Biology and hemodynamics of aneurismal vasculopathies

European Journal of Radiology 82 (2013) 1606–1617 Contents lists available at ScienceDirect European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad Biology and hemodynamics of aneurismal vasculopathies Vitor Mendes Pereira ∗ , Olivier Brina, Ana Marcos Gonzalez, Ana Paula Narata, Rafik Ouared, Karl-Olof Lovblad Interventional Neuroradiology Unit, Service of Neuroradiology, University Hospital of Geneva, Switzerland a r t i c l e i n f o Article history: Received 30 November 2012 Received in revised form 11 December 2012 Accepted 13 December 2012 Keywords: Intracranial aneurysm Hemodynamics Rupture Cerebral vasculopathy Wall disease and angiography a b s t r a c t Aneurysm vasculopathies represents a group of vascular disorders that share a common morphological diagnosis: a vascular dilation, the aneurysm. They can have a same etiology and a different clinical presentation or morphology, or have different etiology and very similar anatomical geometry. The biology of the aneurysm formation is a complex process that will be a result of an endogenous predisposition and epigenetic factors later on including the intracranial hemodynamics. We describe the biology of saccular aneurysms, its growth and rupture, as well as, current concepts of hemodynamics derived from application of computational flow dynamics on patient specific vascular models. Furthermore, we describe different aneurysm phenotypes and its extremely variability on morphological and etiological presentation. © 2013 Elsevier Ireland Ltd. All rights reserved. 1. Introduction 2. Biology of the vessel wall and vascular remodeling Aneurismal disease consists of vascular wall pathologies sharing common morphological vascular changes [1]. It is characterized by a dilatation of the vessel wall (aneurysm) which produces the classical phenotype that is shared by all entities from different etiologies [2]. It corresponds to a spectrum of disorders that varies from a true iatrogenic cause such as a traumatic aneurysm from a high-energy accident to a complete spontaneous arterial rupture in the context of a genetic disorder [3–5]. In between these extremes, arterial dissections of different origins are included which include saccular aneurysms with their complexity including familial cases related or not to genetic diseases [1,3]. The development of intracranial aneurysms (IAs) is often related to the specificity of each etiology, morphology and presentation. In this review, we present the etiological distinction and describe current biological and hemodynamic aspects that can be found and shared among different vascular disorders. Blood vessels are tubular organs disposed into a network and that contain three layers. These layers interact with each other in order to promote vascular integrity and functionality [6]. Intracranial vessels present some particularities compared to other vessels. The adventitia is the outer layer composed of connective tissue, nutrient vessels (vasa vasorum) and autonomic nerves. When the cerebral vessels enter the subarachnoid space their adventitia is made up of leptomeningeal cells [7]. The muscular or middle layer is the thickest because it consists of smooth muscle cells and elastic tissue taking care of the local vasoregulation. The leptomeningeal arteries lose their external elastic limitans as soon as they cross the dura mater. The inner layer is the endothelium, which has anti-atherogenic properties and inhibiting platelet aggregation. Additionally, it controls vasodilatation, cell adhesion and inflammatory activity (smooth muscle proliferation, and adherence/pro-inflammatory cascades) [8]. Blood vessels are formed by the alignment of endothelial cells and their branches, by sprouting, sending out cells at the forefront of the sprout to a determined target [9]. Vasculogenesis is the formation of the first primitive vascular plexus inside the embryo and angiogenesis is the remodeling and expansion this network. Vasculogenesis refers to in situ differentiation and growth of blood vessels from mesodermal derivatives and angiogenesis is a combination of two mechanisms: endothelial sprouting and intussiseptive micro vascular growth [10]. These are complex processes mediated by hormones, chemoreceptors and inflammatory substances that will modulate the formation of the vasculature according to hemodynamics. Assuming that a vascular disease can Abbreviations: SAH, subarachnoid hemorrhage; CT, computed tomography; MRI, magnetic resonance imaging; MCA, middle cerebral artery; ACA, anterior communicating artery; CFD, computational flow dynamics. ∗ Corresponding author at: Department of Neuroradiology, University Hospitals of Geneva, Switzerland, Rue Gabrielle-Perret-Gentil, 4, Geneva 1211, Switzerland. Tel.: +41 79 553 25 35; fax: +41 22 372 70 72. E-mail addresses: vitormpbr@hotmail.com (V.M. Pereira), olivier.brina@hcuge.ch (O. Brina), ana.marcosgonzalez@hcuge.ch (A.M. Gonzalez), ana.p.narata@hcuge.ch (A.P. Narata), rafik.ouared@unige.ch (R. Ouared), Karl-olof.lovblad@hcuge.ch (L. Karl-Olof). 0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2012.12.012 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 Fig. 1. Spectrum of aneurismatic disease and distributions of different diseases and its relationship with related factors. Sample of diseases: 1, Ehlers–Danlos aneurysm; 2, neurofibromatosis type 1; 3, polycystic kidney disease; 4, spontaneous arterial dissections; 5, familial aneurysms; 6, multiple aneurysms; 7, isolated saccular aneurysm; 8, high-energy traumatic arterial dissection. Modified from a draw of Pierre Lasjaunias, Chiang-Mai, November, 2006. be initiated at any stage of the vascular development, a determined targeted defect will affect the vascular network from that point of the development. So, considering topography of an abnormality, the earlier (in embryonic development) it occurs, the larger (or diffuse) will be the involvement of vessels derived from that segment. Therefore, a sporadic, non-familial, isolated saccular aneurysm would correspond to a very late defect on the vessel wall compared to a giant aneurysm that involves one segment of the intracranial carotid or middle cerebral artery for example. From this point, the formation of IAs will be a conjunction of genetics, development and molecular factors that will be affected during patient’s life by other phenomena like remodeling process, external risk factors and hemodynamics. Different phenotypes will be the result of the interaction between all these elements and the biological predisposition. This concept correspond to the congenital nature of vascular diseases and can explain some of the multiple, familial or segmental distribution of IAs [11] (Fig. 1). The formation of an aneurysm is a complex process that involves a number of biological, physical and external factors associated to genetic predisposition that will interact among them to produce an initial vascular lesion, which will not be corrected by the healing process of the vessel wall, and, therefore, it will progress and grow influenced by the vascular remodeling process. 