New-Onset Atrial Arrhythmias Are Independently Associated With In-Hospital Mortality in Veno-Venous Extracorporeal Membrane Oxygenation

To explore if atrial arrhythmias are associated with in-hospital mortality in veno-venous extracorporeal membrane oxygenation (VV-ECMO) patients. Retrospective observational cohort study. Quaternary care academic medical center. Patients with respiratory failure requiring VV-ECMO for >24 hours between January 1, 2016, and January 1, 2019. None, observational study. Two hundred nineteen VV-ECMO patients were included. Patients were stratified by absence or presence of clinically significant atrial arrhythmias during the VV-ECMO run. Atrial arrhythmias were defined as either atrial fibrillation or atrial flutter that occurred during VV-ECMO and required pharmacologic or electrical intervention. The primary outcome was in-hospital mortality. Secondary outcomes included a composite of thrombotic events, which included ischemic stroke and on-pump arterial thrombosis. Other objectives of this analysis included characterization of atrial arrhythmia incidence, risk factors, and management. A total of 67 patients (30.5%) experienced new-onset atrial arrhythmias post–ECMO cannulation. Age, male sex, and norepinephrine use were independently associated with atrial arrhythmia development. In-hospital mortality was significantly higher in the atrial arrhythmia group (38.8% v 19.1%; p = 0.003). In the multivariate logistic regression analysis, atrial arrhythmias during VV-ECMO were independently associated with increased odds of in-hospital mortality (odds ratio, 2.21; 95% confidence interval, 1.08-4.55; p = 0.03), after controlling for Respiratory Extracorporeal Membrane Oxygenation Survival Prediction score, acute renal failure, total norepinephrine dose, and total cannulation time. New-onset atrial arrhythmias are a frequent complication during VV-ECMO and are independently associated with excessive in-hospital mortality. Thus, their presence may serve as an important prognostic tool in this patient population.