Cardiac Magnetic Resonance Feature Tracking Biventricular Two-Dimensional and Three-Dimensional Strains to Evaluate Ventricular Function in Children After Repaired Tetralogy of Fallot as Compared with Healthy Children

Cardiac magnetic resonance imaging is an important tool to evaluate cardiac anatomy and ventricular size and function after repaired tetralogy of Fallot. Magnetic resonance tissue tagging is the gold standard for evaluation of myocardial strain. However, myocardial tagging strain requires tagged images to be obtained prospectively, during the scan and with limited temporal resolution. Cardiac magnetic resonance feature tracking is a new tool that allows the retrospective analysis of cine images. There is limited experience with cardiac magnetic resonance feature tracking strain analysis in children. The medical records of patients with repaired tetralogy of Fallot that had a cardiac magnetic resonance (CMR) study from December 2013 to June 2015 were reviewed. The control group included patients who underwent a CMR with normal cardiac anatomy and ventricular function. Global longitudinal, circumferential and radial strain parameters (2D and 3D) were obtained by retrospectively contouring cine images from ventricular short axis, two chamber and four chamber views using post-processing software (Circle CVi(42), Calgary, Canada). The correlation between conventional ventricular function parameters and ventricular strain was performed using Pearson's correlation. The mean age of tetralogy of Fallot and control subjects was 12.4 and 14.1 years, respectively. In patients after repaired tetralogy of Fallot, the mean left ventricular global 2D and 3D circumferential strains were -17.4 +/- 2.9 and -10.1 +/- 3, respectively. The mean indexed right ventricular end-diastolic volume was 135.4 cc m(2) +/- 46 compared to 75.7 cc m(2) +/- 17 in control subjects (P = 0.0001, CI 95%). Left ventricular global circumferential 3D strain showed a statistically significant difference in patients after TOF repair compared to normal subjects (-10.1 +/- 3 vs. -14.71 +/- 1.9, P = 0.00001). A strong correlation between left ventricular global circumferential 3D strain and right ventricular end-diastolic volume (P <= 0.0001) was noted. We found a strong correlation between left ventricular circumferential 3D strain and indexed right ventricular end-diastolic volume, as well as a strong correlation between left ventricular longitudinal 2D strain and right ventricular ejection fraction. Circumferential 3D strain may be a suitable tool to detect early abnormalities of ventricular myocardium even before the ejection fraction becomes compromised. Large-scale prospective studies are recommended.