ABSTRACT
Venous thromboembolism (VTE) is a life-threating condition that is common in patients with adult-type diffuse gliomas, yet thromboprophylaxis is controversial because of possible intracerebral hemorrhage. Effective VTE prediction models exist for other cancers, but not glioma. Our objective was to develop a VTE prediction tool to improve glioma patient care, incorporating clinical, blood-based, histologic, and molecular markers. We analyzed preoperative arterial blood, tumor tissue, and clinical-pathologic data (including next-generation sequencing data) from 258 patients with newly diagnosed World Health Organization (WHO) grade 2 to 4 adult-type diffuse gliomas. Forty-six (17.8%) experienced VTE. Tumor expression of tissue factor (TF) and podoplanin (PDPN) each positively correlated with VTE, although only circulating TF and D-dimers, not circulating PDPN, correlated with VTE risk. Gliomas with mutations in isocitrate dehydrogenase 1 (IDH1) or IDH2 (IDHmut) caused fewer VTEs; multivariable analysis suggested that this is due to IDHmut suppression of TF, not PDPN. In a predictive time-to-event model, the following predicted increased VTE risk in newly diagnosed patients with glioma: (1) history of VTE; (2) hypertension; (3) asthma; (4) white blood cell count; (5) WHO tumor grade; (6) patient age; and (7) body mass index. Conversely, IDHmut, hypothyroidism, and MGMT promoter methylation predicted reduced VTE risk. These 10 variables were used to create a web-based VTE prediction tool that was validated in 2 separate cohorts of patients with adult-type diffuse glioma from other institutions. This study extends our understanding of the VTE landscape in these tumors and provides evidence-based guidance for clinicians to mitigate VTE risk in patients with glioma.
Subject(s)
Brain Neoplasms , Glioma , Venous Thromboembolism , Humans , Adult , Venous Thromboembolism/genetics , Venous Thromboembolism/diagnosis , Anticoagulants/therapeutic use , Glioma/complications , Glioma/genetics , Glioma/drug therapy , Biomarkers , Brain Neoplasms/complications , Brain Neoplasms/genetics , Brain Neoplasms/pathology , Isocitrate Dehydrogenase/genetics , MutationABSTRACT
AIMS/HYPOTHESIS: Physiological gestational diabetes mellitus (GDM) subtypes that may confer different risks for adverse pregnancy outcomes have been defined. The aim of this study was to characterise the metabolome and genetic architecture of GDM subtypes to address the hypothesis that they differ between GDM subtypes. METHODS: This was a cross-sectional study of participants in the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study who underwent an OGTT at approximately 28 weeks' gestation. GDM was defined retrospectively using International Association of Diabetes and Pregnancy Study Groups/WHO criteria, and classified as insulin-deficient GDM (insulin secretion <25th percentile with preserved insulin sensitivity) or insulin-resistant GDM (insulin sensitivity <25th percentile with preserved insulin secretion). Metabolomic analyses were performed on fasting and 1 h serum samples in 3463 individuals (576 with GDM). Genome-wide genotype data were obtained for 8067 individuals (1323 with GDM). RESULTS: Regression analyses demonstrated striking differences between the metabolomes for insulin-deficient or insulin-resistant GDM compared to those with normal glucose tolerance. After adjustment for covariates, 33 fasting metabolites, including 22 medium- and long-chain acylcarnitines, were uniquely associated with insulin-deficient GDM; 23 metabolites, including the branched-chain amino acids and their metabolites, were uniquely associated with insulin-resistant GDM; two metabolites (glycerol and 2-hydroxybutyrate) were associated with the same direction of association with both subtypes. Subtype differences were also observed 1 h after a glucose load. In genome-wide association studies, variants within MTNR1B (rs10830963, p=3.43×10-18, OR 1.55) and GCKR (rs1260326, p=5.17×10-13, OR 1.43) were associated with GDM. Variants in GCKR (rs1260326, p=1.36×10-13, OR 1.60) and MTNR1B (rs10830963, p=1.22×10-9, OR 1.49) demonstrated genome-wide significant association with insulin-resistant GDM; there were no significant associations with insulin-deficient GDM. The lead SNP in GCKR, rs1260326, was associated with the levels of eight of the 25 fasting metabolites that were associated with insulin-resistant GDM and ten of 41 1 h metabolites that were associated with insulin-resistant GDM. CONCLUSIONS/INTERPRETATION: This study demonstrates that physiological GDM subtypes differ in their metabolome and genetic architecture. These findings require replication in additional cohorts, but suggest that these differences may contribute to subtype-related adverse pregnancy outcomes.
