ABSTRACT: From 2010 to 2012, women were recruited into The Infant Development and the Environment Study (TIDES) from obstetrical clinics affiliated with academic medical centers in four U.S. cities. Using previously collected first and second trimester urine samples from this prospective cohort of nearly 800 pregnancies, we designed a nested case control study aimed to determine whether maternal metabolic abnormalities/differences are associated with FGR. To find the patients with fetal growth restriction (FGR), we reviewed de-identified questionnaires and de-identified previously collected data, and performed a growth potential formula considering gestational age, infant gender, maternal and paternal height, and interaction of gestational age with maternal weight. Using a case:control ratio of 1:2, matched on study site, maternal age (± 2 years), parity, and infant's sex, 53 cases were matched to 106 controls for a total of 159 patients, and 318 samples (one in each ttrimester). The samples were analyses by NMR spectroscopy, using Bruker IVDr platform. Identification of FGR cases and controls: FGR cases were determined using the AUDIPOG formula for the average predicted birthweight for an infant with specific characteristics of: maternal height, age, and prenatal weight, as well as infant sex, gestational age, and birth rank. AUDIPOG formula: avg_pred_bw = 10,228066774 - 0,646727171*GA + 0,0259713417*GA² - 0,000291122*GA³ - 0,045467351*sex + 0,0606013862*rank - 0,013592585*rank² + 0,0009109473*rank³ + 0,0003976103*MA + 0,0019992269*MH + 0,0169049061*MW - 0,000171266*MW² + 5,8340462E-7*MW³ The natural logarithm of this average predicted birthweight was used with the observed birthweight to determine a z-score and percentile for each infant. If an infant fell into the 10th percentile for his or her given characteristics, then they were considered as having FGR. Otherwise they were considered in the pool of controls.
Ivan Vuckovic 
