Clinical prediction model for gestational diabetes mellitus utilizing thyroid function indicators: a retrospective cohort study

Clinical prediction model for gestational diabetes mellitus utilizing thyroid function indicators: a retrospective cohort study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Clinical prediction model for gestational diabetes mellitus utilizing thyroid function indicators: a retrospective cohort study Xiaoqing Li, Yanjun Hu, Haiying Chen, Chunling Chen, Chanchan Liao, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7192354/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Our objective is to explore the relationship between gestational diabetes mellitus (GDM) and thyroid hormones, and construct a clinical prediction model based on the clinical features of GDM and thyroid parameters. Methods A population-based retrospective cohort study, including 1,035 GDM patients and 4,194 healthy control. Statistical tests were conducted to evaluate the associations between primary risk factors, including age, family history of diabetes, gestational hypertension, hypertension family history, thyroglobulin antibody (TGAb), and thyroid hormones (TT3, TSH, FT3, TPOAb), with GDM risk. Results In this study, age, family history of diabetes, gestational hypertension, hypertension family history, and TGAb concentration were identified as primary risk factors. The first four risk factors showed a positive associated with GDM, while height and TGAb concentration were significantly negatively correlated with GDM risk. Additionally, lower levels of total triiodothyronine (TT3) were associated with an increased risk of GDM in all patients, while consistently lower levels of thyroid-stimulating hormone (TSH) also heightened GDM risk. In the TGAb-negative group, higher levels of TT3 and TSH were linked to reduced risk of GDM, whereas lower levels of free triiodothyronine (FT3) were associated with an increased risk. In the TGAb-positive group, thyroid peroxidase antibody (TPOAb) had a strong positive association with GDM. Conclusions Thyroid hormones play a crucial role in pregnancy and may counteract insulin, affecting blood glucose balance. Therefore, changes in thyroid parameters should be appropriately considered in the prevention and screening of GDM. Gestation Diabetes mellitus Thyroid Hormones Machine learning Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Gestational diabetes mellitus (GDM) is defined as the first-time detection or occurrence of varying degrees of glucose intolerance during pregnancy, making it the most common metabolic disorder in the later stages of pregnancy ( 1 – 3 ). According to the International Diabetes Federation (IDF), the global prevalence of GDM is approximately 14.0%, with rates exceeding 20% in regions like the Middle East, North Africa, and Southeast Asia ( 4 ). In a study conducted in East Asia and Southeast Asia, China ranked third in terms of incidence, with nearly one in nine women experiencing GDM ( 5 ). Data from Chinese mainland indicate an overall incidence of 14.8% ( 6 ), with evidence suggesting a continuous increase in recent years. Factors such as rising obesity rates among women of childbearing age and increased maternal age contribute to the annual increase in GDM incidence ( 7 ). Research reveals that GDM affects approximately 15% of pregnancies worldwide each year, impacting around 180,000 newborns ( 8 ). GDM not only increases the likelihood of cesarean section and complications like hypertension during pregnancy but also poses risks to infants, including preterm birth, respiratory distress, and jaundice ( 9 ). Follow-up studies on women with postpartum GDM show that their offspring have a higher risk of developing diabetes, obesity, and cardiovascular diseases ( 10 , 11 ). Currently, GDM is typically diagnosed in the late stages of pregnancy (24–28 weeks) through the Oral Glucose Tolerance Test (OGTT). However, this method limits the early diagnosis, monitoring, and prediction of disease outcomes ( 12 , 13 ). As pregnancy is a dynamic and complex process, relying solely on a single indicator at a specific time point cannot fully capture the risk factors influencing GDM in expectant mothers. During pregnancy, hormonal fluctuations, including changes in glucose homeostasis and thyroid function, are common ( 14 ). Thyroid hormones directly affect insulin secretion and glucose metabolism. Early pregnancy thyroid dysfunction (TD) may predict impaired glucose tolerance later in pregnancy ( 15 ). A significant positive correlation has been observed between TD and GDM in women ( 16 ). Subclinical hypothyroidism with positive anti-thyroid autoantibodies significantly increases the risk of GDM ( 17 , 18 ), while pregnant women with subclinical hyperthyroidism are less likely to develop GDM ( 19 ). Recent studies indicate that even elderly pregnant women with normal thyroid function, but the presence of thyroid autoimmunity are at a higher risk of developing GDM ( 20 ). Some scholars have proposed focusing on TD patients in the early and mid-stages of pregnancy to predict GDM ( 21 , 22 ). Evidence suggests that using thyroid indicators to predict GDM and post-OGTT blood glucose levels is effective ( 23 ). However, current studies often consider only a few thyroid indicators ( 24 – 28 ), lacking comprehensive analysis and an investigation into the combined effects of multiple parameters. Clinical predictive models built on machine learning algorithms in the field of artificial intelligence can help comprehend or explain the causal mechanisms underlying exposure-outcome relationships ( 29 – 32 ). Increasingly, research is utilizing machine learning algorithms to identify risk factors for GDM and develop early prediction models ( 33 ). For instance, Artzi et al. achieved accurate and rapid stratification and prediction of GDM using machine learning methods with retrospective cohort data from 588,622 Israeli pregnant women ( 34 ). Beyond conventional clinical features like routine examination indicators and medical history ( 35 ), GDM classification prediction models have also been constructed using bacterial biomarkers from the oral microbiome ( 36 ) and endogenous metabolites ( 37 ). However, the exploration of the correlation between thyroid indicators and GDM has largely been confined to statistical data analysis ( 38 , 39 ), with limited research focused on constructing clinical predictive risk models for GDM using machine learning algorithms. 2. Materials and Methods 2.1 Study population This retrospective case study was conducted on Chinese pregnant women utilizing data gathered from the People's Hospital of Wenzhou City in Zhejiang Province. Participants included pregnant women who gave birth between January 2010 and June 2022, with all prenatal examinations registered at the hospital. The study focused on native females from southern China, with demographic and clinical characteristics extracted from the Hospital Information System (HIS). A total of 59,713 pregnant women were screened for inclusion. According to the American Diabetes Association (ADA) criteria ( 40 ), individuals with a history of GDM or pre-existing diabetes were excluded. Inclusion criteria were established for maternal basic information and hospital examination records (Fig. 1 ), which were independently screened by two researchers, using only intersected data. Ultimately, 5,229 eligible women were selected, comprising 1,035 cases in the GDM group and 4,194 controls in the comparison group, achieving a case-to-control ratio of approximately 1:4. The study was approved by the Ethics Committee of Wenzhou People's Hospital, and all participants provided informed consent (No: KY-2022-277). 2.2 GDM ascertainment The diagnosis of GDM in pregnant women was determined through retrospective medical record review. Women included in the study were required to fast for at least eight hours before consuming a 300 ml solution containing 75g of glucose between the 24th and 28th week of pregnancy. Based on the 2010 diagnostic criteria published by the International Association of Diabetes and Pregnancy Study Groups (IADPSG) ( 41 ), obstetricians diagnosed GDM using OGTT results. Blood glucose concentrations were measured before drinking the glucose solution, at 1 hour, and at 2 hours. Women with blood glucose levels below the thresholds of 5.1 mmol/L, 10.0 mmol/L, and 8.5 mmol/L (92mg/dL, 180mg/dL, 153mg/dL) were considered to have normal glucose tolerance. Those with glucose levels equal to or exceeding any of these thresholds were diagnosed with GDM. 2.3 Testing for thyroid function Serum samples were collected during the first prenatal examination. In this cohort study, a thyroid function analyzer was used to measure thyroid function markers, following the guidelines for diagnosis and treatment of thyroid diseases in China ( 42 ), and considering the specific laboratory conditions and testing methods used at our hospital. Normal ranges for thyroid function tests included total triiodothyronine (TT3) concentrations of 0.80 − 2.00 ng/ml and total thyroxine (TT4) concentrations of 51.3 − 140.6 ng/ml. The TT3/TT4 ratio was calculated by dividing the of TT3 concentration (ng/ml) by TT4 concentration (ng/ml). Free triiodothyronine (FT3) concentrations ranged from 2.8 − 7.1 pmol/L, and free thyroxine (FT4) concentrations ranged from 7.9 − 16.4 pmol/L. The FT3/FT4 ratio was calculated by dividing FT3 (pmol/L) by FT4 (pmol/L). Normal ranges for thyroid-stimulating hormone (TSH) concentrations were 0.38 − 5.33 mIU/L, and for thyroglobulin (TG) concentrations, 0.8 − 68.0 ng/ml. Thyroid peroxidase antibodies (TPOAb) and thyroglobulin antibodies (TGAb), which are associated with autoimmune thyroid conditions, were considered positive if concentrations were ≥ 35 IU/ml and ≥ 4.0 IU/ml, respectively. 2.4 Statistical analysis Real-world clinical data analysis was conducted using IBM SPSS Statistics 22. For normally distributed continuous variables such as age and height, means ± standard deviations (SD) were calculated, and independent samples t-tests were used to compare group means. For binary and categorical data, frequencies and percentages were used, with differences between groups analyzed using chi-square tests, such as for family history of diabetes, hypertension, thyroid function, and other pregnancy complications. Non-normally distributed continuous data were described using quartiles, and differences between groups were inferred using rank-sum tests, such as for thyroid indicators including TT3, TT4, and FT3. Seven thyroid indicators and two calculated ratios (TT3/TT4, FT3/FT4) underwent quartile-based continuous analysis. Logistic regression analysis was first conducted to evaluate the odds ratio (OR), 95% confidence intervals (CI), and P-values for each thyroid marker. Four factors showing significant differences in baseline analysis (age, family history of diabetes, pregnancy-induced hypertension, and family history of hypertension) were then included as covariates to obtain adjusted OR values (aORs). Since thyroid antibody TGAb levels may affect thyroid function and glucose homeostasis ( 43 ), TGAb was categorized into negative and positive subsets and included in the model. Logistic regression analysis of thyroid markers was performed again on both case and control groups to observe differences in ORs, aORs, confidence intervals, and P-values. A significance level of p < 0.05 was considered statistically significant. 2.5 Screening predictor variables and constructing GDM models Pearson and Spearman correlation coefficients were calculated for the 21 included features to assess their individual impact on GDM outcomes. These features included age, height, family history of diabetes, scarred uterus, gestational hypertension, family history of hypertension, family history of cancer, re-pregnancy, polycystic ovary syndrome, ovarian cysts, thyroid function, TT3, TT4, TT3/TT4, FT3, FT4, FT3/FT4, TSH, Tg, TPOAb, and TGAb. Both Pearson and Spearman coefficients were used to analyze the relationship between pairs of data columns. Further data analysis employed the Random Forest (RF) algorithm to calculate the contribution of included risk factors to clinical outcomes, assessing the relative importance of these features in relation to GDM outcomes. Grey Relational Analysis ( 44 ) was utilized to observe the influence of multifactor interactions on outcome indicators by eliminating low-contributing factors and incorporating clinically impactful ones,. The final cohort included only the most influential factors affecting clinical outcomes. The data were randomly divided into training and a testing sets with a ratio of 7:3. Four clinical prognostic models, including stepwise regression (SR), best subset selection (BSS), RF, and support vector machines (SVM), were constructed using R software (version 4.1.3). These models aimed to identify the most important clinical features and risk factors, with each algorithm applying its own criteria to determine output results. To evaluate the models, the area under the curve (AUC, 95% CI), sensitivity, specificity, true positive rate (TPR), true negative rate (TNR), concordance index (C-index, 95% CI), X-squared, degree of freedom (DF), and Hosmer and Lemeshow goodness-of-fit (GOF) test (P-value) were calculated. The calibration function was used to generate continuous recalibration plots in order to assess the consistency between predicted values and actual observations. Additionally, nomogram graphs were constructed for each of the four models to integrate multiple predictive indicators and illustrate the relationships between variables in the prediction models. Nomogram graphs assigned scores to different levels of each influential factor based on their contribution to the outcome variable (size of regression coefficients). These scores were summed to obtain a total score, and a mathematical transformation was applied to the total score to predict the probability of the outcome event. Decision curves and clinical impact curves were also plotted to evaluate the predictive abilities of different models. 3. Results 3.1 Baseline characteristics between GDM and control groups After screening 59,713 samples that met the study criteria, a total of 5,229 pregnant women were included in the final analysis. Based on the research data, there were 1,035 women in the GDM group and 4,194 women in the control group (non-GDM), resulting in a GDM prevalence of approximately 19.8% (1,035/5,229). The demographic and clinical characteristics of the overall study population in both groups were presented (Table 1). The analysis showed that, compared to the control group, women in the GDM group were significantly older (34±4.3 years vs 32±4.4 years, P < 0.001). Women with a family history of diabetes were more likely to develop GDM compared to those without such a history (4.7% vs 0.9%, P < 0.001). Pregnant women diagnosed with gestational hypertension also had a higher risk of GDM compared to those without this condition (8.4% vs 3.1%, P < 0.001). Additionally, a family history of hypertension increased the risk of GDM (6.2% vs 4.2%, P = 0.006), highlighting the need for preventive measures. There was a significant statisticant difference in the distribution of TGAb between the GDM and control groups (P = 0.001). Lastly, we portrayed the actual distribution of these five clinical factors ( Figure 2A-E ). No significant differences were observed in the remaining 17 characteristic factors between the two groups. 3.2 Univariate logistic regression analysis for thyroid function indicators To explore the relationship between thyroid function indicators and the risk of GDM, the indicators, except for TGAb, were divided into four quartiles (P > 0.05). The quartiles, based on data distribution at 25%, 50%, 75%, and 100% intervals, served as the basis for logistic regression analysis. Considering significant differences in age, family history of diabetes, gestational hypertension, and family history of hypertension between the GDM and the control groups, these four factors were adjusted as covariates in the multifactor logistic regression. According to the analysis results (Table 2), there was a significant difference in GDM incidence among the TT3 level quartiles (P < 0.013), with lower TT3 levels associated with an increased risk of GDM (ORQ1 = 1.255, 95% CI 1.040–1.514, p = 0.018). After adjusting for covariates, no significant difference was found in TT3 levels. Analysis of FT3 indicated that higher FT3 levels were negatively associated with GDM risk (ORQ3 = 0.784, 95% CI 0.646–0.951, p = 0.013), a relationship that remained significant after adjustment (aORQ1 = 0.800, 95% CI 0.656–0.975, p = 0.027). Regarding TSH levels, lower levels were linked to an increased risk of GDM (ORQ1 = 1.211, 95% CI 1.001–1.465, p = 0.049), with this association becoming more pronounced after adjustment (aORQ1 = 1.288, 95% CI 1.059–1.565, p = 0.011). These findings suggest that low levels of TSH increase GDM risk from both univariate and multivariate perspectives. Other factors, such as TT4, TT3/TT4, FT3, FT4, FT3/FT4, TPOAb, and Tg, did not show significant differences before and after adjustment (P > 0.05). 3.3 Univariate logistic regression analysis of TGAb negative and positive groups To examine the association between TGAb levels, other thyroid indicators, and GDM risk, the study population was divided into TGAb-positive (+, ≥4.0 IU/ml) and TGAb -negative (-,<4.0 IU/ml) groups. Other thyroid indicators were grouped into quartiles for logistic regression analysis, with four covariates included for model adjustment as described in section 3.2. The analysis sample comprised 3,275 TGAb- negative women and 1,954 TGAb+ positive women (Table 3). In the TGAb- negative group, significant differences were observed between the GDM and control groups in the Q2 range of TT3 levels (1.05-1.2), indicating that higher TT3 levels were associated with a reduced risk of GDM (ORQ2 = 0.684, 95% CI 0.549–0.852, p = 0.001; aORQ2 = 0.719, 95% CI 0.574–0.899, p = 0.004). Conversely, lower FT3/FT4 levels were linked to a higher risk of GDM (ORQ3 = 1.275, 95% CI 1.043–1.558, p = 0.018; aORQ3 = 1.323, 95% CI 1.077–1.625, p = 0.008). The impact of TSH levels on GDM risk aligned with the TT3 results, where higher TSH concentrations reduced the risk of GDM (ORQ2 = 0.735, 95% CI 0.580–0.932, p = 0.010; aORQ2 = 0.766, 95% CI 0.629–0.932, p = 0.008; aORQ4 = 0.744, 95% CI 0.639–0.944, p = 0.011). Similar findings were noted in the Q3 group (1.4-12) for TPOAb, which showed a significant association with GDM (ORQ3 = 0.735, 95% CI 0.588–0.995, p = 0.046). In the TGAb+ positive group, the quartiles of thyroid indicators did not show significant significance (P > 0.05). However, it is worth noting that when the TPOAb variable was included in the regression equation, all results exhibited a confidence interval of 0 and a large OR value, suggesting a strong positive impact of TPOAb on the likelihood of pregnant women developing GDM. 3.4 Screening for clinical risk factors Pearson correlation analysis revealed a predominantly positive association between demographic information of the participants and the incidence of GDM, whereas thyroid examination indicators tended to show a negative correlation ( Figure 3A ). Among different subgroups of GDM, five factors (age, family history of diabetes, family history of hypertension, gestational hypertension, and TGAb) exhibited significant difference (P < 0.05). Notably, maternal height and thyroid function demonstrated relatively mild correlations with GDM risk, even though their P-values were not significantly different. Spearman correlation analysis highlighted height as an evident negative factor, implying that shorter height correlates with higher GDM risk ( Figure 3B) . Interesting, most thyroid indicators showed negative correlation, while Tg had a positive correlation, suggesting the need for further investigation. Random Forest (RF) analysis of relative importance identified age and all thyroid test indicators as having high contributions to GDM risk ( Figure 3C ). Factors like family history of cancer, polycystic ovary syndrome, and ovarian cysts had minimal impact and were excluded from further analysis. The latest calculations showed that age, TSH, TT4, Tg, and FT3/FT4 were the top five influential factors ( Figure 4A ). A total of 18 factors were selected to construct the prognostic model, with grey correlation analysis indicates indicating a strong association (coefficient of association ≥ 0.6) among the factors (Figure 4B) . 