Osteopathy Journals and Research by Darren Chandler


 RSS Feed

  1. Introduction

    Movement of CSF contained within the dural (thecal) sac in the subarachnoid space is determined by:

    • Compliance of the dura mater as it forms the outer covering of the dural (thecal) sac. This is determined by the venous distension of the surrounding verterbal venous plexus.
    • A decrease in volume and pressure of CSF in the surrounding subarachnoid space --> decreases pressure in central canal (syrinx can develop) and decreases venous blood flow and volume in the epidural vertebral veins (as they aren't being compressed by filling of the venous plexus) --> reduces dural compliance.
    • Cross sectional area of the subarachnoid space.
    • Movement of the brain tissue.
    • Pressure gradients i.e. respiration and body movement.
    • Arterial pressure in systole and venous pressure in diastole.

    This movement of CSF in the dural (thecal) sac is (i) external to the spinal cord in the subarachnoid space. This movement is characterised by CSF moving cephalic and caudad as well as a complex mixing pattern; (ii) internal from the subarachnoid space within the spinal cord along perivascular pathways mixing with local interstitial fluid.

    The systems determining local movement of CSF is related to systemic body systems and mechanics. This relationship is explored from the perspective of:

    • The movement of blood between the systemic venous system and the vertebral venous system; as well as the movement of blood between the vertebral venous system and the dural venous sinus. It is the vertebral venous plexus that determines the compliance of the dural (thecal) sac at a segmental level.
    • The effects of respiration on CSF flow and movement of venous blood. 
    • Effects of brain mobility on CSF flow determined by systole/diastole and cervical spine movement.
    • Soft tightness transmitted to the dura mater in the upper cervical spine by myodural bridges, epidural ligaments (of Hoffman), meningodural ligaments and the first (intracranial) denticulate ligament.
    • Cross sectional area of the upper spinal canal-foramen magnum determined by local anatomical features and physiological influences.

    A review is carried out of the anatomy and physiology of the CNS including updates on CSF production, absorbtion and transportation.

    Dural anatomy

    From superificial to deep ...

    Epidural space

    The epidural space lies between the bone (tissues that line the vertebral canal) and dura mater. Its borders are:

    • Superior border: sealed by fusion of the dura mater to the foramen magnum.
    • Inferior border: sacrococcygeal ligament that closes the sacral hiatus.
    • Anterior border: posterior longitudinal ligament, posterior margins of vertebral bodies and intervertebral discs.
    • Posterior border: ligamentum flavum, facet joint capsules and the laminae.
    • Lateral border: the pedicles and intervertebral foramina.

    The epidural space is a real space at the intervertebral level, however, it is a potential space at the vertebral level as the dura fuses with the posterior longitudinal and annular ligament. 

    Its contents are:

    • Posterior longitudinal ligament.

    C2 body (continuous with the tectorial membrane --> sacrum.

    Attaches to the posterior aspect of the intervertebral discs and to the adjacent margins of the vertebral bodies. Primarily to the annulus fibrosis. It is separated from the vertebral body by the basivertebral veins, and the paravertebral venous plexus (--> internal vertebral venous plexus).

    It is attached intimately to the dura mater and the the connective tissue accompanying the dural extensions into the intervertebral foramina (Lubeck et al 1992).

    Common attachment of the posterior longitudinal ligament covered almost all of the lower disc area, but some of the upper lateral disc area was not covered by the posterior longitudinal ligament. Lee et al (2018) divided the posterior longitudinal ligament into superfcial and deep layers:

    Superficial layers: extends between three and four vertebra. Consists of (i) central portion of a strong midline strap of longitudinal fibers. These fibers decrease in width from L1-5. Restrains disc herniation. (ii) Transverse fan like portion which attaches to the disc is more membranous. 

    Deep layers: extend between one or two vertebræ. Deep layer is difficult to distinguish from the dura mater.

    The meningovertebral ligaments extend mainly from the ligamentum nuchae but also the laminae to the dura (thecal) sac. In the lumbar spine these ligaments additionally fix the dural (thecal) sac to the posterior longitudinal ligament.

    • Loosely packed connective tissue.

    This connective tissue extends for a short distance through the intervertebral foramina along the sheaths of the spinal nerves. These nerve root sheaths are partially tethered to the walls of the foramina by fine meningovertebral ligaments.

    In the lumbar region connective tissue in the epidural space permits movement of the dural (thecal) sac (Standring 2015).

    • Internal vertebral venous plexus.

    The internal vertebral venous plexus is part of the vertebral venous plexus. It lies between the dura mater and the vertebrae/posterior longitdinal ligament in the epidural space.

    The internal vertebral venous plexus consists of the anterior and posterior plexi. The anterior plexus is more pronounced than the posterior especially at C1 and C2 (Ruiz et al 2002)

    The veins of this plexus communicate with the dural sinuses and the veins of the systemic circulation. Like the veins of the dural sinus they are valveless allowing for an ebb and flow between these two continuous systems.

    Ruiz et al (2002) found at the craniocervical junction the internal vertebral venous plexus communications with the venous sinuses in the posterior cranial fossa e.g. the marginal and basilar sinus via the:

    • Vertebral artery venous plexus (suboccipital cavernous sinus): at the O/A space.
    • Condylar veins: posterior condylar emissary vein and anterior condylar (venous plexus of the hypoglossal canal), lateral condylar vein.
    • Occipital and mastoid emissary veins.
    • Anterior condylar confluence lies at the base of the skull external to the hypoglossal canal. It communicates with the dural sinuses of the posterior cranial fossa and internal jugular vein. 

    Martins et al (1972) found the filling of blood in this venous plexus restricted movement of the mobile spinal dura mater. 

    As the veins fill with blood they swell placing a tighter 'wedge' between the vertebrae and the dura restricting its movement. Consequently the emptying of blood from these veins places a smaller 'wedge' between the vertebrae and the dura allowing increased dural flexibility.

    To illustrate this these authors found a partial collapse of the thoracolumbar dural (thecal) sac as it expelled its CSF during valsalva.  They hypothesised the rise in pressure within the thorax and abdomen caused venous blood to be forced rapidly from the vena cava and other large veins within the body cavities into the thoracolumbar epidural venous plexus.

    This engorgement of blood in the veins pressed on the dura ensuring the dural (thecal) sac couldn't expand. The CSF crammed in this restricted dural (thecal) space caused a local increase in intrathecal pressure in the subarachnoid space. This local increased pressure shot the CSF up the spinal cord consequently causing a 'collapse' of the thoracolumbar dural (thecal) sac and a dilation higher up where the CSF pooled.

    The redirection of venous blood to the vertebral venous plexus due to changes in thoracic and abdominal cavity pressure was later confirmed by Aktas et al (2019).

    Flanagan (2015) found CSF volume and pressure also determined venous blood flow and volume in the epidural veins. A decrease volume and pressure of CSF in the subarachnoid space decreases its squeezing pressure internally leading to a decrease pressure in the central canal (which would normally oppose a syrinx) and decreases it squeezing pressure externally reducing venous blood flow and volume in the epidural vertebral veins. This intern further reduces compliance in the dural (thecal) sac.

    The vertebral venous plexus is continuous with the dural venous sinus that drains the CSF via the arachnoid villi. As the vertebral venous plexus is valveless Tobinick (2017) postulated the possibility of a valveless, ebb and flow system between the vertebral venous plexus and the dural sinuses. Flanagan (2015) found back pressure against the abdominal and thoracic veins (e.g. valsalva) is transmitted via the vertebral veins to the dural sinuses, which increases intracranial pressure.

    To extend this concept further the connections of the vertebral venous plexus with the systemic venous circulation show potentially an interdependent systemic venous network affecting dural mobility and CSF circulation.

    • Fat.

    The epidural fat protects the spinal cord and spinal nerves. It buffers the pulsatile movements of the dural (thecal) sac, facilitates dural (thecal) sac movement over the periosteum during flexion and extension and creates a reservoir for lipophilic substances.

    A reduction or absence of epidural fat has been associated with chronic spinal conditions e.g. spinal stenosis (Sions et al 2018).

    Epidural fat also has a different distribution in the cervical, thoracic, lumbar and sacral areas (Reina et al 2009). 

