Cardiac Magnetic Resonance Imaging in Coronavirus Disease 2019 (COVID-19): A Systematic Review of Cardiac Magnetic Resonance Imaging Findings in 199 Patients : Journal of Thoracic Imaging

Secondary Logo

Journal Logo

Original Articles

Cardiac Magnetic Resonance Imaging in Coronavirus Disease 2019 (COVID-19)

A Systematic Review of Cardiac Magnetic Resonance Imaging Findings in 199 Patients

Ojha, Vineeta MD*; Verma, Mansi MD*; Pandey, Niraj N. DM*; Mani, Avinash MD; Malhi, Amarinder S. DM*; Kumar, Sanjeev MD*; Jagia, Priya MD*; Roy, Ambuj DM; Sharma, Sanjiv MD*

Author Information
Journal of Thoracic Imaging: March 2021 - Volume 36 - Issue 2 - p 73-83
doi: 10.1097/RTI.0000000000000574

Abstract

KEY POINTS

  • Myocarditis was the most prevalent diagnosis on cardiac magnetic resonance imaging in patients with Coronavirus disease 2019 (COVID-19).
  • Mapping abnormalities were the most common imaging findings, followed by edema and late gadolinium enhancement (LGE). Subepicardial LGE in the basal to mid left ventricle was the most prevalent pattern of LGE.
  • Ventricular functions were normal in most of the patients.

INTRODUCTION

The rapid emergence of COVID-19 caused by novel coronavirus (SARS-Cov-2) has led to an unprecedented global health crisis.1 Although the clinical course is primarily characterized by respiratory symptoms, cardiac involvement in COVID-19 has been documented and is seen to cause substantial morbidity and mortality. Adverse outcomes have been reported especially in patients with preexisting cardiovascular disease. COVID-19 has been implicated in a wide gamut of cardiac manifestations including heart failure, cardiogenic shock, arrhythmias, myocardial inflammation, and coronary involvement, among others.2

The putative mechanisms for myocardial injury in COVID-19 include exaggerated immune response or direct viral damage. It has been hypothesized that SARS-CoV-2 binds to angiotensin-converting enzyme-2 (ACE-2) receptors on cardiac myocytes followed by its incorporation and replication resulting in direct damage to the cardiac tissue.3 Other possible mechanisms include activation of interleukins and interferons especially interleukin-6 and the subsequent cytokine storm, microcirculatory endothelial dysfunction due to systemic inflammatory response and hypoxic injury.4 A hypercoagulable state created by this virus can also cause thrombosis of coronary arteries leading to ischemia.2

Cardiac complications can be diagnosed by a variety of modalities available. Cardiac magnetic resonance (CMR) imaging has the unique ability of providing morphologic and functional information and tissue characterization and is recommended by the American Heart Association to detect myocardial insult.5 It is imperative for the health care workers to be aware of the spectrum of CMR findings in COVID-19 to provide timely diagnosis and prompt institution of appropriate treatment. Our knowledge pertaining to cardiac complications is still evolving and the literature regarding cardiac manifestations of this disease entity is sparse and scattered with lack of a cohesive compilation. To our knowledge, this is the first systematic review to compile the data in the current published literature regarding cardiac involvement in COVID-19 as depicted on CMR.

MATERIALS AND METHODS

Search Strategy

We sought to conduct a narrative synthesis of the reported CMR findings in patients with COVID-19 (synthesis without meta-analysis [SWiM]). We developed our search strategy according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines.6 The study protocol was registered in PROSPERO (CRD42020199104). Comprehensive electronic search of the PubMed, Embase, Google Scholar, and World Health Organization Library databases was performed on August 4, 2020 using the search terms: (“covid” or “covid-19” or “coronavirus” or “SARS-CoV-2” or “2019-nCoV” or “n-CoV”) AND (“MRI” OR “MR” OR “CMR” OR “Magnetic Resonance”). The search was limited to articles published in the year 2020. Additional search of the gray literature and the reference lists of the extracted studies selected was done to extract other relevant studies. Duplicates were excluded.

Study Selection

The published articles (including case reports or series) describing CMR findings in patients with confirmed COVID-19 infection were included in the review. Additional inclusion criteria were articles conducted on human beings, published in English and with extractable full text. No restrictions were applied based on the country of research. Reviews, preprints, editorials, guidelines, and recommendations were excluded. Two independent reviewers screened the titles and abstracts of the included articles according to the criteria mentioned above. Any disagreements were solved mutually and by the senior author, if required.

Quality of Study Assessment

Two independent reviewers rated all the included studies for their quality based on the National Institutes of Health (NIH) Quality Assessment Tool for Case Series Studies.7 The included studies had a small sample size due to rarity of reported cases. The methodological quality of the studies was generally rated as fair indicative of the limited and low-quality data available pertaining to CMR findings.

Data Extraction

We retrieved the full texts of the articles included for the final review and further screened them for their eligibility. After careful scrutiny, articles included for the systematic review and final analysis were shortlisted. Two independent reviewers extracted the relevant data from the full texts of the included articles into a Microsoft Excel database using the following fields: author, study design, journal, country, demographics, sample size, clinical features, CMR imaging features, and follow-up. To extract the granular data, various subfields were also used such as biomarker elevation, distribution of lesions, etc. Discrepancies were resolved by mutual discussion between the 2 reviewers. Data was analyzed using Microsoft Excel.

CMR Data Analysis

Substantial heterogeneity existed within the data. Many studies described findings according to the standard magnetic resonance imaging (MRI) definitions for conditions such as myocarditis (Lake Louise criteria 2009 in 4 studies and modified Lake Louise criteria 2018 in 6). For those studies that did not give the definition, the analysis was done in accordance with the Lake Louise criteria 2018.8,9 For most of the studies, edema was defined to be present when the ratio of myocardium: skeletal muscle signal intensity was >2.10 Myocarditis-like LGE was defined as the one not corresponding to any vascular territory and sparing the subendocardium.11

RESULTS

Characteristics of Included Studies

After removing the duplicate studies, a total of 289 unique records were identified from the 4 databases (Fig. 1). After initial screening, a total of 63 records met the criteria for a full text review. Out of these, 34 studies were finally included for analysis after careful scrutiny. Table 1 describes the demographic information pertaining to the study population. In 34 included studies, a total of 221 patients underwent 224 MRI scans (including 3 follow-ups). However, the second largest study in this systematic review described in detail the findings on CMR in only 29 patients (with unknown etiology) out of a total of 51 patients.11 So, we excluded the remaining 22 patients from the final analysis, giving a total of 199 patients. Most of the studies were case reports except for 5 retrospective and 1 prospective observational studies. CMR findings were reported in all these studies. The findings were reported from many countries across the globe; however, Germany, England, and China constituted maximum proportion of the sample size. Methodologic quality of the studies was evaluated using the NIH Quality Assessment Tool for Case Series and was fair for all the included studies (Supplementary Table 1, Supplemental Digital Content 1, https://links.lww.com/JTI/A182). The diagnosis of COVID-19 was confirmed by real-time reverse transcriptase polymerase chain reaction (RT-PCR) in all the patients included in this systematic review, except 5 (in 2 studies) who were positive on serology.10,19 Most of the patients had recovered from COVID-19, rather than harboring active disease. All the studies reported raised troponin levels. Thirteen studies comprising 80 patients described raised NT-proBNP levels (Table 2).

