Mapping versus source methods for quantifying myocardial T1 in controls and in repaired tetralogy of Fallot: interchangeability and reproducibility in children

Background: Myocardial T1 relaxometry can be performed by contouring on individual T1-weighted source images (source method) or on a single T1 map (mapping method). Objective: This study compares (a) agreement between native T1 and extracellular volume results of the two methods and (b) interobserver reproducibility of the two methods in children without heart disease and those with tetralogy of Fallot (TOF). Materials and methods: We retrospectively analyzed pediatric patients (controls and those with repaired TOF) with cardiac magnetic resonance examinations including extracellular volume quantification using the modified Look-Locker inversion recovery (MOLLI) sequence. We compared native T1 and extracellular volume of the entire left ventricle and interventricular septum derived using the source and the mapping approaches. Results: In the control group (n=25, median age 14.0 years, interquartile range [IQR] 11.5–16.5 years), the mapping method produced lower native T1 values than the source method in the interventricular septum (mean difference ± standard deviation [SD] = 12±15 ms, P<0.001). In the TOF group (n=50, median age 13.3 years, IQR 9.9–15.0 years), the mapping method produced lower values for native T1 and extracellular volume in the interventricular septum (mean difference 9±14 ms and 0.6±1.1%, P<0.001). In 6–12% of the children, differences were >3 standard deviations from the mean difference. Interobserver reproducibility between the two methods by intraclass correlation coefficients were clinically equivalent. Conclusion: T1 and extracellular volume values generated by the source and mapping methods show systematic differences and can vary significantly in an individual child, and thus cannot be used interchangeably in clinical practice. The source method might allow for easier detection and, in some cases, mitigation of artifacts that are not infrequent in children and can be difficult to appreciate on the T1 map.