(1388) A Quality Improvement Initiative to Decrease the Prevalence of Hyperoxia in Critically Ill Children

Introduction: Studies have shown that in children receiving supplemental oxygen therapy, excessive oxygen exposure is associated with increased risk of multiple organ dysfunction and mortality. Our quality improvement (QI) initiative aims to describe the prevalence of both hypoxia and hyperoxia in children admitted to our pediatric intensive care unit (PICU) and reduce excess oxygen exposure.

Methods: Hyperoxia was defined as FiO2 >21% while saturating >96%. Hypoxia was defined as SpO2 < 90% with an FiO2 of 21-60%, or SpO2 < 88% with an FiO2 >60%. PICU staff including nurses, respiratory therapists, resident, fellows, and attending physicians were provided with a presentation describing an SpO2 goal of 92-96% in patients receiving supplemental oxygen (unless medically contraindicated, e.g. unrepaired cyanotic heart disease), and encouraging timely titration of FiO2 to meet that goal range. Baseline data was collected for all patients admitted to the PICU in the 6 months prior to the intervention, and for 2 months after during which time reminders were periodically provided by QI project leaders. Patients were included in data analysis if they received respiratory support via high-flow nasal cannula (HFNC), supplemental flow via tracheostomy, non-invasive positive pressure ventilation (NIV), or invasive mechanical ventilation (IMV). Cumulative excess oxygen exposure (CEOE) and hyperoxia prevalence were calculated for the first 24 hours of PICU stay. Data sets were compared using two-sample t-test.

Results: Pre-intervention data included 28,398 hourly SpO2 values obtained during 445 admissions. Post-intervention data included 5,331 SpO2 values from 112 admissions. Pre-intervention, the average hyperoxia prevalence across all respiratory support modes was 73.2% in the first 24 hours of admission, which decreased to 55.4% post-intervention (p < 0.001); average CEOE similarly decreased from 17.7 to 13.0 (p < 0.001). The rate of hypoxia remained the same at less than one percent.

Conclusions: Staff education improved the prevalence of hyperoxia without increasing the rate of hypoxia. However, further interventions and data collection cycles are required to determine the degree of success and sustainability of this QI initiative.