Good morning, everybody. My name is Ivana Kapan. I am the PGY-5 at Mount Sinai Kravis Children's Hospital, where I am pursuing my Pediatric Critical Care Fellowship. This morning, I'll be discussing my research project entitled Extracorporeal Membrane Oxygenation, a bridge to palliation in single ventricle physiology. Extracorporeal Membrane Oxygenation, more commonly known as ECMO, is a form of invasive mechanical support that can serve as a bridge to recovery, decision, or transplant. It is frequently used in the single ventricle population during or after palliation. There is literature to suggest the prognosis of patients who require ECMO during stages 1 and 2 of palliation. However, a subset of patients with single ventricle physiology experience myocardial or respiratory decompensation prior to the first stage of palliation. Little is known regarding the use of preoperative ECMO in this specific cohort. Bautista Hernandez-Eyal from Boston looked at ECMO as a bridge to definitive versus palliative repair in patients with congenital heart disease. They found that ECMO could successfully be utilized prior to palliative or definitive repair. In their study, they found that 62% of patients requiring ECMO survived to hospital discharge. Those who died were more likely to have longer ECMO runs. The study did not show a difference in outcome between single versus two ventricle congenital heart defects. So our group set out to investigate the single ventricle population specifically. Our objective was thereby to determine the outcomes of patients with single ventricle physiology who had undergone ECMO support as a bridge to first stage palliation. From a method standpoint, this study was a retrospective registry database study. We utilized the ELSO registry. We obtained data from patients from January 2016 to January 2021 who were supported by ECMO pre-palliation. We looked at two cohorts, survivors versus non-survivors. Non-survivors were defined as patients who did not survive to hospital discharge. Patients were included in the study if they were zero to two months of age, had single ventricle physiology, and had received ECLS. Patients were included if they had no documented surgical code in the ELSO database pre-ECLS. Surgical codes included those for PA banding, the Norwood procedure, DKS, or systemic to PA shunts. Patients with single ventricle physiology were identified using ELSO cardiac codes, ICD-9, and ICD-10 codes. These included hypoplastic left heart syndrome, tricuspid atresia, pulmonary atresia, double inlet left ventricle, among others. Patients were excluded if the patient had gone on cardiopulmonary bypass pre-ECLS. From a statistical standpoint, we conducted descriptive statistics for all variables. Continuous variables were assessed using the Mann-Whitney U-test, and dichotomous variables were assessed using the Chi-square and Fisher's exact test. Given the small sample size, a p-value of less than 0.1 was used to identify variables of interest. Variables that were found to have a p-value of less than 0.1 were then placed into the multivariate analysis through a logistic regression model. A p-value of less than 0.05 was used to determine statistical significance in the multivariate analysis. Moving forward into our results section, here is a table describing the demographics of all patients, those that survived, and those that died. There were 65 patients in total, and 38 patients died. No patient characteristics between the two cohorts were found to be statistically significant. Generally speaking, this cohort was an older cohort, with the median age being one to two weeks old on average. It doesn't appear that prematurity was a factor in this group either. In both groups, the most common indication for ECMO appeared to be myocardial decompensation, with a small portion of patients in each cohort requiring eCPR. Given our time constraint, we only presented data comparing survivors vs. non-survivors. In summary, there were 65 patients in the study who went on to ECMO. Of the 65 patients, 24 died on ECMO, and 41 survived the initial ECMO run. 14 patients died before discharge, leaving 27 patients who survived to hospital discharge. Here is a pie chart demonstrating the distribution of single ventricle diagnoses. The most common diagnosis was hypoplastic left heart syndrome, accounting for nearly 40% of our patients. With regards to pre- and initial ECMO support, patients who survived were more likely to have a higher oxygen requirement, lower peak inspiratory pressure, and higher PEEP prior to ECMO than those who died. The only inference we can make was that non-survivors may have worse compliance than survivors based on their peak inspiratory pressures. Patients who survived were found to have a significantly shorter ECMO run than non-survivors. Hours on ECMO, then, may be a surrogate marker of better or worse candidacy for surgical palliation. Patients who died were significantly more likely to suffer from mechanical complications, surgical bleeding, and renal injury requiring renal replacement therapy. As expected, patients who died were significantly more likely to suffer from four or more complications. Multivariate analysis using logistic regression was completed. The following variables were variables of interest in the multivariate model, renal replacement therapy, PEEP, and hours on ECMO. Backward selection was done, and after adjusting for renal replacement therapy and PEEP, the odds of surviving versus dying decreased by 1% for every hour on ECMO. Moving forward into our discussion, 54% of patients who required ECMO pre-palliation died before hospital discharge. There were no significant differences between patient characteristics when comparing survivors and non-survivors. The utilization of ECMO pre-palliation appears to be feasible in the single ventricle population. However, in our study, we found that greater than 50% of patients who required this form of mechanical support died before hospital discharge. Patients who died were more likely to have more complications on ECMO and have longer ECMO runs, which may indicate that they, in fact, are poorer candidates for surgical palliation. This study may provide insight to both physicians and families on how to set realistic expectations with regards to prognosis and patient outcomes. It may also help guide challenging conversations with families of these critically ill neonates. In comparison to Bautista-Hernandez's study, where 62% of patients survived to hospital discharge, only 42% of our patients survived. Bautista-Hernandez's group had 10 two-ventricle survivors in contrast to 7 non-survivors, and 6 single ventricle survivors in contrast with 3 non-survivors. Although Bautista-Hernandez's group did not appreciate a significant difference in outcome between single and two-ventricle patients, this may be a reason as to why we saw greater mortality, as we exclusively looked at the single ventricle population. A paper by Sherwin et al. looked at outcomes of neonates requiring ECMO after stage 1 palliation for hypoplastic left heart syndrome. Sherwin's group found a 31% survival to hospital discharge, a survival rate similar to our study. Therefore, patients requiring ECMO before and after first stage palliation may carry a similar risk of mortality. From a limitation standpoint, this was a retrospective study. In addition, we were limited by only being able to analyze the variables that were already collected in the ELSO registry. Furthermore, the sample size was small, with limited power. Moving forward, as we prepare our manuscript, we will be looking at other subpopulations of our cohort, specifically patients who died on ECMO, those who may have survived the ECMO run but did not survive to hospital discharge, and the hypoplastic left heart syndrome subset specifically. Thank you all for your time, and special thanks to the contributing authors, specifically my mentor, Dr. Scott Aiden, who has helped me immensely while we were creating this project over the past couple of years. On a final note, I would like to dedicate this project to Dr. George Aforia-Mambo, who was a contributing author, a dear friend, the chief of our pediatric critical care division at Mount Sinai, and an outstanding clinician, leader, teacher, and mentor to many who we unfortunately lost at the end of 2021. These are a list of my references that I used. And again, I would like to thank everybody for their attention, and I would love to entertain any comments or questions from the audience.

ECMO

single ventricle physiology

patient outcomes

prognosis

complications