See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/316520924

Role of choroid plexus in cerebrospinal fluid hydrodynamics

Article  in  Neuroscience · April 2017

DOI: 10.1016/j.neuroscience.2017.04.025

CITATIONS READS

57 12,674

3 authors, including:

Darko Orešković Milan Rados

Ruđer Bošković Institute University of Zagreb
81 PUBLICATIONS   1,107 CITATIONS    80 PUBLICATIONS   710 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Subplate zone of the human brain: unsolved problems View project

All content following this page was uploaded by Darko Orešković on 05 July 2017.

The user has requested enhancement of the downloaded file.

Neuroscience 354 (2017) 69–87

REVIEW
ROLE OF CHOROID PLEXUS IN CEREBROSPINAL
FLUID HYDRODYNAMICS
DARKO OREŠKOVIĆ, a*y MILAN RADOŠ b AND capillaries and interstitial fluid. Ó 2017 IBRO. Published by
MARIJAN KLARICA b*y Elsevier Ltd. All rights reserved.
a
Ruder Bošković Institute, Department of Molecular Biology,
Zagreb, Croatia
b
Department of Pharmacology and Croatian Institute for Key words: cerebrospinal fluid, choroid plexus, classical
Brain Research, School of Medicine University of Zagreb, Zagreb,
hypothesis of cerebrospinal fluid physiology, hydrocephalus,
Croatia
Intracranial pressure, Bulat-Klarica-Orešković hypothesis.

Abstract—The classic hypothesis presents the cere- Contents

brospinal fluid (CSF) as the ‘‘third circulation,” which flows Introduction 00
from the brain ventricles through the entire CSF system to The choroid plexus 00
the cortical subarachnoid space to eventually be passively The role of the choroid plexus as a CSF pump 00
absorbed into the superior sagittal sinus through arachnoid Is the choroid plexus really a CSF pump? 00
granulations. The choroid plexus (CP) represents a key The choroid plexus and entry of water into CSF system 00
organ in the classic CSF physiology and a powerful biolog- The choroid plexus and perfusion of the CSF system 00
ical pump, which exclusively secretes CSF. Thereby, the CP The choroid plexus as
is considered to be responsible for CSF pressure regulation a generator of hydrocephalus development 00
and hydrocephalus development. This article thoroughly The role of the choroid plexus in the CSF homeostasis 00
analyzes the role of the CP in the CSF dynamics, presenting The choroid plexus and regulation of CSF volume 00
arguments in favor of the thesis that the CPs are neither bio- The choroid plexus and biochemical composition of the
logical pumps nor the main site of CSF secretion; that they CSF 00
do not participate in regulation of ICP/CSF pressure; are not The choroid plexus and CSF pressure 00
the reason for the existence of hydrostatic pressure gradient Conclusions 00
in the CSF system and that this gradient is not permanent Acknowledgments 00
(disappeared in the horizontal position); and that they do References 00
not generate imagined unidirectional CSF circulation,
hydrocephalus development and increased ICP/CSF pres-
sure. The classic hypothesis cannot provide an explanation
for these controversies but the recently formulated Bulat-K
larica-Orešković hypothesis can. According to this hypothe-
sis, CSF production and absorption (CSF exchange) are
constant and present everywhere in the CSF system, and
although the CSF is partially produced by the CP, it is mainly INTRODUCTION
formed as a consequence of water filtration between the
The hydrodynamics of cerebrospinal fluid (CSF) has so
far been almost exclusively explained using the classic
hypothesis of the CSF. According to this hypothesis, the
*Corresponding authors at: Ruder Bošković Institute, Department of CSF is actively produced inside the brain ventricles,
Molecular Biology, Bijenička 54, 10 000 Zagreb, Croatia. Fax: +385- circulates unidirectionally within the entire CSF space
1-456-1-177 (D. Orešković). Department of Pharmacology and (from formation to absorption site) and is passively
Croatian Institute for Brain Research, School of Medicine University absorbed into the venous blood on the brain surface
of Zagreb, Zagreb, Croatia, Šalata 11, 10 000 Zagreb, Croatia. Fax:
+385-1-4596-942 (M. Klarica). and/or into the lymph via paraneural sheaths of nerves
E-mail addresses: doresk@irb.hr (D. Orešković), mklarica@mef.hr (Fig. 1A; Brierly and Field, 1948; Cserr, 1971; Bradbury,
(M. Klarica). 1981; Davson et al., 1987; Johnston et al., 2005; Koh
y
Both these authors are corresponding authors and equally partic- et al., 2006; Pollay, 2010; Sakka et al., 2011; Brinke
ipate in presented work.
Abbreviations: AQP, aquaporin; BV, brain ventricle; CM, cisterna
et al., 2014; Hladky and Barrand, 2014). The CSF is
magna; CNS, central nervous system; CP, choroid plexus; CPC, represented as a river that flows slowly from its source
choroid plexus carcinomas; CPP, choroid plexus papillomas; CSF, to its mouth. In physiological conditions, the CSF flow is
cerebrospinal fluid; ICP, intracranial pressure; ISF, interstitial fluid; LV, steady, constant, pulsating and without barriers.
lateral ventricle; SAS, subarachnoid space.

http://dx.doi.org/10.1016/j.neuroscience.2017.04.025
0306-4522/Ó 2017 IBRO. Published by Elsevier Ltd. All rights reserved.