F1
FIGURE 1:
PRISMA 2009 flowchart describing selection of studies included in the systematic review. (Adapted from Moher et al.6 Therefore, in order to reprint this adapted figure, authorization must be obtained both from the owner of the copyright in the original work and from the owner of copyright in the translation or adaptation.)
TABLE 1 - Overview of the Included Studies and the Demographic Profile of the Population
References Country of Study Study Design Number of Patients With CMR Findings Male Age (y) (Mean or Median)
Puntmann et al12 Germany Prospective observational 100 53/100 (53%) 49 (45-53)
Knight et al11 England Retrospective observational, letter 29 24 /29 (83%) 64±9
Blondiaux et al10 France Retrospective case series 4 1/4 (25%) 9±3 (range 6-12)
Esposito et al13 Italy Case series 10 2/10 (20%) 52±6
Huang et al14 China Retrospective observational 26 10/26 (38.5%) Median=38; [IQR: 32-45]
Caballeros Lam et al15 Spain Scientific letter, case series 2 1/2 (50%) 26, 13
Coyle et al16 US Case report 1 1 57
Beşler et al17 Turkey Case report 1 1 20
Inciardi et al18 Italy Case report 1 0 53
Gravinay et al19 France Case report 1 1 51
Trogen et al20 US Case report 1 1 17
Luetkens et al21 Germany Case report 1 1 79
Manka et al22 Switzerland Case report 1 1 75
Pavon et al23 Switzerland Case report 1 1 64
Sardari et al24 Iran Case report 1 1 31
Gnecchi et al25 Italy Case report 1 1 16
Paul et al26 France Case report 1 1 35
Sala et al27 Italy Case report 1 0 43
Bonnet et al28 France Case report 1 1 27
Kim et al29 Korea Case report 1 0 21
Doyen et al30 France Case report 1 1 69
Warchoł et al31 Poland Case report 1 1 74
Sassone et al32 Italy Case report 1 1 38
Salamanca et al33 Spain Case report 1 1 44
Bernardi et al34 Italy Case report 1 1 74
Fischer et al35 France Case report 1 1 15
Bernal-Torres et al36 Spain Case report 1 0 38
Weinsaft et al37 Case report 1 0 36
Madamanchi et al38 US Case report 1 1 41
Yuan et al39 China Case report 1 1 33
Garot et al40 France Case report 1 1 18
Monmeneu et al41 Spain Case report 1 1 43
Oberweis et al42 Luxembourg Case report 1 1 8
Frédéric et al43 France Case report 1 1 39
IQR indicates interquartile range.

TABLE 2 - Pooled Incidence of Various Abnormalities on CMR in COVID-19 Patients
Number of Studies Included Pooled Incidence (as Per Total Number of MRI Performed)
CMR diagnosis
 Myocarditis (4 with Kawasaki like manifestation, 1 with reverse Takotsubo) 34 80/199 (40.2%)
 Myopericarditis 34 3/199 (1.5%)
 Takotsubo 34 3/199 (1.5%)
 Ischemia 34 5/199 (2.5%)
 Dual ischemic plus nonischemic 34 4/199 (2.0%)
 Normal CMR 34 42/199 (21.1%)
Mean ejection fractions on CMR
 Mean LVEF* 24 51.6% (6609.7/128)
 Mean RVEF 24 56.2% (6126.3/109)
Major abnormalities on cardiac MRI
 Regional wall motion abnormality (RWMA) 20 13/32 (40.6%)
 Edema on T2 or STIR 28 46/90 (51.11%)
 Perfusion deficit 3 18/21 (85.71%)
 LGE 34 85/199 (42.7%)
 T1 mapping abnormality 13 109/150 (72.7%)
 T2 mapping abnormality 10 91/144 (63.2%)
 ECV mapping abnormality 6 21/40 (52.5%)
 Pericardial effusion 11 43/175 (24.6%)
 Pericardial LGE 1 22/100 (22%)
 Left ventricle hypertrophy 6 6
 Pericardial thickening 2 2
 LV apical thrombus 2 2
 Early gadolinium enhancement 2 1/2
 Other findings:
  Diastolic dysfunction 1 1
  Left atrial enlargement 1 1
  LV wall thinning 1 1
  Reduced global longitudinal strain 1 1
*1 study did not mention exact LVEF but it was <40% in 2 patients, >55% in 5 patients, 40-55% in 3 patients.13
Cumulative percentages for some of the findings have not been calculated as they were mentioned in a very few studies/number of patients and are not truly representative.
LVEF indicates left ventricular ejection fraction; RVEF, right ventricular ejection fraction; STIR, short tau inversion recovery.

Common Imaging Findings on CMR

Various CMR imaging findings have been described across the included studies as described in Table 2. Table 3 provides the pooled incidence of various imaging findings. The mean duration of CMR from symptom onset varied widely, ranging from day 2 to day 71. Myocarditis (80/199; 40.2%) was the most prevalent diagnosis, whereas normal CMR was seen in 21.1% (42/199) of the patients. Uncommon CMR diagnoses included inducible ischemia in 2.5% (5/199), acute dual ischemic plus nonischemic pattern in 2% (4/199), Takotsubo syndrome, and myopericarditis in 1.5% (3/199) of the patients, each.

