Association of Mode of Delivery and Birth Order with Hyperinsulinemia in Neonatal Cord Blood

Association of Mode of Delivery and Birth Order with Hyperinsulinemia in Neonatal Cord Blood | 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 Association of Mode of Delivery and Birth Order with Hyperinsulinemia in Neonatal Cord Blood Gábor Firneisz, Botond A. Nemes, László Németh, Ákos Nádasdi, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7676142/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 Importance: Cesarean delivery (CD) is associated with higher risk of obesity development from childhood and with type 2 diabetes mellitus (T2DM) later in life. Higher neonatal cord C-peptide (CP) levels were reported to be associated with increased risk of offspring metabolic disorders in childhood as well. Objective: To assess the association between mode of delivery as the primary exposure and 27 additional maternal/neonatal clinical factors and the risk of neonatal hyperinsulinemia. Design: Retrospective data analysis. Setting: Multinational, multi-ancestry observational population-based Hyperglycemia Advanced Pregnancy Outcome (HAPO) cohort study gene-environment interaction (GEI) substudy. Participants: A total of 4967 mother-neonate pairs from 4 major ancestry groups (Afro-Caribbean, European, Hispanic, Thai) were assessed. Exposure: Data on mode of delivery and 27 additional (18 maternal + 9 neonatal) variables were utilized from the NIH dbGaP database (accession phs000096.v4.p1). Main outcome and measures: The primary outcome was neonatal hyperinsulinemia herein defined as cord serum CP level of ancestry specific > 90th percentile. χ 2 -test was used to assess the difference between CD and vaginal delivery groups and standardized multivariate logistic regression with Akaike information criterion-based model selection and interaction analyses were used to model the primary outcome. Missing data were addressed using Bayesian joint modeling approach. Results: Out of the total 4967 pregnancies assessed CD was performed in 1154 mothers (23.2%). Cord blood hyperinsulinemia occurred in 447 cases of which 178 in the CD (15.4%) and 269 in vaginal delivery (7.1%) group (OR crude, 2.4; 95% CI, 1.96 to 2.94). We identified novel factors individually associated with neonatal hyperinsulinemia: CD (OR in firstborns: 4.25; 95% CI, 2.66-6.78, OR in non-firstborns: 4.25; 95% CI, 2.59-6.99) and non-firstborn vaginal delivery (OR, 1.95; 95% CI, 1.34-2.88) when compared to vaginally delivered firstborns. Sensitivity analyses confirmed the stability of these results and subgroup analyses revealed the consistency of the two novel associations across different ancestries. Conclusion : To our best knowledge we first report that neonatal hyperinsulinemia risk is more than quadrupled in neonates born via cesarean delivery independently of birth order when compared to vaginally delivered firstborns. Non-firstborns are also at increased risk. Translational Medicine Maternal & Fetal Medicine Cesarean delivery Mode of delivery Birth order Neonatal hyperinsulinemia HAPO study Cord blood C-peptide Figures Figure 1 Figure 2 Figure 3 Introduction Obesity and related diseases are posing a serious global public health challenge in both child and adulthood. 1 , 2 A link between cesarean delivery (CD) and childhood obesity has been reported as early as 2 years of age with associations persisting in toddlers at 3 and in preschool children at 3–6 years of age. 3 – 6 A recent meta-analysis also concluded that children born by CD were more commonly developed obesity compared to those delivered vaginally. 7 The analysis of a large prospective cohort consistently reported that birth by CD was associated with offspring obesity in individuals aged from 9–28 years even after accounting for confounding factors, including maternal body mass index (BMI). 8 In the HAPO FUS study involving 4155 children, cord blood C-peptide (CP) levels obtained at birth, examined for their potential association with future metabolic disorders, were inversely correlated with the later assessed (mean age, 11.4 years) Matsuda index. 9 However the effect of CD on neonatal cord blood CP levels has never been directly assessed in any larger study. A small study (n < 50) reported that large amounts of dextrose infusion administered to pregnant women before CD increased the venous cord blood glucose concentrations and subsequently elevated cord blood insulin levels that resulted in increased risk of neonatal hypoglycemia. 10 Consistently, the association between CD and cord CP levels was reported in very few and small studies prior. 10 We hypothesized that neonatal hyperinsulinemia herein defined by the umbilical cord CP levels could be associated with mode of delivery as the primary exposure and we aimed to assess this association in a database derived from a large, multicenter, multi-ancestry clinical study. Methods Study Design We retrospectively analyzed data obtained from the HAPO multinational, multiethnic and multi-ancestry, population-based observational study 11 , maternal glycemia and birthweight gene-environment interaction (GEI) substudy dataset. 12 The enrollment and data collection period of the original HAPO study lasted from 2000 to 2006 and it included pregnant women of 4 major ancestry groups (Hispanic, Afro-Caribbean, South-Asian and European) from 9 countries and 15 study centers. 11 Data Sources We accessed the GENEVA GWA mapping Maternal Metabolism-Birth Weight Interactions dataset via the NCBI dbGaP (the Database of Genotypes and Phenotypes) system (dbGaP accession phs000096.v4.p1) that contained data on a total of 4994 mother-neonate pairs from the original HAPO cohort study. 11,12 Study Cohort After initial restriction to mandatory data in our analysis (cord CP level and ancestry was required for the primary outcome and mode of delivery as the main exposure) the dataset contained data on 4967 of 4994 mother-neonate pairs. The complete case analysis consisted of 3492 mother-neonate pairs where all variables employed in final model were available. Exposure The primary exposure was the mode of delivery (cesarean delivery or vaginal delivery). We considered additional variables potentially associated with cord CP levels including maternal variables: (1) age at oral glucose tolerance test (OGTT), (2) pre-pregnancy weight (kg), (3) family history of diabetes, (4) family history of hypertension, (5) ethnicity, (6-7-8) fasting, 60min, 120min plasma glucose (PG) values at OGTT (mmol/L), (9) HbA1c at OGTT (%), (10) fasting CP levels at OGTT (ug/L), (11) weight at OGTT (kg), (12) hypertension (any type), (13) proteinuria highest urine dipstick result >=3+, (14) height (cm), (15) prior pregnancy delivered at >= 20 weeks, (16) drinker status, (17) smoker status, (18) mean systolic blood pressure at OGTT (mmHg) and also neonatal variables measured at delivery: (19) gestational age, (20) birth weight (kg), (21) mean flank skinfold (mm), (22) head circumference (cm), (23) length (cm), (24) subscapular skinfold (mm), (25) triceps skinfold (mm), (26) sex, (27) venous cord blood PG concentration (mmol/L). Outcome The primary outcome of the analysis was neonatal hyperinsulinemia herein defined as cord serum CP level of > 90 th percentile. Ancestry specific CP 90 th percentile cutoff values were 1.6 ng/mL in the Afro-Caribbean and South-Asian populations and 1.7 ng/mL in those with European and Hispanic origins, respectively. Covariates Clinical variables included gestational diabetes mellitus (GDM), defined from maternal plasma glucose (PG) values during the 24–28-week OGTT using International Association of the Diabetes and Pregnancy Study Groups criteria, as adopted by WHO in 2013. 13,14 GDM was used as a categorical variable; individual OGTT PG values were not included simultaneously. HOMA2-IR was calculated from fasting maternal PG and CP values using the HOMA2 calculator and modeled as a continuous predictor; corresponding fasting maternal PG and CP values were excluded to avoid redundancy. 15 Although HOMA2-IR determination is not routinely recommended for use during pregnancy, it is still an acceptable estimate of insulin resistance under study settings at OGTT. 16 An alternative model used fasting, 1-, and 2-hour PG values as continuous variables, excluding GDM and HOMA2-IR. Additional covariates were maternal pre-pregnancy BMI, gestational weight gain to OGTT, and preeclampsia (binary). When derived predictors were included, their component variables were excluded to reduce collinearity and overfitting. Statistical Analyses χ 2 -test assessed differences in the primary endpoint between CD and vaginal delivery groups. Neonatal hyperinsulinemia was modeled using standardized multivariate logistic regression. A backward variable selection procedure was iteratively applied, whereby in each cycle the variable whose removal yielded the greatest improvement in the model’s Akaike Information Criterion (AIC) was excluded. This process was performed using a complete-case dataset, comprising only individuals with no missing values for any of the candidate variables. Once no further improvement in AIC could be reached by variable removal, the optimal model was established for this reduced subset of the data. To maximize the sample size without compromising the integrity of the variable selection process, we implemented an iterative approach as presented in Figure 1 . Initially, the dataset was restricted to complete cases. AIC-based variable selection was then performed, and the variables retained in the resulting optimal model were identified. The dataset was subsequently redefined to include complete cases only for this reduced set of variables. This process was repeated iteratively until the set of selected variables remained stable across successive iterations. At each step, the sample size was preserved or increased relative to the previous iteration, ensuring convergence toward an optimal model that incorporated the maximum number of participants possible under the complete-case constraint. The algorithm converged after three iterations in our analysis. Details of the iterative model selection steps, including sample sizes and retained variables at each stage, are provided in eTable 1 and eTable 2 in Supplement 1. Model Refinement and Interaction Terms In the final modeling step, interaction terms between the most significant variables were incorporated, followed by an additional round of AIC-based variable selection. The inclusion of interaction terms was selectively restricted; specifically, interactions between ethnicity and maternal pre-pregnancy BMI were constrained due to established differences in healthy BMI ranges across ethnic subgroups, which could confound clinical interpretation. 