Abstract
Background
The Lake Louise Criteria for the diagnosis of myocarditis by cardiac MRI (CMR) was published in 2009 (JACC April 2009; 53(17): 1475–1487) utilizing T2 STIR, “early” gadolinium enhancement (EGE) and late gadolinium enhancement (LGE). In 2012, our CMR unit adopted a chest pain protocol to enhance detection of inflammatory heart disease, using all 3 pulse sequences. In 2017, T1 relaxing mapping and T2 relaxation mapping were added to the chest pain protocol.
Purpose
This study was designed to assess the impact of pulse sequence changes in the CMR chest pain protocol on the detection rate of myocarditis.
Methods
An institutional cardiac imaging database was queried for all patients with chest pain who underwent a CMR study from 2009 thru 2018. The number of newly diagnosed cases of myocarditis per year was computed, based on the prevailing CMR diagnostic criteria at that time. The detection rate of new myocarditis cases per year, was calculated by dividing the number of new cases by the total of CMR studies performed for evaluation of chest pain, during each calendar year.
Results
Of the 4,946 patients in the cardiac imaging database, 2,126 patients underwent CMR imaging for complaints of chest pain. Detection rates of myocarditis by year are shown in Figure 1. Prior to inclusion of T1 STIR pulse sequences, CMR detection of myocarditis was nearly nil. Use of T2 STIR imaging allowed for an increased rate of myocarditis detection. However, addition of T1 mapping and T2 mapping in 2017 resulted in another increase in myocarditis detection rates from baseline.
Figure 1. Detection rate of myocarditis.
Conclusions
Detection rates of myocarditis using CMR were improved by using T2 STIR pulse sequences, per the initial Lake Louise Criteria. Further enhancement of myocarditis detection rates is achieved by addition of T1 and T2 mapping. These CMR impact data provide additional support for the adoption of the recently modified Lake Louise Criteria (JACC Dec 2018; 72(24): 3158–3176).
TOPPE: A framework for rapid prototyping of MR pulse sequences
