Characterization of Propofol Use in a Pediatric ICU

Introduction: SCCM 2022 guidelines recommend propofol as an adjunct sedative for pediatric ICU (PICU) patients due to associations with earlier extubation. However, prior warnings against use in children likely influence practice variability. We aimed to compare outcomes in PICU patients requiring mechanical ventilation with and without propofol for continuous sedation. We hypothesized that patients receiving propofol will have more ventilator-free days (VFDs) than those who do not.


Methods: We queried EHR from a single tertiary center for PICU patients requiring invasive mechanical ventilation from June 2018 to December 2022. We compared demographics, admission diagnoses, and clinical variables between patients who did and did not receive propofol. Surrogate markers of illness severity included Pediatric Index of Mortality score (PIM3), continuous renal replacement therapy, extracorporeal membrane oxygenation, vasoactive infusions, and continuous neuromuscular blockade. Markers with significant differences on univariate comparison were retained in the final model. Multivariable modelling investigated associations of propofol with VFDs and mortality, adjusting for admission diagnosis and retained markers of illness.


Results: Of 1,048 unique patient admissions analyzed: 483 (46%) received propofol and 565 (54%) did not. Patients receiving propofol were older than those who did not (94 months [23, 177] vs. 26 months [4, 136], p< 0.0001). Patients receiving propofol were less likely to be intubated for respiratory failure, cardiac arrest, or septic shock (p < 0.01) and more likely to be intubated for trauma and traumatic brain injury (p < 0.01). Those receiving propofol had 23 VFDs [13.2, 26.7] with 10% mortality, compared to 23.3 VFDs [14.3, 26.5] with 17.5% mortality in the propofol-free group. After adjusting for admission diagnoses and markers of illness severity, there was no difference in VFDs between groups (p=0.6), but the propofol group had lower mortality (p=0.0006).


Conclusions: There was no difference in VFDs between those who did or did not receive propofol after adjusting for underlying illness. A difference in mortality suggests that illness severity may be a possible confounder for the potential benefit of earlier extubation with propofol.