    At cervical level epidural fat is absent or almost inexistent. 

    T1-7 epidural fat is continuous.  This is opposite to the arrangement in the lumbar spine. Here the epidural fat is segmented and discontinuous. This patchy arrangement in the lumbar spine means the epidural fat does not usually adhere to these structures. This allows for mobility of the dura within the vertebral canal. Therefore could the continuous epidural fat from T1-7 have an oppposite effect reducing mobility of the dura within the vertebral canal?

    From T8-12 the distribution of epidural fat becomes patchy and discontinuous possibly acting as a transition from T8 to the lumbar spine. 

    This change in epidural fat not being so adherent at T8, possibly allowing for more movement of the dura, could parallel its location near the costodiaphragmatic recess. The most pronounced effect of respiraton on CSF flow at T8 is though to be from the opening of the costodiaphragmatic recess which is in close proximity to T8 (Aktas et al 2019). This would suggest the dura would have to move more freely at this level to accomodate the mechanical movement from the diaphragm and movement of the CSF. Refer to 'CSF movement'; 'respiration'.

    • Small arterial branches.
    • Lymphatics.
    • Epidural ligaments (of Hoffman): dura mater (dural sac and dural extension of the nerve root sleeve) --> posterior longitudinal ligament, spinal canal, and the ligamentum flavum. They provide support and protection by stabilizing and anchoring the dural sac (spinal cord and spinal nerves) to the bony vertebral canal. Hofmann’s ligaments prevent stretching of the spinal nerve roots when the intervertebral disc places pressure on the anterior part of the dural (thecal) sac. They have been postulated as a cause of low back and sciatic pain (Tardieu et al 2016).  

    Dura mater

    The spinal dura mater forms a tube. It’s upper end is attached to the base of the skull around the level of the foramen magnum and posterior surface of the C3 body. Its lower end is attached to the coccyx.

    It is attached along its course to the posterior longitudinal ligament (by the epidural ligaments of Hoffman < caudal end of the vertebral canal) and its continuation the tectorial membrane (C2 --> clivus* --> intracranial dura) (Husemeyer & White 1980). It also attached to the posterior antloccipital membrane in cases (Krakeness et al 2001). 

    *: also attached to the clivus is thick fibrocartilaginous tissue and the prevertebral fascia (Komune et al 2019).

    The dura mater is only partially attached to the vertebrae (apart fron C3), being separated from it by the epidural space. Oreskovic and Klarica (2014) found this permits a level of compliance of the dura mater resulting in the intradural volume being significantly altered in different physiological states e.g. hyperventilation, hypoventilation, pressure applied to the abdomen.

    This fluctuating compliance based on different physiological conditions makes the dural (thecal) sac a dynamic structure, readily changing its capacity in response to prevailing pressure gradients.

    These dynamic changes in the dura are both dependent upon and result in adaptive filling and emptying of the venous and lymphatic vessels; movement of CSF in the subarachnoid space of the dural (thecal) sac; perivascular transport of CSF and ISF into the spinal cord (refer to 'relationship between the external and internal movement of CSF').

    The dura mater receives myofascial attachments in the upper cervical spine:

    • Myodural bridges (Scali et al 2013): bands of fascia crossing the epidural space link the suboccipital muscles and the dura mater. This suboccipital fascia is continuous with the galea aponeurotica (Dacey et all 2018). Myodural bridges reach the dura via the PAO membrane from the Rectus Capitis Posterior Minor (RCPMi) and via the atlantoaxial interspace from the Rectus Capitis Posterior Major (RCPMa), Obliquus Capitis Inferior (OCI) and the Nuchal Ligament (NL). Zheng et al (2019) described the vertebrodural ligaments (refer below) as the final common pathway for the myodural bridge (RCPMa, OCI & NL) in the posterior atlanto-axial interspace.

    The myodural bridges may function in: (i) protect the spinal cord from dural enfolding (ii) transmission of proprioception from the spinal dura (iii) allows the suboccipital muscles to exert a reflexive myostatic response in placing the dura under tension (iv) myodural biofeedback might participate in maintaining the integrity of the subarachnoid space and posterior cerebellomedullary cistern (iv) these muscles exert a pumping effect by deforming the dura providing an important force required to move CSF in the spinal canal (Zheng et al 2014 & Xu et al 2016). Xu et al (2016) found CSF circulation at the craniocervical junction is significantly propelled in a cranial direction by head rotations attributed to the effects of these muscles on the dural sac (as well as changes in heart rate and respiratory rate).

    • Posterior atlanto-occipital membrane (PAO): The PAO membrane extends between the posterior arch of C1 and the occiput. Connective tissue fibres create a PAO membrane-spinal dura complex (Nash et al 2005). 

      The PAO membrane is mainly formed by the rectus capitis posterior minor tendinous fibers, the deep layer of the rectus capitis posterior minor fascia, and arterial and venous perivascular connective tissue sheathes. The rectus capitis posterior minor attaches to the dura indirectly via the PAO (Zheng et al 2019).

    The PAO attachments are:

    Laterally: the membrane extends between the rectus capitis posterior minor tendon (or aponeurosis) --> vertebral artery and vertebral cavernous sinus vascular sheath. Via the vascular sheath this membrane was continuous with the spinal dura.

    Medially: can not be identified but was formed from the rectus capitis posterior minor fascia and the perivascular sheathes of vertebral venous vessels.

    Inferiorly: fuses with the spinal dura and the perivascular sheath of the internal vertebral plexus.

    Superiorly: continues with the deep layer of the rectus capitis posterior minor fascia fusing to the occiput. 

    • First (intracranial) denticulate ligament: its attachments extend from (lateral): outer layer of the dura mater of the marginal sinus at the foramen magnum --> pierces the posterior atlantooccipital membrane and dura mater --> (medial): spinomedullary junction (Tubbs et al 2011). Refer to 'Denticulate Ligament'; 'First (intracranial) dentiuclate ligament'.
    • Epidural ligaments (of Hoffman): dura mater (dural sac and dural extension of the nerve root sleeve) --> posterior longitudinal ligament, the vertebral canal and the ligamentum flavum (Tardieu et al 2016).
    • Dorsal meningodural ligaments (Shi et al 2014): connective tissue bands running mainly from the ligamentum nuchae but also the laminae to the posterior dural (thecal) sac. Most commonly from C5 up with the strongest attachments being at C1-C2. They only exist in the cervical spine.
    • Vertebrodural ligaments include the verterebrodural ligament (VDL), to be named ligament (TBNL) and the craniale durae matris spinalis (CDMS) (Zheng et al 2014). Even though they are described seprarately all the vertebrodural ligaments are anatomical and functional linked with each other and the myodural bridges. Zheng et al (2019) described the VDL as the final common pathway for the myodural bridge (RCPMa, OCI & NL) in the posterior atlanto-axial interspace:

    i. Vertebrodural ligament: connects C1-2 ligamentum nuchae and posterior wall of the C1 & C2 vertebral canal to the cervical dura mater. The dura mater is distinctly thick at the VDL insertion. 

    VDL subdivided into three parts: Superior portion: C1: posterior arch of the atlas. Middle portion (TBNL’s part): C1-2 interspace. It is continuous with the TBNL. Inferior portion (axial part): C2 lamina. It is continuous with the TBNL which adheres to the lamina. 

    Suboccipital myodural bridge becomes part of the VDL as it passes through the atlantoaxial interspace.

    ii. TBNL: local enhancement of the ligamentum nuchae that forms part of the myodural bridge. It enters the epidural space through the atlanto-axial interspace and terminates at the posterior cervical spinal dura mater.

    Posterior border of ligamentum nuchae* --> atlantoaxial interspace --> C1-2 dura mater.

    *: Ligament nuchae attachments of the TBNL are: main attachment is below C3 SP at the common origins of the splenius capitis, superior posterior serratus and rhomboid minor muscles. At the C2 SP.

    As well as its more indirect muscular attachments at C3 it receives direct attachments from the rectus capitis posterior minor, rectus capitis posterior major and obliquus capitis inferior (Yuan et al 2017). 

    Zheng et al (2014) speculated that cervical flexion draws the cervical dura mater posteriorly creating a bulge via the TBNL and VDL passage through the atlantooccipital interspace.

    iii. Ligamentum craniale durae matris spinalis (CDMS): continuous with the deep part of the ligamentum nuchae.

