Microbial Metabolites and Outcomes of Pregnancy Study

Overview

Emerging data connect diet, the gut microbiota and its metabolites in cardiometabolic disease. Hypertensive disorders of pregnancy (HDP) are common and are a leading cause of maternal and neonatal morbidity. HDP likely share similar pathophysiology as cardiometabolic disease in non-pregnant people with a yet unrevealed role of diet and the gut microbiota, including systemic inflammation and endothelial dysfunction. Despite high biological plausibility that nutrition, the gut microbiota and its metabolites may play a role in health and disease in pregnancy, there is a paucity of data regarding these associations, thus limiting advancement of the field. Similar to the proposed pathogenesis for diet, gut microbiota and the microbial metabolite trimethylamine-N-oxide (TMAO) in cardiovascular disease, we hypothesize that the interplay between maternal diet, the gut microbiota and its associated microbial metabolites play a mechanistic role in HDP. We propose to test this hypothesis in a racially-diverse US cohort to determine association with adverse pregnancy outcomes, specifically future development of HDP. We propose to prospectively collect plasma and urine TMAO throughout pregnancy from a cohort of 200 pregnant participants. Through 1) characterizing plasma and urine TMAO levels across each trimester of pregnancy, and 2) assessment of this microbial metabolite as a predictor of development of HDP, we have the potential to identify a biomarker that would allow us to identify people at risk of HDP early in pregnancy and provide new opportunities for therapeutic interventions to improve maternal and neonatal outcomes.

Overall Plan: To evaluate maternal markers of metabolism including the maternal microbial metabolite, Trimethylamine N-oxide (TMAO), across pregnancy, and its association with development of hypertensive disorders of pregnancy. Research Design and Methods: This prospective longitudinal cohort will enroll pregnant patients in their first trimester of pregnancy (10-14 weeks gestation). Plasma, serum, blood ribonucleic acid (RNA) and urine samples will be collected at the time of enrollment, along with dietary assessment, and self-reported demographic information. The total volume of blood for research lab purposes will be 22.5 cc (one 10 ml lavender top tube, one 10 ml serum separator tube, and one 2.5 ml PAXgene blood RNA tube). If research labs are drawn in coordination with other necessary prenatal labs, the anticipated maximum volume at blood draw would be 42-50 cc of blood, depending on combination of commonly ordered prenatal labs. The total amount of blood drawn from pregnant subjects (including clinical and research draws) will not exceed 50ml in a two-week period. Plasma, serum, blood RNA and urine samples will again be collected in the second trimester, at 24-28 weeks. In addition, maternal plasma, serum, blood RNA and urine will be collected at the time of delivery admission. A 10 cc sample of cord blood will be obtained after delivery and the placenta will be sent to pathology for evaluation. The investigators will measure TMAO levels and other markers of metabolism on all three maternal samples and that of the cord blood. Deoxyribonucleic acid (DNA) from the buffy coat of plasma collection and RNA from a PAXgene blood tube will be obtained for molecular testing of metabolic biomarkers. The placenta will be examined pathologically with attention to vascular pathology. In addition, a dietary history will be obtained at each time point using the interviewer-guided ASA24 (Automated self-administered 24-hour dietary assessment tool)-a web-based survey provided at no cost through the National Institutes of Health website under a researcher-only password protected account (https://epi.grants.cancer.gov/asa24/). Data and Sample Collection: Phlebotomy and midstream urine self-collection will be performed at pre-specified time points (10-14 weeks, 24-28 weeks, and at time of admission for delivery). Placenta and cord blood samples will be collected following delivery by delivering clinician. Data and Sample Storage: Data security will be ensured by limiting access to those team members involved in the collection and analysis of patient data. Medical record data will be stored securely through Cleveland Clinic REDCap database. Dietary assessment data will be similarly stored within the web-based NIH Automated Self-Administered 24-hour (ASA24) Dietary Assessment Tool researcher password-protected account. Safety monitoring will include ensuring safe and clean clinical environment for specimen collection, following appropriate safety protocol for venipuncture, and proper disposal of supplies in designated clinical areas. Sample Size: Since the investigators have limited information regarding TMAO in pregnancy, the power calculation was estimated to detect mean TMAO differences between groups using a two sample T-test at a significance level of 0.05, 80% power and total sample size of 200 patients with 10% dropout rate. Standard deviations of TMAO estimated from the literature are ranging from 0.61 to 10.15 micromole/L, assuming the ratio of hypertensive disorders of pregnancy group to normal group is 1:10, the detectable mean TMAO differences were calculated to be 0.451 to 7.497 micromole/L. Analysis plan: Patients with hypertensive disorders of pregnancy will be compared to those without. Approximately normally-distributed continuous measures will be summarized using means and standard deviations and will be compared using two-sample t-tests. Continuous measures that show departure from normality and ordinal measures will be summarized using medians and quartiles and will be compared using Wilcoxon rank sum tests. Categorical factors will be summarized using frequencies and percentages and will be compared using Pearson's chi-square tests or Fisher's Exact tests. If groups differ on baseline characteristics, linear regression adjusting for factors of clinical importance will be performed. Depending on the distribution of final data, Pearson or Spearman correlation will be calculated between TMAO level and other continuous variables. For each level of categorical factors, TMAO level will be summarized using mean and standard deviation or median and quartiles and will be compared using analysis of variance (ANOVA) or Kruskal-Wallis tests. All analyses will be done using SAS (version 9.4, The SAS Institute, Cary, NC) and a p \< 0.05 will be considered statistically significant.