ABSTRACT
Background Patients with COVID-19 can develop myocardial injury and arrhythmia during the course of their illness. However, the underlying risk factors for the development of cardiovascular related manifestations are unclear.
Methods Using a register-based multi-center cross-sectional design, we analyzed 80 patients with myocardial injury and 401 controls, as well as 71 patients with arrhythmia and 409 controls, all admitted with COVID-19. Putative risk factors for myocardial injury and arrhythmia were evaluated with logistic regression with adjustment for potential confounders.
Results COVID-19 patients with myocardial injury had fatigue (66.2%) and dyspnea (63.7%), while those with arrhythmia had dyspnea (71.8%). Patients with myocardial injury and arrhythmia had a significant mortality of 92.5% and 94.4%, respectively. A history of chronic obstructive pulmonary disease (COPD) or heart diseases was associated with an increased risk of myocardial injury (odds ratio [OR] = 1.94, 95% confidence interval [CI]: 1.01-3.71; OR = 7.43, 95% CI: 3.99-13.83) and arrhythmia (OR = 1.94, 95% CI: 1.00-3.75; OR = 13.16, 95% CI: 6.75-25.68). In addition, we found that gamma glutamyltranspeptidase (GGT) >50U/L (OR = 2.14, 95% CI: 1.37-3.32; OR = 1.85, 95% CI: 1.19-2.85), serum creatinine >111μmol/L (OR = 8.96, 95% CI: 4.4-18.23; OR = 3.71, 95% CI: 2.01-6.85), serum sodium <136 mmol/L (OR = 4.68, 95% CI: 2.46-8.91; OR = 2.06; 95% CI: 1.06-4.00) were all associated with increased risk of myocardial injury and arrhythmia, respectively.
Conclusion Our reported clinical characteristics and identified risk factors are important for clinical study of COVID-19 patients developing myocardial injury and arrhythmia.
INTRODUCTION
The outbreak of new coronavirus disease (COVID-19) was characterized as a pandemic by World Health Organization on March 11, 20201, 2. COVID-19 may cause pneumonia with severe respiratory symptoms that may progress to acute respiratory distress syndrome (ARDS) and even death. Common signs of COVID-19 are fever, cough, muscle soreness, weakness, breathing difficulty and dyspnea3. During an infection, COVID-19 patients may develop varying degrees of myocardial injury3-6. In a previous study by Huang et al.1, 5 out of 41 study participants were diagnosed with acute myocardial injury. The main manifestations of acute myocardial injury have been hypersensitive cardiac aggregates, troponin I (hs-cTnI) and elevated blood pressure. Most patients with significant myocardial injury required treatment in the Intensive Care Unit (ICU)3-6. Other studies have shown that plasma biomarkers of myocardial injury, including creatine kinase (CK-MB), and troponin I were higher in ICU patients than those not treated in the ICU, providing evidence that ICU patients are more likely to develop cardiac complications such as myocardial injury3.
It has been suggested that systemic inflammatory responses and immune system disorders lead to increased rates of cardiovascular symptoms1, 7-9. SARS-CoV-2 causes a severe inflammatory response, which may play a role in the development of myocardial infarction and arrhythmias. In a retrospective study of 138 hospitalized COVID-19 patients by Wang et al, nearly 17% of patients developed arrhythmia and 7.2% experienced acute cardiac injury3.
The clinical presentation of COVID-19 patients with myocardial injury or arrhythmias is different in many ways from patients without cardiac involvement. Fatigue, anorexia, dyspnea, vomit and diarrhea are were more commonly observed for COVID-19 patients with myocardial injury or arrhythmias than the latter patients. To date, there has not been any clinical research investigating the risk factors related to myocardial injury and arrhythmia in patients with COVID-19. Furthermore, there are no studies which explore whether or not the development of myocardial injury and arrhythmia influences the prognosis of COVID-19 patients, although intuitively, one would expect patients with cardiac manifestations to have a worse prognosis. Here, we have recruited 481 COVID-19 patients used a registry-based multi-center cross-sectional design from a retrospective study. We evaluated clinical characteristics of 80 patients with myocardial injury and 71 patients with arrhythmias. We further evaluated the associations of potential risk factors with myocardial injury and arrhythmia caused by COVID-19.
METHODS
Study design and data collection
This was a retrospective cross-sectional study of medical records from three medical centers in Hubei province. Oral informed consent was obtained from all subjects, and the institutional review boards of all participating institutions approved the study protocol. We recruited 481 admitted COVID-19 patients from Wuhan Union West Hospital (n=265), Wuhan Hannan General Hospital (n=153) and Enshi General Hospital (n=63) from January 8th, 2020 to April 20th, 2020. The end-of-study date was April 24th, 2020. We reviewed nursing charts, clinical records, laboratory findings, and imaging results for patients with a COVID-19 diagnosis in the three hospital systems based on the National Health Commission and CDC, China protocol (the 5th edition). These data have not yet been published.