    Edge of the foramen magnum, the posterior border of the O/A joints, nuchal ligament posterior arch of the atlas, and the arch of the axis --> cervical dura. 

    Superior attachments of the suboocipital fascia comes from the galea aponeurotica. Dacey et al (2018) found the galea aponeurotica transitions to the suboccipital fascia by dense fibrous attachments. Therefore we have myofascial continuity from Tenon's capsule running posteriorly via the occipitofrontalis to the galea aponeurotica/suboccipital fascia terminating in the upper cervical complex.

    Arachnoid mater

    The arachnoid mater is closely applied to the deep aspect of the dura mater.

    The nerves and blood vessels are invested by the inner layer of the meninges the pia mater. As the nerves and blood vessels drag the pia mater across the subarachnoid space the arachnoid mater reflects on to the surface of these nerves and vessels. This fusion of pia mater and arachnoid mater forms a leptomeningeal covering.

    At the level of the cauda equina, the arachnoid mater surrounds all the rootlets together as a sac and not individually (Dauleac et al 2019).

    As the rootlets continue to pass more laterally through the dura a fibrous ring of arachnoidodural adhesions is produced (Dauleac et al 2019).

    Within or slightly beyond the intervertebral foramen the dura mater continues as the epineurium. In the cervical spine, due to the increased movement, the dural sheaths are tethered to the adjacent transverse processes. In the lumbosacral region there is less tethering to the verterbrae although there maybe some attachment to the facet joint capsule.

    The leptomeninges (arachnoid and pia mater) don’t extend as far distally as the dura but continues as the perineurium. This seals the subarachnoid space preventing particulate matter spreading along the spinal nerves.

    Intermediate layer of leptomeninges (subarachnoid space)

    The intermediate leptomeningeal layer contains loosely arranged leptomeningeal cells that extend from the arachnoid mater to the spinal cord. Traversing the subarachnoid space they are loosely arranged forming a lattice work of fibers.

    This lattice work of fibers becomes compacted in places to form the dorsal, dorsolateral and ventral ligaments of the spinal cord.

    Weller et al (2018) hypothesised the leptomeningeal fibers may play a role in regulating the flow of CSF within the spinal subarachnoid space. A similar role was attributed by Pahlavian et al (2014) to the denticulate ligaments and nerve roots as they too traverse the subarachnoid space. Refer to 'movement of CSF; nerve roots, denticulate ligaments and arachnoid trabeculae'.

    Dauleac et al (2019) found other connections between the arachnoid mater and pia mater in the subarachnoid space at the level of the spinal cord:

    • Entrance point of dorsal rootlets in the spinal cord: arachnoid trabeculations were described at this level. There were numerous arachnoid adhesions between the the dorsal rootlets and the pia mater over the spinal cord.
    • Ventral part of the subarachnoid space: there were some arachnoid trabeculations between the arachnoid and pia mater.

    Pia Mater

    The spinal pia mater closely invests the surface of the spinal cord and passes into the ventral median fissure.

    The pia mater is reflected on to arteries and veins as they enter or leave the surface of the CNS (Weller et al 2018). The relationship of the pia mater around the nerves and the arachnoid mater and dura mater is discussed under ‘arachnoid mater’ 

    The subpial space extends between the pia mater (pial cellular layer) and the spinal cord (basal membrane in contact with neuroglial cells). The subpial compartment contains large amounts of collagen fibers, amorphous fundamental substance, fibroblasts, and a small number of macrophages and blood vessels.

    The collagen fibers in the subpial space is continuous with the denticular ligaments (subplial connective tissue --> dura mater).

    Ozawa et al (2004) found the pia mater highly elastic. This provides a constraint on the spinal cord surface. It prevents elongation of the circumference and produces a large strain energy that is responsible for shape restoration following decompression.

    Denticulate ligaments

    Ceylan et al (2012) identified denticulate ligaments in the cervical and thoracic spine. They end at the level of the conus medullaris, where the last pair of ligaments are broader and fibrous and turn downward and join with the filum terminale in a fork-like manner (Adeeb et al 2013). Overall they are stronger in the cervical region and decrease in strength as they move to the lower spinal levels (Ceylan et al 2012). 

    These ligaments extend from the pia mater (subpial collagen) --> dura mater (dural collagen). They are surrounded by a leptomeningeal layer that is continuous with the cellular layer of the pia and arachnoid mater (Adeeb et al 2013). 

    In addition to their attachments to the pia mater the denticulate ligaments in the cervical cord has collagen fibers that penetrate the spinal cord directly (Ceylan et al 2012).

    The denticulate ligaments extend traversley dividing the spinal canal into anterior and posterior compartments.

    Each denticulate ligament is composed of a single narrow fibrous strip with lateral triangular extensions.

    At cervical levels these triangular extensions are attached to the dura via short fibrous bands. In the thoracic spine the apices of the extensions are attached directly to the dura.

    This difference in attachment may be related to the mobility of the vertebrae at different levels. Due to increase neck movement the denticulate ligaments in the cervical canal need to permit more movement of the spinal cord than in the more rigid thoracic spine (Ceylan et al 2012).

    The denticulate ligaments are more resistant when stress is applied to the cord in the caudal as opposed to the cephalad direction. Tubbs et al (2001) found these ligaments also constrained anterior and posterior motion of the spinal cord.

    Reid (1960) identified several biomechanical phenomena of the denticulate ligaments during neck and trunk movement:

    • In the upper cervical region the dura doesn’t chang its position relative to the cord with flexion-extension movements of the neck. The reason why cord and dura moved together was because of the number and nature of the denticulate ligaments.
    • Movement of the nerve roots transmitted its effects to the cord via the dural sheath and the denticulate ligaments rather than the rootlets.

    The denticulate ligaments (with the nerve roots and arachnoid trabeculae) contribute to increasing the velocity of CSF flow by narrowing the spinal subarachnoid space. 

    This narrowed space between the denticulate ligament and ventral and dorsal nerve roots creates a flow jet. The altered velocity and flow patterns from these flow jets creates more of a complex mixing phenomena rather than just a straight up or down movement of the CSF (Pahlavian et al 2014). Refer to 'CSF movement; nerve roots, denticulate ligaments and arachnoid trabeculae'.

    The first (intracranial) denticulate ligament (Tubbs et al 2011):

    The first (intracranial) denticulate ligament consists of extensions of the arachnoid and pia mater.

    Its attachments extend from (lateral): outer layer of the dura mater of the marginal sinus at the foramen magnum --> pierces the posterior atlantooccipital membrane and dura mater --> (medial): spinomedullary junction.

    The ligament can blend with the vertebral artery and adhere to the first cervical rootlets.

    The connections between the dorsal rootlets of C1 and the spinal accessory nerve pierce the first (intracranial) denticulate ligament.

    It restricts rotation of the spinomedullary junction. As the spinomedullary junction flexes with cervical spine flexion (Malis et al 2016) could rotation of the cervical spine produce a rotatory movement at the spinomedullary junction? 

    Clinically the first (intracranial) denticulate ligament has been shown to compress the spinal accessory nerve causing a torticollis.

    As the vertebral venous plexus (as well as other venous routes) drains into the marginal sinus, which is the attachment of the intracranial (first) denticulate ligament, could dilation or constriction of the marginal sinus affect tension in this ligament?

    Intrathecal ligaments

    The intrathecal ligaments  are possible remnants from the fetal development of the denticulate ligaments. They are found in the cauda equina, below L1 --> S2.  They are surrounded by leptomeningeal cells that become continuous with the pia mater of the spinal roots (Adeeb et al 2013).

    These ligaments attach the dorsal spinal roots to the surrounding dura mater. Sometimes they vertically bind the ventral roots to the dorsal roots.

    The filum terminale

    The filum terminale (nervus impar) is a fibrous band that extends from the tip of the conus medullaris to blend with the periosteum of the posterior surface of the coccyx.

    It’s divided into:

    • Intradural compartment (filum terminale internum): reaches as far as the lower border of the S2. It is continuous above with the pia mater and is contained within the dural (thecal) sac. It is surrounded by the cauda equina. 
    • Extradural compartment (filum terminale externum or coccygeal ligament): closely adheres to the dura mater. It extends downward from the apex of the dural (thecal) sac and is attached to the back of the Co1.

