Hemodynamics and cerebrovascular disease

1. Normal vascular remodeling 

Although there are many factors that regulate the vascular remodeling, the endothelial cell and its interaction with the fluid-induced wall shear stress play a key role in the adaptive response of vessels. The fluid-induced wall shear stress is a very small frictional force induced by a viscous fluid flow moving on the surface of solid materials. The endothelial cell is capable of sensing the level and the direction of the wall shear stress induced by the blood flow (Fig. 1A, B), and the biomechanical properties of endothelial cell are partially regulated by the wall shear stress [10], [14]. There may be numerous explanations for the physiologic meaning of the endothelial capability of sensing wall shear stress. The author's understanding is that our blood vessels indirectly monitor the blood flow velocity and the blood flow direction by their endothelial cells sensing the wall shear stress.

View Large Image Download to PowerPoint

Fig. 1. The cultured endothelial cells were elongated and aligned into the flow direction after the unidirectional shear stress (2 Pa) exposure, indicating that the endothelial cells are capable of sensing the flow direction. (A) Before flow exposure. (B) After flow exposure for 48 hours. Scale bar = 100 μm. Arrow indicates flow direction.

The fluid-induced wall shear stress (τ) along a vessel wall was calculated from the following equation: τ = μ dv/dx, where μ is the kinematic viscosity of the blood, dv is the blood flow velocity, and dx is the distance between the arterial wall and the point where the blood flow velocity was sampled (Fig. 2) [10], [52], [53]. The blood flow velocity profile in the human artery tends to be laminar and parabolic because of the friction (wall shear stress) between the viscous blood and the inner surface of blood vessel [31]. Therefore, the faster the blood velocity is, the greater the velocity gradient (dv/dx) becomes, and then the greater the value of wall shear stress becomes.

View Large Image Download to PowerPoint

Fig. 2. A schematic cross-sectional view of vessel and laminar parabolic flow. τ indicates wall shear stress; μ, kinematic viscosity; dv, blood flow velocity; dx, the distance from the inner surface of arterial wall. Wall shear stress (τ) becomes greater as flow velocity (dv) becomes faster.

For example, when the demand of blood in a distal tissue increases, the small arteries in the tissue dilate and the peripheral resistance decreases. Then the blood flow velocity going through the proximal vessel increases and the blood flow volume supplying the distal tissue also increases. The resultant elevation of the blood flow velocity induces higher wall shear stress than the original level, which elicits the endothelium-dependent vasodilatation in the proximal vessel. This increase in vessel diameter counteracts the effects of increased flow velocity and reduces the level of wall shear stress while maintaining the increased level of blood flow volume. The reverse is also true when the blood demand in a distal tissue is decrease.

In the short term, vessels either dilate or constrict to accommodate changes in blood flow velocity, usually owing to the local release of vasoactive peptides such as NO. However, in the case of a persistent increase of blood flow, a different process occurs involving the adaptive remodeling of the vessel wall with concomitant reorganization of cellular and extracellular components. An animal study investigated the remodeling of the dog carotid artery in response to increased flow produced by carotid to jugular anastomoses [24]. The flow overload elicited by the arteriovenous anastomoses induced arterial enlargement to the point where wall shear stress returned to the original physiologic baseline values [24]. Chronic increase in blood flow in dog femoral arteries resulting from an AVF enhances the tonic and receptor-stimulated production of NO [36]. Microscopic and ultrastructural studies of an arterial wall that has been exposed to increased blood flow have shown extensive tears and fragmentation of the internal elastic lamina of arterial wall [15], [34]. The fenestrations and elastic fiber degeneration in the internal elastic lamina after increased blood flow would tend to augment arterial distensibility, leading to an increase in arterial diameter. Matrix metalloproteinases (MMPs) are likely instigators of the internal elastic lamina degradation in the arterial wall [45]. In general, MMP family are also involved in the degeneration of extracellular matrix proteins. The controlled and balanced degeneration and synthesis of vascular extracellular matrix allows vessels to adapt their diameter in a very wide range for sustained increase or decrease of the blood flow.

2. Hemodynamics, vascular remodeling, and atherosclerosis 

Atherosclerosis is a systemic proliferative and inflammatory vascular disease, which predominantly involves the endothelial layer and vascular smooth muscle layer. Although the etiology of atherosclerosis involves a complex multifactorial process, hemodynamics, particularly flow-induced shear stress, has emerged as an essential feature of atherogenesis [31]. The inability of vessels to remodel appropriately (pathologic remodeling or vascular failure) is associated with the etiology of atherosclerosis. The incidence of atherosclerotic plaque is related to the vessel geometry because it determines the local hemodynamic condition. Numerous studies have been performed during the past 3 decades, correlating hemodynamic factors with the development of atherosclerotic plaques [26], [32]. The endothelial dysfunction induced by a certain hemodynamic condition is an elemental feature in the formation of atherosclerotic plaque.

