Analgosedative prescribing and loading dose strategies in the neonatal intensive care unit: a retrospective cohort study with a focus on hypothermia-treated and extremely premature neonates

Abstract Aim Treating discomfort in neonates is complex, and protocols often lack tailored advice, risking inadequate pain management. This study aims to examine analgosedative prescribing behavior in the Neonatal Intensive Care Unit (NICU) with a focus on the use and impact of loading doses on reducing total cumulative dose and especially in the particularly vulnerable neonates receiving therapeutic hypothermia and those born extremely premature. Methods This single-center, retrospective study analyzed baseline characteristics and analgosedative administration data from all neonates admitted to our level III-IV NICU between 2017 and 2021. Results Of the 2633 admissions, 825 neonates (31%) received analgosedation. Among them, 126 (15%) received therapeutic hypothermia and 142 (17%) were born extremely premature. Loading doses were infrequently administered, ranging from 0 to 61% per continuous infusion (re)initiation and dose escalation. Across the cohort, loading doses were associated with a lower total cumulative doses for morphine, but higher for midazolam and fentanyl. Extremely premature neonates receiving morphine continuous infusion with loading doses were followed by lower median starting and maintenance dosages in the most critical first 36 h. Conclusion Loading doses did not significantly reduce total cumulative exposure to analgosedatives, even increasing it in some cases. However, in extremely premature neonates receiving morphine continuous infusion with loading doses, both the total cumulative dose and the maintenance dose in the first 36-hours were reduced. This indicates that structurally administering loading doses may improve pain management through faster therapeutic effects. Possibly, it could reduce cumulative exposure, if combined with lower initial maintenance dosages and smaller dose-escalation increments.