    Marginal Sinus  

    The vertebral venous plexus drains into the marginal sinus. The marginal (dural) sinus rims the internal aspect of the foramen magnum. It lies between the layers of the dura mater.

    The outer coat of dura mater makes the dural sinuses stronger than normal veins. This makes them better suited to resist deformation by the brain, as well as collapse due to sudden drops and negative pressures caused by upright posture (Flanagan 2015).

    The first (intracranial) denticulate ligament extends from this dura mater, pierces the posterior atlantooccipital membrane and dura mater and attaches on to the spinomedullary junction (Tubbs et al 2011).

    Vertebral artery pierces the marginal sinus (Tubbs et al 2006) as it traverses the posterior atlantooccipital membrane and dura mater (Flanagan 2015).

    The dural venous sinuses are valveless and communicate with an extensive system of valveless emissary, scalp, face, and diploic and vertebral veins.

     The marginal sinus communicates with (Tubbs et al 2006):

    • Basilar venous plexus.
    • Occipital sinus.
    • Sigmoid sinuses: distal portions via small connections that that communicate with the veins of the hypoglossal canal.
    • Vertebral venous plexus: through the craniocervical junction. The most abundant vertebral veins are found in the anterior aspect of the canal.

    Flanagan (2015) found the most efficient and well developed emissary venous outlets used to drain the brain during an upright posture pass through the foramen magnum and hypoglossal and condylar canals of the craniocervical junction.

    Cerebrospinal fluid

    CSF functions in (Berliner et al 2019).:

    • Maintenance of the homeostatic environment for neurons and glia (Berliner et al 2019).
    • Transport of neuroactive substances around the CNS (Berliner et al 2019).
    • Acting as a drainage system for CNS interstitial fluid (ISF) by creating a CSF-ISF solution (Berliner et al 2019).
    • Buffer and buoyancy effects against external forces (Matsumae et al 2019).
    • Transmitting vascular pulsations (Matsumae et al 2019).
    • Buffer function of excess brain pulsation (Matsumae et al 2019).

    Production and absorbtion

    CSF is secreted by the Choroid plexus located within the ventricles of the brain < lateral ventricles. Through this mechanism it is essentially an extra-fine filtrate of the blood.

    Discussed later is the relative proportion of CSF that is in fact ISF. Therefore if we was to collect a sample of CSF it would come from two sources (i) the choroid plexus producing 'pure CSF'; (ii) CSF-ISF mix.

    CSF flows throughout the ventricular system unidirectionally in a rostral to caudal manner. CSF produced in the lateral ventricles travels through the interventricular foramina to the third ventricle. It then travels through the cerebral aqueduct to the fourth ventricle. The central canal of the spinal cord is continuous with the fourth ventricle. CSF flows from the fourth ventricle both down the central canal and through the median aperture (of Magendie) and lateral aperture (of Lushka) into the subarachnoid space.

    Once in the subarachnoid space, the CSF begins have a gentle multidirectional flow over the surface of the brain and down the length of spinal cord while in the subarachnoid space.

    It leaves the subarachnoid space through the arachnoid villi found in the dural sinuses (< superior sagittal sinus and venous lacunae) and diploic veins. CSF also leaves the subarachnoid space by the capillary veins, lymphatic pathways as well as around the roots of spinal nerves.

    Absorbtion of CSF from the subarachnoid space into the dural venous sinuses is dependent upon venous pressure (Flanagan 2015). To maintain constant pressure this emptying of CSF into the dural sinuses is balanced by reflex production of CSF by the choroid plexus.

    CSF is an extracellular fluid. Its absorption back into the circulation is influenced by venous pressure (Williams 2008). Increase venous pressure in the dural sinus decreases absorbtion of CSF in them (i.e. you can't empty CSF into them because they're already too full). This creates a back log of CSF increasing intracranial pressure and predisposing to conditions such as hydrocephalus.

    Passive production of CSF occurs during upright posture which decreases venous pressure in the vertebral veins and dural sinuses:

    A decrease venous pressure in the superior sagittal sinus means the venous sinus has greater capacity to absorb CSF into it from the subarachnoid space. This increase flow of CSF out of the brain via the dural sinus decreases intracranial CSF volume. To address this reduced CSF volume additional blood gets drawn from the choroid plexus to produce more CSF to make up for that that has been lost. This is called passive production of CSF (Flanagan 2015).

    Superior sagittal sinus venous pressure and flow is affected by posture, respiration, and movement. It is also affected by downstream pressure in the internal jugular and vertebral veins.

    Therefore to maintain constant intracranial pressure there is a balance between the influx of fluid into and out of the cranium:

    DECREASE INTRACRANIAL PRESSURE: increase venous outflow (decreases venous pressure) --> increase CSF absorbtion --> INCREASE INTRACRANIAL PRESSURE: increase arterial inflow --> increase CSF production.

    INCREASE INTRACRANIAL PRESSURE: decrease venous outflow (increases venous pressure) --> decrease CSF absorbtion --> DECREASE INTRACRANIAL PRESSURE: decrease arterial inflow --> decrease CSF production.

    Oreskovic and Klarica (2014) hypothesised an additional route for CSF production and absorbtion.

    These authors proposed that CSF is being permanently produced and absorbed inside the entire CSF system. This is by water filtration and reabsorption through the capillary walls into the ISF of the surrounding brain tissue. Yiming et al (2017) found 20% of all CSF originates from the brain's ISF.

    This ISF freely communicates with CSF. The exchanges between ISF and CSF occur at the walls of the ventricle through ependymal cells, pia-glial membranes at the surface of the brain/spinal cord, perivascular pathways along the spinal cord and via the g-lymphatic system. 

    Klarica et al (2019) and Lam et al (2017) also found absorbtion of substances from the CSF and ISF through transport mechanisms in the microvessels within the CNS parenchyma. Refer to 'internal movement of the CSF'.

    As 99% of CSF is composed of water and absorbtion of substances from the CSF is determined by transport mechanisms in the microvessels this may yield CSF and ISF more susceptible to fluid mechanics.

    Oreskovic and Klarica (2014) found because of the loose attachments of the dura mater to the vertebrae, as opposed to into the cranium where it is firmly attached, it is more distensible and compliant in response to different physiological conditions.

    This means in response to e.g. body position and respiration it has the potential to exert more of a pumping action on the CSF and surrounding venous and lymphatic vessels.

    This pumping action of the dura mater was meant to be permitted by the internal vertebral venous plexus (Martin et al 1972). When engorged with blood these vessels were meant to ‘wedge’ the dura mater against the bone increasing intrathecal pressure expelling CSF from the dural (thecal) sac at that particular level. Conversely when these veins were emptied it allowed the dura to expand filling the dural (thecal) sac with CSF.

    This intrathecal pressure gradient not only effects how CSF travels up and down the dural (thecal) sac at the subarachnoid space, but also how CSF travels in an out of the spinal cord along perivascular pathways (Berliner et al 2019).

    These perivascular pathways are responsible for the absorbtion of CSF from the subarachnoid space and ISF into the spinal cord and possibly drainage of the spinal cord. Refer to 'CSF movement; relationship between the external and internal movement of CSF'.

    Lam et al (2017) suggested a venous and lymphatic clearance route of CSF-ISF from the spinal cord.

    These authors found the continuity of the subarachnoid space with the dorsal and ventral spinal nerve rootlets suggested the existence of an additional outflow pathway for fluid along lymphatic or venous systems to the peripheral tissue.

    This lymphatic clearance could occur along cranial and spinal nerves to lymphatics outside of the CNS

    Could pressure changes in the dural (intrathecal) sac effect venous and lymphatic drainage of CSF-ISF from the spinal cord? 

    CSF movement

    CSF moves externally to the spinal cord in the dural (thecal) sac within the subarachnoid space, traversing the spinal cord via perivascular pathways and through the central canal of the spinal cord.

    External movement of CSF in the dural (thecal) sac within the subarachnoid space

    Traditionally CSF flow is described as being directed cephalad and caudad. However there is also a local mixing motion of the CSF at a segmental level.

    Broadly speaking movement of the CSF in the dural (thecal) sac is determined by five factors.