Study outcomes
Acute myocardial injury was defined by serum levels of troponin T (TnT) at the 99th percentile or greater, as previously described1. Malignant arrhythmia was defined as rapid ventricular tachycardia lasting more than 30 seconds, inducing hemodynamic instability and/or ventricular fibrillation, as per the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards (ACC/AHA/HRS). Health services utilization patterns were analyzed as binary variables (yes or no) to ventilation, antibiotics, antifungals, antiviral agents and hemodialysis.
The following demographic and clinical features were identified from patients’ medical records: socio-demographic variables included sex (male and female), race (Han, others), age (continuous variable), and smoking. Disease severities included the number of symptoms, unilateral or bilateral infection, and imaging characteristics (ground-glass opacification/opacity, yes/no). Comorbidity variables included cardiovascular diseases such as hypertension, underlying heart disease and coronary artery disease, chronic kidney and/or liver disease, chronic obstructive airways disease (COPD), diabetes and others. Laboratory tests analyzed included complete blood count, comprehensive metabolic panel (CMP) and clotting factors.
Statistical Analysis
Kruskal-Wallis tests were applied for continuous variables among patients from the three different regions. Categorical variables were displayed as counts or percentages. Chi-square tests were applied for categorical variables among patients from different regions. The associations of risk factors with myocardial injury and arrhythmia were evaluated using logistic regression models with an adjustment of potential confounders including after adjusting for age, gender, smoking status, and disease history. Continuous variables were summarized as medians. All statistical analyses were implemented using SAS, version 9.4 (SAS Institute, Inc.). A P value <0.05 for a two-tailed test was considered statistically significant.
RESULTS
Clinic characteristics of study participants with myocardial injury
Socio-demographic and clinical characteristics of the 481 study participants including 80 cases of myocardial injury and 401 controls, were shown in Table 1. The median age (SD) of the study participants was 56. Of the patients with myocardial injury, 66.2% had fatigue and 63.7% had dyspnea. Disease severity was more likely to be higher in patients with myocardial injury (2.5% were mild, 2.5% were moderate and 95% were severe, P<0.001). In contrast, 2/3 of the control subjects had mild disease. Of the patients with myocardial injury, 7.5% recovered and were discharged from hospital, while 92.5% died (P<0.001).
We observed that age was associated with an increased risk of myocardial injury case (P<0.01; Table 1).We also observed that there was a male predominance for myocardial injury (70% vs 30%, P=0.003; Table 1). There was no such male predominance in the control group. No significant associations with myocardial injury were observed for smoking status and maximum temperature during fever.
Multivariate analysis between laboratory parameters and myocardial injury
Patients with an abnormal GGT were associated with an increased risk of developing myocardial injury (OR = 2.14, 95% CI: 1.37-3.32; Table 2). Elevated serum creatinine (adjusted OR = 8.96, 95% CI: 4.4-18.23) and reduced serum sodium (adjusted OR = 4.68, 95% CI: 2.46-8.91) were also associated with an increased risk of myocardial injury. However, white blood cell count, lymphocyte count, total bilirubin, alanine aminotransferase, aspartate aminotransferase (AST) and blood urea nitrogen (BUN) did not appear to be associated with the risk of myocardial injury. In addition to these laboratory parameters, we observed that the uses of antibiotics (adjusted OR = 3.07, 95% CI: 1.58-5.95) and antiviral therapy (OR = 2.18, 95% CI: 1.2-3.93) were associated with an increased risk of myocardial injury. However, AST, albumin, BUN did not appear to be associated with the risk of malignant arrhythmia caused by COVID-19.
Clinic characteristics of study participants with arrythmias
Socio-demographic and clinical characteristics of 71 cases of arrhythmia and 409 controls, were analyzed (Table 1). Patients with arrhythmia (Median age=61) had significantly older than controls (Median age=56) – 66.2% of cases were male. No significant differences were found for smoking status and highest fever between cases and controls.
Of the 71 arrythmia cases, 71.8% had dyspnea – 95.8% participants with arrhythmia were classified as having severe COVID-19 disease. The outcome of patients with arrhythmia was poor, with only 5.6% recovering and being discharges and 94.4% dead (P<0.001).
Multivariate analysis of laboratory parameters with arrythmias
Table 2 illustrates the association between different laboratory parameters and arrythmias. Patients with increased GGT had a 1.85-fold increase in risk of arrhythmias when compared with those with normal GGT (95% CI:1.19-2.85). In addition, the elevated serum sodium was associated with increased risk of arrhythmia (adjusted OR = 2.06, 95% CI:1.06-4.00). In addition to these laboratory parameters, we observed that the uses of oxygen therapy (adjusted OR = 3.11, 95%CI: 1.65-5.83) and antibiotic therapy (OR = 2.03, 95% CI: 1.1-3.76) were associated with an increased risk of arrhythmia. However, AST, albumin, and BUN did not appear to be associated with the risk of arrhythmia caused by COVID-19 as well.
DISCUSSION
In a large study of Covid-19 patients with and without myocardial injury and arrhythmia we found that older age, male sex, history of COPD and heart diseases, electrolyte disturbance and renal dysfunction were associated .With the exception of asymptomatic carriers, most COVID-19 patients may develop pneumonia with severe respiratory symptoms1, 10. The more severe cases progress to multiple organ failure or even death.