Subject(s)
Diabetes, Gestational , Hyperglycemia , Insulin Resistance , Female , Pregnancy , Humans , Blood Glucose/metabolism , Insulin Resistance/genetics , Pregnancy Outcome , Glucose Tolerance Test , Genome-Wide Association Study , Cross-Sectional Studies , Retrospective Studies , Insulin/metabolism , Glucose/metabolismABSTRACT
BACKGROUND: Higher birthweight is associated with higher adult body mass index (BMI). Alleles that predispose to greater adult adiposity might act in fetal life to increase fetal growth and birthweight. Whether there are fetal effects of recently identified adult metabolically favorable adiposity alleles on birthweight is unknown. AIM: We aimed to test the effect on birthweight of fetal genetic predisposition to higher metabolically favorable adult adiposity and compare that with the effect of fetal genetic predisposition to higher adult BMI. METHODS: We used published genome wide association study data (n = upto 406 063) to estimate fetal effects on birthweight (adjusting for maternal genotype) of alleles known to raise metabolically favorable adult adiposity or BMI. We combined summary data across single nucleotide polymorphisms (SNPs) with random effects meta-analyses. We performed weighted linear regression of SNP-birthweight effects against SNP-adult adiposity effects to test for a dose-dependent association. RESULTS: Fetal genetic predisposition to higher metabolically favorable adult adiposity and higher adult BMI were both associated with higher birthweight (3 g per effect allele (95% CI: 1-5) averaged over 14 SNPs; P = 0.002; 0.5 g per effect allele (95% CI: 0-1) averaged over 76 SNPs; P = 0.042, respectively). SNPs with greater effects on metabolically favorable adiposity tended to have greater effects on birthweight (R2 = 0.2912, P = 0.027). There was no dose-dependent association for BMI (R2 = -0.0019, P = 0.602). CONCLUSIONS: Fetal genetic predisposition to both higher adult metabolically favorable adiposity and BMI is associated with birthweight. Fetal effects of metabolically favorable adiposity-raising alleles on birthweight are modestly proportional to their effects on future adiposity, but those of BMI-raising alleles are not.
Subject(s)
Adiposity , Genome-Wide Association Study , Adiposity/genetics , Adult , Alleles , Birth Weight/genetics , Body Mass Index , Genetic Predisposition to Disease , Humans , Obesity/genetics , Polymorphism, Single Nucleotide/geneticsABSTRACT
BACKGROUND: Higher maternal pre-pregnancy body mass index (BMI) is associated with adverse pregnancy and perinatal outcomes. However, whether these associations are causal remains unclear. METHODS: We explored the relation of maternal pre-/early-pregnancy BMI with 20 pregnancy and perinatal outcomes by integrating evidence from three different approaches (i.e. multivariable regression, Mendelian randomisation, and paternal negative control analyses), including data from over 400,000 women. RESULTS: All three analytical approaches supported associations of higher maternal BMI with lower odds of maternal anaemia, delivering a small-for-gestational-age baby and initiating breastfeeding, but higher odds of hypertensive disorders of pregnancy, gestational hypertension, preeclampsia, gestational diabetes, pre-labour membrane rupture, induction of labour, caesarean section, large-for-gestational age, high birthweight, low Apgar score at 1 min, and neonatal intensive care unit admission. For example, higher maternal BMI was associated with higher risk of gestational hypertension in multivariable regression (OR = 1.67; 95% CI = 1.63, 1.70 per standard unit in BMI) and Mendelian randomisation (OR = 1.59; 95% CI = 1.38, 1.83), which was not seen for paternal BMI (OR = 1.01; 95% CI = 0.98, 1.04). Findings did not support a relation between maternal BMI and perinatal depression. For other outcomes, evidence was inconclusive due to inconsistencies across the applied approaches or substantial imprecision in effect estimates from Mendelian randomisation. CONCLUSIONS: Our findings support a causal role for maternal pre-/early-pregnancy BMI on 14 out of 20 adverse pregnancy and perinatal outcomes. Pre-conception interventions to support women maintaining a healthy BMI may reduce the burden of obstetric and neonatal complications. FUNDING: Medical Research Council, British Heart Foundation, European Research Council, National Institutes of Health, National Institute for Health Research, Research Council of Norway, Wellcome Trust.
Subject(s)
Diabetes, Gestational , Hypertension, Pregnancy-Induced , Pre-Eclampsia , Female , Humans , Infant, Newborn , Pregnancy , Body Mass Index , Cesarean Section , Hypertension, Pregnancy-Induced/epidemiology , Pre-Eclampsia/epidemiology , Mendelian Randomization AnalysisABSTRACT
Gaussian graphical models (GGMs) are a popular form of network model in which nodes represent features in multivariate normal data and edges reflect conditional dependencies between these features. GGM estimation is an active area of research. Currently available tools for GGM estimation require investigators to make several choices regarding algorithms, scoring criteria, and tuning parameters. An estimated GGM may be highly sensitive to these choices, and the accuracy of each method can vary based on structural characteristics of the network such as topology, degree distribution, and density. Because these characteristics are a priori unknown, it is not straightforward to establish universal guidelines for choosing a GGM estimation method. We address this problem by introducing SpiderLearner, an ensemble method that constructs a consensus network from multiple estimated GGMs. Given a set of candidate methods, SpiderLearner estimates the optimal convex combination of results from each method using a likelihood-based loss function. K $$ K $$ -fold cross-validation is applied in this process, reducing the risk of overfitting. In simulations, SpiderLearner performs better than or comparably to the best candidate methods according to a variety of metrics, including relative Frobenius norm and out-of-sample likelihood. We apply SpiderLearner to publicly available ovarian cancer gene expression data including 2013 participants from 13 diverse studies, demonstrating our tool's potential to identify biomarkers of complex disease. SpiderLearner is implemented as flexible, extensible, open-source code in the R package ensembleGGM at https://github.com/katehoffshutta/ensembleGGM.
