevalence of overweight or obesity in the preeclampsia group. Similarly, before the 14th week of gestation—prior to the clinical onset of preeclampsia—the CRP–triglyceride–glucose index, a composite marker reflecting metabolic and inflammatory stress, was significantly elevated in the preeclampsia group (1.74 [IQR: 1.26–2.24]) compared with the nonpreeclampsia group (1.35 [IQR: 0.92–1.77]; Z = − 11.79, P <.001) (Fig. 1). Gravidity did not differ significantly between the two groups (both medians = 2.00; Z = − 0.76, P =.450). However, parity was slightly lower in the preeclampsia group (Z = − 3.56, P <.001), suggesting a higher proportion of nulliparous women among those who developed the condition (Table 1). Association Between First-Trimester CTGI and Risk of Preeclampsia With respect to the categorical variables, the proportion of multifoetal pregnancies was substantially greater among women with preeclampsia (10.70%) than among those without preeclampsia (1.11%) (χ² = 289.67, P <.001). Additionally, the use of IVF was more frequent in the preeclampsia group (25.10%) than in the nonpreeclampsia group (8.47%) (χ² = 155.21, P <.001), reflecting a greater reliance on assisted reproductive technology in this population. Logistic regression analysis of factors associated with preeclampsia According to the univariate analysis, multiple pregnancy (OR: 10.66; 95% CI: 7.62–14.93), IVF (OR: 3.62; 95% CI: 2.92–4.49), maternal age (OR per year increase: 1.05; 95% CI: 1.03–1.08), BMI (OR per unit increase: 1.23; 95% CI: 1.20–1.26), and the C-reactive protein–triglyceride–glucose (CRP–TG–glucose) index (OR per unit increase: 2.54; 95% CI: 2.21–2.93) were all significantly associated with increased odds of preeclampsia. In contrast, parity was associated with reduced odds (OR: 0.74; 95% CI: 0.61–0.89) (Table 2). Table 2Univariate and multivariate logistic regression analyses of maternal and Pregnancy-Related factors associated with preeclampsiaVariablesUnivariate analysisMultivariate analysisβS.EZ P OR (95%CI)βS.EZ P OR (95%CI)Parity−0.310.10−3.16 0.002 0.74 (0.61 ~ 0.89)−0.580.12−5.01 <.001 0.56 (0.45 ~ 0.70)Multifetal pregnancy No1.00 (Reference)1.00 (Reference) Yes2.370.1713.78 <.001 10.66 (7.62 ~ 14.93)1.880.199.73 <.001 6.56 (4.49 ~ 9.59)IVF pregnancies No1.00 (Reference)1.00 (Reference) Yes1.290.1111.71 <.001 3.62 (2.92 ~ 4.49)0.710.135.56 <.001 2.04 (1.59 ~ 2.62)Age0.050.014.43 <.001 1.05 (1.03 ~ 1.08)0.050.013.47 <.001 1.05 (1.02 ~ 1.08)BMI0.200.0115.86 <.001 1.23 (1.20 ~ 1.26)0.160.0111.22 <.001 1.18 (1.14 ~ 1.21)CTGI0.930.0712.95 <.001 2.54 (2.21 ~ 2.93)0.580.087.00 <.001 1.78 (1.51 ~ 2.09)OR Odds Ratio, CI Confidence Interval, Bold indicates P value < 0.05 Univariate and multivariate logistic regression analyses of maternal and Pregnancy-Related factors associated with preeclampsia OR Odds Ratio, CI Confidence Interval, Bold indicates P value < 0.05 In the multivariate model, after adjusting for potential confounders, the CRP–TG–glucose index remained significantly associated with preeclampsia (adjusted OR: 1.78; 95% CI: 1.51–2.09; P <.001), indicating that it is an independent risk factor. Multiple pregnancy (adjusted OR: 6.56; 95% CI: 4.49–9.59; P <.001), IVF (adjusted OR: 2.04; 95% CI: 1.59–2.62; P <.001), maternal age (adjusted OR per year: 1.05; 95% CI: 1.02–1.08; P <.001), and BMI (adjusted OR per unit: 1.18; 95% CI: 1.14–1.21; P <.001) also remained independently associated with preeclampsia. Parity continued to indicate a protective effect (adjusted OR: 0.56; 95% CI: 0.45–0.70; P <.001) (Table 2). These findings suggest that higher systemic inflammatory and metabolic stress, as indicated by the CRP–TG–glucose index, is significantly associated with the risk of preeclampsia, independent of maternal age, BMI, parity, use of IVF, and multifoetal gestation. Association between the CRP–triglyceride–glucose index (CTGI) and the risk of preeclampsia The association between increasing quartiles of the CTGI and the odds of preeclampsia was evaluated across three logistic regression models of progressive adjustment (Table 3). Table 3Multivariable logistic regression analyses assessing the association between First-Trimester CTGI and subsequent risk of preeclampsia (Bold indicates P value < 0.05)VariablesModel 1Model 2Model 3OR (95%CI) P OR (95%CI) P OR (95%CI) P CTGI2.54 (2.21 ~ 2.93) <.001 2.04 (1.74 ~ 2.39) <.001 1.81 (1.54 ~ 2.12) <.001 CTGI (Quartile) 11.00 (Reference)1.00 (Reference)1.00 (Reference) 21.25 (0.90 ~ 1.74)0.1811.15 (0.82 ~ 1.60)0.4231.11 (0.79 ~ 1.55)0.552 31.78 (1.31 ~ 2.43) <.001 1.51 (1.10 ~ 2.07) 0.010 1.36 (0.99 ~ 1.87)0.059 43.68 (2.78 ~ 4.87) <.001 2.48 (1.83 ~ 3.36) <.001 2.06 (1.52 ~ 2.81) <.001 P for trend< 0.001< 0.001< 0.001 Multivariable logistic regression analyses assessing the association between First-Trimester CTGI and subsequent risk of preeclampsia (Bold indicates P value < 0.05) In the unadjusted model (Model 1), higher CTGI quartiles were significantly associated with increased odds of preeclampsia. Compared with the lowest quartile (reference), the odds ratios (ORs) for preeclampsia were 1.25 (95% CI: 0.90–1.74) for quartile 2 (P =.181), 1.78 (95% CI: 1.31–2.43) for quartile 3 (P <.001), and 3.68 (95% CI: 2.78–4.87) for quartile 4 (P <.001), demonstrating a dose–response relationship. After adjusting for maternal age, body mass index (BMI), and parity in Model 2, the association remained significant for quartile 3 (OR: 1.51; 95% CI: 1.10–2.07; P =.010) and quartile 4 (OR: 2.48; 95% CI: 1.83–3.36; P <.001). Further adjustment in Model 3 for additional confounders, including multifoetal gestation and the use of IVF, slightly attenuated the associations. The ORs were 1.11 (95% CI: 0.79–1.55; P =.552) for quartile 2, 1.36 (95% CI: 0.99–1.87; P =.059) for quartile 3, and 2.06 (95% CI: 1.52–2.81; P <.001) for quartile 4. These findings indicate that higher CTGI levels are independently associated with an elevated risk of preeclampsia, particularly among individuals in the highest quartile, even after adjustment for multiple maternal and obstetric factors. Threshold effect of the CRP–triglyceride–glucose index (CTGI) on the risk of preeclampsia To explore the potential nonlinear relationship between the CTGI and the odds of preeclampsia, restricted cubic spline (RCS) regression models were fitted. Visual inspection of the RCS curve (Fig. 2) revealed a nonlinear increasing trend in the risk of preeclampsia with increasing CTGI levels, particularly above the median reference point. After adjusting for maternal age, BMI, parity, multifoetal pregnancy, and IVF status, this positive nonlinear association remained evident. Nonlinear Association Between First-Trimester CTGI and Risk of preeclampsia Based on Restricted Cubic Spline Regression. A shows the unadjusted model, in which a strong nonlinear association was observed (P for overall < 0.001; P for nonlinearity < 0.001). B displays the fully adjusted model, controlling for maternal age, body mass index (BMI), parity, in vitro fertilization (IVF), and multifetal pregnancy. The nonlinear association persisted after adjustment (P for overall < 0.001; P for nonlinearity = 0.001), although attenuated in magnitude To further characterize the relationship between the CTGI and preeclampsia risk, a threshold effect analysis was performed using two-piecewise linear regression based on the segmented R package. Evidence of a threshold effect was identified, as indicated by the likelihood ratio test (P =.005), suggesting a nonlinear relationship with a significant inflection point at a CTGI value of 2.244. For CTGI values less than 2.244, the association remained positive but was more modest (OR: 1.45; 95% CI: 1.16–1.82; P =.001). For CTGI values equal to or above 2.244, the association was substantially stronger (OR: 3.93; 95% CI: 2.09–7.39; P <.001) (Table 4). These findings suggest a threshold-dependent effect of the CTGI on preeclampsia risk, with disproportionately higher risk observed beyond a CTGI value of 2.244. This nonlinear pattern underscores the clinical relevance of identifying high-risk individuals on the basis of CTGI stratification. Table 4Threshold effect of the First-Trimester CTGI on risk of preeclampsia based on standard and Two-Piecewise linear regression models (Bold indicates P value < 0.05)Outcomeeffect P Model 1 Fitting model by standard linear regression1.78 (1.51–2.09) <.001 Model 2 Fitting model by two-piecewise linear regressionInflection point2.24 <2.241.45 (1.16–1.82) 0.001 ≥2.243.93 (2.09–7.39) <.001 P for likelihood test 0.005 Threshold effect of the First-Trimester CTGI on risk of preeclampsia based on standard and Two-Piecewise