69
70 D. Orešković et al. / Neuroscience 354 (2017) 69–87

Fig. 1. (A) The scheme of the cerebrospinal fluid system with location of the choroid plexuses, the CSF distribution, and arachnoid granulations.
CSF is represented by the light blue area and the direction of CSF circulation and the sites of CSF absorption are represented by red arrows. (B)
The scheme of the human brain ventricles with emphasized choroid plexuses in red color and CSF shown as the light blue area. (C) The choroid
plexus structure, with emphasized branched choroid plexus structure and the villi projecting into the brain ventricle. The plexus network is presented
with fenestrated capillaries and the bold blue arrows indicate the direction and active nature (secretion) of CSF formation.

Consequently, along with blood and lymph, the CSF
dynamics in mammals was presented as the ‘‘third circu-
lation” (Cushing, 1914, 1925; Taketomo and Saito, 1965;
Milhorat, 1975). According to Cushing’s (1925) interpreta-
tion (‘‘that at about the fifth week of embryonic life fluid
begins to percolate through the remaining fragment of
the roof plate overlying the fourth ventricle; that it finds
its way through leptomeningeal channels which are
thereby formed in the mesenchyme; and finally, that it
escapes by way of persisting leptomeningeal communica-
tions, the villi, that project through the dura into its major
venous synuses. All of which assuredly indicates the birth
of an actual circulation under the pressure of secretion.”),
the site of CSF absorption, the direction of CSF circulation
and the energy that stimulates the circulation (pressure of
secretion) are strictly defined. If ‘‘the third circulation” or
CSF circulation is similar to blood circulation through the
body, driven by the pumping action of the heart, it is nec-
essary that an organ similar to the heart (biological pump
which constantly pumps blood in the same direction)
exists in the CSF system. The working principle of each
pump in a circulatory system, including the heart, is the
production of constant gradient of hydrostatic pressure
between the highest pressure value at the beginning
and the lowest pressure value at the end of the circula-
tion. According to the classic hypothesis, this role in the
CSF system belongs to the choroid plexuses (CPs). In
other words, CPs secrete the CSF in the brain ventricles
(Fig. 1; Davson et al., 1987; Brown et al., 2004; Damkier
et al., 2013; Lun et al., 2015; Spector et al., 2015). Since
the CSF formation is an active process (secretion), mod-
erate changes in intracranial pressure (ICP) should not
significantly change the CSF formation rate (Heisey
et al., 1962; Rubin et al., 1966; Cutler et al., 1968; Sklar
et al., 1980; Pollay et al., 1983).
Consequently the CP is imagined as an organ which
represents a biological pump in the CSF system, i.e. as
an organ responsible for ‘‘pressure of secretion”. A
hundred years ago Weed (1914) assumed that the intra-
ventricular secretion of the CSF by the CP imparted the
energy required for fluid circulation from the ventricles,
out over the brain convexities, and back into the blood
(Fig. 1). If it is considered that the CSF is passively
absorbed from the subarachnoid space into the venous
sinuses due to a lower hydrostatic pressure in the sinuses
than in the subarachnoid space (Weed, 1935; Brodbelt
and Stoodley, 2007; Pollay, 2010), it seems that the con-
ditions for CSF circulation exist. Therefore, the most
important role in CSF physiology was given to the CP,
and it is necessary to re-elaborate Weed’s assumption.
Hence, it could be concluded that the relationship
between active CSF formation by the CSF pumps (CPs)
 D. Orešković et al. / Neuroscience 354 (2017) 69–87
and passive CSF absorption is what creates the total CSF
volume. Namely, under physiological conditions, there
has to be a balance inside the CSF system between
CSF secretion (Vf) by the CPs and absorption (Va) of
the CSF by dural venous sinuses:
Vf ¼ Va
Consequently, the rate of CSF formation (inside the
brain ventricles) must be equal to the rate of CSF
absorption (into the venous sinuses). If they are not
equal, an imbalance in the CSF volume may arise,
leading to a pathological condition. The same should be
valid for the rate of unidirectional circulation (QCSF). The
CSF secretion and absorption CSF take place at two
separate and distant sites in the CSF system and,
therefore, the rate of unidirectional circulation (QCSF)
between the secretion and absorption site should be the
same, so that Vf remains equal to Va;
Vf ¼ QCSF ¼ Va
Changes in the CSF secretion rate by the CPs should
influence the rates of both the CSF flow and CSF
absorption to maintain physiological CSF volume. It is,
therefore, logical that CPs (CSF secretion) are the main
generators (pumps) of CSF circulation and the
regulators of the CSF volume inside the CSF space.
Since the CSF is secreted exclusively by the CPs
(Spector et al., 2015), the CPs obviously play a crucial
role in the CSF dynamics, i.e. in its secretion, circulation
and absorption. Consequently, the rates of physiological
CSF secretion, circulation and absorption directly depend
on the CPs’ metabolic function.
According to the presented dynamics of the CSF, the
regulation of CSF/ICP is also in its essence defined by
function of CPs. This is clearly shown by Marmarou’s
equation (1978), which connects the classical concept
of CSF physiology and ICP:
ICP ¼ VfxRo þ Pv
where Vf equals the rate of CSF secretion, Ro equals the
resistance to the CSF movement (circulation) inside the
CSF system, and Pv equals the pressure in dorsal
sinuses and the resistance to the CSF absorption into
the dural sinuses (blood circulation). This means that the
rate of CSF secretion is the main factor in ICP regulation
(increase of the CSF secretion rate directly results in an
increase of ICP). In addition, the CSF secretion will
generate higher ICP if there is higher resistance to CSF
circulation and absorption, caused by different pathologic
processes in CNS which affect CSF system (scars,
bleeding, tumors, edema, cysts and so on). Since the
CSF formation and CSF circulation, according to this
classical concept, are the consequence of CP function, it
is strongly suggested that the CPs as biological pumps
should have a determining impact on the ICP.
Furthermore, the circulation theory of the
hydrocephalus development was also based on the
function of CPs. Namely, hydrocephalus is assumed to
result from an imbalance between CSF secretion and
absorption, with a subsequent increase in the CSF
volume in the cranium and enlargement of the brain
ventricles. Since the CSF is continuously secreted by
71
the CPs (CSF pumps) it should be formed against
increased CSF/ICP (Pollay et al., 1983; Davson et al.,
1987). Pathological blockade of the normal CSF pathway
should result in a CSF over-accumulation in front of the
block (Dandy, 1919). For example, in some hypertensive
hydrocephalus patients, the ICP/CSF pressure is so high
that it requires (urgent) neurosurgical intervention, i.e. the
opening of CSF system to save the patient’s life. Such eti-
ology of hydrocephalus unquestionably presents the CP
as a biological pump and shows how this pump could
have an important role in the CSF system.
Therefore, the classically interpreted dynamics of CSF