TABLE 3 - Pooled Incidence of Various Imaging Finding Across the Included Studies
References Troponin N-terminal Pro-B-type Natriuretic Peptide CK-MB CRP Clinical Presentation CMR Diagnosis Normal MRI Gap Between Symptom Onset and CMR LVEF on MRI RVEF MRI RWMA (Dyskinesia/Hypokinesia) LV Wall Thickness Edema on T2 or STIR (Y/N) Segments T2 Adenosine Stress Perfusion EGE (Y/N) LGE Pattern of LGE Segments LGE (Y/N) T1 Mapping Abnormality (Y/N) T1 Mapping Value (Segment) T2 Mapping Abnormality (Y/N) T2 Mapping Value (Segment) ECV Pericardial Thickening/ Effusion Other Cardiac Findings Lung Findings (Chest X-ray and/or CT)
Puntmann et al12 Raised in 76 Raised in 68 NA Raised Recoverd patients with increased serological markers chest pain (17) palpitation (20) shortness of breath (36) Myocarditis (78) 22/100 71 d (from positive testing) 56% 56% NA NA NA NA NA NA Y Myocardial (32), nonischemic (20), ischemic (12) NA Y (73) 1130 ms Y(60) NA NA Effusion (20) pericardial LGE (22) NA
Knight et al11 Raised NA NA NA Discharged patients with elevated troponin NI: 11, I: 5, Dual:4, No cause: 9 (13/29 myocarditis, 9/29 ischemic, 7-ischemia; 1 prior MI) 9/29 (31%) 46±15 d 67.7±11.4% 63.7%±9.5% NA NA N (51 ms in myocarditis as well as non myocarditis) Myocarditis like: 2 (1-2.5) Done in 19 pts Ischemia: 9 Inducible ischemia: 8 NA NI: 11, I: 5, Dual:4, No cause: 9, Myocarditis like LGE: 13 non sp mid wall: 2 NA NA NA NA NA NA Effusion: 2/29 with unknown etiology (7%) Lung parenchymal changes 20/29 (69%); pleural effusion 4/29 (14%)
Blondiaux et al10 Raised in all NA NA NA MIS-C and Kawasaki like: pain (4), vomiting (2), diarrhea (2), and fever. Myocarditis 0 3: acute, 1: recovery phase 68, 51, 52, 56 63, 53, 57, 55 No NA Y (3) N: (1) 3: ratio >2. 2: 47.62 ms NA NA N N N 3:Y 3:>1100 ms 1: 1050 ms NA NA NA Effusion: 3 Peripheral opacities on CT: 3, normal: 1
Esposito et al13 Raised NA NA NA 8/10 (80%) experienced oppressive chest pain. 2/10 (20%)had dyspnea 2:Takutsubo 8: myocarditis like 0 1 wk 2:<40%, 5:>55%, 3: 40-55% NA No NA Y (8): ratio 2.3 2:apical 8:diffuse NA NA 7:N, 3: Y Thin sunepicardial striae Lateral wall Y (8) 1.5T:1,156 ms, 3 T:1378 ms Y (8) 62 ms 2 patients: 30 and 36% Effusion: 6/8 (75%); Thickening: None NA
Huang et al14 At admission, 13/26 patients: median [IQR] peak value of 2.2 [1.9-2.6] pg/mL; normal at time of CMR NA NA NA Precordial chest pain 3/26 (12%), palpitation 23/26 (88%), and chest distress 6/26 (23%); history of hypertension before COVID-19 2/26 (8%) 15/26 had myocardial edema and/or LGE 11/26 (42%) 47 d (IQR: 36-58) 60.7 (45% in one) 36.5 NA NA Y (14) 54% 33% (137/416) LV the majority of T2 signal hyperintensity was located in the interventricular septum, anterior, anterior-lateral, and inferior wall at base and mid-chamber NA NA Y(8) 31% Focal linear subepicardial and patchy mid-wall LGE Most LGE (9/15) [60%] lesions were located at inferior and inferior-lateral segments at base and mid-LV Y (15) 1271ms [IQR: 1243-1298] vs. 1237 ms [IQR: 1216-1262] in patients with and without positive MRI, respectively Y (15) 42.7 [IQR: 3.1] vs. 38.1 [IQR: 2.4] in patients with and without positive MRI, respectively 28.2% [IQR: 24.8-36.2] vs. 24.8% [IQR: 23.1-25.4] in patients with and without positive MRI, respectively Effusion: 7 NA
Caballeros Lam et al15 Raised, raised NA NA NA, raised Chest pain, mild cough and fever (2 patient) Myocarditis 0 7 d, NA 59%, normal function in second NA No NA NA NA NA NA Y, N Mesocardial and subepicardial Basal and mid-inferoseptal and inferior myocardial segments Y 1303.1110 ms Y 53.54 ms NA Effusion in second patient NA
Coyle et al16 Raised NA NA NA Shortness of breath, fever, cough, myalgia Myocarditis 0 25 d 82% NA NA NA Y Diffuse biventricular and biatrial NA NA Y Focal mid wall Basal inferolateral segments NA NA NA NA No effusion Bilateral patchy interstitial opacities (CT and CXR)
Beşler et al17 Raised Raised Raised Raised Fever, chest pain Myocarditis 0 14 d 64% NA NA NA Y Subepicardial mid posterolateral LV NA NA Y Subepicardial Mid PL NA NA NA NA NA NA Subpleural consolidation in left upper lobe (CXR and CT )
Inciardi et al18 Raised Raised Raised Raised Fatigue (fever and dry cough a week before) Acute myopericarditis with systolic dysfunction 0 2 d 35% NA Diffuse biventricular hypokinesis, especially in the apical segments LVH Y Diffuse biventricular NA NA Y Diffuse biventricular NA Y NA NA NA NA Effusion (12mm) Mild LV diastolic dysfunction Normal
Gravinay et al19 Raised NA NA NA Fever, atypical chest pain Myocarditis 0 8 d Normal NA No NA Y Subepicardial lateral and inferior LV NA NA Y Subepicardial Inferior and lateral LV NA NA NA NA NA NA Apical LV thrombus CT normal
Trogen et al20 Raised NA NA NA Fever Myocarditis 0 NA 40% 39% Yes (focal) NA Y Mid wall inferior RV-LV junction NA NA Y Mid wall Inferior RV-LV junction NA NA NA NA NA NA Hazy GGOs at bilateral lower lobes (CXR)
Luetkens et al21 Raised Raised NA Raised Fatigue, SOB Myocarditis 0 10 d 49% Normal Global hypokinesis NA Y T2 ratio 2.2 Diffuse edema (image: basal) NA NA N N N Y 1035 Y 62 ms NA Effusion (10 mm) GGOs in the left upper lobe and pleural and pericardial effusions (CT)
Manka et al22 Raised Raised NA Raised Fever, dyspnea Diffuse myocardial injury 0 6 d 59% 72% No NA Y Diffuse edema NA NA N N N Y 1090 Y 56 ms NA NA NA
Pavon et al23 Raised NA NA NA Chest pain, dyspnea Late acute myocarditis 0 6 wk 42% NA Mild hypokinesia NA NA NA NA NA Y Subepicardial Anterior IVS, inferior and inferolateral wall base. Mid cavity apex NA NA Y 55-57 ms NA NA GGOs in the right lung
Sardari et al24 Normal at time of CMR NA NA NA Dyspnea, fever Myocarditis 0 3 wk 50% NA NA NA Y Mid inferoseptal and inferior wall NA NA Y Subepicardial Mid inferior wall NA NA NA NA NA NA NA
