Abstract
Background: Differentiating between immune checkpoint inhibitor (ICI) myocarditis and non-ICI myocarditis is clinically important. Cardiovascular magnetic resonance (CMR) is a well-established method for diagnosing acute myocarditis. The value of CMR for distinguishing ICI myocarditis from non-ICI myocarditis remains unclear, which this study sought to determine. Methods: A total of 54 patients (n = 26 ICI myocarditis; n = 28 non-ICI myocarditis) underwent clinical CMR for the assessment of cardiac function (cines), myocardial fibrosis (native T1-mapping, extracellular volume [ECV] fraction, late gadolinium enhancement [LGE]) and myocardial oedema (native T2-mapping). Results: ICI myocarditis patients were older than non-ICI myocarditis patients (75 years [71-78] vs. 39 years [30-64]; p < 0.001). Both groups had similar left ventricular (LV) ejection fraction (58 ± 11% vs. 58 ± 6%; p = 0.970). ICI myocarditis and non-ICI myocarditis patients also had similar native myocardial T1 values (1041 ± 84 ms vs. 1063 ± 60 ms; p = 0.281), native myocardial T2 values (59 ± 6 ms vs. 59 ± 6 ms; p = 0.943) and ECV (0.32 ± 0.07 vs. 0.31 ± 0.04; p = 0.403). Native myocardial T1 values (Rho = -0.553) and ECV (Rho = -0.502) were significantly associated with LVEF in non-ICI myocarditis patients (both p < 0.05). There was no significant association between myocardial T1 values, T2 values or ECV, with LVEF, in ICI myocarditis patients (all p < 0.05). Non-ICI myocarditis patients had a greater frequency of LGE in the LV compared to ICI myocarditis patients (89% vs. 52% p = 0.005). However, the pattern of LGE was similar between the two patient groups (mostly subepicardial and/or mid-wall). Conclusions: In this single centre retrospective cohort, the findings suggest that quantitative parametric mapping methods by CMR may not differentiate between ICI vs. non-ICI myocarditis. Further work is needed to assess the value of CMR for diagnosing standalone ICI myocarditis.