Outcomes in Neonates with Multiple Runs of ECMO – Analysis of ELSO Registry Data

Background: For neonates with severe respiratory or cardiac failure that is not responsive to conventional treatment, ECMO is a lifesaving support mechanism that has been used since the mid 1970’s with increasing success. In the past decade, there have been fewer neonatal ECMO runs due to improved intensive care management, however it remains a critical strategy. Some patients deteriorate after a successful run of ECMO, and consideration is given to re-cannulation and repeat ECMO support, despite unclear results. The aim of this report was to understand outcomes, morbidities, and causes for mortality or survival in neonates with more than one run of ECMO.

Methods: Data regarding neonates who underwent ECMO were obtained from the Extracorporeal Life Support Organization for the years 2009-2019. The cohort was stratified into patients who had multiple ECMO runs (MR) and those who had single runs (SR). The cohorts were compared with regards to demographics, type of ECMO, complications and mortality. Comparisons were made using Fishers exact test for categorical variables and either Mann Whitney U test or T-test for continuous data.

Results: Overall, 14,773 neonates were found in the ELSO registry during the study period. 562 (3.8%) cases had MR, while 14, 211 had SR. Of the MR patients, there was no change in frequency/year over the time period. Males comprised 59.2%, 58.6 were white, and 93.8% had two runs, while 6% had three. Cardiac support type was noted in 50% MR, while ECPR comprised 20% of the total MR group. MR cases had a significantly shorter first run of ECMO compared to SR. Survival in MR was best in pulmonary support, and of these CDH had a 47% survival. Table 1 notes the differences between MR and SR and Table 2 between MR survived and died. Third runs were associated with only 14% survival.

Conclusion: Multiple ECMO runs in neonates are uncommon, with an expected significantly reduced survival compared to single run cases. However, a second run of ECMO did salvage a third of the cases, and careful patient selection may be critical in improving survival. Higher survival was associated with pulmonary support and in particular a CDH diagnosis; Cardiac and ECPR support had poorer outcomes. Further analysis may help to create an algorithm that can prognosticate which patients should be offered a second ECMO run.