3.5 Development of a clinical predictive model for GDM The stepwise regression (SR) model was developed using a bidirectional progressive regression method, examining all variables in the model with each addition and eliminating those with insignificant effects to find the optimal combination. After twelve iterations, seven core factors were identified, resulting in an Akaike information criterion (AIC) value of 4988.9. The best subset selection (BSS) technique evaluated all combinations of feature variables, fitting 131,071 models to identify the optimal set. In comparison to the SR model, the BUS model included FT4 as an additional crucial variable, bringing the total to eight core factors. The RF model constructed 50 decision trees and evaluated three variable at each splitting point, ultimately selecting 10 important factors. The average squared residuals were 0.166. Meanwhile, the SVM model explores linear and functional, polynomial kernel function, radial basis kernel function, and sigmoid kernel function constructions to identify the most suitable model and outputs the top five influential factors (Table 4). 3.6 Model validation and evaluation Evaluation metrics for the clinical prediction models included AUC (95% CI), sensitivity, specificity, TPR, TNR, C-index (95% CI), X-squared, DF, and P-value. The SVM model exhibited the highest accuracy with an AUC of 0.824 (0.8137-0.8345), though its C-index value was moderate at 0.645 (0.6263, 0.6430) (Table 5). The SVM model also had superior TPR (0.867) and TNR (0.823) compared to the other three models. Additionally, the SVM model demonstrated the superior calibration (0.629). Despite moderate C-index values across all models, highlighting the complexity of GDM prediction, the models were ranked by AUS as follows: SVM > SR > BSS > RF. Calibration curves were constructed to assess the disparity between predicted and actual outcomes ( Figure 5A-D ). Nomograms for each model were created based on independent risk factors, illustrating the relationships between prediction factors and GDM risk ( Figure 5E-H ). Risk scores for each factor were summed to calculate the total GDM risk score, which was then transformed into the probability of developing GDM. Decision curve indicated that the intervention measures are most beneficial when the GDM risk probability is between 0.1 and 0.4 ( Figure 6A-D ). The clinical impact curve depicted the number of true positives at various threshold probabilities, suggesting that intervention when GDM risk ranges from 10% to 40% yields favorable clinical outcomes. Ultimately, the SUM model achieved an accuracy of 81.52% on the test set ( Figure 6E-H ). 4. Discussion In our study, age, family history of diabetes, pregnancy-induced hypertension, and family history of hypertension were identified as significant risk factors for increased susceptibility to GDM. Numerous studies have already shown that both age and a family history of diabetes contribute to an increased risk of GDM ( 45 – 48 ), which aligns with our research findings. Therefore, it is crucial to prioritize GDM prevention in pregnant women of advanced age and those with a family history of diabetes. Close attention should be paid to the health responses of these individuals during pregnancy, allowing timely intervention measures. GDM-induced hyperglycemia is known to affect systemic arteries when insulin resistance occurs, resulting in elevated blood pressure ( 49 ). Conversely, if gestational hypertension develops first, it can trigger inflammatory responses or cause abnormalities in vascular function, potentially interfering with the normal action of insulin and thereby increasing the risk of GDM. It is worth noting that these two pregnancy complications share some common risk factors, such as obesity, age, and family history, which should be considered by women of childbearing age who are planning for pregnancy or are already pregnant. Some studies have indicated a higher risk among pregnant women with a history of hypertension ( 50 , 51 ). However, in reality, a family history of hypertension is a risk factor that cannot be separated from blood pressure and blood glucose considerations. In our Spearman correlation analysis, we discovered a negative relationship between height and GDM risk, which might be attributed to excessive weight gain during pregnancy. As height decreases, the impact of weight gain during pregnancy becomes more pronounced, affecting maternal metabolism and blood glucose levels. This finding suggests that shorter women may require extra attention to weight control while ensuring proper nutrition during pregnancy. Thyroid hormones play a crucial role in glucose metabolism and maintaining internal balance. They influence both the growth and development of the fetus and the adverse pregnancy outcomes for mothers, making thyroid disorders prevalent among pregnant women ( 52 ). However, the underlying mechanisms linking abnormal thyroid function markers to the outcome of GDM remain unclear ( 53 ). TGAb serves as a serologic marker for autoimmune thyroid disease. If women have a family history of diabetes and thyroid disease, as well as a positive TGAb result ranging from 8–16%, they are more likely to develop GDM ( 22 ). Elevated levels of early-pregnancy FT3 and TT3 are associated with an increased risk of GDM ( 24 ), a finding consistent with our study regarding FT3 levels during the early pregnancy. However, our results do not support a strong association between TT3 levels and early detection of GDM. Similarly, higher levels of TSH in early pregnancy indicate a higher risk of GDM ( 54 ), whereas low TSH levels during the later stages of pregnancy become a risk factor. Further research is needed to understand the reasons behind these phenomena. Some studies suggest a lack of average TSH levels during early pregnancy ( 55 ), whereas our research demonstrates that higher TSH levels effectively reduce the risk of GDM in pregnant women. Women who test positive for TGAb and TPOAb are more susceptible to GDM ( 22 ), which aligns with our findings, establishing a positive correlation. These results emphasize the importance of monitoring thyroid antibodies and hormone levels in pregnant women to effectively prevent the occurrence of GDM. Regarding feature selection for clinical prediction models, we embarked on a comprehensive analysis, looking beyond univariate correlations to recognize the collective influence of multiple factors due to the complex nature of GDM. We proposed a novel approach to assess the correlation between exposure and outcome, using Pearson and Spearman correlation coefficients to ascertain the direction of associations, the RF method to determine the relative importance of each exposure on the outcome, and ultimately calculating the grey relational degree to capture multidimensional interplay. To construct the clinical prediction model, we compared four different methods. The SR model has been utilized since 2010 to identify GDM predictive factors ( 56 ), while the BSS model uncovered a relationship between alterations in gut microbiota and GDM status ( 57 ). Notably, in a study focusing on the early prediction of GDM in the Chinese population ( 46 ), the SVM model achieved a maximum accuracy of only 77%. However, our search revealed that the BSS model has not yet been applied in the field of GDM clinical prediction, making it a novel approach to prognostic model construction. In our retrospective cohort study, the SVM model exhibited commendable predictive ability due to its robustness in tackling complex classification problems ( 58 ). Future investigations into clinical prediction models and the identification of hazardous exposures would integrate multiple methods as proposed in our study, with an emphasis on using the SVM approach for model development. 5. Conclusions The findings from this retrospective cohort study indicate a significant correlation between thyroid function parameters and GDM outcomes. Our research underscores the necessity of thyroid hormone screening as part of early GDM risk assessment. We propose the concept of multiple risk factors interact to influence clinical outcomes and offer screening methods. The goal is to develop a clinical prediction model that prompts obstetricians to pay more attention to these risk factors during pregnancy, enabling early screening and intervention to reduce the incidence of GDM. However, it is important to acknowledge the limitations of our study. Future research should focus on designing prospective cohort studies, utilizing time series to monitor thyroid hormone changes throughout the entire pregnancy, and forming dynamic real-time feedback. Additionally, increasing the sample size for training and testing the model will help make it more comprehensive. Lastly, it is crucial to include external cohorts for training and validation to avoid regional bias in the samples and support the generalizability of this model. Abbreviations Acronym Full name GDM gestational diabetes mellitus TGAb thyroglobulin antibody IDF International Diabetes Federation OGTT Oral Glucose Tolerance Test TD thyroid dysfunction HIS Hospital Information System ADA American Diabetes Association IADPSG International Association of Diabetes and Pregnancy Study Groups TT3 total triiodothyronine TT4 total thyroxine FT3 Free triiodothyronine FT4 Free thyroxine TSH thyroid-stimulating hormone TG thyroglobulin TPOAb Thyroid peroxidase antibodies RF Random Forest SR stepwise regression BSS best subset selection AIC Akaike information criterion SVM support vector machines TPR true positive rate TNR true negative rate DF degree of freedom GOF goodness-of-fit Declarations Acknowledgements Not applicable. Authors' contributions Conceptualization, XL and HZ; Data curation, CC and WW; Formal analysis, HC and CL; Funding acquisition, XL; Investigation, HC; Methodology, XL; Project administration, HZ; Resources, YH, HC, CL, Li Chen and WW;; Software, YH; Supervision, HZ; Validation, YH, CC and LC; Visualization, CC; Writing – original draft, XL; Writing – review & editing, HZ. Funding This study was supported by the Zhejiang Provincial Natural Science Foundation of China (LBY23H200008), the National Natural Science Foundation of China (T2341010), the Medical Health Science and Technology Project of Zhejiang Provincial (2023RC272), and the Science and Technology Planning Project of Wenzhou (ZY2021025 and Y2023088). Availability of data and materials The data are available from the corresponding author upon reasonable request. Ethics approval and consent to participate All studies were approved by the Ethical Committee of Wenzhou People's Hospital (No: KY-2022-277). All participants in this study signed an informed consent form. Consent for publication Written informed consent has been obtained from the patients to publish this paper. Competing interest The authors declare that they have no competing interests. References Saravanan P. Gestational diabetes: opportunities for improving maternal and child health. Lancet Diabetes Endocrinol. 2020;8(9):793-800. Szmuilowicz ED, Josefson JL, Metzger BE. Gestational Diabetes Mellitus. Endocrinol Metab Clin North Am. 2019;48(3):479-93. Coustan DR. Gestational diabetes mellitus. Clin Chem. 2013;59(9):1310-21. Wang H, Li N, Chivese T, Werfalli M, Sun H, Yuen L, et al. IDF Diabetes Atlas: Estimation of Global and Regional Gestational Diabetes Mellitus Prevalence for 2021 by International Association of Diabetes in Pregnancy Study Group's Criteria. Diabetes Res Clin Pract. 2022;183:109050. Nguyen CL, Pham NM, Binns CW, Duong DV, Lee AH. Prevalence of Gestational Diabetes Mellitus in Eastern and Southeastern Asia: A Systematic Review and Meta-Analysis. J Diabetes Res. 2018;2018:6536974. Gao C, Sun X, Lu L, Liu F, Yuan J. Prevalence of gestational diabetes mellitus in mainland China: A systematic review and meta-analysis. J Diabetes Investig. 2019;10(1):154-62. Sweeting A, Wong J, Murphy HR, Ross GP. A Clinical Update on Gestational Diabetes Mellitus. Endocr Rev. 2022;43(5):763-93. Modzelewski R, Stefanowicz-Rutkowska MM, Matuszewski W, Bandurska-Stankiewicz EM. Gestational Diabetes Mellitus-Recent Literature Review. J Clin Med. 2022;11(19). Ye W, Luo C, Huang J, Li C, Liu Z, Liu F. Gestational diabetes mellitus and adverse pregnancy outcomes: systematic review and meta-analysis. BMJ. 2022;377:e067946. Retnakaran R, Ye C, Hanley AJ, Connelly PW, Sermer M, Zinman B. Treatment of Gestational Diabetes Mellitus and Maternal Risk of Diabetes After Pregnancy. Diabetes Care. 2023;46(3):587-92. Grunnet LG, Hansen S, Hjort L, Madsen CM, Kampmann FB, Thuesen ACB, et al. Adiposity, Dysmetabolic Traits, and Earlier Onset of Female Puberty in Adolescent Offspring of Women With Gestational Diabetes Mellitus: A Clinical Study Within the Danish National Birth Cohort. Diabetes Care. 2017;40(12):1746-55. Zhang C, Catalano P. Screening for Gestational Diabetes. JAMA. 2021;326(6):487-9. Jin J. Screening for Gestational Diabetes. JAMA. 2021;326(6):577. Pinto S, Croce L, Carlier L, Cosson E, Rotondi M. Thyroid dysfunction during gestation and gestational diabetes mellitus: a complex relationship. J Endocrinol Invest. 2023. Milovanović Z, Filimonović D, Soldatović I, Karadžov Orlić N. Can Thyroid Screening in the First Trimester Improve the Prediction of Gestational Diabetes Mellitus? J Clin Med. 2022;11(13). Yang Y, Li Q, Wang Q, Ma X. Thyroid antibodies and gestational diabetes mellitus: a meta-analysis. Fertil Steril. 2015;104(3). Jia M, Wu Y, Lin B, Shi Y, Zhang Q, Lin Y, et al. Meta-analysis of the association between maternal subclinical hypothyroidism and gestational diabetes mellitus. Int J Gynaecol Obstet. 2019;144(3):239-47. Kent NL, Young SL, Akison LK, Cuffe JSM. Is the link between elevated TSH and gestational diabetes mellitus dependant on diagnostic criteria and thyroid antibody status: a systematic review and meta-analysis. Endocrine. 2021;74(1):38-49. Luo J, Wang X, Yuan L, Guo L. Association of thyroid disorders with gestational diabetes mellitus: a meta-analysis. Endocrine. 2021;73(3):550-60. Sitoris G, Veltri F, Ichiche M, Kleynen P, Praet J-P, Rozenberg S, et al. Association between thyroid autoimmunity and gestational diabetes mellitus in euthyroid women. Eur Thyroid J. 2022;11(2). Sert UY, Buyuk GN, Engin Ustun Y, Ozgu Erdinc AS. Is There Any Relationship Between Thyroid Function Abnormalities, Thyroid Antibodies and Development of Gestational Diabetes Mellitus (GDM) in Pregnant Women? Medeni Med J. 2020;35(3):195-201. Tang L, Li P, Zhou H, Li L. A longitudinal study of thyroid markers during pregnancy and the risk of gestational diabetes mellitus and post-partum glucose metabolism. Diabetes Metab Res Rev. 2021;37(4):e3441. Mennickent D, Ortega-Contreras B, Gutiérrez-Vega S, Castro E, Rodríguez A, Araya J, et al. Evaluation of first and second trimester maternal thyroid profile on the prediction of gestational diabetes mellitus and post load glycemia. PLoS One. 2023;18(1):e0280513. Wu J-N, Pérez-López FR, Peng T, Xie F, Li M-Q. Association of Early Pregnancy Free and Total Triiodothyronine With the Subsequent Risk of Gestational Diabetes Mellitus. Endocr Pract. 2023;29(1):40-7. Raets L, Minschart C, Van den Bruel A, Van den Bogaert E, Van Crombrugge P, Moyson C, et al. Higher Thyroid fT3-to-fT4 Ratio Is Associated with Gestational Diabetes Mellitus and Adverse Pregnancy Outcomes. J Clin Med. 2022;11(17). Demiral Sezer S, Topaloglu O. Effects of thyroid hormones in women with gestational diabetes. Gynecol Endocrinol. 2022;38(7):588-91. Zhang S, Wu Y, Pan W, Li G, Zhang D, Li S, et al. Free-Triiodothyronine to Free-Thyroxine Ratio Mediated the Effect of Prepregnancy Body Mass Index or Maternal Weight Gain During Early Pregnancy on Gestational Diabetes Mellitus. Endocr Pract. 2022;28(4):398-404. Li P, Cui J, Li L, Chen X, Ouyang L, Fan J, et al. Association between isolated maternal hypothyroxinemia during the first trimester and adverse pregnancy outcomes in Southern Chinese women: a retrospective study of 7051 cases. BMC Pregnancy Childbirth. 2022;22(1):866. Eloranta S, Boman M. Predictive models for clinical decision making: Deep dives in practical machine learning. J Intern Med. 2022;292(2):278-95. Li JK, Tang T, Zong H, Wu EM, Zhao J, Wu RR, et al. Intelligent medicine in focus: the 5 stages of evolution in robot-assisted surgery for prostate cancer in the past 20 years and future implications. Mil Med Res. 2024;11(1):58. Li LS, Yang L, Zhuang L, Ye ZY, Zhao WG, Gong WP. From immunology to artificial intelligence: revolutionizing latent tuberculosis infection diagnosis with machine learning. Mil Med Res. 2023;10(1):58. Zhang YP, Zhang XY, Cheng YT, Li B, Teng XZ, Zhang J, et al. Artificial intelligence-driven radiomics study in cancer: the role of feature engineering and modeling. Mil Med Res. 2023;10(1):22. Huang Q-F, Hu Y-C, Wang C-K, Huang J, Shen M-D, Ren L-H. Clinical First-Trimester Prediction Models for Gestational Diabetes Mellitus: A Systematic Review and Meta-Analysis. Biol Res Nurs. 2023;25(2):185-97. Artzi NS, Shilo S, Hadar E, Rossman H, Barbash-Hazan S, Ben-Haroush A, et al. Prediction of gestational diabetes based on nationwide electronic health records. Nat Med. 2020;26(1):71-6. Wu Y, Ma S, Wang Y, Chen F, Zhu F, Sun W, et al. A risk prediction model of gestational diabetes mellitus before 16 gestational weeks in Chinese pregnant women. Diabetes Res Clin Pract. 2021;179:109001. Li X, Zheng J, Ma X, Zhang B, Zhang J, Wang W, et al. The oral microbiome of pregnant women facilitates gestational diabetes discrimination. J Genet Genomics. 2021;48(1):32-9. Koos BJ, Gornbein JA. Early pregnancy metabolites predict gestational diabetes mellitus: implications for fetal programming. Am J Obstet Gynecol. 2021;224(2):215.e1-.e7. Rawal S, Tsai MY, Hinkle SN, Zhu Y, Bao W, Lin Y, et al. A Longitudinal Study of Thyroid Markers Across Pregnancy and the Risk of Gestational Diabetes. J Clin Endocrinol Metab. 2018;103(7):2447-56. Li X, Zuo J, Li Y-H, Tang Y-P, Bao Y-R, Ying H. Association between thyroid function and risk of gestational diabetes mellitus in assisted pregnancies: A retrospective cohort study. Diabetes Res Clin Pract. 2021;171:108590. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2021. Diabetes Care. 2021;44(Suppl 1):S15-S33. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33(3):676-82. Miao Z, Li-Xin SJCJoPIM. A clinical interpretation of 2012 Chinese guidelines for the diagnosis and treatment of thyroid disease during pregnancy and postpartum. 2012. Chidakel A, Mentuccia D, Celi FS. Peripheral metabolism of thyroid hormone and glucose homeostasis. Thyroid. 2005;15(8):899-903. Gao D, Guo L, Wang F, Zhang Z. Study on the Spontaneous Combustion Tendency of Coal Based on Grey Relational and Multiple Regression Analysis. ACS Omega. 2021;6(10):6736-46. Liu H, Li J, Leng J, Wang H, Liu J, Li W, et al. Machine learning risk score for prediction of gestational diabetes in early pregnancy in Tianjin, China. Diabetes Metab Res Rev. 2021;37(5):e3397. Wu Y-T, Zhang C-J, Mol BW, Kawai A, Li C, Chen L, et al. Early Prediction of Gestational Diabetes Mellitus in the Chinese Population via Advanced Machine Learning. J Clin Endocrinol Metab. 2021;106(3):e1191-e205. Benhalima K, Van Crombrugge P, Moyson C, Verhaeghe J, Vandeginste S, Verlaenen H, et al. Estimating the risk of gestational diabetes mellitus based on the 2013 WHO criteria: a prediction model based on clinical and biochemical variables in early pregnancy. Acta Diabetol. 2020;57(6):661-71. Gao S, Leng J, Liu H, Wang S, Li W, Wang Y, et al. Development and validation of an early pregnancy risk score for the prediction of gestational diabetes mellitus in Chinese pregnant women. BMJ Open Diabetes Res Care. 2020;8(1). Mate A, Blanca AJ, Salsoso R, Toledo F, Stiefel P, Sobrevia L, et al. Insulin Therapy in Pregnancy Hypertensive Diseases and its Effect on the Offspring and Mother Later in Life. Curr Vasc Pharmacol. 2019;17(5):455-64. Wang Y, Ge Z, Chen L, Hu J, Zhou W, Shen S, et al. Risk Prediction Model of Gestational Diabetes Mellitus in a Chinese Population Based on a Risk Scoring System. Diabetes Ther. 2021;12(6):1721-34. Donovan BM, Breheny PJ, Robinson JG, Baer RJ, Saftlas AF, Bao W, et al. Development and validation of a clinical model for preconception and early pregnancy risk prediction of gestational diabetes mellitus in nulliparous women. PLoS One. 2019;14(4):e0215173. Lee SY, Pearce EN. Assessment and treatment of thyroid disorders in pregnancy and the postpartum period. Nat Rev Endocrinol. 2022;18(3):158-71. Shahbazian H, Shahbazian N, Rahimi Baniani M, Yazdanpanah L, Latifi SM. Evaluation of thyroid dysfunction in pregnant women with gestational and pre-gestational diabetes. Pak J Med Sci. 2013;29(2):638-41. Fernández Alba JJ, Castillo Lara M, Jiménez Heras JM, Moreno Cortés R, González Macías C, Vilar Sánchez Á, et al. High First Trimester Levels of TSH as an Independent Risk Factor for Gestational Diabetes Mellitus: A Retrospective Cohort Study. J Clin Med. 2022;11(13). Shahid MM, Rahman KMT, Gomes RR, Ferdous M, Ferdousi S, Zahan T. Association of gestational diabetes mellitus and thyroid status during pregnancy: a cross-sectional study in a tertiary health care center of Bangladesh. Gynecol Endocrinol. 