Cardiac arrhythmias have been observed in patients hospitalized with coronavirus disease (COVID-19). Most analyses of rhythm disturbances to date include cases of sinus tachycardia, which may not accurately reflect true cardiac dysfunction. Furthermore, limited data exist regarding the development of conduction disturbances in patients hospitalized with COVID-19. Hence, we performed a retrospective review and compared characteristics and outcomes for patients with versus without incident arrhythmia, excluding sinus tachycardia, as well as between those with versus without incident conduction disturbances. There were 27 of 173 patients (16%) hospitalized with COVID-19 who developed a new arrhythmia. Incident arrhythmias were associated with an increased risk of intensive care unit admission (59% vs 31%, p = 0.0045), intubation (56% vs 20%, p <0.0001), and inpatient death (41% vs 10%, p = 0.0002) without an associated increase in risk of decompensated heart failure or other cardiac issues. New conduction disturbances were found in 13 patients (8%). Incident arrhythmias in patients hospitalized with COVID-19 are associated with an increased risk of mortality, likely reflective of underlying COVID-19 disease severity more than intrinsic cardiac dysfunction. Conduction disturbances occurred less commonly and were not associated with adverse patient outcomes.
Mounting evidence shows that patients with coronavirus disease (COVID-19) are at risk of developing arrhythmias. Several early publications include sinus tachycardia in their definitions. Given that sinus tachycardia is often reflective of systemic disease rather than intrinsic cardiac dysfunction, including it in the definition of arrhythmia may bias such analyses. It is also still unclear if mortality associated with arrhythmias in COVID-19 is driven by disease severity versus primary cardiac dysfunction. Furthermore, limited data exist on conduction disturbances and other electrocardiographic (ECG) abnormalities in COVID-19. Hence, the purpose of this study is to report the incidence of atrioventricular arrhythmias and other ECG abnormalities in patients hospitalized with COVID-19 and to examine their associations with mortality.
All inpatient adults aged 18 years or older hospitalized for COVID-19 from March 2020 through June 2020 were included in this retrospective study. Only those with a confirmed nasopharyngeal polymerase chain reaction test were considered positive for COVID-19. Charts were reviewed for demographic information and medical comorbidities present on admission. ECG and telemetry data were reviewed for abnormalities throughout hospitalization. Any arrhythmia or other abnormality was defined according to standard accepted criteria. Atrial arrhythmias included atrial fibrillation, atrial flutter, frequent atrial premature complexes, and other supraventricular tachycardias such as atrioventricular nodal re-entrant tachycardia or atrioventricular re-entrant tachycardia. Ventricular arrhythmias were defined as ventricular fibrillation, ventricular tachycardia (VT), and ventricular premature complexes in a bigeminal or trigeminal pattern. Sinus bradycardia was considered significant at a heart rate <40 beats/min. Conduction disturbances were defined and analyzed per standard criteria and included atrioventricular conduction defects such as first, second, and third-degree atrioventricular block and intraventricular conduction delays such as right bundle branch block and left bundle branch block. Lesser degrees of intraventricular conduction delays, such as left anterior fascicular block or right anterior fascicular block, were also included. This study was approved by the Institutional Review Board of the Baylor Scott & White Research Institute.
Categorical variables are presented as frequencies and percentages. Continuous variables are presented as median (quartile 1, quartile 3). We compared continuous characteristics of patients with versus without incident arrhythmia using the two-sample t test and the Wilcoxon rank sum test, depending on normality. We compared discrete characteristics using the chi-square and Fisher’s exact test, depending on the expected cell count. Variables found to be associated with incident arrhythmia were considered for inclusion in a multivariable logistic regression model to assess the joint effect on the outcome of inpatient mortality. We created the adjusted (multivariable) logistic regression model using a data-driven stepwise selection process, which resulted in the minimum Akaike information criterion, indicating that the final model was both information-rich and simplistic. This resultant data-driven model provides insight into the relation between incident arrhythmia and inpatient mortality whereas jointly accounting for age and congestive heart failure (HF). Analyses were performed in SAS version 9.4 (Cary, North Carolina).
