Sex differences in DNA methylation across gestation: a large scale, cross-cohort, multi-tissue analysis.

Biological sex is a key variable influencing many physiological systems. Disease prevalence as well as treatment success can be modified by sex. Differences emerge already early in life and include pregnancy complications and adverse birth outcomes. The placenta is a critical organ for fetal development and shows sex-based differences in the expression of hormones and cytokines. Epigenetic regulation, such as DNA methylation (DNAm), may underlie the previously reported placental sexual dimorphism. We associated placental DNAm with fetal sex in three cohorts. Individual cohort results were meta-analyzed with random-effects modelling. CpG-sites differentially methylated with sex were further investigated regarding pathway enrichment, overlap with methylation quantitative trait loci (meQTLs), and hits from phenome-wide association studies (PheWAS). We evaluated the consistency of findings across tissues (CVS, i.e. chorionic villus sampling from early placenta, and cord blood) as well as with gene expression. We identified 10,320 epigenome-wide significant sex-differentially methylated probes (DMPs) spread throughout the epigenome of the placenta at birth. Most DMPs presented with lower DNAm levels in females. DMPs mapped to genes upregulated in brain, were enriched for neurodevelopmental pathways and significantly overlapped with meQTLs and PheWAS hits. Effect sizes were moderately correlated between CVS and placenta at birth, but only weakly correlated between birth placenta and cord blood. Sex differential gene expression in birth placenta was less pronounced and implicated genetic regions only marginally overlapped with those associated with differential DNAm. Our study provides an integrative perspective on sex-differential DNAm in perinatal tissues underscoring the possible link between placenta and brain.

Parity modulates impact of BMI and gestational weight gain on gut microbiota in human pregnancy.

Dysregulation of maternal adaptations to pregnancy due to high pre-pregnancy BMI (pBMI) or excess gestational weight gain (GWG) is associated with worsened health outcomes for mothers and children. Whether the gut microbiome contributes to these adaptations is unclear. We longitudinally investigated the impact of pBMI and GWG on the pregnant gut microbiome. We show that the gut microbiota of participants with higher pBMI changed less over the course of pregnancy in primiparous but not multiparous participants. This suggests that previous pregnancies may have persistent impacts on maternal adaptations to pregnancy. This ecological memory appears to be passed on to the next generation, as parity modulated the impact of maternal GWG on the infant gut microbiome. This work supports a role of the gut microbiome in maternal adaptations to pregnancy and highlights the need for longitudinal sampling and accounting for parity as key considerations for studies of the microbiome in pregnancy and infants. Understanding how these factors contribute to and shape maternal health is essential for the development of interventions impacting the microbiome, including pre/probiotics.

Alterations of materno-placento-fetal glucose homeostasis after a single course of antenatal betamethasone.

INTRODUCTION: Intrauterine growth impairment is associated with long-term metabolic changes (perinatal programming). We recently demonstrated that antenatal betamethasone (BET) decreased head circumference in term born females. Since glucose is the main energy source for fetal growth, BET-induced changes in maternal glucose homeostasis, a reduced transplacental glucose transfer or an altered fetal glucose utilization may be linked with the observed growth impairment. METHODS: 86 pregnant women exposed to BET (single course, <34 + 0 weeks of gestation (wks)) were compared to 92 gestational-age/sex-matched controls. Glucose, insulin, leptin, insulin-like growth factors (IGF-1, IGF-2) and their binding proteins (IGFBP-1, IGFBP-3) were measured in maternal and umbilical cord blood samples. Homeostasis Model Assessment (HOMA-IR) was calculated. Placental glucose transporter 1 and 3 (GLUT1, GLUT3) protein levels were determined. Statistics were performed for overall and subgroup analysis (gestational age, sex). RESULTS: After BET maternal HOMA-IR was elevated, IGFBP-1 reduced. In female pregnancies, glucose levels ≥37 + 0 wks and IGF-1 levels <37 + 0 wks were tendentially increased. Placental GLUT1 and GLUT3 protein levels were not significantly altered. Fetal umbilical venous glucose levels ≥37 + 0 wks were increased. HOMA-IR tended to be elevated in females. DISCUSSION: Growth impairment after BET appears neither caused by maternal nor fetal hypoglycemia nor changes of GLUT1 and GLUT3 total protein levels. Nonetheless, glucose homeostasis of mothers and daughters was altered even beyond the BET time frame (hyperglycemia, enhanced insulin resistance). Despite glucose supply was sufficient, an anabolic effect was apparently absent. Overall, our results highlight the relevance of adequate glucose management after BET and peripartum.

