A unifying hypothesis for hydrocephalus, Chiari malformation, syringomyelia, anencephaly and spina bifida
Cerebrospinal Fluid Research
A unifying hypothesis for hydrocephalus, Chiari malformation, syringomyelia, anencephaly and spina bifida Helen Williams
0 Address: 19 Elibank Road, Eltham, London, SE9 1QQ , UK
This work is a modified version of the Casey Holter Memorial prize essay presented to the Society for Research into Hydrocephalus and Spina Bifida, June 29th 2007, Heidelberg, Germany. It describes the origin and consequences of the Chiari malformation, and proposes that hydrocephalus is caused by inadequate central nervous system (CNS) venous drainage. A new hypothesis regarding the pathogenesis, anencephaly and spina bifida is described. Any volume increase in the central nervous system can increase venous pressure. This occurs because veins are compressible and a CNS volume increase may result in reduced venous blood flow. This has the potential to cause progressive increase in cerebrospinal fluid (CSF) volume. Venous insufficiency may be caused by any disease that reduces space for venous volume. The flow of CSF has a beneficial effect on venous drainage. In health it moderates central nervous system pressure by moving between the head and spine. Conversely, obstruction to CSF flow causes localised pressure increases, which have an adverse effect on venous drainage. The Chiari malformation is associated with hindbrain herniation, which may be caused by low spinal pressure relative to cranial pressure. In these instances, there are hindbrain-related symptoms caused by cerebellar and brainstem compression. When spinal injury occurs as a result of a Chiari malformation, the primary pathology is posterior fossa hypoplasia, resulting in raised spinal pressure. The small posterior fossa prevents the flow of CSF from the spine to the head as blood enters the central nervous system during movement. Consequently, intermittent increases in spinal pressure caused by movement, result in injury to the spinal cord. It is proposed that posterior fossa hypoplasia, which has origins in fetal life, causes syringomyelia after birth and leads to damage to the spinal cord in spina bifida. It is proposed that hydrocephalus may occur as a result of posterior fossa hypoplasia, where raised pressure occurs as a result of obstruction to flow of CSF from the head to the spine, and cerebral injury with raised pressure occurs in anencephaly by this mechanism. The current view of dysraphism is that low central nervous system pressure and exposure to amniotic fluid, damage the central nervous system. The hypothesis proposed in this essay supports the view that spina bifida is a manifestation of progressive hydrocephalus in the fetus. It is proposed that that mesodermal growth insufficiency influences both neural tube closure and central nervous system pressure, leading to dysraphism.
Hydrocephalus involves degrees of raised central nervous
system (CNS) pressure and extracellular fluid
accumulation. It is thought to result from many unrelated disease
processes. The relationship between hydrocephalus and
spina bifida has been the subject of prolonged debate. The
Chiari I malformation is related to posterior fossa
hypoplasia and causes spinal injury in syringomyelia by
obstruction to cerebrospinal fluid (CSF) flow at the
foramen magnum [1,2]. The Chiari II malformation is
found in association with spina bifida and is thought by
many to be unrelated to Chiari I, with neural injury and
reduced posterior fossa size caused by failure of neural
tube closure and its consequences, including toxicity
caused by exposure to amniotic fluid [3,4]. It is proposed
that there is an alternative explanation for the varied
manifestations of anencephaly and spina bifida that represents
a combination of defective neural tube closure and
hydrocephalus in the fetus. The hypothesis argues that:
Hydrocephalus progresses because of venous
insufficiency. This occurs with all primary abnormalities that
cause a pathological increase in CNS pressure.
Hindbrain herniation is caused by an abnormal
craniocervical pressure gradient or hindbrain compression
resulting from posterior fossa hypoplasia. These
mechanisms frequently act together to give rise to the Chiari
Chiari malformations cause obstruction to CSF flow that
elevates CNS pressure and damages neural tissue by
ischemic and mechanical forces.
Chiari-related syringomyelia, spina bifida and
anencephaly form a spectrum of disease related to restricted
growth of the posterior fossa.
The terms pressure and volume may often be used
interchangeably when describing CNS pressure, because
pressure depends upon volume . The hypothesis depends
upon the relationship between pressure and volume,
which includes the phenomenon of compliance.
Compliance enables a volume increase to occur in the intrathecal
CNS space without causing a pressure increase , and
occurs when a corresponding amount of venous blood is
displaced. Compliance affects the ability of the CNS to
accommodate volume fluctuations that occur with
movement and so avoid ischemia.
Blood from the CNS is drained by a venous plexus that is
extensive, anastamotic and valveless. It allows blood to
flow in a retrograde direction, away from the heart and
into the CNS with postural movements [6,7]. CSF
pressure fluctuations with movement have been observed
directly  and small movements such as occur during
speech, may cause detectable CNS pressure fluctuations
. Veins situated between the dura and surrounding
bone are highly compressible and capable of large volume
fluctuations . Such veins are prominent in the spine,
and are susceptible to volume fluctuation with alterations
in body cavity pressure . As CNS venous volume
increases with retrograde flow veins may be distended due
to 'back pressure' or may be compressed due to elevation
of overall pressure, resistance to venous outflow then
increases, outflow is reduced, and pressure will tend to
rise. Pressure increase in response to increasing intrathecal
volume corresponds to a phase of reduced compliance
. If volume increases further, a pressure may be reached
which compromises arterial flow, resulting in ischemia of
neural tissue . The relationship between overall CNS
pressure and added volume is shown in Fig. 1. The
pressure volume index (PVI) is the volume that when added
to the CSF space results in a ten-fold increase in pressure.
The normal adult range is 1326 ml . Adults will have
a larger PVI than infants because of a larger central
nervous system with a correspondingly greater venous
CSF flow contributes to compliance by regulating the
transluminal pressure of veins and therefore their patency
and flow. Poiseuille's law states that laminar flow in a
tube is inversely proportional to the fourth power of the
radius. Hence a decrease in vessel diameter will cause a
four-fold increase in resistance. If any pressure increase
associated with increase in CNS volume is distributed
throughout the intrathecal space, there will be a small
iTFnhiggeuthirneetrea1fcfercatnioanlpinretrsasucraenviaollupmreessruerlaetioofnisnhcirpe:aasignrgapvholduempeictThe intracranial pressure volume relationship: a
graph depicting the effect on intracranial pressure of
increasing volume. Central nervous system compliance
depends upon intrathecal volume. Redrawn from , with
permission [Additional file 1].
average reduction in the diameter of many veins. If the
whole CNS venous system is affected by a CNS volume
increase, the overall reduction in venous outflow will be
minimised, thus free flow of CSF facilitates venous
drainage. Veins inside the head do not collapse when moving
to an upright position because the movement of CSF with
gravitational force allows both fluids to experience the
same pressure gradient , maintaining the patency of
veins. Obstruction to CSF flow at the foramen magnum is
an important cause of reduced compliance. Movement of
the body appears to contribute to foramen magnum
obstruction, so that CSF flow may be affected by posture
. The head and spine may be transiently separated into
compartments with reduced pressure volume indices.