2021;37(4):312-4. Savvidou M, Nelson SM, Makgoba M, Messow C-M, Sattar N, Nicolaides K. First-trimester prediction of gestational diabetes mellitus: examining the potential of combining maternal characteristics and laboratory measures. Diabetes. 2010;59(12):3017-22. Kuang Y-S, Lu J-H, Li S-H, Li J-H, Yuan M-Y, He J-R, et al. Connections between the human gut microbiome and gestational diabetes mellitus. Gigascience. 2017;6(8). Van Belle V, Lisboa P. White box radial basis function classifiers with component selection for clinical prediction models. Artif Intell Med. 2014;60(1):53-64. Tables Table 1: Demographic and Clinical characteristic of enrolled pregnant woman Total (n=5229) GDM (n=1035,19.8%) Control (n=4194,80.2%) Number(%) Median±SD Number(%) Median±SD Number(%) Median±SD P a Age(years) - 32.4±4.4 - 34±4.3 - 32±4.4 <0.001 Height(cm) - 159.7±18.7 - 159.2±4.8 - 159.9±20.7 0.278 Family history of diabetes 135 (2.6%) - 49 (4.7%) - 36 (0.9%) - <0.001 S carred uterus 2824 (54%) - 540 (52.2%) - 2284 (556%) - 0.187 Gestational hypertension 217 (4.1%) - 87 (8.4%) - 130 (3.1%) - <0.001 Family history of hypertension 240 (4.6%) - 64 (6.2%) - 176 (4.2%) - 0.006 Family history of cancer 90 (1.7%) - 17 (1.6%) - 73 (1.8%) - 0.828 Re-pregnancy 2359 (45.1%) - 470 (45.4%) - 1889 (45%) - 0.830 Polycystic ovary syndrome 9 (0.2%) - 2 (0.2%) - 7 (1.7%) - 0.855 Ovarian cysts 25 (0.5%) - 4 (0.4%) - 21 (0.5%) - 0.633 Thyroid f unction - - - Hypothyroidism 220 (4.2%) - 52 (5%) - 168 (4%) - 0.236 Normal thyroid function 4978 (95.2%) - 975 (94.2%) - 4003 (95.4%) - Hyperthyroidism 31 (0.6%) - 8 (0.8%) - 23 (0.6%) - Thyroid function indicators Median(Quartile) Median(Quartile) Median(Quartile) TT3(ng/ml) - 1.2 (1.04,1.39) - 1.19 (1.02,1.38) - 1.2(1.04,1.4) 0.068 TT4(ng/ml) - 86.8 (75.9,98) - 85.7 (74.7,97.5) - 86.9 (76.2,98.13) 0.058 TT3/TT4 - 0.014(0.012,0.016) - 0.014(0.012,0.017) - 0.014(0.012,0.016) 0.985 FT3(pmol/L) - 4.2(3.8,4.6) - 4.2(3.8,4.6) - 4.2(3.8,4.6) 0.127 FT4(pmol/L) - 8.9(7.7,11) - 8.9(7.8,10.9) - 8.9(7.7,11) 0.926 FT3/FT4 - 0.479(0.359,0.572) - 0.479(0.358,0.563) - 0.478(0.359,0.575) 0.435 TSH(mIU/L) - 1.81(1.27,2.52) 1.81(1.2,2.49) - 1.81(1.29,2.53) 0.250 TPOAb(IU/ml) - 1.3(0.4,12) - 1.3(0.4,12.1) - 1.3(0.4,12) 0.470 TGAb(IU/ml) - 0.9(<0.9,10) - 0.9(<0.9,10) - 0.9(=35 IU/ml 267 (5.1%) - 46 (4.4%) - 221 (5.3%) - 0.280 =4.0 IU/ml 1954 (37.4%) - 394 (38%) - 1560 (37.2%) - 0.604 <4.0 IU/ml 3275 (62.6%) - 641 (62%) - 2634 (62.8%) - Note: P a : P-values were obtained from the statistical results. Continuous data are normally distributed using the independent samples t-test, continuous data are not normally distributed using the rank sum test, and categorical data use the chi-square test. Table 2: GDM Risk Analysis of thyroid function in all enrolled woman GDM (n=1035) Control (n=4194) P for trend OR (95% CI) P aOR (95% CI) P TT3(ng/ml) 0.013 Quartile 1 (~1.04) 310 (30.0%) 1056 (25.2%) 1.255 (1.040,1.514) 0.018 1.208 (0.996,1.466) 0.056 Quartile 2 (1.05-1.2) 229 (22.1%) 1051 (25.0%) 0.932 (0.763,1.137) 0.485 0.940 (0.767,1.153) 0.555 Quartile 3 (1.21-1.39) 249 (24.0%) 1031 (24.6%) 1.033 (0.849,1.256) 0.749 1.049 (0.859,1.282) 0.632 Quartile 4 (1.40-3.68) 247 (23.9%) 1056 (25.2%) Ref - Ref - TT4(ng/ml) 0.378 Quartile 1 (~75.9) 278 (26.9%) 1033 (24.6%) Ref - Ref - Quartile 2 (76-86.8) 254 (24.5%) 1063 (25.3%) 0.888 (0.734,1.074) 0.221 0.938 (0.772,1.139) 0.516 Quartile 3 (86.9-98) 261 (25.2%) 1039 (24.8%) 0.933 (0.772,1.128) 0.476 1.003 (0.826,1.218) 0.976 Quartile 4 (98.1-235.6) 242 (23.4%) 1059 (25.3%) 0.849 (0.700,1.029) 0.096 0.958 (0.786,1.168) 0.674 TT3/TT4 0.708 Quartile 1 (~0.012) 259 (25.0%) 1047 (25.0%) Ref - Ref - Quartile 2 (0.013-0.014) 248 (24.0%) 1061 (25.3%) 0.945 (0.778,1.147) 0.567 0.982 (0.805,1.197) 0.856 Quartile 3 (0.015-0.016) 256 (24.7%) 1048 (25.0%) 0.987 (0.814,1.197) 0.898 0.981 (0.806,1.196) 0.852 Quartile 4 (0.017-0.070) 272 (26.3%) 1038 (24.7%) 1.059 (0.875,1.282) 0.554 1.036 (0.852,1.259) 0.726 FT3(pmol/L) 0.099 Quartile 1 (~3.8) 302 (29.2%) 1088 (26.0%) Ref - Ref - Quartile 2 (3.9-4.2) 278 (26.9%) 1120 (26.7%) 0.894 (0.745,1.074) 0.231 0.920 (0.763,1.110) 0.383 Quartile 3 (4.3-4.6) 221 (21.3%) 1016 (24.2%) 0.784 (0.646,0.951) 0.013 0.800 (0.656,0.975) 0.027 Quartile 4 (4.7-14.3) 234 (22.6%) 970 (23.1%) 0.869 (0.718,1.052) 0.151 0.895 (0.735,1.089) 0.269 FT4(pmol/L) 0.783 Quartile 1 (~7.7) 255 (24.6%) 1075 (25.6%) Ref - Ref - Quartile 2 (7.8-8.9) 267 (25.8%) 1031 (24.6%) 1.092 (0.901,1.322) 0.370 1.157 (0.951,1.407) 0.146 Quartile 3 (9.0-11) 267 (25.8%) 1061 (25.3%) 1.061 (0.876,1.285) 0.545 1.148 (0.943,1.396) 0.169 Quartile 4 (11.1-37.1) 246 (23.8%) 1027 (24.5%) 1.010 (0.831,1.227) 0.922 1.097 (0.898,1.339) 0.364 FT3/FT4 0.150 Quartile 1 (~0.359) 260 (25.1%) 1047 (25.0%) Ref - Ref - Quartile 2 (0.360-0.479) 256 (24.7%) 1051 (25.1%) 0.981 (0.809,1.189) 0.844 0.971 (0.797,1.183) 0.769 Quartile 3 (0.480-0.572) 283 (27.4%) 1025 (24.4%) 1.112 (0.920,1.343) 0.272 1.081 (0.890,1.312) 0.432 Quartile 4 (0.573-1) 236 (22.8%) 1071 (25.5%) 0.887 (0.730,1.079) 0.231 0.834 (0.683,1.020) 0.077 TSH(mIU/L) 0.047 Quartile 1 (~1.27) 288 (27.8%) 1024 (24.4%) 1.211 (1.001,1.465) 0.049 1.288 (1.059,1.565) 0.011 Quartile 2 (1.28-1.81) 234 (22.6%) 1074 (25.6%) 0.938 (0.769,1.143) 0.526 0.986 (0.805,1.208) 0.891 Quartile 3 (1.82-2.52) 267 (25.8%) 1037 (24.7%) 0.297 (0.914,1.345) 0.297 1.12 (0.919,1.365) 0.263 Quartile 4 (2.53-22.89) 246 (23.8%) 1059 (25.3%) Ref - Ref - TPOAb(IU/ml) 0.623 Quartile 1 (~0.4) 304 (29.4%) 1166 (27.8%) Ref - Ref - Quartile 2 (0.5-1.3) 220 (21.3%) 950 (22.7%) 0.888 (0.732,1.078) 0.230 0.848 (0.696,1.034) 0.103 Quartile 3 (1.4-12) 249 (24.0%) 1041 (24.8%) 0.917 (0.761,1.106) 0.367 0.876 (0.723,1.061) 0.174 Quartile 4 (12.1-994) 262 (25.3%) 1037 (24.7%) 0.969 (0.805,1.166) 0.739 0.925 (0.765,1.118) 0.419 Tg(ng/ml) 0.268 Quartile 1 (~5.9) 250 (24.2%) 1072 (25.6%) Ref - Ref - Quartile 2 (6.0-10.5) 244 (23.6%) 1049 (25.0%) 0.997 (0.820,1.213) 0.979 0.979 (0.802,1.196) 0.838 Quartile 3 (10.6-17.9) 262 (25.3%) 1059 (25.2%) 1.061 (0.875,1.287) 0.548 1.033 (0.848,1.258) 0.748 Quartile 4 (18.0-298.9) 279 (26.9%) 1014 (24.2%) 1.180 (0.975,1.428) 0.090 1.101 (0.905,1.339) 0.335 Note: OR = Odds Ratio, aOR = adjusted Odds(adjusted Age, Family history of diabetes, Gestational hypertension and Family history of hypertension). Table 3: GDM varies with levels with other thyroid function in patients with different levels of TPOAb TGAb- (n=3275) P for trend TGAb+ (n=1954) P for trend GDM/Control OR (95% CI) P aOR (95% CI) P GDM/Control OR (95% CI) P aOR (95% CI) P TT3(ng/ml) 0.007 0.456 Q1 (~1.04) 256/878 (34.6%) Ref - Ref - 54/178 (11.9%) Ref - Ref - Q2 (1.05-1.2) 160/802 (22.1%) 0.684(0.549,0.852) 0.001 0.719(0.574,0.899) 0.004 69/249 (16.3%) 0.913(0.609,1.369) 0.661 0.940(0.620,1.426) 0.773 Q3 (1.21-1.39) 150/629 (24.0%) 0.818(0.652,1.025) 0.081 0.847(0.671,1.070) 0.164 99/402 (25.6%) 0.812(0.558,1182) 0.277 0.882(0.600,1.298) 0.524 Q4 (1.40-3.68) 75/325 (23.9%) 0.791(0.594,1.055) 0.110 0.833(0.619,0.120) 0.226 172/731 (46.2%) 0.776(0.548,1.097) 0.151 0.805(0.563,1.151) 0.235 TT4(ng/ml) 0.823 0.160 Q1 (~75.9) 173/666 (25.6%) Ref - Ref - 105/367 (24.2%) Ref - Ref - Q2 (76-86.8) 163/672 (25.5%) 0.934 (0.735,1.186) 0.575 0.976 (0.764,1.246) 0.844 91/391 (24.7%) 0.813 (0.594,1.114) 0.199 0.874 (0.633,1.207) 0.414 Q3 (86.9-98) 156/672 (25.3%) 0.894 (0.702,1.138) 0.362 0.940 (0.734,1.203) 0.621 105/367 (24.1%) 1.000 (0.736,1.359) 1.000 1.122 (0.818,1.538) 0.476 Q4 (98.1-235.6) 149/624 (23.6%) 0.919 (0.720,1.174) 0.500 0.990 (0.770,1.273) 0.937 93/435 (27.0%) 0.747 (0.547,1.021) 0.747 0.907 (0.658,1.251) 0.554 TT3/TT4 0.944 0.548 Q1 (~0.012) 220/872 (33.3%) Ref - Ref - 39/175 (11.0%) Ref - Ref - Q2 (0.013-0.014) 179/756 (28.6%) 0.938(0.753,1.169) 0.572 0.966 (0.771,1.210) 0.763 69/305 (19.1%) 1.015 (0.657,1.567) 0.946 1.023(0.656,1.595) 0.919 Q3 (0.015-0.016) 140/587 (22.2%) 0.945 (0.746,1.197) 0.641 0.950 (0.746,1.210) 0.677 116/461 (29.5%) 1.129 (0.755,1.688) 0.554 1.059(0.701,1.601) 0.784 Q4 (0.017-0.070) 102/419 (15.9%) 0.965 (0.742,1.254) 0.789 0.979 (0.749,1.281) 0.880 170/619 (40.4%) 1.232 (0.837,1.813) 0.289 1.104 (0.742,1.641) 0.625 FT3(pmol/L) 0.307 0.194 Q1 (~3.8) 111/391 (15.3%) Ref - Ref - 191/697 (45.4%) Ref - Ref - Q2 (3.9-4.2) 171/671 (25.7%) 0.898 (0.686,1.175) 0.432 0.943 (0.715,1.244) 0.679 107/449 (28.5%) 0.870 (0.667,1.133) 0.301 0.895 (0.682,1.175) 0.424 Q3 (4.3-4.6) 166/740 (27.7%) 0.790 (0.603,1.035) 0.087 0.829 (0.629,1.093) 0.184 55/276 (16.9%) 0.727 (0.522,1.012) 0.059 0.727 (0.518,1.021) 0.066 Q4 (4.7-14.3) 193/832 (31.3%) 0.817 (0.628,1.063) 0.132 0.857(0.654,1.122) 0.261 41/138 (9.2%) 1.084 (0.739,1.591) 0.680 1.157 (0.777,1.723) 0.472 FT4(pmol/L) 0.771 0.365 Q1 (~7.7) 239/997 (37.7%) Ref - Ref - 16/78 (4.8%) Ref - Ref - Q2 (7.8-8.9) 236/921 (35.4%) 1.069(0.874,1.307) 0.516 1.131 (0.921,1.390) 0.241 31/110 (7.2%) 1.374 (0.703,2.683) 0.352 1.424 (0.717,2.829) 0.312 Q3 (9.0-11) 150/656 (24.6%) 0.954(0.760,1.197) 0.683 1.058 (0.838,1.335) 0.635 117/405 (26.7%) 1.408 (0.792,2.505) 0.244 1.503 (0.832,2.714) 0.177 Q4 (11.1-37.1) 16/60 (2.3%) 1.112(0.630,1.966) 0.714 1.198 (0.665,2.156) 0.548 230/967 (61.3%) 1.160 (0.664,2.023) 0.602 1.302 (0.735,2.307) 0.365 FT3/FT4 0.089 0.885 Q1 (~0.359) 13/49 (1.9%) 1.210 (0.645,2.269) 0.533 1.161 (0.608,2.217) 0.608 247/998 (63.7%) 1.059 (0.622,1.802) 0.833 1.136 (0.657,1.965) 0.649 Q2 (0.360-0.479) 150/661 (24.8%) 1.035 (0.822,1.302) 0.771 1.132 (0.894,1.434) 0.302 106/390 (25.4%) 1.163 (0.667,2.028) 0.595 1.161 (0.655,2.058) 0.609 Q3 (0.480-0.572) 260/930 (36.3%) 1.275 (1.043,1.558) 0.018 1.323 (1.077,1.625) 0.008 23/95 (6.0%) 1.036(0.521,2.057) 0.920 1.050 (0.520,2.123) 0.891 Q4 (0.573-1) 218/994 (37.0%) Ref - Ref - 18/77 (4.9%) Ref - Ref - TSH(mIU/L) 0.081 0.501 Q1 (~1.27) 192/687 (26.8%) Ref - Ref - 96/337 (22.1%) Ref - Ref - Q2 (1.28-1.81) 150/730 (26.9%) 0.735 (0.580,0.932) 0.010 0.766 (0.629,0.932) 0.008 84/344 (21.9%) 0.857 (0.617,1.191) 0.359 0.831 (0.593,1.165) 0.283 Q3 (1.82-2.52) 163/643 (24.6%) 0.907 (0.717,1.147) 0.415 0.870 (0.718,1.054) 0.154 104/394 (26.0%) 0.927 (0.677,1.267) 0.633 0.857 (0.621,1.182) 0.348 Q4 (2.53-22.89) 136/574 (21.7%) 0.848 (0.663,1.084) 0.188 0.777 (0.639,0.944) 0.011 110/485 (30.0%) 0.796 (0.586,1.082) 0.146 0.759 (0.554,1.039) 0.086 TPOAb(IU/ml) 0.331 0.175 Q1 (~0.4) 304/1148 (44.4%) Ref - Ref - 0/18 (0.9%) Ref - Ref - Q2 (0.5-1.3) 214/920 (34.6%) 0.878 (0.723,1.068) 0.193 0.833 (0.628,1.018) 0.075 6/30 (1.8%) 323152995 (0,0) 0.998 305225056 (0,0) 0.998 Q3 (1.4-12) 94/427 (15.9%) 0.831 (0.643,1.075) 0.158 0.765 (0.588,0.995) 0.046 155/614 (39.4%) 407888552 (0,0) 0.998 358188050 (0,0) 0.998 Q4 (12.1-994) 29/139 (5.1%) 0.788 (0.518,1.199) 0.265 0.738 (0.480,1.135) 0.166 233/898 (57.9%) 419235233 (0,0) 0.998 358253152 (0,0) 0.998 Tg(ng/ml) 0.307 0.860 Q1 (~5.9) 152/667 (25.0%) Ref - Ref - 98/405 (25.7%) Ref - Ref - Q2 (6.0-10.5) 171/757 (28.3%) 0.991 (0.778,1.263) 0.943 0.940 (0.734,1.204) 0.626 73/292 (18.7%) 0.997 (0.820,1.213) 0.950 1.072 (0.758,1.517) 0.694 Q3 (10.6-17.9) 171/684 (26.1%) 1.097 (0.860,1.399) 0.455 1.041 (0.812,1.335) 0.751 91/375 (23.8%) 1.061 (0.875,1.287) 0.986 1.017 (0.735,1.409) 0.917 Q4 (18.0-298.9) 147/526 (20.6%) 1.226 (0.951,1.581) 0.115 1.142 (0.881,1.481) 0.317 132/488 (31.8%) 1.180 (0.975,1.428) 0.456 1.0531 (0.779,1.422) 0.738 Note: Q: Quartile, OR = Odds Ratio, aOR = adjusted Odds (adjusted Age, Family history of diabetes, Gestational hypertension and Family history of hypertension ). Table 4: Construction of clinical prediction models for GDM using different methods Screening methods Selection criteria Feature numbers Feature Stepwise regression AIC 7 Age, Height, Family history of diabetes, Gestational hypertension, ‘FT3/FT4’, TSH, TGAb Best s ubset s election AIC 8 Age, Height, Family history of diabetes, Gestational hypertension, ‘TT3/TT4’, FT4, TSH, TGAb Random f orest OOB error rate 10 Age, Height, TSH, TT4, Tg, TT3,FT4, TPOAb,FT3,TGAb S upport vector machines Best of four kernel functions 5 Age, Height, Family history of diabetes, Gestational hypertension, ‘FT3/FT4’ Notes: AIC: Akaike information criterion. OOB: Out of bag. Four kernel functions:linear, polynomial, radial and sigmoid. Table 5: Evaluation of different clinical prediction models Models AUC (95% CI) Sensitivity Specificity TPR TNR C-index(95% CI) X-squared DF P-value Stepwise regression 0.647 (0.6284,0.6656) 0.540 0.660 0.540 0.660 0.647 (0.6288,0.6660) 2.043 2 0.360 Best s ubset s election 0.646 (0.6276,0.6647) 0.691 0.211 0.529 0.691 0.646 (0.6276,0.6647) 1.832 2 0.400 Random f orest 0.569 (0.5494,0.5884) 0.632 0.485 0.632 0.485 0.632 (0.6139,0.6510) 0.949 2 0.622 S upport vector machines 0.824 (0.8137,0.8345) 0.132 0.995 0.867 0.823 0.645 (0.6263,0.6430) 0.926 2 0.629 Note: AUC: Area Under Curve TPR: True Positive Rate, TNR: True Negative Rate C-index: Concordance Index, DF : Degree of Freedom P-value：Hosmer and Lemeshow goodness of fit (GOF) test Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7192354","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":499149534,"identity":"595377b6-434b-4e7b-8b7c-735f81a368b8","order_by":0,"name":"Xiaoqing Li","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xiaoqing","middleName":"","lastName":"Li","suffix":""},{"id":499149535,"identity":"f08a5fc2-1f21-4964-a8e3-6504e15f8a08","order_by":1,"name":"Yanjun Hu","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yanjun","middleName":"","lastName":"Hu","suffix":""},{"id":499149536,"identity":"96488739-4616-4ec0-9e8a-b20eb7d2694d","order_by":2,"name":"Haiying Chen","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Haiying","middleName":"","lastName":"Chen","suffix":""},{"id":499149537,"identity":"88ba4b9e-faba-41b4-8d80-feb08ba447c7","order_by":3,"name":"Chunling Chen","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chunling","middleName":"","lastName":"Chen","suffix":""},{"id":499149538,"identity":"0cd59450-7bee-4e25-a2da-be04540bfcd7","order_by":4,"name":"Chanchan Liao","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chanchan","middleName":"","lastName":"Liao","suffix":""},{"id":499149539,"identity":"c94db029-b7ec-45b7-b8f2-cecc03dcfc77","order_by":5,"name":"Li Chen","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Chen","suffix":""},{"id":499149540,"identity":"3a590083-77a5-454d-9423-e36df489401e","order_by":6,"name":"Wenhuan Wang","email":"","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wenhuan","middleName":"","lastName":"Wang","suffix":""},{"id":499149541,"identity":"90ec4000-741f-4a56-8302-39357011f69f","order_by":7,"name":"Hongping Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsUlEQVRIiWNgGAWjYFCCA2wMDBUScvwkajljYSzZQIIeNgbGtorEDURrkW88/uxx4TwJxg0MzA8f3SBGC2PDgXTjmdskmM0Z2IyNc4jRwsxw4Jg07zYJNssGHjZporSwMRxsk+adI8FjcIBYLTwMh9mkeRskJIjXIsFwjE16xjEJA8lmYv0iP+P4M+mCmrr6fvbmh4+J0sIgcQAYBCDATJRyEOBvIEHxKBgFo2AUjEwAAJt+Kq/alW6OAAAAAElFTkSuQmCC","orcid":"","institution":"Wenzhou People's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Hongping","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-07-23 05:38:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7192354/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7192354/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89272407,"identity":"6e3c099f-dc38-4068-8d24-c1ed18138110","added_by":"auto","created_at":"2025-08-18 09:05:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1987117,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of this study.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/4198c8ab6d9591873378a1bf.png"},{"id":89272409,"identity":"1093f72a-ff8b-4998-b627-32576f67c87d","added_by":"auto","created_at":"2025-08-18 09:05:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1116861,"visible":true,"origin":"","legend":"\u003cp\u003eDifferences in clinical factors between GDM and control groups (P\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003eBox-violin plot of age.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003eBox-violin plot of family history of diabetes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003eBox-violin plot of gestational hypertension.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003eBox-violin plot of family history of hypertension.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003eBox-violin plot of TGAb.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/f57dbd3c4020eb6538b840aa.png"},{"id":89272411,"identity":"4aaa23ed-a77f-4dce-a174-b7b888f0f97b","added_by":"auto","created_at":"2025-08-18 09:05:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":178575,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation analysis of clinical factors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Pearson correlation coefficients for important clinical factors (21 features).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Spearman correlation coefficient of important clinical factors (21 features).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003eImportance of twenty-one features in random forest analysis.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/e03fec701f468d9af0dd1cf9.png"},{"id":89274054,"identity":"8ff7b26e-1fca-4254-a599-90bbcba0e761","added_by":"auto","created_at":"2025-08-18 09:13:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":276418,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation analysis of clinical factors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003eImportance of eighteen features in random forest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Grey relational analysis for important clinical factors (18 features).\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/49d0db1d314213d89a4c2e95.png"},{"id":89272414,"identity":"e5594540-2dbd-47a6-9ca5-14462ea1460c","added_by":"auto","created_at":"2025-08-18 09:05:47","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":142094,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of calibration curves and nomogram to evaluate four clinical prediction models.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003eCalibration curves for the stepwise regression model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003eCalibration curves for the best subset selection model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003eCalibration curves for the random forest model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003eCalibration curves for the support vector machines model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e Nomogram for the stepwise regression model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e Nomogram for the best subset selection model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e Nomogram for the random forest model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e Nomogram for the support vector machines model.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/63ce9b2e43aa7c39c4c40f81.png"},{"id":89274056,"identity":"90ade2bf-a208-4217-bfd9-91ae6145a801","added_by":"auto","created_at":"2025-08-18 09:13:47","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1016125,"visible":true,"origin":"","legend":"\u003cp\u003eDecision curve analysis and clinical impact curve for evaluating four clinical prediction models.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Decision curve analysis for the stepwise regression model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e Decision curve analysis for the best subset selection model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e Decision curve analysis for the random forest model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e Decision curve analysis for the support vector machines model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e Clinical impact curve for the stepwise regression model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e Clinical impact curve for the best subset selection model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e Clinical impact curve for the random forest model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e Clinical impact curve for the support vector machines model.