Myocardial parametric mapping in patients with suspected acute myocarditis: a prognosis study

Abstract Background T1 mapping and T2 mapping have become essential components of the 2018 Lake Louise criteria and obtained excellent results in diagnosing myocarditis. However, their prognostic value in acute myocarditis has not yet been well recognized or established. Purpose Our study aims to investigate the prognostic value of T1 mapping, T2 mapping, and extracellular volume fraction (ECV) in a large cohort of patients with suspected acute myocarditis. Materials Patients meeting the recommended clinical criteria for suspected acute myocarditis were consecutively enrolled. The primary endpoint is the composite of cardiac death, heart failure hospitalization, heart transplantation, sustained ventricular arrhythmia, and recurrent myocarditis, defined as major adverse cardiovascular events (MACE). All patients underwent CMR at 3.0 T scanner using a standardized, routine imaging protocol, and the three short-axis view slices (basal, mid-ventricular, and apical) were divided into 16 segments according to the American Heart Association 17-segment model (apex excluded). We also averaged the native T1 values, ECV, and T2 values of all 16 segments to get global T1 and ECV fraction values, respectively. Statistically significant parameters from univariate Cox analyses were put into multivariate Cox analysis. Results A total of 235 patients (150 men; 32±13 years) were enrolled in this study. During a mean follow-up of 43.0±3.6 months, MACE occurred in 37 patients (15.7%) and was univariably associated with heart failure presentation, left ventricular ejection fraction, left ventricular end-systolic volume index, left ventricular end-diastolic volume index, native T1 and ECV. Patients with MACE showed higher global native T1, ECV, and T2 values (1348±59 vs 1266±51, p=<0.001; 40.0±8.7 vs 32.8±5.9, p<0.001; 63.4±12.1 vs. 55.5±9.1, p=0.011), and were more likely to have tissue changes in the interventricular septum and anterior wall (Figure 1). In a series of nesting multivariable Cox regression models, the addition of native T1 (per 10-ms increase: HR, 1.128; 95% CI: 1.043, 1.220; p = 0.003; Harrell’s C-index = 0.833) or ECV (per 5% increase: HR, 1.382; 95% CI: 1.060, 1.802; p = 0.017; Harrell’s C-index = 0.847) improved prognostication compared with the model based on clinical variables, left ventricular ejection fraction, and septal late gadolinium enhancement (Harrell’s C-index = 0.762) (Figure 2). T2 in multivariate Cox regression analysis didn’t show predictive value, but his inclusion increased the C-index of the model to 0.814. Conclusions Myocardial parametric mapping not only holds substantial diagnostic value but also plays a critical role in the prognostic assessment and risk prediction of myocarditis. Their utilization in these contexts enhances the accuracy and effectiveness of clinical evaluations and decision-making processes.