Evidence of obstruction may be present without an obvious
hindbrain hernia [11,12] or with cerebellar deformity.
Pressure gradients between the CSF spaces of the head and
spine are normally those due to gravity . In the
presence of an obstruction associated with spina bifida or
syringomyelia, pressure may fluctuate independently in
the head and spine [8,14,15]. This phenomenon has been
termed 'craniospinal pressure dissociation'. With normal
CSF pathways at the foramen magnum, retrograde flow of
venous blood into the CNS may be compensated by
compression of distant veins. If a volume increase occurs in
either the head or the spine in the presence of foramen
magnum obstruction, it will, according to this hypothesis
cause a pressure response which will be greater than if CSF
pathways were unobstructed. It is proposed therefore that
with Chiari malformation retrograde venous flow causes
an enhanced pressure response, as illustrated in Fig. 2.
In a relaxed subject with normal compliance, small
volume fluctuations caused by blood flow will not be
detected by overall pressure measurement. As compliance
is reduced, minor volume fluctuations may result in
measurable pulsations and physical movement will cause the
greatest pressure fluctuations and peaks. The elevation of
CSF pressure in normal pressure hydrocephalus  and
in hydrocephalic children observed during continuous
pressure monitoring  supports the hypothesis.
Reduction of pulsatility of CSF pressure following cranial
expansion as a treatment for hydrocephalus  is also
consistent with the proposed theory.
The mechanism of extracellular fluid accumulation
It is proposed that when compliance is reduced:
Venous volume increase with postural movements will
cause delay of venous and arterial flow in the
At peaks of pressure, vessels in the parenchyma of the
brain or spinal cord may be sufficiently compressed to
result in ischemia.
A hypothetical graph showing the relation between
spinal or cranial pressure, and venous volume, with
and without Chiari malformation. CSF obstruction at
the foramen magnum divides the CSF space into cranial and
spinal compartments. The sum of the pressure volume
indices of the two spaces would approximate that of the
unobstructed CNS. The pressure response to influx of venous
volume either in the head or the spine may be enhanced by
the presence of Chiari malformation.
Continuous unidirectional flow in venous plexus vessels
is not essential, venous blood may be diverted via
anastamoses to lower resistance vessels whereas continuous flow
in parenchymal vessels is necessary to provide oxygen.
Full compliance benefits flow in all vessels [5,19]. The
autonomic nervous system, by controlling vessel diameter
and flow, is known to influence extracellular fluid volume
in many tissues, with filtration or absorption taking place
in venules depending on local circumstances .
Autonomic regulation of flow in CNS vessels is likely to
influence the rate of filtration into the parenchyma. If,
however, autoregulation is unable to compensate for the
effect of venous pressure fluctuations, extracellular fluid
will accumulate causing reduction in compliance.
Consequently, venous outflow will be further compromised and
hydrocephalus will progress. Ischemia is likely to be a part
of the pathological process with a hyperemic response
enhancing filtration of fluid into the parenchyma.
Volume increases that do not cause ischemia are
accommodated, but progressive volume increase will cause
ischemia at peaks of CNS pressure.
The observation that removing CSF from patients with
symptomatic normal pressure hydrocephalus improves
both arterial supply and venous drainage  supports
the hypothesis proposed here. It is suggested that artificial
increase in cranial pressure by balloon insufflation inside
a lateral ventricle mimics the mechanism for progression
of hydrocephalus. In this model, intracranial pressure is
intermittently raised  in a manner that is comparable
to that which will occur with movement in the presence of
any cause of loss of compliance. Signs of a hyperemic
response may have been observed during direct
measurement of CNS pressure following pressure increase with
movement in Chiari I . Experimentally induced
foramen magnum blockage has lead to subjective
observations of venous insufficiency in the cord , and the
effect of pressure on venous flow, causing localised
oedema that resolves with improved perfusion of veins
may be seen during surgery .
Chiari I and II malformations are characterised by degrees
of hindbrain herniation. Adults with Chiari I
malformation have a reduced posterior fossa size, relative to head
size [12,24-26], providing evidence for posterior fossa
restriction as a cause of cerebellar herniation. Cerebellar
herniation is a feature of posterior fossa restriction with
known causes, for example craniosynostosis . The
familial tendency for Chiari I indicates that there are
genetic determinants for posterior fossa growth [28,29].
Chiari II is characterised by a hypoplastic posterior fossa
with compression of hindbrain structures . Associated
bone abnormalities in the face, provide additional
evidence of a primary maldevelopment of bone as part of the
malformation . Facial and posterior fossa bones have
common embryological origins and their responsiveness
to fibroblast growth factor suggests that they have related
growth mechanisms . The greater incidence of Chiari
I and II in females supports the view that posterior fossa
size is genetically determined, with males having larger
posterior fossa CSF spaces than females, so that restriction
of posterior fossa size leads to hindbrain herniation more
readily in females than males.
Spinal CSF and venous blood provide buoyancy to the
central nervous system , maintaining the position of
the hindbrain. The phenomenon of coning in response to
lumbar puncture illustrates how hindbrain herniation can
occur in response to loss of spinal fluid volume. The use
of lumboperitoneal shunts is associated with cerebellar
herniation , and is likely to be related to loss of CSF
from the spinal compartment. Reduction of posterior
fossa CSF space may also cause an abnormal cranio-spinal
pressure gradient by retaining CSF in the head, which
would otherwise be free to move through the foramen
magnum. Direct measurement of pressure in the presence
of Chiari I support the hypothesis that low CSF pressure
may cause the cerebellar tonsils to be drawn into the spine
. It is suggested here that this may occur when venous
blood leaves the spinal venous plexus so that excessively
free drainage from the spinal plexus may also contribute
to hindbrain herniation. In these instances, hindbrain
symptoms may predominate over those caused by raised
spinal pressure. In animal studies the hindbrain descends
following creation of an artificial spina bifida lesion ,
and may elevate following intrauterine repair . This
allows for the common assertion that low spinal pressure
contributes to the hindbrain herniation of Chiari II.
