Key result
A novel ECG-based machine learning algorithm predicted the presence of a spontaneous pulse during uninterrupted CPR with an area under the receiver operating characteristic curve of 0.84.
Why the study?
Resuscitation protocols require pausing chest compressions to check for a pulse, but pauses during a pulseless rhythm can worsen patient outcomes.
Does an ECG-based machine learning algorithm accurately predict pulse status during uninterrupted CPR in patients with out-of-hospital cardiac arrest?
Observational (n=383)
Does an ECG-based machine learning algorithm accurately predict pulse status during uninterrupted CPR in patients with out-of-hospital cardiac arrest?
Effect estimate: AUC 0.84 (95% CI 0.797-0.88)
A novel ECG-based machine learning algorithm can accurately predict pulse status during uninterrupted CPR, potentially allowing resuscitation to proceed without pauses for pulse checks.
No takes yet. Share an insight, caveat, or question.
Promising accuracy for pulse detection during uninterrupted CPR; leaves open clinical impact and requires prospective validation before practice change.
Sashidhar et al. (2020) conducted an observational in Out-of-hospital cardiac arrest (n=383). ECG-based machine learning algorithm vs. Clinical pulse status (audio annotation and blood pressure) was evaluated on Prediction of pulse status during CPR (AUC) (AUC 0.84, 95% CI 0.797-0.88). A novel ECG-based machine learning algorithm predicted the presence of a spontaneous pulse during uninterrupted CPR with an area under the receiver operating characteristic curve of 0.84.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: