Author + information
- Julian A. Luetkens, MD,
- Jonas Doerner, MD,
- Hans Schild, MD and
- Claas P. Naehle, MD∗ ()
- ↵∗University of Bonn, Department of Radiology, Sigmund-Freud-Strasse 25, 53127 Bonn, Germany
With interest we read the article by Radunski et al. (1) reporting on cardiac magnetic resonance (CMR) using quantitative tissue parameters in patients suspected of having severe myocarditis. However, several aspects of this study differ from previous studies (2,3) and therefore require careful discussion.
First, viral clearance usually is completed within the first days after infection during the natural course of myocarditis, and mean duration of disease activity lasts between 2 and 4 weeks (4). In the present study (1), median interval between onset of symptoms and CMR was 2 weeks, indicating that approximately 25% of the patients underwent CMR after more than 7 weeks (interquartile range: 1 to 7 weeks).
Second, no information is given on what the definition of severe myocarditis was based on, and no comparison group of patients with “less severe” myocarditis is presented. In addition, disease severity is not listed as an inclusion criterion (1), also suggesting that patient inclusion in the study was retrospective and not prospective. Therefore, it remains unclear to what degree the results were influenced by disease severity and by the time interval between disease onset and date of CMR. All factors cause an inhomogeneity of the patient population, which may be responsible for the relatively low diagnostic accuracy for native T1 mapping of 69%, compared with 2 previous studies (2,3). In these studies, native T1 mapping yielded a diagnostic accuracy of 91%, respectively, and showed a superior diagnostic performance compared with single conventional CMR parameters (Lake Louise Criteria) (2,3).
Third, quantitative T2 relaxation times yielded a lower diagnostic accuracy compared with edema-sensitive black-blood T2-weighted ratio (63% vs. 70%) (1). Here, 2 more factors may have hampered the diagnostic performance of T2 mapping: 1) the accuracy of the T2 mapping sequence has not been validated with appropriate phantom studies; and 2) T2 mapping was performed with free breathing, making it more susceptible to motion artifacts compared with breath-hold sequences.
Fourth, diagnostic accuracy is reported to be significantly higher compared with classic Lake Louise Criteria when using extracellular volume (ECV) quantification in combination with late gadolinium enhancement (90% vs. 79%, p = 0.0053) (1). Increased ECV has also been reported in patients with several cardiac risk factors, for example, diabetes (5). Interestingly, the diagnostic value of ECV was less favorable in another recent study on patients with acute myocarditis with an area under the curve of only 0.71 (2) compared with 0.86 in this study. Unfortunately, no detailed information is provided on the distribution of comorbidities in the study population (1), because an unequal distribution of comorbidities between patients and controls might have falsely influenced the diagnostic performance to the advantage of ECV.
We believe that in a setting of acute myocardial injury and inflammation, quantitative CMR (using native T1 and T2 mapping) may improve diagnostic performance of CMR as reported previously (2,3). However, carefully defined patient populations with well-defined disease stages are necessary to obtain reliable results that allow for introduction of these diagnostic techniques into clinical routine.
- 2015 American College of Cardiology Foundation
- Radunski U.K.,
- Lund G.K.,
- Stehning C.,
- et al.
- Ferreira V.M.,
- Piechnik S.K.,
- Dall'Armellina E.,
- et al.