depends on the active CSF formation (secretion), i.e. on
the proper function of the CPs. Without the CPs as
biological pumps, CSF formation, circulation and
passive absorption, as well as regulation of CSF
pressure would not be possible. In other words,
according to Weed’s assumptions (1914) and the
classic hypothesis, the CP has a key position in the
physiology and pathophysiology of CSF. However,
although the above-mentioned arguments seem
reasonable and easily acceptable, there are numerous
experimental and clinical data that cannot be explained
by the classic hypothesis.
Bering has clearly shown in his experiments on dogs
(1962) that CPs are not the major sites of CSF
formation, but suggested that they are CSF pumps.
These pumps should act as unvalved pulse pumps,
using transfer of choroid arterial pulsation, imparting a
to-and-fro motion to the CSF (Bering, 1955). Hence, the
CPs were shown as pistons which produce a force for
CSF circulation, but not as secretory organs whose action
should be the driving force of CSF circulation. Using MRI
methods, Greitz (1993) did not mention CPs at all as a
driving force of CSF circulation, but this role was intended
for pulsating flow produced by the alternative pressure
gradient. This gradient is a consequence of the systolic
intracranial arteries’ expansion causing movement of
CSF from cranium into flexible spinal SAS. That way,
the mentioned force during systole is responsible for
remodeling the brain by its piston action in a chimney-
like fashion (Greitz et al., 1992). However, in spite of such
proposed CSF movement, the net flow at the cervical
SAS could not be simultaneously identified (Greitz, 1993).
Furthermore, using the generally accepted hypothesis
of CSF secretion, unidirectional circulation and absorption
many clinical states cannot be explained (for example
occurrence of unilateral hydrocephalus with patent CSF
pathways, the need for shunt or external drainage after
plexectomy, occurrence of hydrocephalus with various
processes present inside the spinal canal or after spina
bifida repair with completely patent cranial CSF
pathways, development of external hydrocephalus,
development of normal or hypotensive pressure
hydrocephalus, development of syringomyelia at
different positions inside the central canal, etc.), and the
treatment methods used so far for the disorders
associated with CSF do not lead to further progress,
showing relatively poor results. All of this indicates that
it is necessary to reevaluate the current understanding
of the CSF physiology and pathophysiology.
72 D. Orešković et al. / Neuroscience 354 (2017) 69–87
As a result of research attempts to overcome the
limitations of the classic hypothesis, a new Bulat-Oreško
vić-Klarica hypothesis of the CSF has recently been
presented. According to it, CSF production and
absorption (CSF exchange) are constant and present
everywhere in the CSF system, partially originating from
the CP but mainly as a consequence of water filtration
between the capillaries and interstitial fluid (Orešković
and Klarica, 2010, 2011; Bulat and Klarica, 2011;
Klarica and Orešković, 2014). The new hypothesis has
so far been strongly supported by independent experi-
mental laboratories (Krishnamurthy et al., 2012, 2014;
Igarashi et al., 2013; Yamada, 2014) and computer-
based models (Buishas et al., 2014; Linninger et al.,
2016). According to this hypothesis, interstitial fluid (ISF)
and CSF (99% of CSF volume is water) are generated
by water filtration across the arterial capillaries in the
entire central nervous system (CNS; brain and spinal cord
tissue), while plasma osmotic active substances are
retained, generating osmotic counter-pressure inside the
capillaries, which is crucial in ISF/CSF water absorption
into the venous microvessel network. Thus, according to
this hypothesis the main regulators of CSF and interstitial
fluid volumes, which are connected and can be observed
as a functional unit, are osmotic and hydrostatic forces at
the vast capillary network of the brain and the spinal cord.
Fluid and substances are constantly exchanged
between the CSF system and the surrounding tissue,
depending on the osmotic and hydrostatic pressure
gradients at capillary levels observed under different
physiological and pathophysiological conditions within
these compartments. It should be emphasized that the
contact surface between the choroid plexus and the
CSF is around 5000 times smaller than that between
the brain capillary network and the brain interstitium.
Therefore, CSF formation does not take place in one
part of the CSF system (brain ventricles) and absorption
in another (cortical SAS), and CPs participate only
partially (according to the size of the contact surface
area) in the exchange processes rather than being a
biological pump and the dominant place of CSF secretion.
This way regulated CSF volume is carried
bidirectionally along all CSF spaces by the systolic-
diastolic to-and-fro displacement (Orešković and Klarica,
2014b). This kind of movement is different in spinal and
cranial space due to anatomical and biophysical charac-
teristics of the cranial (unchangeable volume; rigid cra-
nium) and spinal (changeable volume; dural sac) part.
All other substances in CSF (1% of total CSF volume)
are distributed in all directions throughout entire CSF sys-
tem, i.e. in direction of ‘‘circulation” (from ventricle to sub-
arachnoid space-SAS) as well as in the opposite direction
(from SAS into ventricles) (Orešković and Klarica, 2010,
2014b; Bulat and Klarica, 2011). The distribution distance
of any substances within the CSF space and ISF depends
on the half-life of these substances in CSF/ISF. The sub-
stance which is removed from the CSF/ISF more slowly,
would eventually reach larger distance. Thus, a new con-
cept of the CSF physiology connects it with the physiology
of every other cranio-spinal fluid (blood, ISF, intracellular
fluid), implying which forces regulate the volumes of CSF
and other fluids, and how the CSF volume and sub-
stances present inside it move.
The aim of this paper is to give an explanation about
the CPs’ role in the CSF hydrodynamics based of
experimental and clinical results and systemic analysis
of published data. Taking into account the presented
facts, we will also try to answer the following questions:
1) Do the CPs play a similar role in CSF circulation as
the heart does in blood circulation, representing an
energy source (biological pumps) that produces the
hydrostatic pressure gradients in the CSF system and
allow the unidirectional flow (circulation) of CSF? 2)
Does the CSF move in all directions, depending on the
anatomical and biophysical relationships, law of fluid
mechanics and systolic and diastolic pulsation of the
CSF, as was suggested by the new hypothesis?
THE CHOROID PLEXUS
The CP is a plexus positioned in the ventricles of the
brain, consisting of modified ependymal cells, which
produces the CSF (Fig. 1). By classic view, the CSF is
undoubtedly formed inside the cerebral ventricular
system and the CPs represent the main site of CSF
production (Dandy and Blackfan, 1914; Dandy, 1919;
Masserman, 1934; de Rougemont et al., 1960; Welch,
1963; Rubin et al., 1966; Davson, 1967; Netsky and
Shuangshoti, 1970; Cserr, 1971; Welch, 1975; Milhorat,
1987; Spector and Johanson, 1989; Johanson et al.,