There were 173 patients hospitalized with COVID-19 included. Baseline traits are listed in Table 1 ; incident arrhythmia and ECG abnormalities are reported in Figure 1 . There were 27 patients (16%) with incident arrhythmia. Such patients were older and had a higher comorbidity burden (e.g., higher rates of type 2 diabetes mellitus, congestive heart failure, and peripheral vascular disease) than those without an incident arrhythmia ( Table 1 ). Patients with an incident arrhythmia were also more likely to have had a previous atrial arrhythmia. Of these 27 patients, 10 (6%) had new atrial fibrillation, 1 (<1%) had new atrial flutter, 2 (1%) had other new supraventricular tachycardias, 6 (4%) had new atrial premature complexes, 5 (3%) had new monomorphic VT, 1 (<1%) had new ventricular fibrillation, and 2 (1%) had new ventricular premature complexes. We did not detect a significant relation between the use of proarrhythmic drugs and the onset of arrhythmia ( Table 2 ). Patients with incident arrhythmia were more likely to be admitted to the intensive care unit (ICU) and be intubated than patients without an incident arrhythmia. Additionally, the inpatient mortality rate was significantly higher for patients who had incident arrhythmia compared with those who did not (41% vs 10%, p = 0.0002; Figure 2 ). After adjusting for age and congestive HF, the effect of incident arrhythmia remained significant, with an adjusted odds ratio 3.2 (95% confidence interval 1.06 to 9.76, p = 0.0393). Additionally, the combination of these 3 risk factors yielded an area under the receiver operating characteristic curve = 0.8, indicating a moderately strong discriminatory ability. Further, the 6-month mortality rates of the patients who survived to discharge showed a similar, but nonsignificant trend (19% vs 9%, p = 0.2180; including all patients: 52% vs 17%, p = 0.0013).
| Variable | Incident arrhythmia | ||
|---|---|---|---|
| Yes (n=27) | No (n=146) | P-value | |
| Age (years) | 79 [66, 91] | 60 [47, 71] | <0.0001 |
| Men | 17 (63%) | 82 (56%) | 0.5119 |
| Body mass index (kg/m 2 ) | 27 [21.2, 30.0] | 29 [24.8, 35] | 0.0382 |
| Type 2 diabetes mellitus | 9 (36%) | 47 (32%) | 0.7077 |
| Coronary artery disease | 5 (19%) | 10 (7%) | 0.0623 |
| Systemic hypertension | 11 (41%) | 42 (29%) | 0.2150 |
| Chronic kidney disease | 2 (7%) | 12 (8%) | 1.0000 |
| Congestive heart failure | 8 (31%) | 9 (6%) | 0.0009 |
| Peripheral vascular disease | 6 (22%) | 5 (3%) | 0.0023 |
| Prior atrial arrhythmia | 7 (26%) | 8 (6%) | 0.0028 |
| Prior ventricular arrhythmia | 0 | 1 (1%) | 1.0000 |
| Prior conduction disturbance | 6 (22%) | 16 (11%) | 0.1194 |
| QTC prolongation | 2 (7%) | 5 (3%) | 0.2999 |
| Prior CIED | 1 (4%) | 5 (3%) | 1.0000 |
| Variable | Incident arrhythmia | ||
|---|---|---|---|
| Yes (n=27) | No (n=146) | P-value | |
| Azithromycin use | 14 (52%) | 70 (48%) | 0.7091 |
| Ciprofloxacin use | 0 | 6 (4%) | 0.5917 |
| Hydroxychloroquine use | 4 (15%) | 19 (13%) | 0.7618 |
| Remdesivir use | 8 (30%) | 34 (23%) | 0.4801 |
| Decompensated heart failure | 2 (7%) | 14 (10%) | 1.0000 |
| New conduction disturbance | 2 (7%) | 11 (8%) | 1.0000 |
| New QT prolongation | 2 (7%) | 2 (1%) | 0.1154 |
| Other EKG abnormality | 0 | 7 (5%) | 0.5978 |
| Intensive Care Unit admission | 16 (59%) | 45 (31%) | 0.0045 |
| Intubation | 15 (56%) | 29 (20%) | <0.0001 |
| Length of stay (days) | 10 [4, 19] | 9 [5, 20] | 0.9716 |
| Death during hospitalization | 11 (41%) | 14 (10%) | 0.0002 |
| Death at 6 months | 14 (52%) | 26 (18%) * | 0.0013 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
