Since March 2020, the coronavirus disease (COVID-19) pandemic has become an important public health issue leading to increased morbidity and mortality1. Although initially it was thought that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes only lung injury (Figure 1), several studies discovered that cardiac complications may occur in 10–20% of patients, making the management of the disease more difficult, with worse outcome2. Moreover, COVID-19 patients with pre-existing cardiovascular disease have a higher risk of in-hospital mortality2. Myocardial involvement could be the cause of cardiovascular complications of the viral disease or exacerbation of preexisting cardiovascular diseases, leading to acute coronary syndromes (ACS), myocarditis, pericarditis, myocardial injury, arrhythmias, and pulmonary embolism (PE)3,4.
Thoracic computed tomography in a female patient with a moderate form of COVID-19. Bilateral infiltrates predominant at the level of the posterior and inferior pulmonary lobes, with a typical ground-glass aspect. Panel A. Acute phase of the disease; Panel B. Follow up at one month.
Electrocardiogram, various imaging techniques, and cardiac biomarkers are essential for the early diagnosis and timely management of cardiac involvement in COVID-19 patients in order to improve survival and long-term prognosis4. Thus, this review will focus on the electrocardiographic (ECG), imaging and biomarkers changes in SARS-CoV-2 disease with cardiac involvement.
ECG abnormalities in COVID-19 patients are nonspecific and dynamic, and reflect a wide spectrum of cardiovascular complications such as ACS, arrhythmias, acute myocarditis, pericarditis or PE (5). Therefore, the main ECG changes are:
- ST-T abnormalities: ST-elevation in leads DII, DIII, aVF and DI, aVL, ST-depression in leads V1-V6, deep T wave inversion in precordial leads or diffuse concave ST-elevation5,6,7; - QTc prolongation due to COVID-19 medication (hydroxychloroquine and azithromycin) that can lead to secondary torsades de pointes8,9; - arrhythmias: atrial fibrillation in 8.5% of patients, ventricular tachycardia in 3.5% of patients and Brugada pattern - two cases reported in the United States9,10,11; - S1Q3T3 pattern and sinus tachycardia: in PE patients12.
Cardiac biomarkers. COVID-19 is a systemic infection characterized by abnormalities of inflammatory, hematologic, thrombotic, and cardiac biomarkers13. The mechanisms underlying acute myocardial injury in SARS-CoV2 disease are incompletely elucidated and include increased myocardial consumption in response to viral infection, inflammatory response due to cytokines, and thrombogenic environment due to platelet activation, endothelial dysfunction and direct cytotoxic myocardial damage13. Thus, increased N-terminal pro-brain natriuretic peptide (NT-pro-BNP) and high-sensitivity cardiac troponin (hs-cTn) may not necessarily diagnose heart failure or myocardial infarction, but they correlate with a worse prognosis of COVID-19 patients13.
a) Troponin. As a quantitative marker of cardiomyocyte injury, increased troponin levels reveal acute myocardial injury, associated or not with pre-existing cardiovascular disease, in 25% of COVID-19 patients hospitalized in the intensive care unit (ICU)13. Moreover, elevated hs-cTn correlates significantly with 28-day mortality, but with lower cut-offs than the cut-off used for cardiac disease in non-COVID-19 patients14.
Isolated mild elevation of hs-cTn, below 3 times the upper limit of normal (ULN), may not require work-up or treatment for myocardial infarction unless strongly suggested by symptoms and ECG changes14. However, the troponin rise is explained by the combination of possible pre-existing cardiac diseases and/or associated acute myocardial injury in SARS-CoV2 disease14. Significant elevation of hs-cTn, more than 5 times the ULN, indicates a severe form of COVID-19, with shock, severe respiratory failure, tachycardia, systemic hypoxemia, myocarditis, Takotsubo syndrome or myocardial infarction13. Echocardiography should be considered for an initial diagnosis and to establish prognosis in COVID-19 patients14.
