Vasa vasora are the blood supply of the artery itself, originating in the adventitia in response to the arterial wall’s nutritional requirements. The presence of a thicker tunica media or neointima requires the development of vasa vasorum to meet the metabolic demands of the arterial wall, since diffusion from the arterial lumen alone is insufficient. Over 40 years ago Wolinsky and Glagov demonstrated that arteries with a medial thickness exceeding 0.47mm required vasa vasorum for nutrition.1 Vasa vasora can be further categorized into first-order vasa, which run longitudinal to the vessel lumen, and second-order vasa, which originate from the first-order vasa and are circumferential around the artery.2
The vasa vasora are necessary to maintain normal vessel wall homeostasis and inadequate perfusion of the vessel wall as deficiencies in the vasa have been shown to lead to intimal hyperplasia.3 The initial hypothesis relating atherosclerosis to increased development of the vasa vasorum was made by Barger et al. in 1984.4 Since then, it has been shown that coronary vasa neovascularization takes place early after induction of experimental hypercholesterolemia, suggesting a role for neovascularization in atherogenesis.5 This neovascularization has been shown to favor second-order vasa.2 Apolipoprotein E-deficient mice given the angiogenic factor vascular endothelial growth factor (VEGF), which leads to increased angiogenesis, show a subsequent increase in plaque area,6 while mice administered the anti-angiogenic factors endostatin and TNP-470 show decreased intimal hyperplasia.7 A decrease in intimal hyperplasia was also observed in a murine study following chronic endothelin receptor antagonism, which decreases VEGF expression and decreases vasa neovascularization.8 These studies provide evidence that neovascularization of the arterial wall is a crucial part of the atherosclerotic process. Abnormalities of the vasa vasorum have also been implicated in the development of neointimal hyperplasia after balloon angioplasty and stenting. In two animal models, local injury to the vascular wall stimulated intimal hyperplasia and adventitial neovascularization that was increased by VEGF and PR39 and tempered by the inhibition of VEGF and fibroblast growth factor, leading Khurana et al. to hypothesize that intimal hyperplasia has both angiogenesis-dependent and -independent phases.9 Indeed, it has previously been shown that following angioplasty injury, the number and density of adventitial microvessels increase in the initial three post-procedural days, then regress.10 Kwon et al. evaluated the spatial pattern of neovascularization, showing that although the number and diameter of the vasa increased after injury, the total vascular area was lower in injured vessels than in control vessels.11
Deployment of an intravascular stent leads to arterial wall compression and increased resistance within the vasa vasorum, resulting in vascular wall ischemia and subsequent neointimal proliferation.12 Further evidence for the importance of neovascularization on neointimal proliferation were studies showing that a tyrosine kinase inhibitor inhibited both neovascularization and neointimal proliferation after coronary stenting, and that the neointimal proliferation was proportional to the number of adventitial microvessels present.13 With the established importance of the coronary vasa vasorum on neointimal proliferation in atherosclerosis, angioplasty injury, and arterial stenting, accurate quantification of vasa vasora number and volume is a fruitful area of research. The traditional method of vasa vasorum quantification uses histology, but this approach requires staining for vasa vasora endothelial cells and suffers from difficulties such as cutting through metallic stents, inaccuracies due to unperfused vasa, and incorrect data due to a limited number of measured cross-sections. An in vivo human method is not plausible, as the coronary microcirculation begins at arterioles of 500μm in diameter and progressively branches into capillaries 5μm in diameter; these blood vessels are too small to visualize using currently available methods.12 Three-dimensional microscopic computed tomography (micro-CT) has emerged as an accurate and accessible method for quantification of the coronary vasa vasorum.
Microscopic Computed Tomography
Micro-CT was developed by Flannery et al. in 1987 as a high-resolution imaging technique with a resolution of 5–30μm per voxel, depending on the X-ray source and the number of two-dimensional images obtained.12,14–15 The technique provides high-resolution three-dimensional images that can be segmentally viewed and analyzed in any plane. Importantly, micro-CT allows for imaging of the entire microvasculature and its relationship to other sections of the growing atherosclerotic or intimal lesion. Quantitatively, the results obtained with micro-CT correlate with histological measures of vasa vasorum quantitation.11,16 It has been postulated that combining micro-CT measurements of the arterial wall with local pressure measurements may allow investigators to accurately estimate spatial patterns of compressive stresses, leading to the development of novel stents that have fewer adverse effects on the vasa vasorum17 and specialized stents for bifurcation or ostial lesions.
We have used three-dimensional micro-CT to quantify the density of vasa vasorum in swine coronary arteries treated with drug-eluting stents. Using micro-CT, the density and number of vasa vasorum were compared in stented and unstented sections of the artery. There were, however, some problems with using micro-CT in the imaging of stented arteries. Owing to the metallic nature of the stents, there is image artifact that has similar density to the contrast material used to fill arterial beds. For this reason, it is necessary to use region-growing, a computerized technique that requires manual selection of vasa vasora and maps regions with similar densities. Vasa vasora are thus assigned a density value greater than that of the stent artifact. Using MicroView software, the number of pixels with this assigned high-density value are quantified and compared with the number of pixels without the high-density value to give a density measurement of the vasa vasorum. This allows for the exclusion of artifact from density calculations. With experience, this technique allows for accurate and efficient quantification of the vasa.
Analysis of static measurements such as quantitation of the vasa are optimal for conventional micro-CT, but some physiological processes are too transient to be analyzed by this modality as the production of an image with an acceptable signal/noise ratio may take several hours. Cryostatic micro-CT, a technique that utilizes ‘snap-freezing’ of specimens before the scanning process, has evolved to remedy this situation.18,19 Using this technique, the time-course of solute transfer from the arterial lumen and adventitial vasa vasorum was elucidated, based on arteries snap-frozen at small intervals of time.19 Micro-CT is also a highly accurate form of imaging in the evaluation of intravascular stent position. Owing to the three-dimensional nature of the imaging and the high resolution, it is possible to accurately determine the position of the stent against the arterial wall in intact arteries, avoiding the inherent imprecision and cutting artifact of cross-sectional analysis by histology. Additionally, the high resolution of the imaging allows for analysis of neointimal proliferation (see Figure 1). Using analysis software, it is possible to accurately quantify the amount of neointima in three dimensions in intact stented arteries. This accurate quantification of neointimal hyperplasia has broad applications in experimental studies of novel stents. In addition to the use of micro-CT for the analysis of the arterial vasa vasorum and stent geometry, the three-dimensional nature of the images generated has found use in analysis of other aspects of arterial anatomy and pathophysiology. Micro-CT has been used to quantify and analyze three-dimensional calcification patterns in coronary arteries with atherosclerotic plaques.20 Micro-CT studies are likely to become increasingly important to visualize spatial patterns in the media and adventitia of diseased arteries that have been subjected to various medications and stents to identify devices or agents that might decrease development of atherosclerosis and in-stent restenosis.