Cardioembolic Stroke in a Patient with Coronavirus Disease of 2019 (COVID-19) Myocarditis: A Case Report

Introduction There is a growing body of literature detailing coronavirus 2019 (COVID-19) cardiovascular complications and hypercoagulability, although little has been published on venous or arterial thrombosis risk. Case Report In this report, we present a single case of cardioembolic stroke in the setting of COVID-19 related myocarditis, diagnosed via cardiac magnetic resonance imaging and echocardiography. COVID-19 infection was confirmed via a ribonucleic acid polymerase chain reaction assay. Conclusion Further research is needed to evaluate the hypercoagulable state of patients with COVID-19 to determine whether prophylactic anticoagulation may be warranted to prevent intracardiac thrombi and cardioembolic disease in patients with COVID-19 related myocarditis.


INTRODUCTION
The first cases of coronavirus of 2019 (COVID-19) were reported in Wuhan, China, in December 2019. 1 While the full spectrum of clinical disease that the virus can cause has yet to be elucidated, a growing body of literature is emerging detailing various cardiovascular complications, ranging from myocardial injury with mild troponin elevations to fulminant myocarditis. 2,3 Elevations in cardiac biomarkers such as B-type natriuretic peptide (BNP, 27.5% of patients) and cardiac troponin (7-17%) are common, with the latter being associated with worsening disease severity, intensive care unit status, and mortality. [4][5][6] Additionally, cardiac dysrhythmias from a variety of etiologies have been reported. 2 In a case series of 150 patients with confirmed COVID-19, 7% of all deaths were attributed to myocarditis with ensuing circulatory collapse. 7 Derangements of coagulation laboratory studies have also been reported from COVID-19, including elevations of D-dimer, modulation of fibrinogen (high in early disease, low UC Davis Health, Department of Emergency Medicine, Sacramento, California in advanced disease), and modulation of prothrombin time and partial thromboplastin time, with over 70% of non-survivors in one study meeting criteria for disseminated intravascular coagulation. 8 While rates of venous thromboembolism have not been reported for COVID-19, severe acute respiratory syndrome, coronavirus 1, (SARS-CoV-1)(2013) was associated with ischemic stroke, deep venous thrombosis and pulmonary embolism, making it probable that similar complications are possible with the 2019 novel virus. 9 A recent case series identified seven cases of acro-ischemia in patients without evidence of shock and not on vasopressor support, providing early evidence of hypercoagulability in COVID-19 infection. 10 In this report, we present a single case of cardioembolic stroke in the setting of COVID-19 related myocarditis.