The long-lasting shadow of litter size in rodents: litter size is an underreported variable that strongly determines adult physiology.

BACKGROUND/PURPOSE: Litter size is a biological variable that strongly influences adult physiology in rodents. Despite evidence from previous decades and recent studies highlighting its major impact on metabolism, information about litter size is currently underreported in the scientific literature. Here, we urge that this important biological variable should be explicitly stated in research articles. RESULTS/CONCLUSION: Below, we briefly describe the scientific evidence supporting the impact of litter size on adult physiology and outline a series of recommendations and guidelines to be implemented by investigators, funding agencies, editors in scientific journals, and animal suppliers to fill this important gap.

Reliability of a novel approach for reference-based cell type estimation in human placental DNA methylation studies.

The placenta is a central organ during early development, influencing trajectories of health and disease. DNA methylation (DNAm) studies of human placenta improve our understanding of how its function relates to disease risk. However, DNAm studies can be biased by cell type heterogeneity, so it is essential to control for this in order to reduce confounding and increase precision. Computational cell type deconvolution approaches have proven to be very useful for this purpose. For human placenta, however, an assessment of the performance of these estimation methods is still lacking. Here, we examine the performance of a newly available reference-based cell type estimation approach and compare it to an often-used reference-free cell type estimation approach, namely RefFreeEWAS, in placental genome-wide DNAm samples taken at birth and from chorionic villus biopsies early in pregnancy using three independent studies comprising over 1000 samples. We found both reference-free and reference-based estimated cell type proportions to have predictive value for DNAm, however, reference-based cell type estimation outperformed reference-free estimation for the majority of data sets. Reference-based cell type estimations mirror previous histological knowledge on changes in cell type proportions through gestation. Further, CpGs whose variation in DNAm was largely explained by reference-based estimated cell type proportions were in the proximity of genes that are highly tissue-specific for placenta. This was not the case for reference-free estimated cell type proportions. We provide a list of these CpGs as a resource to help researchers to interpret results of existing studies and improve future DNAm studies of human placenta.

Dosage escalation of antenatal steroids in preterm twin pregnancies does not improve long-term outcome.

OBJECTIVES: To analyze long-term effects of antenatal betamethasone (≤16 mg, =24 mg and >24 mg) in preterm twins on infant and childhood morbidity. METHODS: Retrospective cohort study among 198 preterm twins. Three follow up time points, including a total of 84 outcomes, were evaluated: first neonatal examination after birth and in the neonatal period up to 10 days after birth using data from the clinic charts; examination from the 21st to the 24th month of life and examination from the 60th to the 64th months, using data from copies of the children's examination booklets sent back by the parents. Dosage-dependent and sex-specific long-term effects of antenatal betamethasone treatment on neonatal, infant and early childhood development and morbidity up to 5.3 years of age were analyzed. RESULTS: Dosage escalation of >24 mg was not associated with improved neonatal, infant or early child hood outcome, independent of twin pair structure. In contrast, higher doses >24 mg were significantly linked to increased rates of congenital infections (OR 5.867, 95% CI 1.895-18.167). Male sex as a factor was obvious for lower rates of apnea-bradycardia-syndrome in neonates, higher rates of no free steps after 15 months in infancy and highest rates of motor clumsiness in early childhood. CONCLUSIONS: Betamethasone dosage escalation >24 mg in twins born between 23+5 and 33+6 weeks of gestation did not improve neonatal, infant or early childhood morbidity. In contrast, higher doses >24 mg total dose resulted in significantly higher rates of congenital infections and are not recommended. For males, 24 mg betamethasone appears to be the preferable dose.

