Orientation
Rapid review
What you should already have
Lesson 1 — organisation of the nervous system; The circulation — flow, pressure and resistance.
About 50 minutes
Plus the time it takes to redraw this lesson’s figures from memory, which is the fastest way to find out what you have not understood.
Where this shows up
This is the anatomy the rest of Part II spends its time doing physiology to. The tentorium is why herniation has a direction; the dural sinuses are why air entrains and why a rise in central venous pressure raises intracranial pressure twice over; the circle of Willis is why one carotid can be clamped in some patients and not in others. It is also the shortest route to marks in a viva, because it is entirely nameable.
Learning outcomes
By the end of this lesson you should be able to:
- Name the three meningeal layers, the four dural infoldings and the contents of the extradural, subdural and subarachnoid spaces, and state which vessel bleeds into each.
- Name the four arteries supplying the brain and the proportion of cerebral blood flow each pair carries.
- Draw the circle of Willis with all nine named vessels, and mark which belong to the anterior and which to the posterior circulation.
- State the territory of the anterior, middle and posterior cerebral arteries, and predict the deficit produced by occlusion of each.
- Trace venous blood from the cerebral cortex and from the deep structures to the internal jugular vein, naming every sinus on the way.
- Explain why the dural venous sinuses are valveless and non-collapsible, and give two anaesthetic consequences of that.
Together these settle one syllabus objective: Cerebral vascular anatomy and the venous drainage. Tick it on the Physiology objective list once you can do all of the above without notes.
The minimum high-yield framework
- Three meninges. Dura mater (two layers — periosteal and meningeal), arachnoid mater, pia mater.
- Four dural infoldings. Falx cerebri, tentorium cerebelli, falx cerebelli, diaphragma sellae. Where the two dural layers separate, they enclose a venous sinus.
- Three spaces, three bleeds. Extradural — middle meningeal artery. Subdural — bridging veins. Subarachnoid — aneurysm of the circle of Willis.
- Four arteries in. Two internal carotids carrying about two-thirds of cerebral blood flow, and two vertebral arteries carrying about one-third.
- The circle of Willis links three arterial stems — the two internal carotids and the basilar — through one anterior and two posterior communicating arteries. Anterior circulation: ACA, ACom, MCA. Posterior circulation: basilar, PCA, PCom.
- Three territories. ACA medial and superior; MCA lateral, plus much of the internal capsule; PCA occipital and medial temporal.
- The circle is often incomplete. That, not the theory of collateral supply, is why carotid disease causes stroke.
- Two venous systems, one exit. Superficial cortical veins into the dural sinuses; deep structures by the internal cerebral veins into the great cerebral vein of Galen. Both reach the transverse, then sigmoid, then internal jugular veins.
- The sinuses are dural, valveless and non-collapsible. That single sentence carries venous air embolism, the effect of head-up position, and the transmission of central venous pressure into the cranium.
The container
The cranial cavity and the folds that divide it
The adult cranium is a closed, non-distensible container with one large opening, the foramen magnum, and a number of small ones. That rigidity is the whole basis of the Monro-Kellie doctrine and of the volume-pressure relationship taught in lesson 9. What is less often said, and is worth saying first, is that the cavity is not a single undivided space. The dura throws four rigid infoldings inward, and these divide the cranial cavity into communicating compartments.
This matters because pressure within the cranium is not uniform once a mass lesion develops. A pressure gradient can exist between compartments, and brain tissue is displaced down that gradient — under the falx, through the tentorial hiatus, or through the foramen magnum. The dural folds are therefore not anatomical trivia; they are the boundaries that make herniation directional and predictable, and the reason a single intracranial pressure reading from one compartment can be reassuring while another compartment is failing.
