Advertisement
Case Report|Articles in Press

Cardiac recovery from COVID-19–associated fulminant myocarditis by extracorporeal biventricular assist

Published:February 22, 2023DOI:https://doi.org/10.1016/j.jccase.2023.02.013

      Abstract

      We report a case of cardiac recovery from coronavirus disease 2019 (COVID-19)-associated fulminant myocarditis in a 48-year-old woman diagnosed with COVID-19 infection 4 days before, whose hemodynamic collapse were resuscitated first with venoarterial extracorporeal membranous oxygenation, followed by escalation to extracorporeal biventricular assist devices (ex-BiVAD) using two centrifugal pumps and an oxygenator. She was likely to be multisystem inflammatory syndrome in adults (MIS-A) negative. Cardiac contractility gradually recovered after the 9th day of ex-BiVAD support, and the patient was successfully weaned from ex-BiVAD on the 12th day of support. Due to postresuscitation encephalopathy, she was transferred to the referral hospital for rehabilitation with recovered cardiac function. The histopathology of the myocardial tissue showed smaller amounts of lymphocytes and more infiltration of macrophages. It is important to recognize two phenotypes of MIS-A+ or MIS-A−, with distinct manifestations and outcomes. It is also important to refer urgently such patients with COVID-19-associated fulminant myocarditis, showing different histopathology from usual viral myocarditis, with evolution toward refractory cardiogenic shock to a center with capability for advanced mechanical support to avoid a too-late cannulation.

      Learning objective

      We should recognize the clinical course and histopathology of the multisystem inflammatory syndrome in adults phenotype of coronavirus disease 2019-associated fulminant myocarditis. We should urgently refer such patients with evolution toward refractory cardiogenic shock to a center with capability for advanced mechanical support, such as venoarterial extracorporeal membrane oxygenation, Impella (Abiomed, Danvers, MA, USA), and extracorporeal biventricular assist devices.

      Keywords

      Introduction

      In an international cohort of 56,963 hospitalized patients with coronavirus disease 2019 (COVID-19), the prevalence of acute myocarditis varied from 2.4 to 4.1 per 1000 patients [
      • Ammirati E.
      • Lupi L.
      • Palazzini M.
      • Hendren N.S.
      • Grodin J.L.
      • Cannistraci C.V.
      • et al.
      Prevalence, characteristics, and outcomes of COVID-19-associated acute myocarditis.
      ], among whom 39 % had fulminant myocarditis requiring inotropic support or mechanical circulatory support. According to the recent largest COVID-19 myocarditis case series using extracorporeal life support [
      • Tonna J.E.
      • Tan C.S.
      • Hryniewicz K.
      • Barbaro R.P.
      • Brodie D.
      • MacLaren G.
      Outcomes after extracorporeal life support for COVID-19 myocarditis: an analysis of the Extracorporeal Life Support Organization Registry.
      ], 1.8 % had fulminant myocarditis, with mortality up to 51 %, among 4792 COVID-19 patients who were supported with extracorporeal membrane oxygenation (ECMO). This mortality was higher than that in patients with COVID-19 without ECMO, as of those on ECMO with only acute respiratory failure. We experienced a survival patient with COVID-19 fulminant myocarditis who required venoarterial ECMO (VA-ECMO), which was escalated to extracorporeal biventricular assist devices (ex-BiVAD) due to no aortic valve opening. The histopathology of the myocardial tissue showed smaller amounts of lymphocytes and more infiltration of macrophages, which was not similar to viral myocarditis.