Betamethasone administration during pregnancy is associated with placental epigenetic changes with implications for inflammation.

BACKGROUND: Glucocorticoids (GCs) play a pivotal role in fetal programming. Antenatal treatment with synthetic GCs (sGCs) in individuals in danger of preterm labor is common practice. Adverse short- and long-term effects of antenatal sGCs have been reported, but their effects on placental epigenetic characteristics have never been systematically studied in humans. RESULTS: We tested the association between exposure to the sGC betamethasone (BET) and placental DNA methylation (DNAm) in 52 exposed cases and 84 gestational-age-matched controls. We fine-mapped associated loci using targeted bisulfite sequencing. The association of placental DNAm with gene expression and co-expression analysis on implicated genes was performed in an independent cohort including 494 placentas. Exposure to BET was significantly associated with lower placenta DNAm at an enhancer of FKBP5. FKBP5 (FK506-binding protein 51) is a co-chaperone that modulates glucocorticoid receptor activity. Lower DNAm at this enhancer site was associated with higher expression of FKBP5 and a co-expressed gene module. This module is enriched for genes associated with preeclampsia and involved in inflammation and immune response. CONCLUSIONS: Our findings suggest that BET exposure during pregnancy associates with few but lasting changes in placental DNAm and may promote a gene expression profile associated with placental dysfunction and increased inflammation. This may represent a pathway mediating GC-associated negative long-term consequences and health outcomes in offspring.

Altered SOCS3 DNA methylation within exon 2 is associated with increased mRNA expression in visceral adipose tissue in gestational diabetes.

Overweight/obesity is the main risk factor for gestational diabetes mellitus (GDM). In our cohort of pregnant women with GDM, n = 19, and without, n = 22, we previously reported a significant increase in SOCS3 mRNA expression (+62%) in visceral adipose tissue (VAT) according to GDM, without altered promoter DNA-methylation. Here, we examined methylation status of additional SOCS3 exon 2 regions in VAT and maternal blood. We found significantly altered methylation at specific CpG sites corresponding to aberrant mRNA expression levels of SOCS3 in VAT. We propose a potential regulatory element/region within exon 2; however, this region does not appear to be a good blood-marker representing VAT.

Long term alterations of growth after antenatal steroids in preterm twin pregnancies.

OBJECTIVES: To compare the long-term effects of antenatal betamethasone (ANS, ≤16 mg, =24 mg and >24 mg) in twins on infant and childhood growth. METHODS: A retrospective cohort follow up study among 198 twins after ANS including three time points: U1 first neonatal examination after birth and in the neonatal period; U7 examination from the 21st to the 24th month of life and U9 examination from the 60th to the 64th month of life using data from copies of the children's examination booklets. Inclusion criteria are twin pregnancies with preterm labor, cervical shortening, preterm premature rupture of membranes, or vaginal bleeding, and exposure to ANS between 23+5 and 33+6 weeks. Outcome measures are dosage-dependent and sex-specific effects of ANS on growth (body weight, body length, head circumference, body mass index and ponderal index) up to 5.3 years. RESULTS: Overall, 99 live-born twin pairs were included. Negative effects of ANS on fetal growth persisted beyond birth, altered infant and childhood growth, independent of possible confounding factors. Overall weight percentile significantly decreased between infancy and early childhood by 18.8%. Birth weight percentiles significantly changed in a dose dependent and sex specific manner, most obviously in female-female and mixed pairs. The ponderal index significantly decreased up to 42.9%, BMI index increased by up to 33.8%. CONCLUSIONS: ANS results in long-term alterations in infant and childhood growth. Changes between infancy and early childhood in ponderal mass index and BMI, independent of dose or twin pair structure, might indicate an ANS associated increased risk for later life disease. SYNOPSIS: First-time report on long-term ANS administration growth effects in twin pregnancies, showing persisting alterations beyond birth in infant and childhood growth up to 5.3 years as potential indicator of later life disease risk.