CASE REPORT
A 53-year-old male with a past medical history significant only for hyperlipidemia, was brought in by ambulance to the  Chest radiograph (CXR) showed a left lower lobe consolidation. His potassium was replaced, he was given an amiodarone load, started on ceftriaxone/ azithromycin for presumed community-acquired pneumonia, and admitted to the cardiology service.
During his hospitalization, he converted to normal sinus rhythm with electrolyte replacement and amiodarone load, but LBBB morphology and prolonged QTc persisted. Azithromycin was changed to doxycycline due to concerns of potentially worsening QTc prolongation from azithromycin. Rapid flu test, respiratory viral panel (RVP), legionella urine antigen, and blood cultures were all negative. Transthoracic echocardiography (TTE) was performed which showed mild left ventricular (LV) dilation with hypokinesis (ejection fraction 15%). There was no comparison TTE available. Cardiac catheterization did not reveal significant coronary artery disease. Cardiac magnetic resonance imaging (MRI) with contrast confirmed LV dilation with global hypokinesis, increased T2 signal, hyperemia, and edema consistent with viral myocarditis. In the setting of these MRI findings, the patient's CXR infiltrate, and the absence of an additional viral etiology for myocarditis (negative RVP), radiology recommended dedicated COVID-19 testing.
The patient was tested using a qualitative ribonucleic acid polymerase chain reaction assay (via nasopharyngeal swab) and resulted positive. However, since LV dilation (defined as LV end diastolic diameter [LVEDD] >3.3 centimeters per meter squared [cm/m 2 ]) is only present in approximately 50% of patients with acute myocarditis, and the range of LVEDD reported in one study of myocarditis patients with LV dilation was 3.4-6.1cm/m 2 , the patient's initial LVEDD (5.96 cm/m 2 ) approached the upper range for LV dilation expected in myocarditis, leading the cardiology service to suspect a chronic undiagnosed cardiomyopathy. 12 The patient had no evidence of ischemic heart disease, history of alcohol or methamphetamine use, or any other obvious etiology of cardiomyopathy. Further investigation revealed that the patient had spent time in Mexico, so he was tested for Chagas disease. This was a send-out lab and would not result for several days. The patient was discharged in stable condition on hospital day four with new prescriptions for metoprolol succinate, losartan and spironolactone.
Three days later, the patient returned to the ED with acute expressive aphasia without other neurological deficits. MRI and MR angiography of the brain revealed an acute left middle cerebral artery stroke involving Broca's area. He was treated with tissue plasminogen activator and admitted to the neurology service. ECG demonstrated a stable LBBB with no acute ST-T wave changes. Troponin T was elevated at 66 nanograms per liter (ng/L) (Ref: <19ng/L), and peaked at 373ng/L, 11 hours later. Emergent computed tomography angiography of the neck, did not reveal any carotid lesions, but did show ground-glass infiltrates in bilateral lung apices, consistent with COVID-19 infection. 4 CXR showed bilateral peripheral airspace opacities (left greater than right). Repeat TTE showed a new LV thrombus (not visualized on TTE or cardiac MRI from previous hospitalization) and worsening LV dilation (diastolic diameter increased from 5.96 cm to 6.53 cm, compared to previous echocardiography six days earlier). He was started on anticoagulation therapy with warfarin with a heparin bridge, and transferred to an outside hospital for continued care and rehabilitation.
The patient's Trypanasoma cruzi (T. cruzi) immunoglobulin G (IgG) was found to be positive at 1.8, suggesting active or past infection. The patient was notified of this finding, and infectious disease follow-up was arranged.

DISCUSSION
While the patient's presentation was consistent with acute viral myocarditis, the presence of T. cruzi IgG antibodies confounded the clinical picture. Since it is possible that the patient had undiagnosed Chagas cardiomyopathy, it is difficult to know whether the patient's presentation was truly related to COVID-19 myocarditis or simply was related to chronic pre-existing heart failure. However, myocardial fibrosis, a marker of Chagas cardiomyopathy that is detected as delayed gadolinium enhancement on cardiac MRI, was not seen in our patient, making Chagas cardiomyopathy less likely. 13 Furthermore, acute worsening of LV dilation and rising troponin levels, suggested an acute, rather than chronic process, making viral (COVID-19) myocarditis more likely.
In the setting of a newly diagnosed LV thrombus, the most likely source of the patient's stroke was cardioembolic. Transesophageal echocardiography (TEE) is the gold standard for diagnosing intracardiac thrombi. However, cardiac MRI has been found to be both more sensitive and specific than TEE for detecting LV thrombus, making it unlikely that the LV thrombus was present during the initial hospitalization, and instead, more likely that the thrombus formed during the three days between discharge and re-hospitalization. 14 While it is possible that the patient had undiagnosed paroxysmal atrial fibrillation that predisposed him toward forming a LV thrombus, this seems unlikely given that his irregular rhythm converted to sinus after electrolyte correction, and he did not re-enter an irregular rhythm during the same hospitalization, or during subsequent re-hospitalization. Furthermore, if he indeed did have atrial fibrillation, his CHA 2 DS 2 -VASc score would have been one (one point for newly diagnosed heart failure, annual stroke risk 0.6%), making it unlikely that a ventricular thrombus could have formed and embolized in three days. 15 The prevalence of LV thrombus in patients with dilated cardiomyopathy with reduced ejection fraction and sinus rhythm, is as high as 13%, with increasing LV size being independently associated with LV thrombus. 16,17 Since complete coagulation parameters were not initially obtained, it is difficult to quantify his level of hypercoagulability. Nonetheless, it is likely that this hypercoaguable state, in conjunction with acute myocarditis and worsening LV dilation, predisposed the patient to LV thrombus formation and cardioembolic stroke.

CONCLUSION
Myocarditis is a serious complication of COVID-19 infection and may predispose patients to further cardiovascular injury, such as cardioembolic stroke. Further research is needed to evaluate the full scope of cardiovascular complications in order to better inform treatment. Prophylactic anticoagulation should be considered in high-risk patients at risk for venous and arterial thromboembolism.