      Case report

      A previously healthy 48-year-old woman presented at the outpatient clinic with fever (37.7 °C). Although she had already received the 2nd COVID-19 mRNA vaccination 11 months before, she was diagnosed as having COVID-19 infection based upon positive polymerase chain reaction (PCR) assay for severe acute respiratory syndrome coronavirus 2 (SARSCoV-2) on her nasopharyngeal swab. Four days later, she developed diarrhea, nausea, myalgia, and general malaise. Her symptoms worsened over the following two days, and she was finally admitted to the local hospital with severe dyspnea and peripheral coldness. She was diagnosed with COVID-19–associated acute myocarditis based upon electrocardiography showing bradycardia with low voltage, elevation of serum myocardial biomarkers, echocardiography showing a nondilated but globally and severely dysfunctional left ventricle (LV), and coronary angiography revealing normal coronary arteries. Computed tomography scans showed pleural effusion and pulmonary congestion but no findings of COVID-19–associated pneumonia (Fig. 1). Soon she presented cardiac arrest and was immediately resuscitated with intubation, peripheral VA-ECMO, and intra-aortic balloon pump (IABP). She was transferred to our hospital for further treatment.
      Fig. 1
      Fig. 1Computed tomography scans at the referring hospital showed pleural effusion and pulmonary congestion but no inflammatory changes of the lungs related to coronavirus disease 2019 infection (A). Electrocardiography at admission to our hospital showed wide and low QRS in all leads and first-degree atrioventricular block (B). Chest X-ray at admission to our hospital demonstrated severe pulmonary congestion despite assistance with venoarterial extracorporeal membrane oxygenation (C). Pulmonary congestion on the chest X-ray immediately after placement of extracorporeal biventricular assist devices (D) improved on the following day (E).
      On admission to our hospital, blood pressure was 70/45 mmHg, pulse rate was 58/min with VA-ECMO flow at 2.0 L/min without any inotropes. Electrocardiography (ECG) showed wide and low QRS in all leads and first-degree atrioventricular block (Fig. 1). Chest X-ray showed severe pulmonary congestion despite assistance with VA-ECMO (Fig. 1). Laboratory tests revealed the elevation of myocardial biomarkers, liver enzymes, lactate, and white blood cell count (Table 1). The serum antibody titers against popular myocarditis-causing viruses including echoviruses, coxsackieviruses, and influenza viruses did not rise. A transthoracic echocardiography demonstrated severe global hypokinesia with increased biventricular wall thickness and no opening of the aortic valve (Video 1).
      Facing the hemodynamic collapse, we decided to place ex-BiVAD with two Rotaflow pumps (Getinge, Hirrlingen, Germany) for the following reasons (Video 1). First, more systemic perfusion flow than the limited VA-ECMO flow to 2.0 L/min due to small femoral and iliac arteries was required to avoid multiple organ failure. Second, the fulminant myocarditis was affecting both right ventricle and LV. Third, it was imperative to unload the LV to improve blood stagnation in the akinetic LV. Although Impella 5.5 (Abiomed, Danvers, MA, USA) was attempted to be placed, her axillary and femoral arteries were too small for anastomosis of a 10 mm vascular graft to place it.
      Under cardiopulmonary bypass, VA-ECMO and IABP were removed, a left VAD was established from the LV apex with a 32 Fr cannula to the ascending aorta with a 10 mm graft, and a right VAD with a membrane oxygenator (Biocube™, Nipro, Tokyo, Japan) was established from the right atrium with a 32/40 Fr cannula to the pulmonary artery with a 10 mm graft. After ex-BiVAD placement, chest X-ray showed improved pulmonary congestion (Fig. 1). With no inotropes, heparin to achieve activated clotting time of 160 s or activated partial thromboplastin time of 40 to 50 s, and inhalation of nitric oxide, we obtained left and right VAD flows of 3.5 to 4.0 L/min, respectively. We also administered corticosteroid pulse therapy with methylprednisolone of 1000 mg/day for 3 days and intravenous immunoglobulin of 1 g/kg for 2 days.
      Table 1Laboratory results.
      At

      admission
      Before

      ex-BiVAD
      Support

      Day 3
      Support

      Day 6
      Support

      Day 9
      Support

      Day 12
      Day 2

      after weaning
      Day 7

      after weaning
      Normal values
      White blood cells (/μL)11,60064009300890011,80015,80012,40089003300–8600
      Hemoglobin (g/dL)10.810.810.010.210.310.710.110.311.3–15.2
      Platelets (×104/μL)5.54.77.35.25.88.29.610.213.0–36.9
      CRP (mg/dL)0.040.131.260.328.387.884.873.57<0.14
      CK (U/L)204227319422171072121098841–153
      CK-MB (U/L)18921514.53.05.19.83.52.9<5
      Troponin I (ng/mL)45.545.519.58.54.32.32.21.9<0.026
      NT-proBNP (pg/mL)30,43721,053223610813566672253234867<125
      Total bilirubin (mg/dL)0.70.70.60.60.90.60.40.50.4–1.5
      AST (U/L)427427103837083453913–30
      ALT (U/L)2412413639273928247–23
      LDH (U/L)10341034504451564451402376124–222
      Lactate (mg/dL)18587131481310820–38
      Creatinine (mg/dL)1.271.270.920.800.630.920.760.617–11
      ex-BiVAD, extracorporeal biventricular assist device; CRP, C-reactive protein; CK, creatine kinase; CK-MB, creatine kinase-myoglobin binding; NT-proBNP, N-terminal pro–B-type natriuretic peptide; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH, lactate dehydrogenase.
      Fig. 2
      Fig. 2Histological findings in the myocardial tissue of the left ventricular apex, which was excised for placement of a left ventricular assist device. The specimens were stained with hematoxylin and eosin (HE) (i), anti-CD3 (ii), anti-CD4 (iii), anti-CD8 (iv), anti-CD20 (v), anti-CD79a (vi), and anti-CD68 antibodies (vii). The HE stains show destructed myocardium. However, unlike in usual viral myocarditis, the HE stains show smaller amounts of inflammatory cell infiltration, including neither T lymphocytes nor B lymphocytes. Moderate amounts of macrophages are shown in the stains with anti-CD68 antibodies.
      Cardiac contractility gradually recovered after the 9th day of ex-BiVAD support, and the patient was successfully weaned from ex-BiVAD on the 12th day of support (Video 1). Instead, IABP was placed again for 2 days. Although cardiac function restored stably, the patient presented with poor consciousness. Magnetic resonance imaging showed bright diffuse high signal intensity in the bilateral cerebral cortex and basal ganglia by diffusion-weighted image, which was compatible with postresuscitation encephalopathy or coma (Video 1). As well as QRS widening on ECG was recovered, inflammatory and myocardial biomarkers also gradually declined (Table 1). Echocardiography one month after weaning from ex-BiVAD showed almost normal right ventricle and LV systolic function (Video 1). Histopathology of the myocardial tissue of the LV apex, which was excised for left VAD cannulation, showed smaller amounts of lymphocytes and more infiltration of macrophages, which was not similar to viral myocarditis with predominantly lymphocytic infiltrates (Fig. 2). The patient was alive uneventfully at 3-month follow-up.