2008; Spector et al., 2015). Some other extrachorioidal
sources of CSF were suggested, such as the ependyma
lining the ventricles (Pollay and Curl, 1967), the blood–
brain barrier, but in a very small extent, having no impor-
tance for the main postulates of the classic hypothesis
(Cserr, 1988; Abbott, 2004) and the subarachnoidal arte-
rioles and veins, for which there was minimum evidence
(Davson et al., 1987). As a result of these alternative
sources, there would be no net active formation of the
CSF, and the term CSF secretion should be used only
for the process performed by the CP epithelial cells
(Spector et al., 2015).
No one has done so much for acceptance of the CPs
are as the biological pumps and dominant sites of CSF
formation, as Dandy in his crucial experiment (1919).
After he performed unilateral plexectomy and obstructed
the foramina of Monro of both lateral ventricles in a dog,
the ventricle containing the CP dilated and the ventricle
without the CP collapsed (Fig. 2). The results obtained
in this way clearly point to the CP as a place of active
and exclusive CSF formation. Namely, if the CSF was
secreted inside the ventricle without the CP, then this
ventricle would not collapse, and if CP is not a biological
pump, the ventricle with the CP could not be dilated.
This experiment has had a crucial influence on
scientists over the past one hundred years and has
provided the main framework for thinking about CSF
physiology. Furthermore, the experiment showed that
the lateral ventricle obviously could not dilate if CSF
absorption took place in the ventricles. Consequently, if
CSF absorption did not take place in the ventricles, the
CSF should circulate from the CPs to the place of
 D. Orešković et al. / Neuroscience 354 (2017) 69–87
Fig. 2. The scheme of Dandy’s crucial experiment (1919) in a dog,
where choroid plexectomy in the right lateral ventricle was performed
and the foramina of Monro of both lateral ventricles was obstructed.
The left ventricle containing the choroid plexus is dilated and the right
ventricle without the choroid plexus collapsed. The light blue area
represents the CSF.
absorption outside of the ventricles. Hence, this crucial
experiment formed a general picture of the classic
hypothesis, with CSF secretion, circulation and
absorption, and with the CPs as the most important
organs. But here it is necessary to stress that the
presented result was obtained in only a single dog and
that unfortunately no one so far has reproduced this
experiment (1919). Namely, no one has been able to
confirm that the plexectomized and obstructed ventricle
collapsed after the obstruction of the foramen of Monro,
either in dogs (Hassin et al., 1937; Bering, 1962) or in rhe-
sus monkeys (Milhorat, 1969). On the contrary, the plex-
ectomized and obstructed lateral ventricle became
markedly dilated if significant time elapsed between the
time of obstruction and the time of death (Milhorat,
1969). This also indicates that hydrocephalus can quickly
develop inside ventricles with surgically removed CPs and
that the CPs are not the main origin of the CSF (Milhorat,
1969). A very similar result was also observed in humans
(Radoš et al., 2014a). Although the classic hypothesis
has emerged from an unrepeatable experiment in only
one animal and clinical data that cannot be explained in
the traditional way (see later), nothing has diminished its
significance and the significance of Dandy’s experiment.
THE ROLE OF THE CHOROID PLEXUS AS A
CSF PUMP
Since the CP is presented as an unquestionable place of
CSF secretion (Spector et al., 2015), we will continue to
focus on the role of the CP as a CSF pump. The CP rep-
resents a highly vascularized glomus supplied by anterior
and posterior sets of choroidal arteries with an extremely
rich venous network joining to form a single large main
choroidal vein (Fig. 3A; Davson, 1967). It is a highly vas-
73
cularized and branched structure consisting of numerous
villi, which project into the ventricles (the lateral, the third
and the fourth) of the brain (Fig. 1A, B). The CSF should
be formed in a two-step process. The first, the CSF is
formed from plasma ultrafiltrate, which passes through
the fenestrated endothelium of the CP capillaries, facili-
tated by hydrostatic pressure and the second, it passes
through the choroidal epithelium into the ventricle. The
second step is an active metabolic process that trans-
forms the ultrafiltrate into a secretion (Segal and Pollay,
1977; Pollay et al., 1985; Davson et al., 1987; Brown
et al., 2004; Johanson et al., 2008). The CSF secretion
depends on the unidirectional transport of ions by the
Na+/K+ pump, which generates an osmotic gradient
inducing the net water movement (Fig. 3). Water move-
ment through the CP epithelium is regulated by aquaporin
(AQP) channels (Fig. 3C). However, although the whole
active process of CSF secretion is still incomplete and
speculative, it was presented in many articles with mini-
mal differences (Pollay, 1977; Davson et al., 1987;
Spector and Johanson, 1989; Lyons and Meyer, 1990;
Bergsneider, 2001; Brown et al., 2004; Brodbelt and
Stoodley, 2007; Johanson et al., 2008; Damkier et al.,
2013; Spector et al., 2015). Aquaporins were discovered
due to the observation that it is not possible that such a
great volume of water can pass so quickly through the cell
membrane in many physiological processes in the organ-
ism if there are no cellular structures that allow this pas-
sage. And AQP do precisely that – they belong to a
family of membrane channel proteins that serve as selec-
tive pores for water and prevent the passage of ions and
other solutes (Agre, 2006). Hence, without the existence
of AQP (water) channels, rapid water movement through
the cells cannot be explained. Since the CSF consists of
99% of water (Orešković and Klarica, 2010), AQPs should
play a very important role in the CSF secretion and the
structure of the CP as a secretory organ. Therefore, the
presence of AQPs in the CP was also used as an impor-
tant indicator that the CP should be the main site of CSF
secretion, i.e. that CP is a biological pump. Even although
there is reasonable suspicion that animal models cannot
determine if such transport is able to form the CSF vol-
ume (Brown et al., 2004), the rapid movement of water
(practically sole content of the CSF) further indicates the
importance of AQP channels. Anyhow, numerous experi-
ments were performed in various animal species showing
that the CP was the site where many substances enter
the CSF from the blood (Kanaka et al., 2001; Li et al.,
2002; Brown et al., 2004; Oshio et al., 2005; Johanson,
2008; Spector et al., 2015) and a place of one way traffic.
Except the experiment by Dandy (1919), substantial
contribution to the belief that the CP is a secretory organ
was obtained by experiments on animal CP itself (de
Rougemont et al., 1960; Welch, 1963; Pollay et al.,
1972). de Rougemont et al. (1960) surgically accessed
the brain ventricles in cats and using a glass pipette col-
lected and measured the newly formed CSF from the
CP surface covered with Pantopaque oil. This experiment
pointed to the role of the CP in the CSF secretion, but it is
important to mention that its results are questionable not
only because it was conducted under non-physiological
74 D. Orešković et al. / Neuroscience 354 (2017) 69–87

