3. Results

3.1. Baseline Characteristics and Outcomes in Overall Cohort
Median age was 64.1 ± 16.7 years, and the majority of patients were male (n = 108, 65.1%). Arterial hypertension constituted the most common risk factor (n = 83, 50.0%). Cardiovascular disease was present in 18.1% (n = 30), cardiomyopathy in 3.0% (n = 5), diabetes mellitus in 17.5% (n = 29), and obesity in 21.7% (n = 36). Other relevant comorbidities comprised pulmonary disease (e.g., chronic obstructive pulmonary disease, pulmonary fibrosis) in 14.5% (n = 24), previous or active cancer disease in 8.4% (n = 14), and conditions associated with immunodeficiency (e.g., due to chronic immunosuppression therapy after organ transplantation or hematological disease) in 8.4% (n = 14). Information on left ventricular ejection fraction (LVEF) was available in 44 patients (26.5%). Mean LVEF in this subgroup was 53.0 ± 12.3%. In three patients a cardiac pacemaker had been implanted (1.8%). Implanted cardioverter-defibrillators (ICDs) were present in two patients (1.2%). Seven patients (4.2%) presented with syncope in association with COVID-19 prior to hospital admission.
Median duration of hospitalization was 10.5 days (P25: 5 days; P75: 22 days, n = 154). In 12 patients, data on duration of hospitalization could not be assessed due to transfer to a different center not participating in this study. The majority of patients required oxygen therapy in the course of hospital stay, and over a third of all patients was admitted to intensive care (ICU) or intermediate care units (IMC) (Figure 1). Median duration of ICU/IMC-therapy was 8 days (P25: 4 days; P75: 22.5 days). High-flow oxygen therapy and/or non-invasive ventilation (NIV) by continuous positive airway pressure (CPAP) was necessary in 39 patients (23.5%), and pharmacological circulatory support by vasopressors was provided in 30 patients (18.1%). Thirty-seven patients (22.3%) received mechanical ventilation with a median duration of 17 days (P25: 7.5 days; P75: 26 days). Of these, six patients had to be re-intubated after initially successful weaning (Figure 1). Only a minority of patients underwent extracorporeal membrane oxygenation (ECMO) or in-hospital cardiopulmonary resuscitation (Figure 1).
With respect to cardiovascular events, myocarditis was suspected in one patient based on cardiac biomarker-kinetics and mildly reduced LVEF in echocardiography. This patient died due to respiratory failure during the hospital stay. However, the diagnosis of myocarditis could not be confirmed upon autopsy. Four patients were diagnosed with myocardial infarction (2.4%) during hospitalization. Percutaneous coronary intervention (PCI) was performed in two cases, of which one patient presented with ST segment elevation myocardial infarction (STEMI) and one patient with non-ST segment elevation myocardial infarction (NSTEMI). In the other two cases diagnosed with NSTEMI, medical therapy alone was preferred due to clinical instability with predominant respiratory symptoms and stable echocardiographic and ECG findings. One patient with NSTEMI and PCI died in the course of hospitalization due to mesenteric ischemia. Stroke or transient ischemic attack (TIA) was seen in three patients (1.8%). Twenty-six patients died during hospitalization, predominantly due to respiratory failure (n = 20, 76.9%) or other non-cardiac reasons (n = 5, 19.2%). Only one death was attributed to cardiac causes (3.8%), i.e., cardiac circulatory failure in a mechanically ventilated patient with pre-existent severe cardiovascular disease.

