Cardiac MRI in Suspected Acute COVID-19 Myocarditis

Keywords: COVID-19; coronavirus; myocarditis; cardiac MRI; T1 mapping; T2 mapping

Although primarily affecting the respiratory system, COVID-19 associated myocardial injury is common and can occur directly due to myocardial viral infection or indirectly due to systemic inflammation, endothelial activation, and/or microvascular thrombosis (1,2). Beside myocardial infarction, myocardial injury can also be a result of myocardial inflammation (2). However, current information about associated myocardial inflammation is mainly limited to case reports or series (3). As cardiac MRI is important for the diagnostic work up of patients with myocarditis, more data about MRI characteristics of COVID-19 associated acute inflammatory injury is needed. The aim of this study was to describe cardiac MRI findings in participants with active COVID-19 infection and suspected acute myocarditis.

Material and Methods
The institutional ethics commission approved this prospective study. All participants gave written informed consent. Participants with COVID-19 without structural heart disease and mechanical ventilatory support were included consecutively during the recruitment period from April 2020 to December 2020.
Participants had a clinical suspicion for COVID-19 associated myocarditis with signs of acute myocardial injury (elevated troponin levels with or without electrocardiographic changes). Acute coronary syndromes were excluded by cardiac catheterization. The control groups consisted of healthy volunteers and participants with suspected acute non-COVID-19 myocarditis (4).
Cardiac MRI was performed at 1.5 Tesla in all participants using previously described acquisition parameters (4). Left ventricular function, average systolic longitudinal strain, T2 signal intensity ratio, T1 relaxation times, T2 relaxation times, extracellular volume, and quantitative late gadolinium enhancement (LGE) were I n p r e s s determined. Focal myocardial edema and LGE were visually assessed. Measurements of the blinded readers (J.A.L. and A.I., with 8 and 4 years of experience, respectively, in cardiac MRI) were conducted as described previously (5).
Continuous variables between two groups were compared with Student's t test.
χ 2 test was used to compare dichotomous variables. One-way analysis of variance followed by Tukey multiple comparison tests was performed to compare variables in the three participant groups. Statistical significance level was set to P < .05.