| Fold | Plane | What it separates | Why it matters |
|---|---|---|---|
| Falx cerebri | Vertical, in the midline sagittal plane | Separates the two cerebral hemispheres | Its free lower edge is what the cingulate gyrus is pushed beneath in a subfalcine shift. The superior sagittal sinus lies in its attached upper edge and the inferior sagittal sinus in its free lower edge. |
| Tentorium cerebelli | Horizontal, roofing the posterior fossa | Separates the occipital lobes above from the cerebellum below | It is the partition that makes the words supratentorial and infratentorial mean anything. Its opening, the tentorial hiatus, transmits the brainstem, and is the aperture through which transtentorial herniation occurs. The straight sinus runs within it. |
| Falx cerebelli | Vertical, below the tentorium | Separates the two cerebellar hemispheres | Small, and rarely of consequence in itself. |
| Diaphragma sellae | Horizontal, a circular fold | Roofs the sella turcica and envelops the pituitary gland | The pituitary stalk passes through its central aperture, which is why a pituitary mass expands upward against the optic chiasm rather than in every direction equally. |
The coverings
The meninges and the three spaces
The meninges are the membranes covering the brain and spinal cord. From outside inward:
- Dura mater — a thick, tough outer membrane of two layers. The outer periosteal layer is applied to and adherent to the inner table of the skull, most firmly at the sutures. The inner meningeal layer is the dura proper. The two are fused except in two situations: where they separate to enclose a dural venous sinus, and where the meningeal layer folds inward to form the four partitions of the previous section. Below the foramen magnum the periosteal layer stops, and only the meningeal layer continues as the spinal dura — which is why an extradural space exists in the spine but only potentially in the cranium.
- Arachnoid mater — a thin, avascular membrane of spider-web appearance, applied to the inner surface of the dura. It bridges across the sulci rather than dipping into them.
- Pia mater — very thin and highly vascular, adherent to the surface of the brain and following every gyrus and sulcus. It continues below the end of the spinal cord as the filum terminale, which with the dura tethers the cord to the coccyx.
The meninges, the three spaces, and the vessel that bleeds into each
Three layers. The dura mater is the thick outer membrane and has two layers — an outer periosteal layer applied to the skull and an inner meningeal layer. Where the two separate they enclose a dural venous sinus, drawn here at the vertex. The arachnoid mater is a thin membrane of spider-web appearance. The pia mater is very thin, highly vascular, and adheres to the surface of the brain, following every gyrus and sulcus.
Three spaces, and this is the part that is examined. The extradural space is between skull and dura and is where the middle meningeal artery bleeds — arterial, so the collection is fast and lens-shaped. The subdural space, between dura and arachnoid, is a potential space: it is opened only when a bridging vein tears, which is venous and therefore slow. The subarachnoid space, between arachnoid and pia, normally contains CSF, and is where a ruptured aneurysm of the circle of Willis bleeds. Note that the pia follows the cortical surface into the sulci while the arachnoid bridges across them — which is why the subarachnoid space is a real space with a real volume and the subdural space is not.
| Space | Boundaries | Normal contents | Vessel that bleeds | Behaviour of the collection |
|---|---|---|---|---|
| Extradural (epidural) | Between the skull and the periosteal layer of dura | A potential space only — the periosteal dura is adherent to bone, most firmly at the sutures | Middle meningeal artery | Arterial pressure, so the collection accumulates quickly and is limited by the sutures, giving the classical biconvex shape |
| Subdural | Between the meningeal dura and the arachnoid | A potential space — the two membranes are normally in contact | Bridging veins crossing from cortex to sinus | Venous pressure, so accumulation is slow, and there is no sutural limit, giving a collection that spreads over the convexity |
| Subarachnoid | Between the arachnoid and the pia | A real space, containing CSF and the cerebral vessels, expanded into cisterns at the base | Vessels of the circle of Willis, usually an aneurysm at a branch point | Blood mixes with CSF and spreads through the whole space, which is why the presentation is diffuse rather than focal |
Two consequences follow directly from the geometry, and both are examinable. First, because the arachnoid bridges over the sulci while the pia dips into them, the subarachnoid space is a real space with a real volume — expanded at the skull base into the cisterns, which is where the cerebral vessels and the cranial nerves actually run. The subdural space, by contrast, is only potential. Second, the bridging veins have to cross that potential space to reach a sinus, without support and without a surrounding cuff of tissue. In the elderly, cerebral atrophy increases the distance they must span and the amplitude of brain movement within the skull, so a trivial deceleration tears them. That is the anatomy behind chronic subdural haematoma after a fall that the patient does not remember.