    • Cross sectional area of the spinal subarachnoid space.
    • Pressure gradients from respiration and posture.
    • Displacement of brain tissue into the spinal subarachnoid space.
    • Systole and diastole: the CSF pressure in the subarachnoid space is dependent upon the arterial pressure during systole and the venous pressure during diastole. CSF moving into the extracellular spaces is dependent on the CSF flow waveform (related to arterial input and venous output). 
    • Compliance in the dura mater covering the subdural (thecal) space: this is dependent upon venous distension. 

    These factors are interdependent and involve a variety of different mechanisms.

    Movement of CSF in the subarachnoid space of the dural (thecal) sac is determined by:

    • Vertebral venous plexus.

    As the dura mater isn’t firmly attached to the vertebrae it can move relatively freely changing the diameter of the dural (thecal) sac. This means the compliance of the dura mater and dural (thecal) sac changes passively in response to the pressure placed upon it externally by the internal vertebral venous plexus. When the veins at a certain level fill with blood and get larger the dural (thecal) sac at this level can't expand. This decreased movement increases intrathecal pressure causing the sac to empty CSF at this level (like squeezing a tube of toothpaste). Vice versa when the veins empty of blood the dural (thecal) sac become more compliant and can expand to let CSF pool in it.

    Martins et al (1972) found a partial collapse of the thoracolumbar dural (thecal) sac by the expulsion of its CSF during Valsalva.  They hypothesised the rise in pressure within the thorax and abdomen caused venous blood to be forced rapidly from the large veins of the bodys cavities into the thoracolumbar vertebral venous plexus. The external pressure from the rapid filling of these veins in the epidural space 'wedges' the dura against the vertebrae preventing it from expanding. This reduces compliance in the dural (thecal) sac at this level, increasing intrathecal pressure and results in CSF emptying out of it much like if you was to squeeze a tube of toothpaste. The redirection of venous blood to the vertebral venous plexus due to changes in thoracic and abdominal cavity pressure was later confirmed by Aktas et al (2019).

    Flanagan (2015) found the volume of venous blood flow in the internal vertebral venous plexus is also dependent on the volume and pressure of CSF in the subarachnoid space. A decrease volume and pressure of CSF in the subarachnoid space means it can't squeeze the spinal cord internally and internal vertebral venous plexus externally. This lack of squeezing on the spinal cord decreases pressure in the central canal (that would normally prevent a syrinx) and the lack of pressure on the compressible internal vertebral venous plexus in the epidural space decreases venous blood flow and volume.

    The vertebral venous plexus is continuous with the dural venous sinus that drains the CSF via the arachnoid villi. Tubbs et al (2018) found the vertebral venous plexuses, connected intracranially to the marginal sinus, basilar venous plexus via the transclival emissary veins, occipital sinus and the veins of the hypoglossal canal. These veins connected intracranially through foramen or the anterior atlanto-occipital membrane.

    As the vertebral venous plexus and dural sinuses are valveless Tobinick (2017) postulated the possibility of a valveless, ebb and flow system between the vertebral venous plexus and the dural sinuses. Flanagan (2015) found back pressure against the abdominal and thoracic veins, such as in a Valsalva maneuver, is transmitted via the vertebral veins to the dural sinuses, which increases intracranial pressure.

    To extend this concept further the connections of the vertebral venous plexus with the systemic venous circulation show an interdependent venous network affecting spinal dural mobility and CSF circulation.

    The intracranial (first) denticulate ligament attaches on to the marginal sinus (Tubbs et al 2011). Could filling or emptying of the marginal sinus from its venous (including vertebral) tributaries affect tension in this ligament influencing CSF flow? Refer to 'nerve roots, denticulate ligaments and arachnoid trabeculae'. 

    • Respiration.

    Aktas et al (2019) found forced respiration the dominant regulator of CSF dynamics. These authors found heart beat to represent a continuous albeit minor component to CSF movement causing CSF to move small distances at high velocities. Respiration however tends to move the CSF more slowly but over longer distances.

    The CSF moves upward to the cranial vault through the aqueduct during forced inspiration. This upward movement is most pronounced at C3, T1 & T8.

    The effects are reversed during exhalation. The most pronounced descent of CSF on expiration was at L3 < T8 > T1. 

    The most pronounced effect of respiraton on CSF flow at T8 is thought to be from the opening of the costodiaphragmatic recess. Especially in abdominal breathing which is in close proximity to T8. This change in CSF flow at T8 could also be mirrored by the more patchy discontinuous distribution of epidural fat that occurs at T8 allowing for increase mobility of the dura mater (refer to 'dural anatomy'; 'epidural space'; 'fat').

    The upward motion of CSF into the head and brain in inspiration is explained by the Monro-Kellie doctrine.

    Monro-Kellie doctorine allows for a manintenance of intracranial pressure. Intracranial pressure is dependent upon the total volume of substances within the cranial vault i.e. the brain, arterial blood, venous blood, CSF, and ISF. Therefore to maintain constant intracranial pressure if so much fluid descends out the head in inspiration e.g. venous blood then so much fluid has to ascend back into it again e.g. CSF. Vice versa for expiration. Obviously the movement of fluids is dependent upon all fluids i.e. arterial blood, lymph and interstitial fluids not just CSF and venous blood. However in the context of respiration CSF and venous blood are the most immediate repsonsive.

    How pressure changes during repsiration (and body position) affects venous blood flow to the vertebral venous plexus and the changes this makes to dural compliance have already been described. Refer 'vertebral venous plexus'.

    These pressure changes in the thorax and abdomen during respiration splits the spinal subarachnoid space into functional zones. This is because during respiration the thorax and abdomen differ in their pressure gradients, therefore the venous response in the thorax and abdomen is different and this places different demands on different levels of the internal vertebral venous plexus, which intern alters the compliance in the dural (thecal) sac at different levels.

    Dreha-Kulaczewski et al (2018) found pressure in the internal vertebral venous plexus in the upper-middle epidural spaces (above the heart) is dependent upon the pressure conditions in the superior caval venous system. Pressure in the internal vertebral venous plexus in the middle-inferior epidural spaces (below the heart) mirrors the pressure conditons in the inferior vena cava system.

    Usubiaga et al (1967) divided the epidural spaces into three regions dependent upon the functional relationship of the internal vertebral venous plexus to the body cavities:

    i. Cervico-thoracic region: the most extensive region. Influenced primarily by changes in the superior vena cava system pressure. Also affected by arterial beats, respiratory oscillations of the superior vena cava type.

    ii. Lumbar region: influenced primarily by changes in intra-abdominal pressure. Is also influenced by diminished arterial pulsations, respiratory oscillations of the inferior vena cava type.

    iii. Sacral region: this area does not exhibit a negative pressure at the time of an epidural puncture, arterial beats are absent, respiratory oscillations cannot be detected, and it is unaffected by abdominal wall compression.                                        

    • Brain and spinal cord motion

    Yiallourou et al (2012) found in cardiac systole the tonsils descend through the foramen magnum reducing the cross sectional area of the spinal subarachnoid space. Pahlavian et al (2015) measured this displacement of the tonsils and cervical cord to be 1mm. The movement of these tissues and altered cross sectional area of the spinal subarachnoid space causes increases CSF velocities in this region between C1 and C2. In diastole this movement is reversed and the tonsils move in a cephalic direction.

    Greitz et al (1992) noted a similar phenomenon. During systole the arteries expand and a relatively large volume of blood enters the brain. This causes a surge in CSF production at the choroid plexus and an increase capillary flow into the brain parenchyma. This influx into the brain parenchyma causes it to expand squashing their inner ventricles. The expansion of the arteries during systole and increase intracranial pressure creates a piston like effect squashing the CSF out of the brain and into the spinal cord. The velocity of this CSF movement through the ventricles accelerates as it descends through the third and fourth ventricle approaching the spinal cord (Matsumae et al 2019).

    Grietz et al (1992) hypothesised this venting of CSF into the spinal cord during systole may explain the funnel-shaped movement of the brain as if it were being pulled down by the spinal cord. This caudad movement was associated with an anterior motion. This caudad-anterior motion increased towards the foramen magnum and towards the midline. A similar movement of the brainstem was noted by Pahlavian et al (2015). 