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/d887734eec6b05c101f07be4.png"},{"id":105924885,"identity":"07e0f193-ba57-479b-accf-f27458b1ec98","added_by":"auto","created_at":"2026-04-01 13:13:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6547585,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7192354/v1/6c375af5-2ae1-479a-9ecb-6b906d81900f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Clinical prediction model for gestational diabetes mellitus utilizing thyroid function indicators: a retrospective cohort study","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eGestational diabetes mellitus (GDM) is defined as the first-time detection or occurrence of varying degrees of glucose intolerance during pregnancy, making it the most common metabolic disorder in the later stages of pregnancy (\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). According to the International Diabetes Federation (IDF), the global prevalence of GDM is approximately 14.0%, with rates exceeding 20% in regions like the Middle East, North Africa, and Southeast Asia (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). In a study conducted in East Asia and Southeast Asia, China ranked third in terms of incidence, with nearly one in nine women experiencing GDM (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Data from Chinese mainland indicate an overall incidence of 14.8% (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e), with evidence suggesting a continuous increase in recent years. Factors such as rising obesity rates among women of childbearing age and increased maternal age contribute to the annual increase in GDM incidence (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Research reveals that GDM affects approximately 15% of pregnancies worldwide each year, impacting around 180,000 newborns (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). GDM not only increases the likelihood of cesarean section and complications like hypertension during pregnancy but also poses risks to infants, including preterm birth, respiratory distress, and jaundice (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Follow-up studies on women with postpartum GDM show that their offspring have a higher risk of developing diabetes, obesity, and cardiovascular diseases (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCurrently, GDM is typically diagnosed in the late stages of pregnancy (24\u0026ndash;28 weeks) through the Oral Glucose Tolerance Test (OGTT). However, this method limits the early diagnosis, monitoring, and prediction of disease outcomes (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). As pregnancy is a dynamic and complex process, relying solely on a single indicator at a specific time point cannot fully capture the risk factors influencing GDM in expectant mothers. During pregnancy, hormonal fluctuations, including changes in glucose homeostasis and thyroid function, are common (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Thyroid hormones directly affect insulin secretion and glucose metabolism. Early pregnancy thyroid dysfunction (TD) may predict impaired glucose tolerance later in pregnancy (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). A significant positive correlation has been observed between TD and GDM in women (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Subclinical hypothyroidism with positive anti-thyroid autoantibodies significantly increases the risk of GDM (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e), while pregnant women with subclinical hyperthyroidism are less likely to develop GDM (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Recent studies indicate that even elderly pregnant women with normal thyroid function, but the presence of thyroid autoimmunity are at a higher risk of developing GDM (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Some scholars have proposed focusing on TD patients in the early and mid-stages of pregnancy to predict GDM (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Evidence suggests that using thyroid indicators to predict GDM and post-OGTT blood glucose levels is effective (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). However, current studies often consider only a few thyroid indicators (\u003cspan additionalcitationids=\"CR25 CR26 CR27\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), lacking comprehensive analysis and an investigation into the combined effects of multiple parameters.\u003c/p\u003e\u003cp\u003eClinical predictive models built on machine learning algorithms in the field of artificial intelligence can help comprehend or explain the causal mechanisms underlying exposure-outcome relationships (\u003cspan additionalcitationids=\"CR30 CR31\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). Increasingly, research is utilizing machine learning algorithms to identify risk factors for GDM and develop early prediction models (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). For instance, Artzi et al. achieved accurate and rapid stratification and prediction of GDM using machine learning methods with retrospective cohort data from 588,622 Israeli pregnant women (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). Beyond conventional clinical features like routine examination indicators and medical history (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e), GDM classification prediction models have also been constructed using bacterial biomarkers from the oral microbiome (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e) and endogenous metabolites (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). However, the exploration of the correlation between thyroid indicators and GDM has largely been confined to statistical data analysis (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e), with limited research focused on constructing clinical predictive risk models for GDM using machine learning algorithms.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1 Study population\u003c/h2\u003e\u003cp\u003eThis retrospective case study was conducted on Chinese pregnant women utilizing data gathered from the People's Hospital of Wenzhou City in Zhejiang Province. Participants included pregnant women who gave birth between January 2010 and June 2022, with all prenatal examinations registered at the hospital. The study focused on native females from southern China, with demographic and clinical characteristics extracted from the Hospital Information System (HIS). A total of 59,713 pregnant women were screened for inclusion. According to the American Diabetes Association (ADA) criteria (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e), individuals with a history of GDM or pre-existing diabetes were excluded. Inclusion criteria were established for maternal basic information and hospital examination records (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), which were independently screened by two researchers, using only intersected data. Ultimately, 5,229 eligible women were selected, comprising 1,035 cases in the GDM group and 4,194 controls in the comparison group, achieving a case-to-control ratio of approximately 1:4. The study was approved by the Ethics Committee of Wenzhou People's Hospital, and all participants provided informed consent (No: KY-2022-277).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2 GDM ascertainment\u003c/h2\u003e\u003cp\u003eThe diagnosis of GDM in pregnant women was determined through retrospective medical record review. Women included in the study were required to fast for at least eight hours before consuming a 300 ml solution containing 75g of glucose between the 24th and 28th week of pregnancy. Based on the 2010 diagnostic criteria published by the International Association of Diabetes and Pregnancy Study Groups (IADPSG) (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e), obstetricians diagnosed GDM using OGTT results. Blood glucose concentrations were measured before drinking the glucose solution, at 1 hour, and at 2 hours. Women with blood glucose levels below the thresholds of 5.1 mmol/L, 10.0 mmol/L, and 8.5 mmol/L (92mg/dL, 180mg/dL, 153mg/dL) were considered to have normal glucose tolerance. Those with glucose levels equal to or exceeding any of these thresholds were diagnosed with GDM.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3 Testing for thyroid function\u003c/h2\u003e\u003cp\u003eSerum samples were collected during the first prenatal examination. In this cohort study, a thyroid function analyzer was used to measure thyroid function markers, following the guidelines for diagnosis and treatment of thyroid diseases in China (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e), and considering the specific laboratory conditions and testing methods used at our hospital. Normal ranges for thyroid function tests included total triiodothyronine (TT3) concentrations of 0.80\u0026thinsp;\u0026minus;\u0026thinsp;2.00 ng/ml and total thyroxine (TT4) concentrations of 51.3\u0026thinsp;\u0026minus;\u0026thinsp;140.6 ng/ml. The TT3/TT4 ratio was calculated by dividing the of TT3 concentration (ng/ml) by TT4 concentration (ng/ml). Free triiodothyronine (FT3) concentrations ranged from 2.8\u0026thinsp;\u0026minus;\u0026thinsp;7.1 pmol/L, and free thyroxine (FT4) concentrations ranged from 7.9\u0026thinsp;\u0026minus;\u0026thinsp;16.4 pmol/L. The FT3/FT4 ratio was calculated by dividing FT3 (pmol/L) by FT4 (pmol/L). Normal ranges for thyroid-stimulating hormone (TSH) concentrations were 0.38\u0026thinsp;\u0026minus;\u0026thinsp;5.33 mIU/L, and for thyroglobulin (TG) concentrations, 0.8\u0026thinsp;\u0026minus;\u0026thinsp;68.0 ng/ml. Thyroid peroxidase antibodies (TPOAb) and thyroglobulin antibodies (TGAb), which are associated with autoimmune thyroid conditions, were considered positive if concentrations were \u0026ge;\u0026thinsp;35 IU/ml and \u0026ge;\u0026thinsp;4.0 IU/ml, respectively.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4 Statistical analysis\u003c/h2\u003e\u003cp\u003eReal-world clinical data analysis was conducted using IBM SPSS Statistics 22. For normally distributed continuous variables such as age and height, means\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviations (SD) were calculated, and independent samples t-tests were used to compare group means. For binary and categorical data, frequencies and percentages were used, with differences between groups analyzed using chi-square tests, such as for family history of diabetes, hypertension, thyroid function, and other pregnancy complications. Non-normally distributed continuous data were described using quartiles, and differences between groups were inferred using rank-sum tests, such as for thyroid indicators including TT3, TT4, and FT3. Seven thyroid indicators and two calculated ratios (TT3/TT4, FT3/FT4) underwent quartile-based continuous analysis. Logistic regression analysis was first conducted to evaluate the odds ratio (OR), 95% confidence intervals (CI), and P-values for each thyroid marker. Four factors showing significant differences in baseline analysis (age, family history of diabetes, pregnancy-induced hypertension, and family history of hypertension) were then included as covariates to obtain adjusted OR values (aORs). Since thyroid antibody TGAb levels may affect thyroid function and glucose homeostasis (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e), TGAb was categorized into negative and positive subsets and included in the model. Logistic regression analysis of thyroid markers was performed again on both case and control groups to observe differences in ORs, aORs, confidence intervals, and P-values. A significance level of p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5 Screening predictor variables and constructing GDM models\u003c/h2\u003e\u003cp\u003ePearson and Spearman correlation coefficients were calculated for the 21 included features to assess their individual impact on GDM outcomes. These features included age, height, family history of diabetes, scarred uterus, gestational hypertension, family history of hypertension, family history of cancer, re-pregnancy, polycystic ovary syndrome, ovarian cysts, thyroid function, TT3, TT4, TT3/TT4, FT3, FT4, FT3/FT4, TSH, Tg, TPOAb, and TGAb. Both Pearson and Spearman coefficients were used to analyze the relationship between pairs of data columns. Further data analysis employed the Random Forest (RF) algorithm to calculate the contribution of included risk factors to clinical outcomes, assessing the relative importance of these features in relation to GDM outcomes. Grey Relational Analysis (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e) was utilized to observe the influence of multifactor interactions on outcome indicators by eliminating low-contributing factors and incorporating clinically impactful ones,.\u003c/p\u003e\u003cp\u003eThe final cohort included only the most influential factors affecting clinical outcomes. The data were randomly divided into training and a testing sets with a ratio of 7:3. Four clinical prognostic models, including stepwise regression (SR), best subset selection (BSS), RF, and support vector machines (SVM), were constructed using R software (version 4.1.3). These models aimed to identify the most important clinical features and risk factors, with each algorithm applying its own criteria to determine output results. To evaluate the models, the area under the curve (AUC, 95% CI), sensitivity, specificity, true positive rate (TPR), true negative rate (TNR), concordance index (C-index, 95% CI), X-squared, degree of freedom (DF), and Hosmer and Lemeshow goodness-of-fit (GOF) test (P-value) were calculated. The calibration function was used to generate continuous recalibration plots in order to assess the consistency between predicted values and actual observations. Additionally, nomogram graphs were constructed for each of the four models to integrate multiple predictive indicators and illustrate the relationships between variables in the prediction models. Nomogram graphs assigned scores to different levels of each influential factor based on their contribution to the outcome variable (size of regression coefficients). These scores were summed to obtain a total score, and a mathematical transformation was applied to the total score to predict the probability of the outcome event. Decision curves and clinical impact curves were also plotted to evaluate the predictive abilities of different models.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 Baseline characteristics between GDM and control groups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter screening 59,713 samples that met the study criteria, a total of 5,229 pregnant women were included in the final analysis. Based on the research data, there were 1,035 women in the GDM group and 4,194 women in the control group (non-GDM), resulting in a GDM prevalence of approximately 19.8% (1,035/5,229). The demographic and clinical characteristics of the overall study population in both groups were presented (Table 1). The analysis showed that, compared to the control group, women in the GDM group were significantly older (34\u0026plusmn;4.3 years vs 32\u0026plusmn;4.4 years, P \u0026lt; 0.001). Women with a family history of diabetes were more likely to develop GDM compared to those without such a history (4.7% vs 0.9%, P \u0026lt; 0.001). Pregnant women diagnosed with gestational hypertension also had a higher risk of GDM compared to those without this condition (8.4% vs 3.1%, P \u0026lt; 0.001). Additionally, a family history of hypertension increased the risk of GDM (6.2% vs 4.2%, P = 0.006), highlighting the need for preventive measures. There was a significant statisticant difference in the distribution of TGAb between the GDM and control groups (P = 0.001). Lastly, we portrayed the actual distribution of these five clinical factors (\u003cstrong\u003eFigure 2A-E\u003c/strong\u003e). No significant differences were observed in the remaining 17 characteristic factors between the two groups.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Univariate logistic regression analysis for thyroid function indicators\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore the relationship between thyroid function indicators and the risk of GDM, the indicators, except for TGAb, were divided into four quartiles (P \u0026gt; 0.05). The quartiles, based on data distribution at 25%, 50%, 75%, and 100% intervals, served as the basis for logistic regression analysis. Considering significant differences in age, family history of diabetes, gestational hypertension, and family history of hypertension between the GDM and the control groups, these four factors were adjusted as covariates in the multifactor logistic regression. According to the analysis results (Table 2), there was a significant difference in GDM incidence among the TT3 level quartiles (P \u0026lt; 0.013), with lower TT3 levels associated with an increased risk of GDM (ORQ1 = 1.255, 95% CI 1.040\u0026ndash;1.514, p = 0.018). After adjusting for covariates, no significant difference was found in TT3 levels. Analysis of FT3 indicated that higher FT3 levels were negatively associated with GDM risk (ORQ3 = 0.784, 95% CI 0.646\u0026ndash;0.951, p = 0.013), a relationship that remained significant after adjustment (aORQ1 = 0.800, 95% CI 0.656\u0026ndash;0.975, p = 0.027). Regarding TSH levels, lower levels were linked to an increased risk of GDM (ORQ1 = 1.211, 95% CI 1.001\u0026ndash;1.465, p = 0.049), with this association becoming more pronounced after adjustment (aORQ1 = 1.288, 95% CI 1.059\u0026ndash;1.565, p = 0.011). These findings suggest that low levels of TSH increase GDM risk from both univariate and multivariate perspectives. Other factors, such as TT4, TT3/TT4, FT3, FT4, FT3/FT4, TPOAb, and Tg, did not show significant differences before and after adjustment (P \u0026gt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Univariate logistic regression analysis of TGAb negative and positive groups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo examine the association between TGAb levels, other thyroid indicators, and GDM risk, the study population was divided into TGAb-positive (+, \u0026ge;4.0 IU/ml) and TGAb -negative (-,\u0026lt;4.0 IU/ml) groups. Other thyroid indicators were grouped into quartiles for logistic regression analysis, with four covariates included for model adjustment as described in section 3.2. The analysis sample comprised 3,275 TGAb- negative women and 1,954 TGAb+ positive women (Table 3). In the TGAb- negative group, significant differences were observed between the GDM and control groups in the Q2 range of TT3 levels (1.05-1.2), indicating that higher TT3 levels were associated with a reduced risk of GDM (ORQ2 = 0.684, 95% CI 0.549\u0026ndash;0.852, p = 0.001; aORQ2 = 0.719, 95% CI 0.574\u0026ndash;0.899, p = 0.004). Conversely, lower FT3/FT4 levels were linked to a higher risk of GDM (ORQ3 = 1.275, 95% CI 1.043\u0026ndash;1.558, p = 0.018; aORQ3 = 1.323, 95% CI 1.077\u0026ndash;1.625, p = 0.008). The impact of TSH levels on GDM risk aligned with the TT3 results, where higher TSH concentrations reduced the risk of GDM (ORQ2 = 0.735, 95% CI 0.580\u0026ndash;0.932, p = 0.010; aORQ2 = 0.766, 95% CI 0.629\u0026ndash;0.932, p = 0.008; aORQ4 = 0.744, 95% CI 0.639\u0026ndash;0.944, p = 0.011). Similar findings were noted in the Q3 group (1.4-12) for TPOAb, which showed a significant association with GDM (ORQ3 = 0.735, 95% CI 0.588\u0026ndash;0.995, p = 0.046). In the TGAb+ positive group, the quartiles of thyroid indicators did not show significant significance (P \u0026gt; 0.05). However, it is worth noting that when the TPOAb variable was included in the regression equation, all results exhibited a confidence interval of 0 and a large OR value, suggesting a strong positive impact of TPOAb on the likelihood of pregnant women developing GDM.