Practice patterns of cardiovascular magnetic resonance use in the diagnosis of pediatric myocarditis: A survey-based study

Abstract

Background: Cardiovascular magnetic resonance (CMR) is used to diagnose myocarditis in adults and children based on the original Lake Louise criteria (LLC) and more recently the revised LLC. The major change included in the revised LLC was the incorporation of parametric mapping, which significantly increases the sensitivity and specificity of diagnosis. Subsequently, scientific statements have recommended the use of parametric mapping in the diagnosis of myocarditis in children. However, there are some challenges to parametric mapping that are unique to the pediatric population. Our goal is to characterize clinical CMR and parametric mapping practice patterns for the diagnosis of myocarditis in pediatric centers.

Methods: The Cardiovascular Magnetic Resonance Evaluation in Return to Athletes for Myocarditis in Coronavirus Disease 2019 and Immunization Consortium (CERAMICi) created a Research Electronic Data Capture (REDCap) survey to evaluate clinical practice patterns for diagnosis of myocarditis in pediatrics. This survey was distributed to the Society for Cardiovascular Magnetic Resonance community.

Results: Fifty-nine responses from 51 centers were received, with only one response from each center being utilized. Only 35% (18/51) of centers (37% (14/38) North America, 31% (4/13) international) reported using CMR routinely in all patients with a suspicion of myocarditis. Diagnostic uncertainty was noted as the most important reason for CMR, while cost was noted as the least important consideration. The majority of centers reported using the revised LLC (37/51, 72%) compared to original LLC (7/51, 14%) or a hybrid criteria (6/51, 12%). When looking at the use of parametric mapping, only 5/47 (11%) for T1 mapping and 11/49 (22%) for T2 mapping reported having scanner-specific pediatric normative data.

Conclusion: Routine CMR imaging for diagnosis of myocarditis in pediatrics is infrequently performed at surveyed centers despite the focus on a group of non-invasive cardiac imagers. While the majority reported using parametric mapping, few centers reported having pediatric scanner-specific normative data. This highlights an important gap in the utilization of CMR that may aid in the diagnosis of myocardial disease.

Keywords: Lake Louise criteria; Myocarditis; Parametric mapping; Pediatric.

Graphical abstract

Fig. 1

Percentage of responses from North American and international sites. North American sites are broken down by the institution’s state of origin plus one Canadian site. International sites are broken down by the institution’s country of origin

Fig. 2

Cardiovascular magnetic resonance (CMR) volume per year broken down by location (North America vs international).

Fig. 3

Cardiovascular magnetic resonance criteria used to diagnosis myocarditis by center volume and center location. Small centers <251 scans per year, medium centers = 251–500 scans per year, large centers >500 scans per year. LLC Lake Louise criteria

Fig. 4

Reported sources for normative values for native T1 mapping, T2 mapping, and ECV fraction calculation based on center location. Ped pediatric, ECV extracellular volume fraction