Sedation monitoring in premature infants

Theoretical framework Several clinical and environmental factors could alter brain development in premature infants. These patients, in fact, are exposed to a consistent number of procedures during their entire time of hospitalization. The use of analgesic and sedative drugs is essential in order to grant them a maximal level of comfort. However, the administration of such drugs is complicated by the level of physiological maturity and by the fact that this special collective of patients is still in a preverbal stage of development. Objective methods like a conventional EEG could help understand more about an infants level of sedation. However, its interpretation is time- consuming and requires a given level of expertise. Today, new methods could be used to automatically detect important EEG features. Machine learning, in fact, gives us the opportunity to recognize important EEG trends that could be further easily interpreted by the care-taking team. Hypotheses The overarching aim of this study is to use deep-machine-learning algorithms for the interpretation of suddenly changes in the EEG-background activity related to the administration of sedation; and to contextualize automatic EEG-background changes related to sedation with the clinical opinion of sedation expressed by the scoring of the Neonatal, Pain, Agitation and Sedation Scale (N-PASS). Methods In this Study, 50 preterm infants undergoing clinical procedures for which sedation is required for a short period of time (e.g. central venous catheter), will be prospectively recruited. Both the deviation from actual gestational age and IBI duration will be used to understand changes in the EEG-background activities during sedation administration. In more, trend-EEG parameters will be correlated to the clinical expert opinion of the level of sedation measured through the N- PASS. Level of originality The topic addressed in this project is of outstanding relevance in neonatology as the use of automatic EEG-trends could be useful for the automatic identification of critical neurological events, to evaluate and ameliorate sedation administration in critically ill infants.