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4 Screening for clinical risk factors\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePearson correlation analysis revealed a predominantly positive association between demographic information of the participants and the incidence of GDM, whereas thyroid examination indicators tended to show a negative correlation (\u003cstrong\u003eFigure 3A\u003c/strong\u003e). Among different subgroups of GDM, five factors (age, family history of diabetes, family history of hypertension, gestational hypertension, and TGAb) exhibited significant difference (P \u0026lt; 0.05). Notably, maternal height and thyroid function demonstrated relatively mild correlations with GDM risk, even though their P-values were not significantly different. Spearman correlation analysis highlighted height as an evident negative factor, implying that shorter height correlates with higher GDM risk (\u003cstrong\u003eFigure 3B)\u003c/strong\u003e. Interesting, most thyroid indicators showed negative correlation, while Tg had a positive correlation, suggesting the need for further investigation. Random Forest (RF) analysis of relative importance identified age and all thyroid test indicators as having high contributions to GDM risk (\u003cstrong\u003eFigure 3C\u003c/strong\u003e). Factors like family history of cancer, polycystic ovary syndrome, and ovarian cysts had minimal impact and were excluded from further analysis. The latest calculations showed that age, TSH, TT4, Tg, and FT3/FT4 were the top five influential factors (\u003cstrong\u003eFigure 4A\u003c/strong\u003e). A total of 18 factors were selected to construct the prognostic model, with grey correlation analysis indicates indicating a strong association (coefficient of association \u0026ge; 0.6) among the factors \u003cstrong\u003e(Figure 4B)\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.5 Development of a clinical predictive model for GDM\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe stepwise regression (SR) model was developed using a bidirectional progressive regression method, examining all variables in the model with each addition and eliminating those with insignificant effects to find the optimal combination. After twelve iterations, seven core factors were identified, resulting in an Akaike information criterion (AIC) value of 4988.9. The best subset selection (BSS) technique evaluated all combinations of feature variables, fitting 131,071 models to identify the optimal set. In comparison to the SR model, the BUS model included FT4 as an additional crucial variable, bringing the total to eight core factors. The RF model constructed 50 \u0026nbsp; decision trees and evaluated three variable at each splitting point, ultimately selecting \u0026nbsp;10 important factors. The average squared residuals were 0.166. Meanwhile, the SVM model explores linear and functional, polynomial kernel function, radial basis kernel function, and sigmoid kernel function constructions to identify the most suitable model and outputs the top five influential factors (Table 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.6 Model validation and evaluation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEvaluation metrics for the clinical prediction models included AUC (95% CI), sensitivity, specificity, TPR, TNR, C-index (95% CI), X-squared, DF, and P-value. The SVM model exhibited the highest accuracy with an AUC of 0.824 (0.8137-0.8345), though its C-index value was moderate at 0.645 (0.6263, 0.6430) (Table 5). The SVM model also had superior TPR (0.867) and TNR (0.823) compared to the other three models. Additionally, the SVM model demonstrated the superior calibration (0.629). Despite moderate C-index values across all models, highlighting the complexity of GDM prediction, the models were ranked by AUS as follows: SVM \u0026gt; SR \u0026gt; BSS \u0026gt; RF. Calibration curves were constructed to assess the disparity between predicted and actual outcomes (\u003cstrong\u003eFigure 5A-D\u003c/strong\u003e). Nomograms for each model were created based on independent risk factors, illustrating the relationships between prediction factors and GDM risk (\u003cstrong\u003eFigure 5E-H\u003c/strong\u003e). Risk scores for each factor were summed to calculate the total GDM risk score, which was then transformed into the probability of developing GDM. Decision curve indicated that the intervention measures are most beneficial when the GDM risk probability is between 0.1 and 0.4 (\u003cstrong\u003eFigure 6A-D\u003c/strong\u003e). The clinical impact curve depicted the number of true positives at various threshold probabilities, suggesting that intervention when GDM risk ranges from 10% to 40% yields favorable clinical outcomes. Ultimately, the SUM model achieved an accuracy of 81.52% on the test set (\u003cstrong\u003eFigure 6E-H\u003c/strong\u003e).\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn our study, age, family history of diabetes, pregnancy-induced hypertension, and family history of hypertension were identified as significant risk factors for increased susceptibility to GDM. Numerous studies have already shown that both age and a family history of diabetes contribute to an increased risk of GDM (\u003cspan additionalcitationids=\"CR46 CR47\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e), which aligns with our research findings. Therefore, it is crucial to prioritize GDM prevention in pregnant women of advanced age and those with a family history of diabetes. Close attention should be paid to the health responses of these individuals during pregnancy, allowing timely intervention measures. GDM-induced hyperglycemia is known to affect systemic arteries when insulin resistance occurs, resulting in elevated blood pressure (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e). Conversely, if gestational hypertension develops first, it can trigger inflammatory responses or cause abnormalities in vascular function, potentially interfering with the normal action of insulin and thereby increasing the risk of GDM. It is worth noting that these two pregnancy complications share some common risk factors, such as obesity, age, and family history, which should be considered by women of childbearing age who are planning for pregnancy or are already pregnant. Some studies have indicated a higher risk among pregnant women with a history of hypertension (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e). However, in reality, a family history of hypertension is a risk factor that cannot be separated from blood pressure and blood glucose considerations. In our Spearman correlation analysis, we discovered a negative relationship between height and GDM risk, which might be attributed to excessive weight gain during pregnancy. As height decreases, the impact of weight gain during pregnancy becomes more pronounced, affecting maternal metabolism and blood glucose levels. This finding suggests that shorter women may require extra attention to weight control while ensuring proper nutrition during pregnancy.\u003c/p\u003e\u003cp\u003eThyroid hormones play a crucial role in glucose metabolism and maintaining internal balance. They influence both the growth and development of the fetus and the adverse pregnancy outcomes for mothers, making thyroid disorders prevalent among pregnant women (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). However, the underlying mechanisms linking abnormal thyroid function markers to the outcome of GDM remain unclear (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). TGAb serves as a serologic marker for autoimmune thyroid disease. If women have a family history of diabetes and thyroid disease, as well as a positive TGAb result ranging from 8\u0026ndash;16%, they are more likely to develop GDM (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e). Elevated levels of early-pregnancy FT3 and TT3 are associated with an increased risk of GDM (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e), a finding consistent with our study regarding FT3 levels during the early pregnancy. However, our results do not support a strong association between TT3 levels and early detection of GDM. Similarly, higher levels of TSH in early pregnancy indicate a higher risk of GDM (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e), whereas low TSH levels during the later stages of pregnancy become a risk factor. Further research is needed to understand the reasons behind these phenomena. Some studies suggest a lack of average TSH levels during early pregnancy (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e), whereas our research demonstrates that higher TSH levels effectively reduce the risk of GDM in pregnant women. Women who test positive for TGAb and TPOAb are more susceptible to GDM (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), which aligns with our findings, establishing a positive correlation. These results emphasize the importance of monitoring thyroid antibodies and hormone levels in pregnant women to effectively prevent the occurrence of GDM.\u003c/p\u003e\u003cp\u003eRegarding feature selection for clinical prediction models, we embarked on a comprehensive analysis, looking beyond univariate correlations to recognize the collective influence of multiple factors due to the complex nature of GDM. We proposed a novel approach to assess the correlation between exposure and outcome, using Pearson and Spearman correlation coefficients to ascertain the direction of associations, the RF method to determine the relative importance of each exposure on the outcome, and ultimately calculating the grey relational degree to capture multidimensional interplay. To construct the clinical prediction model, we compared four different methods. The SR model has been utilized since 2010 to identify GDM predictive factors (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e), while the BSS model uncovered a relationship between alterations in gut microbiota and GDM status (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e). Notably, in a study focusing on the early prediction of GDM in the Chinese population (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e), the SVM model achieved a maximum accuracy of only 77%. However, our search revealed that the BSS model has not yet been applied in the field of GDM clinical prediction, making it a novel approach to prognostic model construction. In our retrospective cohort study, the SVM model exhibited commendable predictive ability due to its robustness in tackling complex classification problems (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e). Future investigations into clinical prediction models and the identification of hazardous exposures would integrate multiple methods as proposed in our study, with an emphasis on using the SVM approach for model development.\u003c/p\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThe findings from this retrospective cohort study indicate a significant correlation between thyroid function parameters and GDM outcomes. Our research underscores the necessity of thyroid hormone screening as part of early GDM risk assessment. We propose the concept of multiple risk factors interact to influence clinical outcomes and offer screening methods. The goal is to develop a clinical prediction model that prompts obstetricians to pay more attention to these risk factors during pregnancy, enabling early screening and intervention to reduce the incidence of GDM. However, it is important to acknowledge the limitations of our study. Future research should focus on designing prospective cohort studies, utilizing time series to monitor thyroid hormone changes throughout the entire pregnancy, and forming dynamic real-time feedback. Additionally, increasing the sample size for training and testing the model will help make it more comprehensive. Lastly, it is crucial to include external cohorts for training and validation to avoid regional bias in the samples and support the generalizability of this model.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAcronym\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFull name\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eGDM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003egestational diabetes mellitus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTGAb\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003ethyroglobulin antibody\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eIDF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eInternational Diabetes Federation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eOGTT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eOral Glucose Tolerance Test\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003ethyroid dysfunction\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eHIS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eHospital Information System\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eADA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eAmerican Diabetes Association\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eIADPSG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eInternational Association of Diabetes and Pregnancy Study Groups\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003etotal triiodothyronine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003etotal thyroxine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eFT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eFree triiodothyronine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eFT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eFree thyroxine\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTSH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003ethyroid-stimulating hormone\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003ethyroglobulin\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTPOAb\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eThyroid peroxidase antibodies\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eRF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eRandom Forest\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eSR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003estepwise regression\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eBSS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003ebest subset selection\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eAIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003eAkaike information criterion\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eSVM\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003esupport vector machines\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTPR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003etrue positive rate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eTNR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003etrue negative rate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eDF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003edegree of freedom\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 179px;\"\u003e\n \u003cp\u003eGOF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 374px;\"\u003e\n \u003cp\u003egoodness-of-fit\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, XL and HZ; Data curation, CC and WW; Formal analysis, HC and CL; Funding acquisition, XL; Investigation, HC; Methodology, XL; Project administration, HZ; Resources, YH, HC, CL, Li Chen and WW;; Software, YH; Supervision, HZ; Validation, YH, CC and LC; Visualization, CC; Writing \u0026ndash; original draft, XL; Writing \u0026ndash; review \u0026amp; editing, HZ. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Zhejiang Provincial Natural Science Foundation of China (LBY23H200008), the National Natural Science Foundation of China (T2341010), the Medical Health Science and Technology Project of Zhejiang Provincial (2023RC272), and the Science and Technology Planning Project of Wenzhou (ZY2021025 and Y2023088).\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data are available from the corresponding author upon reasonable request.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll studies were approved by the Ethical Committee of Wenzhou People\u0026apos;s Hospital (No: KY-2022-277). All participants in this study signed an informed consent form.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWritten informed consent has been obtained from the patients to publish this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSaravanan P. Gestational diabetes: opportunities for improving maternal and child health. Lancet Diabetes Endocrinol. 2020;8(9):793-800.\u003c/li\u003e\n\u003cli\u003eSzmuilowicz ED, Josefson JL, Metzger BE. Gestational Diabetes Mellitus. Endocrinol Metab Clin North Am. 2019;48(3):479-93.\u003c/li\u003e\n\u003cli\u003eCoustan DR. Gestational diabetes mellitus. Clin Chem. 2013;59(9):1310-21.\u003c/li\u003e\n\u003cli\u003eWang H, Li N, Chivese T, Werfalli M, Sun H, Yuen L, et al. IDF Diabetes Atlas: Estimation of Global and Regional Gestational Diabetes Mellitus Prevalence for 2021 by International Association of Diabetes in Pregnancy Study Group\u0026apos;s Criteria. Diabetes Res Clin Pract. 2022;183:109050.\u003c/li\u003e\n\u003cli\u003eNguyen CL, Pham NM, Binns CW, Duong DV, Lee AH. Prevalence of Gestational Diabetes Mellitus in Eastern and Southeastern Asia: A Systematic Review and Meta-Analysis. J Diabetes Res. 2018;2018:6536974.\u003c/li\u003e\n\u003cli\u003eGao C, Sun X, Lu L, Liu F, Yuan J. Prevalence of gestational diabetes mellitus in mainland China: A systematic review and meta-analysis. J Diabetes Investig. 2019;10(1):154-62.\u003c/li\u003e\n\u003cli\u003eSweeting A, Wong J, Murphy HR, Ross GP. A Clinical Update on Gestational Diabetes Mellitus. Endocr Rev. 2022;43(5):763-93.\u003c/li\u003e\n\u003cli\u003eModzelewski R, Stefanowicz-Rutkowska MM, Matuszewski W, Bandurska-Stankiewicz EM. Gestational Diabetes Mellitus-Recent Literature Review. J Clin Med. 2022;11(19).\u003c/li\u003e\n\u003cli\u003eYe W, Luo C, Huang J, Li C, Liu Z, Liu F. Gestational diabetes mellitus and adverse pregnancy outcomes: systematic review and meta-analysis. BMJ. 2022;377:e067946.\u003c/li\u003e\n\u003cli\u003eRetnakaran R, Ye C, Hanley AJ, Connelly PW, Sermer M, Zinman B. Treatment of Gestational Diabetes Mellitus and Maternal Risk of Diabetes After Pregnancy. Diabetes Care. 2023;46(3):587-92.\u003c/li\u003e\n\u003cli\u003eGrunnet LG, Hansen S, Hjort L, Madsen CM, Kampmann FB, Thuesen ACB, et al. Adiposity, Dysmetabolic Traits, and Earlier Onset of Female Puberty in Adolescent Offspring of Women With Gestational Diabetes Mellitus: A Clinical Study Within the Danish National Birth Cohort. Diabetes Care. 2017;40(12):1746-55.\u003c/li\u003e\n\u003cli\u003eZhang C, Catalano P. Screening for Gestational Diabetes. JAMA. 2021;326(6):487-9.\u003c/li\u003e\n\u003cli\u003eJin J. Screening for Gestational Diabetes. JAMA. 2021;326(6):577.\u003c/li\u003e\n\u003cli\u003ePinto S, Croce L, Carlier L, Cosson E, Rotondi M. Thyroid dysfunction during gestation and gestational diabetes mellitus: a complex relationship. J Endocrinol Invest. 2023.\u003c/li\u003e\n\u003cli\u003eMilovanović Z, Filimonović D, Soldatović I, Karadžov Orlić N. Can Thyroid Screening in the First Trimester Improve the Prediction of Gestational Diabetes Mellitus? J Clin Med. 2022;11(13).\u003c/li\u003e\n\u003cli\u003eYang Y, Li Q, Wang Q, Ma X. Thyroid antibodies and gestational diabetes mellitus: a meta-analysis. Fertil Steril. 2015;104(3).\u003c/li\u003e\n\u003cli\u003eJia M, Wu Y, Lin B, Shi Y, Zhang Q, Lin Y, et al. Meta-analysis of the association between maternal subclinical hypothyroidism and gestational diabetes mellitus. Int J Gynaecol Obstet. 2019;144(3):239-47.\u003c/li\u003e\n\u003cli\u003eKent NL, Young SL, Akison LK, Cuffe JSM. Is the link between elevated TSH and gestational diabetes mellitus dependant on diagnostic criteria and thyroid antibody status: a systematic review and meta-analysis. Endocrine. 2021;74(1):38-49.\u003c/li\u003e\n\u003cli\u003eLuo J, Wang X, Yuan L, Guo L. Association of thyroid disorders with gestational diabetes mellitus: a meta-analysis. Endocrine. 2021;73(3):550-60.\u003c/li\u003e\n\u003cli\u003eSitoris G, Veltri F, Ichiche M, Kleynen P, Praet J-P, Rozenberg S, et al. Association between thyroid autoimmunity and gestational diabetes mellitus in euthyroid women. Eur Thyroid J. 2022;11(2).\u003c/li\u003e\n\u003cli\u003eSert UY, Buyuk GN, Engin Ustun Y, Ozgu Erdinc AS. Is There Any Relationship Between Thyroid Function Abnormalities, Thyroid Antibodies and Development of Gestational Diabetes Mellitus (GDM) in Pregnant Women? Medeni Med J. 2020;35(3):195-201.\u003c/li\u003e\n\u003cli\u003eTang L, Li P, Zhou H, Li L. A longitudinal study of thyroid markers during pregnancy and the risk of gestational diabetes mellitus and post-partum glucose metabolism. Diabetes Metab Res Rev. 2021;37(4):e3441.\u003c/li\u003e\n\u003cli\u003eMennickent D, Ortega-Contreras B, Guti\u0026eacute;rrez-Vega S, Castro E, Rodr\u0026iacute;guez A, Araya J, et al. Evaluation of first and second trimester maternal thyroid profile on the prediction of gestational diabetes mellitus and post load glycemia. PLoS One. 2023;18(1):e0280513.\u003c/li\u003e\n\u003cli\u003eWu J-N, P\u0026eacute;rez-L\u0026oacute;pez FR, Peng T, Xie F, Li M-Q. Association of Early Pregnancy Free and Total Triiodothyronine With the Subsequent Risk of Gestational Diabetes Mellitus. Endocr Pract. 2023;29(1):40-7.\u003c/li\u003e\n\u003cli\u003eRaets L, Minschart C, Van den Bruel A, Van den Bogaert E, Van Crombrugge P, Moyson C, et al. Higher Thyroid fT3-to-fT4 Ratio Is Associated with Gestational Diabetes Mellitus and Adverse Pregnancy Outcomes. J Clin Med. 2022;11(17).\u003c/li\u003e\n\u003cli\u003eDemiral Sezer S, Topaloglu O. Effects of thyroid hormones in women with gestational diabetes. Gynecol Endocrinol. 2022;38(7):588-91.\u003c/li\u003e\n\u003cli\u003eZhang S, Wu Y, Pan W, Li G, Zhang D, Li S, et al. Free-Triiodothyronine to Free-Thyroxine Ratio Mediated the Effect of Prepregnancy Body Mass Index or Maternal Weight Gain During Early Pregnancy on Gestational Diabetes Mellitus. Endocr Pract. 