238 - Ventricular Function, ECMO Mode, and Outcome in Neonates with Respiratory Failure Treated with ECMO

Poster Number: 238
Publication Number: 238.33

Shazia Bhombal, Emory University and Children's Healthcare of Atlanta, Atlanta, GA, United States; Elizabeth K. Sewell, Emory University; Children's Healthcare of Atlanta, Atlanta, GA, United States; Joel Davis, Children's Healthcare of Atlanta, Loganville, GA, United States; Sarah Keene, Emory Childrens Healthcare of Atlanta, Decatur, GA, United States

Background: Neonates with severe respiratory failure requiring extracorporeal membrane oxygenation (ECMO) often have concomitant cardiac failure, in the form of decreased systolic function of one or both ventricles.  Cardiac dysfunction impacts ECMO mode selection, with frequent preclusion of veno-venous ECMO since it offers no direct cardiac support.

Objective: Describe our experience with cardiac dysfunction and ECMO modality in this population.

Design/Methods: This retrospective chart review included all neonates (< 30 days of life) who underwent ECMO for primary respiratory failure between January 1, 2010 and October 31, 2022 and had an echocardiogram prior to ECMO cannulation.  Demographic data, short term outcomes and mode of ECMO was documented.  Echocardiogram reports were reviewed for description of moderate-severe systolic dysfunction.  The echocardiogram prior to cannulation was compared with the final before decannulation for changes over time.  Data was analyzed with descriptive statistics and univariate analysis.

Results: Of the 176 neonates treated with ECMO for respiratory failure, 157 had an echo prior to ECMO initiation. Sixty infants (39%) had initial systolic dysfunction: 39 (25%) only right dysfunction, 5 (3%) only left dysfunction, and 16 (10%) biventricular dysfunction.  Primary diagnosis was CDH 26 (43%), MAS 8 (13%), and other 26 (43%). Of these 60 infants, 39 (65%) were treated with VV ECMO with one conversion, and 21 (35%) were initially placed on VA.  There was no statistically significant difference in dysfunction among patients placed on VV compared with the VA ECMO (39/110 [35%] vs. 21/47 [45%]; p=0.276). VA ECMO was indicated due to inability to place VV cannula or lack of availability of VV cannula from 2019-2021.  Survival to discharge did not differ between patients with dysfunction and those without dysfunction (44/60 [73%] vs. 73/97 [75%]; p=0.788). Of those patients with dysfunction, 85% of patients on VV ECMO survived compared with 52% for VA ECMO, p=0.017.  The majority of patients on either mode showed improvement over time.

Conclusion(s): The majority of cardiac dysfunction in patients with primary respiratory failure will improve over time, regardless of ECMO mode.  Surprisingly, cardiac dysfunction was not a predictor of mortality.