Describing heart rate variability in patients with chronic atrial fibrillation during hospitalization for COVID‐19

Abstract Introduction Myriad cardiovascular manifestations have been reported with COVID‐19. We previously reported that failure of PR interval shortening with increasing heart rate (HR) in patients with COVID‐19 is associated with adverse outcomes. Here, we report on heart rate variability (HRV) and clinical outcomes in patients with chronic atrial fibrillation (cAF) hospitalized for COVID‐19. Methods A retrospective review of admitted COVID‐19 patients with cAF between 1 March 2020 to 30 June 2020 was performed. HRV in cAF was compared during pre‐COVID‐19 and COVID‐19 admissions; we selected pre‐COVID‐19 ECGs with HRs that were within 10 beats per minute of the COVID‐19 ECGs. Mean HR and each RR interval were recorded. Time‐domain measurements of HR variability were then calculated (SDSD, RMSSD, pNN50). Clinical outcomes during COVID‐19 were correlated to indices of HRV. Results A total of 184 ECGs (95 pre‐COVID‐19, 89 COVID‐19) from 38 cAF in‐patients were included. Mean age 78.6 ± 11.4 years, male 44.7%. The mean number of ECGs analyzed per patient pre‐COVID‐19 was 2.50 and during COVID‐19 was 2.34. Comparing pre‐COVID‐19 versus COVID‐19 ECGs showed: mean HR (95.9 ± 24.3 vs. 101.6 ± 22.8 BPM; P = .10), SDSD (109.0 ± 50.6 vs. 90.3 ± 37.2 ms; P < .01), RMSSD (184.1 ± 80.4 vs. 147.3 ± 59.8 ms; P < .01), pNN50 (73.8 ± 16.3 vs. 65.6 ± 16.6%; P < .01). Patients who had a smaller pNN50 during a COVID‐19 admission had increased mortality (50.0% vs. 14.3%; log‐rank test P = .02). Conclusion In patients with cAF, the HRV was reduced during COVID‐19 compared with prior illnesses at similar average heart rates. Patients with the most depressed HRV as measured by pNN50 had an associated increase in mortality compared with patients whose HRV was preserved.


| INTRODUC TI ON
The coronavirus disease 2019 (COVID- 19) pandemic has resulted in considerable morbidity and mortality. Although the disease predominantly affects the respiratory system, multiorgan dysfunction, including cardiac injury, is widely reported. Cardiac injury is common in COVID-19 patients, with studies reporting a prevalence of 19.7%, 29%, 77%, in hospitalized, critically ill, and deceased patients, respectively. [1][2][3] Various disease manifestations, including myocarditis, acute coronary syndrome, arrhythmias, heart failure, and venous thromboembolism have all been reported. 4 Importantly, cardiac involvement is associated with a higher risk for in-hospital mortality, and patients with pre-existing cardiovascular abnormalities have an increased risk for severe illness. 1,5 It has been proposed that the virus causes direct myocardial injury, whereas the release of cardiotoxic cytokines further exacerbates this injury. 6 Currently, there is still a paucity of literature detailing the effect of COVID-19 on the cardiac conduction system and autonomic nervous system (ANS). We have previously reported that in sinus rhythm, the PR interval in patients with COVID-19 may fail to shorten appropriately with increasing rate and that this observation is associated with adverse clinical outcomes. 7 Electrophysiologic effects of COVID-19 in patients with chronic atrial fibrillation (cAF) have not been described.
Anecdotally, we observed that patients hospitalized for COVID-19 with comorbid cAF demonstrated more regularized ventricular rates. This observation prompted further analysis of heart rate variability (HRV) and clinical outcomes in patients with cAF hospitalized for COVID-19.

| Study population and design
We performed a retrospective review of all admitted COVID-19 patients with the diagnosis of cAF between 1 March 2020 and 30 June 2020 at Thomas Jefferson University Hospital in Philadelphia, USA.
Clinical and electrocardiographic (ECG) data were collected from a shared electronic health records system and the hospital ECG database. This study was determined to be exempt from review by the institutional review board in accordance with institutional policy.
Patients were required to have at least one recorded ECG in AF during a previous non-COVID-19 hospitalization, in addition to at least one ECG during a COVID-19 hospitalization to be included in the study. Patients with implanted electronic pacing devices were excluded from this study, even if intermittent native atrioventricular (AV) conduction was present.