2022;28(4):398-404.\u003c/li\u003e\n\u003cli\u003eLi P, Cui J, Li L, Chen X, Ouyang L, Fan J, et al. Association between isolated maternal hypothyroxinemia during the first trimester and adverse pregnancy outcomes in Southern Chinese women: a retrospective study of 7051 cases. BMC Pregnancy Childbirth. 2022;22(1):866.\u003c/li\u003e\n\u003cli\u003eEloranta S, Boman M. Predictive models for clinical decision making: Deep dives in practical machine learning. J Intern Med. 2022;292(2):278-95.\u003c/li\u003e\n\u003cli\u003eLi JK, Tang T, Zong H, Wu EM, Zhao J, Wu RR, et al. Intelligent medicine in focus: the 5 stages of evolution in robot-assisted surgery for prostate cancer in the past 20 years and future implications. Mil Med Res. 2024;11(1):58.\u003c/li\u003e\n\u003cli\u003eLi LS, Yang L, Zhuang L, Ye ZY, Zhao WG, Gong WP. From immunology to artificial intelligence: revolutionizing latent tuberculosis infection diagnosis with machine learning. Mil Med Res. 2023;10(1):58.\u003c/li\u003e\n\u003cli\u003eZhang YP, Zhang XY, Cheng YT, Li B, Teng XZ, Zhang J, et al. Artificial intelligence-driven radiomics study in cancer: the role of feature engineering and modeling. Mil Med Res. 2023;10(1):22.\u003c/li\u003e\n\u003cli\u003eHuang Q-F, Hu Y-C, Wang C-K, Huang J, Shen M-D, Ren L-H. Clinical First-Trimester Prediction Models for Gestational Diabetes Mellitus: A Systematic Review and Meta-Analysis. Biol Res Nurs. 2023;25(2):185-97.\u003c/li\u003e\n\u003cli\u003eArtzi NS, Shilo S, Hadar E, Rossman H, Barbash-Hazan S, Ben-Haroush A, et al. Prediction of gestational diabetes based on nationwide electronic health records. Nat Med. 2020;26(1):71-6.\u003c/li\u003e\n\u003cli\u003eWu Y, Ma S, Wang Y, Chen F, Zhu F, Sun W, et al. A risk prediction model of gestational diabetes mellitus before 16 gestational weeks in Chinese pregnant women. Diabetes Res Clin Pract. 2021;179:109001.\u003c/li\u003e\n\u003cli\u003eLi X, Zheng J, Ma X, Zhang B, Zhang J, Wang W, et al. The oral microbiome of pregnant women facilitates gestational diabetes discrimination. J Genet Genomics. 2021;48(1):32-9.\u003c/li\u003e\n\u003cli\u003eKoos BJ, Gornbein JA. Early pregnancy metabolites predict gestational diabetes mellitus: implications for fetal programming. Am J Obstet Gynecol. 2021;224(2):215.e1-.e7.\u003c/li\u003e\n\u003cli\u003eRawal S, Tsai MY, Hinkle SN, Zhu Y, Bao W, Lin Y, et al. A Longitudinal Study of Thyroid Markers Across Pregnancy and the Risk of Gestational Diabetes. J Clin Endocrinol Metab. 2018;103(7):2447-56.\u003c/li\u003e\n\u003cli\u003eLi X, Zuo J, Li Y-H, Tang Y-P, Bao Y-R, Ying H. Association between thyroid function and risk of gestational diabetes mellitus in assisted pregnancies: A retrospective cohort study. Diabetes Res Clin Pract. 2021;171:108590.\u003c/li\u003e\n\u003cli\u003eClassification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2021. Diabetes Care. 2021;44(Suppl 1):S15-S33.\u003c/li\u003e\n\u003cli\u003eMetzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33(3):676-82.\u003c/li\u003e\n\u003cli\u003eMiao Z, Li-Xin SJCJoPIM. A clinical interpretation of 2012 Chinese guidelines for the diagnosis and treatment of thyroid disease during pregnancy and postpartum. 2012.\u003c/li\u003e\n\u003cli\u003eChidakel A, Mentuccia D, Celi FS. Peripheral metabolism of thyroid hormone and glucose homeostasis. Thyroid. 2005;15(8):899-903.\u003c/li\u003e\n\u003cli\u003eGao D, Guo L, Wang F, Zhang Z. Study on the Spontaneous Combustion Tendency of Coal Based on Grey Relational and Multiple Regression Analysis. ACS Omega. 2021;6(10):6736-46.\u003c/li\u003e\n\u003cli\u003eLiu H, Li J, Leng J, Wang H, Liu J, Li W, et al. Machine learning risk score for prediction of gestational diabetes in early pregnancy in Tianjin, China. Diabetes Metab Res Rev. 2021;37(5):e3397.\u003c/li\u003e\n\u003cli\u003eWu Y-T, Zhang C-J, Mol BW, Kawai A, Li C, Chen L, et al. Early Prediction of Gestational Diabetes Mellitus in the Chinese Population via Advanced Machine Learning. J Clin Endocrinol Metab. 2021;106(3):e1191-e205.\u003c/li\u003e\n\u003cli\u003eBenhalima K, Van Crombrugge P, Moyson C, Verhaeghe J, Vandeginste S, Verlaenen H, et al. Estimating the risk of gestational diabetes mellitus based on the 2013 WHO criteria: a prediction model based on clinical and biochemical variables in early pregnancy. Acta Diabetol. 2020;57(6):661-71.\u003c/li\u003e\n\u003cli\u003eGao S, Leng J, Liu H, Wang S, Li W, Wang Y, et al. Development and validation of an early pregnancy risk score for the prediction of gestational diabetes mellitus in Chinese pregnant women. BMJ Open Diabetes Res Care. 2020;8(1).\u003c/li\u003e\n\u003cli\u003eMate A, Blanca AJ, Salsoso R, Toledo F, Stiefel P, Sobrevia L, et al. Insulin Therapy in Pregnancy Hypertensive Diseases and its Effect on the Offspring and Mother Later in Life. Curr Vasc Pharmacol. 2019;17(5):455-64.\u003c/li\u003e\n\u003cli\u003eWang Y, Ge Z, Chen L, Hu J, Zhou W, Shen S, et al. Risk Prediction Model of Gestational Diabetes Mellitus in a Chinese Population Based on a Risk Scoring System. Diabetes Ther. 2021;12(6):1721-34.\u003c/li\u003e\n\u003cli\u003eDonovan BM, Breheny PJ, Robinson JG, Baer RJ, Saftlas AF, Bao W, et al. Development and validation of a clinical model for preconception and early pregnancy risk prediction of gestational diabetes mellitus in nulliparous women. PLoS One. 2019;14(4):e0215173.\u003c/li\u003e\n\u003cli\u003eLee SY, Pearce EN. Assessment and treatment of thyroid disorders in pregnancy and the postpartum period. Nat Rev Endocrinol. 2022;18(3):158-71.\u003c/li\u003e\n\u003cli\u003eShahbazian H, Shahbazian N, Rahimi Baniani M, Yazdanpanah L, Latifi SM. Evaluation of thyroid dysfunction in pregnant women with gestational and pre-gestational diabetes. Pak J Med Sci. 2013;29(2):638-41.\u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez Alba JJ, Castillo Lara M, Jim\u0026eacute;nez Heras JM, Moreno Cort\u0026eacute;s R, Gonz\u0026aacute;lez Mac\u0026iacute;as C, Vilar S\u0026aacute;nchez \u0026Aacute;, et al. High First Trimester Levels of TSH as an Independent Risk Factor for Gestational Diabetes Mellitus: A Retrospective Cohort Study. J Clin Med. 2022;11(13).\u003c/li\u003e\n\u003cli\u003eShahid MM, Rahman KMT, Gomes RR, Ferdous M, Ferdousi S, Zahan T. Association of gestational diabetes mellitus and thyroid status during pregnancy: a cross-sectional study in a tertiary health care center of Bangladesh. Gynecol Endocrinol. 2021;37(4):312-4.\u003c/li\u003e\n\u003cli\u003eSavvidou M, Nelson SM, Makgoba M, Messow C-M, Sattar N, Nicolaides K. First-trimester prediction of gestational diabetes mellitus: examining the potential of combining maternal characteristics and laboratory measures. Diabetes. 2010;59(12):3017-22.\u003c/li\u003e\n\u003cli\u003eKuang Y-S, Lu J-H, Li S-H, Li J-H, Yuan M-Y, He J-R, et al. Connections between the human gut microbiome and gestational diabetes mellitus. Gigascience. 2017;6(8).\u003c/li\u003e\n\u003cli\u003eVan Belle V, Lisboa P. White box radial basis function classifiers with component selection for clinical prediction models. Artif Intell Med. 2014;60(1):53-64.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1: Demographic and Clinical characteristic of enrolled pregnant woman\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"567\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" colspan=\"3\" style=\"width: 26.3889%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003cstrong\u003eTotal (n=5229)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" colspan=\"3\" style=\"width: 26.2626%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eGDM (n=1035,19.8%)\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" colspan=\"3\" style=\"width: 26.2626%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eControl (n=4194,80.2%)\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber(%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber(%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber(%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eMedian\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge(years)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e32.4\u0026plusmn;4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e34\u0026plusmn;4.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e32\u0026plusmn;4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHeight(cm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e159.7\u0026plusmn;18.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e159.2\u0026plusmn;4.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e159.9\u0026plusmn;20.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.278\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFamily history of diabetes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e135 (2.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e49 (4.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e36 (0.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003ecarred uterus\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2824 (54%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e540 (52.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2284 (556%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.187\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eGestational hypertension\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e217 (4.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e87 (8.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e130 (3.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFamily history of hypertension\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e240 (4.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e64 (6.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e176 (4.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.006\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFamily history of cancer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e90 (1.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e17 (1.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e73 (1.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.828\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRe-pregnancy\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2359 (45.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e470 (45.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1889 (45%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.830\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePolycystic ovary syndrome\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9 (0.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2 (0.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (1.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.855\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eOvarian cysts\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e25 (0.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (0.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e21 (0.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.633\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eThyroid\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;f\u003c/strong\u003e\u003cstrong\u003eunction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHypothyroidism\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e220 (4.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e52 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e168 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003e0.236\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNormal thyroid function\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4978 (95.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e975 (94.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4003 (95.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHyperthyroidism\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e31 (0.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (0.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e23 (0.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eThyroid function indicators\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;Median(Quartile)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMedian(Quartile)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMedian(Quartile)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTT3(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.2 (1.04,1.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.19 (1.02,1.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.2(1.04,1.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.068\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTT4(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.8 (75.9,98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e85.7 (74.7,97.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e86.9 (76.2,98.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.058\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTT3/TT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.014(0.012,0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.014(0.012,0.017)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.014(0.012,0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.985\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFT3(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4.2(3.8,4.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4.2(3.8,4.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4.2(3.8,4.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.127\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFT4(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8.9(7.7,11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8.9(7.8,10.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8.9(7.7,11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.926\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFT3/FT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.479(0.359,0.572)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.479(0.358,0.563)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.478(0.359,0.575)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.435\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTSH(mIU/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.81(1.27,2.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.81(1.2,2.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.81(1.29,2.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.250\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTPOAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.3(0.4,12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.3(0.4,12.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.3(0.4,12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.470\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTGAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.9(\u0026lt;0.9,10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.9(\u0026lt;0.9,10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.9(\u0026lt;0.9,10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTg(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10.5(5.9,17.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11(6.2,18.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10.4(5.8,17.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.052\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTPOAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026gt;=35 IU/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e267 (5.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e46 (4.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e221 (5.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e0.280\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt;35 IU/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4962 (94.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e989 (95.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3973 (94.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTGAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026gt;=4.0 IU/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1954 (37.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e394 (38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1560 (37.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e0.604\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026lt;4.0 IU/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e3275 (62.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e641 (62%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2634 (62.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eNote: P\u003csup\u003ea\u003c/sup\u003e : P-values were obtained from the statistical results. Continuous data are normally distributed using the independent samples t-test, continuous data are not normally distributed using the rank sum test, and categorical data use the chi-square test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2: GDM Risk Analysis of thyroid function in all enrolled woman\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" align=\"left\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eGDM (n=1035)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eControl (n=4194)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP for trend\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eaOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTT3(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.013\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~1.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e310 (30.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1056 (25.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e1.255 (1.040,1.514)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.018\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.208 (0.996,1.466)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.056\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (1.05-1.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e229 (22.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1051 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.932 (0.763,1.137)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.940 (0.767,1.153)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.555\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (1.21-1.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e249 (24.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1031 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.033 (0.849,1.256)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.749\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.049 (0.859,1.282)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.632\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (1.40-3.68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e247 (23.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1056 (25.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTT4(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.378\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e278 (26.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1033 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (76-86.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e254 (24.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1063 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.888 (0.734,1.074)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.221\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.938 (0.772,1.139)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.516\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (86.9-98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e261 (25.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1039 (24.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.933 (0.772,1.128)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.476\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.003 (0.826,1.218)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.976\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (98.1-235.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e242 (23.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1059 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.849 (0.700,1.029)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.096\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.958 (0.786,1.168)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.674\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTT3/TT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.708\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~0.012)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e259 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1047 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (0.013-0.014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e248 (24.