Gnecchi et al25 Raised NA Raised Raised Chest pain, fever Myocarditis 0 11 d NA NA Hypokinesia of inferior and inferoseptal segment (echo) NA Y Patchy lateral wall NA NA Y Subepicardial Lateral wall NA NA NA NA NA NA NA
Paul et al26 Raised NA NA NA Chest pain, fatigue Myocarditis 0 NA NA NA NA NA NA NA NA NA Y Subepicardial Lateral and inferior wall NA NA NA NA NA NA NA
Sala et al27 Raised Raised NA Raised Chest pain, dyspnea for 3 d Acute virus-negative lymphocytic myocarditis associated with SARS-CoV-2 0 7 d 64% NA Mild hypokinesia at basal and mid left ventricular segments NA Y Diffuse basal and mid level, IVS NA NA N N N Y 1188 ms Y 61 ms NA NA Bilateral GGOs; no pleural effusion
Bonnet et al28 Raised Raised NA NA Respiratory distress Myocarditis with underlying isolated ventricular noncompaction 0 30 d NA NA NA NA NA NA NA NA Y Subepicardial Inferior wall mid cavity (image) NA NA NA NA NA NA Consolidation
Kim et al29 Raised Raised NA NA Fever, dyspnea Myocarditis 0 NA NA NA NA LVH, LV mass index: 111.3 g/m2 Y ratio 2.2 Diffuse lateral LV wall NA NA Y Extensive transmural Diffuse lateral LV Y Mid-septum, 1431 ms; lateral wall, 1453 ms NA NA NA NA Bilateral multifocal consolidation
Doyen et al30 Raised NA NA NA Vomiting, diarrhea, fever, dyspnea (history of hypertension) Myocarditis 0 NA NA NA No (echo) LVH (echo) (chronic htn) NA NA NA NA Y Subepicardial Apex and inferolateral wall (mid cavity in figure) NA NA NA NA NA NA Bilateral GGOs and condensations
Warchoł et al31 Raised NA Raised NA VT Myocarditis 0 NA 20% NA Global LV hypokinesia NA N N NA NA Y Large, patchy, and linear localized subepicardially and intramurally Basal and mid-cavity segments of the inferior and inferolateral wall and in the apical segments of the inferior wall NA NA NA NA NA NA Left atrial enlargement NA
Sassone et al32 Raised NA NA Raised Chest pain Acute myocarditsi 0 NA NA NA NA NA Y Mid-basal LV lateral wall NA NA Y Subepicardial (image) Mid-basal LV lateral wall NA NA NA NA NA NA Bilateral GGOs,consolidationS (CT)
Salamanca et al33 Raised NA NA NA Dyspnea, syncope Myocarditis 0 14 d 75% NA No NA Y Diffuse with less involvement of inferolateral wall NA NA N N N Y 1120 ms NA NA 36% NA Bilateral pneumonia (CXR)
Bernardi et al34 Raised NA NA NA Chest pain, fever Takutsubo 0 NA 22% NA Hypokinesia of medio-apical segments of the left ventricle with the typical apical ballooning pattern NA Y Mid-apical segments of the left ventricle NA NA N N N NA NA Y NA NA NA Apical thrombus
Fischer et al35 Raised Raised NA Raised Chest pain, fever Acute myocarditsi 0 4 d 48% Normal Mild diffuse hypokinesia (echo) NA Y Posterolateral LV NA NA Y Subepicardial Posterolateralbasal LV wall NA NA NA NA NA Effusion Normal (CT)
Bernal-Torres et al36 Raised NA NA NA Papitations, no respiratory symptoms Myocarditis 0 18 d 60% NA NA NA Y Transmural biventricular NA NA Y Subepicardial (image) Inferobasal NA NA NA NA NA NA Alveolar opacities; GGOs (CT)
Weinsaft et al37 Raised NA NA NA Chest pain Myocarditis 0 NA 38% NA Global hypokinesia (echo) NA Y Lateral wall NA NA Y Subepicardial Lateral wall mid n basal NA NA NA NA NA NA
Madamanchi et al38 Raised NA NA NA Syncope Myocarditis 0 NA 33% NA Dyskinesis of the inferolateral wall NA N N NA NA Y Extensive subepicardial Entire anterior, inferior, and lateral walls, and mid anteroseptal wall NA NA Normal NA NA NA Ct: inferolateral myocardial wall thinning, consolidation Consolidation (CT)
Yuan et al39 NA NA NA NA Chest pain, fever and muscle ache Myocarditis 0 5 d Decreased slightly NA NA NA Y Apical LV NA N N N N NA NA NA NA NA NA Nodular calcification in left upper lobe and local thickening of the right pleura
Garot et al40 Raised Raised NA Raised Cough, fever, fatigue, and myalgias Myocarditis 0 Day 7 and 14 33% NA Marked diffuse hypokinesis 14 mm Y LV basal posterolateral wall NA Y (LV basal posterolateral wall indicating hyperemia) Y Nodular subepicardial enhancement Of the LV basal posterolateral wall Y Anteroseptal 1102 ms; posterolateral 1209 ms Y 57 ms in anteroseptal and 69 ms in posterolateral 33% in anteroseptal; 39% in posterolateral NA No perfusion defect Crazy paving
Monmeneu et al41 tNt raised raised NA Raised Fever, dry cough, and haemoptoic sputum Subacute myopericarditis 0 Day 15 53% NA Diffuse hypokinesia LVH Y Subepicardial pattern was seen in the lateral, anterior, inferior, and apical segments NA NA Y Extensive, patchy intramyocardial subepicardial enhancement Affecting the entire lateral, anterior, inferior, and apical septal walls and the pericardium Y Average, 1110 ms; mid-septum, 1047 ms; lateral wall, 1204 ms Y Average, 60 ms; mid-septum, 53 ms; lateral wall, 67 ms Average, 33%; mid-septum, 29%; lateral wall, 39% Pericardial edema without associated effusion GLS decreased Diffuse bilateral opacities; right pleural effusion
Oberweis et al42 Raised hs troponin T: 0.044 ng/mL Raised 5112 pg/mL) NA Raised Fever, cough, fatigue Myocarditis 0 Day 3 41% 46% NA NA NA NA NA NA Y Subepicardial Lateral LV wall NA 1148±67 ms NA NA NA Mild thickening of pericardium Bilateral lower lobe pneumonias; bilateral pleural effusions (CT)
Frédéric et al43 Raised 15.4 μg/L Raised NA Raised Chest pain, dyspnea Myopericarditis 0 5 d NA NA NA NA Y Basal inferolateral NA NA Y Subepicardial Basal inferolateral NA NA NA NA NA Effusion Pleural effusion, atelectasis
EGE indicates early gadolinium enhancement; GGO, ground-glass opacities; GLS, global longitudinal strain; LVEF, left ventricular ejection fraction; LVH, left ventricle hypertrophy; N, normal; NA, data not available; RVEF, right ventricular ejection fraction; RWMA, regional wall motion abnormality; STIR, short tau inversion recovery; Y, yes.

