Perfusion territories subtended by penetrating coronary arteries increase in size and decrease in number toward the subendocardium

Abstract

Blood flow distribution within the myocardium and the location and extent of areas at risk in case of coronary artery disease are dependent on the distribution and morphology of intramural vascular crowns. Knowledge of the intramural vasculature is essential in novel multiscale and multiphysics modeling of the heart. For this study, eight canine hearts were analyzed with an imaging cryomicrotome, developed to acquire high-resolution spatial data on three-dimensional vascular structures. The obtained vasculature was skeletonized, and for each penetrating artery starting from the epicardium, the dependent vascular crown was defined. Three-dimensional Voronoi tessellation was applied with the end points of the terminal segments as center points. The centroid of end points in each branch allowed classification of the corresponding perfusion territories in subendocardial, midmyocardial, and subepicardial. Subendocardial regions have relatively few territories of about 0.5 ml in volume having their own penetrating artery at the epicardium, whereas the subepicardium is perfused by a multitude of small perfusion territories, in the order of 0.01 ml. Vascular volume density of small arteries up till 400 μm was 3.2% at the subendocardium territories but only 0.8% in the subepicardium territories. Their higher volume density corresponds to compensation for flow impeding forces by cardiac contraction. These density differences result in different scaling law properties of vascular volume and tissue mass per territory type. This novel three-dimensional quantitative analysis may form the basis for patient-specific computational models on coronary perfusion and aid the interpretation of image-based clinical methods for assessing the transmural perfusion distribution.

Influence of segmented vessel size due to limited imaging resolution on coronary hyperemic flow prediction from arterial crown volume

“…8). For stems of transmural penetrating arteries, perfused layers were assigned by their myocardial penetration depth (23). Subendocardial trees penetrated transmurally to the endocardium, and subepicardial trees reached no further than one-third into the myocardium.…”

“…Although the results are only from a single heart, a clear differential effect of truncation in these two layers can be observed, with a more pronounced effect on subepicardial branches. This difference is probably related to the higher vascular density at the subendocardium (23). Therefore, the choice of ␤ for regional flow estimation may also have to include the transmural location of the vascular tree.…”

“…Five healthy canine hearts were obtained from a series of physiological experiments carried out according to a protocol that was approved by the Institutional Animal Care and Use Committee of the University of Utrecht Medical Center. Details on experimental procedures and animal handling are described elsewhere (22,23). After premedication (medetomidine 0.03 mg/kg im, ketamine 0.03 mg/kg im, and atropine 0.5 mg im), the animals were anesthetized (sufentanil 10 g·kg Ϫ1 ·h Ϫ1 iv), intubated, and mechanically ventilated with supplemental oxygen.…”

Functional and anatomical measures for outflow boundary conditions in atherosclerotic coronary bifurcations

“…Next, the clinical validation of the used volume segmentation tool is currently limited to reliably segment myocardial area with low flow compared to single-photon emission computed tomography data. Nonetheless, the Voronoi algorithm on which it is based has proven to work for the myocardium (Termeer et al, 2010), but also for arterial networks (Karch et al, 2003) and smaller vessel in the subendocardium (van Horssen et al, 2014), making it likely that representative volume segmentations were obtained in this study as well.…”

Selective subepicardial localization of monocyte subsets in response to progressive coronary artery constriction

“…Sequential episcopic fluorescence image capture yielded multispectral 3D virtual reconstructions of the hearts with an in-plane resolution of 14 to 28 m. Isometric voxel size was achieved by average intensity projection of every two adjacent images in the stack. Processing of fluorescent vascular cast and cell images, including correction for spectral overlap of fluorescent cell tracker probes, was conducted as described previously (11,30).…”