Maternal but Not Paternal High-Fat Diet (HFD) Exposure at Conception Predisposes for 'Diabesity' in Offspring Generations.

While environmental epigenetics mainly focuses on xenobiotic endocrine disruptors, dietary composition might be one of the most important environmental exposures for epigenetic modifications, perhaps even for offspring generations. We performed a large-scale rat study on key phenotypic consequences from parental (F0) high-caloric, high-fat diet (HFD) food intake, precisely and specifically at mating/conception, focusing on 'diabesity' risk in first- (F1) and second- (F2) generation offspring of both sexes. F0 rats (maternal or paternal, respectively) received HFD overfeeding, starting six weeks prior to mating with normally fed control rats. The maternal side F1 offspring of both sexes developed a 'diabesity' predisposition throughout life (obesity, hyperleptinemia, hyperglycemia, insulin resistance), while no respective alterations occurred in the paternal side F1 offspring, neither in males nor in females. Mating the maternal side F1 females with control males under standard feeding conditions led, again, to a 'diabesity' predisposition in the F2 generation, which, however, was less pronounced than in the F1 generation. Our observations speak in favor of the critical impact of maternal but not paternal metabolism around the time frame of reproduction for offspring metabolic health over generations. Such fundamental phenotypic observations should be carefully considered in front of detailed molecular epigenetic approaches on eventual mechanisms.

Visceral Adipose Tissue Inflammatory Factors (TNF-Alpha, SOCS3) in Gestational Diabetes (GDM): Epigenetics as a Clue in GDM Pathophysiology.

Gestational diabetes (GDM) is among the most challenging diseases in westernized countries, affecting mother and child, immediately and in later life. Obesity is a major risk factor for GDM. However, the impact visceral obesity and related epigenetics play for GDM etiopathogenesis have hardly been considered so far. Our recent findings within the prospective 'EaCH' cohort study of women with GDM or normal glucose tolerance (NGT), showed the role, critical factors of insulin resistance (i.e., adiponectin, insulin receptor) may have for GDM pathophysiology with epigenetically modified expression in subcutaneous (SAT) and visceral (VAT) adipose tissues. Here we investigated the expression and promoter methylation of key inflammatory candidates, tumor necrosis factor-alpha (TNF-α) and suppressor of cytokine signaling 3 (SOCS3) in maternal adipose tissues collected during caesarian section (GDM, n = 19; NGT, n = 22). The mRNA expression of TNF-α and SOCS3 was significantly increased in VAT, but not in SAT, of GDM patients vs. NGT, accompanied by specific alterations of respective promoter methylation patterns. In conclusion, we propose a critical role of VAT and visceral obesity for the pathogenesis of GDM, with epigenetic alterations of the expression of inflammatory factors as a potential factor.

Sex-specific epigenetic alterations of the hypothalamic Agrp-Pomc system do not explain 'diabesity' in the offspring of high-fat diet (HFD) overfed maternal rats.