Subject(s)
Algorithms , Normal Distribution , Humans , Likelihood Functions , Software , Gene Expression , Ovarian Neoplasms/geneticsABSTRACT
BACKGROUND : Construction of networks from cross-sectional biological data is increasingly common. Many recent methods have been based on Gaussian graphical modeling, and prioritize estimation of conditional pairwise dependencies among nodes in the network. However, challenges remain on how specific paths through the resultant network contribute to overall 'network-level' correlations. For biological applications, understanding these relationships is particularly relevant for parsing structural information contained in complex subnetworks. RESULTS: We propose the pair-path subscore (PPS), a method for interpreting Gaussian graphical models at the level of individual network paths. The scoring is based on the relative importance of such paths in determining the Pearson correlation between their terminal nodes. PPS is validated using human metabolomics data from the Hyperglycemia and adverse pregnancy outcome (HAPO) study, with observations confirming well-documented biological relationships among the metabolites. We also highlight how the PPS can be used in an exploratory fashion to generate new biological hypotheses. Our method is implemented in the R package pps, available at https://github.com/nathan-gill/pps . CONCLUSIONS: The PPS can be used to probe network structure on a finer scale by investigating which paths in a potentially intricate topology contribute most substantially to marginal behavior. Adding PPS to the network analysis toolkit may enable researchers to ask new questions about the relationships among nodes in network data.
Subject(s)
Blood Glucose , Hyperglycemia , Cross-Sectional Studies , Female , Humans , Normal Distribution , Pregnancy , Pregnancy OutcomeABSTRACT
Gaussian graphical models (GGMs) provide a framework for modeling conditional dependencies in multivariate data. In this tutorial, we provide an overview of GGM theory and a demonstration of various GGM tools in R. The mathematical foundations of GGMs are introduced with the goal of enabling the researcher to draw practical conclusions by interpreting model results. Background literature is presented, emphasizing methods recently developed for high-dimensional applications such as genomics, proteomics, or metabolomics. The application of these methods is illustrated using a publicly available dataset of gene expression profiles from 578 participants with ovarian cancer in The Cancer Genome Atlas. Stand-alone code for the demonstration is available as an RMarkdown file at https://github.com/katehoffshutta/ggmTutorial.
Subject(s)
Genomics , Humans , Normal DistributionABSTRACT
BACKGROUND: Ventricular assist device simulation-based mastery learning (SBML) results in better patient and caregiver self-care skills compared with usual training. OBJECTIVE: The aim of this study was to evaluate the effect of SBML on driveline exit site infections. METHODS: We compared the probability of remaining infection free at 3 and 12 months between patients randomized to SBML or usual training. RESULTS: The SBML-training group had no infections at 3 months and 2 infections at 12 months, yielding a Kaplan-Meier estimate of the probability of remaining infection free of 0.857 (95% confidence interval [CI], 0.692-1.00) at 12 months. The usual-training group had 6 infections at 3 months with no additional infections by 12 months. Kaplan-Meier estimates of remaining infection free at 3 and 12 months were 0.878 (95% CI, 0.758-1.00) and 0.748 (95% CI, 0.591-0.946), respectively. Time-to-infection distributions for SBML versus usual training showed a difference in 12-month infection rates of 0.109 (P = .07). CONCLUSIONS: Ventricular assist device self-care SBML resulted in fewer 12-month infections.
Subject(s)
Heart Failure , Heart-Assist Devices , Prosthesis-Related Infections , Heart Failure/therapy , Humans , Pilot Projects , Self CareABSTRACT
AIMS/HYPOTHESIS: We aimed to examine associations of newborn anthropometric measures with childhood glucose metabolism with the hypothesis that greater newborn birthweight, adiposity and cord C-peptide are associated with higher childhood glucose levels and lower insulin sensitivity. METHODS: Data from the international, multi-ethnic, population-based Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study and the HAPO Follow-Up Study were used. The analytic cohort included 4155 children (mean age [SD], 11.4 [1.2] years; 51.0% male). Multiple linear regression was used to examine associations of primary predictors, birthweight, newborn sum of skinfolds (SSF) and cord C-peptide, from HAPO with continuous child glucose outcomes from the HAPO Follow-Up Study. RESULTS: In an initial model that included family history of diabetes and maternal BMI during pregnancy, birthweight and SSF demonstrated a significant, inverse association with 30 min and 1 h plasma glucose levels. In the primary model, which included further adjustment for maternal sum of glucose z scores from an oral glucose tolerance test during pregnancy, the associations were strengthened, and birthweight and SSF were inversely associated with fasting, 30 min, 1 h and 2 h plasma glucose levels. Birthweight and SSF were also associated with higher insulin sensitivity (Matsuda index) (ß = 1.388; 95% CI 0.870, 1.906; p < 0.001; ß = 0.792; 95% CI 0.340, 1.244; p < 0.001, for birthweight and SSF higher by 1 SD, respectively) in the primary model, while SSF, but not birthweight, was positively associated with the disposition index, a measure of beta cell compensation for insulin resistance (ß = 0.034; 95% CI 0.012, 0.056; p = 0.002). Cord C-peptide levels were inversely associated with Matsuda index (ß = -0.746; 95% CI -1.188, -0.304; p < 0.001 for cord C-peptide higher by 1 SD) in the primary model. CONCLUSIONS/INTERPRETATION: This study demonstrates that higher birthweight and SSF are associated with greater childhood insulin sensitivity and lower glucose levels following a glucose load, associations that were further strengthened after adjustment for maternal glucose levels during pregnancy. Graphical abstract.
