Background: The detection of ongoing inflammation in myocarditis patients has prognostic relevance and might be adequately evaluated by cardiac MRI with its ability to noninvasively characterize myocardial tissue. To date, there are only limited data on the detection of chronic myocarditis and its differentiation from healed myocarditis based on a cardiopathological reference standard.

Purpose: To assess the performance of cardiac MRI for 1) the detection of chronic myocarditis and 2) the discrimination of biopsy-proven chronic myocarditis from healed myocarditis.

Materials and Methods: A total of 80 consecutive participants with persistent (>30 days) symptoms suggestive of myocarditis were prospectively enrolled from a single-center (university hospital) between January 2020 and August 2022. All participants underwent both endomyocardial biopsy for reference standard diagnosis and a multiparametric 1.5 T cardiac MRI protocol including mapping and late gadolinium enhancement (LGE). Mann‒Whitney U test, Fisher exact test and receiver operating characteristic (ROC) curves were used for group comparison and evaluation of diagnostic performance.

Results: Endomyocardial biopsy (performed within a median of 1 day [IQR, 1-7] of cardiac MRI) revealed 40 of 80 participants (50%) with chronic lymphocytic myocarditis and 40 of 80 participants (50%) with healed myocarditis (median age, 48 years [IQR, 39-59]; 55 men, 25 women), median symptom duration 3.5 months. At the time of diagnostic workup, the most common symptom was dyspnea: 27 of 40 participants (68%) with chronic myocarditis, and 18 of 40 participants (45%) with healed myocarditis, P =0.07. Specifically, 9 of 40 participants (23%) with chronic myocarditis suffered from dyspnea at rest (NYHA IV) vs. none of the healed myocarditis group. Viral genomes and elevated C-reactive protein as inflammatory marker were exclusively detected in participants with chronic myocarditis. Troponin was elevated in 14 of 40 participants (35%) in the chronic group vs. in 3 of 40 participants (8%) in the healed group, P =0.005. NT-proBNP was elevated in 22 of 40 participants (55%) in the chronic group vs. in 6 of 40 participants (15%) in the healed group, P <.001. LV-EF was lower in chronic myocarditis than in healed myocarditis patients (38% vs. 50%), p=0.01. T₂ mapping as a single parameter 1) showed the best sensitivity in detecting chronic myocarditis (39 of 40, 98%; 95% CI: 87, 100) and 2) provided the best discrimination from healed myocarditis as defined by the area under the ROC curve (AUC, 0.86; 95% CI: 0.77, 0.93; P <.001), followed by LGE (AUC, 0.63), extracellular volume fraction (AUC, 0.62), and T₁ mapping (AUC, 0.60).

Conclusion: A multiparametric cardiac MRI protocol including T₂ mapping allows detection of ongoing myocardial inflammation and discrimination of chronic myocarditis from healed myocarditis in participants with biopsy-proven myocarditis and symptoms persisting for >30 days.