Intracranial collateral pathways and anastomosis
Collateral circulation can compensate for obstruction of arteries or veins in the brain. Descriptions of collateral circulation date back to the dawn of neurology. Centuries after Thomas Willis described arterial collateral circulation and its role in disease, angiography revealed these collateral pathways. Subsequent hemodynamic studies also highlighted the crucial role of collateral circulation; however, with advancements in imaging technology, its connection to angiography, neuroprotection, and tissue ischemia has been largely overlooked. In clinical practice of interventional neuroradiotherapy, intracranial arteriovenous collateral circulation can influence the diagnosis, treatment, and prevention of various cerebrovascular diseases.
Knowledge of the anatomy and pathophysiology of collateral circulation can deepen our understanding of a wide range of diseases. Imaging features and angiography can assist in disease diagnosis and broaden our perspectives in new treatment approaches. In this chapter, we will introduce the latest research progress in arteriovenous collateral circulation and focus on its specific roles in various diseases.
Anatomy
Intravascular flow between different regions is primarily influenced by the anatomical structure of intracranial collateral circulation, with vessel diameter being the most crucial factor. The anatomy of cerebral collateral circulation can be subdivided into two types: Willisian collateral circulation located at the circle of Willis or leptomeningeal anastomoses, and atypical collateral circulation occurring in specific lesions. Similarly, cerebral venous collateral circulation can be divided into typical anastomotic pathways and complex, variable atypical anastomotic pathways. Some anastomotic pathways (such as the posterior communicating artery) are embryonic remnants, while others are only found in lesions. Many other intracranial and extracranial arteriovenous shunts exist, which will not be elaborated upon here. Records indicate that anatomical knowledge of the main trunks of intracranial collateral circulation has existed for over a century. Only in recent years, with the development of angiography and modern imaging techniques, have the related functions of these blood flow pathways been confirmed. It has been reported that collateral pathways can be divided into two functional levels (primary and secondary) based on anatomical location; however, this classification method is overly simplistic due to the potential for significant variation. In humans, knowledge of the anatomy of intracranial collaterals is crucial for understanding ischemic stroke and other cerebrovascular diseases, and animal studies have shown that numerous anatomical variations can hinder the successful implementation of treatments. Different species, individuals, and populations exhibit different collateral circulation structures.
The Circle of Willis provides numerous pathways for blood flow distribution. The segments of the Willisian collateral circulation (including the anterior communicating artery segment, the proximal segment of the anterior cerebral artery, the posterior communicating artery segment, and the proximal segment of the posterior communicating artery) can aid in blood flow distribution based on intraluminal pressure gradients. These segments may be atretic or underdeveloped, but they still possess a significant ability to increase cerebral vascular capacity. The size and wall characteristics of these arterial segments are remarkably similar to their generative arteries. This structural feature contributes to the formation of collateral circulation along the longitudinal fissure of the brain and compensates for the pressure gradient between the anterior and posterior circulations. The role of the posterior communicating artery has been emphasized by many. Various terms, including permanent posterior communicating artery or fetal posterior communicating artery, have been used to describe the morphology of these arteries. However, to truly distinguish between different morphologies, it is necessary to rely on endoscopic diameter measurements or imaging studies such as MRA. Furthermore, the morphology and primary function of these vessels are inextricably linked to the proximal posterior cerebral artery. There are also some contrasting cases, such as the posterior communicating artery originating from the internal carotid artery in the embryonic period, which can develop into an malformed posterior communicating artery that, even with a small diameter, can provide sufficient blood flow. Currently, based on our understanding of the anatomical structure, the arteries of the Circle of Willis have been classified, but these classifications are not entirely accurate, as the anatomical structure varies depending on disease, age, and population.