Subject(s)
Adiposity , Birth Weight , Blood Glucose/metabolism , C-Peptide/blood , Fetal Blood/metabolism , Hyperglycemia/blood , Insulin Resistance , Prenatal Exposure Delayed Effects , Adult , Age Factors , Biomarkers/blood , Child , Female , Follow-Up Studies , Humans , Hyperglycemia/diagnosis , Hyperglycemia/physiopathology , Infant, Newborn , Male , Pregnancy , Prospective Studies , Risk Assessment , Risk Factors , Skinfold Thickness , Young AdultABSTRACT
AIMS/HYPOTHESIS: Higher maternal BMI during pregnancy is associated with higher offspring birthweight, but it is not known whether this is solely the result of adverse metabolic consequences of higher maternal adiposity, such as maternal insulin resistance and fetal exposure to higher glucose levels, or whether there is any effect of raised adiposity through non-metabolic (e.g. mechanical) factors. We aimed to use genetic variants known to predispose to higher adiposity, coupled with a favourable metabolic profile, in a Mendelian randomisation (MR) study comparing the effect of maternal 'metabolically favourable adiposity' on offspring birthweight with the effect of maternal general adiposity (as indexed by BMI). METHODS: To test the causal effects of maternal metabolically favourable adiposity or general adiposity on offspring birthweight, we performed two-sample MR. We used variants identified in large, published genetic-association studies as being associated with either higher adiposity and a favourable metabolic profile, or higher BMI (n = 442,278 and n = 322,154 for metabolically favourable adiposity and BMI, respectively). We then extracted data on the metabolically favourable adiposity and BMI variants from a large, published genetic-association study of maternal genotype and offspring birthweight controlling for fetal genetic effects (n = 406,063 with maternal and/or fetal genotype effect estimates). We used several sensitivity analyses to test the reliability of the results. As secondary analyses, we used data from four cohorts (total n = 9323 mother-child pairs) to test the effects of maternal metabolically favourable adiposity or BMI on maternal gestational glucose, anthropometric components of birthweight and cord-blood biomarkers. RESULTS: Higher maternal adiposity with a favourable metabolic profile was associated with lower offspring birthweight (-94 [95% CI -150, -38] g per 1 SD [6.5%] higher maternal metabolically favourable adiposity, p = 0.001). By contrast, higher maternal BMI was associated with higher offspring birthweight (35 [95% CI 16, 53] g per 1 SD [4 kg/m2] higher maternal BMI, p = 0.0002). Sensitivity analyses were broadly consistent with the main results. There was evidence of outlier SNPs for both exposures; their removal slightly strengthened the metabolically favourable adiposity estimate and made no difference to the BMI estimate. Our secondary analyses found evidence to suggest that a higher maternal metabolically favourable adiposity decreases pregnancy fasting glucose levels while a higher maternal BMI increases them. The effects on neonatal anthropometric traits were consistent with the overall effect on birthweight but the smaller sample sizes for these analyses meant that the effects were imprecisely estimated. We also found evidence to suggest that higher maternal metabolically favourable adiposity decreases cord-blood leptin while higher maternal BMI increases it. CONCLUSIONS/INTERPRETATION: Our results show that higher adiposity in mothers does not necessarily lead to higher offspring birthweight. Higher maternal adiposity can lead to lower offspring birthweight if accompanied by a favourable metabolic profile. DATA AVAILABILITY: The data for the genome-wide association studies (GWAS) of BMI are available at https://portals.broadinstitute.org/collaboration/giant/index.php/GIANT_consortium_data_files . The data for the GWAS of body fat percentage are available at https://walker05.u.hpc.mssm.edu .