linear regression models (Bold indicates P value < 0.05) Subgroup analyses The association between the CTGI and the risk of preeclampsia was consistently observed across multiple clinically relevant subgroups (Table 5). In the overall population, a higher CTGI value was significantly associated with increased odds of preeclampsia (OR: 2.54; 95% CI: 2.21–2.93; P <.001). No significant interaction was observed between the CTGI and any of the examined subgroups, indicating that the association between elevated CTGI values and increased risk of preeclampsia was consistent across maternal age, obesity status, parity, plurality, and conception methods. Table 5Subgroup analyses of the association between First-Trimester CTGI and risk of preeclampsiaVariablesn (%)OR (95%CI) P P for interactionAll patients11,916 (100.00)2.54 (2.21 ~ 2.93) <.001 Age0.435 <359427 (79.11)2.60 (2.20 ~ 3.07) <.001 ≥352489 (20.89)2.29 (1.74 ~ 3.00) <.001 BMI0.344 <3011,695 (98.15)2.23 (1.92 ~ 2.60) <.001 ≥30221 (1.85)3.03 (1.62 ~ 5.67) <.001 Parity0.600 Primipara(0)7873 (66.07)2.84 (2.40 ~ 3.36) <.001 Multipara(> 0)4043 (33.93)3.11 (2.32 ~ 4.15) <.001 Multifetal pregnancy0.343 No11,737 (98.50)2.48 (2.14 ~ 2.88) <.001 Yes179 (1.50)1.91 (1.14 ~ 3.19) 0.014 IVF pregnancies0.146 No10,826 (90.85)2.55 (2.16 ~ 3.00) <.001 Yes1090 (9.15)2.00 (1.51 ~ 2.65) <.001 OR Odds Ratio, CI Confidence Interval, Bold indicates P value < 0.05 Subgroup analyses of the association between First-Trimester CTGI and risk of preeclampsia OR Odds Ratio, CI Confidence Interval, Bold indicates P value < 0.05 Comparison of first-trimester CTGI, PlGF, and MAP values for the early prediction of PE PlGF and MAP levels were measured in 6,524 pregnant women during the first trimester. To evaluate the diagnostic utility of the CTGI as an early biomarker for PE, its discriminative performance was compared with that of PlGF and the clinical indicator MAP. The goal was to determine whether the CTGI, a simple inflammation-based index derived from routine laboratory tests, could serve as an independent or complementary tool to existing predictors. With respect to diagnostic performance, ROC analysis revealed that the CTGI demonstrated modest diagnostic ability, with an AUC of 0.658 (95% CI: 0.623–0.694). However, its performance was inferior to that of both PlGF (AUC: 0.735; 95% CI: 0.689–0.781; P =.011 by the DeLong test compared with the CTGI) and MAP (AUC: 0.768; 95% CI: 0.746–0.790, P = 2.7e-07 by the DeLong test compared with the CTGI) (Fig. 3). When optimal cut-off values (CTGI: 1.682; PlGF: 0.859; and MAP: 0.959) were applied, the CTGI achieved a sensitivity of 70.3% and specificity of 55.9%, whereas PlGF demonstrated higher sensitivity (91.8%) but moderate specificity (66.4%). In contrast, MAP yielded a specificity of 87.3% but only moderate sensitivity (64.8%). In summary, the CTGI reflects a complementary inflammatory indicator in the pathophysiology of PE. Owing to its accessibility, low cost, and reliance on standard laboratory measurements, the CTGI may represent a scalable biomarker for early pregnancy PE screening, particularly in resource-limited settings or as a component of multimodal risk-prediction strategies. Comparison of diagnostic performance of first-trimester biomarkers for early prediction of preeclampsia Discussion Preeclampsia is now understood as a multisystem disorder with significant metabolic and inflammatory underpinnings [17]. In this study, the C-reactive protein–triglyceride–glucose index (CTGI)—a composite marker incorporating systemic inflammation and insulin resistance—has attracted attention as a potential predictor of preeclampsia risk. Although the CTGI has been studied in cardiovascular and metabolic contexts, its application in obstetrics remains limited. To our knowledge, this is the first large-scale study that evaluates the CTGI specifically for preeclampsia, underscoring its novelty. These findings demonstrate a significant, independent, and nonlinear association between first-trimester CTGI values and subsequent preeclampsia risk, which can be interpreted through well-established pathophysiological pathways. The rationale behind the CTGI is grounded in known pathophysiology. Preeclampsia is characterized by endothelial dysfunction and abnormal placental perfusion and is often preceded by maternal