Fig. 3. (A) Schematic representation of the choroid plexus supplied by choroidal artery (red color) and choroidal vein (blue color) with highly
developed vascularization. (B) Scheme of the choroidal villus covered by a single layer of cuboidal epithelial cells and fenestrated capillaries. (C) An
explanation of choroid plexus function in the CSF secretion in terms of Na+-K+ pump on the CSF-facing (apical) membrane. Transport of Na+, Cl,
and HCO 3 ions regulates the net osmotic gradient across the choroid plexus. Na
+
is pumped out of the choroid cells by the Na+/K+ pump and its
intracellular concentration is driven down. The Na+ gradient activated in this way on the plasma-facing membrane (basolateral membrane)
subsequently activates via Na+/Cl- cotransport the transport of Na+ into the cell. Cl- exchange takes place across the apical and basolateral
membranes. On this membrane, the influx of Cl- into the choroid cell is allowed by an active transport Cl-/HCO-3 pump and cotransport Na+/K+/2 Cl-
mechanism. On the apical side, the intracellular Cl- efflux takes place possibly by transport proteins and through channels. The regulation of
intracellular pH is performed by basolateral Na+/H+ antiport and Cl-/HCO-3 pumps. Cl- accumulation on the Cl-/HCO-3 exchanger on the basolateral
membrane is stimulated by intracellular HCO-3 produced by the hydration of carbon dioxide, which is catalyzed by carbonic anhydrase. Aquaporin-1
channels, which exist on the apical and basolateral membrane regulate the water movement across the choroid epithelia.