The leptomeningeal anastomoses connecting the distal ends of the major cerebral arteries are small arteries, ranging from 50 to 400 μm in diameter, and provide retrograde blood flow to adjacent areas . These anastomoses exhibit morphological diversity, including end-to-end anastomoses, end-to-side anastomoses, and shunt variations. The connection between the middle cerebral artery and the anterior and posterior cerebral arteries is achieved through these small arterial anastomoses. Anastomoses originating from the anterior cerebral artery supply the distribution area of the superior branch of the middle cerebral artery, while the distribution area of the inferior branch is supplied by collateral shunts from the posterior cerebral artery. Anastomoses between the anterior and posterior cerebral arteries are relatively rare. Following a groundbreaking study of 20 cadavers, Vander Eecken and Adams described the main characteristics of leptomeningeal collateral anastomoses and clarified that they exhibit significant variation in size, number, and location. Such a large amount of variation is likely to affect our past learnings of many specific anatomical structures, and similarly impacts the interpretation of many controversial issues in studies of age-related collateral circulation function. Anastomosing vessels are also concentrated in the cerebellar cortex, such as at the confluence of the posterior inferior cerebellar arteries (PICAs), anterior inferior cerebellar arteries, and superior cerebellar arteries (SCAs) . During PICA occlusion, the posterior meningeal arteries also participate in the anastomosis. These anastomoses, which facilitate the establishment of collateral circulation between cerebral hemispheres and their distribution proximal and distal to the basal fossa, often benefit from the symmetry of the posterior fossa anatomy.
Atypical collateral pathways can develop when blood flow demand and pressure gradients exceed the capacity of the main arteries, Willisian collateral circulation, or pia mater collateral circulation. Some collateral pathways may utilize smaller pathways with normal variations, such as asymmetric vascular connections between the ACAs. Blood flow from the anterior and posterior choroidal arteries can exist in either direction of the anterior or posterior cerebral circulation. In patients with moyamoya disease, this network of choroidal collateral circulation is commonly formed, with other small arteries intruding into subcortical structures and meandering around the occluded middle cerebral artery. At the margins, anastomoses may shunt between the posterior cerebral artery and the superior cerebellar artery. More atypical collateral pathways may also occur, often accompanied by significant EI-CI collaterals. Atypical collaterals can appear in any form and are limited to some extent only by physical barriers, such as the falx cerebri or tentorium cerebelli.
The anatomy of cerebral venous circulation can be better understood through typical venous return patterns. Venous return is balanced by the superficial/deep venous system, allowing the transverse cerebral veins to shunt in any direction. In the superficial venous system of the brain (including the cortical veins and superior sagittal sinus), the typical pattern of return is that most of the refluxed blood empties along the right transverse and sigmoid sinuses and flows into the jugular venous system. If the jugular vein is occluded, most of the blood flow returns through the vertebral venous plexus. Anastomotic veins of the Trolard and Labbe shunts cross the cerebral hemispheres and return to the sub-pressure area. Similarly, the direction of blood flow is altered between cortical veins through vascular anastomoses. The deep venous system primarily collects venous return from the choroid plexus and thalamus, striatum, periventricular white matter, limbic system, and medulla oblongata; higher-level veins in this system include the basal vein, great cerebral vein, and straight sinus. The probing venous system returns to the left transverse sinus system via the straight sinus or flows anteriorly into the basal vein. The abundant venous anastomoses on the base of the brain facilitate the return of blood from the deep venous system; the deep middle cerebral vein, inferior and superior petrosal veins, and basal venous plexus are shunted through these anastomoses. Due to the diversity of venous outflow patterns and potential vascular anastomoses, some venous hypertension can be relieved. However, we rarely pay attention to the systemic features of the anatomy of venous collateral circulation.
Epidemiology
Some literature occasionally mentions epidemiological studies of intracranial collateral circulation, introduces some anatomical variations of the Willisian circulation, and cites some still-developing theories, such as the relationship between different populations (e.g., the elderly) and collateral circulation formation. Most of the anatomical data in these studies comes from autopsies. Unfortunately, for living individuals with altered collateral hemodynamics, this use of autopsy data to describe potential collateral circulation function is of little significance. Studies of Willisian circulation architecture in healthy individuals are also limited because the role of collateral circulation is not readily apparent in healthy individuals. In specific clinical conditions, such as acute ischemic stroke, some functional analyses, including collateral angiography, can provide more detailed information (Figure 5); however, we still do not know a range of changes and the formation and development of collateral circulation. Other literature reports some speculations about the relationship between collateral circulation in the coronary or peripheral arteries and cerebral circulation, but these have not been confirmed. Many scholars have explored the effects of age on collateral circulation blood flow; however, intracranial collateral circulation may exhibit significant variations with age. For arterial collateral blood flow via the Willisian circulation and the leptomeningeal pathway, even in some specific diseases, the relevant epidemiological data are still scarce, and the epidemiological characteristics of venous collateral circulation are often unknown.

