Introduction: Studies have shown the benefit of intensive care unit (ICU) bundled protocols; however, they are primarily derived from medical patients. We hypothesized that patients and their medication profiles are different between critically ill medical, surgical, and trauma patients.
Methods: The Pediatric Health Information System 2017 dataset was used to perform a retrospective cohort study of critically ill children. The pediatric medical, surgical, and trauma cohorts were separated based on ICD-10 codes. Data collected included demographics, secondary diagnoses, outcomes, and medication data. Medications were grouped as opiates, GABA-agonists, alpha-2 agonists, anti-psychotics, paralytics, and "other" sedatives. A non-parametric Kolmogorov-Smirnov test (KS test) and odds ratios (reference group: medical cohort) were calculated to compare medication administration between the study cohorts for the first 30 ICU days.
Results: A total of 4488 critically ill children (medical 2078, surgical 1650, and trauma 760) were identified. The trauma cohort had increased incidence of delirium (medical 10.8%, surgical 11.5%, trauma 13.8%; p < 0.01) and mortality (medical 5.4%, surgical 2.4%, trauma 11.7%; p < 0.01). For all study cohorts, > 50% received GABA-agonists on ICU days 0-30. With the KS test, there was a significant difference in administration of opiates, GABA-agonists, alpha-2 agonists, anti-psychotics, and "other" sedatives over the first 30 days in the ICU. Relative to medical patients, trauma patients had significantly higher odds of receiving anti-psychotics on ICU days 10-20 and 22-24.
Conclusion: Critically ill pediatric trauma, medical, and surgical patients are distinctly different patient populations with differing pharmacologic profiles for analgesia, sedation, and delirium.
Level of evidence: Level III (Retrospective Comparative Study).
Keywords: Analgesia; Children; Critically ill; Delirium; Sedation; Trauma.
© 2021. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.