That the degree of hindbrain herniation does not correlate
with spinal injury in syringomyelia  and spina bifida,
and posterior fossa size does not correlate with the degree
of cerebellar herniation in Chiari I [11,12,36], indicates
that factors other than posterior fossa hypoplasia are
influencing the degree of neural injury. Herniation will
relate to the degree to which skeletal defects lead to
pressure changes and the extent to which low spinal pressure
causes hindbrain herniation in any individual case. Both
herniation and neural injury will depend upon an
interaction between the anatomy of the posterior fossa and
physiological influences on CSF pressure and flow. Although
low pressure may contribute to the cause of hindbrain
herniation, once herniation becomes more established
CNS pressure will tend to vary more widely as the head
and spine become separate compartments. Pressure in the
head may tend to rise  due to a larger volume of neural
tissue in the head than the spine, with a greater arterial
supply, as illustrated by the caudal flow of CSF across the
foramen magnum in systole and the rostral flow in
diastole . The potential range in pressure is arguably
greatest in the spine because of the capacity and
compressibility of the intraspinal venous plexus vessels.
This theory asserts that the highest spinal pressures lead to
cord ischemia and the lowest pressures contribute to
maintaining the hindbrain herniation. Loss of neural
tissue by ischemic atrophy will tend to benefit remaining
cord or brain, whereas growth of neural tissue may
perpetuate reduced compliance. A syrinx cavity forms a
spaceoccupying lesion and so reduces compliance. The effect of
a space-occupying lesion on spinal pressure will be as in
Fig. 3. Benign lesions may be accommodated without
pressure increase if neural or CSF volumes fall. The
combination of pressure effects illustrated in Figs. 2 and 3 will
be a left shift in the pressure volume curve in
It is proposed that hydrocephalus is an oedema of the
central nervous system. Rapidly developing venous
insufficiency leads to increase in parenchyma water content and
slower processes lead to increase in water that is
distributed between the parenchyma and larger extracellular
spaces. Symptoms, signs and morphology of
hydrocephalus will depend upon the pathology that causes the
pressure increase, the rate of fluid accumulation, and the
developmental stage at which hydrocephalus develops.
pFHriygepsuosrutehre3twiciatlhgmraopvhesmheonwtinagndthaescphaacneg-oecicnusppyiinnagl lveosliuomne and
Hypothetical graph showing the change in spinal
volume and pressure with movement and a
space-occupying lesion. Lines a and b represent volume fluctuation
with movement, with 'a' representing the normal range and
'b' the range with an uncompensated space occupying lesion.
Points 'a' and 'b' on the curve indicate maximal CSF pressures
that occur with maximal venous volume.
The internal dimensions of CNS bones
Fontanelles and surgically created bone defects
The presence of any space occupying lesion
Free flow of CSF around the CNS
The integrity of CNS veins and central venous pressure
Growth and atrophy of neural tissue
Autonomic regulation of blood pressure and flow
Fluid flows down pressure gradients from areas of higher
to lower pressure . The passage of fluid into the
ventricles and subarachnoid spaces represents flow to an area of
lower mean pressure than the parenchyma. Maximal CSF
pressures are achieved during physical movement and
arterial pressures will tend to exceed this. Movement of
tracers from the subarachnoid space to the parenchyma
has been demonstrated to occur rapidly  and might
appear to contradict this theory. If movement of fluid
occurs along perivascular channels, such movement
might depend upon the amplitude of the arterial pulse
and the energy that it imparts . According to the
proposed hypothesis rapid accumulation of fluid would tend
to remain in the parenchyma. Slow accumulation that
occurs with intermittent periods of relaxation, when CSF
pressure falls, would encourage fluid flow into CSF spaces.
Ventricle size is not a good indicator of intracranial
pressure. Lateral ventricle size is determined by ventricle wall
tension, brain turgor and the ability of CSF to flow
through the aqueduct and fourth ventricle. Small
ventricles may be found with raised, normal, or low intracranial
pressure if CSF can exit the head . Brain oedema may
lead to small ventricles if CSF can pass into the spine.
Fluid accumulation in the parenchyma will increase brain
turgor and cause it to resist ventricle enlargement 
whereas tension in the ventricle walls will exert a force on
brain tissue, favouring enlargement of the ventricles. Wall
tension is proportional to the internal radius of the cavity
, as well as fluid pressure. Pressure waves generated by
venous volume fluctuation in the spine may influence
lateral ventricle wall tension without the requirement for
flow , so physical movement may maintain or
increase ventricle size when posterior fossa CSF pathways
are patent. In chronic cases of raised intracranial pressure
there will be atrophy related to ischemia, causing the
brain to shrink. The posterior fossa CSF spaces and the
aqueduct will remain open, giving the clinical
presentation of normal pressure hydrocephalus. Rapid pressure
increase will tend to cause obstructions to CSF flow. When
the overall volume occupied by the brain increases with
space occupying lesions, enlarged ventricles, or oedema,
there may be obstruction of CSF flow at the aqueduct .
This is due to the narrowness of the aqueduct and its
compressibility in comparison to the ventricles. As overall
brain (including fluid) volume increases, CSF may be
displaced from the subarachnoid spaces into the spine and
the hindbrain may be displaced towards the foramen
magnum. This may enhance the rate of intracerebral
pressure increase by preventing the normal to and fro flow
movement of CSF across the foramen magnum  that
facilitates CNS venous drainage. These two major
obstructive effects on CSF flow will contribute to the
self-perpetuating nature of hydrocephalus, minor obstructions will
also be detrimental.
A balance between brain turgor and ventricle pressure
may occur, this is a feature of normal health and is also
illustrated by cases where normal lateral ventricle size may
occur with raised pressure and papilloedema.
Alternatively, very rapid increase in ventricle pressure may cause
symptoms of raised CNS pressure, with little time for the
development of clinically detectable oedema of the optic
nerve. A phenomenon of reactive enlargement of the
ventricles following lumbar puncture has been observed in
the presence of cerebral oedema . In these cases a
decrease in overall CNS pressure with lumbar puncture
allows improved venous drainage, water passes out of the
oedematous brain and CSF may then move out of the
spine into the cranial cavity as venous blood enters the
spinal venous plexus. If the ventricles are small or
obstructed and intracranial pressure is high the patient's
condition will be critical. This will be indicated by
obliteration of subarachnoid CSF spaces.