2.1. Hemodynamics The brain receives 20% of the circulating blood volume through the four cervical vessels (carotid and vertebral arteries). There are anatomical particularities like presence of curves and bifurcations at the level of the circle of Willis that create a peculiar development of the blood flow. Circadian variations as the systolic and diastolic cycle, resting and exercising status or chronic conditions like hypertension increase the complexity of the evaluation of the intracranial hemodynamics. The interaction between the circulation and the vessel wall follow the Poiseuille’s law [12], which basically describe the laminar displacement of a viscous fluid in a pipe. Among different mechanical forces applied on cerebral vasculature, the tangential force applied to the vessel wall by the blood flow, namely wall shear stress (WSS), was identified to influence endothelial cells and its role in vascular diseases was largely studied [13]. For instance, high levels of WSS are considered as a prerequisite for the initiation of IAs [14]. On the other side, the implication of the WSS in growth and rupture is still unknown and subject to diverging opinion [15]. Indeed, unlike in other anatomical regions, hemodynamic in IAs present certain characteristics that complicate studies attempting to investigate the impact of flow in the aneurismal disease. For instance, the absence of safe in vivo measurement methods 1607 precludes any direct measurements for clinical validation of noninvasive methods. The models available, either demonstrate the vascular changes after dramatic hemodynamic changes without presenting similar morphology or they present geometrical resemblance but a damaged vessel wall due to the technique employed to build the animal model [16]. If we consider only the hemodynamic aspect out of the biologic context, in vitro studies took then an important place to understand the intra-aneurismal flow, thanks to significant improvements in models conception that reproduce patient’s anatomy. Different techniques of visualization have used material like ink, particles and even more sophisticated quantitative measurements techniques as particles image velocimetry (PIV). There are however limitations to these experiments like the accuracy of the pump to simulate physiological blood flow, constraints with the vessels wall and the criteria and tool for measurements [17]. To overcome those issues, computational flow dynamics (CFD), largely used in engineering and industry, started to be used to evaluate intracranial vessels and IAs [18]. It consists of a simulation method based on Navier–Stokes equations that model the blood flow through patient specific vascular geometry usually generate from patient imaging data. CFD can provide WSS quantities and distributions in IAs [18], but also many fluid dynamics relevant parameters as Oscillating Shear Index (OSI) which reflected the aneurysm wall solicitation and fatigue. In addition, velocity fields could be computed and visualized as streamlines or vectors pathway and can be used for qualitative assessment of the flow behavior. CFD is helping in the understanding of IAs flows but the relationship with aneurysm formation, growth or rupture is still to be proven. Most of the IAs are located on bifurcations (MCA, ACA, Basilar and carotid bifurcations) though they have different outcomes concerning rupture risk and growth [2]. They have a particular flow behavior with a high-pressure zone located on the artery close to the neck border and a high WSS on the branches of the bifurcation. The inflow streamlines behavior and direction will influence the distribution of the WSS on the aneurysm surface and it has been demonstrated very dependent on the inflow angle and the spatial position of the aneurysm according to the artery. Sidewall aneurysms present a flow related in part to its geometry and part on the inflow angle inside the aneurysm compared to the artery. The high pressure is concentrated on the distal border of the neck as well as the high WSS zone. The complexity of the flow vectors or streamlines of both types of IAs has been studied and although is subjective, some “patterns” are suppose to predispose IAs rupture [18] or growth. Cebral et al. studied retrospectively 210 IAs using qualitative parameters to define the intra aneurismal flow behavior; they found that concentrated inflow jets, small impingement areas complex and unstable flow patterns were correlated with the clinical history of aneurysm rupture. Additionally, they also used quantitative parameters related to flow behavior on the same patient cohort, and confirmed these trends [19]. Basically, these findings highlight that maximum WSS and concentrated inflow streams are statistically correlated with aneurysm rupture (Fig. 2). By contrast, some recent studies have pointed out that ruptured aneurysms seem to present a larger area of low WSS, or lower global WSS than un-ruptured ones. Xiang et al. showed in a retrospective study on 119 aneurysms that low WSS area and OSI are independent significant variables in a multivariate analysis [20]. Jou et al. also found in 26 IAs that ruptured aneurysms had greater area under low WSS whereas maximum WSS was similar for both ruptured and unruptured aneurysms groups [21]. Lu et al. used 9 mirrors IAs and showed that the ruptured aneurysms manifested lower WSS compared to their parent artery. Low WSS induced by under-stimulation of the arterial wall is correlated with wall degradation by degeneration of endothelial cells via the 1608 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 Fig. 2. Graphic representation of CFD results of two patients. Colors represent values in a range from the lowest (blue) to highest (red) in the correspondent unit of the parameter. A, B, intracranial carotid bifurcation aneurysm; A, WSS representation; B, pressure