Maternal high-fat diet (HFD) overfeeding pre- and during pregnancy and lactation may 'program' a 'diabesity' predisposition in the offspring, for inconclusive reasons. Acquired alterations of the hypothalamic promoter methylation and mRNA expression of the satiety neurohormone Pomc are possibly of critical importance here. We investigated within one developmental approach, including male and female rats, the sex-specific DNA methylation pattern and corresponding mRNA expression of both Pomc and its endogenous functional antagonist Agrp in the hypothalamus of adult HFD offspring. Obesity and diabetic disturbances occurred in both male and female HFD offspring, accompanied by altered Pomc promoter methylation pattern. However, this was not related to significant Pomc mRNA expression alterations. In contrast, male-specific alterations of Agrp promoter methylation were found, even associated with reduced mRNA expression of this orexigenic/anabolic Pomc antagonist. In conclusion, acquired epigenetic alterations of the hypothalamic Agrp-Pomc system hardly explain the 'diabesity' phenotype in HFD offspring, while distinct vulnerability and functionality of Agrp promoter and related genomic regions methylation should be further investigated.

Visceral adipose tissue alteration of PI3KR1 expression is associated with gestational diabetes but not promoter DNA methylation.

Obesity and diabetes are at an epidemic rate, as well as growing incidences of gestational diabetes mellitus (GDM) which causes pregnancy risks, and harm in both maternal and child health. It remains unclear which molecular mechanisms are driving the functional differences between visceral and subcutaneous fat and how these types directly affect an individual's health outcome. Paired abdominal subcutaneous and omental visceral adipose tissue were collected from women with GDM (n = 20) and with normal glucose tolerance (NGT, n = 22) during planned caesarian section. Both groups had similar maternal age (average 32.5 years) and BMI at delivery (average 33.3 kg/m2). Adipose tissue mRNA expression analyses of insulin signalling genes: PI3KCA, PI3KR1, IRS1 and IRS2 showed significantly decreased PI3KR1 expression (-23%) in visceral fat in GDM with no association to promoter DNA methylation. Reduced visceral fat PI3KR1 expression appears to be a pathogenic factor in GDM but not through altered promoter methylation.

Hypothalamic insulin receptor expression and DNA promoter methylation are sex-specifically altered in adult offspring of high-fat diet (HFD)-overfed mother rats.

Maternal overnutrition around reproduction has been shown to increase the offspring's risk for "diabesity," mediated by altered hypothalamic neuropeptide expression. In this report, a possible contribution of altered hypothalamic sensing capacity for the peripheral satiety signals glucose, insulin and leptin will be addressed, taking into account potential sex differences. Specifically, we evaluated the effects a maternal high-fat diet (HFD) overfeeding has in rats pre- and during pregnancy and lactation on the hypothalamic gene expression patterns of insulin and leptin receptors (InsR, ObRb) and glucose transporter 3 (Glut3) as well as DNA methylation in the offspring at adult age (day 200 of life). Maternal HFD consumption resulted in a metabolic syndrome phenotype, i.e., obesity, hyperleptinemia, hyperinsulinemia, impaired glucose tolerance and increased homeostatic model assessment of insulin resistance. Interestingly, in turn, insulin resistance was more pronounced in male offspring, accompanied by decreased hypothalamic InsR-mRNA. This was linked with hypermethylation of an activating transcription factor binding site within the hypothalamic InsR promoter. The degree of methylation correlated inversely with respective InsR expression, while InsR expression itself was inversely related to phenotypic "diabesity." Expression of ObRb and Glut3 mRNA was not significantly changed. In conclusion, sex-specific alterations of hypothalamic InsR expression and DNA promoter methylation in adult offspring of HFD-overfed dams may lead to hypothalamic insulin resistance and "diabesity," with males predisposed to this epigenetic malprogramming.

Reduced Insulin Receptor Expression and Altered DNA Methylation in Fat Tissues and Blood of Women With GDM and Offspring.