The supply
Four arteries in
The entire arterial supply of the brain arrives through four vessels: the right and left internal carotid arteries, which together carry about two-thirds of cerebral blood flow, and the right and left vertebral arteries, which carry about one-third. That asymmetry is worth stating explicitly in an answer, because it is the reason carotid disease dominates the clinical picture while vertebrobasilar disease produces a smaller number of much more dangerous syndromes.
| Vessel | Paired? | Share of flow | Principal branches | Note |
|---|---|---|---|---|
| Internal carotid artery | Paired | about two-thirds of cerebral blood flow | Anterior cerebral, middle cerebral, posterior communicating, ophthalmic | Forms the anterior circulation. Its termination is the branch point where the anterior and middle cerebral arteries arise. |
| Vertebral artery | Paired | about one-third of cerebral blood flow | Posterior inferior cerebellar; the two vertebrals then unite as the basilar; contributions to the anterior spinal artery | Forms the posterior circulation. It also supplies the upper cord — the anterior spinal artery of the cervical cord is formed from branches of both vertebrals. |
| Basilar artery | Single | Carries the whole vertebral contribution | Anterior inferior cerebellar, superior cerebellar, pontine perforators; divides into the two posterior cerebral arteries | Its perforating branches supply the brainstem, which is why a basilar occlusion is catastrophic out of proportion to the volume of tissue infarcted. |
The vertebral arteries ascend through the transverse foramina of the cervical vertebrae and enter the skull through the foramen magnum. Each gives off a posterior inferior cerebellar artery before the two unite at the pontomedullary junction to form the single basilar artery. The basilar gives off the anterior inferior cerebellar and superior cerebellar arteries and a series of small pontine perforating branches to the brainstem, then divides at its tip into the two posterior cerebral arteries.
The internal carotid artery enters through the carotid canal, traverses the cavernous sinus, and terminates by dividing into the middle cerebral artery, which continues the line of the parent vessel laterally, and the anterior cerebral artery, which turns medially. It gives off the posterior communicating artery before that division, and the ophthalmic artery as it emerges from the cavernous sinus — a branch worth remembering, because it is the anatomical basis of amaurosis fugax as a warning sign of carotid disease, and because it makes the retinal circulation a window onto the cerebral one.
The anastomosis
The circle of Willis
The unique feature of the cerebral circulation is that its four feeding vessels are joined into a ring at the base of the brain. The circle of Willis is formed by three arterial stems — the two internal carotids and the basilar — linked by three communicating vessels: a single anterior communicating artery joining the two anterior cerebral arteries across the midline, and paired posterior communicating arteries, each joining the internal carotid of its own side to the posterior cerebral artery of that side.
The circle of Willis
Four vessels in, nine vessels named. The brain is supplied by two internal carotid arteries, carrying about two-thirds of cerebral blood flow, and two vertebral arteries carrying about one-third. The vertebral arteries give off the posterior inferior cerebellar arteries before joining to form the basilar artery, which gives off the anterior inferior cerebellar and superior cerebellar arteries and then divides into the two posterior cerebral arteries. Each internal carotid gives off a middle cerebral artery, which continues its line laterally, and an anterior cerebral artery medially.
What closes the ring. A single anterior communicating artery joins the two anterior cerebral arteries; paired posterior communicating arteries join each internal carotid to the posterior cerebral artery of the same side. So three arterial stems — the two carotids and the basilar — are linked by three communicating vessels. Everything fed by the carotids (ACA, ACom, MCA) is the anterior circulation; everything fed by the vertebrals (basilar, PCA, PCom) is the posterior circulation.
The caveat that carries the marks. In theory the circle is a complete collateral pathway, and where it is complete three of the four feeding vessels can be occluded gradually without infarction. In practice it is anatomically incomplete or hypoplastic in a large minority of people, which is why carotid occlusion causes stroke at all, and why the back-pressure measured at the carotid stump during endarterectomy — collateral flow arriving through the circle from the other carotid and from the vertebrobasilar system — is so variable between patients.
The division into two circulations follows from which stem feeds what. The anterior circulation is everything supplied through the internal carotids: the anterior cerebral arteries, the anterior communicating artery and the middle cerebral arteries. The posterior circulation is everything supplied through the vertebral arteries: the basilar, the posterior cerebral arteries and the posterior communicating arteries. The posterior communicating artery is the hinge between the two, and it is the vessel by which the vertebrobasilar system can supply a hemisphere when a carotid fails.