17,18 Bayesian Joint Modeling for Handling Missing Data To address the missing data—particularly for maternal pre-pregnancy BMI and number of prior pregnancies—a Bayesian joint modeling approach was applied using the JointAI package in R. The logistic regression model, containing the same variables as the complete-case analysis, was fitted with simultaneous imputation of missing covariates via Markov Chain Monte Carlo (MCMC) sampling (yielding 1,060 samples across three independent chains). This approach accounts for uncertainty due to missing data and provides posterior estimates with corresponding 95% credible intervals and Bayesian tail probabilities. The mean posterior estimates were compared with those from the complete-case analysis to evaluate the sensitivity of the results to missing data. Sensitivity Analyses Several additional sensitivity analyses were also conducted. First, the GDM and HOMA-R variables were replaced with the exact maternal PG and CP values in the final model to examine the sensitivity of this type of grouping. In an additional analysis, birth order was excluded, due to that it was missing in a large proportion of mothers with Thai ancestry. Additionally, the final complete-case model was stratified by ancestry to assess the consistency of effects across subgroups. Finally, the complete case model was reconsidered using the mode of delivery and birth order predictors combined into a single, multilevel variable. No further variable selection was applied within these stratified models; and the previously determined optimal model was re-fitted within each ethnic subgroup. This approach allowed the identification of both ancestry-specific and generally consistent associations. Results Demographic and Clinical Characteristics Among the 4967 women included in the analysis after the initial restriction to mandatory data, 447 (9%) delivered neonates who presented with cord blood hyperinsulinemia. Of the 4,967 pregnancies assessed, CD was performed in 1,154 (23.2%) cases. Additional demographic and clinical characteristics of the complete-case mother–neonate pairs are presented in the Table. Comparative maternal and neonatal characteristics stratified by the mode of delivery are provided in eTable3 in Supplement 1. Neonatal Hyperinsulinemia and Cesarean Delivery Cord blood hyperinsulinemia occurred more frequently among neonates delivered by cesarean delivery: 178 cases (15.4%) compared with 269 cases (7.1%) among those born via vaginal delivery (unadjusted odds ratio [OR], 2.4; 95% CI, 1.96–2.94). In a multivariable model based on the analysis of 3,492 complete-case mother–neonate pairs incorporating interaction terms CD was in significant interactions with the first-time delivery (interaction β = 0.5681; 95% credible interval, 0.0556–1.0834) and with cord blood PG levels (interaction β = 0.2113; 95% credible interval, 0.0088–0.4205). However, most examined interaction terms were eliminated during the iterative AIC-based variable selection process. The interaction terms between ancestries and maternal BMI were retained in the model due to established differences in healthy BMI ranges across ethnic subgroups, marked BMI differences across ancestry groups, although these did not exert a statistically significant effect. Following Bayesian joint modeling with imputation, 4,967 mother–neonate pairs were analyzed. Cesarean delivery remained significantly associated with an increased risk of neonatal hyperinsulinemia (OR, 2.66; 95% CI, 1.90–3.77) in the final model ( Figure 2 ). Additional factors associated with neonatal hyperinsulinemia All results refer to models adjusted for known and novel covariates associated with neonatal hyperinsulinemia ( Figure 2 ). Maternal factors including GDM (β = 0.4192; 95% credible interval, 0.1379 – 0.6892 OR: 1.5207 1.1479 – 1.9921), HbA1c level at OGTT (β = 0.1764, 95% credible interval 0.0524 – 0.3017), weight gain till the OGTT (β = 0.1270, 95% credible interval 0.0099 – 0.2462) HOMA2-IR at OGTT (β = 0.2941, 95% credible interval 01868 – 0.4041), and African ancestry (β = 0.6794, 95% credible interval 0.3399 – 1.0195) were associated with higher, while 1st time delivery (β = -0.5442, 95% credible interval -0.8749 – -0.2252) was associated with lower risk of neonatal cord blood hyperinsulinemia in the final model. In contrast, maternal age and pre-pregnancy BMI that were constrained in the model due to their clinical significance and had no effect ( Figure 2 ). Neonatal factors, such as gestational age at delivery (β = -0.1950, 95% credible interval -0.3179 – -0.0733), birthweight (β = 0.7423; 95% credible interval, 0.5749 – 0.9050), venous cord blood PG (β = 0.6549; 95% credible interval, 0.5210 – 0.7887), female sex (β = 0.8862; 95% credible interval, 0.6649 – 1.1110), were also significantly associated with the outcome. ( Figure 2 ). Model Accuracy, Sensitivity and Subgroup Analyses We assessed the performance of both the complete case analysis models using the available data sets. The performance was graphically displayed on a receiver operating characteristics (ROC) curve. ( Figure 3 ). The area under the curve (AUC) value was 0.822, respectively, indicating effective model discrimination. Although validation using an independent test set would be ideal, the primary objective of this study was an association analysis; therefore, these AUC values are considered adequate assessments. In contrast, ROC AUC could not be calculated for the Bayesian model because its probabilistic outputs and handling of missing data prevent generation of single-point predictions required for ROC analysis. To test the stability of our findings, we conducted multiple sensitivity analyses. First, in the complete case model, we substituted the GDM and HOMA-IR variables with measured maternal PG and CP levels at OGTT. Cesarean delivery and birth order continued to exhibit similar associations (eFigure 1 in Supplement 1). Second, we excluded birth order from the final complete case model due to substantial missing data, particularly within the Thai subpopulation (missing birth order n = 524 out of 1241). The associations of CD and other covariates remained largely unchanged, suggesting that the missing data were likely missing at random and did not introduce any significant bias (eFigure 2 in Supplement 1). Finally, we stratified the cohort by ethnicity (European, Afro-Caribbean, Hispanic, and Thai) and fitted the final model separately within each subpopulation. While the magnitude of effects varied across groups, the direction of associations for CD and birth order remained consistent, despite smaller sample sizes. These results are presented in eFigure 3, eFigure 4, eFigure 5, and eFigure 6 in Supplement 1. To explore the interaction between mode of delivery and birth order we further modified the final model and employed these predictors combined into single, multilevel variables as presented in eFigure 7 in Supplement 1. Discussion In this retrospective analysis of a multinational, multiethnic, and multi-ancestry population-based observational substudy of the HAPO study, we found that neonates delivered by CD were at higher risk of hyperinsulinemia in cord blood. This association remained significant after adjustment for a broad set of predictors and interactions. Sensitivity analyses confirmed the robustness of this finding, and subgroup analyses demonstrated its consistency across different ancestral groups. Previous studies have reported that children born via planned CD have an increased risk of obesity in preadolescence, with evidence that these metabolic associations persist into adulthood. 19 Women born by CD were found to have a higher risk of type 2 diabetes than those born vaginally (adjusted hazard ratio, 1.46; BMI-adjusted risk ratio, 1.34). 20 Proposed mechanisms focus on altered gut microbiota, as infants born by CD may acquire less diverse microbiomes. 21 , 22 The HAPO Follow-Up Study reported an inverse association between neonatal cord C-peptide (CP) levels and the Matsuda index of whole-body insulin sensitivity in 4,155 children (mean age, 11.4 years), however did not adjust for delivery mode. 9 In our analysis, we also assessed umbilical venous cord plasma glucose (PG), which correlates strongly with maternal glucose and reflects maternal glycemia at birth. 23 , 24 Maternal dextrose infusion before CD can raise fetal insulin. 10 In contrast, we found lower cord PG levels in CD compared vaginal births, suggesting that elevated hyperinsulinemia risk is not due to increased fetal glucose exposure. As expected, our final model demonstrated a significant interaction between CD and cord PG levels, alongside independent significant associations of both variables with the primary outcome of neonatal hyperinsulinemia. We also identified a novel association between first delivery and lower hyperinsulinemia risk, which contrasts with smaller studies reporting reduced insulin sensitivity in first-borns. 23 However, this study had a substantially smaller sample size (n = 87) 23 , compared with our analysis that employed the data of 3492 and 4,967 mother–neonate pairs in the complete case analysis and the final model, respectively. We found that this effect only in vaginal births, suggesting potential interactions with labor-related factors. Taken together, these findings may provide a mechanistic link between CD and later obesity, insulin resistance, and type 2 diabetes, and highlight the need for further investigation of birth order effects. The additional predictors independently associated with neonatal hyperinsulinemia in our final model largely reflected previously reported factors or well-established pathophysiological conditions, including maternal glycemic status and insulin-resistance (GDM, HbA1c, and HOMA2-IR at OGTT), maternal weight gain in pregnancy till OGTT, and neonatal characteristics such as gestational age, birthweight, and sex. 24 – 32 The World Health Organization (WHO) recommends that population-level CD rates ideally range between 10% and 15%. 33 Contrary to these guidelines, CD rates worldwide have tripled to over 21% in between 1999 and 2023. 34 This trend, in combination with the rising prevalence of childhood obesity and associated health conditions, underscores the potential clinical significance of our findings. 35 , 36 . An integrated interpretation of our findings suggests that the increased risk of neonatal hyperinsulinemia with cesarean delivery and the decreased risk with first delivery may relate to the absence or prolongation of labor, respectively, with natural labor potentially protective. Furthermore, we report here that age of gestation at delivery is inversely associated with the risk of neonatal hyperinsulinemia that is consistent with a meta-analysis which reported that neonates born by preterm delivery had a higher risk of childhood obesity. 