conditions, i.e. during direct exposure of the CP, but also
because Pantopaque oil has been shown to cause acute
ventriculitis and inflammatory changes in the CPs. (Clark
et al., 1971).
In rabbits, Welch (1963) developed a new model for
CSF secretion measurement (micropuncture of the princi-
pal vein of the CP) assuming that water loss during CSF
secretion increased hemoglobin and hematocrit concen-
trations in the venous blood. He compared hematocrit or
hemoglobin concentrations of choroidal venous and
mixed arterial blood and calculated the CSF secretion rate
by involving the parameter of blood flow rate (Welch,
1963, 1975). Based on a similar assumption, Pollay
et al. (1972) perfused the sheep CP after rapid removal
of the brain by ligation of the anterior choroidal artery
and cannulation of the internal cerebral vein. They calcu-
lated the CSF secretion based on hematocrit changes in
the inflowing and outflowing fluid and the rate of perfusion.
However, these methods have several important limita-
tions, and the hematocrit levels of aortic and choroidal
blood are probably not equal (Cserr, 1971). And finally,
according to Spector et al. (2015) the direct evidence that
the CSF was actively secreted by the CP was the exper-
iment on cultured CP epithelial cells ‘‘in vitro” (Hakvoort
et al., 1998).
The main criticism, besides the already mentioned,
directed to all these experiments is that they were
performed under non-physiological conditions. Namely,
it is generally accepted that fluid exchange at the
capillary level in living organisms is regulated by
Starling’s forces (oncotic and hydrostatic) operating
across the microvascular walls. The same should apply
to the CP.
According to this view, fluid exchange is determined
by surface area (S), and hydrostatic (DP) and oncotic
(Dp) pressure gradients. At the capillary level, DP
equals the difference between the pressure in the
capillary (Pc) and that in the interstitium (Pt),
DP ¼ Pc  Pt
while Dp equals the oncotic pressure difference between
the capillary and interstitial compartments
Dp ¼ pc  pt
Interstitial oncotic pressure and capillary hydrostatic
cause the fluid movement from the vascular to the
interstitial compartment, while plasma oncotic pressure
pulls the fluid into the vascular space. In addition, under
physiological conditions the interstitial hydrostatic
pressure in many tissues is similar to the atmospheric
 D. Orešković et al. / Neuroscience 354 (2017) 69–87
pressure or slightly subatmospheric (negative), which
causes the fluid movement into the interstitial space.
However, if tissue hydration increases, the interstitial
hydrostatic pressure becomes positive, which changes
the fluid movement into the opposite direction, i.e. into
the vascular compartment. It is obvious that the fluid
exchange depends on the fine-tuning of hydrostatic (DP)
and oncotic (Dp) pressure gradients, regardless of
whether the substances enter or exit the capillaries. And
such gentle and fragile physiological balance is
suddenly exposed to rough experimental approach such
as ‘‘ex vivo” experimental approach, artificial perfusion
of the CP blood stream, interaction between the CP and
aggressive oil, exposure of the CP to atmospheric
pressure, or culturing the CP epithelial cells ‘‘in vitro”. It
is rather naı̈ve to expect that the experimental results
obtained in this way will reflect the real CP exchange of
substances in living organisms under physiological
conditions.
Ionic transport in organism is known to take place bi-
directionally (in and out of the cell), but the current
technology is not able to determine which direction is
dominant. This is especially the case in the mentioned
experiments, since they were designed to deliberately
collect new CSF by glass pipette from the CP surface
covered with oil, or by CP perfusion technique, without
intention (possibility) to simultaneously monitor the
eventual return of substances into the CP blood stream.
But anyhow, these experiments have had a key role in
accepting the CP as an active biological pump and the
main place of CSF formation.
IS THE CHOROID PLEXUS REALLY A CSF
PUMP?
The choroid plexus and entry of water into CSF
system
To answer the question whether the CP is a pump, we
should pay full attention to water. Namely, as was
mentioned, 99% of the CSF volume is water, and all
other substances account for only 1%. So, if we want to
observe the CP as the biological pump that secreted all
the CSF, the key question is how 99% of CSF volume
(water) passes throughout the pump. Because of such a
CSF composition, if the CPs are the main place of CSF
secretion they should also be the main place of water
entry into the CSF system.
However, the experiments that studied water transit
from the blood to CSF did not recognize the CPs as the
main point of water entry into the CSF system
(Orešković and Klarica, 2015; Orešković et al., 2016).
The most illustrative example is a recent study by
Igarashi et al. (2013) in control and in AQP-1 or AQP-4
knocked out mice. As was mentioned, AQP enables the
water movement across biological membranes, which is
why it is very important in the CP transport system of
water. In the mammalian CNS there are several AQP iso-
forms, but only AQP-1 and AQP-4 are relevant for mam-
mals in vivo. AQP-4 is the most widespread isoform,
uniquely restricted to the subpial and perivascular endfeet
of astrocytes (Huber et al., 2012; Igarashi et al., 2013)
75
while AQP-1 is expressed in the CP epithelium (Nielsen
et al., 1993; Dolman et al., 2005). Although AQP-4 was
also partially observed in CPs (Speake et al., 2003), the
authors have suggested that AQP-1 is responsible for
water transport across the CP epithelium during CSF
secretion. If water movement in the CPs is critical for
the CSF volume, AQP-1 rather than AQP-4, should be
expected to play an important role in maintaining the
CSF volume. For this reason, water flux from blood to
CSF of the third ventricle has been examined by MRI
technique (JJVCPE imaging) in highly sophisticated
experiments in control mice and AQP-1 or AQP-4 knock-
out mice (Igarashi et al., 2013). The results have clearly
demonstrated that water influx into CSF is controlled by
AQP-4 (outside the CP), which is responsible for water
homeostasis of the pericapillary space, rather than by
AQP-1, which is located in the CP, i.e. the CP does not
serve as a biological pump in the mouse. Consequently,
it cannot be expected to play a major role in CSF forma-
tion, since no significant contribution of water/CSF by
the CP was observed (no difference in AQP-1 knocked
out mice and control animals). The similar experiments
in mice were performed by ventriculo-cisternal perfusion
as a method for measurement of CSF formation (Oshio
et al., 2005). In these experiments CSF formation was
calculated in wild-type mice (control) and null AQP-1
mice. In contrast to the experiment of Igarashi et al.
(2013), where the difference in entrance of CSF/water into
CSF system between wild-type and AQP-1 free mice had
not been noticed, Oshio et al. (2005) have obtained a
decrease in CSF formation of about 20% in AQP-1 free
mice. The most important difference between those two
experiments is the methodology. While Igarashi et al.
(2013) used new, sophisticated, physiological method,
Oshio et al. (2005) used venticulo-cisternal perfusion
method, which has been strongly criticized as indirect,
unreliable and non-physiological; i.e. one cannot be cer-
tain what the true subject of measurement really was
(see later; Orešković and Klarica, 2014a,b). But even if
these results are to be considered reliable, the question
arises what the observed 20% reduction in CSF formation
really means. It only means that the entrance (formation)