      Discussion

      Pathogenesis of COVID-19–associated fulminant myocarditis include direct myocardial injury by SARS-CoV-2 targeting angiotensin-converting enzyme 2 (ACE 2) receptors and inflammatory-mediated myocardial dysfunction [
      • Barhoum P.
      • Pineton de Chambrun M.
      • Dorgham K.
      • Kerneis M.
      • Burrel S.
      • Quentric P.
      • et al.
      Phenotypic heterogeneity of fulminant COVID-19-related myocarditis in adults.
      ].
      The former pathogenesis, direct myocardial injury by SARS-CoV-2, is characterized histopathologically by scattered myocyte necrosis without significant brisk lymphocytic inflammatory infiltrates [
      • Chen L.
      • Li X.
      • Chen M.
      • Feng Y.
      • Xiong C.
      The ACE2 expression in human heart indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2.
      ,
      • Fox S.E.
      • Akmatbekov A.
      • Harbert J.L.
      • Li G.
      • Quincy Brown J.
      • Vander Heide R.S.
      Pulmonary and cardiac pathology in African American patients with COVID-19: an autopsy series from New Orleans.
      ], as shown in our patient. In our patient, although SARS-CoV-2 was not evidenced, destructed myocardium on hematoxylin and eosin stains resulted from direct viral injury. These histopathological findings may suggest the different pathogenesis of COVID-19-associated myocarditis from common viral myocarditis. In COVID-19-associated myocarditis, the pericytes with increased expression of ACE2 might be infected by SARS-CoV-2 and cause capillary endothelial cell or microvascular dysfunction that could cause myocarditis [
      • Basso C.
      • Leone O.
      • Rizzo S.
      • De Gaspari M.
      • van der Wal A.C.
      • Aubry M.C.
      • et al.
      Pathological features of COVID-19-associated myocardial injury: a multicentre cardiovascular pathology study.
      ,
      • Hanson P.J.
      • Liu-Fei F.
      • Ng C.
      • Minato T.A.
      • Lai C.
      • Hossain A.R.
      • et al.
      Characterization of COVID-19-associated cardiac injury: evidence for a multifactorial disease in an autopsy cohort.
      ].
      The latter pathogenesis, immunologic dysregulation, leads to the multisystem inflammatory syndrome in adults (MIS-A), for whose diagnosis the U.S. Centers for Disease Control and Prevention (CDC) has developed definition criteria [

      Centers for Disease Control and Prevention. Multisystem inflammatory syndrome in adults (MIS-A) case definition information for healthcare providers. Accessed November 1, 2021https://www.cdc.gov/mis/mis-a/hcp.html.