There are many disease processes that can lead to a state of
CNS venous insufficiency by causing an increase in
arterial supply, filtration of fluid into the parenchyma or
venous pressure. An incomplete list includes CNS
infection, hypoxia with altitude, carbon monoxide poisoning,
water intoxication, renal impairment, obesity, and
anaemia. Hydrocephalus may be caused by abnormalities of
CSF production in the choroid plexus or absorption at the
arachnoid granulations. That cardiac failure is an
infrequent cause of hydrocephalus  may be an illustration
of the efficiency of normal autoregulation of CNS
extracellular fluid volume. Restriction of internal bone space
through multiple suture craniosynostosis with a normally
developing brain will tend to cause hydrocephalus  by
compressing venous channels and CSF spaces. Single
lambdoid suture fusion will be associated with raised
pressure , because of the relative importance of
posterior fossa size to ventricle emptying and foramen
magnum flow. Direct and indirect measurements of venous
pressure demonstrate variable correlation with
hydrocephalus because pressure  and venous drainage 
are dynamic processes. Current methods of quantifying
these pressures are not representative of fluctuation with
Syringomyelia and spina bifida have been described as
part of a disease continuum, with more severe
manifestations in the fetus . The mechanism for neural injury in
the original theory had a requirement for hydrocephalus
with CSF flow from the head into the cord parenchyma
via the fourth ventricle. Hydrocephalus is not always
present in the two conditions and fluid flow from the
fourth ventricle into spinal cord cavities in syringomyelia
is uncommon. It is proposed that the disease continuum
relates to posterior fossa hypoplasia that causes reduced
CNS compliance before or after birth.
Anencephaly and Spina Bifida
Features of spina bifida that need to be addressed for a
unifying hypothesis to be plausible include:
The nature of progressive neural injury
An association with hydrocephalus
A higher incidence of Chiari malformation in females
Small head size in the fetus
Intrauterine growth retardation
Syringomyelia after birth
Animal models indicate that spina bifida may result from
failure of neural tube closure . Evidence suggests that
mesodermal growth supports and shapes the neural tube
to facilitate the closure process [49,50]. Vitamin A impairs
growth of the mesoderm and may induce dysraphic
disorders by this mechanism . Bones of the basichondrium
arise from mesoderm early in fetal development, prior to
completion of neural tube closure . Neural tube
closure in humans proceeds in a rostral direction from the
hindbrain region . It is proposed that genetic and
environmental influences have variable effects on
mesodermal growth influencing posterior fossa size and the
ability of the para-axial mesoderm, which forms vertebral
bone, to facilitate neural tube closure. Growth reduction
in the posterior fossa will tend to increase CNS pressure
which opposes neural tube closure. A fine balance
between growth and pressure may be required from the
earliest stages to achieve normal development of central
nervous system and surrounding bone. Where raised
pressure interacts with mesodermal restriction a failure of
neural tube closure will occur. A simplified representation of
the mechanism for dysraphism is shown in Fig. 4.
The more severe the mesodermal impairment, the more
hypoplastic the posterior fossa will tend to be, with earlier
and greater potential for separation of the spinal and
cerebral CSF spaces and subsequent neural injury.
Anencephaly represents the highest and most severe lesion. A more
normal posterior fossa size will lead to more normal
cerebral development. If posterior fossa size is not
significantly restricted and a small deficit in the vertebrae occurs,
a meningocele will tend to result at any level. The effect of
timed dosing of vitamin A in the creation of neural tube
defects adds strength to the argument that abnormal
mesodermal growth at different stages may result in different
morphological features. This teratogen results in more
frequent anencephaly when given early in gestation and
more frequent spina bifida if dosing is delayed . There
will, according to this theory, be a tendency for higher
lesions to be more severe, which accords with some
observations on spina bifida [54,55].
CSF Obstr uction
In the Poster ior Fossa
and For amen Magnum
FAigsiumrpelif4ied flow diagram illustrating the causes of spina bifida and anencephaly
A simplified flow diagram illustrating the causes of spina bifida and anencephaly.
In anencephaly and spina bifida, early neural
development may be relatively normal , with progressive
damage during gestation. Posterior fossa restriction occurs
at an early stage with progressive hindbrain herniation
[57,58]. It is suggested that any movement may cause
mechanical force on neural tissue. Fetal cardiac
movement is present from five post-menstrual weeks, trunk
movements are detectable from seven weeks and the fetus
is active by ten weeks . It is proposed that disturbances
in blood flow, accumulation of CSF and stretching forces
act on tissue, which lacks mechanical support of the
mesoderm. As fetal movement becomes stronger forces will be
magnified around an open lesion, but the whole CNS may
be affected. Widening of the whole vertebral column may
be found in anencephaly, suggesting severe distending
forces  and vertebral widening is found in the cervical
spine with syringomyelia suggesting a mild distending
force during growth .
Spina bifida is associated with reduced head size in the
fetus . This may be because as gestation progresses
skull growth depends upon pulsatility of pressure. This
view is supported by the rapidity of response to tensile
forces in skull suture fibroblasts . Pressure pulsations
generated in the spinal venous plexus are normally
transmitted to the head  but foramen magnum obstruction
attenuates pressure transmission across the foramen
magnum, between the two compartments [8,15]. Posterior
fossa hypoplasia may, by compressing the hindbrain,
block transmission of pressure pulsations and so reduce
the normal stimulus for skull growth in the fetus. Pressure
changes with movement will tend to be confined to the
spine, dissipated by the vertebral defect and prevent the
vertebral canal from closing. The lowest spinal pressures
will occur during relaxation of the fetus and may be
abnormally low or low for excessive periods, as a result of
the vertebral defect. It is proposed that these mechanisms
explains the majority of the progressive cord injury that is
observed in spina bifida. Brain growth in the presence of
skull restriction would tend to decrease ventricle size and
worsen the hindbrain herniation until foramen magnum
obstruction impairs fourth ventricle emptying sufficiently
that the ventricles enlarge. The skeletal abnormalities and
abnormal pressure gradient lead to a progressively
worsening hindbrain herniation that tends to become
moulded and impacted . It is proposed that ventricle
size in the affected fetus fluctuates with pressure according
to Fig. 5. The phase of small ventricle size represents the
early stage of relatively raised intracranial pressure. As
hydrocephalus progresses the development of a severe
foramen magnum obstruction is potentially lethal before
or after birth.