This is also what is being measured when the internal carotid stump pressure is recorded during carotid endarterectomy. With the artery cross-clamped, the pressure in the distal stump is the back-pressure produced by collateral flow arriving through the circle — from the contralateral carotid and from the vertebrobasilar system. A high stump pressure means the circle is doing its job in that patient; a low one means it is not, and that a shunt is needed. The technique is easy to perform and correspondingly poor as a predictor, precisely because it is reporting an anatomy that varies from person to person.
What each vessel owns
Arterial territories, and the deficit each loss produces
Which artery supplies which surface, and what its loss costs
The three cerebral arteries divide the hemisphere between them along lines that are worth knowing as a rule rather than as a picture. The middle cerebral artery takes the lateral surface — most of what you see looking at the brain from the side — and, critically, much of the internal capsule, where a large number of cortical afferent and efferent fibres are gathered into a small volume. That concentration is why it has been called the artery of cerebral haemorrhage, and why a lacunar infarct in the posterior limb of the internal capsule produces a hemiparesis out of all proportion to its size. The anterior cerebral artery takes the superior and medial portions of the hemisphere, and the posterior cerebral artery the occipital lobe and the medial temporal lobe.
The band widths below indicate extent, not measured area — no source in the library states a numeric territory area, so none is claimed here.
| Artery | Territory | Deficit produced by occlusion |
|---|---|---|
| Anterior cerebral | Superior and medial portions of the cerebral hemisphere, including the leg area of the motor and sensory cortex | Contralateral weakness affecting the leg more than the arm; there may be personality change and incontinence from medial frontal involvement |
| Middle cerebral | Lateral aspect of the cerebral hemisphere, and much of the internal capsule | Contralateral weakness affecting the face and arm more than the leg; a homonymous hemianopia if the optic radiation is involved; dysphasia if the dominant hemisphere is affected, or visuospatial neglect if the non-dominant is |
| Posterior cerebral | Occipital lobe and the medial portion of the temporal lobe | Homonymous hemianopia, characteristically with macular sparing; memory disturbance if the medial temporal lobe is involved |
The pattern that makes these easy to reconstruct rather than memorise is the homunculus. The motor and sensory strips run over the top of the hemisphere and down its lateral surface, with the leg represented medially at the top and the face laterally at the bottom. The anterior cerebral artery supplies the medial surface, so it takes the leg. The middle cerebral artery supplies the lateral surface, so it takes the face and arm. That single fact generates the two commonest stroke patterns without any further learning.
The middle cerebral artery deserves its own sentence. Beyond the cortical surface it supplies much of the internal capsule, where the corticospinal, corticobulbar and thalamocortical fibres are gathered from a whole hemisphere into a structure a few millimetres across. A very small infarct there produces a dense hemiparesis, which is why lacunar strokes are so disproportionate to their size, and it is the reason the vessel has been described as the artery of cerebral haemorrhage.
| Category | Vessel affected | Clinical features |
|---|---|---|
| Anterior circulation infarct | Thrombus or embolus in the anterior circulation, most often the middle cerebral artery | Three categories together: contralateral weakness from motor cortex involvement, homonymous hemianopia from optic tract involvement, and higher cerebral dysfunction — dysphasia or visuospatial disorder according to hemisphere |
| Posterior circulation infarct | Thrombus or embolus in the vertebrobasilar circulation | Cerebellar dysfunction with ataxia and nystagmus; cranial nerve palsies or loss of consciousness from brainstem involvement; homonymous hemianopia from occipital involvement |
| Lacunar infarct | Occlusion of a single small subcortical vessel | A deficit out of proportion to the size of the infarct — a pure motor hemiparesis, for instance — because the lesion sits where fibres are densely packed, such as the lateral thalamus or the posterior limb of the internal capsule. Higher cerebral function is spared, and the prognosis is correspondingly better. |
The drainage
Venous drainage: two systems, one exit
The cerebral venous system does not accompany the arteries. It divides instead into a superficial system draining the cortex and a deep system draining the central structures, and both converge on the same exit.
