Cardiac arrhythmias among hospitalized Coronavirus 2019 (COVID-19)
patients: prevalence, characterization, and clinical algorithm to
classify arrhythmic risk
Objectives: A significant proportion of COVID-19 patients may have
cardiac involvement including arrhythmias. Although arrhythmia
characterization and possible predictors were previously reported, there
are conflicting data regarding the exact prevalence of arrhythmias.
Clinically applicable algorithms to classify COVID patients’ arrhythmic
risk are still lacking, and are the aim of our study. Methods: We
describe a single center cohort of hospitalized patients with a positive
nasopharyngeal swab for COVID-19 during the initial Israeli outbreak
between 1/2/2020 –30/5/2020. The study’s outcome was any documented
arrhythmia during hospitalization, based on daily physical examination,
routine ECG’s, periodic 24-hour Holter, and continuous monitoring.
Multivariate analysis was used to find predictors for new arrhythmias
and create classification trees for discriminating patients with high
and low arrhythmic risk. Results: Out of 390 COVID-19 patients included,
28 (7.2%) had documented arrhythmias during hospitalization, including:
23 atrial tachyarrhythmias, combined atrial fibrillation (AF) and
ventricular fibrillation, ventricular tachycardia storm, and 3
bradyarrhythmias. Only 7/28 patients had previous arrhythmias. Our study
showed significant correlation between disease severity and arrhythmia
prevalence (p<0.001) with a low arrhythmic prevalence among
mild disease patients (2%). Multivariate analysis revealed background
heart failure (CHF) and disease severity are independently associated
with overall arrhythmia while age, CHF, disease severity, and arrhythmic
symptoms are associated with tachyarrhythmias. A novel decision tree
using age, disease severity, CHF, and troponin levels was created to
stratify patients into high and low risk for developing arrhythmia.
Conclusions: Dominant arrhythmia among COVID-19 patients is AF.
Arrhythmia prevalence is dependent on age, disease severity, CHF, and
troponin levels. A novel simple Classification tree, based on these
parameters, can discriminate between high and low arrhythmic risk