550 - Cardiac Arrest in Infants with Bronchopulmonary Dysplasia

Publication Number: 3531.550

Laura Sillers, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Yi Sun, Guangzhou Women and Children‘s Medicial Center, Guangzhou Medical University, Guangzhou, Guangdong, China (People's Republic); Mary Haggerty, Children's Hospital of Philadelphia, haddonfield, NJ, United States; Kathleen Gibbs, Childrens Hospital of Philadelphia, Philadelphia, PA, United States

Background: Bronchopulmonary dysplasia (BPD) is a major driver of mortality among older infants in the neonatal intensive care unit (NICU). There are no current studies that specifically examine cardiac arrest, resuscitation practices, and outcomes following cardiac arrest in the BPD population.

Objective: To describe clinical characteristics and outcomes of infants with BPD who experience cardiac arrest at a single academic referral center. A secondary objective was to evaluate for an association between lung hyperinflation and primary cardiac arrest, which has been anecdotally observed within our BPD program but has not been formally described.

Design/Methods: Retrospective cohort study of cardiac arrest occurring in BPD patients between January 2020 and January 2025 at a single quaternary NICU with a large, primarily referral-based BPD program. Cardiac arrest was defined as events receiving ≥1 minute of chest compressions. Eligible events were identified using a unit-based codes database, with additional patient and code-level data obtained by review of the electronic medical record. The most recent chest x-ray prior to each code event was assessed for the presence of severe hyperinflation, defined as lung expansion ≥11 ribs with diaphragmatic flattening or inversion. Statistical analysis was performed using Stata/IC 15.1 (StataCorp, 2020). Comparisons of categorical data were performed with Fisher's exact test; continuous variables were compared using Kruskal-Wallis test as data was not normally distributed.

Results: 68 eligible events occurred in 50 infants; 11 infants had multiple eligible events (range 2-4). 80% of infants with cardiac arrest had grade 3 BPD, as defined by invasive ventilation at 36 weeks postmenstrual age (Table 1). Most events occurred in patients on invasive mechanical ventilation (98.5%) and were preceded by respiratory decompensation (88.2%) (Table 2). Primary cardiac arrest was associated with lower rates of ROSC (50.0% vs 91.7%, p=0.008) and had a higher frequency of severe hyperinflation on the most recent imaging prior to arrest, although this did not meet statistical significance (87.5% vs 46.7%, p=0.06). While ROSC was achieved in 86.8% of events, only 62% of BPD patients who experienced cardiac arrest ultimately survived to hospital discharge (Figure 1). A high rate of tracheostomy and home ventilation was observed among those infants who survived to discharge (58.1%).

Conclusion(s): Morbidity and mortality are common among infants with BPD who experience cardiac arrest in the NICU. Severe hyperinflation may be associated with risk of primary cardiac arrest in this patient population.

Table 1: Infant Demographics (n=50) Variable n (%) or median [IQR]


Table 2: Cardiac arrest events by return of spontaneous circulation
ROSC = return of spontaneous circulation; PMA = postmenstrual age; HFOV = high frequency oscillatory ventilation; CMV = conventional mechanical ventilation; NO = nitric oxide; ARC = acute respiratory compromise; CPA = cardiopulmonary arrest; PEA = pulseless electrical activity; *PEA vs. sinus bradycardia; **asystole vs. sinus bradycardia

Figure 1: Survival to Discharge After Cardiac Arrest Events
ROSC = return of spontaneous circulation; dc = discharge