representation; C, D, posterior communicating artery aneurysm; C, WSS with superposed streamlines; D, pressure. apoptotic cell [22]. However, in IAs the exact threshold to determine the range of low WSS under which endothelial cell dysfunction causing arterial wall remodeling is not known. Many reports linked low WSS and rupture using different thresholds (0.4, 1, 1.5 Pa) or used a parent vessel WSS reference at various locations but the absence of consensus in low WSS definition between CFD studies precludes any general conclusions and only allows the discussion of general trends. Another approach is applied by Goubergrits et al. when they use statistical WSS mapping, evoked the possibility of high and low WSS cohabitation with large areas of steep WSS transition in ruptured IAs [23]. In the same idea, Shojima et al. also discussed the proximity of high and low WSS in a small area of the aneurysm that might increase degenerative changes at the wall [22]. However, even if many valuable studies, including small prospective studies and multivariate analyses have been published, CFD in IAs studies still resist to be transposed to the clinical side [20,24]. Fluid structure interaction, constant viscosity, geometric modeling, Newtonian flow behavior and boundary conditions are technical challenges that would need to be addressed in the near future. Furthermore, as explained by Taylor et al. in a large review of image-based modeling, the CFD studies are based on a mechanical model and not on a biomechanical model and underline that there is a real need to improve the biological relevance of the CFD studies. If we add the large IAs morphological variability, the relatively low amount of cases available and the ethical issue to conduct large-scale prospective studies on unruptured IAs, the need to focus essentially on the practical clinical utility of CFD becomes obvious. Essentially, CFD technique would gain in a better understanding of the large amount of data in term of clinically relevant parameters. 2.2. Biological mechanisms of the rupture In general, saccular IAs are vascular lesions formed probably after a local injury of the vessel wall that progressively develop a dilatation of this wall. After completely developed, an aneurysm has, at least, three possible outcomes: the first is to remain unchanged and stable over the years; the second is to grow until it reaches a size that is considered large or giant associated or not with an increased ruptured risk or compressive symptoms; and the third possibility, which is the most threatening one, the aneurysm rupture [2]. Aneurysm growth remains a not fully understood phenomenon. It has been reported to be 6 or 7% during a follow-up period of to 41 months [25]. The rupture risk of growing aneurysms is not known but it seems that there is instability at level of the vessel wall promoting or permitting the growth. There is no means to predict what will happen with an unruptured aneurysm. Recent studies have stated that small aneurysms carry a small risk of rupture although, in daily practice, we observe significant rupture in this small aneurysms group [26]. In fact, probably, these studies presented low patient numbers or a methodological bias and where a correct estimation of the real environment is compromised. Are ruptured and unruptured aneurysms the same entity or we are dealing with two different processes? Histologically, ruptured and unruptured aneurysms are different [27]. Small-ruptured aneurysms present less or no collagen or smooth cells layer but a hyaline-like structure covering the aneurysm sac [28]. Thus, these ruptured small lesions are recently formed without time for a sufficient vascular remodeling? In contrast, unruptured small aneurysms contain both dense collagen and smooth muscle cells, which is in favor of remodeling and vascular wall reinforcement possibly part of a healing process. V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 1609 Fig. 3. 42 y old female patient presenting SAH (A) after a rupture of a saccular posterior communicating artery. Angiogram demonstrated the ruptured lesion but also a false aneurysm corresponding to the site of rupture (B). CFD results from the 3DRA (C) with representation of the streamlines and velocity. Note the inflow jet entering through the neck and diverging on the opposite wall. The rupture point is located at the end of the sac on a lower WSS zone. Some studies also describe the presence of inflammatory activity with hypercellularity and complement activation [29], apoptosis [30], variations in smooth muscle cells patterns [31] and augmented protease activity. Factors distinguishing unruptured and ruptured IAs are: decellularization, apoptosis, and degeneration of wall matrix; de-endothelialization; thrombus organization; proliferation; and inflammatory infiltration (T-cell and macrophage infiltration as well as SMC proliferation) [32,33]. Furthermore, atherosclerotic calcifications of associated plaques are not a common feature in saccular aneurysms [27,28,33]. Histological variations or specific patterns have not been related to location, size (on saccular cases only) or presence of secondary pouches [33] (Fig. 3). It seems that before a rupture of an IA, there is histological activity at the level of the aneurismal wall expressing the described changes. The secretion of the matrix metalloproteinases during remodeling and myointimal hyperplasia can affect part of the wall matrix and permit the migration of smooth muscle cells. Conversely, the gene expression of inflammation is augmented in the aneurysm wall compared to arterial sample controls, but it is higher in the unruptured group, indicating that inflammation can also be related to a protective mechanism against rupture [34]. Recently, more evidence has been showed in favor of a common genetic association: indeed, a connection between IA risk and a risk allele at immediate proximity to endothelin receptor type A (EDNRA) has been found [35]. EDNRA has been associated with various other vascular conditions and it seems to be activated whenever there is a vascular lesion [36]. EDNRA mediates a vascular mitogenic effect of EDN1 by promoting cell cycle progression and proliferation that might play a role in IA progression and rupture. 