The most prevalent MRI findings (described out of >100 patients) included T1 mapping abnormalities (109/150; 72.7%), T2 mapping abnormalities (91/144; 63.2%), and LGE (85/199; 42.7%). Other common findings (described out of <100 patients) included perfusion deficits (18/21; 85.71%), edema on T2-weighted sequences (46/90; 51.11%), and extracellular volume mapping (ECV) abnormalities (21/40; 52.5%). Findings on perfusion imaging were mentioned for a total of 21 patients (Supplementary Table 2, Supplemental Digital Content 2, https://links.lww.com/PAS/B52). Knight et al11 described ischemia in 9/19 and inducible ischemia in 8/19 patients. Pericardial effusion and pericardial LGE were also noted in 24% and 22% of the patients, respectively. One patient was having preexisting ventricular noncompaction (Table 3).28 Mean left ventricular ejection fraction across the studies was 52% and right ventricular ejection fraction was 56%. Right ventricle (RV) dysfunction was described in 4 of 9 studies which mentioned RV function. Information regarding regional wall motion abnormalities was provided in 13 case reports and diffuse hypokinesia was the most common pattern (9/13; 69.2%) (Supplementary Table 2, Supplemental Digital Content 2, https://links.lww.com/PAS/B52).

Distribution of LGE and Mapping Abnormalities

LGE, when present, was most commonly seen involving the basal (60%) and mid-ventricular (67%) part of the left ventricle (LV) and involving the inferior, inferolateral, and inferoseptal LV (26/41; 63.4%). A subepicardial nonischemic pattern of LGE typical for myocarditis was the most prevalent pattern (43/53; 81.1%). Mid-wall LGE was seen in 33.9% (15/53) and ischemic pattern of LGE (subendocardial LGE in coronary distribution) was seen in ~17% (21/53). Diffuse biventricular and transmural LGE were seen in one case each (Supplementary Table 2, Supplemental Digital Content 2, https://links.lww.com/PAS/B52).

Average T1 and T2 mapping values across all the studies were 1165.59 and 54.65 ms. Very few studies described the segments with mapping abnormalities. Whereas T1 mapping values were variable, T2 mapping value was higher in lateral segments (67 ms: lateral wall in 1 case and 69 ms: posterolateral wall in 1 case). Average ECV values were also higher (34.24 ms) in COVID-19 survivors (Supplementary Table 2, Supplemental Digital Content 2, https://links.lww.com/PAS/B52).

Uncommon Imaging Findings on CMR

Some uncommon imaging findings were also described. For example, LV hypertrophy (including pseudohypertrophy due to inflammation) was described in 6 case reports (Table 3). Two case reports mentioned the presence of apical left ventricular thrombus including one in the presence of the Takotsubo syndrome.19,34 Pericardial thickening was also found in 2 cases.41,42 Reduced global longitudinal strain, left atrial enlargement, and diastolic dysfunction of the LV were described in one case each.18,31,41 In addition, lymphadenopathy was conspicuously absent in all the patients.

Imaging Findings in Children

A case series of four children admitted in the intensive care described Kawasaki-like clinical features in these children. Acute myocarditis occurred within a week of symptom onset. CMR (3 in acute stage, 1 in recovery phase) demonstrated diffuse myocardial edema (on T2/STIR sequences and native T1 mapping) without any LGE, suggesting myocardial inflammation without necrosis/fibrosis (Fig. 2).10 However, 2 case reports in children also described the presence of subepicardial LGE suggestive of typical myocarditis pattern along with edema.35,42 In the report by Oberweis et al,42 the repeat CMR showed complete resolution of inflammation and LGE on immunomodulatory treatment.

F2
FIGURE 2:
Cardiac MRI of 4 children with Kawasaki-like symptoms due to COVID-19 and with clinical diagnosis of acute myocarditis. The top panel shows the cine images with minimal pericardial effusion. The second panel demonstrates edema in the form of increased T2-STIR signal intensity ratio between the myocardium and the skeletal muscle (>2:1) in patient 2, 3, and 4. The third panel demonstrates abnormal native-T1 mapping (>1100 ms) in patients 2, 3, and 4 and normal native T1 in patient 1. The bottom panel demonstrates absence of late gadolinium enhancement in patients 2 and 3 (myocardial null times were recognized as too short in patient 4 but could not be repeated due to lack of further patient cooperation). (Reproduced with permission from Blondiaux et al.10)

Endomyocardial Biopsy (EMB) Findings

EMB data were available in 2 studies.12,27 Three patients with severe disease were referred for EMB in the study by Puntmann et al12 and all of them showed active lymphocytic inflammation without any viral genome. In the case report of a patient with Takotsubo-like syndrome, there was presence of diffuse lymphocytic inflammatory infiltrates, interstitial edema, and patchy necrosis without any replacement fibrosis. No viral genome was found on molecular analysis.27

Findings on Follow-up CMR

Follow-up CMR findings were available for 3 patients (in 3 different case reports). Whereas myocardial edema (T2 hyperintensity) resolved in all the 3 cases, 1 case with subepicardial LGE remained constant, suggesting irreversible fibrosis. Decrease in LV wall thickness and improvement in left ventricular ejection fraction were also seen in 1 case (Table 4).39,40,42