Venous drainage of the brain: two systems, one exit
The superficial system drains the cerebral and cerebellar cortex through thin-walled cortical veins into the dural venous sinuses — channels lying between the two layers of dura. The superior sagittal sinus runs in the attached edge of the falx cerebri and drains usually into the right transverse sinus. The deep system drains the thalamus, basal ganglia and choroid plexus through paired internal cerebral veins, which unite as the great cerebral vein of Galen; this joins the inferior sagittal sinus, running in the free edge of the falx, and the two continue as the straight sinus within the tentorium cerebelli, draining usually into the left transverse sinus. Both systems then follow the same exit: transverse sinus, sigmoid sinus, and out of the skull as the internal jugular vein. The cavernous sinuses, one either side of the pituitary fossa, drain eventually into the transverse sinuses.
Three properties of the sinuses, and why each matters. They are formed from dura, not from vein wall; they are valveless; and they are non-collapsible, because the dura holds them open. A vein that cannot collapse cannot close when the pressure inside it falls below atmospheric — which is the anatomical reason venous air embolism is a risk whenever the operative site is above the level of the heart. The same rigidity means sinus pressure is transmitted almost directly to CSF absorption at the arachnoid granulations, so a rise in central venous pressure raises intracranial pressure by two routes at once.
The superficial system collects from the cerebral and cerebellar cortices into thin-walled cortical veins, which cross the subdural plane and empty into the dural venous sinuses. The deep system drains the thalamus, basal ganglia and choroid plexus through paired internal cerebral veins, which unite to form the great cerebral vein of Galen. That vein and the inferior sagittal sinus meet, and the two continue as the straight sinus. From there both systems follow one path out: transverse sinus, sigmoid sinus, and through the jugular foramen as the internal jugular vein.
| Sinus | Location | Receives | Drains to |
|---|---|---|---|
| Superior sagittal | Attached (upper) edge of the falx cerebri | Cortical veins of the superolateral surface; CSF through the arachnoid granulations | Usually the right transverse sinus |
| Inferior sagittal | Free (lower) edge of the falx cerebri | Medial surface of the hemispheres | Joins the great cerebral vein; the two continue as the straight sinus |
| Straight | Within the tentorium cerebelli, in its junction with the falx | The inferior sagittal sinus and the great cerebral vein of Galen | Usually the left transverse sinus |
| Transverse | Attached margin of the tentorium cerebelli, paired | Superior sagittal and straight sinuses; the cavernous sinuses | The sigmoid sinus of the same side |
| Sigmoid | An S-shaped groove in the posterior fossa, paired | The transverse sinus | Leaves the skull at the jugular foramen as the internal jugular vein |
| Cavernous | Either side of the pituitary fossa, paired | Ophthalmic veins and superficial middle cerebral vein | Eventually the transverse sinuses, by the petrosal sinuses |
There is a second consequence of the same rigidity that connects this lesson directly to CSF physiology. Absorption of CSF through the arachnoid granulations is pressure-dependent bulk flow into the superior sagittal sinus, and it requires CSF pressure to exceed sinus pressure. Because the sinus cannot collapse, its pressure tracks central venous pressure closely. So a rise in central venous pressure raises intracranial pressure by two mechanisms at once: it impedes venous outflow and therefore raises cerebral blood volume, and it closes the gradient for CSF absorption. The consequences of that are followed through in lesson 7 and lesson 9.
Application
Where this anatomy changes what you do
Positioning
The head-up position lowers intracranial pressure by improving venous drainage, and this is one of the few interventions that lowers pressure without any cost to metabolic rate. The same position, by making cerebral venous pressure subatmospheric in non-collapsible sinuses, creates the conditions for venous air embolism. The two are the same piece of anatomy read in two directions, and the sitting craniotomy is where they meet. Note also that a neutral head position matters as much as elevation: rotation or flexion of the neck kinks the internal jugular vein, and a tight collar or tracheal tube tie compresses it, and either abolishes the benefit of head-up positioning entirely.
Carotid endarterectomy
Cross-clamping the internal carotid is tolerated only to the extent that the circle of Willis delivers collateral flow, and that is an individual anatomical question rather than a physiological one. This is why cerebral monitoring is used during the clamp period at all — stump pressure, transcranial Doppler of the middle cerebral artery, electroencephalography or near-infrared spectroscopy — and why no single one of them is a reliable predictor.
