A systems biology approach: new insights into fetal growth restriction using Bayesian Networks.

IL-6, IGF-II and IGFBP-2 concentrations in placental lysates were previously shown to be associated with foetal growth. This study aimed to apply a Bayesian Network (BN) model in order to investigate complex dependencies among biochemical and clinical factors and fetal growth outcome. Twenty-one Intra-Uterine Growth Restricted (IUGR) and 25 Appropriate for Gestational Age (AGA) pregnancies were followed throughout pregnancy. Information was collected on maternal and gestational age, neonatal gender, previous gynaecological history. Total protein content, IGF-II, IGFBP-1, IGFBP-2, IL-6, and TNF-alpha concentrations in placental lysates were measured, and IGF-I, IGF-II, IGFBP-1, IGFBP-2 and IL-6 relative gene expression in placenta assessed. A BN and a hybrid forecasting system were implemented: BN revealed a key role of maternal age and TNF-alpha on IUGR and confirmed a close relationship among IGF-II, IL-6 and foetal growth. A relationship between duration of gestation, appropriateness for gestational age, and placental IL-6 concentration was also confirmed. Compared with other techniques, BN showed a better accuracy. Findings confirmed a major role of maternal age in addition to IGF-II, IL-6 and TNF-alpha in IUGR. A direct role of IGFBP-2 was not shown. BN confirmed to be useful in understanding the system's biology and graphically representing variable relationships and hierarchy, particularly where, as in IUGR, many interactions among predictors exist.