Study population and setting
This was an international case series that evaluated the characteristics of cardiac arrhythmias among patients aged 18 years and older with confirmed SARS-CoV-2 infection who were admitted to 29 institutions in 12 countries across 4 continents (Asia, Europe, North America, South America), from January 4 to August 7, 2020. Cardiac electrophysiologists from participating institutions contributed deidentified data from hospitalized COVID-19 patients, including demographics, comorbidities, electrocardiography (ECG) and echocardiogram, treatment with oseltamivir or remdesivir, hydroxychloroquine (HCQ) or azithromycin medications, presence and type of cardiac arrhythmias (atrial fibrillation [AF], atrial flutter [AFL], supraventricular tachycardia [SVT], non-sustained ventricular tachycardia [NSVT], monomorphic or polymorphic ventricular tachycardia [VT], atrioventricular block [AVB], sinus bradycardia below 40bpm [bradyarrhythmia]). Standardized data elements were uploaded through research electronic data capture from each center and analyzed by the Columbia University Medical Center, which served as the coordinating center. Chi-square tests and analysis of variance (ANOVA) were used to compare proportions and means respectively between different groups of COVID-19 subjects.
Summary of Main Findings
A total of 4526 adults hospitalized with COVID-19 were included in the study database (mean age 62.8 years old, 57% male), among whom 827 had cardiac arrhythmias during their hospitalization. The prevalence of cardiac arrhythmia was 12.9% among 2762 COVID-19 patients from the four institutions who included data from all hospitalized COVID-19 patients over the study period. Among the study sample of 827 patients with cardiac arrhythmia, the mean age was 71.1 years old, 64.7% were male, and 43.7% were white. A significant proportion had hypertension (68.6%), diabetes (41.6%), congestive heart failure (30.8%), coronary artery disease (24.3%), lung disease (18.3%), and chronic kidney disease (21%). Compared to the overall sample of 4526 patients with COVID-19, the sample who developed cardiac arrhythmia were older, had more comorbidities including hypertension, diabetes, coronary artery disease, chronic kidney and lung disease, and reduced left ventricular ejection fraction. There was considerable heterogeneity in the treatment of SARS-CoV-2 across the continents. Antiviral treatments were commonly given to subjects with COVID-19 and arrhythmia in Asia and Europe (93.2% and 50.9% respectively). On the other hand, HCQ and azithromycin were commonly given as an anti-SARS-CoV-2 treatment in Europe, North and South America. About 14% of the subjects had received anti-IL-6 treatment, while 61.3% received anticoagulation therapy. The majority of HCQ (87%) and azithromycin was given early in the pandemic (January to May 2020), while there was no difference in the prescription of antiviral treatment across the study period. Data pertaining to dexamethasone treatment was not captured. The prevalence of death among COVID-19 patients with arrhythmias was consistently high across the study period; 51% early in the pandemic (January to April 2020) and 48% from April to August, 2020 The most common cardiac arrhythmia was atrial arrhythmia (SVT, AF, AFL), observed in 81.8% of those with arrhythmia. In addition, 21% and 22.6% developed ventricular arrhythmia (NSVT, VT, VF) and bradyarrhythmia, respectively. Ventricular arrhythmia was associated with worse survival compared to atrial arrhythmia and bradyarrhythmia (37.8% vs. 50.9%). Patients with ventricular arrhythmia tended to have slightly higher baseline corrected QT interval compared to others (overall sample mean QTc 446.5±43.2 ms) and the majority of those who developed monomorphic (n=30) or polymorphic VT (n=33) had reduced baseline left ventricular ejection fraction, LVEF, by echocardiography (mean 36.1% and 46.4% respectively). Overall, the mean LVEF was normal among subjects with any type of cardiac arrhythmia who underwent echocardiography (267/827, 32%). Treatment with HCQ or antiviral medications and regional locations did not affect baseline QT interval. 50% of subjects with ventricular arrhythmia were either hypoxic or hypotensive and 1/3 had renal failure or acidosis at the time of the arrhythmia. Pulseless electrical activity (PEA) and asystole were the most common cardiac rhythm among COVID-19 subjects who died. A secondary aim to document the effect of the COVID-19 pandemic on the volume of electrophysiologic procedures noted a significant decline across the institutions, with more than a 50% decrease in both elective and urgent device implantations and cardiac ablation procedures during the study period (January 1 to August 7, 2020) compared to the
The study was able to collect data on hospitalized COVID-19 patients from 29 institutions across the globe, which enhances generalizability of the results. Data were collected by electrophysiologists at each institution, which decreases the chance of misclassification of cardiac arrhythmia types. Also, the study included data across an 8-month period during the pandemic to capture any changes in cardiac arrhythmia types that could be attributed to changes in medical care and treatment for COVID-19 patients.
The sample was not inclusive of all patients admitted for COVID-19 at each institution (except for four sites); it is possible that patients who were monitored for cardiac arrhythmia were sicker with a higher likelihood of arrhythmia compared to all hospitalized COVID-19 patients. 60% of those with cardiac arrhythmia were from North American institutions, with the majority from two New York City hospitals, further limiting generalizability of the findings on arrhythmia types and correlates measured among the 827 patients with arrhythmia. Though the authors estimated arrhythmia prevalence only among the subset of institutions who submitted data for all hospitalized COVID-19 patients rather than the full sample from all sites, unfortunately, the characteristics of this denominator were not described, making it difficult to evaluate the representativeness of the sample or to directly compare those with and without arrhythmia. Echocardiogram results were available for just a third of the COVID-19 patients with arrhythmia, and those results should be interpreted with caution given the smaller sample which may not be representative of hospitalized COVID-19 patients. Lastly, the study could not differentiate whether the cardiac arrhythmia observed among COVID-19 patients is specific to the pathological effects of SARS-CoV-2 virus or more likely, a marker of critical illness.
The study added significant incremental new information on the characteristics of patients with COVID-19 who develop cardiac arrhythmia, and the relative types of arrhythmia and their associations with patient morbidity and mortality.
This review was posted on: 12 March 2021