Excessive pressure in the head or spine as a result of
posterior fossa hypoplasia may be viewed as a progressive
process occurring during or after gestation. In the fetus
impairment to CSF flow may adversely influence
neuronal migration . Animal studies demonstrate that
ischemia damages developing neural tissue and
neurobehavioural abnormalities, such as memory deficits may
result from such injury . Abnormalities of migration
have been described in human neurospheres transplanted
Increasing posterior fossa size or decreasing lesion level
Hypothetical graph depicting the relation between
fetal lateral ventricle size and intracranial pressure
and the effect of hindbrain impaction on ventricle
size. Hindbrain impaction is caused by a pressure gradient
across the foramen magnum. A relatively small intracranial
pressure gradient will cause the lateral ventricles to empty.
Progressive increase in cerebral pressure will then cause the
ventricles to enlarge.
into rat cerebral cortex following ischemic injury .
Such studies give clues as to the possible origin of
complex cerebral abnormalities that may be found in
anencephaly and spina bifida.
The proposed hypothesis depends upon restriction of
posterior fossa growth as a cause of neural injury with
females having a smaller, genetically-determined, average
posterior fossa size and cisterna magna CSF space than
males. Posterior fossa size will be normally distributed.
The genders represented separately will form overlapping
curves with males to the right. If genetic or environmental
factors that restrict growth of the posterior fossa interact
with a normal variation in posterior fossa CSF space, and
if the pathogenesis of spina bifida is related to hindbrain
compression due to posterior fossa hypoplasia, there will
tend to be a difference in severity of lesions between the
sexes. Anencephaly will predominate in females, with
lower spinal lesions being more common in males. This
concept is represented in Fig. 6, and accords with
observations on differences between lesion prevalence between
the genders [49,67]. This hypothesis allows speculation
that smaller CSF spaces in the female are part of a
regulatory mechanism leading to smaller head size. With similar
fluctuations in spinal venous volume a more spacious
posterior fossa in the male may facilitate greater pressure
wave transmission into the head resulting in larger fetal
head size .
Acute worsening of obstructive hydrocephalus after birth
has been observed with spina bifida . Repair of the
spinal lesion may increase CNS pressure, whereas loss of
Hypothetical graph showing the effect of posterior
fossa size on the frequency of dysraphic lesions at
different levels in males and females. Higher, more severe,
lesions tend to be more frequent in females who have a
smaller posterior fossa as illustrated by the curve on the left.
Lower, less severe, lesions are more frequent in males with a
larger posterior fossa, as illustrated by the curve on the right.
CSF from a lesion may be beneficial for the fetus as
gestation progresses in circumstances where CSF tends to
accumulate. Breathing air and expansion of the thorax will
increase capacitance in pulmonary vessels and reduce
thoracic pressure . This will tend to improve CNS venous
drainage and lower CNS pressure. Improved venous
drainage may contribute to the low CSF pressure found in
normal neonates . It may also encourage herniation
of the hindbrain due to lowering of pressure in the spinal
venous plexus. Loss of brain turgor associated with weight
loss that tends to occur in the neonatal period  may
also contribute to ventricle enlargement. These factors in
combination will tend to favour foramen magnum
obstruction, whereas taping of CSF in the neonatal period
may avert an obstruction to CSF flow.
Abdominal growth tends to be reduced in fetuses with
open spina bifida as gestation progresses . If degrees
of CNS hypoxia are a feature of spina bifida that also
progresses during gestation then it is possible that this
pattern of abnormal abdominal growth would be expected.
One mechanism for accommodating a chronic increase in
central nervous system pressure may be decreased cardiac
output  suggesting a mechanism for growth
retardation. Areas of brain that are most damaged in anencephaly
are also the most susceptible to ischemia in chronic
hydrocephalus [72,73]. The relative preservation of basal
brain structures in anencephaly will relate to their blood
supply. Cervical meningoceles are sometimes found, but
are not associated with significant CSF obstruction at the
foramen magnum and neural development appears to be
normal . Support for a theory of ischemic damage in
spina bifida has been obtained from histological
Surgery that improves CSF pathways at the foramen
magnum may cause collapse of the syrinx cavity, improve cord
blood flow and the clinical features of syringomyelia. This
indicates that neural injury results from impaired flow of
CSF at the foramen magnum [1,2]. The central canal of
the cord forms a potential space, into which fluid from the
parenchyma may pass, allowing the formation of separate
cavities that characterise syringomyelia . During
physical exertion, movement of fluid within the syrinx cavity
may weaken its walls causing it to have a greater capacity
to enlarge [79,80] and during relaxation, the cavity will
tend to fill. This may be facilitated by a hyperaemic
response. Syringomyelia occurs in association with
posterior fossa restriction of Chiari I and other causes of
reduced CNS compliance, particularly if they directly
affect the spine. Tumours, vertebral deformity and
arachnoiditis may reduce space and CSF flow and may lead to
the onset of cavity formation. Vertebral injury causing
abnormal function of the spinal venous plexus may
impair spinal venous drainage as a result of autonomic
dysfunction . Decompressing the cord ameliorates
syringomyelia, and reducing the volume of fluid inside a
syrinx cavity at operation may enhance the rate of any
post-operative improvement . Normal pressure
hydrocephalus is an analogous disease process to
syringomyelia, as has been argued elsewhere,  with differing
distribution of excess extracellular fluid. Spinal cord
oedema would, according to this theory, represent a
presyrinx state . Syringobulbia is the manifestation of
raised spinal pressure with the unusual situation of fluid
being able to track into the brain stem. Communicating
syringomyelia would occur with pressure gradients and
anatomy allowing flow from the fourth ventricle into the
spine. A simplified representation of the mechanism for
FAigsiumrpelif7ied flow diagram representing the causes of hydrocephalus and syringomyelia
A simplified flow diagram representing the causes of hydrocephalus and syringomyelia.
tions on the spine, movement analysis of neonates with
spina bifida, [75,76] and the presence of metabolites
associated with ischemia in the CSF of affected neonates .
hydrocephalus and syringomyelia is shown in Fig. 7.
Posterior fossa hypoplasia makes hydrocephalus and
syringomyelia features of spina bifida after birth.