b) NT-pro-BNP, a quantitative biomarker of hemodynamic myocardial stress and heart failure, is frequently elevated among patients with severe inflammatory and respiratory diseases14. Increased level of NT-pro-BNP in COVID-19 patients represents a combination between presence or extent of pre-existing cardiac disease and acute hemodynamic stress related to SARS-CoV2 disease14. Similar to troponin, NT-pro-BNP is an independent risk factor for adverse outcome in patients with severe forms of COVID-1915. A NT-pro-BNP cutoff value of 88.64 pg/ml, lower than the threshold used to diagnose heart failure, predicts in-hospital death of COVID-19 patients with a sensitivity of 100% and a specificity of 66.67%15. c) High D-dimer level suggests thrombin formation, disseminated intravascular coagulation associated with shock or an acute response in systemic infections or inflammations16. Thus, markers of activated coagulation or impaired fibrinolysis might contribute to acute myocardial injury, affecting coronary capillaries16. Although D-dimer has a low specificity for the diagnosis of thrombosis, patients with COVID-19 pneumonia and increased D-dimer levels have a greater probability of PE, regardless of the clinical suspicion16,17. Moreover, elevated D-dimer level is associated with poor outcome and increased mortality in SARS-CoV2 disease, regardless of the occurrence of thromboembolic disease17. A value of more than 2.01 μg/mL is associated with high in-hospital mortality, especially in elderly COVID-19 patients with diabetes17.
Current guidelines recommend a rigorous selection of COVID-19 patients requiring cardiac ultrasound, to which the management strategy could be modified by the results, providing clinical benefit20. In addition, the guideline highlights the utility of specific echocardiographic modalities, such as focused cardiac ultra-sound, point-of-care cardiac ultrasound, and critical care echocardiography for the evaluation of cardiac involvement in SARS-CoV-2 disease20. These techniques evaluate in a very short time, important insight of chamber geometry, cardiac function and presence of pericardial effusion20. The use of hand-held device might be beneficial to decrease the risk of infection, as they are easier to disinfect, and the use of telemedicine to evaluate acquired echocardiographic images by an expert is also recommended20.
Transthoracic 12 leads electrocardiogram and bulls eye images for the left ventricular strain in a young patient presenting with a moderate form of COVID 19 and chest pain. In the acute phase, troponin levels were increased, coronary angiography showed normal coronary arteries, and strain revealed left ventricular wall motion abnormalities. Follow-up showed the regression of the “STEMI-like” aspect of the ECG, and the regression of the wall motion abnormalities by strain analysis.
Two-dimensional echocardiographic images of the left ventricular walls edema in a patient with acute myocarditis and a severe form of COVID-19. Panel A. Parasternal long-axis view; Panel B. Apical 4 chamber view; Panel C. Short axis view.
LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.
Approximately 5–7% of COVID-19 patients develop pericardial effusion, without correlation with the degree of myocardial involvement; cases with cardiac tamponade are scarce28. Pericardial fluid is often exudative and without virus, secondary to an inflammatory response rather than infectious28.
Withal, during COVID-19 pandemic, abnormal echocardiographic findings like apical ballooning and basal hyperkinesia from Takotsubo cardiomyopathy is constantly increasing, comparing to pre-pandemic period. Possible mechanisms are high psychological distress, elevated sympathetic nervous system activity, cytokine storm, and endothelial dysfunction29.
PE occurs in 33% of COVID-19 patients and is associated with a higher mortality risk and cardiogenic shock comparing with non-PE patients34. When clinical suspicion is present, echocardiographic parameters, such as pulmonary ejection acceleration time <60ms, peak systolic tricuspid valve gradient <60mmHg, impaired contractility of the RV free wall compared to the RV apex, have a high predictive value in the diagnosis and management of PE34,35.
