Prognostic Risk Analyses for Postcardiotomy Extracorporeal Membrane Oxygenation in Children: A Review of Early and Intermediate Outcomes

We evaluated the morbidity and mortality of children requiring postcardiotomy extracorporeal membrane oxygenation (ECMO) to determine independent factors affecting early and intermediate outcomes. Between January 2002 and December 2015, 79 instances of ECMO after cardiac surgery in 73 children were retrospectively reviewed. Follow-up was completed in December 2016. Predictive risk analyses were employed concerning weaning of ECMO, hospital discharge, and mortality after discharge. Age and weight were 14.9 ± 25.6 months and 7.0 ± 5.3 kg, respectively. Median support time was 8.3 ± 4.4 days. Sixty-seven (85%) were successfully weaned off ECMO and 48 (61%) survived to hospital discharge. Multi-variate logistic regression analysis identified the first day to obtain negative fluid balance after initiation of support (adjusted odds ratio = 0.42), high serum lactate levels (0.97), and high total bilirubin (0.84) during support as significant independent factors associated with successful separation from ECMO. The first day of negative fluid balance (0.65) after successful decannulation was an independent risk factor for survival to hospital discharge. After hospital discharge, actuarial 1-year, 5-year, and 10-year survival rates were 94%, 78%, and 78%, respectively. Low weight increased the risk of death after hospital discharge by a multi-variate Cox hazard model. High serum lactate, high serum bilirubin, and unable to obtain early negative fluid balance during support impacted mortality of decannulation. Obtaining a late negative fluid balance in post-ECMO were independent risk factors for death after successful weaning. Low weight affected intermediate outcomes.