A hypothesis to unify all causes of hydrocephalus is
possible by considering the effects of physical movement on
CNS pressure when compliance is reduced. The
hypothesis argues that hydrocephalus is a self-perpetuating
problem caused by loss of compliance and also causing loss of
compliance by the accumulation of excessive extracellular
fluid which may obstruct CSF flow. The hypothesis allows
for the suggestion that a combination of cranial
expansion and relief of obstruction to CSF flow, including
plastic surgery to the posterior fossa, may reduce the necessity
for some shunt procedures. Continuous pressure
monitoring during activity causing reversible ischemia would
provide direct evidence for the proposed mechanism of
hydrocephalus and if this predicted effect could be
demonstrated, tests for reversible ischemia may be of
diagnostic use. An absence of reversible ischemia would suggest a
compensated phase of the disease lacking the potential for
a favourable surgical outcome.
Chiari malformations are primarily caused by congenital
posterior fossa hypoplasia sufficient to cause obstruction
to CSF flow, which damages the CNS. Abnormal
mesodermal growth leads to abnormalities of central nervous
system pressure. The manifestations of Chiari related
syringomyelia, spina bifida and anencephaly form a
spectrum of disease. If fetal posterior fossa CSF flow can be
improved there is the potential for reducing the impact of
Chiari malformation before birth.
CSF Obstr uction
In the Poster ior Fossa
and For amen Magnum
This is the work of one author. The author has read and
approved the final version of the manuscript.
Additional file 1
Permission to redraw Fig. 3 from . The CNS pressure volume curve.
The curve has three zones, a flat zone expressing good compensatory
reserve, an exponential zone, depicting poor compensatory reserve and a
final zone seen at very high ICP depicting derangement of normal
cerebrovascular responses. In  it is shown that pulse amplitude increases
linearly with mean intracranial pressure in the zone of poor compensatory
reserve. The graph is redrawn to suggest that this phase corresponds to a
phase of reduced compliance when pulsation caused by blood flow under
the influence of autoregulation is detectable. At low pressure pulsations are
less evident because of compliance and at the highest pressure pulsatility
decreases as arterial and venous flow is compromised.
Click here for file
I am grateful to Queen Elizabeth Hospital Woolwich for extensive library
assistance. The work of many authors has contributed to the formulation
of ideas presented in this essay; it has not been possible to recognise all of
these contributions. I am grateful to Dr S Demonchaux, Dr J Dickson, Dr
H Jones, Prof T Kohl, Prof K Nicolaides and Dr JMS Pearce.
1. Klekamp J : The pathophysiology of Syringomyelia-Historical overview and current concept . Acta Neurochir 2002 , 144 ( 7 ): 649 - 664 .
2. Levine DN : The pathogenesis of syringomyelia associated with lesions at the foramen magnum: a critical review of existing theories and a proposal of a new hypothesis . J Neurol Sci 2004 , 220 : 3 - 21 .
3. McLone DG , Knepper PA : The cause of Chiari II malformation: a unified theory . Pediatr Neurosci 1989 , 15 : 1 - 12 .
4. Mc Lone DG , Dias MS : The Chiari II malformation: cause and impact . Childs Nerv Syst 2003 , 19 : 540 - 550 .
5. Czosnyka M , Pickard JD : Monitoring and interpretation of intracranial pressure . J Neurol Neurosurg Psychiatr 2004 , 75 : 813 - 821 .
6. Tobinick E , Vega CP : The cerebrospinal venous system: anatomy, physiology and clinical implications . Med Gen Med (epub) 2006 , 8 ( 153 [http://www.medscape.com/viewarticle/522597].
7. Eckenhoff JE : The physiologic significance of the vertebral venous plexus . Surg Gynaecol Obstet 1970 , 131 ( 1 ): 72 - 78 .
8. Williams B : Simultaneous cerebral and spinal fluid pressure recordings . Cerebrospinal dissociation with lesions at the foramen magnum . Acta Neurochir 1981 , 59 ( 1-2 ): 123 - 142 .
9. Czosnyka M , Czosnyka Z , Momjian S , Pickard JD : Cerebrospinal fluid dynamics . Physiol Meas 2004 , 25 : R51 - 76 .
10. Levick RJ : An introduction to cardiovascular physiology London: Hodder Arnold ; 2003 .
11. Tubbs RS , Elton S , Grabb P , Dockery SE , Bartolucci AA , Oakes WJ : Analysis of the posterior fossa in children with the Chiari 0 malformation . Neurosurgery 2001 , 48 : 1050 - 1055 .
12. Sekula R , Jannetta PJ , Casey KF , Marchan EM , Sekula LK , McCrady CS : Dimensions of the posterior fossa in patients symptomatic for Chiari I malformation but without cerebellar tonsillar descent . Cerebrospinal Fluid Res 2005 , 2 : 11 .
13. Williams B : Simultaneous cerebral and spinal fluid pressure recordings . Technique , physiology and normal results . Acta Neurochir 1981 , 58 ( 3-4 ): 167 - 85 .
14. Williams B : Cerebrospinal fluid pressure gradients in spina bifida cystica, with special reference to the Arnold-Chiari malformation and aqueductal stenosis . Dev Med Child Neurol 1975 : 138 - 150 .
15. Hackel M , Benes V , Mohapl M : Simultaneous cerebral and spinal fluid pressure recordings in surgical indications of the Chiari malformation without myelodysplasia . Acta Neurochir 2001 , 143 ( 9 ): 909 - 918 .
16. Di Rocco C , Mairi G , Rossi GF , Vignati A : Cerebrospinal fluid pressure studies in normal pressure hydrocephalus and cerebral atrophy . Eur Neurol 1976 , 14 : 119 - 128 .
17. Di Rocco C , McLone DG , Shimoji T , Raimondi AJ : Continuous intraventricular cerebrospinal fluid pressure recording in hydrocephalic children during wakefulness and sleep . J Neurosurg 1975 , 42 : 683 - 689 .
18. Eide PK , Helseth E , Due-Tonnessen B , Lundar T : Changes in intracranial pressure after calvarial expansion surgery in children with slit ventricle syndrome . Pediatr Surg 2001 , 35 ( 4 ): 195 - 204 .
19. Bateman G : Vascular compliance in normal pressure hydrocephalus . AJNR 2000 , 21 : 1574 - 1585 .
20. Di Rocco C , Di Trapani G , Pettorossi VE , Caldarelli M : On the pathology of experimental hydrocephalus induced by artificial increase in endoventricular CSF pulse pressure . Childs Brain 1979 , 5 : 81 - 95 .
21. Avezaat C , Comment on, Hackel M , Benes V , Mohapl M : Simultaneous cerebral and spinal fluid pressure recordings in surgical indications of the Chiari malformation without myelodysplasia . Acta Neurochir 2001 , 143 : 918 .