3.2. Arrhythmias During Hospitalization for COVID-19
In our cohort 34 patients (20.5%) displayed arrhythmias during hospitalization. In 17 cases (10.2%), the arrhythmia type occurring during hospital stay had already been previously diagnosed in the respective patients prior to SARS-CoV-2 infection. Specifically, in 16 patients previously diagnosed AF occurred, in one patient previously observed bradycardia was recorded. Twenty-two patients (13.3%) displayed new-onset arrhythmia that had not been diagnosed before-either without any previous arrhythmia history or in addition to other previously diagnosed arrhythmia. Of these, 16 patients (9.6%) had never been diagnosed with any type of arrhythmia prior to hospitalization for COVID-19.
With regard to arrhythmia diagnosis during hospitalization for COVID-19, AF was the most common type of arrhythmia recorded (Figure 2A). Similarly, in patients diagnosed with a new arrhythmia type during hospitalization (Figure 2B) and in patients without any previous arrhythmia diagnosis prior to hospitalization (Figure 2C), AF constituted the most common incident arrhythmia. Bradycardia was recorded in four cases, of which one patient already had previously documented asymptomatic bradycardia. No pacemaker implantation was necessary in this subgroup. In addition, patients with frequent PVCs and ventricular tachycardia (VT) constituted relevant subgroups. All recorded VT-episodes were non-sustained. Ventricular fibrillation occurred in one patient with cardiovascular disease who had previously received an ICD due to ventricular tachycardia. In three patients (1.8%) electrical cardioversion was performed for termination of hemodynamically compromising AF. Amiodarone for pharmacological cardioversion was administered in four patients (2.4%), and chronic antiarrhythmic medical therapy was initiated in three patients (1.8%). Patients with arrhythmias more often received therapeutic anticoagulation therapy (Table 1) in accordance with the high proportion of patients with AF in this subgroup. Reasons for therapeutic anticoagulation in patients without arrhythmias were previous or new diagnosis of venous thrombosis or pulmonary embolism.

3.3. Clinical Predictors for New-Onset Arrhythmias
We analyzed predisposing factors associated with arrhythmias during hospitalization for COVID-19 comparing baseline characteristics of patients with arrhythmia and patients without arrhythmia during hospitalization for COVID-19 (Table 1). Patients with arrhythmias were older and more often had been diagnosed with hypertension and cardiovascular disease (Table 1). QTc duration at baseline was longer in patients with arrhythmia, however, median values were within the physiological range in both groups (Table 1). Furthermore, we analyzed the prevalence of potentially proarrhythmic medication administered in the context of COVID-19. In both subgroups, hydroxychloroquine was used in 44.1% and 44.6% of patients, respectively (Table 1). A smaller fraction of patients in both groups additionally received azithromycin. There was no statistically significant difference in the use of QT-prolonging drugs between both groups (Table 1). Median QTc-duration at baseline was within normal range in patients who later received hydroxychloroquine (409.0 ms; P25: 390.5 ms, P75: 421.5 ms).
Regularized logistic regression led to the selection of the variables age, cardiovascular disease and hypertension with respect to the prediction of arrhythmia incidence. The subsequently fitted multiple logistic regression model revealed significant association of age (OR 1.036; 95% CI 1.004–1.074; p = 0.036) and cardiovascular disease (OR 3.307; 95% CI 1.329–8.232; p = 0.01) with incident arrhythmia in COVID-19, whereas the effect of hypertension was not significant (OR 2.08; 95% CI 0.794–5.796; p = 0.144). As measure of the model’s performance an area under the curve (AUC) value of 0.74 (95% CI: 0.65; 0.84) was estimated by applying 5-fold cross-validation (Supplementary Figure S1). Left ventricular ejection fraction was documented in 47.1% of patients with incident arrhythmia and only 21.2% of cases without arrhythmia during hospitalization. Attempts at imputing LVEF and including the imputed dataset in the final logistic regression model hinted at a potential role of LVEF as an additional predictor for arrhythmia incidence (Table S1). However, due to the high number of missing values, LVEF was omitted from the final logistic regression model to ensure reliability of statistical analyses.
With regard to peak levels of cardiac and inflammatory biomarkers assessed during hospitalization, patients with arrhythmia displayed higher levels of hsTnT and NTproBNP (Figure 3A, B). Additionally, a more pronounced increase in IL-6 and LDH could be detected in the arrhythmia subgroup, whereas there was no statistically significant difference in peak levels of CRP between groups (Table 2, Figure 3C, D).