The transcranial Doppler window
Transcranial Doppler is described in lesson 8 as a measure of flow velocity. The reason the middle cerebral artery through the temporal window is almost always the vessel chosen is anatomical: it is large, it runs in a predictable direction almost directly toward or away from the probe, it carries a high proportion of hemispheric flow, and the squamous temporal bone is thin enough to transmit ultrasound.
Intracranial haemorrhage after trauma
The three spaces produce three different clinical courses from the same mechanism. Arterial bleeding into the extradural space is rapid and is classically preceded by a lucid interval. Venous bleeding from torn bridging veins into the subdural space is slow, and in an atrophic brain may present weeks later with confusion rather than acutely. Subarachnoid bleeding from an aneurysm of the circle of Willis is neither, and presents as sudden severe headache with meningism because blood is in contact with the whole subarachnoid space at once.
Sinus and cortical vein thrombosis
Because the sinuses are a low-pressure, valveless, dural-walled system, they thrombose in states of stasis and hypercoagulability. Cavernous sinus thrombosis, sagittal sinus thrombosis and cortical vein thrombosis all occur, and cortical venous thrombosis has a recognised association with pregnancy. It presents as headache, seizures or a focal deficit that does not respect an arterial territory — which is the clue, and which is only recognisable if you know that arterial territories exist and what their boundaries are.
Consolidation
The lesson in one paragraph
The brain is covered by three meninges. The dura mater is a thick outer membrane of an outer periosteal and an inner meningeal layer, which separate to enclose the dural venous sinuses and fold inward as four partitions: the falx cerebri between the hemispheres, the tentorium cerebelli between the occipital lobes and the cerebellum, the falx cerebelli between the cerebellar hemispheres, and the diaphragma sellae around the pituitary. The arachnoid mater is thin and avascular and bridges the sulci; the pia mater is thin, vascular and adherent to the brain surface, following every sulcus. Between skull and dura is the potential extradural space, where the middle meningeal artery bleeds; between dura and arachnoid the potential subdural space, where bridging veins tear; between arachnoid and pia the true subarachnoid space, which contains CSF and the cerebral vessels and into which aneurysms of the circle of Willis rupture.
Arterial supply comes from four vessels: the two internal carotid arteries carrying about two-thirds of cerebral blood flow, and the two vertebral arteries carrying about one-third. Each vertebral gives a posterior inferior cerebellar artery before the two unite as the basilar, which gives the anterior inferior cerebellar and superior cerebellar arteries and pontine perforators before dividing into the two posterior cerebral arteries. Each internal carotid gives the posterior communicating artery and then divides into the middle cerebral artery laterally and the anterior cerebral artery medially. The circle of Willis is the ring formed by three arterial stems — the two carotids and the basilar — linked by one anterior communicating artery between the anterior cerebral arteries and two posterior communicating arteries joining each carotid to the posterior cerebral artery of its side. The anterior circulation is the anterior cerebral, anterior communicating and middle cerebral arteries; the posterior circulation is the basilar, posterior cerebral and posterior communicating arteries. The anterior cerebral artery supplies the superior and medial hemisphere, the middle cerebral the lateral hemisphere and much of the internal capsule, and the posterior cerebral the occipital lobe and medial temporal lobe. The circle is a theoretical collateral pathway that is frequently anatomically incomplete, which is why carotid disease causes stroke and why carotid stump pressure predicts clamp tolerance so poorly.
Venous drainage is superficial and deep. Superficially, cortical veins cross the subdural plane into the dural venous sinuses; the superior sagittal sinus lies in the attached edge of the falx and drains usually into the right transverse sinus. The deep system drains the thalamus, basal ganglia and choroid plexus by the paired internal cerebral veins into the great cerebral vein of Galen, which meets the inferior sagittal sinus of the free edge of the falx, the two continuing as the straight sinus within the tentorium and draining usually into the left transverse sinus. Both systems then pass through the transverse and sigmoid sinuses to leave the skull as the internal jugular veins. The cavernous sinuses lie either side of the pituitary fossa and drain eventually to the transverse sinuses. The sinuses are formed from dura, are valveless and are non-collapsible: therefore pressure downstream is transmitted directly into the cranium, raising intracranial pressure both by impeding venous outflow and by closing the gradient for CSF absorption; and therefore a sinus opened above the level of the right atrium entrains air rather than bleeding, which is the anatomical basis of venous air embolism.