Cardiac Magnetic Resonance in Suspected Myocarditis

Study Questions:

How does cardiac magnetic resonance (CMR) compare with endomyocardial biopsy (EMB) in suspected myocarditis?


In the Myo-Racer (Magnetic Resonance Imaging in Myocarditis) study, patients with suspected myocarditis simultaneously underwent EMB, cardiac catheterization (to exclude coronary artery disease), and CMR including T1 and T2 mapping (on 1.5-T and 3-T scanners). The CMR protocol utilized standard Lake Louise criteria (LLC) for myocarditis and calculation of extracellular volume fraction (ECV). The investigators divided the patients into two groups: acute ≤14 days, and chronic >14 days. The investigators performed receiver-operating curve analysis to calculate optimal thresholds and areas-under-the-curves (AUCs). The AUCs were compared with the DeLong method, and Youden index was used to depict optimal cutoff values.


A total of 129 patients (mean age 44 years) of the study cohort underwent 1.5-T imaging. Of these patients, 61 presented with acute and 68 with chronic symptoms. The investigators found that in patients with acute symptoms, native T1 was the best diagnostic modality as defined by the AUCs (0.82; p = 0.002 when compared to LLC), followed by T2 (0.81, p = 0.001), ECV (0.75; p = 0.04), and LLC (0.56). Native T mapping yielded the highest diagnostic accuracy of 815, followed by T2 mapping (80%), ECV (75%), and LLC (59%). However, in those with chronic patients, only T2 mapping yielded an acceptable AUC (0.77; p = 0.002 when compared to LLC). T2 mapping achieved the highest diagnostic accuracy of 73%, followed by ECV 67%, LLC 59%, and native T1 mapping of 45%. When the investigators utilized 3.0-T scanners, they found that AUCs of native T1, ECV, and LLC were comparable to 1.5-T with no significant differences—the AUC was significantly higher for native T1 (p = 0.004) and ECV (p = 0.002) as compared to LLC. They found no significant differences between T1 and ECV (p = 0.78).


The study authors concluded that T1 mapping is superior to LLC in those with acute symptoms, whereas T2 mapping is adequate for those with chronic symptoms.


The main strength of CMR, especially when using the LLC, is its high specificity and positive predictive value, whereas the sensitivity of CMR to detect myocarditis is variable and depends on the time point and the protocol of the scan. This is an important study because the findings suggest that the sensitivity can be improved by the choice of CMR technique.

< Back to Listings