Elemental Metabolomics for Prediction of Term Gestational Outcomes Utilising 18-Week Maternal Plasma and Urine Samples

A normal pregnancy is essential to establishing a healthy start to life. Complications during have been associated with adverse perinatal outcomes and lifelong health problems. The ability to identify risk factors associated with pregnancy complications early in gestation is vitally important for preventing negative foetal outcomes. Maternal nutrition has been long considered vital to a healthy pregnancy, with micronutrients and trace elements heavily implicated in maternofoetal metabolism. This study proposed the use of elemental metabolomics to study multiple elements at 18 weeks gestation from blood plasma and urine to construct models that could predict outcomes such as small for gestational age (SGA) (n = 10), low placental weight (n = 18), and preterm birth (n = 13) from control samples (n = 87). Samples collected from the Lyell McEwin Hospital in Adelaide, South Australia, were measured for 27 plasma elements and 37 urine elements by inductively coupled plasma mass spectrometry. Exploratory analysis indicated an average selenium concentration 20 mu g/L lower than established reference ranges across all groups, low zinc in preterm (0.64 mu g/L, reference range 0.66-1.10 mu g/L), and higher iodine in preterm and SGA gestations (preterm 102 mu g/L, SGA 111 mu g/L, reference range 40-92 mu g/L). Using random forest algorithms with receiver operating characteristic curves, low placental weight was predicted with 86.7% accuracy using plasma, 78.6% prediction for SGA with urine, and 73.5% determination of preterm pregnancies. This study indicates that elemental metabolomic modelling could provide a means of early detection of at-risk pregnancies allowing for more targeted monitoring of mothers, with potential for early intervention strategies to be developed.

Automated summary

A study using elemental metabolomics was conducted to predict adverse outcomes in pregnancy, including small for gestational age, low placental weight, and preterm birth. It found that certain elements, such as selenium, zinc, and iodine, were associated with these outcomes. Utilizing random forest algorithms, the study showed high accuracy in predicting these adverse outcomes, indicating the potential for early detection and intervention strategies in at-risk pregnancies.