3. Aneurysm phenotypes 3.1. Saccular aneurysms Affecting 2–4% of the population, saccular aneurysms are clinically silent and can present subarachnoid hemorrhage (SAH) around fifties (±10 years) with, sometimes, significant mortality and morbidity [2,26]. Classically, epidemiological risk factors are hypertension, smoke, familial history of SAH, female sex and drug abuse [2]. Previous SAH in a context of multiple aneurysms is also related to increased rupture risk. Some patients can present different symptoms before the rupture that signify either minor leaks or rapid growth and inflammatory activity. Sentinel symptoms in unruptured aneurysms were associated to the same histological pattern observed of ruptured aneurysms (extremely thin thrombosis-lined hypo cellular wall) before diagnosed SAH [33]. A large amount of previously described histology and biology involved essentially saccular aneurysms [30,33,36]. They are, by far, the most common aneurismal disease and they are mostly located 1610 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 Fig. 4. 32 y old female patient presenting frequent retro auricular headaches. Imaging set-up demonstrated bilateral paraclinoid saccular aneurysms in a mirror configuration. Fig. 5. 56 y old male patient presenting multiple saccular aneurysms. Angiogram (A, left ICA – AP view; B, left ICA – oblique view; C, left vertebral – oblique view). Presence of a anterior communicating aneurysm, left MCA, Left posterior communicating and a left superior cerebelar aneurysm. at bifurcations (anterior communicating artery, middle cerebral artery and basilar bifurcation) [2]. They can occur sporadically in one location as consequence of a local predisposition and external factors or it can be associated with early defects on developmental phase and be multiple or segmental [11] (Figs. 4 and 5). Frosen found no differences on IAs walls of familial patients compared to sporadic ones [33]. They can be also associated to a large number of genetic diseases as described on a previous chapter of this special supplement (Fig. 6). 3.1.1. Giant saccular aneurysms Prevalence is expected to be around 5% of all aneurysms [37] but this number contains probably not only saccular aneurysms but also some fusiform lesions. The natural history can have a very high morbidity with a 25% annual risk of hemorrhage and a mortality of 60% over 2 years [38]. In this part, we will focus on saccular aneurysms that grow over time and, which, depending on their size and location the can be clinically manifest with compressive symptoms or rupture. They are considered large when they exceed 25 mm in maximum diameter [39]. They can present a partial thrombosis inside the giant sac, calcifications of the aneurysm’s wall or they can present an intrinsic vascularization via the vasa-vasorum [40]. The morphology of more complex lesions can be undistinguishable from other pathological conditions like chronic dissection or other fusiform aneurysms but this pathophysiological differentiation will be important for the management [1,3,40]. Studies have shown that the aneurysm wall of large ruptured lesions present a different histopathological pattern to small ruptured ones [30,32,33,38]. Small lesions present a hyaline-like structure meaning early on development and unstable, instead large lesions rarely present such a type of composition in their wall when they rupture indicating that they are unlikely to be newly formed [28]. Thus, aneurysms that grow to a large size seem to behave differently from small ruptured aneurysms (Fig. 7). Another mechanism of growth seems to be related to the partial thrombosis that can be formed inside the aneurysm due to a stagnated intra-saccular flow. The mural thrombus is believed to injure the endothelium and internal elastic lamina followed by scarring of the wall, invasion of fibroblasts, collagen formation and deposition of fibrous material [38]. Platelet aggregation and stagnation of flow under the scarred rigid aneurysm wall create intra-luminal clots. A lattice-Boltzmann model of thrombosis in giant aneurysms demonstrated that the relation between the wall shear stress area and the aspect ratio of the aneurysms would determine partial thrombosis [41]. Platelet adhesion and aggregation are primarily driven by changes in blood flow rheology and stabilized soluble agonists. The level of WSS inside the aneurysm can influence the activation of platelet aggregation and initiate the thrombus inside the aneurysm [41]. This clot can present instabilities and presents intra-thrombus repetitive bleeding and produce an inflammatory reaction inside the clot and on the aneurysm wall and promote its growth [42]. Some authors reported the presence of multiple layers on histological analysis of intra-aneurysm thrombus presenting different aspects indicating different formation periods [42]. Independent of the pathomechanism, the outcome can be threatening either due to its extremely high rupture risk or due to compression and ischemic symptoms derived from a mass effect [26,38]. The management of these lesions has been quite V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 1611 Fig. 6. 72 y old female patient with a left ICA cavernous aneurysm (A, B and D) that has been growing after 4 years of follow up compressing cranial nerves. Angiogram before treatment demonstrated associated significant cervical ICA tortuosity and fibromuscular dysplasia (C). Example of association of saccular intracranial aneurysms and chronic vascular wall diseases. controversial over the years. Parent vessel sacrifice after occlusion test or associated to by-pass surgery when required was the most common treated for many years. Other treatment approaches have been proposed since like clipping, coiling or stent assisted coiling but the results could be irregular and the morbidity would remain high. Recently, the introduction of flow diverter devices (FDDs) has proven itself to be a valid therapeutic option for these complex lesions. The principle of parent vessel reconstruction with intra-aneurismal flow modification and consequent aneurysm thrombosis and vascular remodeling is especially interesting for giant lesion with mass effect. Clinical results are quite compelling and the re-absorption and remodeling of the aneurysm itself and complete endothelial covertures around the neck of the aneurysm guarantee the long-term stability [43,44]. Reports of complete occlusion rates vary from 80 to 100% over a 1-year period with an associated morbidity of 4–8%. However, giant saccular aneurysms have been associated with the most life-threatening complication related to FDDs treatment, which is the spontaneous post-procedure rupture. This rupture occurs from hours until several weeks after FFD placement and it is probably related either to intra-aneurismal flow changes induced by the device or by an inflammatory activity related to a partial and unstable thrombus formed inside the aneurysm sac [45,46]. Furthermore, the cases presenting with mass effect can present a significant worsening after procedure that may improve or not after treatment. Still, this treatment remains the best therapeutic option for such lesions. 