metabolic disturbances and systemic inflammation. Insulin resistance (captured by TyG) and chronic inflammation (captured by CRP) can together drive the cascade of events leading to preeclampsia. This biological plausibility is consistent with our epidemiologic observation of a dose–response gradient, where higher CTGI quartiles were associated with progressively increased preeclampsia risk. The identified threshold effect, with a markedly sharper increase in risk above a CTGI value of 2.244, is particularly mechanistically informative. This implies the existence of a critical level of metabolic–inflammatory stress, beyond which homeostatic mechanisms are overwhelmed, potentially triggering more pronounced pathophysiological derangement. However, this cut-off should be considered exploratory, as it may be overfit to our dataset and requires external validation before any clinical translation. Insulin resistance (IR) during pregnancy promotes a cluster of pro-preeclampsia effects; it is associated with hyperglycaemia, dyslipidaemia, and overweight—all of which are independent risk factors for preeclampsia [18]. IR can adversely affect the vascular endothelium by promoting oxidative stress and altering nitric oxide bioavailability, exacerbating endothelial dysfunction. Moreover, IR is accompanied by elevated secretion of inflammatory cytokines and adipokines. In women with insulin resistance, adipose tissue (and even the placenta) releases higher amounts of tumour necrosis factor-α, leptin, and other mediators. This chronic proinflammatory milieu contributes to the maternal systemic inflammation observed before overt preeclampsia [19, 20]. Both mechanisms—endothelial injury and inflammation—converge to impair placental development (e.g., shallow trophoblast invasion and insufficient spiral artery remodelling) and to provoke the clinical manifestations of preeclampsia. The CTGI essentially serves as a surrogate indicator of these two synergistic processes. A high CTGI value early in pregnancy may indicate that a patient has significant underlying insulin resistance and an enhanced inflammatory response, which create an environment for the later development of hypertension, proteinuria, and organ dysfunction, which are characteristics of preeclampsia. This finding is consistent with epidemiologic links: women with metabolic syndrome or gestational diabetes (extreme insulin resistance states) have higher rates of preeclampsia, and women with elevated inflammatory markers are likewise at increased risk [21, 22]. This suggests that these factors are not merely consequences of disease but also part of its pathogenesis. Furthermore, the consistent association observed across all subgroups, despite the limited ability to detect interactions, suggests that the fundamental pathophysiological pathway captured by the CTGI, metabolic–inflammatory stress, is a universal contributor to preeclampsia risk and is applicable across diverse maternal characteristics. Compared with any single biomarker, the CTGI offers a more comprehensive risk assessment for preeclampsia by integrating inflammatory and metabolic statuses. Another practical advantage of the CTGI is its reliance on routine laboratory measurements (fasting glucose, triglyceride, and CRP levels), which are widely available. Unlike more specialized biomarkers (e.g., PlGF or uterine artery Doppler indices used in formal first-trimester screening), the CTGI can be calculated using standard prenatal blood tests, facilitating its incorporation in diverse clinical settings. Although the CTGI demonstrated lower discriminatory performance than PlGF and MAP did, it captures distinct pathophysiological processes that are not fully represented by these established markers. Thus, the CTGI may serve as a complementary tool within multimodal prediction models, particularly in resource-constrained environments where accessibility and cost-effectiveness are critical. Study strengths and limitations This study provides a large-scale, retrospective cohort analysis investigating the novel C-reactive protein–triglyceride–glucose index (CTGI) as an early predictor of preeclampsia. The research leverages a substantial clinical population and employs advanced statistical modelling, including restricted cubic spline and threshold effect analysis, to