37 It could be argued that the observed associations may be influenced by unmeasured confounders. However, the presence of an unmeasured confounder during pregnancy that would plausibly affect birth order can be considered unlikely, even from a theoretical standpoint. With regard to the mode of delivery, the 27 variables assessed likely capture the main mechanisms underlying the currently proposed pathophysiology. 38 Although the possibility of residual confounding cannot be entirely ruled out, the likelihood of a meaningful unmeasured confounder is considered low. Strengths First, it is a retrospective analysis of large, population-based cohort study encompassing data from 4967 mother-neonate pairs in the final model. Such a sample size enhances the statistical power and reliability of clinically relevant predictors, while minimizing recall bias and allowing for the assessment of multiple exposures in relation to neonatal hyperinsulinemia as the primary outcome. Furthermore, the multi-ancestry design of both the HAPO study and the specific data subset analyzed enhances the generalizability and applicability of our findings across diverse populations. Limitations Our study did not explore the biological mechanisms underlying the increased risk of neonatal hyperinsulinemia associated with cesarean delivery, nor the protective effect observed with primiparity. As such, our interpretations of these novel associations remain speculative and require further investigation through mechanistic and prospective studies. Additionally, despite rigorous statistical procedures and adjustment for confounders, the possibility of residual, unmeasured confounding cannot be excluded, limiting the ability to infer definitive causal relationships. Conclusions We observed an association between cesarean delivery and increased risk of neonatal hyperinsulinemia defined as the uppermost decimal of the umbilical cord blood C-peptide levels. To our knowledge, we also first report a protective effect of first delivery on neonatal hyperinsulinemia. The associations between cesarean delivery, first time delivery and neonatal hyperinsulinemia both remained significant in analyses using rigorous, multi-step adjustments for an extensive set of the most important maternal and neonatal covariates that exceeds the available data in routine pregnancy care. The multi-ancestry design of the HAPO study provided the generalizability of our findings across diverse populations. This suggests that this association may indicate a true effect of CD, and a potential benefit might be reached on neonatal hyperinsulinemia when preferring a vaginal delivery in the absence of medical or obstetric indication of CD. Additional evidence from large, prospective cohort studies is needed to determine whether CD associated neonatal cord blood hyperinsulinemia might contribute to increased risk of obesity and related metabolic disorders later in life. Declarations Acknowledgements We are grateful to Healthware Consulting Ltd., Budapest, Hungary for helping us to select and adapt the appropriate statistical methods. References Collaborators, G.B.D.R.F. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 403 , 2162-2203 (2024). Okunogbe, A. , et al. Economic impacts of overweight and obesity: current and future estimates for 161 countries. BMJ Glob Health 7 (2022). Pei, Z. , et al. Cesarean delivery and risk of childhood obesity. J Pediatr 164 , 1068-1073 e1062 (2014). Huh, S.Y. , et al. Delivery by caesarean section and risk of obesity in preschool age children: a prospective cohort study. Arch Dis Child 97 , 610-616 (2012). Rutayisire, E. , et al. Cesarean section may increase the risk of both overweight and obesity in preschool children. BMC Pregnancy Childbirth 16 , 338 (2016). Terashita, S. , et al. Caesarean section and childhood obesity at age 3 years derived from the Japan Environment and Children's Study. Sci Rep 13 , 6535 (2023). Slabuszewska-Jozwiak, A., Szymanski, J.K., Ciebiera, M., Sarecka-Hujar, B. & Jakiel, G. Pediatrics Consequences of Caesarean Section-A Systematic Review and Meta-Analysis. Int J Environ Res Public Health 17 (2020). Yuan, C. , et al. Association Between Cesarean Birth and Risk of Obesity in Offspring in Childhood, Adolescence, and Early Adulthood. JAMA Pediatr 170 , e162385 (2016). Bianco, M.E. , et al. Hyperglycemia and Adverse Pregnancy Outcome Follow-Up Study: newborn anthropometrics and childhood glucose metabolism. Diabetologia 64 , 561-570 (2021). Kenepp, N.B. , et al. Fetal and neonatal hazards of maternal hydration with 5% dextrose before caesarean section. Lancet 1 , 1150-1152 (1982). HapoStudyCooperativeResearchGroup , et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 358 , 1991-2002 (2008). Geoff Hayes; Boyd E. Metzger, A.R.D., Lynn P. Lowe, Margrit Urbanek, Nancy Cox, Dan Nicolae, William L. Lowe, Jr GWA mapping: Maternal Metabolism-Birth Weight Interactions in GENEVA GWA mapping: Maternal Metabolism-Birth Weight Interactions. (https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000096.v4.p1, 2012). InternationalAssociationofDiabetes. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 33 , 676-682 (2010). WorldHealthOrganization. Diagnostic Criteria and Classification of Hyperglycaemia First Detected in Pregnancy. Online document WHO/NMH/MND/13.2 , 1-62 (2013). Wallace, T.M., Levy, J.C. & Matthews, D.R. Use and abuse of HOMA modeling. Diabetes Care 27 , 1487-1495 (2004). Jones, D.L., Kusinski, L.C., Gillies, C. & Meek, C.L. A critique of measurement of defective insulin secretion and insulin sensitivity as a precision approach to gestational diabetes. Diabetologia 68 , 752-765 (2025). Consultation, W.H.O.E. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 363 , 157-163 (2004). Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser 894 , i-xii, 1-253 (2000). Sitarik, A.R. , et al. Association between cesarean delivery types and obesity in preadolescence. Int J Obes (Lond) 44 , 2023-2034 (2020). Chavarro, J.E. , et al. Association of Birth by Cesarean Delivery With Obesity and Type 2 Diabetes Among Adult Women. JAMA Netw Open 3 , e202605 (2020). Shao, Y. , et al. Stunted microbiota and opportunistic pathogen colonization in caesarean-section birth. Nature 574 , 117-121 (2019). Martinez, K.A., 2nd , et al. Increased weight gain by C-section: Functional significance of the primordial microbiome. Sci Adv 3 , eaao1874 (2017). Ayyavoo, A., Savage, T., Derraik, J.G., Hofman, P.L. & Cutfield, W.S. First-born children have reduced insulin sensitivity and higher daytime blood pressure compared to later-born children. J Clin Endocrinol Metab 98 , 1248-1253 (2013). Niknam, A. , et al. Umbilical cord blood concentration of connecting peptide (C-peptide) and pregnancy outcomes. BMC Pregnancy Childbirth 22 , 764 (2022). Yin, B. , et al. Combining HbA1c and insulin resistance to assess the risk of gestational diabetes mellitus: A prospective cohort study. Diabetes Res Clin Pract 199 , 110673 (2023). Rifas-Shiman, S.L. , et al. First and second trimester gestational weight gains are most strongly associated with cord blood levels of hormones at delivery important for glycemic control and somatic growth. Metabolism 69 , 112-119 (2017). Ma, R. , et al. Interpregnancy interval and early infant neurodevelopment: the role of maternal-fetal glucose metabolism. BMC Med 22 , 2 (2024). Garcia-Flores, J. , et al. Weight-related and analytical maternal factors in gestational diabetes to predict birth weight and cord markers of diabetic fetopathy. Gynecol Endocrinol 32 , 548-552 (2016). Regnault, N. , et al. Higher cord C-peptide concentrations are associated with slower growth rate in the 1st year of life in girls but not in boys. Diabetes 60 , 2152-2159 (2011). Kuchay, M.S., Kudyar, R.P., Gupta, A., Pandita, K.K. & Ganie, M.A. Gender differences in insulin and C-peptide concentrations at birth using cord blood collection. Arch Endocrinol Metab 60 , 264-266 (2016). Lee, I.L. , et al. Cord blood metabolic markers are strong mediators of the effect of maternal adiposity on fetal growth in pregnancies across the glucose tolerance spectrum: the PANDORA study. Diabetologia 63 , 497-507 (2020). Shields, B.M. , et al. Measurement of cord insulin and insulin-related peptides suggests that girls are more insulin resistant than boys at birth. Diabetes Care 30 , 2661-2666 (2007). Betran, A.P., Torloni, M.R., Zhang, J.J., Gulmezoglu, A.M. & Section, W.H.O.W.G.o.C. WHO Statement on Caesarean Section Rates. BJOG 123 , 667-670 (2016). Angolile, C.M., Max, B.L., Mushemba, J. & Mashauri, H.L. Global increased cesarean section rates and public health implications: A call to action. Health Sci Rep 6 , e1274 (2023). Collaborators, G.B.o.D.A.B. Global, regional, and national prevalence of child and adolescent overweight and obesity, 1990-2021, with forecasts to 2050: a forecasting study for the Global Burden of Disease Study 2021. Lancet 405 , 785-812 (2025). WorldObesityFederation. World Obesity Atlas 2023. Ou-Yang, M.C. , et al. Accelerated weight gain, prematurity, and the risk of childhood obesity: A meta-analysis and systematic review. PLoS One 15 , e0232238 (2020). Josefson, J.L. , et al. Newborn Adiposity and Cord Blood C-Peptide as Mediators of the Maternal Metabolic Environment and Childhood Adiposity. Diabetes Care 44 , 1194-1202 (2021). Table Table. Demographic and clinical characteristics of mother–neonate pairs in the complete-case analysis Maternal Characteristics Complete case (n=3492) Mode of delivery, No. (%) Cesarean delivery 831 (23.8) Vaginal 2661 (76.2) Age, median (IQR), y 29.63 (25-33.4) Pre-pregnancy weight, median (IQR), kg 61.29 (54-72) Height, median (IQR), cm 161.29 (156.2-166.3) Body mass index, median (IQR) a 23.58 (20.9-27.2) Ancestry, No. (%) European 1355 (38.8) Afro - Caribbean 719 (20.6) Thai 618 (17.7) Hispanic 800 (22.9) Plasma glucose at OGTT, median (IQR), mmol/L, [mg/dL] Fasting 4.5 (4.3-4.8), [81 (77.4-86.4)] 1 - hour 7.3 (6.3-8.6), [131.4 (113.4-154.8)] 2 - hour 6.1 (5.3-6.9), [109.8 (95.4-124.2)] HbA1c, median (IQR), % 4.7 (4.5-5) Fasting CP levels at OGTT, median, (IQR), ug/L 1.8 (1.3-2.4) GDM, No. (%) 618 (17.7) HOMA2-IR, median (IQR) 1.25 (0.93-1.7) Weight gain until OGTT, median (IQR), kg 8.97 (6.1-12.1) Hypertension (any type), No. (%) 421 (12.1) Systolic blood pressure at OGTT, median (IQR), mm Hg 106 (99.5-113) 1 st delivery, No. (%) 1480 (42.4) Drinker, No. (%) 312 (8.9) Smoker, No. (%) 191 (5.5) Neonatal Characteristics Complete case (n=3492) Gestational age, median (IQR), weeks 39.71 (38.9-40.6) Birth weight, median (IQR), g 3295(3010-3632.5) Flank skinfold, median (IQR), mm 3.95 (3.4-4.7) Head circumference, median (IQR), cm 34.3 (33.5-35.1) Length, median (IQR), cm 50 (49-51.5) Subscapular skinfold, median (IQR), mm 4.3 (3.7-5.1) Sex, Male No. (%) 1768 (50.6) Venous cord blood glucose, median (IQR), mmol/L [mg/dL] 4,4 (3.8-5.1), [79.2 (68.4-91.8)] Venous cord C-peptide, median (IQR), ug/L 0.9 (0.7-1.2) Hyperinsulinemia, No. (%) 299 (8.6) Additional Declarations The authors declare no competing interests. Supplementary Files Supplementarymaterial.docx Supplemental Online Content 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-7676142","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":518757612,"identity":"d5e8b5f1-e21e-4064-b963-094d3921709b","order_by":0,"name":"Gábor Firneisz","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABJklEQVRIie2QMUvEMBTHnxTSJWc2iVToV0gRTofj8lUaDroF3NzKgXAu/QCVG/0CykFcHAKBu0HRVejSqXNd5IZDbe9EUXJ1FcyP5PH45/14EACH4w/in7WVrXvU3MEmLgHw+ljA5ruSbOK4S9FffauYTwW2Kn6vKuuTlAO5rMrn20dO9u5EKSaDg3Dcm1Ob4vlHUc6MB7RqmqoQF1M5Y2KSYKZ3E5vCPYSC5hUB1f0A6yJmhVRUKIMZ4L59S6OsWIqBLl4a5YHzQt4shXrD4bhDAeZRIFm7Re9cBVKBUBqD3q7sZ8wwRPFplOuRyAs5o+J1hJnBh8c2hcwRXa5STshClbUecjKV13V9P+TheRY9Wb/5A0Tjn5HXNd9C9G8TDofD8V95B6OqWFLpsxJzAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-9123-8131","institution":"Semmelweis University, Institute of Translational Medicine","correspondingAuthor":true,"prefix":"","firstName":"Gábor","middleName":"","lastName":"Firneisz","suffix":""},{"id":518757613,"identity":"0b1d59d4-f409-4fbd-bb2a-48e1d4d5f810","order_by":1,"name":"Botond A. Nemes","email":"","orcid":"https://orcid.org/0009-0000-8607-418X","institution":"Semmelweis University, Institute of Translational Medicine","correspondingAuthor":false,"prefix":"","firstName":"Botond","middleName":"A.","lastName":"Nemes","suffix":""},{"id":518757614,"identity":"f0fec10e-33aa-47de-beef-7bd439bf4260","order_by":2,"name":"László Németh","email":"","orcid":"https://orcid.org/0000-0003-1011-9032","institution":"Semmelweis University, Institute of Translational Medicine","correspondingAuthor":false,"prefix":"","firstName":"László","middleName":"","lastName":"Németh","suffix":""},{"id":518757615,"identity":"5fb35ea3-167e-4f66-a460-40127f576ab2","order_by":3,"name":"Ákos Nádasdi","email":"","orcid":"https://orcid.org/0000-0002-7016-6474","institution":"Semmelweis University, Institute of Translational Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ákos","middleName":"","lastName":"Nádasdi","suffix":""},{"id":518757616,"identity":"8ac951de-6f5a-444d-9cdd-c6710ba64efd","order_by":4,"name":"Zoltán Benyó","email":"","orcid":"https://orcid.org/0000-0001-6015-0359","institution":"Semmelweis University, Institute of Translational Medicine","correspondingAuthor":false,"prefix":"","firstName":"Zoltán","middleName":"","lastName":"Benyó","suffix":""}],"badges":[],"createdAt":"2025-09-22 13:08:38","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7676142/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7676142/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92158091,"identity":"d1346d60-e494-4c1b-925a-5ccb9a008717","added_by":"auto","created_at":"2025-09-25 09:30:43","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":162866,"visible":true,"origin":"","legend":"","description":"","filename":"Maintextfilewithtableandfigurelegendstoupload.docx","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/56c52c5ff2c26273b50dbdb3.docx"},{"id":92159404,"identity":"d9045f1c-c38b-4d60-b07e-98320e57bbb4","added_by":"auto","created_at":"2025-09-25 09:46:42","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7676142.json","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/0d1b736f5ef61fc6509cf504.json"},{"id":92158621,"identity":"088acb9e-90ac-4ae6-8db5-eda8f422327d","added_by":"auto","created_at":"2025-09-25 09:38:43","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84575,"visible":true,"origin":"","legend":"","description":"","filename":"rs76761420enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/435d43b6fd0b804d376e2f40.xml"},{"id":92158092,"identity":"c2be20ae-5487-47da-8db5-e7bc484fe334","added_by":"auto","created_at":"2025-09-25 09:30:43","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":83405,"visible":true,"origin":"","legend":"","description":"","filename":"rs76761420structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/0a4efa2d7dae561b055cb0f3.xml"},{"id":92158093,"identity":"644bad9d-0cc4-4a97-bc0e-83c659a227e9","added_by":"auto","created_at":"2025-09-25 09:30:43","extension":"html","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":89667,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/17ca6410992249a36a2bba1f.html"},{"id":92158090,"identity":"af655ea2-130d-4bd8-ba6a-c19314ec4297","added_by":"auto","created_at":"2025-09-25 09:30:42","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":168646,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlowchart of iterative backward variable selection process based on Akaike Information Criterion\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure1AICciklus.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/9faea0585e590f883f1b73df.jpg"},{"id":92158088,"identity":"3a6ac772-fd0a-47e0-9876-5f1f6576baaa","added_by":"auto","created_at":"2025-09-25 09:30:42","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":816867,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForrest plot of neonatal cord blood hyperinsulinemia predictors in the complete case analysis and in the final model\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eComplete case analysis with generalized linear model (GLM) n = 3 492 (marked as blue). Final model (Bayesian imputation) n = 4\u0026nbsp;967 (marked as red).\u003c/p\u003e\n\u003cp\u003e*Odds ratios are reported with 95% confidence and credibitlity intervals (CIs) according to the model. Standardized effect sizes are indicated for continuous variables. Effects of categorical predictors are indicated against the following corresponding reference categories: European ethnicity, vaginal delivery, non-first time delivery, male neonatal sex. Interaction terms for CD *neonatal cord PG cc; CD*1\u003csup\u003est\u003c/sup\u003e time delivery; maternal BMI at OGTT*ethnicity are also indicate\u003c/p\u003e","description":"","filename":"Figure2forestplotmodels.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/804b9b61b7976ec918ec54f7.jpg"},{"id":92158618,"identity":"fca12033-ef6e-4d09-9301-c864bc54582f","added_by":"auto","created_at":"2025-09-25 09:38:42","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":608916,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eReceiver operating characteristics (ROC) curve for the complete case analysis (GLM)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eArea under curve (AUC) value = 0.822\u003c/p\u003e","description":"","filename":"Figure3ROC.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/fb54d76271e8c7d69ef4e621.jpg"},{"id":92207075,"identity":"e4fbd695-7626-4d2c-96b3-2f338db33a3a","added_by":"auto","created_at":"2025-09-25 19:02:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2570171,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/07a1ce1f-e3d2-4d9c-ab95-787c22e6f9c0.pdf"},{"id":92158620,"identity":"7a8a92e6-982c-466b-a665-6c1a0d3671b4","added_by":"auto","created_at":"2025-09-25 09:38:43","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1969399,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplemental Online Content\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Supplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-7676142/v1/9a2a49c29e0cd96110089634.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eAssociation of Mode of Delivery and Birth Order with Hyperinsulinemia in Neonatal Cord Blood\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eObesity and related diseases are posing a serious global public health challenge in both child and adulthood.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eA link between cesarean delivery (CD) and childhood obesity has been reported as early as 2 years of age with associations persisting in toddlers at 3 and in preschool children at 3\u0026ndash;6 years of age.\u003csup\u003e\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e A recent meta-analysis also concluded that children born by CD were more commonly developed obesity compared to those delivered vaginally.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e The analysis of a large prospective cohort consistently reported that birth by CD was associated with offspring obesity in individuals aged from 9\u0026ndash;28 years even after accounting for confounding factors, including maternal body mass index (BMI).\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIn the HAPO FUS study involving 4155 children, cord blood C-peptide (CP) levels obtained at birth, examined for their potential association with future metabolic disorders, were inversely correlated with the later assessed (mean age, 11.4 years) Matsuda index.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e However the effect of CD on neonatal cord blood CP levels has never been directly assessed in any larger study.\u003c/p\u003e\u003cp\u003eA small study (n\u0026thinsp;\u0026lt;\u0026thinsp;50) reported that large amounts of dextrose infusion administered to pregnant women before CD increased the venous cord blood glucose concentrations and subsequently elevated cord blood insulin levels that resulted in increased risk of neonatal hypoglycemia.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Consistently, the association between CD and cord CP levels was reported in very few and small studies prior.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eWe hypothesized that neonatal hyperinsulinemia herein defined by the umbilical cord CP levels could be associated with mode of delivery as the primary exposure and we aimed to assess this association in a database derived from a large, multicenter, multi-ancestry clinical study.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Design\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe retrospectively analyzed data obtained from the HAPO multinational, multiethnic and multi-ancestry, population-based observational study\u003csup\u003e11\u003c/sup\u003e, maternal glycemia and birthweight gene-environment interaction (GEI) substudy dataset.\u003csup\u003e12\u003c/sup\u003e The enrollment and data collection period of the original HAPO study lasted from 2000 to 2006 and it included pregnant women of 4 major ancestry groups (Hispanic, Afro-Caribbean, South-Asian and European) from 9 countries and 15 study centers.\u003csup\u003e11\u003c/sup\u003e \u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe accessed the GENEVA GWA mapping Maternal Metabolism-Birth Weight Interactions dataset via the NCBI dbGaP (the Database of Genotypes and Phenotypes) system (dbGaP accession phs000096.v4.p1) that contained data on a total of 4994 mother-neonate pairs from the original HAPO cohort study.