Quick Fire Session
SCMR 22nd Annual Scientific Sessions
Helix angle (HA) measurements in the left ventricle (LV) first reported by Streeter et al.  have become a widely-accepted standard for describing cardiac fiber architecture and are used to generate rule-based mathematical fiber geometries in computational models of the heart. Streeter’s HA results were limited to the LV, where an inverse-sine distribution from the endocardium to the epicardium was evident. RV measurements were not reported. The objective of this study was to describe regional variation in LV and RV HA using diffusion tensor (DT) MRI.
Fresh normal swine hearts (N=14) were arrested and underwent ex vivo DT MRI with spatial resolution 1x1x1 mm3. HAs were calculated as the angle between the primary eigenvector direction and the short-axis plane of the heart (Fig.1). The transmural coordinate of each point in the myocardium was found by solving Laplace’s equation with boundary conditions of -1 at the endocardium and +1 at the epicardium. The LV was divided according to the AHA 16-segment model, and two-parameter linear, tan-1, and sin-1 curve fits were performed for each segment. Wilcoxon rank rum test and four-way Kruskal-Wallis non-parametric ANOVA were used for two-group and multi-group comparisons, respectively.
HA pitch (range) (Fig.2 A1-2): The RV has significantly smaller transmural HA pitch than the LV (-28°±13° vs. -54°±14°, p<0.001). In both ventricles, there are high variations across regions with low variations within regions. There exist significant differences among the four regions of the LV (p = 0.003), with the highest HA pitch seen in the posterior free wall, compared to the anterior and lateral free walls and the septum (p = 0.02, <0.001, =0.008). In the RV, the basal anterior free wall exhibited lower HA pitches than did either the posterior free wall or the apical anterior free wall (p=0.03).
Mid-wall HA (Fig.2 B1-2): We did not analyze the mid-wall HAs for the RV because the very thin wall makes the concept of "mid-wall" problematic. There exist significant differences among the four regions of LV (p < 0.001). The posterior and anterior free walls had significantly higher mid-wall HAs than did the septum (p < 0.001) and borderline higher than did the lateral free wall (p = 0.02).
The sin-1 distribution observed by Streeter is predominant in the free wall, while the septum has a predominantly tan-1 distribution (Fig.3).
These findings are helpful in establishing the normal differences between RV and LV and defining the form of transmural HA change. They can be used to compare to pathological conditions, and potentially improve mathematical models by providing more evidence-based microstructural inputs. We acknowledge the support from the Sarnoff Foundation, NIH K25 HL135408 and UCLA Radiological Sciences.