Subject(s)
Adiposity , Genome-Wide Association Study , Adiposity/genetics , Birth Weight , Body Mass Index , Female , Humans , Infant, Newborn , Pregnancy , Reproducibility of ResultsABSTRACT
MOTIVATION: High-throughput reporter assays dramatically improve our ability to assign function to noncoding genetic variants, by measuring allelic effects on gene expression in the controlled setting of a reporter gene. Unlike genetic association tests, such assays are not confounded by linkage disequilibrium when loci are independently assayed. These methods can thus improve the identification of causal disease mutations. While work continues on improving experimental aspects of these assays, less effort has gone into developing methods for assessing the statistical significance of assay results, particularly in the case of rare variants captured from patient DNA. RESULTS: We describe a Bayesian hierarchical model, called Bayesian Inference of Regulatory Differences, which integrates prior information and explicitly accounts for variability between experimental replicates. The model produces substantially more accurate predictions than existing methods when allele frequencies are low, which is of clear advantage in the search for disease-causing variants in DNA captured from patient cohorts. Using the model, we demonstrate a clear tradeoff between variant sequencing coverage and numbers of biological replicates, and we show that the use of additional biological replicates decreases variance in estimates of effect size, due to the properties of the Poisson-binomial distribution. We also provide a power and sample size calculator, which facilitates decision making in experimental design parameters. AVAILABILITY AND IMPLEMENTATION: The software is freely available from www.geneprediction.org/bird. The experimental design web tool can be accessed at http://67.159.92.22:8080. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
Subject(s)
Software , Alleles , Bayes Theorem , Gene Frequency , Humans , Linkage DisequilibriumABSTRACT
PURPOSE: We examined the demographic and clinicopathological parameters associated with the time to convert from active surveillance to treatment among men with prostate cancer. MATERIALS AND METHODS: A multi-institutional cohort of 7,279 patients managed with active surveillance had data and biospecimens collected for germline genetic analyses. RESULTS: Of 6,775 men included in the analysis, 2,260 (33.4%) converted to treatment at a median followup of 6.7 years. Earlier conversion was associated with higher Gleason grade groups (GG2 vs GG1 adjusted hazard ratio [aHR] 1.57, 95% CI 1.36-1.82; ≥GG3 vs GG1 aHR 1.77, 95% CI 1.29-2.43), serum prostate specific antigen concentrations (aHR per 5 ng/ml increment 1.18, 95% CI 1.11-1.25), tumor stages (cT2 vs cT1 aHR 1.58, 95% CI 1.41-1.77; ≥cT3 vs cT1 aHR 4.36, 95% CI 3.19-5.96) and number of cancerous biopsy cores (3 vs 1-2 cores aHR 1.59, 95% CI 1.37-1.84; ≥4 vs 1-2 cores aHR 3.29, 95% CI 2.94-3.69), and younger age (age continuous per 5-year increase aHR 0.96, 95% CI 0.93-0.99). Patients with high-volume GG1 tumors had a shorter interval to conversion than those with low-volume GG1 tumors and behaved like the higher-risk patients. We found no significant association between the time to conversion and self-reported race or genetic ancestry. CONCLUSIONS: A shorter time to conversion from active surveillance to treatment was associated with higher-risk clinicopathological tumor features. Furthermore, patients with high-volume GG1 tumors behaved similarly to those with intermediate and high-risk tumors. An exploratory analysis of self-reported race and genetic ancestry revealed no association with the time to conversion.
Subject(s)
Prostatectomy/statistics & numerical data , Prostatic Neoplasms/therapy , Watchful Waiting/statistics & numerical data , Aged , Biopsy, Large-Core Needle/statistics & numerical data , Disease Progression , Follow-Up Studies , Humans , Kallikreins/blood , Male , Middle Aged , Neoplasm Grading , Neoplasm Staging , Prostate/pathology , Prostate/surgery , Prostate-Specific Antigen/blood , Prostatic Neoplasms/blood , Prostatic Neoplasms/diagnosis , Prostatic Neoplasms/pathology , Risk Assessment/statistics & numerical data , Risk Factors , Time Factors , Tumor BurdenABSTRACT
BACKGROUND: The American Heart Association's formal characterization of cardiovascular health combines several metrics in a health-oriented, rather than disease-oriented, framework. Although cardiovascular health assessment during pregnancy has been recommended, its significance for pregnancy outcomes is unknown. OBJECTIVE: The purpose of this study was to examine the association of gestational cardiovascular health-formally characterized by a combination of 5 metrics-with adverse maternal and newborn outcomes. STUDY DESIGN: We analyzed data from the Hyperglycemia and Adverse Pregnancy Outcome study, including 2304 mother-newborn dyads from 6 countries. Maternal cardiovascular health was defined by the combination of the following 5 metrics measured at a mean of 28 (24-32) weeks' gestation: body mass index, blood pressure, lipids, glucose, and smoking. Levels of each metric were categorized using pregnancy guidelines, and the total cardiovascular health was scored (0-10 points, where 10 was the most favorable). Cord blood was collected at delivery, newborn anthropometrics were measured within 72 hours, and medical records were abstracted for obstetrical outcomes. Modified Poisson and multinomial logistic regression were used to test the associations of gestational cardiovascular health with pregnancy outcomes, adjusted for center and maternal and newborn characteristics. RESULTS: The average age of women at study exam was 29.6 years old, and they delivered at a mean gestational age of 39.8 weeks. The mean total gestational cardiovascular health score was 8.6 (of 10); 36.3% had all ideal metrics and 7.5% had 2+ poor metrics. In fully adjusted models, each 1 point higher (more favorable) cardiovascular health score was associated with lower risks for preeclampsia (relative risk, 0.67 [95% confidence interval, 0.61-0.73]), unplanned primary cesarean delivery (0.88 [0.82-0.95]), newborn birthweight >90th percentile (0.81 [0.75-0.87]), sum of skinfolds >90th percentile (0.84 [0.77-0.92]), and insulin sensitivity <10th percentile (0.83 [0.77-0.90]). Cardiovascular health categories demonstrated graded associations with outcomes; for example, relative risks (95% confidence intervals) for preeclampsia were 3.13 (1.39-7.06), 5.34 (2.44-11.70), and 9.30 (3.95-21.86) for women with ≥1 intermediate, 1 poor, or ≥2 poor (vs all ideal) metrics, respectively. CONCLUSION: More favorable cardiovascular health at 24 to 32 weeks' gestation was associated with lower risks for several adverse pregnancy outcomes in a multinational cohort.