3.4. Prognostic Implications of Arrhythmia on Clinical Outcome
Overall duration of hospitalization was longer in patients with arrhythmia associated with COVID-19 (Table 3). Univariate analysis showed an increase of hospitalization duration of 11.4 days with the presence of incident arrhythmia (95% CI 6.05–16.7 days; p <0.001). Additionally, patients with arrhythmia were more often admitted to ICU or IMC wards (OR 2.37; 95% CI 1.10–5.09; p = 0.03), and incident arrhythmia was associated with a longer duration of hospitalization on ICU/IMC wards (Table 3). Patients with incident arrhythmia more often received vasopressors for circulatory support and non-invasive ventilation or high-flow oxygen-therapy (Table 3). Importantly, patients with arrhythmia more often presented with severe respiratory failure requiring mechanical ventilation (OR 6.69; 95% CI 2.92–15.35; p < 0.001). Duration of mechanical ventilation was not significantly different between patients with and without arrhythmia (Table 3).
With regard to cardiac events, myocardial infarction was more common in the patient group with arrhythmia, however, with a low overall number of events (Table 3). Stroke or transient ischemic attack (TIA) occurred in one case in the patient group with arrhythmia who had a prior diagnosis of AF and frequent PVCs, and in two patients without arrhythmia. All patients with myocardial infarction or stroke/TIA had received anticoagulation therapy with low-molecular-weight-heparin, however, in one patient heparin-therapy was paused after coronary angiography due to severe bleeding complications. In-hospital mortality was significantly elevated in COVID-19 patients with incident arrhythmia during hospitalization (OR 3.02; 95% CI 1.22–7.46; p = 0.02). Multiple regression analyses adjusting for differences in baseline parameters revealed that the incidence of arrhythmia constitutes a more powerful prognostic factor regarding hospitalization duration and the need for mechanical ventilation than age and prevalence of cardiovascular disease (Table 4). Finally, previous diagnosis of cardiovascular disease in our cohort was significantly associated with mortality in these analyses.

3.5. Subgroup Analysis of Patients with Atrial Fibrillation
Atrial fibrillation constituted the most common incident arrhythmia during hospitalization for COVID-19. In the subgroup analysis of patients with incident AF, age, hypertension and cardiovascular disease were associated with incidence of the arrhythmia (Table 5). Both cardiac and inflammatory markers showed a stronger increase in patients with AF. Similar to the effects of incident arrhythmia in the overall cohort, AF itself was associated with longer overall hospitalization times and longer duration of ICU/IMC care. Additionally, an increased need for high-flow oxygen therapy or non-invasive ventilation, mechanical ventilation and pharmacological circulatory support could be seen in this subgroup. Information regarding initiation of anticoagulation therapy was available in 10 patients with AF during hospitalization for COVID-19. In six cases, low-molecular-weight heparin (LMWH) in therapeutic doses was applied, two other patients received NOACs. Two patients only received prophylactic doses of LMWH, in one case due to a low CHA2DS2-Vasc-Score of 1 and a self-limiting AF-episode, in one case due to delayed diagnosis of AF. Two patients in the AF-subgroup were diagnosed with myocardial infarction. In both cases, anticoagulation therapy had been initiated with LMWH. One patient underwent PCI and died in the later course of the hospitalization due to mesenteric ischemia. In this patient anticoagulation therapy had been paused for 22 days after coronary angiography due to severe bleeding complications requiring transfusion therapy. The other patient received medical treatment as a type-II myocardial infarction was suspected due to stable echocardiography und ECG-findings. All cases of death in patients with AF were attributed to non-cardiac causes: in addition to the patient described above three patients deceased due to respiratory failure.