IntroductionElectrocardiographic (ECG) rhythms are used during advanced cardiac life support (ACLS) to guide resuscitation management. Survival to hospital discharge has been reported to be better for patients with pulseless electrical activity (PEA) than asystole in out-of-hospital arrests. Despite this, treatment for these two (non-shockable) rhythms is combined in ACLS guidelines. This study examines if the recorded cardiac rhythm of asystole or PEA during ACLS accurately predicts mechanical cardiac activity as determined by point-of-care ultrasound (PoCUS).MethodsA database review was completed for patients (> 19 years without a do not resuscitate (DNR) order) who presented to a tertiary emergency department in PEA or asystolic cardiac arrest between 2010 and 2014. Patients were separated into two groups: those with electrical cardiac activity (PEA) and those without (asystole). We compared ECG rhythm and PoCUS-documented cardiac activity results (both initial and any) for each case.Results A total of 186 patients met the study criteria. The 46 patients with PEA on ECG were more likely to have cardiac activity than the 140 patients with asystole (odds ratio 7.22 (95% confidence intervals 2.79-18.7) for activity on initial PoCUS; odds ratio 5.45 (2.49-12.0) for activity on any PoCUS during arrest).ECG alone was poorly sensitive for initial cardiac activity (63.64%; 40.66% to 82.80%) and any cardiac activity (54.29%; 36.65% to 71.17%), with specificity marginally better at 80.49% (73.59% to 86.25%) for initial and 82.12% (75.06% to 87.87%) for any activity.ConclusionOur results suggest that ECG rhythm alone is not an accurate predictor of cardiac activity. This supports the use of PoCUS during cardiac arrest, in addition to ECG, to identify patients with ongoing mechanical cardiac activity and to help determine appropriate treatment for this group.