Subject(s)
Birth Weight , Blood Glucose/metabolism , Blood Pressure , Body Mass Index , Cesarean Section/statistics & numerical data , Pre-Eclampsia/epidemiology , Smoking/epidemiology , Triglycerides/metabolism , Adult , Cohort Studies , Female , Glucose Tolerance Test , Heart Disease Risk Factors , Humans , Infant, Newborn , Insulin Resistance , Male , Pregnancy , Pregnancy Outcome , Pregnancy Trimester, Second , Pregnancy Trimester, Third , Skinfold Thickness , Young AdultABSTRACT
Importance: Pregnancy may be a key window to optimize cardiovascular health (CVH) for the mother and influence lifelong CVH for her child. Objective: To examine associations between maternal gestational CVH and offspring CVH. Design, Setting, and Participants: This cohort study used data from the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study (examinations: July 2000-April 2006) and HAPO Follow-Up Study (examinations: February 2013-December 2016). The analyses included 2302 mother-child dyads, comprising 48% of HAPO Follow-Up Study participants, in an ancillary CVH study. Participants were from 9 field centers across the United States, Barbados, United Kingdom, China, Thailand, and Canada. Exposures: Maternal gestational CVH at a target of 28 weeks' gestation, based on 5 metrics: body mass index, blood pressure, total cholesterol level, glucose level, and smoking. Each metric was categorized as ideal, intermediate, or poor using pregnancy guidelines. Total CVH was categorized as follows: all ideal metrics, 1 or more intermediate (but 0 poor) metrics, 1 poor metric, or 2 or more poor metrics. Main Outcomes and Measures: Offspring CVH at ages 10 to 14 years, based on 4 metrics: body mass index, blood pressure, total cholesterol level, and glucose level. Total CVH was categorized as for mothers. Results: Among 2302 dyads, the mean (SD) ages were 29.6 (2.7) years for pregnant mothers and 11.3 (1.1) years for children. During pregnancy, the mean (SD) maternal CVH score was 8.6 (1.4) out of 10. Among pregnant mothers, the prevalence of all ideal metrics was 32.8% (95% CI, 30.6%-35.1%), 31.7% (95% CI, 29.4%-34.0%) for 1 or more intermediate metrics, 29.5% (95% CI, 27.2%-31.7%) for 1 poor metric, and 6.0% (95% CI, 3.8%-8.3%) for 2 or more poor metrics. Among children of mothers with all ideal metrics, the prevalence of all ideal metrics was 42.2% (95% CI, 38.4%-46.2%), 36.7% (95% CI, 32.9%-40.7%) for 1 or more intermediate metrics, 18.4% (95% CI, 14.6%-22.4%) for 1 poor metric, and 2.6% (95% CI, 0%-6.6%) for 2 or more poor metrics. Among children of mothers with 2 or more poor metrics, the prevalence of all ideal metrics was 30.7% (95% CI, 22.0%-40.4%), 28.3% (95% CI, 19.7%-38.1%) for 1 or more intermediate metrics, 30.7% (95% CI, 22.0%-40.4%) for 1 poor metric, and 10.2% (95% CI, 1.6%-20.0%) for 2 or more poor metrics. The adjusted relative risks associated with 1 or more intermediate, 1 poor, and 2 or more poor (vs all ideal) metrics, respectively, in mothers during pregnancy were 1.17 (95% CI, 0.96-1.42), 1.66 (95% CI, 1.39-1.99), and 2.02 (95% CI, 1.55-2.64) for offspring to have 1 poor (vs all ideal) metrics, and the relative risks were 2.15 (95% CI, 1.23-3.75), 3.32 (95% CI,1.96-5.62), and 7.82 (95% CI, 4.12-14.85) for offspring to have 2 or more poor (vs all ideal) metrics. Additional adjustment for categorical birth factors (eg, preeclampsia) did not fully explain these significant associations (eg, relative risk for association between 2 or more poor metrics among mothers during pregnancy and 2 or more poor metrics among offspring after adjustment for an extended set of birth factors, 6.23 [95% CI, 3.03-12.82]). Conclusions and Relevance: In this multinational cohort, better maternal CVH at 28 weeks' gestation was significantly associated with better offspring CVH at ages 10 to 14 years.
Subject(s)
Adolescent Health , Cardiovascular System , Child Health , Heart Disease Risk Factors , Maternal Health , Pregnancy , Adolescent , Adult , Cardiovascular Diseases/epidemiology , Cardiovascular Diseases/prevention & control , Child , Cohort Studies , Female , Health Behavior , Humans , Male , PrevalenceABSTRACT
AIMS/HYPOTHESIS: Our study aimed to integrate maternal metabolic and genetic data related to insulin sensitivity during pregnancy to provide novel insights into mechanisms underlying pregnancy-induced insulin resistance. METHODS: Fasting and 1 h serum samples were collected from women in the Hyperglycemia and Adverse Pregnancy Outcome study who underwent an OGTT at â¼28 weeks' gestation. We obtained targeted and non-targeted metabolomics and genome-wide association data from 1600 and 4528 mothers, respectively, in four ancestry groups (Northern European, Afro-Caribbean, Mexican American and Thai); 1412 of the women had both metabolomics and genome-wide association data. Insulin sensitivity was calculated using a modified insulin sensitivity index that included fasting and 1 h glucose and C-peptide levels after a 75 g glucose load. RESULTS: Per-metabolite and network analyses across the four ancestries identified numerous metabolites associated with maternal insulin sensitivity before and 1 h after a glucose load, ranging from amino acids and carbohydrates to fatty acids and lipids. Genome-wide association analyses identified 12 genetic variants in the glucokinase regulatory protein gene locus that were significantly associated with maternal insulin sensitivity, including a common functional missense mutation, rs1260326 (ß = -0.2004, p = 4.67 × 10-12 in a meta-analysis across the four ancestries). This SNP was also significantly associated with multiple fasting and 1 h metabolites during pregnancy, including fasting and 1 h triacylglycerols and 2-hydroxybutyrate and 1 h lactate, 2-ketoleucine/ketoisoleucine and palmitoleic acid. Mediation analysis suggested that 1 h palmitoleic acid contributes, in part, to the association of rs1260326 with maternal insulin sensitivity, explaining 13.7% (95% CI 4.0%, 23.3%) of the total effect. CONCLUSIONS/INTERPRETATION: The present study demonstrates commonalities between metabolites and genetic variants associated with insulin sensitivity in the gravid and non-gravid states and provides insights into mechanisms underlying pregnancy-induced insulin resistance. Graphical abstract.
Subject(s)
Insulin Resistance/genetics , Metabolomics , Pregnancy/genetics , Adaptor Proteins, Signal Transducing/genetics , Adult , Asian People , Black People , Diabetes, Gestational/genetics , Diabetes, Gestational/metabolism , Female , Genome-Wide Association Study , Glucose Tolerance Test , Humans , Insulin Resistance/physiology , Mediation Analysis , Mexican Americans , Mutation, Missense , Polymorphism, Single Nucleotide , Pregnancy/metabolism , White People , Young AdultABSTRACT
Genome-wide association studies of birth weight have focused on fetal genetics, whereas relatively little is known about the role of maternal genetic variation. We aimed to identify maternal genetic variants associated with birth weight that could highlight potentially relevant maternal determinants of fetal growth. We meta-analysed data on up to 8.7 million SNPs in up to 86 577 women of European descent from the Early Growth Genetics (EGG) Consortium and the UK Biobank. We used structural equation modelling (SEM) and analyses of mother-child pairs to quantify the separate maternal and fetal genetic effects. Maternal SNPs at 10 loci (MTNR1B, HMGA2, SH2B3, KCNAB1, L3MBTL3, GCK, EBF1, TCF7L2, ACTL9, CYP3A7) were associated with offspring birth weight at P < 5 × 10-8. In SEM analyses, at least 7 of the 10 associations were consistent with effects of the maternal genotype acting via the intrauterine environment, rather than via effects of shared alleles with the fetus. Variants, or correlated proxies, at many of the loci had been previously associated with adult traits, including fasting glucose (MTNR1B, GCK and TCF7L2) and sex hormone levels (CYP3A7), and one (EBF1) with gestational duration. The identified associations indicate that genetic effects on maternal glucose, cytochrome P450 activity and gestational duration, and potentially on maternal blood pressure and immune function, are relevant for fetal growth. Further characterization of these associations in mechanistic and causal analyses will enhance understanding of the potentially modifiable maternal determinants of fetal growth, with the goal of reducing the morbidity and mortality associated with low and high birth weights.
Subject(s)
Birth Weight/genetics , Genome-Wide Association Study/methods , Polymorphism, Single Nucleotide/genetics , Actins/genetics , Adaptor Proteins, Signal Transducing , Alleles , Birth Weight/physiology , Cytochrome P-450 CYP3A/genetics , DNA-Binding Proteins/genetics , Female , Genetic Variation/genetics , Genotype , Germinal Center Kinases , Gestational Age , HMGA2 Protein/genetics , Humans , Intracellular Signaling Peptides and Proteins , Kv1.3 Potassium Channel/genetics , Protein Serine-Threonine Kinases/genetics , Proteins/genetics , Receptor, Melatonin, MT2/genetics , Trans-Activators/genetics , Transcription Factor 7-Like 2 Protein/geneticsABSTRACT
AIMS/HYPOTHESIS: We aimed to determine the association of maternal metabolites with newborn adiposity and hyperinsulinaemia in a multi-ethnic cohort of mother-newborn dyads. METHODS: Targeted and non-targeted metabolomics assays were performed on fasting and 1 h serum samples from a total of 1600 mothers in four ancestry groups (Northern European, Afro-Caribbean, Mexican American and Thai) who participated in the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study, underwent an OGTT at ~28 weeks gestation and whose newborns had anthropometric measurements at birth. RESULTS: In this observational study, meta-analyses demonstrated significant associations of maternal fasting and 1 h metabolites with birthweight, cord C-peptide and/or sum of skinfolds across ancestry groups. In particular, maternal fasting triacylglycerols were associated with newborn sum of skinfolds. At 1 h, several amino acids, fatty acids and lipid metabolites were associated with one or more newborn outcomes. Network analyses revealed clusters of fasting acylcarnitines, amino acids, lipids and fatty acid metabolites associated with cord C-peptide and sum of skinfolds, with the addition of branched-chain and aromatic amino acids at 1 h. CONCLUSIONS/INTERPRETATION: The maternal metabolome during pregnancy is associated with newborn outcomes. Maternal levels of amino acids, acylcarnitines, lipids and fatty acids and their metabolites during pregnancy relate to fetal growth, adiposity and cord C-peptide, independent of maternal BMI and blood glucose levels.