3.2. Fusiform aneurysms Fusiform aneurysms correspond to a morphological syndrome diagnosis of a group of disorders presenting the same arterial geometrical appearance: concentric vascular dilation or aneurysm formation without a defined neck [39]. These lesions range from acute or chronic spontaneous dissections to lesions related to connective tissue diseases like Ehlers–Danlos, Marfan or neurofibromatosis type 1 [3,4]. The development of these lesions will depend on the etiology involved originally but they will share probably some mechanisms and also some possible clinical manifestations. These lesions are usually more often encountered on longer segments of intracranial large arteries rather than on bifurcations or distal small vessels [37,38,42]. The process starts with an injury on the vessel wall that will either be acutely developed and symptomatic or chronically progressive over a long period before being clinically manifested [47]. The fusiform morphology 1612 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 Fig. 7. 73 y old female patient presenting transitory ischemic lesions. CT angio demonstrate a large basilar tip aneurysm (A, B). Angiogram (C, D) confirmed the diagnosis. Note the presence of calcifications on the aneurysm sac. Lesions that growth present a stable vessel wall compared to lesions that rupture with smaller sizes. is referred to by different terms in the literature that can cause confusion because some aneurysms can present a similar appearance but a complete different natural history. We describe this class of aneurysm using an etiological approach to better describe their specificities. 3.2.1. Spontaneous arterial dissections Spontaneous dissection corresponds to the most common etiology related to fusiform aneurysms. The initial mechanism of lesion is a sudden disruption of the endothelium, the intimae and the internal elastic lamina with subsequent invasion of circulating blood into the media [3]. The diagnosis of a dissection is classically made by the typical angiographic appearance with a double lumen, a focal vessel wall irregularity, a pre-aneurismal narrowing and a fusiform dilatation [3]. On MRI, intramural thrombus, an irregular vessel wall and a vessel wall flap can be seen. Furthermore, the presentation and pathological evolution will depend on the degree of vascular damage and the remodeling and reparation processes [3]. Acute and sub acute presentations will depends on the involvement of the vascular wall and be essentially two mains groups: hemorrhagic and ischemic presentations. Intracranial hemorrhage can occur when vascular wall damage reaches the adventitia and ischemic symptoms can be related to hypo perfusion due to an arterial occlusion or to a distal embolism of a clot that can dislocate from the dissected segment. Most of these lesions, independent of the presentation will regress spontaneous if a specific treatment is not necessary. The healing mechanism may be delayed under several conditions such as aneurysms with extensive defects of the aneurismal wall in the ruptured portion (i.e. large aneurysms), aneurysms with abundant thrombus in the ruptured segment, or aneurysms in which the media is completely separated from the adventitia [48]. In addition, the healing response may be insufficient in underlying vessel wall diseases [1,3]. Then, the classic fusiform morphology of aneurysms will be formed. They can vary from large partially thrombosed aneurysms to long serpentine segments [47] (Fig. 8). Idiopathic dissections, when an underlying disease is not found, are the most common disorders but a significant number of connective tissue or genetic diseases can increase the vascular fragility and predispose for dissections with a minimum traumatism. Ehlers–Danlos syndrome, Marfan disease, Neurofibromatosis type 1, pseudoxantoma elasticum, fibromuscular dysplasia are among the most frequent disorders associated with acute or chronic arterial dissections [3,4]. V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 1613 Fig. 8. 75 y old male patient presented at the hospital with hemiparesis. MRI showed a pontine infarction (A) and a dilated basilar artery with partially thrombosed lumen (B). Patient was treated with anticoagulation and a CT angiogram (C, D) two years later showed a remodeling and clot partial regression. Note the presence of calcifications and the large vessel wall in some segments that are compatible with a long and dynamic process of lesion, remodeling and regression. 3.2.2. Infectious aneurysms Bacteria and fungi can produce intrinsic arterial wall lesions while virus are more associated to vasculitis. The mechanism consists in an infiltration of the vessel wall via intraluminal pathogenic implantation from systemic infections or direct invasion in case of contiguous located infections [49]. Therefore, a secondary and reactive inflammatory activity will corroborate to the proteolytic and enzymatic pathogenic damage to the vascular structure and cause infectious aneurysms. This is a rare condition and its incidence is probably underestimated and the diagnosis is made, most of the time, after an intracranial hemorrhage in a context of a chronic or acute infection [50]. A recent review summarized previous infectious aneurysm reported in the literature and the ruptured infectious aneurysms were related mostly to systemic acute bacterial infection but also to bacterial meningitis [51]. Chronic bacterial conditions like tuberculosis have also been described with a more insidious presentation like fungal infectious. The treatment of the specific infection is, for sure, the first and most important therapeutic measure. The management of ruptured lesions has been mostly endovascular parent vessel occlusion with no reported further infectious complications although spontaneous regression with antibiotic therapy has been described [51]. This seems to be the most judicious approach when infectious aneurysms presented with intracranial hemorrhage and the risk of re-bleeding can be extremely high because of the nature of the arterial damage [50,51]. A particular condition is a lesion associated to adjacent infections. Deep sinusitis, meningitis or skull base bone infections will directly affect the vessels [50,52]. The presentations vary with the topography of the artery involved and can be from an ischemia due to an associated vascular occlusion to intracranial or extracranial bleeding. When the infections involve bone, they require prolonged antibiotics and the arterial infectious lesion should be carefully followed to avoid recurrences (Fig. 9). 