Diagnosis of Myocarditis at Cardiac MRI: The Continuing Quest for Improved Tissue Characterization
2019; Radiological Society of North America; Volume: 292; Issue: 3 Linguagem: Inglês
10.1148/radiol.2019191476
ISSN1527-1315
Autores Tópico(s)Antimicrobial Resistance in Staphylococcus
ResumoHomeRadiologyVol. 292, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialDiagnosis of Myocarditis at Cardiac MRI: The Continuing Quest for Improved Tissue CharacterizationAlbert de Roos Albert de Roos Author AffiliationsFrom the Department of Radiology, Leiden University Medical Center, Albinusdreef 2, Leiden 2333 ZA, the Netherlands.Address correspondence to the author (e-mail: [email protected]).Albert de Roos Published Online:Jul 30 2019https://doi.org/10.1148/radiol.2019191476MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Baessler et al in this issue.Download as PowerPointOpen in Image Viewer Albert de Roos, MD, is professor of radiology at Leiden University Medical Center, Leiden, the Netherlands. His scientific work relates to cardiovascular imaging using CT and MRI. He has served as a deputy editor for Radiology since 2008 and more recently as senior deputy editor for Radiology until 2019. Currently he is senior consultant to the editor for Radiology: Cardiothoracic Imaging.IntroductionThe diagnosis of myocarditis with use of clinical and imaging criteria is challenging due to its extremely heterogeneous manifestations (1). Because the clinical presentation of myocarditis is frequently nonspecific, it may mimic more common causes of chest pain, dyspnea, and palpitations. Furthermore, myocarditis may manifest at variable time points after the initial insult (eg, acute heart failure with infarct-like presentation or chronic dilated cardiomyopathy). Moreover, a wide spectrum of infectious pathogens, toxic agents, and hypersensitivity reactions may trigger the disease process, further complicating the clinical presentation (1). Viral infections appear to be the most common pathogens causing myocarditis. These pathogens initially insult the myocardium. This is followed by autoimmune myocardial injury, which causes myocardial remodeling that may ultimately result in dilated cardiomyopathy. The prevalence of undiagnosed myocarditis in postmortem series ranges from 9% to 44%, illustrating the complexity of initial diagnosis (2). Many diagnostic features may have limited accuracy, and only histologic or immunohistologic evidence of an inflammatory cell infiltrate with or without myocyte damage provides a definite diagnosis.Direct myocardial injury combined with immune-mediated cardiac injury may result in interstitial cellular infiltrates, cell necrosis, myocardial edema, and diffuse myocardial fibrosis. It has also been suggested that activation of matrix metalloproteinases can digest the interstitial collagen and elastin framework of the heart, further stimulating the inflammatory process (1). As a consequence, complex inflammatory reactions may result in variable myocardial tissue changes that may be challenging to define with noninvasive imaging such as cardiac MRI.However, MRI pulse sequences offer a wide range of options to characterize the possible underlying myocardial tissue alterations that may occur in myocarditis. For example, T2-weighted MRI techniques may depict myocardial edema, early gadolinium enhancement may depict myocardial hyperemia and capillary leak, and late gadolinium enhancement may depict necrosis in a nonischemic pattern (eg, subepicardial or midwall pattern). A combination of these MRI features has been used to estimate the likelihood of myocarditis (so-called Lake Louise criteria), although with variable and suboptimal sensitivity and specificity. One problem is that T2-weighted MRI may suffer from artifacts, which decreases the diagnostic accuracy of the MRI examination. Another limitation in diagnosing myocarditis is the evolving nature of myocardial edema and necrosis over time after the initial insult.More recently, T1 and T2 mapping techniques have been explored to provide more quantitative and objective markers of inflammation in myocarditis to improve the diagnostic accuracy (3). However, these techniques may have limited diagnostic accuracy due to the large overlap and heterogeneity of myocardial T1 and T2 values in healthy and inflammatory myocardial disease. The combination of nonenhanced (native) and contrast material–enhanced T1 mapping with hematocrit data allows quantification of the contrast agent within the myocardium to derive the extracellular volume fraction. Myocardial edema will prolong both T1 and T2 relaxation times and may be detected with both T1 and T2 mapping techniques. Native T1 mapping is well suited for the detection of inflammation in acute myocarditis, outperforming standard