Maintenance of laminar shear stress within the physiologic range is essential for normal vascular functioning, which includes the regulation of vascular caliber and tone, inhibition of cellular proliferation, endothelial permeability, thrombosis, and inflammation in the vessel wall [28], [32], [33], [49]. Mediators or factors regulated by wall shear stress include NO, angiotensin-converting enzyme, prostacyclin, endothelin-1, platelet-derived growth factor B, basic fibroblast growth factor, epidermal growth factor, thrombomodulin, tissue factor, tissue plasminogen activator, MMP, monocyte chemoattractant protein, vascular cell and intercellular adhesion molecules, and so on. There are also many other factors regulated by wall shear stress; nevertheless, NO is the core mediator of the vessel behavior.

Physiologic laminar flow condition inducing an adequate level of wall shear stress is the most important stimulus for the continuous production of endothelial NO. Attenuation of NO is one of the earliest changes preceding endothelial dysfunction. There is a strong correlation between endothelial dysfunction and areas of low and/or oscillatory shear stress because of slow and turbulent flow condition [3]. Low or oscillating shear stress has been shown to reduce endothelial production of vasodilating mediators such as NO and prostaglandin I2. Both of these also inhibit platelet aggregation and attenuate smooth muscle proliferation [2], [12], [31]. Manifestation of the endothelial dysfunction is frequently observed in the carotid bulb because of its out-pouching shape that causes blood flow separation and turbulent flow [13], [31], [57]. The endothelial dysfunction influences the site selectivity of atherosclerotic plaque formation and vessel wall remodeling, which can also affect plaque vulnerability [8].

The magnitude and direction of wall shear stress influence inflammatory processes in the vessels. Although the interaction of wall shear stress and inflammation involves a complicated cascade, slow and turbulent flow condition has been shown to induce gene expression of proinflammatory molecules in the endothelial cells [4], [42]. Slow and turbulent flow condition also enhances the adhesion of leukocytes to the endothelial cells [20].

The use of appropriate experimental animal model is necessary for the better understanding of the progression of atherosclerosis. Most of the previously introduced atherosclerotic models in experimental animals used the method of mechanical endothelial injury to induce the endothelial dysfunction [38], [50]. Ishii et al [21] introduced a unique swine model of carotid atherosclerosis, in which atherosclerosis plaques could be induced by combining a high-fat and high-cholesterol diet with surgically created stenosis of a carotid artery to elicit local hemodynamic changes (Fig. 3A-D). Their experimental swine, including a control group, was fed with a high-fat diet until death. Advanced vulnerable atherosclerotic plaques were observed in the surgically created stenosis group without mechanical endothelial injury, and no atherosclerotic changes were confirmed in the control group despite the high-fat diet [21]. Their model suggests that the endothelial dysfunction solely induced by hemodynamic modification is an essential feature of atherosclerotic plaque formation.

View Large Image Download to PowerPoint

Fig. 3. Carotid atherosclerotic plaque induced by hemodynamic alteration in experimental swine. (A) Intraoperative view of carotid stenosis induced by surgical ligation. (B) Carotid arteriogram immediately after the surgical ligation. (C) Carotid arteriogram 3 months after the surgical ligation showing liminal narrowing proximal to the surgically induced stenosis and marked poststenosis dilatation distally. (D) Macroscopic view of the carotid atherosclerotic plaque (courtesy of Dr Akira Ishii).

3. Hemodynamics and saccular brain aneurysms 

The role of blood flow physiologic parameters regulating aneurysm morphology and natural history is very poorly understood. Recent research on animal models suggests that local hemodynamic forces including fluid-induced wall shear stress contribute to the initiation of saccular brain aneurysms [11], [43]. Some studies suggest that focal high wall shear stress induced by a certain flow pattern is a predisposing factor for brain aneurysm formation in healthy arteries [35]. The distribution of wall shear stress in a complex anatomy is not uniform because it is in a straight tube [39].

