TOP 66 - Avoid Risk, Remember Oxygen Wean (ARROW): A Quality Improvement Initiative to Reduce Excess Oxygen Use

Publication Number: 1825.TOP 66

Danielle M. Severns, Riley Hospital for Children at Indiana University Health, Indianapolis, IN, United States; Aimee Ealy, Riley Hospital for Children at Indiana University Health, PLAINFIELD, IN, United States; Shyloh Flaherty, Riley Hospital for Children at Indiana University Health, Indianapolis, IN, United States; Erin M. Kramer, Riley Hospital for Children at IU Heath, Brownsburg, IN, United States; Leah E. Bramlett, Riley Hospital for Children at Indiana University Health, Brownsburg, IN, United States; Jason Stemple, Indiana University School of Medicine, Indianapolis, IN, United States; Kristen Ely, Indiana University School of Medicine, Fishers, IN, United States; Lee Murphy, Riley Hospital for Children at Indiana University Health, Indianapolis, IN, United States; Daniel Cater, Riley Hospital for Children at Indiana University Health, Indianapolis, IN, United States; Samer Abu-Sultaneh, Riley Hospital for Children at Indiana University Health, Indianapolis, IN, United States

Background: Hypoxemic respiratory failure is a common reason for admission to the pediatric intensive care unit (PICU). The use of supplemental oxygen is an essential intervention. However, children often receive supplemental fraction of inspired oxygen (FiO2) despite having oxygen saturations (SpO2) > 97%. This contributes to hyperoxia, which causes oxidative stress and cell damage, and is associated with increased mortality. Additionally, it is imperative to monitor impaired oxygenation in intubated patients with acute respiratory distress syndrome. An SpO2 88-97% is required to accurately assess their illness severity using the oxygen saturation index.

Objective: The SMART aim of this project is to improve the compliance with weaning FiO2 to a minimum of 21% to keep SpO2 < 97%, in intubated patients in the PICU, from 70% to 80% by January 2026.

Design/Methods: This is a single center quality improvement (QI) project that includes a multidisciplinary team of registered nurses, respiratory therapists, nurse practitioners, and physicians. It was reviewed and deemed exempt by institutional IRB as a QI study. Project implementation occurred in January 2025. Bedside providers were asked to wean FiO2 to a minimum of 21% to keep SpO2 < 97% in all intubated patients. Patients with a tracheostomy, cyanotic congenital heart disease, pulmonary hypertension, or acute traumatic brain injury were excluded. A Key Driver Diagram identified key drivers associated with our SMART aim (Figure 1). Our process measure was compliance with FiO2 weaning. Outcome measures included average SpO2, FiO2, and frequency of SpO2 > 97%.
Baseline data was collected from the electronic medical record twice weekly for three months prior to project implementation and is ongoing post-implementation. The first intervention included educational slides distributed to bedside providers, flyers posted in staff common areas, and arrow signs on patient monitors reminding providers to wean FiO2. Additional PDSA cycle interventions are listed in Figure 1.
We have recently experienced improvement in our process measure and reached the compliance goal of 80% (Figure 2). In addition, we have improved each outcome measure (Figure 3A-C) with decreases in SpO2, FiO2, and frequency of SpO2 > 97%. However, the process, outcome, and balancing measures are currently being assessed with more granular data using the Etiometry T3 platform. This captures patient data every five seconds and will better evaluate our progress. The data is currently being evaluated, with results forthcoming.

Figure 1: Key Driver Diagram
Our four key drivers include: standardized SpO2 goal ranges for patients, bedside provider engagement in weaning FiO2, standardized documentation of respiratory support and settings, and standardized FiO2 weaning protocols. A number of interventions have been completed or successfully implemented, but several are still ongoing.

Figure 2: Compliance run chart
The x-axis is groups of patient data in blocks of 10, and each point on the graph is the average compliance of that group. The y-axis represents percentage of average compliance. The light blue horizontal line represents baseline median compliance rate starting at 70%, and the gray dotted line represents goal median compliance rate of 80%. The arrow labels correspond to the timing of each PDSA cycle. There has been an upward trend in the desired direction of change, and beginning with block 53 there is a significant increase in the median to goal median compliance.

Figure 3 A-C: Outcome measures run charts
Groups of patient data are represented in blocks of 10 on the x-axis, and each point on the graph is the average value of that group. The y-axis represents percentage of average SpO2 (3A), FiO2 (3B), and SpO2 > 97% (3C), respectively. Each run chart has demonstrated improvement in the median in the desired direction of change.