3.2.3. Neoplastic aneurysms Neoplasic proliferation on the vascular wall of intracranial is the pathomechanism of tumor related aneurysms [53]. When a metastatic tumor or viable embolic cells invades the blood vessels, these cancerous cells will then spread through other organs. They then penetrate the vascular wall, proliferate, cause vascular damage and progressive dilatation [54]. The typical morphology consists of multiple fusiform aneurysms distally located and they can present intracranial hemorrhage [55]. Tumors from different origins have been associated to neoplasic aneurysms and the most described lesions are lung carcinoma and choriocarcinoma [54–56]. A particular tumor that is frequently associated with embolic ischemic or arterial lesions is the myxoma [53]. They present, typically, with a benign behavior but due to its most frequent location 1614 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 Fig. 9. Basilar trunk dissection with SAH (A) in a 2 y old child with leukemia presenting an infection at level of the clivus (MRI – T1 with gadolinium – F). Arterial lesion with false aneurysm (angiogram – B, C) treated with coiling and balloon remodeling (angiogram – D, E). Patient was treated with large antibiotic therapy and finished the chemotherapy with complete remission of both infection and neoplasia. in the left atrium, tumoral embolic complications and vascular neoplasic lesions are often associated [53,57]. 3.2.4. Dolichoectasias Natural large sized vessels can be observed spontaneously or in association with connective tissues diseases [39,47]. They can represent extreme outliers of the normal mean vascular diameter or they can be related to a baseline vascular disease that produces an inherent instability of the vessel wall that will permit a progressive increase of its size [58]. They are associated with other risk cardiovascular risk factors and their physiopathology is probably a combination of genetic predisposition with external risk factors. They represent a fragility that will predispose to a spontaneous dissection whenever at least an external traumatic cause is involved [58]. They can present with ischemic symptoms related to arterial wall chronic dissection or to embolic distal events. Compressive symptoms have been also reported like trigeminal neuralgia or obstructive hydrocephalus [59,60]. The best therapeutic management seems to be restricted to the control of external risk factors [58]. 3.2.5. Iatrogenic and traumatic aneurysms Traumatic intracranial aneurysms are lesions directly related to a brain trauma which is a completely external etiological factor [61]. We include in this category all cases of vascular lesions associated to brain trauma or a direct vascular injury related to a procedure [62]. At the cervical level, the most susceptible sites of traumatic injury are the first portion of the cervical internal carotid artery or the second segment of the vertebral artery that runs inside the transverse process of cervical vertebra [63]. Intracranially, most common affected vascular segments are located close to rigid structures like bone surfaces or dura mater structures [64]. Distal segments of anterior cerebral, middle cerebral or posterior inferior cerebellar arteries correspond to the most affected vessels in brain trauma mainly because they are located close to bone eminences or the pericallosal artery because of its proximity to the inter-cerebral falx [39,61,64]. Intracranial aneurysms related to brain trauma usually present with subarachnoid hemorrhage and, rarely, with neurological deficits due to vessel occlusion. Sometimes, false aneurysm can be identified and it requires acute management to avoid secondary rupture. They are classically associated with an acute event of delayed intracranial bleeding with a mean time from initial trauma to aneurysm hemorrhage of approximately 21 days [61]. The mortality rate for patients harboring these aneurysms may be as high as 50%. Prompt diagnosis and aggressive surgical management are associated with better outcome than conservative treatment. Cervical lesions can manifest differently with an acute occlusion, neck mass with large pseudoaneurysm formation, parachordal V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 1615 Fig. 10. 30 y old male patient present a car accident with multiple fractures including a skull base and sphenoid fracture (A). He presented a SAH 2 months later (B). CT Angio demonstrated a presence of an aneurysm at level of anterior communicating artery (C) confirmed with angiography (D). Comparing the CT angio with the previous (A) at the day of the accident the lesion was not presence. We described that lesion as a traumatic lesion and it was treated with coiling at the day of the SAH. arteriovenous fistulas or with a distal embolization of a thrombus formed into the dissected false aneurysm sac [4,47]. The management of lesions in this location will depend on the clinical presentation but it can vary from a conservative and observational management to a vascular reconstruction with multilayer stents in order to correct a symptomatic non-compensated intracranial hypo perfusion. Anti-aggregation or anticoagulation can be associated to this conservative attitude although their initial presentation will depend on the extent and severity of brain damage (Fig. 10). Iatrogenic aneurysms are associated to surgical procedures and it will depend more on the relationship between the surgical access and the affected vessel [63,65]. Cavernous carotid iatrogenic lesions are associated to trans-sphenoidal approaches, distal basilar locations are related to third ventriculostomies and second and third vertebral artery segments are linked to surgical instrumentation of the cervical region [65]. The presentation can be an acute or subacute hemorrhage in the surgical bed or access route or even an ischemic consequence of a distal embolization of a related thrombus or due an arterial occlusion. The management will depend on the presentation but will consist of a vascular reconstruction in case of hypo perfusion or parent vessel occlusion in case of acute bleeding [63,65]. 