ARDS is a complication of severe COVID-19, associated with high mortality; it requires mechanical ventilation with high positive end-expiratory pressure (PEEP) and prone positioning36. Patients with ARDS treated with PEEP are prone to developing RV failure, acute cor pulmonale with systolic and diastolic overload37. The main echocardiographic findings are right heart dilatation, end-systole paradoxical septal motion, and reduced RV global function37.
Currently, imaging data from follow-up of the patients recovered from COVID-19 and cardiac involvement is limited. However, cardiovascular assessment and standard echocardiography in the first six months after SARS-CoV2 disease is recommended for these patients in order to screen for post-residual myocardial damage, to establish the burden of long-term cardiac diseases, and to early initiate protective therapeutic measurements24.
I. Risks and guideline recommendations CMR is an essential tool for diagnosis and monitoring myocardial injury in COVID-19 patients41. Based on functional sequences, like cine white blood steady state free precession on the short and long axis of two-, three-, and four-chambers views, and tissue morphological characterization sequences such as T2 short tau inversion recovery (T2 STIR), T1 pre-contrast, and post-contrast mappings, T2 mapping and late gadolinium enhancement (LGE), CMR allows a differential diagnosis of ischemic and non-ischemic acute cardiovascular injury41. Moreover, in acute myocarditis, CMR is the method of choice that identifies with a high sensitivity focal or diffuse myocardial edema through T2 STIR, necrosis areas and fibrosis by LGE, diffuse expansion of extracellular volume fraction and hyperemia41,42.
In order to reduce exposure risk during the COVID-19 pandemic, current guidelines recommend short CMR examinations, only when strongly indicated, and adapted to the patients’ capacity of breath hold43. The main indications for CMR in active or convalescent phase of COVID-19 patients with cardiac involvement are heart failure, myocarditis, pericarditis, myocardial infarction, Takotsubo cardiomyopathy, and ventricular arrhythmias43.
II. Main CMR findings in adults. CMR examination in patients recovered from SARS-CoV-2 disease with cardiovascular symptoms reveals myocardial edema in 54% of patients, and LGE in 31% of patients (Figure 4); more interesting, in all patients with CMR pathological findings, RV EF, cardiac index and stroke volume/body surface are impaired, suggestive of RV dysfunction44. On the contrary, even in patients recovered from COVID-19 and with no evidence of cardiac involvement during hospitalization, CMR shows increased T1, T2 and extracellular volume possibly explained by ongoing inflammation or fibrosis44,45. In addition, in the discharge day of twenty-nine COVID-19 patients with high troponin level at admission, using LGE and stress perfusion imaging, Knight et al. identifies non-ischemic heart disease in 38% of patients, ischemic heart disease in 17% of patients and both pathologies in 14% of patients, with preserved LV and RV EF46. The non-ischemic etiology is diagnosed based on the non-myocardial infarction LGE pattern, sparing the endocardium and without any correspondence with a coronary artery region46. In this group, LGE-myocarditis pattern is present in only 45% of the cases and a non-specific mid-wall LGE pattern is found in 18% of patients46.
Cardiac magnetic resonance in a patient with COVID-19 and myocarditis. CMR shows late gadolinium enhancement (LGE) with endocardial spearing at the level of the lateral and infero-lateral walls in an apical 2 chamber view (panel A) and short axis view (panel B).
CMR, cardiac magnetic resonance; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.
When compared to matched controls, competitive athletes that recovered from asymptomatic or mild forms of COVID-19 infection have increased myocardial T2 relaxation times in all segments, with inflammation or fibrosis in 9% of cases, without any changes on ECG or myocardial deformation47. Furthermore, they have increased mid-septal extracellular volume, similar with athletic controls47. Also, mild regional increase in T1 and T2 are found in 39% of COVID-19 athletes, 13% of healthy athletes and only 8% of normal subjects47. More than that, Rajpal et al. report a high rate of LGE (42%). This may correlate an increased risk of ventricular arrhythmias, poor outcome and suspended competitional activity48.
