Background/Objectives: Conventional cardiopulmonary resuscitation (CCPR) has been the foundational resuscitation approach for decades. Where CCPR is unsuccessful, extracorporeal membrane oxygenation-assisted CPR (ECPR) may improve outcomes. Predicting failure of CCPR and immediate need for ECPR is difficult, and data are lacking. In this retrospective analysis, we analysed both factors that are associated with conversion from CCPR to ECPR and survival outcomes for each event.

Methods: Patients having a CPR event that occurred in the PICU between 2016 and 2022 were included. Pre-CPR-event clinical and laboratory data were collected. We recorded whether CPR was converted to ECPR and documented patient outcomes. Results: 201 CPR events occurred in 164 children, with 45 events converted from CCPR to ECPR. Time to ROSC or time to ECMO flow was (median [IQR]) 2 (1.5) min for CCPR events and 37 (21.60) min for ECPR events. The maximum pre-CPR-event lactate values were 1.8 mmol/L for CCPR and 4.5 mmol/L for ECPR events. Respiratory arrest preceded 35.3% of CCPR and 4.4% of ECPR events. PICU mortality was 27.8% for CCPR and 50% for ECPR events. Most deaths occurred because of withdrawal of life-sustaining treatments. In a multivariable analysis, cardiac surgical diagnosis, pre-CPR-event lactate, as well as duration of CPR were associated with conversion from CCPR to ECPR.

Conclusions: Our study demonstrates that pre-CPR-event lactate concentrations and duration of arrest should alert clinicians to a high likelihood of needing ECPR, while a preceding respiratory arrest may indicate a low likelihood. Mortality post CCPR is significant, mainly due to overall illness severity rather than the consequences of the CPR event.