462 - Cross-site comparisons of early exposure to analgesic medications in preterm neonates

Poster Number: 462
Publication Number: 462.342

Pearl Zaki, University of Toronto Temerty Faculty of Medicine, Toronto, ON, Canada; Thiviya Selvanathan, The Hospital for Sick Children, Toronto, ON, Canada; Stephanie Au-Young, The Hospital for Sick Children, Toronto, ON, Canada; Giselle Da Rocha, The Hospital for Sick Children, Toronto, ON, Canada; Cecil M Y Chau, University of British Columbia Faculty of Medicine, Vancouver, BC, Canada; Mark LePine, The Hospital for Sick Children, TORONTO, ON, Canada; Vann Chau, The Hospital for Sick Children, Toronto, ON, Canada; Steven Ufkes, The Hospital for Sick Children, Toronto, ON, Canada; Mia McLean, University of British Columbia Faculty of Medicine, Vancouver, BC, Canada; Natalie Chan, University of British Columbia Faculty of Medicine, Vancouver, BC, Canada; Linh Ly, University of Toronto Temerty Faculty of Medicine, Toronto, ON, Canada; Edmond Kelly, Mount Sinai Hospital, Toronto,, ON, Canada; Steven P. Miller, The Hospital for Sick Children, Toronto, ON, Canada; Ruth E. Grunau, University of British Columbia Faculty of Medicine, Vancouver, BC, Canada

Background: Exposure to pain-related stress, especially early in life, may contribute to altered brain development in preterm neonates, making adequate analgesia critical during this period. There is a lack of consensus around neonatal pain management.

Objective: Our study aims to characterize cross-site differences in analgesic practices across 3 tertiary NICUs.

Design/Methods: Clinical data were collected for 276 very preterm neonates at 3 sites as part of a prospective cohort study of early-life MRI that included day-by-day counts of invasive procedures. In this substudy, we focus on the time between birth and early-life MRI, quantifying number of invasive procedures as an index of pain exposure. Kruskal-Wallis and Fisher exact tests were used to compare clinical characteristics and cumulative analgesic dose between sites. Multivariate linear regression was used to examine whether morphine and fentanyl use differed by site, with the outcome of cumulative exposure to analgesia and the primary predictor as the interaction between site and number of invasive procedures. All models accounted for gestational age, days from birth to early MRI, major surgery, mechanical ventilation, infection and cumulative sucrose dose.

Results: Clinical characteristics, number of invasive procedures and analgesia dose relative to number of invasive procedures differed across sites. Infants at site 1 were not exposed to sucrose. Infants at site 3 received higher cumulative doses of morphine and fentanyl. Infants at site 2 received higher cumulative doses of midazolam, although only a small proportion of all infants received midazolam. Multivariate linear regression models were performed for morphine and fentanyl only as these had greatest practice variation across sites. There was a significant interaction between site and number of invasive procedures (p < 0.001) in predicting early cumulative morphine use with no other significant predictors. Early cumulative fentanyl exposure was predicted by the interaction between site and number of invasive procedures (p=0.03). Fentanyl use was also associated with major surgery (b=0.2, 95%CI 0.1-0.3, p< 0.001).Conclusion(s): Morphine and fentanyl use significantly varies across 3 NICUs in Canada for a given exposure to early-life pain, even when accounting for surgery, infection, and mechanical ventilation. Greater sucrose use was not associated with less use of morphine and fentanyl. These findings highlight important differences in clinical practice and the need to develop clinical guidelines to standardize neonatal pain management.
Pearl Zaki - CVCV P ZAKI_2021.pdf
Figure 1Total invasive procedures by study site (birth to early MRI) invasive procedures (including skin-breaking and non-skin-breaking) by site.