4. Conclusion Vessel wall biology and its relationship to a specific vascular disease is a complex and controversial topic and more investigation is necessary. A congenital and or genetical predisposition may be the initial process but they will be manifested when triggered features are associated later on. The interaction between genotype and risk factors will then determine the morphological presentation of most of the different types of aneurysms. High levels of WSS have been related to aneurysm initiation and combination of high and low WSS are related to its growth. Endothelium is sensible to hemodynamic conditions. The interaction between the flow and the vessel wall can influence the remodeling process. Inflammation seems to play an important role in the destabilization of the aneurysm vessel wall and to predispose rupture. Analysis of an aneurysm must consider its etiology and different pathomechanisms. Future research needs to focus on the evaluation of biological and biomechanical 1616 V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 information on intracranial vessels integrating hemodynamic assessment into clinical studies. References [1] Krings T, Mandell DM, Kiehl TR, et al. Intracranial aneurysms: from vessel wall pathology to therapeutic approach. Nature Reviews Neurology 2011;7:547–59. [2] Brisman JL, Song JK, Newell DW. Cerebral aneurysms. New England Journal of Medicine 2006;355:928–39. [3] Krings T, Lasjaunias PL, Geibprasert S, et al. The aneurysmal wall. The key to a subclassification of intracranial arterial aneurysm vasculopathies? Journal of Interventional Neuroradiology 2008;14(Suppl 1):39–47. [4] Pereira VM, Geiprasert S, Krings T, et al. Extracranial vertebral artery involvement in neurofibromatosis type I. Report of four cases and literature review. Journal of Interventional Neuroradiology 2007;13:315–28. [5] Peyre M, Ozanne A, Bhangoo R, et al. Pseudotumoral presentation of a cervical extracranial vertebral artery aneurysm in neurofibromatosis type 1: case report. Neurosurgery 2007;61:E658 [discussion E658]. [6] Raybaud C. Normal and abnormal embryology and development of the intracranial vascular system. Neurosurgery Clinics of North America 2010;21: 399–426. [7] Kathuria S, Gregg L, Chen J, et al. Normal cerebral arterial development and variations. Seminars in Ultrasound, CT and MR 2011;32:242–51. [8] Chalouhi N, Ali MS, Jabbour PM, et al. Biology of intracranial aneurysms: role of inflammation. Journal of Cerebral Blood Flow and Metabolism: Official Journal of the International Society of Cerebral Blood Flow and Metabolism 2012;32:1659–76. [9] Carmeliet P, Jain RK. Molecular mechanisms and clinical applications of angiogenesis. Nature 2011;473:298–307. [10] Patan S. Vasculogenesis and angiogenesis as mechanisms of vascular network formation, growth and remodeling. Journal of Neuro-oncology 2000;50:1–15. [11] Lasjaunias P. A revised concept of the congenital nature of cerebral arteriovenous malformations. Journal of Interventional Neuroradiology 1997;3:275–81. [12] Bonnefous OPV, Ouared R, Brina O, et al. Quantification of arterial flow with digital subtracted angiography (DSA). Medical Physics 2012;39:6264–75. [13] Malek AM, Alper SL, Izumo S. Hemodynamic shear stress and its role in atherosclerosis. JAMA: The Journal of the American Medical Association 1999;282:2035–42. [14] Gao LHY, Swartz DD, Kolega J, Siddiqui A, Meng H. Nascent aneurysm formation at the basilar terminus induced by hemodynamics. Stroke 2008;39:2085–90. [15] Nixon AMGM, Sumpio BE. The critical role of hemodynamics in the development of cerebral vascular disease. Journal of Neurosurgery 2010;112:1240–53. [16] Lewis DA, Ding YH, Dai D, et al. Morbidity and mortality associated with creation of elastase-induced saccular aneurysms in a rabbit model. American Journal of Neuroradiology 2009;30:91–4. [17] Steinman DA. Computational modeling and flow diverters: a teaching moment. American Journal of Neuroradiology 2011;32:981–3. [18] Castro M, Putman C, Radaelli A, et al. Hemodynamics and rupture of terminal cerebral aneurysms. Academic Radiology 2009;16:1201–7. [19] Cebral JR, Mut F, Weir J, et al. Quantitative characterization of the hemodynamic environment in ruptured and unruptured brain aneurysms. American Journal of Neuroradiology 2011;32:145–51. [20] Xiang J, Natarajan SK, Tremmel M, et al. Hemodynamic-morphologic discriminants for intracranial aneurysm rupture. Stroke 2011;42:144–52. [21] Jou LD, Lee DH, Morsi H, et al. Wall shear stress on ruptured and unruptured intracranial aneurysms at the internal carotid artery. American Journal of Neuroradiology 2008;29:1761–7. [22] Shojima M, Oshima M, Takagi K, et al. Magnitude and role of wall shear stress on cerebral aneurysm: computational fluid dynamic study of 20 middle cerebral artery aneurysms. Stroke 2004;35:2500–5. [23] Goubergrits L, Schaller J, Kertzscher U, et al. Statistical wall shear stress maps of ruptured and unruptured middle cerebral artery aneurysms. Journal of the Royal Society, Interface/the Royal Society 2012;9:677–88. [24] Takao H, Murayama Y, Otsuka S, et al. Hemodynamic differences between unruptured and ruptured intracranial aneurysms during observation. Stroke 2012;43:1436–9. [25] Matsubara S, Hadeishi H, Suzuki A, et al. Incidence and risk factors for the growth of unruptured cerebral aneurysms: observation using serial computerized tomography angiography. Journal of Neurosurgery 2004;101:908–14. [26] International Study of Unruptured Intracranial Aneurysms Investigators. Unruptured intracranial aneurysms – risk of rupture and risks of surgical, intervention. New England Journal of Medicine 1998;339:1725–33. [27] Kataoka K, Taneda M, Asai T, et al. Difference in nature of ruptured and unruptured cerebral aneurysms. Lancet 2000;355:203. [28] Kataoka K, Taneda M, Asai T, et al. Structural fragility and inflammatory response of ruptured cerebral aneurysms. A comparative study between ruptured and unruptured cerebral aneurysms. Stroke 1999;30:1396–401. [29] Chyatte D, Bruno G, Desai S, et al. Inflammation and intracranial aneurysms. Neurosurgery 1999;45:1137–46 [discussion 1146-1137]. [30] Hara A, Yoshimi N, Mori H. Evidence for apoptosis in human intracranial aneurysms. Neurological Research 1998;20:127–30. [31] Nakajima N, Nagahiro S, Sano T, et al. Phenotypic modulation of smooth muscle cells in human