22. Taylor AR , Byrnes DP : Foramen magnum and high cervical cord compression . Brain 1974 , 97 : 473 - 80 .
23. Milhorat TH : Cerebrospinal Fluid and the Brain Edemas New York: Neuroscience Society of New York ; 1987 .
24. Milhorat TH , Chou MW , Trinidad EM , Kula RW , Mandell MM , Wolpert CM , Speer MC : Chiari I malformation redefined: clinical and radiographic findings for 364 symptomatic patients . Neurosurgery 1999 , 44 : 1005 - 1017 .
25. Nishikawa M , Sakamoto H , Hakuba A , Nakanishi N , Inoue Y : Pathogenesis of Chiari malformation: a morphometric study of the posterior cranial fossa . Neurosurg 1997 , 86 ( 1 ): 40 - 47 .
26. Aydin S , Hanimoglu H , Tanriverdi T , Yentur E , Yasar Kaynar M : Chiari type malformations in adults: a morphometric analysis of the posterior cranial fossa . Surg Neurol 2005 , 64 : 237 - 241 .
27. Cinalli G , Spennato P , Sainte-Rose C , Arnaud E , Aliberti F , Cianciulli E , Renier D : Chiari malformations in craniosynostosis . Childs Nerv Syst 2005 , 21 : 889 - 901 .
28. Atkinson JLD , Kokmen E , Miller G : Evidence of posterior fossa hypoplasia in the familial variant of adult Chiari I malformation: case report . Neurosurgery 1998 , 42 : 401 - 403 .
29. Speer MC , George TM , Enterline DS , Franklin A , Chantelle M , Wolpert CM , Milhorat TH : A genetic hypothesis for Chiari I malformation with or without syringomyelia . Neurosurg Focus 2000 , 8 : E12 .
30. Marin-Padilla M : Cephalic axial skeletal-neural dysraphic disorders: embryology and pathology . Can J Neurol Sci 1991 , 18 : 153 - 169 .
31. Sarkar S , Petiot A , Copp A , Ferretti P , Thorogood P : FGF2 promotes skeletogenic differentiation of cranial neural crest cells . Development 2001 , 128 : 2143 - 2152 .
32. Chumas PD , Armstrong DC , Drake JM , Kulkarni AV , Hoffman HJ , Humphrey RP , Rutka JT , Hendrick EB : Tonsillar herniation: the rule rather than the exception after lumboperitoneal shunting in the pediatric population . J Neurosurg 1993 , 78 : 568 - 573 .
33. Williams B : Cerebrospinal fluid pressure changes in response to coughing . Brain 1976 , 99 ( 2 ): 331 - 346 .
34. Paek BW , Farmer DL , Wilkinson CC , Albanese CT , Peacock W , Harrison MR , Jennings RW : Hindbrain herniation develops in surgically created myelomeningocele but is absent after repair in fetal lambs . Am J Obstet Gynecol 2000 , 183 : 1119 - 1123 .
35. Elster AD , Chen MY : Chiari I malformations: clinical and radiologic reappraisal . Radiology 1992 , 183 : 347 - 353 .
36. Sgouros S , Kountouri M , Natarajan K : Posterior fossa volume in children with Chiari malformation type I . J Neurosurg 2006 , 105 : 101 - 106 .
37. Quigley MF , Iskandar B , Quigley MA , Nicosia M , Haughton V : Cerebrospinal fluid flow in foramen magnum: temporal and spatial patterns at MR imaging in volunteers and in patients with Chiari I malformation . Radiology 2004 , 232 : 229 - 236 .
38. Stoodley MA , Jones NR , Brown CJ : Evidence for rapid flow from the subarachnoid space into the spinal cord central canal in the rat . Brain Res 1996 , 707 : 155 - 164 .
39. Olivero WC : Pathogenesis of syringomyelia . Letter. AJNR 1999 , 20 : 2024 - 2025 .
40. Rekate HL : The slit ventricle syndrome: advances based on technology and understanding . Pediatr Neurosurg 2004 , 40 : 259 - 263 .
41. Rekate HL : Brain turgor (Kb): intrinsic property of the brain to resist distortion . Pediatr Neurosurg 1992 , 18 : 257 - 262 .
42. Berkouk K , Carpernter PW , Lucey AD : Pressure wave propagation in fluid-filled co-axial elastic tubes . Part 1: Basic theory . J Biomech Eng 2003 , 125 : 852 - 856 .
43. Williams B : Is aqueduct stenosis a result of hydrocephalus? Brain 1973 , 96 : 399 - 412 .
44. Greitz D , Hannerz J : A proposed model of cerebrospinal fluid circulation: observations with radionuclide cisternography . AJNR 1996 , 17 : 431 - 438 .
45. Caplan LR : Cardiac encephalopathy and congestive heart failure: a hypothesis about the relationship . Neurology 2006 , 66 : 99 - 101 .
46. Collmann H , Srensen N , Krauss J : Hydrocephalus in craniosynostosis: a review . Childs Nerv Syst 2005 , 21 : 902 - 912 .
47. Gisolf J , van-Lieshout JJ , van Heusden K , Pott F , Stok WJ , Karemaker KM : Human cerebral venous outflow pathway depends on posture and central venous pressure . J Physiol 2004 , 560 : 317 - 327 .
48. Gardener WJ : The Dysraphic States . From Syringomyelia to Anencephaly Amsterdam: Excerpta Medica ; 1973 .
49. Copp AJ : Neurulation in the cranial region, normal and abnormal . J Anat 2005 , 207 : 623 - 635 .
50. Marin-Padilla M : The closure of the neural tube in the golden hamster . Teratology 1970 , 3 : 39 - 45 .
51. Marin-Padilla M : Mesodermal alterations induced by hypervitaminosis A . J Embryol Exp Morphol 1966 , 15 : 261 - 269 .
52. Berry CL : A view of neurospinal dysraphism . Virchows Archiv A Pathol Anat Histopath 1992 , 420 ( 5 ): 375 - 376 .
53. Marin-Padilla M , Marin-Padilla TM : Morphogenesis of experimentally induced Arnold-Chiari malformation . J Neurol Sci 1981 , 50 : 29 - 55 .
54. Garabedian BH , Fraser FC : Upper and lower neural tube defects: an alternative hypothesis . J Med Genet 1993 , 30 : 849 - 851 .
55. Fletcher J , Copeland K , Frederick JA , Blaser SE , Kramer LA , Northrup H , Hannay HJ , Brandt ME , Francis DJ , Villarreal G , Drake JM , Laurent JP , Townsend I , Inwood S , Boudousquie A , Dennis M : Spinal lesion level in spina bifida: a source of neural cognitive heterogeneity . J Neurosurg 2005 , 102 ( 3 Suppl): 268 - 279 .
