Metabolomic analysis for the study of maturation in pediatrics: Effect of confounding factors in a pilot study

A pilot study for the investigation of the maturation grade of children has been carried out using plasma samples already analyzed in a previous pharmacokinetic study. By using a meticulous data treatment, possible confounding factors that may hinder the obtained results were identified. By doing so, it was possible to obtain enough evidence to support the feasibility of performing a larger study eluding some unwanted variability and minimizing not only the number of subjects involved but also the time and money spent on the study. In the pilot study the metabolic profiles obtained using UHPLC-TOF-MS technique of plasma samples from 14 newborn piglets (<5 days) were compared with the plasma profiles of 16 infant piglets (8 weeks). The type of anaesthesia administered, gender, vein or artery of blood extraction and time of sampling were studied as possible confounding factors. Unsupervised analysis by principal component analysis (PCA) clearly differentiated between neonates and children. During the data treatment and the statistical analysis, the effect of confounding factors such as the anaesthetic regimen was identified and removed, while the effect of the rest of studied factors was not considered relevant, and the discrimination between the two groups based on the age was maintained. This allowed extracting relevant conclusions for a future study design while avoiding the unnecessary sacrifice of animals. Furthermore, the results obtained demonstrate the utility of metabolomics in the discovery of novel putative plasma biomarkers such as carnitines that can be correlated with the maturation state of paediatric patients.