Right Ventricular Myofiber Architecture
Benoit Scherrer1, Amara Majeed2, Onur Afacan1, Jolene M. Singh1, Simon K. Warfield1, and Stephen P. Sanders2

1Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, United States, 2Departments of Cardiology and Pathology, Boston Children's Hospital, Harvard Medical School, Boston, MA, United States

Synopsis

While the left ventricle (LV) myofiber architecture has been studied extensively, there have been fewer studies of right ventricle (RV) myofiber architecture, and there is no established convention for reporting RV myofiber anatomy. Establishing the normal RV myofiber architecture is critical to facilitate our understanding of the impact of congenital microstructural defects in diseased hearts. In this work, we fixed a juvenile swine heart and imaged it with magnetic resonance diffusion compartment imaging (MR-DCI). We describe the detailed myofiber pattern of the RV with both MR-DCI and stained histological microscopy, and propose a standardized convention for describing the myofiber anatomy.

Purpose

There are few studies of right ventricular (RV) myofiber architecture and no established convention for describing and reporting RV myofiber anatomy.1 This study aims to define the detailed myofiber pattern of the right ventricle using novel MR diffusion compartment imaging protocols and analysis methods, and to propose a convention for describing the myofiber anatomy.

Methods

A juvenile swine heart was fixed (5% formalin and 50% ethanol) under mild distending pressure (~10cm H2O), immersed in Galden®, and imaged using a 3T Siemens Skyra. Diffusion-weighted MRI was acquired with a pulsed gradient spin echo sequence, single shot EPI readout and the following parameters: FOV=96x96mm2, matrix=96x96, 84 slices, 1x1x1mm3 spatial resolution, TE=109ms, TR=19s. Our gradient scheme contained 4 b=0s/mm2 images and 3 shells (b=1000, 1500, 2000s/mm2) of 30 gradients each. Eight averages were acquired with both LR and RL phase encoding direction. The total acquisition time was lower than 8h.

DW images were compensated for distortion using the RL/LR images and Topup.2 Diffusion compartment imaging (DCI)3 was computed by considering one isotropic free diffusive component and one anisotropic compartment per voxel. The diffusion data displayed in glyph format were used for qualitative description of myofibers. Helix angle was measured using the short axis of the RV as a reference plane (Figure 1). The compartment mean diffusivity (cMD) and fractional anisotropy (cFA) of the anisotropic compartment were measured and displayed as 2D sectional maps.

Blocks of ventricular myocardium were removed after imaging, sectioned parallel to the epicardial surface, photographed, and stored digitally for estimation of fiber angle. A CT scan of the heart was put in registration with the diffusion images allowing precise correlation between diffusion-derived fiber maps and histological sections.

Results

Myofiber arrangement – The RV free wall and septal wall, both convex anteriorly, form a crescent in cross section (Figure 2). At the apex the RV cones down to a small whorl and at the base it is divided by the parietal band (PB) into inflow and outflow. A single layer of free wall myofibers run circumferentially from the inferior to the superior junction with the septal wall (Figure 1, 2). A second oblique longitudinal layer is added to the endocardial side of the free wall from the moderator band to the base (Figure 2). This layer is composed of trabeculae and papillary muscles. Septal wall endocardial fibers run obliquely longitudinal while deeper fibers are circumferential (Figure 1,2). Myofibers within papillary muscles and other trabeculations run approximately parallel to the long axis of the muscle bundle and smoothly join endocardial fibers where trabeculations join the RV wall (Figure 3). Exceptionally, fibers in the parietal and moderator bands run obliquely (~45ᵒ to long axis) and moderator band fibers intersect those of the septal band at nearly right angles (Figure 2C).

RV reference system – We have used the short-axis plane of the RV as the reference for estimating the helix angle (Figure 1). Fibers in the short-axis plane are assigned a helix angle of 0ᵒ, those directed from inferior to superior toward the base (viewed from outside the RV free wall) a negative angle, and those directed from inferior to superior toward the apex a positive angle.

Diffusion characteristics – cMD is higher in the RV myocardium compared with the left ventricle (Figure 4) consistent with the higher percentage of fibrosis in the RV.

Histology – Fiber orientation by DCI closely matched histology (Figure 5).

Discussion

The RV walls are composed of an external circumferential layer (considering the mid septal wall as an external layer) and an internal oblique longitudinal layer for the most part. The portion of the RV free wall between the moderator band and the apex has only the circumferential layer. The longitudinal fibers are mostly contained in trabeculations, especially on the free wall, and run parallel to the trabeculations. This is consistent with the contraction pattern of apex-to-base shortening produced by the longitudinal fibers coupled with a bellows effect of the free wall against the septal wall produced by shortening of the circumferential fibers.

Muscle bands that cross the cavity from free wall to septum (moderator and parietal bands) have obliquely running fibers. This fiber orientation plus the right-angle junction of the parietal and septal bands likely reflect their embryological development4.

Previous reference frames for fiber orientation have been based solely on the left ventricle. We have proposed a separate reference frame for the RV because it is often the only ventricle present in congenital heart defects.

Acknowledgements

This work was supported in part by NIH grants R01 NS079788, R01 EB018988, U01 NS082320, Intel (c) IPCC, BCH CTREC K-to-R Merit Award and BCH TRP Pilot.

References

1. Nielsen E, Smerup M, Agger P, Frandsen J, Ringgard S, Pedersen M, Vestergaard P, Nyengaard JR, Andersen JB, Lunkenheimer PP, Anderson RH, Hjortdal V. Normal right ventricular three-dimensional architecture, as assessed with diffusion tensor magnetic resonance imaging, is preserved during experimentally induced right ventricular hypertrophy. Anat Rec (Hoboken) 2009;292:640-51,

2. Andersson, J.L., S. Skare, and J. Ashburner, How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. Neuroimage, 2003. 20(2): p. 870-88.

3. Scherrer, B., A. Schwartzman, M. Taquet, M. Sahin, S.P. Prabhu, and S.K. Warfield, Characterizing brain tissue by assessment of the distribution of anisotropic microstructural environments in diffusion-compartment imaging (DIAMOND). Magn Reson Med, 2015: p. to appear.

4. Sizarov A, Lamers WH, Mohun TJ, Brown NA, Anderson RH, Moorman AFM. Three-dimensional and molecular analysis of the arterial pole of the developing human heart. J Anat 2012;220:336–349.

Figures

RV reference system – Reference short-axis plane is constructed perpendicular to RV long axis between the center of the tricuspid valve and apex. Helix angle is assigned as shown. Free wall epicardial fibers are 0+/- 10ᵒ and endocardial fibers negative. Septal endocardial fibers are positive and rapidly transition to ~0.

Short-axis (A - mid ventricle, B - base) and long-axis (C) views of the RV showing the crescent shape, inflow and outflow separated by the parietal band (PB), and the moderator band (MB). Fiber orientation is indicated by the linear glyphs.

Fibers of the posterior papillary muscle of the tricuspid valve are aligned parallel to the long axis and curve smoothly into the endocardial oblique longitudinal fibers of the septal wall.

Compartment mean diffusivity (cMD) and compartment fractional anisotropy (cFA) maps. cMD is higher in the RV compared with the left ventricle reflecting greater percentage of fibrous tissue.

Comparison of myofiber map with histology. The blank spot in the RV free wall on the CT image indicates the sampling site with overlying myofiber map. Representative histology sections are shown to the right with corresponding depth into the wall indicated by the white lines.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
3117