TABLE 4 - Follow Up MRI Findings in Cases Where Repeat MRI was Performed
References Number of Patients Duration of Follow Up MRI Follow Up MRI Findings
Yuan et al39 1 Day 12 T2WI hyperintensity resolved, indicating myocarditis
Garot et al40 1 Day 14 Significant reverse LV remodelling (wall thickness decreased to 11 mm from 14 mm), LV end diastolic index decreased to 88 mL/m2 from 127 mL/m2 and LVEF improved to 54% from 33%), decrease of focal myocardial edema and EGE in the posterolateral wall, and stable LGE lesions in the subepicardium of the posterolateral wall
Oberweis et al42 1 Day 7 Normal systolic function (initially 53%) and resolution of myocardial edema. Native T1 mapping showed slightly decreased T1 values at 1048±78 ms (from initial 1110 ms)
EGE indicates early gadolinium enhancement; LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction.

DISCUSSION

CMR may depict various imaging manifestations of myocardial injury caused by COVID-19. In this systematic review, we have compiled the data from the existing literature regarding the various common and uncommon imaging findings in patients with COVID-19 on cardiac MRI. Most of the studies in this review were of fair quality suggesting risk of some bias. However, the scarcity of data in the literature on this subject in this emergent pandemic situation makes this bias unavoidable. Most of the data is descriptive, nonblinded, and describes the preliminary experience in this less well-known entity. However, we aimed to evaluate the imaging findings on CMR and these shortcomings were not strong enough to invalidate our results.

Myocarditis was the most common imaging diagnosis (~40%) on CMR in recovered/active patients with COVID-19. More than three-fourth of the cases from the largest cohort till date had findings of myocarditis on CMR, demonstrating that the prevalence of myocardial injury in COVID-19 is higher than previously thought.12 The common imaging findings on CMR included increased T1 and T2 mapping values and edema on T2/STIR sequences. LGE was seen in less than half of the patients. When present, LGE was most commonly seen in the subepicardial location in inferior, inferoseptal, and inferolateral segments, similar to viral myocarditis.9 The absence or small amounts of LGE observed in many cases is in agreement with the limited number of histologic findings published in the literature for this disease, reporting limited or absent myocyte necrosis.12,27 EMB data in 2 of these studies showed presence of lymphocytic infiltrates without evidence of viral genome, again suggesting that immune-mediated myocardial inflammation is the principal mechanism of cardiac involvement in COVID-19.12,27 This inflammation (myocarditis) is evident on CMR as raised T1 and T2 mapping values and edema.8 In most of the cases, the involvement was diffuse rather than regional.

Mapping abnormalities were more common than T2 hyperintensity (edema) in our pooled analysis. The proposed reason may be that tissue characterization using only the signal intensities may not be possible or accurate in cases of diffuse inflammation. Because of diffuse increase in signal intensity and lack of reference “normal” myocardium, discrete lesions may not be identified. Further, even if diffusely inflamed myocardium shows raised signal intensity ratio compared with the skeletal muscle, the coexisting skeletal muscle edema may give rise to false-negative results.8 Mapping techniques can allow for direct quantitative pixel by pixel measurement of myocardial relaxation times in acute inflammation, thus avoiding the limitations of semiquantitative techniques. Inflamed myocardium shows raised T1, T2, and ECV values. Previous studies have shown excellent diagnostic accuracy of mapping techniques for suspected myocarditis. Pooled area under the curve (AUC) for the detection of acute myocarditis, from the available literature, are 89, 80, and 74 for T1, T2, and ECV mapping, respectively, compared with 73, 73, and 83 for T2, early gadolinium enhancement and LGE, respectively.8 The MyoRacer myocarditis trial also demonstrated native T1 mapping to be most accurate (diagnostic accuracy: 81%) of all CMR parameters for acute myocarditis.44

In the largest studies included in this review, ventricular functions were predominantly normal. These studies have demonstrated that tissue abnormalities precede functional abnormalities and that the patients could be in relatively earlier phase of cardiac involvement.14 Indeed, our results showed that mapping abnormalities were more prevalent than ventricular dysfunction. This further strengthens the case for mapping techniques as a sensitive tool for detecting early myocardial involvement in COVID. However, it is important that patients are further followed up longitudinally for possible adverse functional remodeling of myocardium. RV dysfunction was described in four studies (18/153 patients; 12%). The proposed mechanism may relate to the fact that even slight increase in pulmonary vascular resistance (due to pulmonary disease) can cause impairment of RV function as it acts as a passive conduit chamber.14

Ischemia was seen in 9 of 29 patients with identifiable CMR findings of COVID-19 in the series by Knight and colleagues, out of which 4 were concomitant with nonischemic pattern. There is an increasing evidence that there is abnormal activation of the coagulation cascade and microcirculatory dysfunction, which happens due to heightened immune response and endothelial dysfunction in COVID-19 and this can cause ischemia and acute coronary syndrome.45 When present, it is associated with poor prognosis. This mechanism is also postulated for the occurrence of ventricular thrombus as a rare complication of COVID-19 infection.19,34

Approximately one-fifth of the patients in this review had normal CMR despite cardiac symptoms. This can have two possible explanations. Considering the fact that normal CMR was more commonly seen in case series with a higher gap between symptoms and time of acquisition, the most likely reason is that patients may have had myocarditis, but were imaged later in the course of the disease when edema had already resolved. Another reason could be that symptoms were because of residual pulmonary involvement rather than the cardiac.12,14

The mean duration of CMR from symptom onset ranged from day 2 to day 71. In 3 of the largest studies, the mean duration was 71, 46, and 47 days, respectively, and there were significant findings on CMR, denoting that cardiac involvement in COVID-19 persists beyond the acute stage and without any trend toward the decrease in these imaging findings through the recovery period. This highlights the importance of clinical surveillance beyond the initial phase in patients with COVID-19. Some additional insight into the course of this disease was also provided by the 3 cases in which follow-up MRI was done (Table 4). Although, edema seemed to decrease in all the 3 cases, LGE lesions did not resolve in 1 study (which described the follow-up LGE findings) suggesting fibrosis. The reversibility of myocardial dysfunction and inflammation puts forth the Takotsubo syndrome as a reasonable differential diagnosis in these cases. Indeed, Takotsubo syndrome has also been described in association with COVID-19.27,34 The putative mechanism is stress-related exaggerated catecholamine response. However, further studies with longitudinal follow-up are warranted to study the evolution of myocardial lesions in COVID-19.