56. Muller F , O'Rahilly R : Development of anencephaly and its variants . Am J Anat 1991 , 190 : 193 - 218 .
57. Osaka K , Tanimura T , Hirayama A , Matsumoto S : Myelomeningocele before birth . J Neurosurg 1978 , 49 : 711 - 724 .
58. Beuls E , Vanormelingen L , van Aalst J , Vandersteen M , Adriaensen P , Cornips E , Vles H , Temel Y , Gelan J : The Arnold-Chiari type II malformation at midgestation . Pediatr Neurosurg 2003 , 39 : 149 - 158 .
59. De Vries JIP , Fong BF : Normal fetal motility: an overview . Ultrasound Obstet Gynaecol 2006 , 27 ( 6 ): 701 - 711 .
60. Lee J , Williams B : Cervical vertebrae measurements in syringomyelia . Clin Radiol 1977 , 28 : 395 - 400 .
61. Van den Hof MC , Nicolaides KH , Campbell J , Campbell S : Evaluation of the lemon and banana signs in one hundred thirty fetuses with open spina bifida . Am J Obstet Gynecol 1990 , 162 : 322 - 327 .
62. Yu JC , Lucas JH , Fryberg K , Borke JL : Extrinsic tension results in FGF-2 release, membrane permeability change, and intracellular Ca++ increase in immature cranial sutures . J Craniofac Surg 2001 , 12 : 391 - 398 .
63. Williams B : Chronic herniation of the hindbrain . Ann R Coll Surg Engl 1981 , 63 : 9 - 17 .
64. Owen-Lynch PJ , Draper CE , Mashayekhi F , Bannister CM , Miyan JA : Defective cell cycle control underlies abnormal cortical development in the hydrocephalic Texas rat . Brain 2003 , 126 : 623 - 631 .
65. Hava G , Huleihel M : The effect of prenatal hypoxia on brain development: short and long-term consequences demonstrated in rodent models . Dev Sci 2006 , 9 : 338 - 349 .
66. Kelly S , Bliss TM , Shah AK , Sun GH , Ma M , Foo WC , Masel J , Yenari MA , Weissman IL , Ichida N , Palmer T , Steinberg GK : Transplanted human fetal neural stem cells survive migrate and differentiate in ischemic rat cerebral cortex . Proc Natl Acad Sci USA 2004 , 101 : 11839 - 11844 .
67. Seller MJ : Neural tube defects and sex ratios . Am J Med Genet 1987 , 26 : 699 - 707 .
68. Williams H : Gender head size and disease; a hypothesis related to posterior fossa growth . Med Hypoth 2008 .
69. Heep A , Bartmann P , Stoffel-Wagner B , Bos A , Hoving E , Brouwer O , Teelken A , Schaller C , Sival D : Cerebrospinal fluid obstruction and malabsorption in human neonatal hydrocephaly . Childs Nerv Syst 2006 , 22 : 1249 - 1255 .
70. Grant DA , Walker AM : Pleural and pericardial pressure limit fetal right ventricular output . Circulation 1996 , 94 : 555 - 561 .
71. Welch K : Intracranial pressure in Infants . J Neurosurg 1980 , 52 : 693 - 699 .
72. Dombrowski SM , Schenk S , Leichliter A , Leibson Z , Fukamachi K , Luciano MG : Chronic hydrocephalus-induced changes in cerebral blood flow: mediation through cardic effects . J Cer Blood Flow Metab 2006 , 26 ( 10 ): 1298 - 1310 .
73. Momjian S , Owler BK , Czosnyka Z , Czosnyka M , Pena A , Pickard J : Pattern of white matter regional cerebral blood flow and autoregulation in normal pressure hydrocephalus . Brain 2004 , 127 : 965 - 972 .
74. Pang D , Dias MS : Cervical myelomeningoceles . Neurosurgery 1993 , 33 ( 3 ): 262 - 373 .
75. Sival DA , van Weerden TW , Vles JSH , Timmer A , den Dunnen WFA , Staal-Schreinemachers AL , Hoving EW , Sollie KM , Kanen-Mastenbroek VJM , Sauer PJJ , Brouwer OF : Neonatal loss of motor function in human spina bifida aperta . Pediatrics 2004 , 114 ( 2 ): 427 - 434 .
76. Sival DA , Brouwer OF , Bruggink JLM , Vles JSH , Staal-Schreinemachers AL , Sollie KM , Sauer PJJ , Bos AF : Movement analysis in neonates with spina bifida aperta . Early Hum Dev 2006 , 82 : 227 - 234 .
77. Pal K , Sharma U , Gupta DK , Pratap A , Jagannathan NR : Metabolite profile of cerebrospinal fluid in patients with spina bifida: a proton magnetic resonance spectroscopy study . Spine 2005 , 30 : E68 - E72 .
78. Milhorat TH , Capocelli AL , Anzil AP , Kotzen RM , Milhorat RH : Pathological basis of spinal cord cavitation in syringomyelia: analysis of 105 autopsy cases . J Neurosurg 1995 , 82 : 802 - 812 .
79. Williams B : Syringomyelia. Neurosurg Clin N Am 1990 , 1 : 653 - 685 .
80. Carpenter PW , Berkouk K , Lucey AD : Pressure wave propagation in fluid filled co-axial elastic tubes . Part 2: mechanisms for the pathogenesis of syringomyelia . J Biomech Eng 2003 , 125 : 857 - 863 .
81. Cassar-Pullicino VN , Colhoun E , McLelland M , McCall IW , El Masry W : Hemodynamic alterations in the paravertebral venous plexus after spinal injury . Radiology 1995 , 197 : 659 - 663 .
82. Holly LT , Johnson P , Masciopinto JE , Batzdorf U : Treatment of posttraumatic syringomyelia with extradural decompressive surgery . Neurosurg Focus 2000 , 8 : E81092 - 0684 .
83. Bateman GA : The role of altered impedance in the pathophysiology of normal pressure hydrocephalus , Alzheimer's disease and syringomyelia. Med Hypoth 2004 , 63 ( 6 ): 980 - 985 .
84. Fischbein NJ , Dillon WP , Cobbs C , Weinstein PR : The presyrinx state: a reversible myelopathic condition that may precede syringomyelia . AJNR 1999 , 20 : 7 - 20 .