Risk factors, thromboembolic events, and clinical course of New‐Onset Atrial Fibrillation among COVID‐19 hospitalized patients: A multicenter cross‐sectional analysis in Iran

Abstract Background and Aims We focused on determining the risk factors, thromboembolic events, and clinical course of New‐Onset Atrial Fibrillation (NOAF) among hospitalized coronavirus disease (COVID‐19) patients. Methods This retrospective study was conducted in the major referral centers in Tehran, Iran. Of 1764 patients enrolled in the study from January 2020 until July 2021, 147 had NOAF, and 1617 had normal sinus rhythm. Univariate and multivariate Logistic regressions were employed accordingly to evaluate NOAF risk factors. The statistical assessments have been run utilizing SPSS 25.0 (SPSS) or R 3.6.3 software. Results For the NOAF patients, the age was significantly higher, and the more prevalent comorbidities were metabolic syndrome, heart failure (HF), peripheral vascular disease, coronary artery disease, and liver cirrhosis. The multivariate analysis showed the established independent risk factors were; Troponin‐I (hazard ratio [HR] = 3.86; 95% confidence interval [CI] = 1.89−7.87; p < 0.001), HF (HR = 2.54; 95% CI = 1.61−4.02; p < 0.001), bilateral grand‐glass opacification (HR = 2.26; 95% CI = 1.68−3.05; p = 0.002). For cases with thromboembolic events, NOAF was the most important prognostic factor (odds ratio [OR] = 2.97; 95% CI = 2.03−4.33; p < 0.001). While evaluating the diagnostic ability of prognostic factors in detecting NOAF, Troponin‐I (Area under the curve [AUC] = 0.85), C‐Reactive Protein (AUC = 0.72), and d‐dimer (AUC = 0.65) had the most accurate sensitivity. Furthermore, the Kaplan‐Meier curves demonstrated that the survival rates diminished more steeply for patients with NOAF history. Conclusion In hospitalized COVID‐19 patients with NOAF, the risk of thromboembolic events, hospital stay, and fatality are significantly higher. The established risk factors showed that patients with older age, higher inflammation states, and more severe clinical conditions based on CHADS2VASC‐score potentially need subsequent preventive strategies. Appropriate prophylactic anticoagulants, Initial management of cytokine storm, sufficient oxygen support, and reducing viral shedding could be of assistance in such patients.


| INTRODUCTION
The new coronavirus disease  pandemic, precipitated by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), 1 enormously affecting worldwide communities, has resulted in exceeding 565 million afflicted individuals and over 6 million mortalities as of late July 2022. 2 The most prevalent clinical symptoms are acute respiratory distress syndrome (ARDS) and Interstitial pneumonia. [3][4][5][6] Also, the cardiovascular system has been reported to be frequently affected in COVID-19 patients, and those with cardiovascular involvement are facing a greater risk of worse outcomes. 7 Atrial fibrillation (AF) is a major etiology for embolism and stroke, particularly if it is not treated with anticoagulative therapy as a preventive stroke measure. 8 Acute respiratory infections are shown to be represented as one of the risk factors of the new-onset or recurring AF Stimulation of the sympathetic nervous system, hypoxia, dehydration, electrolyte abnormalities, metabolic dysfunction, and of course, myocardial injuries and inflammation, as consequences of viral pneumonia, increase the likelihood of New-Onset Atrial Fibrillation (NOAF) and following thromboembolic events. [8][9][10] It is well understood that the thrombogenic phenomena in AF are not confined to local causes such as defective atrial contraction or stasis. In addition, a generalized hypercoagulative condition has been postulated. 11 This raises the possibility that additional procoagulant and proinflammatory conditions, such as COVID-19 pneumonia, might synergistically influence cardiovascular mishaps. 8,11 Moreover, the virus SARS-CoV-2, regardless of causing NOAF, might increase susceptibility to thrombotic illness in both the atrial and venous circulations due to stasis, endothelial dysfunction, platelet activation, and severe inflammation. 12 Notwithstanding, it is currently uncertain whether SARS-CoV-2 causes hemostatic alterations or, as seen in other viral illnesses, are the product of a cytokine cascade that precedes the start of systemic inflammatory response syndrome. 8,13 Since understanding arrhythmic complications in COVID-19 is still evolving, we have aimed to assess the clinical characteristics and prognostic factors of NOAF among hospitalized COVID-19 cases. Also, we evaluated the clinical outcomes of such patients, particularly the thromboembolic events, as a retrospective observational multicenter analysis.