elucidate a significant, independent, and nonlinear association between first-trimester CTGI and preeclampsia risk. The findings underscore the potential clinical utility of CTGI as an accessible, low-cost composite biomarker for early risk stratification, which could facilitate timely interventions in high-risk pregnancies, particularly in resource-constrained settings or as part of multimodal prediction strategies. Nevertheless, this study is subject to certain limitations. First, the retrospective, single-center design may introduce selection bias and limits generalizability; the findings and the identified risk threshold require validation in prospective, multicenter cohorts. Second, while key confounders were adjusted for, residual confounding from unmeasured lifestyle and socioeconomic factors cannot be excluded. Third, the CTGI can be influenced by acute inflammation from non-obstetric causes, potentially affecting its specificity. Finally, the exclusion of women with other obstetric complications to reduce heterogeneity may limit the applicability of the results to broader pregnant populations. Conclusions This large retrospective cohort study evaluated the novel C-reactive protein–triglyceride–glucose index (CTGI) as an early predictor of preeclampsia in 11,916 pregnant women. Advanced modelling identified a significant, independent, and nonlinear association between first-trimester CTGI and preeclampsia risk, with a critical threshold at 2.244. These findings highlight the potential of CTGI, a composite index derived from routine blood tests, to serve as a practical, low-cost tool for early identification of high-risk pregnancies. This may enable more targeted surveillance and timely intervention, ultimately contributing to the prevention of severe maternal complications and improved health outcomes for mothers and newborns. The accessibility of the CTGI could support broader and more equitable implementation of early risk assessment, particularly in diverse healthcare settings. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements The authors express their gratitude to the participants for their invaluable contributions to this study. Abbreviations Preeclampsia C-reactive protein-triglyceride-glucose Placental growth factor C-reactive protein Fasting plasma glucose Serum triglycerides American College of Obstetricians and Gynecologists Body mass index In vitro fertilization Interquartile ranges Adjusted odds ratios Confidence intervals Restricted cubic spline Insulin resistance Authors’ contributions Yuting Liang: conceptualization, formal analysis, writing—original draft, supervision, funding acquisition; Yanqiu Zhang: methodology, data curation, formal analysis, writing—review and editing; Yujing Li: investigation, writing—review and editing; Jun Cao: investigation, writing—review and editing; Bin Feng: investigation, writing—review and editing; Jieyu Jin: investigation, writing—review and editing; Sheng Zhang: investigation, writing—review and editing; Qingqin Tang: investigation, writing—review and editing; Longwei Qiao: conceptualization, writing—review and editing, funding acquisition; Zhixing Jin: formal analysis, writing—original draft, supervision, funding acquisition. Funding This study was supported by Science Foundation of Jiangsu Province Grant BK20240371, Suzhou Health Talent Program (GSWS2024046, GSWS2021004), Suzhou Key Clinical Technology Research (SKY2023001), the National Natural Science Foundation of China (Grant No. 82001576, No.82001523), the Primary Research & Development Plan of Jiangsu Province (BE2022736), Suzhou Medical College of Soochow University - Qilu Medical Research Fund (24QL200205), and the Jiangsu Province College Students’ Innovation and Entrepreneurship Training Program Project (202410285273Y). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate This study protocol was reviewed and approved by the Medical Ethics Committee of the Affiliated Suzhou Hospital of Nanjing Medical University (Suzhou, China). All participants provided broad informed consent. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Longwei Qiao, Email: qiaolongwei1@126.com. Zhixing Jin, Email: jinzhixing131@hotmail.com. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement No datasets were generated or analysed during the current study.