It is believed that a cytokine release syndrome or cytokine storm which is triggered by the host response to the virus is responsible for acute injury to multiple organ systems including the heart, liver, and kidneys. It is becoming apparent that cardiovascular disease resulting from COVID-19 is not as uncommon as previously thought, and that the involvement of the heart prone to more severe disease and increased mortality. Thus, the identification of risk factors for arrythmias and acute myocardial injury is important for prognostication, in addition to early intervention, early diagnosis and treatment, which may help reduce morbidity and mortality.
A previous study showed that 53.3% of COVID-19 patients with heart damage were men. In that study, myocardial injury was negatively correlated with age and the incidence of myocardial injury was lowest over the age of eighty. In our study, indicators of cardiac injury were generally negative in the patients with mild and moderate disease. AST, LDH, CTN I and MYO showed a strong correlation with myocardial injury. This is consistent with the results of Zhong et al. 11. We also found that patients who were not treated with glucocorticoids (GCs) and immunoglobulin therapy appeared to be at higher risk for heart damage. The protective role of GCs may be related to their immunosuppressive effects on leukocytes and endothelial cells to reduce the generation and release of pro-inflammatory cytokines that targetcardiomyocytes12. The glucocorticoid receptor (GR) is expressed abundantly in cardiomyocytes, and the variability of acute cardiac injury may be related to polymorphisms of the glucocorticoid receptor. In addition, the GR can be selectively activated and cyclooxygenase (COX)-2 and prostaglandin (PG) biosynthesis in cardiomyocytes can be induced, which may help mediate cardioprotection13, 14. Biosynthesis of prostaglandin can inhibit COX-2, and, therefore, relieve inflammation derived myocardial damage in COVID-19 patients.
We found that white blood cell numbers, lymphocyte numbers, neutrophil numbers, albumin, total bilirubin, alanine aminotransferase, AST, serum sodium, BUN, serum creatinine, GGT and myoglobin were positively correlated with myocardial injury. Elevated neutrophil counts led to a 3.35-fold increase in acute cardiac injury. Neutrophils are often the first type of leukocyte that infiltrate the heart in times of hyperimmune status. Their activation produces large amounts of inflammatory mediators, which mediate injury to the heart through the action of reactive oxygen species and proteolytic enzymes, leading to reduced oxygenation of the heart tissue15-17. Neutrophils can also induce migration of They are also involved in macrophage migration to the infarct site in order to clear debris. The inflammatory process induced by neutrophils involves a process known as netosis through the generation of neutrophil extracellular traps (NETs).
We observed a negative correlation between albumin levels and risk of myocardial injury. This relationship has never been reported18. Serum albumin (SA) may exert a cardio-protective effect by functioning as a natural inhibitor of the angiotensin-converting enzyme (ACE). ACE activity has also been associated with coronary vascular resistance in recent animal studies. The cardio-protective effect of ACE inhibition in cardiac surgery has been previously demonstrated19-23. Within the physiological range of SA concentrations (35-52 mg/ml), almost all ACE is bound to albumin (and thus inactive). Below the physiological range of SA concentrations there is a rapid increase in the fraction of unbound (and thus active) ACE. The drop in SA in patients with heart damage is an important observation with potential therapeutic implications, including the close monitoring of albumin in patients admitted for COVID-19 pneumonia. Myoglobin, a useful marker of cardiac damage, has also been positively associated with a risk of heart damage24.
In our study, the arrythmia group had a higher number of elderly subjects (except for age >80), whereas in the control group, there was an equal distribution in the <40, 40-60- and 60-80-year age categories. AST, LDH, CTN I and MYO also showed a strong correlation with arrhythmia. Patients who were not treated with glucocorticoids (GCs) and immunotherapy had a higher incidence of arrhythmias, in concert with the observations in myocardial injury, most likely through the same anti-inflammatory mechanism.
LIMITATIONS
One important limitation of our study is that it does not fully elaborate on whether patients were admitted to hospital with cardiac disease already, which could mean that we would be studying the consequences of cardiac disease rather than risk factors for cardiac disease. Although, we do attempt to elucidate this via each respective patient’s medical history. In our study, timing is also crucial: our study is not strictly a cross-sectional study where variables present at the same time were observed - this study is about patients who all had COVID-19, and some also had heart disease around that time. It is important to know whether heart disease began before or after admission for COVID-19.
CONCLUSION
In conclusion, our study produced valuable clinical data related to the study of myocardial injury and arrhythmias in COVID-19 patients. We identified several risk factors associated with myocardial injury and arrhythmias in COVID-19 patients. Our findings provide potential markers for the further development of clinical guidance and treatment for COVID-19 patients with a considerable risk of developing severe cardiac manifestations.
Data Availability
n/a
Declaration of interests
All authors declare no competing interests.
Acknowledgments
We thank all the patients and their families, and the clinical staff who treated the patients in Hannan, Enshi and Union medical centres. We would also like to thank Troy Gharibani at University of Maryland, USA for editing and providing feedback on this article.
Footnotes
Disclosure: the authors have nothing to disclose.