\u003csup\u003e11,12\u003c/sup\u003e \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy Cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAfter initial restriction to mandatory data in our analysis (cord CP level and ancestry was required for the primary outcome and mode of delivery as the main exposure) the dataset contained data on 4967 of 4994 mother-neonate pairs. The complete case analysis consisted of 3492 mother-neonate pairs where all variables employed in final model were available.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExposure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe primary exposure was the mode of delivery (cesarean delivery or vaginal delivery).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe considered additional variables potentially associated with cord CP levels including maternal variables: (1) age at oral glucose tolerance test (OGTT), (2) pre-pregnancy weight (kg), (3) family history of diabetes, (4) family history of hypertension, (5) ethnicity, (6-7-8) fasting, 60min, 120min plasma glucose (PG) values at OGTT (mmol/L), (9) HbA1c at OGTT (%), (10) fasting CP levels at OGTT (ug/L), (11) weight at OGTT (kg), (12) hypertension (any type), (13) \u0026nbsp;proteinuria highest urine dipstick result \u0026gt;=3+, (14) height (cm), (15) prior pregnancy delivered at \u0026gt;= 20 weeks, (16) drinker status, (17) smoker status, (18) mean systolic blood pressure at OGTT (mmHg) and also neonatal variables measured at delivery: \u0026nbsp;(19) gestational age, (20) birth weight (kg), (21) mean flank skinfold (mm), (22) head circumference (cm), (23) length (cm), (24) subscapular skinfold (mm), (25) triceps skinfold (mm), (26) sex, (27) venous cord blood PG concentration (mmol/L).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOutcome\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe primary outcome of the analysis was neonatal hyperinsulinemia herein defined as cord serum CP level of \u0026gt; 90\u003csup\u003eth\u003c/sup\u003e percentile. Ancestry specific CP 90\u003csup\u003eth\u003c/sup\u003e percentile cutoff values were 1.6 ng/mL in the Afro-Caribbean and South-Asian populations and 1.7 ng/mL in those with European and Hispanic origins, respectively. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCovariates\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClinical variables included gestational diabetes mellitus (GDM), defined from maternal plasma glucose (PG) values during the 24\u0026ndash;28-week OGTT using International Association of the Diabetes and Pregnancy Study Groups criteria, as adopted by WHO in 2013.\u003csup\u003e13,14\u003c/sup\u003e GDM was used as a categorical variable; individual OGTT PG values were not included simultaneously. HOMA2-IR was calculated from fasting maternal PG and CP values using the HOMA2 calculator and modeled as a continuous predictor; corresponding fasting maternal PG and CP values were excluded to avoid redundancy.\u003csup\u003e15\u003c/sup\u003e Although HOMA2-IR determination is not routinely recommended for use during pregnancy, it is still an acceptable estimate of insulin resistance under study settings at OGTT.\u003csup\u003e16\u003c/sup\u003e An alternative model used fasting, 1-, and 2-hour PG values as continuous variables, excluding GDM and HOMA2-IR. Additional covariates were maternal pre-pregnancy BMI, gestational weight gain to OGTT, and preeclampsia (binary). When derived predictors were included, their component variables were excluded to reduce collinearity and overfitting.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026chi;\u003csup\u003e2\u003c/sup\u003e-test assessed differences in the primary endpoint between CD and vaginal delivery groups. Neonatal hyperinsulinemia was modeled using standardized multivariate logistic regression. \u0026nbsp;A backward variable selection procedure was iteratively applied, whereby in each cycle the variable whose removal yielded the greatest improvement in the model\u0026rsquo;s Akaike Information Criterion (AIC) was excluded. This process was performed using a complete-case dataset, comprising only individuals with no missing values for any of the candidate variables. Once no further improvement in AIC could be reached by variable removal, the optimal model was established for this reduced subset of the data. To maximize the sample size without compromising the integrity of the variable selection process, we implemented an iterative approach as presented in \u003cstrong\u003eFigure 1\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInitially, the dataset was restricted to complete cases. AIC-based variable selection was then performed, and the variables retained in the resulting optimal model were identified. The dataset was subsequently redefined to include complete cases only for this reduced set of variables. This process was repeated iteratively until the set of selected variables remained stable across successive iterations. At each step, the sample size was preserved or increased relative to the previous iteration, ensuring convergence toward an optimal model that incorporated the maximum number of participants possible under the complete-case constraint. The algorithm converged after three iterations in our analysis.\u0026nbsp;Details of the iterative model selection steps, including sample sizes and retained variables at each stage, are provided in eTable 1 and eTable 2 in Supplement 1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModel Refinement and Interaction Terms\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the final modeling step, interaction terms between the most significant variables were incorporated, followed by an additional round of AIC-based variable selection. The inclusion of interaction terms was selectively restricted; specifically, interactions between ethnicity and maternal pre-pregnancy BMI were constrained due to established differences in healthy BMI ranges across ethnic subgroups, which could confound clinical interpretation.\u003csup\u003e17,18\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBayesian Joint Modeling for Handling Missing Data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo address the missing data\u0026mdash;particularly for maternal pre-pregnancy BMI and number of prior pregnancies\u0026mdash;a Bayesian joint modeling approach was applied using the \u003cem\u003eJointAI\u003c/em\u003e package in R. The logistic regression model, containing the same variables as the complete-case analysis, was fitted with simultaneous imputation of missing covariates via Markov Chain Monte Carlo (MCMC) sampling (yielding 1,060 samples across three independent chains). This approach accounts for uncertainty due to missing data and provides posterior estimates with corresponding 95% credible intervals and Bayesian tail probabilities. The mean posterior estimates were compared with those from the complete-case analysis to evaluate the sensitivity of the results to missing data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSensitivity Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSeveral additional sensitivity analyses were also conducted. First, the GDM and HOMA-R variables were replaced with the exact maternal PG and CP values in the final model to examine the sensitivity of this type of grouping. In an additional analysis, birth order was excluded, due to that it was missing in a large proportion of mothers with Thai ancestry. Additionally, the final complete-case model was stratified by ancestry to assess the consistency of effects across subgroups. Finally, the complete case model was reconsidered using the mode of delivery and birth order predictors combined into a single, multilevel variable.\u003c/p\u003e\n\u003cp\u003eNo further variable selection was applied within these stratified models; and the previously determined optimal model was re-fitted within each ethnic subgroup. This approach allowed the identification of both ancestry-specific and generally consistent associations.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eDemographic and Clinical Characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong the 4967 women included in the analysis after the initial restriction to mandatory data, 447 (9%) delivered neonates who presented with cord blood hyperinsulinemia. Of the 4,967 pregnancies assessed, CD was performed in 1,154 (23.2%) cases. Additional demographic and clinical characteristics of the complete-case mother\u0026ndash;neonate pairs are presented in the Table. Comparative\u0026nbsp;maternal and neonatal characteristics stratified by the mode of delivery are provided in eTable3 in Supplement 1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNeonatal Hyperinsulinemia and Cesarean Delivery\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCord blood hyperinsulinemia occurred more frequently among neonates delivered by cesarean delivery: 178 cases (15.4%) compared with 269 cases (7.1%) among those born via vaginal delivery (unadjusted odds ratio [OR], 2.4; 95% CI, 1.96\u0026ndash;2.94).\u003c/p\u003e\n\u003cp\u003eIn a multivariable model based on the analysis of 3,492 complete-case mother\u0026ndash;neonate pairs incorporating interaction terms CD was in significant interactions with the first-time delivery (interaction \u0026beta; = 0.5681; 95% credible interval, 0.0556\u0026ndash;1.0834) and with cord blood PG levels (interaction \u0026beta; = 0.2113; 95% credible interval, 0.0088\u0026ndash;0.4205). However, most examined interaction terms were eliminated during the iterative AIC-based variable selection process. The interaction terms between ancestries and maternal BMI were retained in the model due to established differences in healthy BMI ranges across ethnic subgroups, marked BMI differences across ancestry groups, although these did not exert a statistically significant effect.