There are several directions for predicting multiple organ dysfunction syndrome (MODS), but almost all of them are poorly tested in neonatology. This review is presented to indicate the problem of the condition severity objectification of newborns and the possibility of predicting the development of MODS. Scales for assessing the severity of MODS in critically ill children have been developed and used since the end of the last century, but their validation in the newborns faces certain difficulties. Prognostic nosospecific scales: NICHD (National Institute of Child Health and Human Development) calculator, CRIB II (Clinical Risk Index for Babies), SNAPPE-II (Score for Neonatal Acute Physiology with Perinatal Extension II) are used in neonatology, however their comparison in this category of patients has not been carried out.Theoretical and practical issues of the short-term and long-term prediction of the MODS onset and its outcomes in newborns is a promising area of neonatology, since it allows a doctor to be warned about an impending catastrophe and opens a “window of opportunity” for timely correction of treatment tactics and complications prevention. Obtaining different phenotypes of critical illness and predicting their outcomes in children may have good predictive potential, but such studies have not been conducted in newborns. A promising direction in predicting MODS is the identification of biomarkers of inflammation, among which endocan, cluster of differentiation 64, cluster of differentiation molecules 11b, “pancreatic stone protein” (PSP), soluble intercellular adhesionmolecule-1 (sICAM-1), progranulin, neopterin, resistin (FIZZ3, presepsin (PSP)) carry a good potential, but their effectiveness in neonatology is still to be investigated.Thus, the prediction of MODS in children and newborns remains an unresolved problem. At the same time, several promising scientific directions are actively being developed today, which may lead to a significant breakthrough in predicting MODS in neonatology.