Outcome of Extracorporeal Cardiopulmonary Resuscitation in Pediatric Patients Without Congenital Cardiac Disease: Extracorporeal Life Support Organization Registry Study

OBJECTIVES: To describe the use of extracorporeal cardiopulmonary resuscitation (ECPR) in pediatric patients without congenital heart disease (CHD) and identify associations with in-hospital mortality, with a specific focus on initial arrest rhythm. DESIGN: Retrospective cohort study using data from pediatric patients enrolled in Extracorporeal Life Support Organization (ELSO) registry between January 1, 2017, and December 31, 2019. SETTING: International, multicenter. PATIENTS: We included ECPR patients under 18 years old, and excluded those with CHD. Subgroup analysis of patients with initial arrest rhythm. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We identified 567 patients: neonates (12%), infants (27%), children between 1 and 5 years old (25%), and children over 5 years old (36%). The patient cohort included 51% males, 43% of White race, and 89% not obese. Most suffered respiratory disease (26%), followed by acquired cardiac disease (25%) and sepsis (12%). In-hospital mortality was 59%. We found that obesity (adjusted odds ratio aOR, 2.28; 95% CI, 1.21-4.31) and traumatic injury (aOR, 6.94; 95% CI, 1.55-30.88) were associated with greater odds of in-hospital mortality. We also identified lower odds of death associated with White race (aOR, 0.64; 95% CI, 0.45-0.91), ventricular tachycardia (VT) as an initial arrest rhythm (aOR, 0.36; 95% CI, 0.16-0.78), return of spontaneous circulation before cannulation (aOR, 0.56; 95% CI, 0.35-0.9), and acquired cardiac disease (aOR, 0.43; 95% CI, 0.29-0.64). Respiratory disease was associated with greater odds of severe neurologic complications (aOR, 1.64; 95% CI, 1.06-2.54). CONCLUSIONS: In children without CHD undergoing ECPR, we found greater odds of in-hospital mortality were associated with either obesity or trauma. The ELSO dataset also showed that other variables were associated with lesser odds of mortality, including VT as an initial arrest rhythm. Prospective studies are needed to elucidate the reasons for these survival differences.