Because of the complex anatomy of brain arteries, hemodynamic research using ideal spherical-shaped brain aneurysm model provides only a limited knowledge [29]. For that reason, hemodynamic researches using geometrically realistic aneurysm model created from clinical images are required to reveal the role of hemodynamics in the initiation and natural progression of brain aneurysms [6], [7], [46], [47], [52], [53], [54], [55]. Saccular brain aneurysm may be formed not as the result of the physical or mechanical force by running blood flow but as the result of the interaction of local hemodynamics with the endothelial cells and arterial wall [11], [35]. Because the endothelial cells regulate local vascular behavior, local increase of wall shear stress theoretically causes local dilatation or local vascular remodeling. Furthermore, local decrease of wall shear stress may induce inflammation in the wall of brain artery and promote degenerative remodeling process [47], [54], [56]. The uneven distribution of hemodynamic forces created by torturous anatomy and the unique histologic feature of brain arteries that they lack the external elastic lamina may also contribute to the formation of saccular brain aneurysms.

To understand the complex mechanism involved in the initiation and progression of brain aneurysms, the use of appropriate animal models with hemodynamic and histologic evaluations is necessary. Assuming it is hemodynamics force that induces the degenerative remodeling of arterial wall and initiate the formation of brain aneurysms, the evaluation of artificially created hemodynamic condition in surgically created aneurysms in experimental animals might have very limited value [9], [48]. Hashimoto et al [17], [18], [19] introduced a remarkable brain aneurysm model in experimental rats. Their animal study has demonstrated that brain aneurysms were formed at the anterior cerebral artery-olfactory artery bifurcation as a result of hemodynamic stress (Fig. 4A-D). The histopathology of their model appears to resemble that of humans [19]. Although the result is still controversial, a study using their model showed that the incidence of brain aneurysm formations in rats was reduced by administration of NO synthase inhibitor despite the same amount of mechanical hemodynamic stress at the anterior cerebral artery-olfactory artery bifurcation [11]. Their study results suggest that high shear stress and NO are related to the initiation of brain aneurysms.

View Large Image Download to PowerPoint

Fig. 4. Progression of arterial wall degeneration toward the formation of brain aneurysm in experimental rats (hematoxylin-eosin–stained sections). (A) Normal anterior cerebral artery-olfactory artery bifurcation. (B) Early stage. Degeneration of internal elastic lamina is observed. (C) Advanced stage. Degeneration and disruption in the internal elastic lamina with thinning in the media. (D) Further advanced stage. Complete disruption of the internal elastic lamina, marked thinning in the media, and apparent spherical-shaped protrusion.

In recent years, more attentions have been given to the role of vascular remodeling not only in the initiation of brain aneurysm but also in the progression. It has already been shown that there are endothelial cells and some smooth muscle cells in the dome of unruptured brain aneurysms [25]. Nevertheless, the role of endothelial cell and vascular remodeling in the progression of brain aneurysm is poorly understood. On the other hand, neurosurgeons, neuroradiologist, and neuropathologist know from their clinical experiences that atherosclerotic plaque can be formed in the parent artery as well as in the dome of saccular aneurysm [27], [40]. Therefore, it is clear that the wall of saccular aneurysm is not an inactive tissue simply behaving like a balloon, and pathologic remodeling processes such as atherosclerosis do take place there [5], [41]. It is assumed that the complex and turbulent flow condition in the dome of brain aneurysms could induce some endothelial dysfunctions, which may result in the formation of atherosclerosis in the wall of brain aneurysms. Because hemodynamic stimuli such as wall shear stress play a significant role in the progression of atherosclerotic plaques in carotid and coronary arteries and descending aorta, the wall of brain aneurysms may also be capable of sensing wall shear stress, and degenerative vascular remodeling process may be involved in their natural progression.

Recent studies have shown the expression of MMPs and tissue inhibitors of metalloproteinases (TIMP) in the dome of human brain aneurysms [23], [41]. Both MMPs and TIMP are essential enzymes in the vascular remodeling. The balance of MMPs and TIMP expressions or activities is one of the important factors that regulate the vascular remodeling process, and their imbalance may result in the destruction of tissues. An overexpression of MMP-9 has been observed in the aortic aneurysm as well as brain aneurysms [23], [58]. Jin et al [23] studied MMPs and TIMP gene expressions in the ruptured and unruptured brain aneurysm harvested during the clipping surgeries. Their study suggested the disproportional expression of MMP-2, MMP-9, and TIMPs contributed to the progression of brain aneurysms.

We have reported that blebs of brain aneurysms were not necessarily exposed to a direct impingement of blood flow [53]. Moreover, our hemodynamic analysis in a growing brain aneurysm showed that there was no impinging blood flow onto the enlarging area of the brain aneurysm throughout its growth, suggesting that the bleb was not exposed to particularly high dynamic pressure compared with the nonenlarging area of the aneurysm (Fig. 5) [54]. There was a flow separation noted at the margin of the bleb, where we found relatively higher wall shear stress [54]. The flow within the aneurysm bleb was relatively stagnant, and the magnitude of wall shear stress was low in the tip of the bleb [54]. Interpretation of these findings is very difficult without correlation between hemodynamic and immunohistochemical findings. The high shear stress at the margin of the bleb might have contributed the degeneration of the aneurysm wall and eventually the growth of the aneurysm. Another possible explanation is that the slow flow condition and low wall shear stress in the tip of the bleb might have induced inflammatory reactions in the wall of the aneurysm wall.

