Neuroprognostication in Extracorporeal Membrane Oxygenation Patients After Cardiac Arrest: A Narrative Review

The use of extracorporeal membrane oxygenation (ECMO) for refractory cardiac arrest and post-cardiac arrest shock is increasing. Still, neuroprognostication in these patients is complicated by sedation duration, altered biomarker clearance, and compromise of usual diagnostic tools. This narrative systematic review assesses the evidence on neuroprognostication strategies in ECMO-supported patients following cardiac arrest, focusing on the accuracy of conventional prognostic tools and novel biomarkers. Literature from 2014 to 2024 was reviewed, with a focus on studies assessing neuroprognostic tools in ECMO patients following cardiac arrest. Available evidence suggests that traditional neuroprognostic tools continue to perform reasonably well in ECMO patients with certain caveats. Neuron-specific enolase is increased as a result of hemolysis, electroencephalography and evoked potentials are unreliable, and magnetic resonance imaging is contraindicated. A multimodal approach using at least two criteria for poor outcome seems justified in ECMO and non-ECMO populations. Although neuroprognostication in ECMO patients is uncertain and difficult, available instruments may provide key prognostic information when used with appropriate caution and by expert interpreters. The potential for research may include prospective validation of available tools and ECMO-specific algorithms, as well as the search for new biomarkers with less circuit-related influence. The field will need to advance further to best treat this challenging population and optimize patient care and resource utilization.