Ex vivo cardiovascular magnetic resonance diffusion weighted imaging in congenital heart disease, an insight into the microstructures of tetralogy of Fallot, biventricular and univentricular systemic right ventricle

Abstract

Purpose: Common types of congenital heart disease exhibit a variety of structural and functional variations which may be accompanied by changes in the myocardial microstructure. We aimed to compare myocardial architecture from magnetic resonance diffusion tensor imaging (DTI) in preserved pathology specimens.

Materials and methods: Pathology specimens (n = 24) formalin-fixed for 40.8 ± 7.9 years comprised tetralogy of Fallot (TOF, n = 10), dextro-transposition of great arteries (D-TGA, n = 8) five with ventricular septal defect (VSD), systemic right ventricle (n = 4), situs inversus totalis (SIT, n = 1) and levo-TGA (L-TGA, n = 1). Specimens were imaged using a custom spin-echo sequence and segmented automatically according to tissue volume fraction. In each specimen T1, T2, fractional anisotropy, mean diffusivity, helix angle (HA) and sheet angle (E2A) were quantified. Pathologies were compared according to their HA gradient, HA asymmetry and E2A mean value in each myocardial segment (anterior, posterior, septal and lateral walls).

Results: TOF and D-TGA with VSD had decreased helix angle gradient by - 0.34°/% and remained symmetric in the septum in comparison to D-TGA without VSD. Helix angle range was decreased by 45°. It was associated with a decreased HA gradient in the right ventricular (RV) wall, i.e. predominant circumferential myocytes. The sheet angle in the septum of TOF was opposing those of the left ventricular (LV) free wall. Univentricular systemic RV had the lowest HA gradient (- 0.43°/%) and the highest HA asymmetry (75%). HA in SIT was linear, asymmetric, and reversed with a sign change at about 70% of the depth at mid-ventricle. In L-TGA with VSD, HA was asymmetric (90%) and its gradients were decreased in the septum, anterior and lateral wall.

Conclusion: The organization of the myocytes as determined by DTI differs between TOF, D-TGA, L-TGA, systemic RV and SIT specimens. These differences in cardiac structure may further enlighten our understanding of cardiac function in these diverse congenital heart diseases.

Keywords: Congenital heart disease; Dextro; Diffusion tensor imaging; Ex vivo; Levo; Microstructure; Situs inversus; Systemic right ventricle; Tetralogy of Fallot; Transposition of the great arteries; Ventricular septal defect.

Fig. 1

Helix angle (HA in red) and sheet angle (E2A in yellow) calculated from the projection of the eigenvectors of the diffusion tensor onto the circumferential and radial planes, respectively

Fig. 2

A specimen with Levo transposition of the great arteries (L-TGA). a spin echo diffusion weighted image at b = 0 s/mm²; b myocardial segment (Anterior, Septum, Posterior, Lateral walls and RV wall); c helix angle; d sheet angle. LV, left ventricle, RV, right ventricle.

Fig. 3

a Fitted line showing helix angle (HA) gradient calculated between 20 and 80% depth in biventricular dextro-tranposition of the great arteries (D-TGA) with and without ventricular septal defect (VSD). b Boxplot for D-TGA without VSD (n = 3). c Boxplot for D-TGA with VSD (n = 5). Middle line corresponds to the mean HA while outer dash lines are the 95% confident interval of the mean HA. a (b, n = 3) and with (c, n = 5). The Boxplot represents all HA values at each percentage of depth with the HA median as the red line, the first quartiles are the boxes and the dashed lines are the whiskers. The maximum whisker length is the interquartile range. HA values beyond the whiskers are displayed using “+”

Fig. 4

Helix angle gradient in tetralogy of Fallot (TOF), dextro-transposition of the great arteries (D-TGA) with and without ventricular septal defect (VSD). Significant differences between HA gradient (p < 0.01) are highlighted in green

Fig. 5

HA asymmetry (intercept of HA) with the 95% confidence interval ([lower limit: upper limit]) in specimen of tetralogy of Fallot and biventricular transposition of the great arteries with or without ventricular septal defect (VSD). The p value measures the significant difference between two HA intercepts from two congenital heart disease groups in each segment

Fig. 6

Sheet angle (E2A) in tetralogy of Fallot (TOF, n = 10), dextro-transposition of the great arteries (D-TGA) with and without ventricular septal defect (VSD). Significant differences (p < 0.01) are highlighted in green