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFollowing Bayesian joint modeling with imputation, 4,967 mother\u0026ndash;neonate pairs were analyzed. Cesarean delivery remained significantly associated with an increased risk of neonatal hyperinsulinemia (OR, 2.66; 95% CI, 1.90\u0026ndash;3.77) in the final model (\u003cstrong\u003eFigure 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional factors associated with neonatal hyperinsulinemia\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll results refer to models adjusted for known and novel covariates associated with neonatal hyperinsulinemia (\u003cstrong\u003eFigure 2\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMaternal factors including GDM (\u0026beta; = 0.4192; 95% credible interval, 0.1379 \u0026ndash; 0.6892 OR: 1.5207 1.1479 \u0026ndash; 1.9921), HbA1c level at OGTT (\u0026beta; = 0.1764, 95% credible interval 0.0524 \u0026ndash; 0.3017), weight gain till the OGTT (\u0026beta; = 0.1270, 95% credible interval 0.0099 \u0026ndash; 0.2462) HOMA2-IR at OGTT (\u0026beta; = 0.2941, 95% credible interval 01868 \u0026ndash; 0.4041), and African ancestry (\u0026beta; = 0.6794, 95% credible interval 0.3399 \u0026ndash; 1.0195) were associated with higher, while 1st time delivery (\u0026beta; = -0.5442, 95% credible interval -0.8749 \u0026ndash; -0.2252) was associated with lower risk of neonatal cord blood hyperinsulinemia in the final model. In contrast, maternal age and pre-pregnancy BMI that were constrained in the model due to their clinical significance and had no effect (\u003cstrong\u003eFigure 2\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003eNeonatal factors, such as gestational age at delivery (\u0026beta; = -0.1950, 95% credible interval -0.3179 \u0026ndash; -0.0733), birthweight (\u0026beta; = 0.7423; 95% credible interval, 0.5749 \u0026ndash; 0.9050), venous cord blood PG (\u0026beta; = 0.6549; 95% credible interval, 0.5210 \u0026ndash; 0.7887), female sex (\u0026beta; = 0.8862; 95% credible interval, 0.6649 \u0026ndash; 1.1110), were also significantly associated with the outcome. (\u003cstrong\u003eFigure 2\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModel Accuracy, Sensitivity and Subgroup Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe assessed the performance of both the complete case analysis models using the available data sets. The performance was graphically displayed on a receiver operating characteristics (ROC) curve. (\u003cstrong\u003eFigure 3\u003c/strong\u003e). The area under the curve (AUC) value was 0.822, respectively, indicating effective model discrimination. Although validation using an independent test set would be ideal, the primary objective of this study was an association analysis; therefore, these AUC values are considered adequate assessments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn contrast, ROC AUC could not be calculated for the Bayesian model because its probabilistic outputs and handling of missing data prevent generation of single-point predictions required for ROC analysis.\u003c/p\u003e\n\u003cp\u003eTo test the stability of our findings, we conducted multiple sensitivity analyses. First, in the complete case model, we substituted the GDM and HOMA-IR variables with measured maternal PG and CP levels at OGTT. Cesarean delivery and birth order continued to exhibit similar associations (eFigure 1 in Supplement 1).\u003c/p\u003e\n\u003cp\u003eSecond, we excluded birth order from the final complete case model due to substantial missing data, particularly within the Thai subpopulation (missing birth order n = 524 out of 1241). The associations of CD and other covariates remained largely unchanged, suggesting that the missing data were likely missing at random and did not introduce any significant bias (eFigure 2 in Supplement 1).\u003c/p\u003e\n\u003cp\u003eFinally, we stratified the cohort by ethnicity (European, Afro-Caribbean, Hispanic, and Thai) and fitted the final model separately within each subpopulation. While the magnitude of effects varied across groups, the direction of associations for CD and birth order remained consistent, despite smaller sample sizes. These results are presented in eFigure 3, eFigure 4, eFigure 5, and eFigure 6 in Supplement 1. To explore the interaction between mode of delivery and birth order we further modified the final model and employed these predictors combined into single, multilevel variables as presented in eFigure 7 in Supplement 1.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this retrospective analysis of a multinational, multiethnic, and multi-ancestry population-based observational substudy of the HAPO study, we found that neonates delivered by CD were at higher risk of hyperinsulinemia in cord blood. This association remained significant after adjustment for a broad set of predictors and interactions. Sensitivity analyses confirmed the robustness of this finding, and subgroup analyses demonstrated its consistency across different ancestral groups.\u003c/p\u003e\u003cp\u003ePrevious studies have reported that children born via planned CD have an increased risk of obesity in preadolescence, with evidence that these metabolic associations persist into adulthood.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e Women born by CD were found to have a higher risk of type 2 diabetes than those born vaginally (adjusted hazard ratio, 1.46; BMI-adjusted risk ratio, 1.34).\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e Proposed mechanisms focus on altered gut microbiota, as infants born by CD may acquire less diverse microbiomes.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eThe HAPO Follow-Up Study reported an inverse association between neonatal cord C-peptide (CP) levels and the Matsuda index of whole-body insulin sensitivity in 4,155 children (mean age, 11.4 years), however did not adjust for delivery mode.\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIn our analysis, we also assessed umbilical venous cord plasma glucose (PG), which correlates strongly with maternal glucose and reflects maternal glycemia at birth.\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e Maternal dextrose infusion before CD can raise fetal insulin.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e In contrast, we found lower cord PG levels in CD compared vaginal births, suggesting that elevated hyperinsulinemia risk is not due to increased fetal glucose exposure. As expected, our final model demonstrated a significant interaction between CD and cord PG levels, alongside independent significant associations of both variables with the primary outcome of neonatal hyperinsulinemia.\u003c/p\u003e\u003cp\u003eWe also identified a novel association between first delivery and lower hyperinsulinemia risk, which contrasts with smaller studies reporting reduced insulin sensitivity in first-borns.\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e However, this study had a substantially smaller sample size (n\u0026thinsp;=\u0026thinsp;87)\u003csup\u003e23\u003c/sup\u003e, compared with our analysis that employed the data of 3492 and 4,967 mother\u0026ndash;neonate pairs in the complete case analysis and the final model, respectively. We found that this effect only in vaginal births, suggesting potential interactions with labor-related factors.\u003c/p\u003e\u003cp\u003eTaken together, these findings may provide a mechanistic link between CD and later obesity, insulin resistance, and type 2 diabetes, and highlight the need for further investigation of birth order effects.\u003c/p\u003e\u003cp\u003eThe additional predictors independently associated with neonatal hyperinsulinemia in our final model largely reflected previously reported factors or well-established pathophysiological conditions, including maternal glycemic status and insulin-resistance (GDM, HbA1c, and HOMA2-IR at OGTT), maternal weight gain in pregnancy till OGTT, and neonatal characteristics such as gestational age, birthweight, and sex.\u003csup\u003e\u003cspan additionalcitationids=\"CR25 CR26 CR27 CR28 CR29 CR30 CR31\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003eThe World Health Organization (WHO) recommends that population-level CD rates ideally range between 10% and 15%.\u003csup\u003e33\u003c/sup\u003e Contrary to these guidelines, CD rates worldwide have tripled to over 21% in between 1999 and 2023.\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e This trend, in combination with the rising prevalence of childhood obesity and associated health conditions, underscores the potential clinical significance of our findings.\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAn integrated interpretation of our findings suggests that the increased risk of neonatal hyperinsulinemia with cesarean delivery and the decreased risk with first delivery may relate to the absence or prolongation of labor, respectively, with natural labor potentially protective.\u003c/p\u003e\u003cp\u003eFurthermore, we report here that age of gestation at delivery is inversely associated with the risk of neonatal hyperinsulinemia that is consistent with a meta-analysis which reported that neonates born by preterm delivery had a higher risk of childhood obesity.\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eIt could be argued that the observed associations may be influenced by unmeasured confounders. However, the presence of an unmeasured confounder during pregnancy that would plausibly affect birth order can be considered unlikely, even from a theoretical standpoint. With regard to the mode of delivery, the 27 variables assessed likely capture the main mechanisms underlying the currently proposed pathophysiology.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e Although the possibility of residual confounding cannot be entirely ruled out, the likelihood of a meaningful unmeasured confounder is considered low.\u003c/p\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eStrengths\u003c/h2\u003e\u003cp\u003eFirst, it is a retrospective analysis of large, population-based cohort study encompassing data from 4967 mother-neonate pairs in the final model. Such a sample size enhances the statistical power and reliability of clinically relevant predictors, while minimizing recall bias and allowing for the assessment of multiple exposures in relation to neonatal hyperinsulinemia as the primary outcome. Furthermore, the multi-ancestry design of both the HAPO study and the specific data subset analyzed enhances the generalizability and applicability of our findings across diverse populations.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eLimitations\u003c/h2\u003e\u003cp\u003eOur study did not explore the biological mechanisms underlying the increased risk of neonatal hyperinsulinemia associated with cesarean delivery, nor the protective effect observed with primiparity. As such, our interpretations of these novel associations remain speculative and require further investigation through mechanistic and prospective studies. Additionally, despite rigorous statistical procedures and adjustment for confounders, the possibility of residual, unmeasured confounding cannot be excluded, limiting the ability to infer definitive causal relationships.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eWe observed an association between cesarean delivery and increased risk of neonatal hyperinsulinemia defined as the uppermost decimal of the umbilical cord blood C-peptide levels. To our knowledge, we also first report a protective effect of first delivery on neonatal hyperinsulinemia.