| ECG analysis
Pre-COVID-19 ECGs with heart rates (HRs) that were within 10 beats per minute of the COVID-19 ECGs were selected for comparison.
Only ECGs obtained during hospitalization were included. For intubated patients, ECGs were collected prior to intubation where patients were not under the influence of sedatives or sympathomimetics. Mean HR and each RR interval in milliseconds (ms) were measured using electronic on-screen calipers. HRV was assessed by calculating time-domain measurements, including standard deviation of successive differences in RR intervals (SDSD), root mean square of successive differences in RR intervals (RMSSD), and the proportion of number of pairs of successive RR intervals that differ by more than 50 ms (pNN50). These measurements were calculated based on formulas previously defined and validated in the literature. 8

| Clinical characteristics and outcomes
The electronic health record of patients was reviewed to collect relevant baseline clinical characteristics, including age, gender, beta-blocker, calcium channel blocker, and antiarrhythmic drug use.

| Statistical analysis
Each patient served as their own control. Means of continuous variables were analyzed using an independent sample t test, and categorical variables were analyzed using the chi-square test. Kaplan-Meier curves and the log-rank test were used to compare survival stratified by HRV. A two-sided P-value of <.05 was used to determine statistical significance. Analyses were performed using STATA/SE 16.1 (College Station, TX, USA).

| RE SULTS
A total of 38 patients were included in the study. There were 184 ECG tracings available for analysis, 95 ECGs from 44 pre-COVID-19 hospitalizations and 89 ECGs from 38 COVID-19 hospitalizations.

| Clinical characteristics and outcomes stratified by degree of heart rate variability
One patient died of unknown etiology. The Kaplan-Meier analysis in Figure 1 demonstrates that those with reduced pNN50 were less likely to have survived after a follow-up of 60 days (log-rank test P =.02). However, mortality was not associated with stratification by SDSD and RMSSD. There was no difference in the need for admission to the ICU, need for nonsurgical intubation, hospital LOS, or ICU LOS between the two groups when stratified by any index of HRV. international normalized ratio, high-sensitivity troponin T, pro-braintype natriuretic peptide, fibrinogen, and absolute neutrophil count; nadir of absolute lymphocyte count). The heart is abundantly innervated by autonomic nerves, with autonomic control consisting of intrinsic and extrinsic ganglia. 9 The extrinsic sympathetic control of the heart is largely mediated by the cervical, cervicothoracic, and thoracic ganglia. The extrinsic parasympathetic control of the heart is largely mediated by the vagus nerve. [10][11][12] Clusters of intrinsic ganglia that are composed of ganglionated plexi are located in the atria, where they are innervated by adrenergic and vagal nerve endings close to the pulmonary vein ostia. [10][11][12][13] These clusters of intrinsic ganglia regulate the interactions between the extrinsic and intrinsic nervous systems.

| D ISCUSS I ON
Heart rate variability analysis is widely used as a noninvasive method to characterize the influence of the ANS on sinus rate. 14 Increases in parasympathetic tone are believed to increase HRV, where vagal withdrawal reduces HRV. 14 Note: CRP C-reactive protein, INR international normalized ratio, hs-TnT high-sensitivity troponin T, pro-BNP pro-brain natriuretic peptide. Data presented as number (%) or mean (standard deviation).
of adverse cardiac outcomes in studies where autonomic modulation of the sinus node has been studied. [16][17][18][19] Lower HRV has been demonstrated in a number of conditions, including coronary artery disease, congestive heart failure, diabetes, and hypertension. [20][21][22] In sinus rhythm, high HRV demonstrates a favorable cardiovascular adaptive response to various endogenous and exogenous factors. 23 AF itself is marked by high HRV, which is secondary to the pattern of AV conduction in AF by influencing the conduction properties of the AV node. [24][25][26] Reduced HRV in patients with AF has been shown to be an independent predictor of cardiovascular and all-cause mortality. 27

Multiple reports have described new-onset AF in patients diagnosed
with COVID-19, but the electrophysiologic impact of COVID-19 in patients with established cAF has not been described. 28

| CON CLUS ION
This is the first study to describe the electrophysiologic effect of COVID-19 on AV conduction in patients with cAF. We found that in patients with cAF, HRV was reduced during COVID-19 compared with prior illnesses. This observation occurred irrespective of the use of AV nodal blocking drugs. Importantly, patients with the most reduced HRV was associated with increased mortality when stratified by pNN50. Depressed HRV in AF may reflect changes in autonomic control of AV conduction, and if confirmed, may be another marker of cardiac injury in COVID-19.

CO N FLI C T O F I NTE R E S T S
The authors declare that they have no competing interests. The results presented in this paper have not been published previously in whole or part, except in abstract form.

AUTH O R S' CO NTR I B UTI O N S
BP design, manuscript, supervision. JJ data collection, analysis, manuscript. JR data collection, analysis, manuscript.

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
This study was determined to be exempt from review by the institutional review board in accordance with institutional policy. This article does not contain any studies with animals performed by any of the authors.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data are safely kept in a password-protected security system at Thomas Jefferson University Hospital. The data sets used and/or analyzed during the current study are deidentified and available from the corresponding author on reasonable request.