Subject(s)
Birth Weight/physiology , Hyperinsulinism/metabolism , Metabolome , Adult , C-Peptide/blood , Female , Glucose Tolerance Test , Humans , Infant, Newborn , Male , Metabolomics , Pregnancy , Pregnancy Outcome , Triglycerides/bloodABSTRACT
AIMS/HYPOTHESIS: Maternal type 2 diabetes during pregnancy and gestational diabetes are associated with childhood adiposity; however, associations of lower maternal glucose levels during pregnancy with childhood adiposity, independent of maternal BMI, remain less clear. The objective was to examine associations of maternal glucose levels during pregnancy with childhood adiposity in the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) cohort. METHODS: The HAPO Study was an observational epidemiological international multi-ethnic investigation that established strong associations of glucose levels during pregnancy with multiple adverse perinatal outcomes. The HAPO Follow-up Study (HAPO FUS) included 4832 children from ten HAPO centres whose mothers had a 75 g OGTT at ~28 weeks gestation 10-14 years earlier, with glucose values blinded to participants and clinical caregivers. The primary outcome was child adiposity, including: (1) being overweight/obese according to sex- and age-specific cut-offs based on the International Obesity Task Force (IOTF) criteria; (2) IOTF-defined obesity only; and (3) measurements >85th percentile for sum of skinfolds, waist circumference and per cent body fat. Primary predictors were maternal OGTT and HbA1c values during pregnancy. RESULTS: Fully adjusted models that included maternal BMI at pregnancy OGTT indicated positive associations between maternal glucose predictors and child adiposity outcomes. For one SD difference in pregnancy glucose and HbA1c measures, ORs for each child adiposity outcome were in the range of 1.05-1.16 for maternal fasting glucose, 1.11-1.19 for 1 h glucose, 1.09-1.21 for 2 h glucose and 1.12-1.21 for HbA1c. Associations were significant, except for associations of maternal fasting glucose with offspring being overweight/obese or having waist circumference >85th percentile. Linearity was confirmed in all adjusted models. Exploratory sex-specific analyses indicated generally consistent associations for boys and girls. CONCLUSIONS/INTERPRETATION: Exposure to higher levels of glucose in utero is independently associated with childhood adiposity, including being overweight/obese, obesity, skinfold thickness, per cent body fat and waist circumference. Glucose levels less than those diagnostic of diabetes are associated with greater childhood adiposity; this may have implications for long-term metabolic health.
Subject(s)
Adiposity , Blood Glucose/analysis , Diabetes, Gestational/blood , Hyperglycemia/blood , Pediatric Obesity/physiopathology , Pregnancy in Diabetics/blood , Prenatal Exposure Delayed Effects/blood , Adult , Body Mass Index , Child , Female , Follow-Up Studies , Glucose Tolerance Test , Humans , Male , Maternal Age , Overweight , Pregnancy , Pregnancy Complications , Pregnancy Outcome , Prenatal Exposure Delayed Effects/physiopathology , Waist CircumferenceABSTRACT
Newborn adiposity, a nutritional measure of the maternal-fetal intra-uterine environment, is representative of future metabolic health. An anthropometric model using weight, length and flank skinfold to estimate neonatal fat mass has been used in numerous epidemiological studies. Air displacement plethysmography (ADP), a non-invasive technology to measure body composition, is impractical for large epidemiological studies. The study objective was to determine the consistency of the original anthropometric fat mass estimation equation with ADP. Full-term neonates were studied at 12-72 h of life with weight, length, head circumference, flank skinfold thickness and ADP measurements. Statistical analyses evaluated three models to predict neonatal fat mass. Lin's concordance correlation coefficient, mean prediction error and root mean squared error between the predicted and observed ADP fat mass values were used to evaluate the models, where ADP was considered the gold standard. A multi-ethnic cohort of 468 neonates were studied. Models (M) for predicting fat mass were developed using 349 neonates from site 1, then independently evaluated in 119 neonates from site 2. M0 was the original anthropometric model, M1 used the same variables as M0 but with updated parameters and M2 additionally included head circumference. In the independent validation cohort, Lin's concordance correlation estimates demonstrated reasonable accuracy (model 0: 0·843, 1: 0·732, 2: 0·747). Mean prediction error and root mean squared error in the independent validation was much smaller for M0 compared with M1 and M2. The original anthropometric model to estimate neonatal fat mass is reasonable for predicting ADP, thus we advocate its continued use in epidemiological studies.