\u003c/p\u003e\u003cp\u003eThe associations between cesarean delivery, first time delivery and neonatal hyperinsulinemia both remained significant in analyses using rigorous, multi-step adjustments for an extensive set of the most important maternal and neonatal covariates that exceeds the available data in routine pregnancy care. The multi-ancestry design of the HAPO study provided the generalizability of our findings across diverse populations.\u003c/p\u003e\u003cp\u003eThis suggests that this association may indicate a true effect of CD, and a potential benefit might be reached on neonatal hyperinsulinemia when preferring a vaginal delivery in the absence of medical or obstetric indication of CD.\u003c/p\u003e\u003cp\u003eAdditional evidence from large, prospective cohort studies is needed to determine whether CD associated neonatal cord blood hyperinsulinemia might contribute to increased risk of obesity and related metabolic disorders later in life.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eWe are grateful to Healthware Consulting Ltd., Budapest, Hungary for helping us to select and adapt the appropriate statistical methods.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCollaborators, G.B.D.R.F. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. \u003cem\u003eLancet\u003c/em\u003e \u003cstrong\u003e403\u003c/strong\u003e, 2162-2203 (2024).\u003c/li\u003e\n\u003cli\u003eOkunogbe, A.\u003cem\u003e, et al.\u003c/em\u003e Economic impacts of overweight and obesity: current and future estimates for 161 countries. \u003cem\u003eBMJ Glob Health\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e(2022).\u003c/li\u003e\n\u003cli\u003ePei, Z.\u003cem\u003e, et al.\u003c/em\u003e Cesarean delivery and risk of childhood obesity. \u003cem\u003eJ Pediatr\u003c/em\u003e \u003cstrong\u003e164\u003c/strong\u003e, 1068-1073 e1062 (2014).\u003c/li\u003e\n\u003cli\u003eHuh, S.Y.\u003cem\u003e, et al.\u003c/em\u003e Delivery by caesarean section and risk of obesity in preschool age children: a prospective cohort study. \u003cem\u003eArch Dis Child\u003c/em\u003e \u003cstrong\u003e97\u003c/strong\u003e, 610-616 (2012).\u003c/li\u003e\n\u003cli\u003eRutayisire, E.\u003cem\u003e, et al.\u003c/em\u003e Cesarean section may increase the risk of both overweight and obesity in preschool children. \u003cem\u003eBMC Pregnancy Childbirth\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 338 (2016).\u003c/li\u003e\n\u003cli\u003eTerashita, S.\u003cem\u003e, et al.\u003c/em\u003e Caesarean section and childhood obesity at age 3 years derived from the Japan Environment and Children\u0026apos;s Study. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6535 (2023).\u003c/li\u003e\n\u003cli\u003eSlabuszewska-Jozwiak, A., Szymanski, J.K., Ciebiera, M., Sarecka-Hujar, B. \u0026amp; Jakiel, G. Pediatrics Consequences of Caesarean Section-A Systematic Review and Meta-Analysis. \u003cem\u003eInt J Environ Res Public Health\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e(2020).\u003c/li\u003e\n\u003cli\u003eYuan, C.\u003cem\u003e, et al.\u003c/em\u003e Association Between Cesarean Birth and Risk of Obesity in Offspring in Childhood, Adolescence, and Early Adulthood. \u003cem\u003eJAMA Pediatr\u003c/em\u003e \u003cstrong\u003e170\u003c/strong\u003e, e162385 (2016).\u003c/li\u003e\n\u003cli\u003eBianco, M.E.\u003cem\u003e, et al.\u003c/em\u003e Hyperglycemia and Adverse Pregnancy Outcome Follow-Up Study: newborn anthropometrics and childhood glucose metabolism. \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 561-570 (2021).\u003c/li\u003e\n\u003cli\u003eKenepp, N.B.\u003cem\u003e, et al.\u003c/em\u003e Fetal and neonatal hazards of maternal hydration with 5% dextrose before caesarean section. \u003cem\u003eLancet\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 1150-1152 (1982).\u003c/li\u003e\n\u003cli\u003eHapoStudyCooperativeResearchGroup\u003cem\u003e, et al.\u003c/em\u003e Hyperglycemia and adverse pregnancy outcomes. \u003cem\u003eN Engl J Med\u003c/em\u003e \u003cstrong\u003e358\u003c/strong\u003e, 1991-2002 (2008).\u003c/li\u003e\n\u003cli\u003eGeoff Hayes; Boyd E. Metzger, A.R.D., Lynn P. Lowe, Margrit Urbanek, Nancy Cox, Dan Nicolae, William L. Lowe, Jr GWA mapping: Maternal Metabolism-Birth Weight Interactions in GENEVA GWA mapping: Maternal Metabolism-Birth Weight Interactions. (https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000096.v4.p1, 2012).\u003c/li\u003e\n\u003cli\u003eInternationalAssociationofDiabetes. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. \u003cem\u003eDiabetes Care\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 676-682 (2010).\u003c/li\u003e\n\u003cli\u003eWorldHealthOrganization. Diagnostic Criteria and Classification of Hyperglycaemia First Detected in Pregnancy. \u003cem\u003eOnline document\u003c/em\u003e \u003cstrong\u003eWHO/NMH/MND/13.2\u003c/strong\u003e, 1-62 (2013).\u003c/li\u003e\n\u003cli\u003eWallace, T.M., Levy, J.C. \u0026amp; Matthews, D.R. Use and abuse of HOMA modeling. \u003cem\u003eDiabetes Care\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1487-1495 (2004).\u003c/li\u003e\n\u003cli\u003eJones, D.L., Kusinski, L.C., Gillies, C. \u0026amp; Meek, C.L. A critique of measurement of defective insulin secretion and insulin sensitivity as a precision approach to gestational diabetes. \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e68\u003c/strong\u003e, 752-765 (2025).\u003c/li\u003e\n\u003cli\u003eConsultation, W.H.O.E. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. \u003cem\u003eLancet\u003c/em\u003e \u003cstrong\u003e363\u003c/strong\u003e, 157-163 (2004).\u003c/li\u003e\n\u003cli\u003eObesity: preventing and managing the global epidemic. Report of a WHO consultation. \u003cem\u003eWorld Health Organ Tech Rep Ser\u003c/em\u003e \u003cstrong\u003e894\u003c/strong\u003e, i-xii, 1-253 (2000).\u003c/li\u003e\n\u003cli\u003eSitarik, A.R.\u003cem\u003e, et al.\u003c/em\u003e Association between cesarean delivery types and obesity in preadolescence. \u003cem\u003eInt J Obes (Lond)\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 2023-2034 (2020).\u003c/li\u003e\n\u003cli\u003eChavarro, J.E.\u003cem\u003e, et al.\u003c/em\u003e Association of Birth by Cesarean Delivery With Obesity and Type 2 Diabetes Among Adult Women. \u003cem\u003eJAMA Netw Open\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e202605 (2020).\u003c/li\u003e\n\u003cli\u003eShao, Y.\u003cem\u003e, et al.\u003c/em\u003e Stunted microbiota and opportunistic pathogen colonization in caesarean-section birth. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e574\u003c/strong\u003e, 117-121 (2019).\u003c/li\u003e\n\u003cli\u003eMartinez, K.A., 2nd\u003cem\u003e, et al.\u003c/em\u003e Increased weight gain by C-section: Functional significance of the primordial microbiome. \u003cem\u003eSci Adv\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, eaao1874 (2017).\u003c/li\u003e\n\u003cli\u003eAyyavoo, A., Savage, T., Derraik, J.G., Hofman, P.L. \u0026amp; Cutfield, W.S. First-born children have reduced insulin sensitivity and higher daytime blood pressure compared to later-born children. \u003cem\u003eJ Clin Endocrinol Metab\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e, 1248-1253 (2013).\u003c/li\u003e\n\u003cli\u003eNiknam, A.\u003cem\u003e, et al.\u003c/em\u003e Umbilical cord blood concentration of connecting peptide (C-peptide) and pregnancy outcomes. \u003cem\u003eBMC Pregnancy Childbirth\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 764 (2022).\u003c/li\u003e\n\u003cli\u003eYin, B.\u003cem\u003e, et al.\u003c/em\u003e Combining HbA1c and insulin resistance to assess the risk of gestational diabetes mellitus: A prospective cohort study. \u003cem\u003eDiabetes Res Clin Pract\u003c/em\u003e \u003cstrong\u003e199\u003c/strong\u003e, 110673 (2023).\u003c/li\u003e\n\u003cli\u003eRifas-Shiman, S.L.\u003cem\u003e, et al.\u003c/em\u003e First and second trimester gestational weight gains are most strongly associated with cord blood levels of hormones at delivery important for glycemic control and somatic growth. \u003cem\u003eMetabolism\u003c/em\u003e \u003cstrong\u003e69\u003c/strong\u003e, 112-119 (2017).\u003c/li\u003e\n\u003cli\u003eMa, R.\u003cem\u003e, et al.\u003c/em\u003e Interpregnancy interval and early infant neurodevelopment: the role of maternal-fetal glucose metabolism. \u003cem\u003eBMC Med\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 2 (2024).\u003c/li\u003e\n\u003cli\u003eGarcia-Flores, J.\u003cem\u003e, et al.\u003c/em\u003e Weight-related and analytical maternal factors in gestational diabetes to predict birth weight and cord markers of diabetic fetopathy. \u003cem\u003eGynecol Endocrinol\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 548-552 (2016).\u003c/li\u003e\n\u003cli\u003eRegnault, N.\u003cem\u003e, et al.\u003c/em\u003e Higher cord C-peptide concentrations are associated with slower growth rate in the 1st year of life in girls but not in boys. \u003cem\u003eDiabetes\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 2152-2159 (2011).\u003c/li\u003e\n\u003cli\u003eKuchay, M.S., Kudyar, R.P., Gupta, A., Pandita, K.K. \u0026amp; Ganie, M.A. Gender differences in insulin and C-peptide concentrations at birth using cord blood collection. \u003cem\u003eArch Endocrinol Metab\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 264-266 (2016).\u003c/li\u003e\n\u003cli\u003eLee, I.L.\u003cem\u003e, et al.\u003c/em\u003e Cord blood metabolic markers are strong mediators of the effect of maternal adiposity on fetal growth in pregnancies across the glucose tolerance spectrum: the PANDORA study. \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 497-507 (2020).\u003c/li\u003e\n\u003cli\u003eShields, B.M.\u003cem\u003e, et al.\u003c/em\u003e Measurement of cord insulin and insulin-related peptides suggests that girls are more insulin resistant than boys at birth. \u003cem\u003eDiabetes Care\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 2661-2666 (2007).\u003c/li\u003e\n\u003cli\u003eBetran, A.P., Torloni, M.R., Zhang, J.J., Gulmezoglu, A.M. \u0026amp; Section, W.H.O.W.G.o.C. WHO Statement on Caesarean Section Rates. \u003cem\u003eBJOG\u003c/em\u003e \u003cstrong\u003e123\u003c/strong\u003e, 667-670 (2016).\u003c/li\u003e\n\u003cli\u003eAngolile, C.M., Max, B.L., Mushemba, J. \u0026amp; Mashauri, H.L. Global increased cesarean section rates and public health implications: A call to action. \u003cem\u003eHealth Sci Rep\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, e1274 (2023).\u003c/li\u003e\n\u003cli\u003eCollaborators, G.B.o.D.A.B. Global, regional, and national prevalence of child and adolescent overweight and obesity, 1990-2021, with forecasts to 2050: a forecasting study for the Global Burden of Disease Study 2021. \u003cem\u003eLancet\u003c/em\u003e \u003cstrong\u003e405\u003c/strong\u003e, 785-812 (2025).