View Large Image Download to PowerPoint

Fig. 5. Intra-aneurysmal hemodynamics before and after growth of a middle cerebral artery aneurysm. Particle imaging velocimetry flow analysis shows that the fast inflow blood stream does not impinge onto the bleb throughout its growth.

Given the high morbidity and mortality of subarachnoid hemorrhage and the high prevalence of small brain aneurysms, it would be very useful to detect certain hemodynamic patterns associated with a higher incidence of rupturing. Cebral et al [7] performed computational fluid dynamics analysis in 62 ruptured and unruptured brain aneurysms. They classified the brain aneurysms into different categories depending on the complexity and stability of the flow pattern, the location and size of the flow impingement region, and the size of the inflow jet. Specifically, 72% of ruptured aneurysms had complex or unstable flow patterns, 80% had small impingement regions, and 76% had small jet sizes. Conversely, unruptured aneurysms accounted for 73%, 82%, and 75% of aneurysms with simple stable flow patterns, large impingement regions, and large jet sizes, respectively. Aneurysms with small impingement sizes were 6.3 times more likely to have experienced rupture than those with large impingement sizes (P = .01). These results have not yet reached to the certain level that can be applied to the clinical practice of brain aneurysms. It, however, shows a possibility that computational fluid dynamics analysis using clinical imaging concomitant with clinical history of the brain aneurysm may bring us a new risk factor of aneurysm rupturing other than the maximum diameter of brain aneurysms that we currently rely on.

Phase-contrast magnetic resonance imaging (PC-MRI) is another useful tool to analyze the complex intra-aneurysmal hemodynamics in the near future [1], [16], [22], [37], [44], [51]. The PC-MRI technique itself has already been used in clinical setting to visualize the cerebrospinal fluid flow in the cisterns and in the ventricular systems. The authors have conducted an in vitro feasibility study of intra-aneurysmal flow visualization using PC-MRI technique [1], [51]. The intra-aneurysmal hemodynamics obtained by PC-MRI was very similar to the one obtained by using laser Doppler velocimetry, one of the most accurate flow measurement tool today. The method and technique are ready to be applied in current MRI suite [16]. The capability to directly visualize the intra-aneurysmal flow structure as well as the aneurysm geometry will bring a positive impact upon the future clinical practice of brain aneurysms.

4. Conclusion 

The endothelial cells sense and integrate hemodynamic and environmental stimuli, and produce various mediators that control vascular functions. The caliber and also the histologic structure of vessel walls are regulated by blood flow, particularly by fluid-induced wall shear stress. In the presence of the endothelial cell and smooth muscle cell, acute or chronic increase of wall shear stress due to increased blood flow elicits an adaptive response of the arterial wall histology, leading to vessel enlargement and reduction in blood flow velocity and wall shear stress to physiologic baseline values. The controlled and balanced degeneration and synthesis of vascular extracellular matrix allows vessels to adapt their diameter in a very wide range for sustained increase or decrease of the blood flow. However, the imbalanced degeneration of extracellular matrix induces the pathologic remodeling, which is related to various cerebrovascular diseases.

Wall shear stress and its interaction with the endothelial cell and smooth muscle cell regulate not only the vessel diameter but also the proliferation of endothelial and smooth muscle cells, endothelial permeability, and inflammation in the vessels. Laminar and physiologic shear stress is one of the crucial factors to maintain the proper level of NO, a key mediator of vessel behavior. Attenuation of NO is one of the earliest changes preceding endothelial dysfunction. Endothelial dysfunction induced by slow or turbulent flow condition promotes atherosclerosis in conjunction with such systemic risk factors as hypercholesteremia, hypertension, diabetes, and smoking.

Local hemodynamic forces including fluid-induced wall shear stress contribute to the initiation of saccular brain aneurysms. Saccular brain aneurysm may be formed as a result of imbalanced degeneration and pathologic remodeling. It is also suggested that the destructive remodeling process takes place in the wall of fully developed brain aneurysms. Hemodynamic analysis using clinical data will improve the current poor predictability of the natural history of small incidental aneurysms and may help us to distinguish incidental aneurysms at higher risk of rupturing from those at low risk in the future clinical practice.