Context: Altered expression of the insulin receptor (IR) in adipose tissue (AT) could contribute to gestational diabetes mellitus (GDM) etiopathogenesis. Transcriptional regulation via epigenetic mechanisms (e.g., DNA methylation) may play a critical role. However, the human IR promoter DNA methylation patterns and involvement in gene expression are unknown. Objective: We evaluated IR mRNA and protein expression accompanied by targeted DNA methylation analyses in AT and blood cells of women with GDM and their offspring. Design: Prospective observational study. Setting: Academic clinic and research unit. Participants: GDM-affected (n = 25) and matched control (n = 30) mother-child dyads. Main Outcome Measures: Maternal IR gene and protein expression in paired subcutaneous (SAT) and visceral adipose tissue samples (VAT). DNA methylation levels in IR promoter and intronic regions in maternal AT and blood cells of mother-offspring pairs. Results: In SAT and VAT, IR mRNA/protein expressions were significantly reduced in women with GDMs (P < 0.05). The decrease in VAT was more pronounced and independent of maternal body mass index. VAT IR protein levels were inversely associated with key maternal and neonatal anthropometric and metabolic parameters (P < 0.05). DNA methylation patterns were similar across tissues, with significant yet small size alterations between groups in mothers and offspring (P < 0.05). Conclusion: Decreased IR levels in AT may be a relevant pathogenic factor in GDM, affecting materno-fetal metabolism. Further investigation of causal factors for IR dysregulation is necessary, especially in VAT. Potential functional and/or clinical roles of altered DNA methylation also should be evaluated.

Alterations of adiponectin gene expression and DNA methylation in adipose tissues and blood cells are associated with gestational diabetes and neonatal outcome.

BACKGROUND: Adiponectin critically contributes to metabolic homeostasis, especially by insulin-sensitizing action. Gestational diabetes mellitus (GDM) is characterized by insulin resistance leading to materno-fetal hyperglycemia and detrimental birth outcomes. By investigating paired subcutaneous (SAT) and visceral adipose tissue (VAT) as well as blood (cell) samples of GDM-affected (n = 25) vs. matched control (n = 30) mother-child dyads of the prospective "EaCH" cohort study, we addressed whether alterations of adiponectin plasma, mRNA, and DNA methylation levels are associated with GDM and offspring characteristics. RESULTS: Hypoadiponectinemia was present in women with GDM, even after adjustment for body mass index (BMI). This was accompanied by significantly decreased mRNA levels in both SAT and VAT (P < 0.05), independent of BMI. Maternal plasma adiponectin showed inverse relations with glucose and homeostatic model assessment of insulin resistance (both P < 0.01). In parallel to reduced mRNA expression in GDM, significant (P < 0.05) yet small alterations in locus-specific DNA methylation were observed in maternal fat (~ 2%) and blood cells (~ 1%). While newborn adiponectin levels were similar between groups, DNA methylation in GDM offspring was variously altered (~ 1-4%; P < 0.05). CONCLUSIONS: Reduced adiponectin seems to be a pathogenic co-factor in GDM, even independent of BMI, affecting materno-fetal metabolism. While altered maternal DNA methylation patterns appear rather marginally involved, functional, diagnostic, and/or predictive implications of cord blood DNA methylation should be further evaluated.

Maternal overweight is not an independent risk factor for increased birth weight, leptin and insulin in newborns of gestational diabetic women: observations from the prospective 'EaCH' cohort study.