of CSF/water by CP (without AQP-1) is diminished as well
as it is evident that the rest of CSF/water volume (domi-
nant) still enters into the CSF system, obviously preva-
lently by AQP-4, as shown in experiment of Igarashi
et al. (2013). Recently, using dynamic PET study by
H152 O on volunteers and patients with idiopathic normal
pressure hydrocephalus, a very fast movement of water
molecules from the arteries into the brain parenchyma
and ventricular CSF was shown, without proof that CP
is a dominant place of water entrance into the CSF sys-
tem (Mase et al., 2016). In other words, mentioned exper-
iments show that CP is not the main site of CSF/water
entry and that CSF/water enters CSF system from all sur-
rounding tissue. All of the above supports the Bulat-Klar
ica-Orešković hypothesis of CSF physiology (see
Introduction).
A very similar result was obtained in humans before
the discovery of AQPs, in a study that investigated the
effect of bilateral choroid plexectomy on the appearance
76 D. Orešković et al. / Neuroscience 354 (2017) 69–87
of water from blood in the CSF system (Bering, 1952).
After Bering surgically removed both ventricular CPs,
contrary to his expectations, there was no difference in
water exchange (D2O appearance) before and after chor-
oid plexectomy. The observed curves of water entrance
(D2O) were almost the same, which confirms that the
CPs play only a small part in the water exchange of the
CSF inside the cerebral ventricles. Namely, if the CP
was responsible for a greater volume of the water
exchange, the postoperative appearance of D2O would
be much slower than preoperative. Hence, the results
were analogous both in humans and mice, i.e. no signifi-
cant contribution of the CPs to water influx into CSF sys-
tem was observed.
The overlapping curves in CSF after i.v. application of
3
H2O were also found in dogs within the CSF in the lateral
brain ventricles and cisterna magna (Klarica et al., 2013).
This suggests that the mechanism of water entry from the
bloodstream into these compartments is the same. How-
ever, if the CP was the main place of CSF/water secretion
these two curves should be considerably different, with
the highest 3H2O ventricular values at the site where the
CP is positioned. Furthermore, oxygen enhancement in
ventricular and subarachnoidal (sulcal) CSF has recently
been found by spin eco MRI sequence in fifteen healthy
volunteers. These results confirm the involvement of the
cerebral vessels in CSF production (Mehemed et al.,
2014). The same study showed that after administration
of oxygen CSF signal was significantly higher in both sul-
cal CSFs than in the ventricular CSF. These results clo-
sely correspond with the experiments by Bering (1952),
who observed that the D2O appearance was faster in
the cisterna magna than in the brain ventricles. The facts
that the water entry from the blood into the CSF system
does not differ between control and plexectomized
patients and between control and AQP-1 knocked out
mice, and that the appearance of 3H2O in a dog and
D2O and oxygen in human are not fastest in the ventricles
indicate that the CP does not behave as biological pump
for water (99% of CSF volume), it obviously cannot be
regarded a CSF pump.
The choroid plexus and perfusion of the CSF system
After Dandy’s experiment (1919; Fig. 2) in the beginning
of 20th century, the next strongest impetus to explore
CSF physiology came in the 1960s with the emergence
of the ventriculo-cisternal perfusion method, as a tool for
study of CSF formation and absorption (Heisey et al.,
1962). The results obtained by the ventriculo-cisternal
perfusion method have been used as important evidence
in favor of the classic hypothesis, and some perfusion
techniques have been used as direct evidence that CP
is a biological pump of CSF (see The choroid plexus as
a CSF pump). While other methods have mostly been
abandoned or used very rarely, ventriculo-cisternal perfu-
sion still represents the generally accepted method for the
study of CSF secretion (Orešković and Klarica, 2010,
2014a,b). Although it does not directly measure the secre-
tion by CPs, it is believed that the calculated CSF secre-
tion rate originates mainly from the CPs. It is an indirect
method, since the CSF secretion rate is determined on
the basis of marker dilution in the outflow perfusate as a
factor for calculation. Therefore, the marker dilution in
the CSF system is a crucial criterion for CSF secretion,
since it is assumed to be a consequence of the newly
secreted CSF, i.e. higher dilution rate is assumed to result
in a higher CSF secretion rate. However, a thorough sci-
entific analysis by Orešković and Klarica (2010, 2014a,b)
has clearly demonstrated that this criterion and presump-
tions are not fulfilled in experimental conditions. This anal-
ysis has shown that marker dilution inside the CSF
system is not a consequence of the newly secreted
CSF, but of a number of other factors (Orešković and
Klarica, 2014a,b). Furthermore, perfusion method, which
has not been adequately validated, can be used to calcu-
late the CSF secretion in an in vitro model and even in
dead animals, as well as in any perfused part of the
CSF system (Orešković and Klarica, 2014a,b): for
instance in subarachnoid space where CPs do not exist
(Sato et al., 1972, 1975). Therefore, ventriculo-cisternal
perfusion generally accepted as prevalent scientific
method for the measurement secretion of CSF, is a
method that can neither confirm the existence of CSF
secretion nor adequately explain the CPs’ function. Its
many methodological errors have been demonstrated,
requiring a re-evaluation of all the results obtained in this
way (Orešković et al., 2000; Orešković and Klarica, 2010,
2014a,b).
The view that the CP is not a biological pump is also
supported by a series of feline experiments using
isolated brain ventricles (Orešković et al., 2001;
Orešković et al., 2002; Klarica et al., 2009). In these
experiments the aqueduct of Sylvius was blocked by a
plastic cannula surrounded by acrylic glue in a watertight
manner. The cannula was filled by artificial CSF and the
external orifice was adjusted to collect the CSF at physi-
ological CSF pressure (Fig. 4A). Thereby, the ventricles
(two lateral ones and the third) were separated from the
rest of CSF system i.e. the CSF secretion site (CPs)
was isolated from the CSF absorption site (subarachnoid
space). If the CPs were indeed the CSF pumps and if
CSF was not absorbed inside the ventricles, then at phys-
iological CSF pressure, all CSF (secreted by CPs) should
eventually be collected through the plastic cannula in a
collection tube (Fig. 4A). However, in all experimental ani-
mals, the CSF was found to pulsate at the external orifice
of the cannula, there was no net flow of the CSF volume
(circulation), and no amount of CSF was collected in the
tube (Orešković et al., 2001, 2002). In the experiments
where the ventricles were completely blocked by a can-
nula and connected to CSF pressure measuring system
(Fig. 4B), there were no significant ICP differences
between the isolated brain ventricles and those in control
conditions, and during the three hours of obstruction the
brain ventricles were not observed to dilate (Klarica
et al., 2009). This result is very similar to earlier results
by Guleke (1930) and Hassin et al. (1937), who in 30 of
38 and 12 of 15 dogs, respectively, found that the ventri-
cles did not dilate and that hydrocephalus did not develop
after the occlusion of the aqueduct of Sylvius. A similar
case has also been reported by Radoš et al. in a human
patient (2014b). They have presented a female patient
 D. Orešković et al. / Neuroscience 354 (2017) 69–87 77

