Extracorporeal Cardiopulmonary Resuscitation Outcomes in Pre-Glenn Single Ventricle Infants: Analysis of A Ten-Year Dataset.

Abstract

Background: While several studies have reported on outcomes of extracorporeal membrane oxygenation (ECMO) in patients with single ventricle physiology, few studies have described outcomes of extracorporeal cardiopulmonary resuscitation (ECPR) in this unique population. The objective of this study was to determine survival and risk factors for mortality after ECPR in single ventricle patients prior to superior cavopulmonary anastomosis, using a large sample from the Extracorporeal Life Support Organization (ELSO) Registry.

Methods: We included single ventricle patients who underwent ECPR for in-hospital cardiac arrest (IHCA) between January 2012 and December 2021. We excluded patients who had undergone a superior cavopulmonary anastomosis, inferior cavopulmonary anastomosis, or who were older than 180 days at the time of ECPR. We collected data on mortality, ECMO course and ECMO complications. Subjects who survived to hospital discharge after ECPR were compared to subjects who did not survive to hospital discharge. We then performed univariate logistic regression followed by multivariable logistic regression analysis for associations with survival to hospital discharge.

Results: There were 420 subjects included who had index ECPR events. Median age was 14 days, median weight was 3.14 kg. Hypoplastic left heart syndrome was the most common diagnosis (354/420; 84.2%), and 47.4% of the cohort (199/420) had undergone a Norwood operation. Survival to hospital discharge occurred in 159/420 (37.9%) of subjects. Median number of hours on ECMO (122 vs. 93 hours; p<0.001), presence of seizures by electroencephalography (24% vs. 15%; p=0.033), and need for renal replacement therapy (45% vs. 34%; p=0.023) were significantly higher among non-survivors compared to survivors. In the subgroup of Norwood patients, survival was 43.2% after ECPR. Presence of Norwood variable was 54% among ECPR survivors in the overall cohort, compared to 43% among non-survivors (p=0.032). In a multivariable logistic regression model to test association with survival to discharge, number of ECMO hours and presence of seizures were associated with decreased odds of survival to hospital discharge [adjusted odds ratio 0.95 (95% C.I. 0.92-0.98) and 0.57 (95% C.I. 0.33-0.97) respectively]. The odds ratio for ECMO hours demonstrated a decrease in odds of survival by 5% for every 12 hours on ECMO. Presence of Norwood operation pre-arrest was associated with increased odds of survival [adjusted odds ratio 1.53 (95% C.I. 1.01-2.32)].

Conclusion: In our cohort of pre-Glenn single ventricle infants, survival after ECPR for in-hospital cardiac arrest was 37.9%. Number of hours on ECMO and seizures post-ECMO cannulation were associated with decreased odds of survival. Single ventricle infants who had undergone Norwood palliation pre-arrest were more likely to survive to hospital discharge.