    Brain motion is not just limited to systole and diastole Doursounian et al (1989) noted there was a downward shift of the lower part of the fourth ventricle during cervical spine flexion but with no change in shape. Malis et al (2016) noted this downward movement in flexion also included the pons and upper part of the cervical spinal cord. To accompany this downward movement in flexion the spinal canal increases in diameter. Conversely a hyperextension-compression force decreases the diameter of the spinal canal due to impingement from the ligamentum flavum (Zeng et al 2016).  

    The lumbar cistern is an enlargement of the subarachnoid space in the lumbar spine. It serves as an expansion tank for excess CSF. Its role as an expansion tank is important in order to maintain proper CSF volume, which is essential to prevent coning of the brainstem. This is how lumbar punctures produce coning of the brain (Flanagan 2015).

    • Arterial pulsations.

    As well as the effects mentioned above on 'brain and spinal cord motion' arterial pulsations has a local effect where the arteries run in the subarachnoid space. Matsumae et al (2014) noted an intense pressure gradient in the subarachnoid space around the anterior aspect of the brainstem from pulsations of the vertebrobasilar artery.

    This phenomena of a local increased velocity of CSF in the subarachnoid space around arteries was noted in other locations around the CNS.

    • Nerve roots, denticulate ligaments and arachnoid trabeculae

    The nerve roots, denticulate ligaments and arachnoid trabeculae contribute to the velocity of CSF flow by narrowing the spinal subarachnoid space. A decrease in the cross-sectional area of the spinal subarachnoid space due to the nerve roots and denticulate ligaments results in an increase in local CSF velocity (Pahlavian et al 2014).

    In areas where the subarachnoid space has a reduced cross sectional area, e.g. lower cervical spine in spondylotic changes or decsent of the tonsils and upper cervical spinal cord in systole or neck flexion, the nerve roots and denticular ligaments by remaining similar in size increase their relative contribution to increasing CSF velocity.

    The nerve roots and denticulate ligaments increase CSF velocity by narrowing the subarachnoid space and producing flow jets. The increased CSF velocity from these flow jets creates altered flow patterns i.e. a complex mixing phenomena rather than just a pure up and down movement of the CSF (Pahlavian et al 2014).

    These jets aren’t anatomical structures but areas of concentrated velocity regions produced between the narrow spaces between the dorsal and ventral nerve roots and denticulate ligament (Pahlavian et al 2015). They are most prevelant at C3-7 (Pahlavian et al 2014).

    • Cardiac systole and diastole

    Bunck et al (2011) noted the CSF flows caudally during systole and cranially during diastole. This was matched by descent of the spinal cord, tonsils and brainstem in systole and ascent in diastole. The effects on systole increasing intracranial pressure was discussed under 'brain and spinal cord motion'. The Monro-Kellie doctrine dictates this rise in intracranial pressure during systole has to be balanced by movement of CSF and venous blood out of the cranium; refer to 'respiration'.

    Another possible mechanism may play a role in the descent of CSF in systole caused by movement of the brain. Flanagan (2015) found the cisterns strategically located to provide extra protection and buoyancy for the brain, especially the brainstem due to its location above the base of the skull and foramen magnum. For example the lower cisterns e.g.cisterna magna, premedullary, and prepontine cisterns, help to prevent the cerebellum from sinking into the foramen magnum. This prevents cerebellar tonsillar ectopia similar to Chiari malformations.

    Could the caudad movement of the brain in systole exert a mechanical pressure causing a pumping action emptying these cisterns?

    • Neck rotation

    Xu et al (2016) found CSF circulation at the craniocervical junction is significantly propelled in a cranial direction by head rotations attributed to the effects of the myodural bridges causing a deformation of the dural sac.

    Internal movement of the CSF through the spinal cord

    The CSF is transported from the subarachnoid space to the central canal of the spinal cord via perivascular spaces. Along this route CSF and ISF can be (i) deposited and picked up from the parenchyma; (ii) transported through the vascular wall directly into the circulation (Lam et al 2017).

    Perivascular spaces are anatomical compartments contained between the walls of the blood vessels and the surrounding astrocytes/glia limitans.

    These compartments are packed with cellular processes of meningeal cells (the pia and arachnoid),

    CSF and solutes move through the subarachnoid space -->  discontinuity of pia and arachnoid layers --> perivascular spaces --> subpial space.

    This anatomical continuity provides a low resistance access pathway for subarachnoid fluid and solutes to rapidly penetrate the nervous tissue of the brain and spinal cord all the way to the central canal and to mix and exchange with interstitial fluid along its way.

    Along their course the perivascular spaces as well as being continuous externally with the extracellular spaces of the surrounding tissue are continuous internally with the basement membranes of the vascular wall. 

    As the fluid has access to the vascular basement membranes there is a possible vesicular transport mechanism across the vascular wall (smooth muscle and endothelial cells) directly to the blood circulation. 

    Currently, peri-arterial spaces are an established route of fluid entry, but there is debate regarding the routes of fluid clearance from the CNS tissue with both peri-arterial and peri-venular outflow suggested. Lam et al (2017) suggested both peri-arterial and peri-venular spaces are in continuity with the subarachnoid space. Although transport is expected to be faster in peri-arterial spaces due to greater width and the mixing effect of arterial pulsations.

    Relationship of the external and internal movement of the CSF

    CSF movement is dependent upon pressure gradients. The point of intersection between the external and internal movement of CSF is the subarachnoid space. Therefore increase pressure in the subarachnoid space will either increase fluid flow into the spinal cord (e.g. like squeezing a tube of toothpaste) or decrease drainage from the spinal cord (e.g. like turning on a hose and stepping on the end of it so water can’t escape).

    Flanagan (2015) found external pressure from the subarachnoid space exerts a squeezing pressure on to the spinal cord preventing a syrinx from developing. This squeezing pressure also extends externally to the epidural vertebral veins helping maintain venous pressure (Flanagan 2015). Conversely these veins themselves play an important role in maintaining compliance of the dural (thecal) sac. 

    Therefore a decrease volume and pressure of CSF in the subarachnoid space --> decreases pressure in central canal (which could lead to a syrinx) and decreases venous blood flow and volume in the internal vertebral venous plexus (as they are compressible).

    This effect is reversed with an increase volume and pressure in the subarachnoid space. Berliner et al (2019) found in the region of subarachnoid space obstruction there is a transient rostro-caudal increase of fluid in the spinal cord white and gray matter, due to an increased inflow along perivascular pathways, a reduced outflow along these pathways, or both. 

    These authors also found localized changes in compliance of the dural (thecal) sac affects CSF flow into the spinal cord.

    A loss of compliance of the dural (thecal) sac can also account for a large amount of perivascular absorption of CSF into the CNS.

    These authors raised the possibility that the fluid in the perivascular pathways may not reach the paremchyma and there could just be a back and forth movement of the CSF tracer between perivascular and subarachnoid space. This wouldn't lead to a significant increase net transport of CSF into the parenchyma itself.


    Standring S. Gray's Anatomy (2015). The Anatomical basis of clinical practice. Forty-first edition

    Dynamics of the cerebrospinal fluid and the spinal dura mater (1972). ALBERT N. MARTINS, JOHN K. WILEY, AND PAUL W. MYERS

    Spinal CSF flow in response to forced thoracic and abdominal respiration (2019). Gökmen AktasJost M. KollmeierArun A. JosephKlaus-Dietmar MerboldtHans-Christoph LudwigJutta GärtnerJens Frahm & Steffi Dreha-Kulaczewski 

    The Cerebrospinal Venous System: Anatomy, Physiology, and Clinical Implications (2017). Edward Tobinick

    The Epidural Ligaments (of Hofmann): A Comprehensive Review of the Literature (2016). Gabrielle G Tardieu, Christian Fisahn, Marios Loukas, Marc Moisi, Jens Chapman, Rod J Oskouian, and R. Shane Tubbs

    Topography of the lumbar epidural space. A study in cadavers using injected polyester resin (1980). Husemeyer RPWhite DC.

    A new look at cerebrospinal fluid movement (2014). Darko Orešković and Marijan Klarica

    Investigation of meningomyovertebral structures within the upper cervical epidural space: A sheet plastination study with clinical implications (2015). Scali F, Pontell ME, Nash LG, Enix DE.

