403 - Implementing a CPR Bundle to Optimize Pediatric Emergency Department CPR

Publication Number: 4395.403

Alexandra Berardi-Bloomfield, Children's Hospital Colorado, Denver, CO, United States; Erin McGonagle, University of Utah School of Medicine, Salt Lake City, UT, United States; Irina Topoz, University of Colorado School of Medicine, Aurora, CO, United States; Andrew Krack, University of Colorado School of Medicine, Highlands Ranch, CO, United States; Cortney Braund, Children's Hospital Colorado, Denver, CO, United States; Kathleen Jacobs, Children's Hospital Colorado, Denver, CO, United States; Joan MacKenzie, Children's Hospital Colorado, aurora, CO, United States; Kristin Vickrey, Children's Hospital, Aurora, CO, United States; Nichole Thompson, Blank Children's Hospital, Cumming, IA, United States; Laura Rochford, Children's Hospital Colorado, Denver, CO, United States

Background: Pediatric cardiac arrest is a rare but devastating occurrence. High-quality cardiopulmonary resuscitation (CPR) remains the cornerstone of intervention, yet there are inconsistencies with adherence to Pediatric Advanced Life Support (PALS) and American Heart Association (AHA) guidelines.

Objective: To improve the quality of pediatric CPR by implementing a standardized CPR bundle with 80% compliance by May 2025.

Design/Methods: This is a single-center, tertiary Emergency Department (ED) quality improvement study. We collected data from September 2023 to May 2025. Inclusion criteria were medical and trauma cardiopulmonary resuscitation with CPR lasting at least 5 minutes. Exclusions were code events with thoracostomy, thoracostomy, and/or Extracorporeal Membrane Oxygenation (ECMO) in the first 5 min of CPR. We used Model for Improvement to identify key drivers for high-quality pediatric CPR. We developed a CPR Bundle based on PALS and AHA guidelines (Figure 1) and implemented the role of CPR coach. We also provided dedicated training with audit and feedback to code participants. The primary outcome was the percentage of bundle items completed for each code (out of 15 available points). The first process measure included the frequency of verbalization of pulse and rhythm checks during each code. The second process measure was the median response on the effectiveness of CPR coaches (on a 1 to 5 scale, with 5 being most effective) from code team members. The balancing measure was the chest compression fraction (percent of time spent on chest compressions during CPR). We used run charts to analyze data over time, the Mann-Whitney U test to compare medians before and after intervention, and Fisher's exact to compare proportions.

Results: We included 21 cardiopulmonary arrests. The first 9 codes were used as a baseline. Completion of the CPR bundle increased from 60% to 80% (p=0.004). Special cause variation was detected on the run chart of bundle completion (Figure 2). The percentage of time pulse and rhythm checks were verbalized during a code went from 39% to 88% (p=0.002) and 11% to 71% (p=0.001), respectively. Median response to CPR coach effectiveness survey was 5. Median chest compression fraction remained at 90% throughout the study.

Conclusion(s): Implementation of a structured CPR Bundle in the first 5 min of CPR and introduction of the CPR Coach improved adherence to PALS and AHA guidelines during pediatric cardiopulmonary resuscitation in the pediatric ED.

Figure 1: CPR Bundle
Total of 15 points available for bundle completion during first 5 minutes of a code event. *Item could be performed more then once.

Figure 2: Run Chart of Percent Bundle Compliance
Two special cause variation detected: too few runs (4 runs instead of expected 8) and a shift 6 points above median).