The CMR findings in children with COVID-19 were also intriguing. In the case series by Blondiaux and colleagues, all 4 children presented with Kawasaki-like multisystem inflammatory syndrome. CMR showed edema and inflammation without LGE, suggesting absence of necrosis, contrary to the findings in many of the adult patients.18,29 This is similar to histopathologic findings in Kawasaki disease where inflammatory infiltrates are found in the myocardium (likely because of cytokine storm), with little myocardial necrosis. This is different from viral myocarditis where the immune response is directed toward the viral infiltration of the myocardium. In Kawasaki-like disease, myocardial inflammation peaks at day 10 after the disease onset and disappears by 20 days and the putative mechanism is the cytokine storm syndrome. Similar findings were noted in this case series.10 This multisystem inflammatory syndrome, with mucocutaneous, dermatologic, and gastrointestinal manifestations along with cardiac dysfunction is being frequently described in children hospitalized with COVID-19 infection.46

As described above, CMR can help in detection, prognostication, management, and follow-up of myocardial injury in COVID-19 survivors and to avoid invasive procedures. Performing EMB or coronary angiograms for the diagnosis of myocardial involvement in this setting could be challenging given the risks associated with the procedure, the critical condition of the patients, and potential viral exposure to health care workers. Recently published clinical scenario-based guidelines for COVID-19 suggest that CMR is indicated in patients with signs of myocardial infarction with nonobstructive coronary arteries or in patients with new-onset LV systolic dysfunction without the evidence of CAD.47 CMR can help detect edema, necrosis, and contractile dysfunction allowing for close monitoring of affected individuals and promptly deciding appropriate therapeutic strategies.

This review has the following limitations. Data search was restricted to articles published in English-language literature. This may have resulted in missing data published in other languages. The number of studies included in this review is less because of paucity of data on this topic. Moreover, the data across the studies is heterogeneous as regards clinical presentation, description of imaging findings, sample sizes, data availability, scanner used, acquisition protocols, and individual experience of the personnel interpreting the data. Hence, the findings from this review should be interpreted in a suitable clinical context and with caution. Some findings such as follow-up MRI features were based on only 3 case reports. Although, a meta-analysis could not be performed in this review due to lack of adequate robust studies, a meta-analysis will be required in the future to address these challenges. Also, the future cardiovascular outcomes of these subclinical finding remain to be studied.

We conclude that the findings on CMR in patients recovered from COVID-19 may provide insights into the prevalence, mechanism, extent, and prognosis of myocardial injury in these patients. The most common imaging diagnosis is myocarditis and the most common imaging findings include T1 and T2 mapping abnormalities and myocardial edema followed by LGE. Our systematic review highlights the utility of CMR with its new quantitative mapping techniques as an essential diagnostic tool to detect diffuse myocardial inflammation associated with COVID-19 infection. It is essential that the physicians and radiologists interpreting CMR are familiar with a myriad of imaging spectrum of COVID, so that they can influence the clinical decision-making in these patients.