T2-weighted imaging (4). Accurate T2 mapping is hampered by biologic variability of myocardial T2 times and distribution of disease (eg, focal vs diffuse) and by technical parameters (eg, field strength and specific pulse sequence).Texture analysis of images from cardiac MRI is a new step in the quest for better tissue characterization. The tool is used to extract and quantify the texture of an image based on pixel signal intensity distributions and relationships of values between neighboring pixels by applying various statistical image analysis techniques. Texture analysis quantifies texture features, reflecting intrinsic heterogeneity of different tissues and lesions that may be imperceptible to the human eye (5).Texture analysis has been applied to myocardial T1 and T2 maps in patients with infarct-like acute myocarditis (6). T2 mapping–derived texture features that reflect tissue inhomogeneity showed high sensitivity (89%) and specificity (92%) in the diagnosis of acute infarctlike myocarditis, outperforming Lake Louise criteria and native T1 and T2 measurements of the myocardium.In this issue of Radiology, Baessler et al (7) applied texture analysis to myocardial T1 and T2 maps of myocarditis with acute and chronic symptoms of heart failure. Their study expands on the experience of using texture analysis in the setting of acute versus chronic myocarditis manifesting with heart failure. In their prospective study, 31 participants with acute ( 14 days of symptoms) heart failure–like myocarditis underwent cardiac MRI to test the diagnostic utility of texture analysis. Endomyocardial biopsy was used as the reference standard (7). For comparison, native T1 mapping, T2 mapping, and Lake Louise criteria were evaluated. Detailed histopathologic analysis of samples from myocardial biopsy revealed subgroups of patients with positive and negative findings consistent with myocarditis. Texture analysis of cardiac MRI T1 and T2 maps was performed with an open-source software package. The output of texture analysis resulted in a large number of features that had to be reduced to a limited data set for final analysis. A number of technical steps in feature selection and dimension reduction are reported; this may constitute a limitation of the analysis when a relatively small number of patients are included. Statistical modeling is required to control for the large number of features compared with the relatively small number of patients. Baessler et al attempted to control for these issues to best advantage, but they acknowledge that overfitting might represent an important limitation of the current analysis using 10-fold cross-validation. The ultimate test to demonstrate diagnostic utility and performance will require larger patient groups with preferably external validation.Although some limitations remain, texture analysis of images from cardiac MRI performed better than Lake Louise criteria and T1 and T2 mapping in this cohort of patients with acute and chronic heart failure–like myocarditis (7). The naming of features of texture analysis is difficult to comprehend, but Baessler et al provide some explanation to understand their significance. In acute heart failure–like myocarditis, the features that may reflect tissue inhomogeneity in the setting of myocardial inflammation were most important (ie, mean T2 and the so-called texture feature T2 gray-level nonuniformity), with an area under the receiver operating characteristic curve of 0.76, sensitivity of 81%, and specificity of 71%. In myocarditis with chronic heart failure–like manifestation, other parameters (ie, T2 kurtosis and T1 gray-level nonuniformity) revealed the best diagnostic performance, with an area under the receiver operating characteristic curve of 0.85, sensitivity of 90%, and specificity of 72%. Kurtosis reflects how values in a histogram are distributed, and lower T2 kurtosis in chronic heart failure–like myocarditis may reflect a broader and more inhomogeneous pixel gray-level distribution in the T2 maps (7).In conclusion, the study by Baessler et al suggests that different underlying histopathologic changes may occur in acute versus chronic settings; these differences may be reflected by the various texture features. In our continuing quest for improved tissue characterization, this proof-of-concept study shows the potential of texture analysis of images from cardiac MRI in the differentiation of acute versus chronic heart failure–like myocarditis. Further validation in larger groups of patients, testing potential utility in other myocardial diseases, and external validation of the utility of texture analysis are still required.Disclosures of Conflicts of Interest: Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: receives lecture fees from IDKD. Other relationships: disclosed no relevant relationships.References1. Sagar S, Liu PP, Cooper LT Jr. Myocarditis. Lancet 2012;379(9817):738–747. Crossref, Medline, Google Scholar2. Priori SG, Blomström-Lundqvist C, Mazzanti A, et al. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: The Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the European Society of Cardiology (ESC)Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC). Europace 2015;17(11):1601–1687. Google Scholar3. Schelbert EB, Messroghli DR. State of the art: clinical applications of cardiac T1 mapping. Radiology 2016;278(3):658–676. Link, Google Scholar4. Ferreira VM, Piechnik SK, Dall’Armellina E, et al. T(1) mapping for the diagnosis of acute myocarditis using CMR: comparison to T2-weighted and late gadolinium enhanced imaging. JACC Cardiovasc Imaging 2013;6(10):1048–1058. Crossref, Medline, Google Scholar5. MacKay JW, Murray PJ, Kasmai B, Johnson G, Donell ST, Toms AP. MRI texture analysis of subchondral bone at the tibial plateau. Eur Radiol 2016;26(9):3034–3045. Crossref, Medline, Google Scholar6. Baessler B, Luecke C, Lurz J, et al. Cardiac MRI texture analysis of T1 and T2 maps in patients with infarctlike acute myocarditis. Radiology 2018;289(2):357–365. Link, Google Scholar7. Baessler B, Luecke C, Lurz J, et al. Cardiac MRI and texture analysis of myocardial T1 and T2 maps in myocarditis with acute versus chronic symptoms of heart failure. Radiology 2019;292:608–617. Link, Google ScholarArticle HistoryReceived: July 1 2019Revision requested: July 10 2019Revision received: July 11 2019Accepted: July 15 2019Published online: July 30 2019Published in print: Sept 2019 FiguresReferencesRelatedDetailsCited ByDilated cardiomyopathy complicated by an intracavitary thrombus and acute heart failure: A rare presentation of systemic lupus erythematosusLamyaeZinoune, MosaabMaazouzi, KarimaBenbouchta, SaidaAmaqdouf, NohaEl ouafi, NabilaIsmaili2022 | Annals of Medicine and Surgery, Vol. 82Endomyocardial biopsy facilitates diagnosis of eosinophilic granulomatosis with polyangiitis (EGPA): a case reportAlyeesha B.Wilhelm, SimonPinsky, ShahzadAhmad, ArwynCunningham, KhaledChatila, Paul J.Boor, Heather L.Stevenson2022 | Cardiovascular Pathology, Vol. 58Integration of multiple imaging platforms to uncover cardiovascular defects in adult zebrafishAnabelaBensimon-Brito, Giulia L MBoezio, JoãoCardeira-da-Silva, AstridWietelmann, SrinathRamkumar, Pia RLundegaard, Christian S MHelker, RadhanRamadass, JanettPiesker, ArnoNauerth, ClemensMueller, Didier Y RStainier2022 | Cardiovascular Research, Vol. 118, No. 12CNN-KCL: Automatic myocarditis diagnosis using convolutional neural network combined with k-means clusteringDanialSharifrazi, RoohallahAlizadehsani, Javad HassannatajJoloudari, Shahab S.Band, SadiqHussain, Zahra AlizadehSani, FereshtehHasanzadeh, AfshinShoeibi, AbdollahDehzangi, MehdiSookhak, HamidAlinejad-Rokny2022 | Mathematical Biosciences and Engineering, Vol. 19, No. 3Laboratory and Imaging Evaluation of Cardiac Involvement in Patients with Post-Acute COVID-19ErsinSarıçam, Ali DoğanDursun, GülçinTürkmen Sarıyıldız, NalanCan, EnginBozkurt, UğurGönüllü, NihalBasay, MeralTürkmen, AbdülazizDenli, MustafaÜnlü2021 | International Journal of General Medicine, Vol. Volume 14Unusual Presentation of Acute Perimyocarditis Following SARS-COV-2 mRNA-1237 Moderna VaccinationFatimaKhogali, RababAbdelrahman2021 | CureusAccompanying This ArticleCardiac MRI and Texture Analysis of Myocardial T1 and T2 Maps in Myocarditis with Acute versus Chronic Symptoms of Heart FailureJul 30 2019RadiologyRecommended Articles The Many Faces of Myocarditis: Role of Cardiac MRIRadiology2021Volume: 302Issue: 1pp. 70-71State of the Art: Clinical Applications of Cardiac T1 MappingRadiology2016Volume: 278Issue: 3pp. 658-676Cardiac MRI and Texture Analysis of Myocardial T1 and T2 Maps in Myocarditis with Acute versus Chronic Symptoms of Heart FailureRadiology2019Volume: 292Issue: 3pp. 608-617Cardiac MRI Assessment of Nonischemic Myocardial Inflammation: State of the Art Review and Update on Myocarditis Associated with COVID-19 VaccinationRadiology: Cardiothoracic Imaging2021Volume: 3Issue: 6Cardiac MRI Texture Analysis of T1 and T2 Maps in Patients with Infarctlike Acute MyocarditisRadiology2018Volume: 289Issue: 2pp. 357-365See More RSNA Education Exhibits MRI Mapping in the study of Myocardial DiseaseDigital Posters2022Myocardial T1 And T2 Mapping: Technique And ApplicationsDigital Posters2021The Game Changer: Cardiac MR Imaging(CMR) in Myocarditis and Its Utility From Diagnosis to PrognosisDigital Posters2020 RSNA Case Collection Acute myocardial InfarctionRSNA Case Collection2021Cardiac sarcoidosisRSNA Case Collection2020Cardiac fibromaRSNA Case Collection2020 Vol. 292, No. 3 Metrics Altmetric Score PDF download
Referência(s)