      ]. SARS-CoV-2 infection evokes inflammatory immune responses, responsible for a cytokine storm with increased interleukin (IL)-6 recruiting macrophages to the myocardium. This pathogenesis coincides with high expression of macrophages in the myocardium of COVID-19 associated myocarditis [
      • Basso C.
      • Leone O.
      • Rizzo S.
      • De Gaspari M.
      • van der Wal A.C.
      • Aubry M.C.
      • et al.
      Pathological features of COVID-19-associated myocardial injury: a multicentre cardiovascular pathology study.
      ,
      • Hanson P.J.
      • Liu-Fei F.
      • Ng C.
      • Minato T.A.
      • Lai C.
      • Hossain A.R.
      • et al.
      Characterization of COVID-19-associated cardiac injury: evidence for a multifactorial disease in an autopsy cohort.
      ], as shown in our patient.
      There are 2 phenotypes of COVID-19–related fulminant myocarditis, MIS-A+ or MIS-A−, with distinct manifestations and outcomes. Compared with MIS-A+ patients with the clinical profiles, including fever, elevated inflammatory markers, rash, conjunctivitis, and gastrointestinal symptoms, MIS-A− patients were characterized by a shorter delay between COVID-19 symptoms onset and myocarditis, more frequent positive PCR, a lower LV ejection fraction, and higher in-hospital mortality up to 31 %, and therefore more likely to require VA-ECMO (92 %). Since C-reactive protein was as low as 0.04 mg/dL and neither ferritin, IL-6, erythrocyte sedimentation rate, nor procalcitonin were elevated, our patient was diagnosed as MIS-A−, according to the definition by CDC. Although MIS-A has a more heterogeneous clinical presentation than previously appreciated and is commonly underdiagnosed [
      • Davogustto G.E.
      • Clark D.E.
      • Hardison E.
      • Yanis A.H.
      • Lowery B.D.
      • Halasa N.B.
      • et al.
      Characteristics associated with multisystem inflammatory syndrome among adults with SARS-CoV-2 infection.
      ], the MIS-A− phenotype may thus represent an exaggerated innate immune response, or cytokine storms, to the COVID-19 infection, whereas the MIS-A+ phenotype may represent a maladaptive acquired immune response, or a postinfectious complication. It is important to differentiate the phenotypes and to urgently refer the MIS-A− patients with the high risk of evolution toward refractory cardiogenic shock to a center with VA-ECMO and VAD capability, closely monitor to avoid a too-late cannulation with early administration of corticosteroids and/or intravenous immunoglobulins.
      The following is the supplementary data related to this article.

      Funding

      None.

      Consent statement

      Written informed consent was obtained from the patient's husband.

      Declaration of competing interest

      The authors have no conflicts of interest to declare.

      Acknowledgments

      Not applicable.

      References

        • Ammirati E.
        • Lupi L.
        • Palazzini M.
        • Hendren N.S.
        • Grodin J.L.
        • Cannistraci C.V.
        • et al.
        Prevalence, characteristics, and outcomes of COVID-19-associated acute myocarditis.
        Circulation. 2022; 145: 1123-1139
        • Tonna J.E.
        • Tan C.S.
        • Hryniewicz K.
        • Barbaro R.P.
        • Brodie D.
        • MacLaren G.
        Outcomes after extracorporeal life support for COVID-19 myocarditis: an analysis of the Extracorporeal Life Support Organization Registry.
        Crit Care. 2022; 26: 235
        • Barhoum P.
        • Pineton de Chambrun M.
        • Dorgham K.
        • Kerneis M.
        • Burrel S.
        • Quentric P.
        • et al.
        Phenotypic heterogeneity of fulminant COVID-19-related myocarditis in adults.
        J Am Coll Cardiol. 2022; 80: 299-312
        • Chen L.
        • Li X.
        • Chen M.
        • Feng Y.
        • Xiong C.
        The ACE2 expression in human heart indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2.
        Cardiovasc Res. 2020; 116: 1097-1100
        • Fox S.E.
        • Akmatbekov A.
        • Harbert J.L.
        • Li G.
        • Quincy Brown J.
        • Vander Heide R.S.
        Pulmonary and cardiac pathology in African American patients with COVID-19: an autopsy series from New Orleans.
        Lancet Respir Med. 2020; 8: 681-686
        • Basso C.
        • Leone O.
        • Rizzo S.
        • De Gaspari M.
        • van der Wal A.C.
        • Aubry M.C.
        • et al.
        Pathological features of COVID-19-associated myocardial injury: a multicentre cardiovascular pathology study.
        Eur Heart J. 2020; 41: 3827-3835
        • Hanson P.J.
        • Liu-Fei F.
        • Ng C.
        • Minato T.A.
        • Lai C.
        • Hossain A.R.
        • et al.
        Characterization of COVID-19-associated cardiac injury: evidence for a multifactorial disease in an autopsy cohort.
        Lab Invest. 2022; 102: 814-825
      1. Centers for Disease Control and Prevention. Multisystem inflammatory syndrome in adults (MIS-A) case definition information for healthcare providers. Accessed November 1, 2021https://www.cdc.gov/mis/mis-a/hcp.html.

        • Davogustto G.E.
        • Clark D.E.
        • Hardison E.
        • Yanis A.H.
        • Lowery B.D.
        • Halasa N.B.
        • et al.
        Characteristics associated with multisystem inflammatory syndrome among adults with SARS-CoV-2 infection.
        JAMA Netw Open. 2021; 4e2110323