REFERENCES

1. WHO Director-General’s opening remarks at the media briefing on COVID-19—March 11, 2020. Available at: https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020. Accessed July 25, 2020.
2. Shafi AMA, Shaikh SA, Shirke MM, et al. Cardiac manifestations in COVID-19 patients—a systematic review. J Cardiac Surg. 2020;35:1988–2008.
3. Zheng Y-Y, Ma Y-T, Zhang J-Y, et al. COVID-19 and the cardiovascular system. Nat Rev Cardiol. 2020;17:259–260.
4. Chen C, Zhou Y, Wang DW. SARS-CoV-2: a potential novel etiology of fulminant myocarditis. Herz. 2020;45:230–232.
5. Kociol Robb D, Cooper Leslie T, Fang, James C, et al. Recognition and initial management of fulminant myocarditis. Circulation. 2020;141:e69–e92.
6. Moher D, Liberati A, Tetzlaff J, et al. PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6:e1000097.
7. Study Quality Assessment Tools. National Heart, Lung, and Blood Institute (NHLBI). Available at: https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools. Accessed April 14, 2020.
8. Ferreira VM, Schulz-Menger J, Holmvang G, et al. Cardiovascular magnetic resonance in nonischemic myocardial inflammation: expert recommendations. J Am Coll Cardiol. 2018;72:3158–3176.
9. Friedrich MG, Sechtem U, Schulz-Menger J, et al. Cardiovascular magnetic resonance in myocarditis: a JACC white paper. J Am Coll Cardiol. 2009;53:1475–1487.
10. Blondiaux E, Parisot P, Redheuil A, et al. Cardiac MRI of Children with Multisystem Inflammatory Syndrome (MIS-C) associated with COVID-19. Radiology. 2020;297:E283–E288.
11. Knight DS, Kotecha T, Razvi Y, et al. COVID-19: myocardial injury in survivors. Circulation. 2020;142:1120–1122.
12. Puntmann VO, Carerj ML, Wieters I, et al. Outcomes of Cardiovascular Magnetic Resonance Imaging in Patients Recently Recovered From Coronavirus Disease 2019 (COVID-19). JAMA Cardiol. 2020;5:1265–1273.
13. Esposito A, Palmisano A, Natale L, et al. Cardiac magnetic resonance characterization of myocarditis-like acute cardiac syndrome in COVID-19. JACC Cardiovasc Imaging. 2020;13:2462–2465.
14. Huang L, Zhao P, Tang D, et al. Cardiac involvement in recovered COVID-19 patients identified by magnetic resonance imaging. J Am Coll Cardiol Img. 2020;13:2330–2339.
15. Caballeros Lam M, de la Fuente Villena A, Hernández Hernández A, et al. Cardiac magnetic resonance characterization of COVID-19 myocarditis. Rev Esp Cardiol. 2020;73:863–864.
    16. Coyle J, Igbinomwanhia E, Sanchez-Nadales A, et al. A recovered case of COVID-19 myocarditis and ARDS treated with corticosteroids, tocilizumab, and experimental AT-001. JACC Case Rep. 2020;2:1331–1336.
      17. Beşler MS, Arslan H. Acute myocarditis associated with COVID-19 infection. Am J Emerg Med. 2020. doi: 10.1016/j.ajem.2020.05.100.
        18. Inciardi RM, Lupi L, Zaccone G, et al. Cardiac involvement in a patient with coronavirus disease 2019 (COVID-19). JAMA Cardiol. 2020;5:819–824.
        19. Gravinay P, Issa N, Girard D, et al. CMR and serology to diagnose COVID-19 infection with primary cardiac involvement. Eur Heart J Cardiovasc Imaging. 2020:jeaa169. doi: 10.1093/ehjci/jeaa169.
        20. Trogen B, Gonzalez FJ, Shust GF. COVID-19-associated myocarditis in an adolescent. Pediatr Infect Dis J. 2020;39:e204–e205.
        21. Luetkens JA, Isaak A, Zimmer S, et al. Diffuse myocardial inflammation in COVID-19 associated myocarditis detected by multiparametric cardiac magnetic resonance imaging. Circ Cardiovasc Imaging. 2020;13:e010897.
          22. Manka R, Karolyi M, Polacin M, et al. Myocardial edema in COVID-19 on cardiac MRI. J Heart Lung Transplant. 2020;39:730–732.
            23. Pavon AG, Meier D, Samim D, et al. First documentation of persistent SARS-CoV-2 infection presenting with late acute severe myocarditis. Can J Cardiol. 2020;36:1326.e5–1326.e7.
              24. Sardari A, Tabarsi P, Borhany H, et al. Myocarditis detected after COVID-19 recovery. Eur Heart J Cardiovasc Imaging. [Epub May 27, 2020]. doi: 10.1093/ehjci/jeaa166.
                25. Gnecchi M, Moretti F, Bassi EM, et al. Myocarditis in a 16-year-old boy positive for SARS-CoV-2. Lancet. 2020;395:e116.
                26. Paul J-F, Charles P, Richaud C, et al. Myocarditis revealing COVID-19 infection in a young patient. Eur Heart J Cardiovasc Imaging. 2020;21:776.
                27. Sala S, Peretto G, Gramegna M, et al. Acute myocarditis presenting as a reverse Tako-Tsubo syndrome in a patient with SARS-CoV-2 respiratory infection. Eur Heart J. 2020;41:1861–1862.
                28. Bonnet M, Craighero F, Harbaoui B. Acute myocarditis with ventricular noncompaction in a COVID-19 patient. JACC Heart Fail. 2020;8:599–600.
                29. Kim I-C, Kim JY, Kim HA, et al. COVID-19-related myocarditis in a 21-year-old female patient. Eur Heart J. 2020;41:1859.
                30. Doyen D, Moceri P, Ducreux D, et al. Myocarditis in a patient with COVID-19: a cause of raised troponin and ECG changes. Lancet. 2020;395:1516.
                31. Warchoł I, Dębska-Kozłowska A, Karcz-Socha I, et al. Terra incognita: clinically suspected myocarditis in a patient with severe acute respiratory syndrome coronavirus 2 infection. Pol Arch Intern Med. 2020;130:446–448.
                32. Sassone B, Muser D, Bruno A, et al. Concealed SARS-CoV-2 interstitial pneumonia unmasked by infarct-like acute myocarditis. Eur Heart J Case Rep. 2020;4:1–2.
                  33. Salamanca J, Díez-Villanueva P, Martínez P, et al. COVID-19 “fulminant myocarditis” successfully treated with temporary mechanical circulatory support. JACC Cardiovasc Imaging. doi: 10.1016/j.jcmg.2020.05.003.
                    34. Bernardi N, Calvi E, Cimino G, et al. COVID-19 pneumonia, Takotsubo syndrome and left ventricle thrombi. JACC Case Rep. 2020;2:1359–1364.
                    35. Fischer Q, Brillat-Savarin N, Ducrocq G, et al. Case report of an isolated myocarditis due to COVID-19 infection in a paediatric patient. Eur Heart J Case Rep. 2020;4:1–5.
                    36. Bernal-Torres W, Herrera-Escandón Á, Hurtado-Rivera M, et al. COVID-19 fulminant myocarditis: a case report. Eur Heart J Case Rep. 2020;4:1–6.
                      37. Weinsaft JW, Kochav J, Kim J, et al. CMR findings in COVID-19 associated myocarditis—Society for Cardiovascular Magnetic Resonance. Available at: https://scmr.org/page/COVID2001. Accessed July 15, 2020.
                        38. Madamanchi C, Steigner M, Kwong R. Prior myocarditis in setting of COVID-19 infection—Society for Cardiovascular Magnetic Resonance. Available at: https://scmr.org/page/COVID2002. Accessed July 15, 2020.
                          39. Yuan W-F, Tang X, Zhao X-X. An “asymptomatic” driver with COVID-19: atypical suspected myocarditis by SARS-CoV-2. Cardiovasc Diagn Ther. 2020;10:242–243.
                          40. Garot J, Amour J, Pezel T, et al. SARS-CoV-2 fulminant myocarditis. JACC Case Rep. 2020;2:1342–1346.
                          41. Monmeneu JV, Dominguez Mafe E, Andres Soler J, et al. Subacute perimyocarditis in a young patient with COVID-19 infection. Eur Heart J Case Rep. 2020;4:1–3.
                          42. Oberweis M-L, Codreanu A, Boehm W, et al. Pediatric life-threatening coronavirus disease 2019 with myocarditis. Pediatr Infect Dis J. 2020;39:e147.
                          43. Frédéric L, Caroline C-N, Nachwan G. Myopericarditis, rhabdomyolysis, and acute hepatic injury. Circulation Cardiovasc Imaging. 2020;13:e010907.
                            44. Lurz P, Luecke C, Eitel I, et al. Comprehensive cardiac magnetic resonance imaging in patients with suspected myocarditis: the MyoRacer-Trial. J Am Coll Cardiol. 2016;67:1800–1811.
                            45. Varga Z, Flammer AJ, Steiger P, et al. Endothelial cell infection and endotheliitis in COVID-19. Lancet. 2020;395:1417–1418.
                            46. Dufort EM, Koumans EH, Chow EJ, et al. Multisystem Inflammatory Syndrome in Children in New York State. N Engl J Med. 2020;383:347–358.
                            47. Rudski L, Januzzi JL, Rigolin VH, et al. Multimodality Imaging in Evaluation of Cardiovascular complications in Patients with COVID-19. J Am Coll Cardiol. 2020;76:1345–1357.
                            Keywords:

                            cardiac magnetic resonance imaging; coronavirus disease 2019; systematic review

                            Supplemental Digital Content

                            Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.