\u003c/li\u003e\n\u003cli\u003eWorldObesityFederation. World Obesity Atlas 2023.\u003c/li\u003e\n\u003cli\u003eOu-Yang, M.C.\u003cem\u003e, et al.\u003c/em\u003e Accelerated weight gain, prematurity, and the risk of childhood obesity: A meta-analysis and systematic review. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e0232238 (2020).\u003c/li\u003e\n\u003cli\u003eJosefson, J.L.\u003cem\u003e, et al.\u003c/em\u003e Newborn Adiposity and Cord Blood C-Peptide as Mediators of the Maternal Metabolic Environment and Childhood Adiposity. \u003cem\u003eDiabetes Care\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 1194-1202 (2021).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003e\u003cstrong\u003eTable.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eDemographic and clinical characteristics of mother\u0026ndash;neonate pairs in the complete-case analysis\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"105%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMaternal Characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComplete case (n=3492)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003eMode of delivery, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003eCesarean delivery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e831 (23.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003eVaginal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e2661 (76.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003eAge, median (IQR), y\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e29.63 (25-33.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003ePre-pregnancy weight, median (IQR), kg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e61.29 (54-72)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHeight, median (IQR), cm\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e161.29 (156.2-166.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003eBody mass index, median (IQR)\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e23.58 (20.9-27.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eAncestry, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eEuropean\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e1355 (38.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eAfro - Caribbean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e719 (20.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eThai\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e618 (17.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHispanic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e800 (22.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003ePlasma glucose at OGTT, median (IQR), mmol/L, [mg/dL]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eFasting\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e4.5 (4.3-4.8), [81 (77.4-86.4)]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003e1 - hour\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e7.3 (6.3-8.6), [131.4 (113.4-154.8)]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003e2 - hour\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e6.1 (5.3-6.9), [109.8 (95.4-124.2)]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHbA1c, median (IQR), %\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e4.7 (4.5-5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eFasting CP levels at OGTT, median, (IQR), ug/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e1.8 (1.3-2.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eGDM, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e618 (17.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHOMA2-IR, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e1.25 (0.93-1.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eWeight gain until OGTT, median (IQR), kg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e8.97 (6.1-12.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHypertension (any type), No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e421 (12.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eSystolic blood pressure at OGTT, median (IQR), mm Hg\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e106 (99.5-113)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003e1\u003csup\u003est\u0026nbsp;\u003c/sup\u003edelivery, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e1480 (42.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eDrinker, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e312 (8.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eSmoker, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e191 (5.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 62px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNeonatal Characteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComplete case (n=3492)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eGestational age, median (IQR), weeks\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e39.71 (38.9-40.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eBirth weight, median (IQR), g\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e3295(3010-3632.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eFlank skinfold, median (IQR), mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e3.95 (3.4-4.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHead circumference, median (IQR), cm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e34.3 (33.5-35.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eLength, median (IQR), cm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e50 (49-51.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eSubscapular skinfold, median (IQR), mm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e4.3 (3.7-5.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eSex, Male No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e1768 (50.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eVenous cord blood glucose, median (IQR), mmol/L [mg/dL]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e4,4 (3.8-5.1), [79.2 (68.4-91.8)]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eVenous cord C-peptide, median (IQR), ug/L\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e0.9 (0.7-1.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 62px;\"\u003e\n \u003cp\u003eHyperinsulinemia, No. (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 37px;\"\u003e\n \u003cp\u003e299 (8.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Semmelweis University","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":"Cesarean delivery, Mode of delivery, Birth order, Neonatal hyperinsulinemia, HAPO study, Cord blood, C-peptide","lastPublishedDoi":"10.21203/rs.3.rs-7676142/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7676142/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eImportance: \u003c/strong\u003eCesarean delivery (CD) is associated with higher risk of obesity development from childhood and with type 2 diabetes mellitus (T2DM) later in life. Higher neonatal cord C-peptide (CP) levels were reported to be associated with increased risk of offspring metabolic disorders in childhood as well.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective: \u003c/strong\u003eTo assess the association between mode of delivery as the primary exposure and 27 additional maternal/neonatal clinical factors and the risk of neonatal hyperinsulinemia.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDesign: \u003c/strong\u003eRetrospective data analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSetting: \u003c/strong\u003eMultinational, multi-ancestry observational population-based Hyperglycemia Advanced Pregnancy Outcome (HAPO) cohort study gene-environment interaction (GEI) substudy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eParticipants: \u003c/strong\u003eA total of 4967 mother-neonate pairs from 4 major ancestry groups (Afro-Caribbean, European, Hispanic, Thai) were assessed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExposure: \u003c/strong\u003eData on mode of delivery and 27 additional (18 maternal + 9 neonatal) variables were utilized from the NIH dbGaP database (accession phs000096.v4.p1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMain outcome and measures: \u003c/strong\u003eThe primary outcome was neonatal hyperinsulinemia herein defined as cord serum CP level of ancestry specific \u0026gt; 90th percentile. χ\u003csup\u003e2\u003c/sup\u003e-test was used to assess the difference between CD and vaginal delivery groups and standardized multivariate logistic regression with Akaike information criterion-based model selection and interaction analyses were used to model the primary outcome. Missing data were addressed using Bayesian joint modeling approach.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eOut of the total 4967 pregnancies assessed CD was performed in 1154 mothers (23.2%). Cord blood hyperinsulinemia occurred in 447 cases of which 178 in the CD (15.4%) and 269 in vaginal delivery (7.1%) group (OR crude, 2.4; 95% CI, 1.96 to 2.94). We identified novel factors individually associated with neonatal hyperinsulinemia: CD (OR in firstborns: 4.25; 95% CI, 2.66-6.78, OR in non-firstborns: 4.25; 95% CI, 2.59-6.99) and non-firstborn vaginal delivery (OR, 1.95; 95% CI, 1.34-2.88) when compared to vaginally delivered firstborns. Sensitivity analyses confirmed the stability of these results and subgroup analyses revealed the consistency of the two novel associations across different ancestries.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: To our best knowledge we first report that neonatal hyperinsulinemia risk is more than quadrupled in neonates born via cesarean delivery independently of birth order when compared to vaginally delivered firstborns. \u0026nbsp;Non-firstborns are also at increased risk.\u003c/p\u003e","manuscriptTitle":"Association of Mode of Delivery and Birth Order with Hyperinsulinemia in Neonatal Cord Blood","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-25 09:30:38","doi":"10.21203/rs.3.rs-7676142/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":"1559f266-d673-47ec-8725-5be72066223f","owner":[],"postedDate":"September 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":55356378,"name":"Translational Medicine"},{"id":55356379,"name":"Maternal \u0026 Fetal Medicine"}],"tags":[],"updatedAt":"2025-09-25T18:54:24+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-25 09:30:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7676142","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7676142","identity":"rs-7676142","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}