    Shunt scissors: technical note (2018). Dacey RG, Flouty OE, Grady MS, Howard MA, Mayberg MR.

    The meninges as barriers and facilitators for the movement of fluid, cells and pathogens related to the rodent and human CNS (2018). Roy O. WellerMatthew M. SharpMyron ChristodoulidesRoxana O. Carare & Kjeld Møllgård 

    Anatomy of the human spinal cord arachnoid cisterns: applications for spinal cord surgery (2019). Dauleac CJacquesson TMertens P.

    The morphology and clinical significance of the dorsal meningovertebra ligaments in the cervical epidural space (2014). Shi B, Zheng X, Min S, Zhou Z, Ding Z, Jin A.

    Characterization of the Discrepancies Between Four-Dimensional Phase-Contrast Magnetic Resonance Imaging and In-Silico Simulations of Cerebrospinal Fluid Dynamics (2015). Soroush Heidari PahlavianAlexander C. BunckFrancis LothR. Shane TubbsTheresia YiallourouJan Robert KroegerWalter Heindel, and Bryn A. Martin

    The Impact of Spinal Cord Nerve Roots and Denticulate Ligaments on Cerebrospinal Fluid Dynamics in the Cervical Spine (2014). Soroush Heidari Pahlavian, Theresia Yiallourou, R. Shane Tubbs, Alexander C. Bunck, Francis Loth, Mark Goodin, Mehrdad Raisee, and Bryn A. Martin 

    The denticulate ligament: anatomical properties, functional 5 and clinical significance (2012). Davut Ceylan & Necati Tatarlı & Tuychiboy Abdullaev & AÅŸkın Åžeker & Sercan D. Yıldız & Evren KeleÅŸ & Deniz Konya & YaÅŸar Bayri & Türker Kılıç & Safiye Çavdar

    The pia mater: A comprehensive review of literature (2013) Adeeb NMortazavi MMDeep AGriessenauer CJWatanabe KShoja MMLoukas MTubbs RS.

    Intracranial Connections of the Vertebral Venous Plexus: Anatomical Study with Application to Neurosurgical and Endovascular Procedures at the Craniocervical Junction. (2018). Tubbs RSDemerdash ALoukas MCuré JOskouian RJAnsari SCohen-Gadol AA.

    The intracranial denticulate ligament: anatomical study with neurosurgical significance Laboratory investigation (2011) R. SHANE TUBBS, MARTIN M. MORTAZAVI, MARIOS LOUKAS, MOHAMMADALI M. SHOJA AND AARON A. COHEN-GADOL

    The ultrastructure of spinal cord perivascular spaces: Implications for the circulation of cerebrospinal fluid (2017). Magdalena A. Lam, Sarah J. Hemley, Elmira Najafi, Nicole G. F. Vella, Lynne E. Bilston, and Marcus A. Stoodley

    Effect of extradural constriction on CSF flow in rat spinal cord (2019). Joel A. Berliner, Thomas Woodcock, Elmira Najafi, Sarah J. Hemley, Magdalena Lam, Shaokoon Cheng, Lynne E. Bilston, and Marcus A. Stoodley

    Kinematic magnetic resonance imaging study of the brain stem and cervical cord by dynamic neck motion (2016)  M. Mališ, B. Georgievski, I. Milić, A. Mijatović, V. Kovačević, D. Lazić, S. Kapor1, S. Marinković 

    Respiration and the watershed of spinal CSF flow in humans (2018). Steffi Dreha-Kulaczewski, Mareen Konopka, Arun A Joseph, Jost Kollmeier, Klaus-Dietmar Merboldt, Hans-Christoph Ludwig, Jutta Gärtner, and Jens Frahm

    The brain interstitial system: Anatomy, modeling, in vivo measurement, and applications (2017). YimingLei Hongbin Han Fan Yuan Aqeel Javeed Yong Zhao

    The Movement of Cerebrospinal Fluid and Its Relationship with Substances Behavior in Cerebrospinal and Interstitial Fluid. (2019). Klarica MRadoš MOrešković D


    Comparison of 4D Phase-Contrast MRI Flow Measurements to Computational Fluid Dynamics Simulations of Cerebrospinal Fluid Motion in the Cervical Spine (2012). Theresia I. Yiallourou, Jan Robert Kröger, Nikolaos Stergiopulos, David Maintz, Bryn A. Martin and Alexander C. Bunck

    Pulsatile brain movement and associated hydrodynamics studied by magnetic resonance phase imaging. The Monro-Kellie doctrine revisited. (1992). Greitz D, Wirestam R, Franck A, Nordell B, Thomsen C, Ståhlberg F.

    Dynamics of the junction between the medulla and the cervical spinal cord: an in vivo study in the sagittal plane by magnetic resonance imaging. (1989). Doursounian LAlfonso JMIba-Zizen MTRoger BCabanis EAMeininger VPineau H

    Magnetic resonance 4D flow characteristics of cerebrospinal fluid at the craniocervical junction and the cervical spinal canal. (2011). Bunck ACKröger JRJüttner ABrentrup AFiedler BSchaarschmidt FCrelier GRSchwindt WHeindel WNiederstadt TMaintz D 

    Definition of the To Be Named Ligament and Vertebrodural Ligament and Their Possible Effects on the Circulation of CSF (2014). Nan Zheng, Xiao-Ying Yuan, Yun-Fei Li, Yan-Yan Chi, Hai-Bin Gao, Xin Zhao, Sheng-Bo YuHong-Jin Sui and John Sharkey 

    The Role of the Craniocervical Junction in Craniospinal Hydrodynamics and Neurodegenerative Conditions (2015). Michael F. Flanagan

    The Craniocervical Venous System in Relation to Cerebral Venous Drainage (2002). Diego San Millán RuízPhilippe GailloudDaniel A. RüfenachtJacqueline DelavelleFrank Henry and Jean H.D. Fasel

    Velocity and pressure gradients of cerebrospinal fluid assessed with magnetic resonance imaging (2014). Mitsunori Matsumae, Akihiro Hirayama, Hideki Atsumi, Satoshi Yatsushiro and Kagayaki Kuroda K

    Changing the Currently Held Concept of Cerebrospinal Fluid Dynamics Based on Shared Findings of Cerebrospinal Fluid Motion in the Cranial Cavity Using Various Types of Magnetic Resonance Imaging Techniques. Mitsunori MATSUMAEKagayaki KURODASatoshi YATSUSHIROAkihiro HIRAYAMANaokazu HAYASHIKen TAKIZAWAHideki ATSUMI, and Takatoshi SORIMACHI


    Mechanical properties and function of the spinal pia mater (2004). Ozawa H1, Matsumoto TOhashi TSato MKokubun S. 

    Epidural Fat and Its Association with Pain, Physical Function, and Disability Among Older Adults with Low Back Pain and Controls (2018). Jaclyn Megan Sions, Christina Angelica Rodriguez, Ryan Todd Pohlig, Gregory Evan Hicks and Peter Charles Coyle

    Clinical implications of epidural fat in the spinal canal. A scanning electron microscopic study (2009). M. A. REINA, C. D. FRANCO, A. LÓPEZ, J. A. DÉ ANDRÉS and A. VAN ZUNDERT 

    Morphometric Study of the Lumbar Posterior Longitudinal Ligament (2018). Sang Beom Lee, Jae Chil ChangGwang Soo Lee, Jae Chan Hwang, Hack Gun Bae and Jae Won Doh.

    MRI assessment of normal ligamentous structures in the craniovertebral junction (2001). Krakenes JKaale BRRorvik JGilhus NE.

    Annals of anatomy. Anatomischer Anzeiger (1992).W Kuhnel Lubeck, D. Drenckhahn, Würzburg J . Fanghänel, Greifswald Ch. Heym, Heidelberg A.F. Holstein, Hamburg W. Linß, Jena W. Platzer, Innsbruck B. Tillmann, Kiel K. Unsicker, Marburg L Vollrath, Mainz K. Zilles, Düsseldorf E. van der Zypen, Bern.