cerebral aneurysmal walls. Acta Neuropathologica 2000;100:475–80. [32] Jonasson L, Holm J, Hansson GK. Smooth muscle cells express Ia antigens during arterial response to injury. Laboratory Investigation: A Journal of Technical Methods and Pathology 1988;58:310–5. [33] Frosen J, Piippo A, Paetau A, et al. Remodeling of saccular cerebral artery aneurysm wall is associated with rupture: histological analysis of 24 unruptured and 42 ruptured cases. Stroke 2004;35:2287–93. [34] Pera J, Korostynski M, Krzyszkowski T, et al. Gene expression profiles in human ruptured and unruptured intracranial aneurysms: what is the role of inflammation? Stroke 2010;41:224–31. [35] Yasuno K, Bakircioglu M, Low SK, et al. Common variant near the endothelin receptor type A (EDNRA) gene is associated with intracranial aneurysm risk. Proceedings of the National Academy of Sciences of the United States of America 2011;108:19707–12. [36] Wang X, Douglas SA, Louden C, et al. Expression of endothelin-1, endothelin-3, endothelin-converting enzyme-1, and endothelin-A and endothelin-B receptor mRNA after angioplasty-induced neointimal formation in the rat. Circulation Research 1996;78:322–8. [37] Peerless SJWM, Drake CG. Giant intracranial aneurysms. In: J.R.Y., editor. Neurological surgery. 3rd ed. Philadelphia, PA: WB Sanders; 1990. [38] Choi IS, David C. Giant intracranial aneurysms: development, clinical presentation and treatment. European Journal of Radiology 2003;46:178–94. [39] Yasargil M. Microsurgical anatomy of the basal cisterns and vessels of the brain, diagnostic studies, general operative techniques and pathological considerations of the intracranial aneurysms. New York: Thieme; 1984. [40] Krings T, Piske RL, Lasjaunias PL. Intracranial arterial aneurysm vasculopathies: targeting the outer vessel wall. Neuroradiology 2005;47:931–7. [41] Ouared RCB, Stahl B, Rufenacht D, Yilmaz H, Courbebaisse G. Thrombosis modeling in intracranial aneurysms: a lattice Boltzmann numerical algorithm. Computer Physics Communications 2008:128–31. [42] Nagahiro S, Takada A, Goto S, et al. Thrombosed growing giant aneurysms of the vertebral artery: growth mechanism and management. Journal of Neurosurgery 1995;82:796–801. [43] Nelson PK, Lylyk P, Szikora I, et al. The pipeline embolization device for the intracranial treatment of aneurysms trial. American Journal of Neuroradiology 2011;32:34–40. [44] Pereira VM, Bonnefous O, Ouared R et al. A DSA-Based Method Using ContrastMotion Estimation for the Assessment of the Intra-Aneurysmal Flow Changes Induced by Flow-Diverter Stents. AJNR American Journal of Neuroradiology 2012 Nov 1. [45] Cebral JR, Mut F, Raschi M, et al. Aneurysm rupture following treatment with flow-diverting stents: computational hemodynamics analysis of treatment. American Journal of Neuroradiology 2011;32:27–33. [46] Chow M, McDougall C, O’Kelly C, et al. Delayed spontaneous rupture of a posterior inferior cerebellar artery aneurysm following treatment with flow diversion: a clinicopathologic study. American Journal of Neuroradiology 2011. [47] Lasjaunias PBA, ter Brugges KG. Surgical neuroangiography. Berlin, Germany: Springer; 2004. [48] Zhao WY, Krings T, Alvarez H, et al. Management of spontaneous haemorrhagic intracranial vertebrobasilar dissection: review of 21 consecutive cases. Acta Neurochirurgica 2007;149:585–96 [discussion 596]. [49] Kanai R, Shinoda J, Irie S, et al. A case of embolic stroke imitating atherothrombotic brain infarction before massive hemorrhage from an infectious aneurysm caused by streptococci. Journal of Stroke and Cerebrovascular Diseases: The Official Journal of National Stroke Association 2012;21: 910 e913–6. [50] Regelsberger J, Elsayed A, Matschke J, et al. Diagnostic and therapeutic considerations for “mycotic” cerebral aneurysms: 2 case reports and review of the literature. Central European Neurosurgery 2011;72:138–43. [51] Gross BA, Puri AS. Endovascular treatment of infectious intracranial aneurysms. Neurosurgical Review 2012. [52] Yu W, Rickert K, Flores B, et al. Infectious mid basilar artery aneurysm from Pseudomonas meningitis. Journal of Neurointerventional Surgery 2012;4:e10. [53] Tamuleviciute E, Taeshineetanakul P, Terbrugge K, et al. Myxomatous aneurysms: a case report and literature review. Journal of Interventional Neuroradiology 2011;17:188–94. [54] Maruki C, Suzukawa K, Koike J, et al. Cardiac malignant fibrous histiocytoma metastasizing to the brain: development of multiple neoplastic cerebral aneurysms. Surgical Neurology 1994;41:40–4. [55] Gliemroth J, Nowak G, Kehler U, et al. Neoplastic cerebral aneurysm from metastatic lung adenocarcinoma associated with cerebral thrombosis and recurrent subarachnoid haemorrhage. Journal of Neurology, Neurosurgery, and Psychiatry 1999;66:246–7. [56] Chang IB, Cho BM, Park SH, et al. Metastatic choriocarcinoma with multiple neoplastic intracranial microaneurysms: case report. Journal of Neurosurgery 2008;108:1014–7. [57] Nucifora PG, Dillon WP. MR diagnosis of myxomatous aneurysms: report of two cases. American Journal of Neuroradiology 2001;22:1349–52. [58] Gutierrez J, Sacco RL, Wright CB. Dolichoectasia—an evolving arterial disease. Nature Reviews Neurology 2011;7:41–50. [59] Campos WK, Guasti AA, da Silva BF, et al. Trigeminal neuralgia due to vertebrobasilar dolichoectasia. Case Reports in Neurological Medicine 2012;2012:367304. [60] Siddiqui A, Chew NS, Miszkiel K. Vertebrobasilar dolichoectasia: a rare cause of obstructive hydrocephalus: case report. The British Journal of Radiology 2008;81:e123–6. V.M. Pereira et al. / European Journal of Radiology 82 (2013) 1606–1617 [61] Larson PS, Reisner A, Morassutti DJ, et al. Traumatic intracranial aneurysms. Neurosurgical Focus 2000;8:e4. [62] Kosty J, Pukenas B, Smith M, et al. Iatrogenic vascular complications associated with external ventricular drain placement: a report of eight cases and review of the literature. Neurosurgery 2012. [63] Zanini MA, Tahara A, Santos GS, et al. Pseudoaneurysm of the internal carotid artery presenting with massive (recurrent) epistaxes: a lifethreatening complication of craniofacial trauma. Arquivos de Neuro-psiquiatria 2008;66:268–71. 1617 [64] Paiva WS, Andrade AF, Sterman Neto H, et al. Traumatic pseudoaneurysm of the superior cerebellar artery. The Journal of Trauma and Acute Care Surgery 2012;72:E115. [65] Ogilvy CS, Tawk RG, Mokin M, et al. Stent-assisted coiling treatment of pediatric traumatic pseudoaneurysm resulting from tumor surgery. Pediatric Neurosurgery 2011;47:442–8.