III. Main CMR findings in children. In a series of four cases of children with multisystem inflammatory syndrome and Kawasaki disease-like due to COVID-19, CMR demonstrates diffuse myocardial edema and hyperemia, but not focal necrosis, fibrosis or coronary artery abnormalities49. On the contrary, Wacker et al. reports a mild reduction of LVEF and no signs of myocarditis in the acute phase of the disease, but one month after COVID diagnosis, coronary artery dilatation is found, suggestive of post-infectious vasculitis50.
Currently, data about CMR findings in acute or recovered COVID-19 patients come from isolated or small cohorts of case reports. Future research is mandatory in order to define adapted diagnostic protocols, prognostic parameters, and management of COVID-19 patients with CMR abnormalities.
When myocardial injury is detected, associated with myocardial thickening and wall motion abnormalities on echocardiography, myocarditis should be considered and investigated. Cardiac computed tomography (CT) might assess myocardial tissue characterization by completing the protocol with delayed-iodine enhanced scan or extracellular mapping and CTA for the exclusion of obstructive CAD53. However, CMR remains the gold standard imaging method for the diagnosis of myocarditis53. In 4.8% of COVID-19 patients, cardiac CT is a useful diagnostic tool for the diagnosis of pericarditis, associated or not with myocarditis54,55. Cardiac tamponade is a rare first manifestation of the SARS-CoV-2 disease, diagnosed by cardiac CT55.
In addition, in COVID-19 patients, a high coronary calcium score, a marker of CAD, correlates with an unfavorable clinical outcome, represented by a severe form of the SARS-CoV-2 infection, transfer to ICU or death56. CTA can combine coronary artery, pulmonary artery, and thoracic aorta assessment by using a triple rule-out protocol for the rapid exclusion of severe acute pathologies with increased mortality, such as ACS, PE or aortic dissection57. In patients with different severity of respiratory symptoms, with high levels of cardiac biomarkers and D-dimer levels, a modified triple rule-out scan protocol with focus on lung parenchyma instead of the thoracic aorta as the third step of examination, may solve different clinical suspicions in just one stage57. Moreover, quadruple rule-out examination may be used, by adding delayed iodine enhanced scan in order to identify areas of myocardial fibrosis or necrosis57.
The use of ICA in patients with SARS-CoV-2 is recommended to those with hemodynamic instability due to acute myocardial infarction, cardiogenic shock and cardiac arrest58. In addition, ICA and primary percutaneous coronary intervention remain the standard of care for COVID-19 patients with ST-elevation myocardial infarction (STEMI), high-risk non-STEMI or unstable angina53,58. Moreover, coronary intravascular imaging or left ventriculography add value to ICA for the differential diagnosis of myocardial infarction with non-obstructive arteries syndrome or Takotsubo cardiomyopathy53,58. Case et al. reports that only 20% of COVID-19 patients with acute myocardial infarction benefit from ICA and primary percutaneous intervention. Moreover, when compared with patients without SARS-CoV-2 disease, COVID-19 patients and ACS are older, with more comorbidities, and have a higher significant in-hospital mortality (27.9% versus 3.7%)59.
Conclusions. SARS-CoV-2 infection has been spreading rapidly worldwide, with a crucial impact on health care services. Although respiratory syndrome prevails, COVID-19 affects also cardiovascular system by several mechanisms, increasing morbidity and mortality of these patients. Integration of cardiac biomarkers, electrocardiographic changes, and multi-modality imaging methods is essential for the early detection of cardiac damage, assessment of cardiovascular involvement extension, supporting the differential diagnosis of myocardial injury patterns and thorough management and follow-up of COVID-19 patients. Clinical and hemodynamic status, the severity of respiratory symptoms, comorbidities, as well as the availability and potential benefit on further management of patients with SARS-CoV-2 infection should be carefully taken into account, when considering the indication, benefit, and risk of different imaging modalities. Efforts should be targeted at collating multicenter experience into registries and elaborating specific assessment algorithms and protocols, in order to minimize exposure.