    The Evaluation and Observation of “Hidden” Hypertrophy of Cervical Ligamentum Flavum, Cervical Canal, and Related Factors Using Kinetic Magnetic Resonance Imaging (2016). Cheng Zeng, Jian Xiong, Jeffrey C. Wang, Hirokazu Inoue, Yanlin Tan, Haijun Tian, and Bayan Aghdasi

    The Fascial Layers Attached to the Skull Base: A Cadaveric Study (2019). Komune N, Matsuo S, Nakagawa T.


    Configuration of the Connective Tissue in the Posterior Atlanto-Occipital Interspace: A Sheet Plastination and Confocal Microscopy Study (2005). Lance Nash 1Helen NicholsonAntonio S J LeeGillian M JohnsonMing Zhang

    The second terminations of the suboccipital muscles: An assistant pivot for the To Be Named Ligament (2017). Xiao-Ying Yuan, Chan Li, Jia-Ying Sui, Qi-Qi Zhao, Xiao Zhang, Na-Na Mou, Zhao Huang-Fu, Okoye Chukwuemeka Samuel, Nan Zheng, Seung-Ho Han, Sheng-Bo Yu, and Hong-Jin Sui

    The Myodural Bridge Complex Defined as a New Functional Structure (2019). Nan ZhengBeom Sun Chung, Yi-Lin LiTai-Yuan LiuLan-Xin ZhangYang-Yang GeNan-Xing WangZhi-Hong ZhangLin CaiYan-Yan ChiJian-Fei ZhangOkoye Chukwuemeka SamuelSheng-Bo YuHong-Jin Sui


  2. Introduction

    Diploic veins are valveless channels traveling intraosseously in the diploic space between the inner and outer tables of the calvaria (diploë). 

    Diploic veins function as:

    • Venous channels within the diploe with connecting to the intra- and extracranial venous systems.

    Emissary veins link the dural sinuses to the diploic and extracranial veins including the veins of the vertebral venous plexus of the spine and spinal canal. The direction of venous blood flow in this vast system of interconnected valveless veins is determined by hydrostatic pressure gradients that are affected by body movements, respiration, Valsalva maneuvers, upright posture, and inversion (Flanagan 2015).

    • Part of the brains cooling mechanism. Venous blood gets cooled at the surface by conduction, convection, radiation, and sweating. This venous blood travels internally to keep the brain 2-3 degs cooler than the rest of the body. The diploic veins are cooled by surface veins and provide an additional layer of thermal protection for the cranial vault. They also with this cool blood cool the suboccipital cavernous sinus (Flanagan 2015).
    • CSF drainage pathways to the extracranial spaces (Tsutsumi et al 2014 & Lankar et al 2019).

    The diploic veins and CSF drainage

    The classic CSF drainage pathways are through:

    • Lymphatic vessels including lymphatics in the nasal mucosa (Engelhardt 2016), perineural spaces (Koh et al 2005) and dural venous sinuses (Benveniste et al 2019).
    • Dural venous sinuses: intracranial arachnoid granulations protrude into the dural venous sinuses. Miyajima and Arai (2015) described this as a secondary pathway but only if there is an increased CSF hydrostatic pressure. 

    In addition to this there is an intraosseous routes of CSF drainage into the diploe vein. In this model CSF is filtered through arachnoid granulations that protrude through defects in the dura mater into the diploë and drain into the diploic veins (Lachkar et al 2019).

    These arachnoid protrusions seem to develop with age. Tsutsumi et al (2014) found the the number of arachnoid protrusions over eighteen years was, except for in the pterional region, more than three-fold higher compared with under eighteens. These authors hypothesised the maturation of the dura mater may be related to the development of these arachnoid protrusions.

    A similar process may occur with spinal CSF whereby it is absorbed through spinal arachnoid villi and granulations into the spinal venous network.

    Absorption of CSF through the spinal cord arachnoid granulations is less than through intracranial arachnoid granulations. This is possibly due to the much smaller size of the spinal arachnoidal structures compared with those in the intracranial cavity (Tsutsumi et al 2015). 

    The diploic veins and venous drainage

    Tsutsumi et al (2014) found arterial blood to reach the arterioles and capillaries of the dura mater before passing into the venules, and then finally pouring into the diploic veins.

    The diploic veins serve as an important connection between the extracranial and intracranial venous systems. They drain either internally to the venous dural sinuses or externally to the veins of the head.

    Lachkar et al (2019) thought it is possible that there is greater communication between the dural venous sinuses and the general circulation than previously believed.

    Standring et al (2015) identified four main trunks of the diploic veins:

    • Frontal diploic veins: drains into (i) supraorbital vein at the bottom of the supraorbital notch; (ii) superior sagittal sinus near the superolateral aspect of the frontal sinus.
    • Anterior temporal diploic veins: situated near the pterion being mainly confined to the frontal bone. It drains into (i) sphenoparietal sinus; (ii) one of the deep temporal veins through an aperture in the greater wing of the sphenoid.
    • Posterior temporal diploic veins: situated near the asterion in the parietal bone. They drain into (i) sphenoparietal sinus. Lachkar et al (2019) found most of the diploic veins in the parietal bone are characterized by directly draining to the sphenoparietal sinus; (ii) Transverse-sigmoid sinus either through an opening at the mastoid angle of the parietal bone or through the mastoid foramen; (iii) superior sagittal sinus; (iv) emissary veins connecting the extracranial and intracranial veins (Lachkar et al 2019).
    • Occipital diploic veins: largest of the four diploic veins and is confined to the occipital bone. It can drain externally or internally. Its external drainage route courses inferiory in the medial occipital bone to drain into the occipital vein until it reaches the suboccipital venous channels. Internally it drains into either the transverse sinus or the confluence of the sinues. Lachkar et al (2019) found the diploic veins of the occipital bone only contributed minutely to venous drainage.

    Tsutsumi et al (2019) identified the diploic veins of the cranial base. Whilst the veins were scarce over the posterior and medial-third regions of the middle cranial base [presumabley the occiput] they were plentiful elsewhere. The pterional veins (intersection of the frontal, temporal, parietal, and sphenoid bones) were generally well defined, as pivotal channels connecting the lateral parts of the anterior and middle cranial base.

    The diploic veins in the frontal bone are more developed during adolescence, as peak growth of the frontal sinus is supported by increased vascularity. This allows for potential intracranial spread of frontal sinus infection (Lachkar et al 2019).

    During the transition from adolescence to adulthood, diploic veins are mostly observed in the parietal bone.

    In adults, as the diploic veins mature they articulate with intracranial dural sinuses most often, in the middle, rather than the anterior or posterior portions of the skull (Lachkar et al 2019)


    The Diploic Veins: A Comprehensive Review with Clinical Applications (2019). Stefan Lachkar, Mary-Margaret DolsBasem Ishak, Joe Iwanaga, and R. Shane Tubbs

    Cerebrospinal fluid drainage through the diploic and spinal epidural veins (2015). Satoshi Tsutsumi, Ikuko Ogino, Masakazu Miyajima, Masanori Ito, Hajime Arai, and Yukimasa Yasumoto

    Cranial Arachnoid Protrusions and Contiguous Diploic Veins in CSF Drainage (2014) S. Tsutsumi, I. Ogino, M. Miyajima, M. Nakamura, Y. Yasumoto, H. Arai, and M. Ito

    Vascular, glial, and lymphatic immune gateways of the central nervous system (2016). Engelhardt B, Carare RO, Bechmann I, Flügel A, Laman JD, Weller RO.

    The glymphatic system and waste clearance with brain aging (2019). Helene Benveniste, Xiaodan Liu, Sunil Koundal, Simon Sanggaard, Hedok Lee, and Joanna Wardlaw

    The glymphatic system and waste clearance with brain aging (2019). Helene Benveniste, Xiaodan Liu, Sunil Koundal, Simon Sanggaard, Hedok Lee, and Joanna Wardlaw

    Integration of the subarachnoid space and lymphatics: Is it time to embrace a new concept of cerebrospinal fluid absorption? (2005). Lena Koh, Andrei Zakharov and Miles Johnston

    Evaluation of the Production and Absorption of Cerebrospinal Fluid (2015). Masakazu MIYAJIMA and Hajime ARAI

    The Role of the Craniocervical Junction in Craniospinal Hydrodynamics and Neurodegenerative Conditions (2015). Michael F. Flanagan