| Data collection
Patients' information was gathered and analyzed from the hospital medical records and was comprised of demographical characteristics, clinical course during hospitalization, prior medical history, therapy, and in-hospital outcome. The data were double-checked independently by four medical researchers. Laboratory tests have been conducted upon admission to the hospital and during the hospitalization period when clinically indicated. Lung spiral computed tomographic scan, and echocardiography have been carried out upon clinical indications.

| NOAF definition
The NAOF was defined as the first occurrence of AF upon or during admission in the hospital wards or intensive care units (ICU) in those hospitalized with COVID-19 and did not have any document or selfstatement regarding previous neglected or managed AF The report of NOAF was conducted in each of the following circumstances: (1) AF lasting for more than 1 h in the electrocardiogram, (2) AF with any duration of existence required cardioversion, (3) AF for which anticoagulation therapeutics were initiated based on CHA2DS2-VASc score criteria. Routine electrocardiogram, pulse rate, and oxygen saturation monitoring were conducted for all patients hospitalized in either wards or ICUs in our mentioned centers. The nursing alarm system report performed the initial suspicion of arrhythmia (nurse/patient ratio: 1/3). The final diagnosis of NOAF, requirement for cardioversion, and initiation of anticoagulants were considered and judged by skilled cardiologists, intensivists, or emergency medical professionals. Further, we excluded patients receiving each of the following high-risk drugs leading to NOAF based on literature 16 : (1) Any antineoplastic medication (we have excluded patients with malignancy), (2) highdose corticosteroids with a dosage of ≥7.5 mg/day equivalents with prednisone, 17 (3) Cardiovascular medications of adenosine, dobutamine, or milrinone, (4) Opiates, cannabis, or methamphetamines.

| Other definitions
The presence of AF and its subtypes were defined based on the recent AF guidelines. NOAF was described as the presence of an AF at ECG throughout the stay in the hospital that was not present upon admission.
Since there was not an organized and common internationally Since alcohol consumption has been shown as an independent factor for NOAF incidence, we added this variable in our history with the definition of >30 cc/day in men and >20 cc/day in women as the remarkable alcohol use. 18

| Statistical analysis
The patients were categorized based on whether they were affected by NOAF or not as a comparison. A total of 1764 patients (147 with NOAF and 1617 normal sinus rhythm [control group]) were evaluated in the analysis. We attempted to base most of this study's analytical and interpretational works on a guideline developed by Assel et al. 19 The Shapiro-Wilk test was used to check the normal distribution of the data. The categorical data have been provided as numbers (%) and compared employing the χ 2 test or Fisher's exact test; Continuous data having normal distribution have been provided as mean (standard deviation) then compared employing the indepen-   Table 1. The age with an average of 66.6 years among all patients was found to have an average of 70.7 years in the NOAF cohort and 66.2 years within the control cohort (p < 0.001). As displayed in Figure 1, for the cases in the NOAF cohort, with the increase in age, the prevalence of AF has increased as well, taking a somewhat exponential trend. This means that the prevalence of AF patients might most probably rise as their age increases. Moreover, the total age distribution of all patients is illustrated by the size of bubbles scattered through the diagram. Most of the patients in both study groups have been of the male gender (76.1% vs. 76.9%; p = 0.81). The mean BMI has been recorded to be 28.9 in the NOAF group and 27.7 in the control group (p = 0.05). The rate of alcohol consumption has been similar in both study groups (p = 0.65). However, the active smoking condition reported as pack-year had an average of 9.4 years in the NOAF cohort, which was markedly greater than 8.2, the average of the control group (p < 0.001). As  Calcium, mg/dl 8.6 (7.9−9.6) 8.5 (7.9−8.9) 8.6 (8.0−9.7) 0.14 Magnesium, mg/dl 1.  The area under the curve (AUC) for each of these five laboratory markers is a measure of the ability of that classifier to make a distinction. Therefore, the higher the AUC, the better the performance of that marker at distinguishing between having or not having AF. Accordingly, as depicted in this figure, our findings showed that the laboratory marker with the highest AUC is

| Within-hospital events
Among the administered therapeutics mentioned in Table 2

| Within-hospital mortality
Eventually, in-hospital mortality, as the ultimate adverse outcome, was recorded for 17.7% of patients in the NOAF group, while for only 6.4% of individuals within the control group (p < 0.001).

| Risk factors of thromboembolic events
As shown in Table 4

ACKNOWLEDGMENTS
The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

CONFLICT OF INTEREST
The authors declare that no conflict of interest.

TRANSPARENCY STATEMENT
The lead author Farzad Esmaeili Tarki affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions ORCID Farzad Esmaeili Tarki http://orcid.org/0000-0002-2265-9995