Are Patients with COVID-19 Dying of or with Cardiac Injury?

To the Editor:

We read with great interest the paper by Du and colleagues presenting the clinical characteristics of 85 patients in Wuhan dying of coronavirus disease (COVID-19) (1). Around 70% presented comorbidities (hypertension, diabetes, and coronary heart disease), and 13 patients (16%) died from cardiac problems, namely cardiac arrest, acute coronary syndrome (ACS), and malignant arrhythmia (1). Cardiac involvement probably complicates severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in patients, but the true incidence (considering specific echocardiographic findings) and the attributable mortality are aspects not yet well clarified.

Very few reports have used echocardiographic criteria beyond biomarkers to diagnose cardiac injury, but none have differentiated between myocarditis, cardiomyopathy (stress or septic), ACS, and acute heart failure in the era of COVID-19. Acute cardiac injury was reported in 44.7% of the fatalities in the report by Du and colleagues, but the specific echocardiographic abnormalities are not presented (1). Did these “cardiac injuries” involve patients with myocarditis? Or were there features indicative of stress or even septic cardiomyopathy, mostly reversible entities? Considering biomarkers, troponin levels are markedly increased in myocarditis and ACS. On the contrary, in Takotsubo and septic cardiomyopathy, there is a disparity between biomarker levels and the extent of myocardial dysfunction. In addition, hypoakinesia usually does not correspond to a specific coronary artery territory (2). Therefore, a reference on the nature of cardiac injury would be worthy.

A diagnosis of “cardiac injury” mainly relying on biomarker levels may be misleading. In a recent report involving 416 hospitalized patients from Wuhan, 19.7% presented with “acute myocardial injury.” The diagnosis relied on increased cardiac biomarker (hypersensitive troponin I) levels, regardless of the electrocardiographic and echocardiographic findings (3). Du and colleagues presented a high percentage of patients with “cardiac injury”; data on lactate dehydrogenase, creatinine kinase, and aspartate aminotransferase are reported but not on cardiac-specific enzymes (1). On the other hand, cardiac-specific biomarkers alone may not be diagnostic of cardiac damage. TnI is elevated in septic shock, pulmonary embolism, and critically ill patients in ICU. In patients with “cardiac injury,” NT-proBNP (N-terminal prohormone of brain natriuretic peptide) levels were found to be elevated (4). However, we have previously found that BNP is a biomarker that correlates with the severity of sepsis (5). BNP may be elevated when patients with SARS-CoV-2 present septic shock resulting from a superinfection, even with normal cardiac function. Additionally, the troponin and BNP levels were normal in a 64-year-old female patient from our ICU, who acutely established pericarditis on the 16th day after COVID-19 diagnosis.

Moreover, in Figure 1C of Du and colleagues, they present a computed tomographic image of a 23-year-old female patient with COVID-19. The cardiac structure seems greatly enlarged; considering the young age of the patient, this finding could correspond to true myocarditis (therefore, ground glass opacities could depict hydrostatic pulmonary edema) (1). It would be informative if the authors provided data on this aspect (increased cardiac dimensions on computed tomographic imaging, a finding beyond the criteria used for “cardiac injury” diagnosis). Inciardi and colleagues reported a 53-year-old woman with COVID-19 who presented acute myopericarditis and cardiogenic shock with severe systolic dysfunction, confirmed with magnetic resonance imaging. Noteworthy, the patient never presented signs of respiratory involvement (6).

Finally, data on the attributable to cardiac injury mortality are totally lacking (1). The proportion of the patients with “cardiac injury” who actually died because of cardiogenic shock is not mentioned. Markers of perfusion, such as low central venous oxygen saturation, would add information on the contribution of cardiac dysfunction to the fatal outcome. Furthermore, did the patients, dying of malignant arrythmia and cardiac arrest, suffer from cardiac comorbidities? Did the arrhythmia occur on a substrate of “myocardial injury,” or was this a complication of the prescribed medications (i.e., chloroquine)? All these issues need to be clarified to thoroughly understand the “myocardial damage” that COVID-19 induces.