BACKGROUND: Both gestational diabetes mellitus (GDM) as well as overweight/obesity during pregnancy are risk factors for detrimental anthropometric and hormonal neonatal outcomes, identified to 'program' adverse health predispositions later on. While overweight/obesity are major determinants of GDM, independent effects on critical birth outcomes remain unclear. Thus, the aim of the present study was to evaluate, in women with GDM, the relative/independent impact of overweight/obesity vs. altered glucose metabolism on newborn parameters. METHODS: The prospective observational 'Early CHARITÉ (EaCH)' cohort study primarily focuses on early developmental origins of unfavorable health outcomes through pre- and/or early postnatal exposure to a 'diabetogenic/adipogenic' environment. It includes 205 mother-child dyads, recruited between 2007 and 2010, from women with treated GDM and delivery at the Clinic of Obstetrics, Charité - Universitätsmedizin Berlin, Germany. Recruitment, therapy, metabolite/hormone analyses, and data evaluation were performed according to standardized guidelines and protocols. This report specifically aimed to identify maternal anthropometric and metabolic determinants of anthropometric and critical hormonal birth outcomes in 'EaCH'. RESULTS: Group comparisons, Spearman's correlations and unadjusted linear regression analyses initially confirmed that increased maternal prepregnancy body-mass-index (BMI) is a significant factor for elevated birth weight, cord-blood insulin and leptin (all P < 0.05). However, consideration of and adjustment for maternal glucose during late pregnancy showed that no maternal anthropometric parameter (weight, BMI, gestational weight gain) remained significant (all n.s.). In contrast, even after adjustment for maternal anthropometrics, third trimester glucose values (fasting and postprandial glucose at 32nd and 36th weeks' gestation, HbA1c in 3rd trimester and at delivery), were clearly positively associated with critical birth outcomes (all P < 0.05). CONCLUSIONS: Neither overweight/obesity nor gestational weight gain appear to be independent determinants of increased birth weight, insulin and leptin. Rather, 3rd trimester glycemia seems to be crucial for respective neonatal outcomes. Thus, gestational care and future research studies should greatly consider late pregnancy glucose in overweight/obese women with or without GDM, for evaluation of critical causes and interventional strategies against 'perinatal programming of diabesity' in the offspring.

DNA methylation and expression of proopiomelanocortin (POMC) gene in the hypothalamus of three-week-old chickens show sex-specific differences.

Increased availability and improved sequence annotation of the chicken (Gallus gallus f. domestica) genome have sparked interest in the bird as a model system to investigate translational embryonic development and health/disease outcomes. However, the epigenetics of this bird genome remain unclear. The aim of this study was to determine the levels of gene expression and DNA methylation at the proopiomelanocortin (POMC) gene in the hypothalamus of 3-week-old chickens. POMC is a key player in the control of the stress response, food intake, and metabolism. DNA methylation of the promoter, CpG island, and gene body regions of POMC were measured. Our data illustrate the pattern, variability, and functionality of DNA methylation for POMC expression in the chicken. Our findings show correlation of methylation pattern and gene expression along with sex-specific differences in POMC. Overall, these novel data highlight the promising potential of the chicken as a model and also the need for breeders and researchers to consider sex ratios in their studies.

Fetal and neonatal outcomes after term and preterm delivery following betamethasone administration in twin pregnancies.

OBJECTIVE: To investigate effects of betamethasone on fetal growth and neonatal outcomes in twins. METHODS: A retrospective cohort study was conducted of twins delivered at one center in Berlin, Germany, between 1993 and 2011. The betamethasone group included twin pregnancies with preterm labor, cervical shortening, preterm premature rupture of membranes, or vaginal bleeding, and exposure to betamethasone between 23(+5) and 33(+6) weeks. The control group included twin pregnancies with no betamethasone exposure matched for length at delivery. Fetal growth and neonatal anthropometric data were analyzed by twin-pair structure, dose, and gestational age (linear mixed model). RESULTS: Overall, 1922 live-born twin pairs (653 betamethasone group, 1269 controls) were included. Compared with controls, late-preterm twins exposed to betamethasone were lighter (mean difference -126g), had a smaller head circumference (-0.4cm), and a shorter body length (-0.8cm) after adjustment for confounders (P<0.05). Female neonates from mixed or same-sex twin pairs had a lower birth weight than controls (betamethasone ≤16mg: -114g; betamethasone 24mg: -124g; betamethasone >24mg: -187g), with no detectable improvement in neonatal morbidity (hyperbilirubinemia, respiratory distress, asphyxia) or mortality. CONCLUSION: Betamethasone reduced birth weight, head circumference, and length of female preterm neonates in twin pairs in a dose-dependent manner. The neonatal mortality and morbidity were not improved by betamethasone.