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1061 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.945 (0.778,1.147)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.567\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.982 (0.805,1.197)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.856\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (0.015-0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e256 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1048 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.987 (0.814,1.197)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.898\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.981 (0.806,1.196)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.852\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (0.017-0.070)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e272 (26.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1038 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.059 (0.875,1.282)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.036 (0.852,1.259)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.726\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT3(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.099\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~3.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e302 (29.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1088 (26.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (3.9-4.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e278 (26.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1120 (26.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.894 (0.745,1.074)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.231\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.920 (0.763,1.110)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.383\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (4.3-4.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e221 (21.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1016 (24.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.784 (0.646,0.951)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.013\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.800 (0.656,0.975)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.027\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (4.7-14.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e234 (22.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e970 (23.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.869 (0.718,1.052)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.151\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.895 (0.735,1.089)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.269\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT4(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.783\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e255 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1075 (25.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (7.8-8.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e267 (25.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1031 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.092 (0.901,1.322)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.370\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.157 (0.951,1.407)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.146\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (9.0-11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e267 (25.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1061 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.061 (0.876,1.285)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.545\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.148 (0.943,1.396)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.169\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (11.1-37.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e246 (23.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1027 (24.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.010 (0.831,1.227)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.922\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.097 (0.898,1.339)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.364\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT3/FT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~0.359)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e260 (25.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1047 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (0.360-0.479)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e256 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1051 (25.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.981 (0.809,1.189)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.844\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.971 (0.797,1.183)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.769\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (0.480-0.572)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e283 (27.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1025 (24.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.112 (0.920,1.343)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.272\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.081 (0.890,1.312)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.432\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (0.573-1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e236 (22.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1071 (25.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.887 (0.730,1.079)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.231\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.834 (0.683,1.020)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.077\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTSH(mIU/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.047\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~1.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e288 (27.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1024 (24.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e1.211 (1.001,1.465)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.049\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e1.288 (1.059,1.565)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.011\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (1.28-1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e234 (22.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1074 (25.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.938 (0.769,1.143)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.526\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.986 (0.805,1.208)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.891\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (1.82-2.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e267 (25.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1037 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.297 (0.914,1.345)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.297\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.12 (0.919,1.365)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.263\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (2.53-22.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e246 (23.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1059 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTPOAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.623\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~0.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e304 (29.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1166 (27.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (0.5-1.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e220 (21.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e950 (22.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.888 (0.732,1.078)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.230\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.848 (0.696,1.034)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.103\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (1.4-12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e249 (24.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1041 (24.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.917 (0.761,1.106)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.367\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.876 (0.723,1.061)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.174\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (12.1-994)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e262 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1037 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.969 (0.805,1.166)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.739\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.925 (0.765,1.118)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.419\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTg(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.268\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 1 (~5.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e250 (24.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1072 (25.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 2 (6.0-10.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e244 (23.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1049 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.997 (0.820,1.213)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.979\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.979 (0.802,1.196)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.838\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 3 (10.6-17.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e262 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1059 (25.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.061 (0.875,1.287)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.548\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.033 (0.848,1.258)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.748\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQuartile 4 (18.0-298.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e279 (26.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1014 (24.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.180 (0.975,1.428)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.090\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.101 (0.905,1.339)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.335\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: OR = Odds Ratio, aOR = adjusted Odds(adjusted Age, Family history of diabetes, Gestational hypertension and Family history of hypertension).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 3: GDM varies with levels with other thyroid function in patients with different levels of TPOAb\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eTGAb- (n=3275)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP for\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003etrend\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\"\u003e\n \u003cp\u003e\u003cstrong\u003eTGAb+ (n=1954)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP for\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003etrend\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eGDM/Control\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eaOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eGDM/Control\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eaOR (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTT3(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.007\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.456\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~1.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e256/878 (34.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e54/178 (11.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (1.05-1.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e160/802 (22.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.684(0.549,0.852)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.719(0.574,0.899)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.004\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e69/249 (16.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.913(0.609,1.369)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.661\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.940(0.620,1.426)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.773\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (1.21-1.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e150/629 (24.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.818(0.652,1.025)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.081\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.847(0.671,1.070)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.164\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e99/402 (25.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.812(0.558,1182)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.277\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.882(0.600,1.298)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.524\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (1.40-3.68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e75/325 (23.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.791(0.594,1.055)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.110\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.833(0.619,0.120)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.226\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e172/731 (46.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.776(0.548,1.097)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.151\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.805(0.563,1.151)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;TT4(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.823\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.160\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~75.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e173/666 (25.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e105/367 (24.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (76-86.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e163/672 (25.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.934 (0.735,1.186)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.575\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.976 (0.764,1.246)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.844\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e91/391 (24.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.813 (0.594,1.114)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.199\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.874 (0.633,1.207)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.414\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (86.9-98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e156/672 (25.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.894 (0.702,1.138)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.362\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.940 (0.734,1.203)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.621\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e105/367 (24.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.000 (0.736,1.359)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.122 (0.818,1.538)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.476\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (98.1-235.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e149/624 (23.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.919 (0.720,1.174)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.990 (0.770,1.273)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.937\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e93/435 (27.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.747 (0.547,1.021)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.747\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.907 (0.658,1.251)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTT3/TT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.944\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.548\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~0.012)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e220/872 (33.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39/175 (11.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (0.013-0.014)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e179/756 (28.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.938(0.753,1.169)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.572\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.966 (0.771,1.210)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.763\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e69/305 (19.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.015 (0.657,1.567)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.946\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.023(0.656,1.595)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.919\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (0.015-0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e140/587 (22.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.945 (0.746,1.197)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.641\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.950 (0.746,1.210)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.677\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e116/461 (29.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.129 (0.755,1.688)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.059(0.701,1.601)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.784\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (0.017-0.070)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e102/419 (15.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.965 (0.742,1.254)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.789\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.979 (0.749,1.281)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.880\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e170/619 (40.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.232 (0.837,1.813)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.289\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.104 (0.742,1.641)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.625\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT3(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.307\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.194\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~3.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e111/391 (15.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e191/697 (45.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (3.9-4.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e171/671 (25.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.898 (0.686,1.175)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.432\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.943 (0.715,1.244)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.679\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e107/449 (28.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.870 (0.667,1.133)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.301\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.895 (0.682,1.175)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.424\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (4.3-4.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e166/740 (27.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.790 (0.603,1.035)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.087\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.829 (0.629,1.093)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.184\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e55/276 (16.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.727 (0.522,1.012)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.059\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.727 (0.518,1.021)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.066\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (4.7-14.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e193/832 (31.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.817 (0.628,1.063)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.857(0.654,1.122)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e41/138 (9.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.084 (0.739,1.591)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.680\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.157 (0.777,1.723)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.472\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT4(pmol/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.771\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.365\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e239/997 (37.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e16/78 (4.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (7.8-8.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e236/921 (35.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.069(0.874,1.307)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.516\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.131 (0.921,1.390)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.241\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e31/110 (7.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.374 (0.703,2.683)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.352\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.424 (0.717,2.829)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (9.0-11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e150/656 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.954(0.760,1.197)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.683\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.058 (0.838,1.335)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.635\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e117/405 (26.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.408 (0.792,2.505)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.244\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.503 (0.832,2.714)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.177\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (11.1-37.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e16/60 (2.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.112(0.630,1.966)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.714\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.198 (0.665,2.156)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.548\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e230/967 (61.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.160 (0.664,2.023)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.602\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.302 (0.735,2.307)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.365\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eFT3/FT4\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.089\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.885\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~0.359)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e13/49 (1.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.210 (0.645,2.269)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.533\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.161 (0.608,2.217)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.608\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e247/998 (63.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.059 (0.622,1.802)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.833\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.136 (0.657,1.965)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.649\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (0.360-0.479)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e150/661 (24.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.035 (0.822,1.302)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.771\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.132 (0.894,1.434)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.302\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e106/390 (25.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.163 (0.667,2.028)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.595\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.161 (0.655,2.058)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.609\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (0.480-0.572)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e260/930 (36.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e1.275 (1.043,1.558)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.018\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003e1.323 (1.077,1.625)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.008\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e23/95 (6.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.036(0.521,2.057)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.920\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.050 (0.520,2.123)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.891\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (0.573-1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e218/994 (37.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e18/77 (4.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTSH(mIU/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.081\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.501\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~1.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e192/687 (26.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e96/337 (22.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (1.28-1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e150/730 (26.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.735 (0.580,0.932)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.010\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.766 (0.629,0.932)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.008\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e84/344 (21.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.857 (0.617,1.191)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.359\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.831 (0.593,1.165)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.283\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (1.82-2.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e163/643 (24.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.907 (0.717,1.147)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.415\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.870 (0.718,1.054)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.154\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e104/394 (26.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.927 (0.677,1.267)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.633\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.857 (0.621,1.182)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.348\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (2.53-22.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e136/574 (21.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.848 (0.663,1.084)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.188\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.777 (0.639,0.944)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.011\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e110/485 (30.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.796 (0.586,1.082)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.146\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.759 (0.554,1.039)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.086\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTPOAb(IU/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.331\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.175\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~0.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e304/1148 (44.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0/18 (0.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (0.5-1.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e214/920 (34.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.878 (0.723,1.068)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.193\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.833 (0.628,1.018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.075\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6/30 (1.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e323152995 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e305225056 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (1.4-12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e94/427 (15.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.831 (0.643,1.075)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.158\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.765 (0.588,0.995)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.046\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e155/614 (39.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e407888552 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e358188050 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (12.1-994)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29/139 (5.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.788 (0.518,1.199)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.265\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.738 (0.480,1.135)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.166\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e233/898 (57.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e419235233 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e358253152 (0,0)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.998\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTg(ng/ml)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.307\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.860\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ1 (~5.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e152/667 (25.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e98/405 (25.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ2 (6.0-10.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e171/757 (28.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.991 (0.778,1.263)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.943\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e0.940 (0.734,1.204)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.626\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e73/292 (18.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.997 (0.820,1.213)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.950\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.072 (0.758,1.517)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.694\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ3 (10.6-17.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e171/684 (26.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.097 (0.860,1.399)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.455\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.041 (0.812,1.335)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.751\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e91/375 (23.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.061 (0.875,1.287)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.986\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.017 (0.735,1.409)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.917\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eQ4 (18.0-298.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e147/526 (20.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.226 (0.951,1.581)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.115\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\"\u003e\n \u003cp\u003e1.142 (0.881,1.481)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.317\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e132/488 (31.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.180 (0.975,1.428)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.456\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.0531 (0.779,1.422)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.738\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eNote: Q: Quartile, OR = Odds Ratio, aOR = adjusted Odds (adjusted Age, Family history of diabetes, Gestational hypertension and Family history of hypertension ).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 4: Construction of clinical prediction models for GDM using different methods\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eScreening methods\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSelection criteria\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFeature numbers\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eFeature\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStepwise regression\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAge, Height, Family history of diabetes, Gestational hypertension, \u0026lsquo;FT3/FT4\u0026rsquo;, TSH, TGAb\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBest\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003cstrong\u003eubset\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003cstrong\u003eelection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAge, Height, Family history of diabetes, Gestational hypertension, \u0026lsquo;TT3/TT4\u0026rsquo;, FT4, TSH, TGAb\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandom\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ef\u003c/strong\u003e\u003cstrong\u003eorest\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eOOB error rate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAge, Height, TSH, TT4, Tg, TT3,FT4, TPOAb,FT3,TGAb\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003eupport vector machines\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBest of four kernel functions\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAge, Height, Family history of diabetes, Gestational hypertension, \u0026lsquo;FT3/FT4\u0026rsquo;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNotes:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eAIC: Akaike information criterion.\u003c/li\u003e\n \u003cli\u003eOOB: Out of bag.\u003c/li\u003e\n \u003cli\u003eFour kernel functions:linear, polynomial, radial and sigmoid.\u003cstrong\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e5:\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Evaluation of different clinical prediction models\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eModels\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAUC (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSensitivity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSpecificity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTPR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eTNR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eC-index(95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eX-squared\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDF\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eP-value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eStepwise regression\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.647 (0.6284,0.6656)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.540\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.660\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.540\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.660\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.647 (0.6288,0.6660)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2.043\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.360\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eBest\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003cstrong\u003eubset\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003cstrong\u003eelection\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.646 (0.6276,0.6647)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.691\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.211\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.529\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.691\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.646 (0.6276,0.6647)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1.832\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.400\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandom\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003ef\u003c/strong\u003e\u003cstrong\u003eorest\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.569 (0.5494,0.5884)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.632\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.632\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.632 (0.6139,0.6510)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.949\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.622\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eS\u003c/strong\u003e\u003cstrong\u003eupport vector machines\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e0.824\u003c/strong\u003e (0.8137,0.8345)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.132\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.995\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.867\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.823\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.645 (0.6263,0.6430)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.926\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.629\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eAUC: Area Under Curve\u003c/li\u003e\n \u003cli\u003eTPR: True Positive Rate, TNR: True Negative Rate\u003c/li\u003e\n \u003cli\u003eC-index: Concordance Index, DF : Degree of Freedom\u003c/li\u003e\n \u003cli\u003eP-value：Hosmer and Lemeshow goodness of fit (GOF) test\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Gestation, Diabetes mellitus, Thyroid, Hormones, Machine learning","lastPublishedDoi":"10.21203/rs.3.rs-7192354/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7192354/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e Our objective is to explore the relationship between gestational diabetes mellitus (GDM) and thyroid hormones, and construct a clinical prediction model based on the clinical features of GDM and thyroid parameters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods \u003c/strong\u003eA population-based retrospective cohort study, including 1,035 GDM patients and 4,194 healthy control. Statistical tests were conducted to evaluate the associations between primary risk factors, including age, family history of diabetes, gestational hypertension, hypertension family history, thyroglobulin antibody (TGAb), and thyroid hormones (TT3, TSH, FT3, TPOAb), with GDM risk.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults \u003c/strong\u003eIn this study, age, family history of diabetes, gestational hypertension, hypertension family history, and TGAb concentration were identified as primary risk factors. The first four risk factors showed a positive associated with GDM, while height and TGAb concentration were significantly negatively correlated with GDM risk. Additionally, lower levels of total triiodothyronine (TT3) were associated with an increased risk of GDM in all patients, while consistently lower levels of thyroid-stimulating hormone (TSH) also heightened GDM risk. In the TGAb-negative group, higher levels of TT3 and TSH were linked to reduced risk of GDM, whereas lower levels of free triiodothyronine (FT3) were associated with an increased risk. In the TGAb-positive group, thyroid peroxidase antibody (TPOAb) had a strong positive association with GDM.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions \u003c/strong\u003eThyroid hormones play a crucial role in pregnancy and may counteract insulin, affecting blood glucose balance. Therefore, changes in thyroid parameters should be appropriately considered in the prevention and screening of GDM.\u003c/p\u003e","manuscriptTitle":"Clinical prediction model for gestational diabetes mellitus utilizing thyroid function indicators: a retrospective cohort study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-18 09:05:42","doi":"10.21203/rs.3.rs-7192354/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"61e4e3bb-07b3-4ac8-91f3-05977ef13641","owner":[],"postedDate":"August 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-01T13:12:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-18 09:05:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7192354","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7192354","identity":"rs-7192354","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}