viously mentioned experiments where

CSF pressure in isolated ventricles
remained within physiological range
and where the ventricles failed to dilate
and the CSF did not escape are prac-
tically inexplicable in the light of the
presumption that the CPs represent
CSF pumps.
An additional confirmation that the
CP does not behave as a biological
pump was obtained in a feline model
of isolated brain ventricles when an
‘‘artificial CP”, i.e. a mechanical
pump, was introduced into the lateral
brain ventricle (Fig. 4C; Orešković
et al., 2002). The infusion rate by the
‘‘artificial CP” (‘‘CSF secretion”) was
adjusted to the expected rate of CSF
secretion (13 mL/min) suggested by
the classic hypothesis (Flexner and
Winters, 1932; Welch, 1963; Plum
Fig. 4. (A) A large pineal cyst with compressed quadrigeminal plate and near-obstruction
stenosis of the aqueduct of Sylvius at sagittal T1 slice on magnetic resonance (MR) imaging and Siesjö, 1975; Pollay, 1975,
exam. (B) Spinal region showing normal morphology of the spinal cord and normal volume of the 1977). Under physiological CSF pres-
spinal CSF at sagittal T2 slices. Presented findings were first detected on an MR exam 6 years sure, mock CSF was infused at a rate
ago, and were without significant changes during that period. Incidental findings of degenerative of 13 mL/min by the ‘‘artificial CP” into
changes of intervertebral disks in the cervical and lumbal region are also visible.
the cat brain ventricle (‘‘CSF secre-
tion”) and the same rate of outflow
fluid volume was obtained in a collect-
ing plastic tube. So, if the CPs in a liv-
ing organism actually secreted CSF
and behaved as a biological pump
then the collected CSF tube should
not be empty (Fig. 4A), but the results
should correspond to the results of the
experiments with the ‘‘artificial CP”
(Fig. 4C). In other words, if the CPs
really secreted CSF, the rate of CSF
secretion would certainly be measured
by the presented method.
All these experimental results in
animals and humans, as well as the
clinical findings in humans, show that
CPs behave neither as biological
pumps nor as the main site of CSF
secretion.

THE CHOROID PLEXUS AS

Fig. 5. (A) A large pineal cyst with compressed quadrigeminal plate and near-obstruction stenosis
of the aqueduct of Sylvius at sagittal T1 slice on the magnetic resonance imaging exam. (B) Spinal
A GENERATOR OF
region showing normal morphology of spinal the cord and normal volume of the spinal CSF at HYDROCEPHALUS
sagittal T2 slice. DEVELOPMENT
In this chapter we will try to answer the
question whether the CPs cause
in whom the aqueduct of Sylvius was obstructed for at hydrocephalus development. The
least 5 years due to a large pineal cyst. In this case, classic hypothesis offers a satisfactory explanation of
despite complete absence of CSF movement through hydrocephalus development, and hydrocephalus
the aqueduct, lateral and third ventricles did not become development confirms the classical hypothesis, as
dilated (Fig. 5; Radoš et al., 2014b). Furthermore, no shown by Dandy, so it is difficult to separately discuss
hydrocephalus development was observed in a male these two concepts (Fig. 2; 1919). Therefore, this
patient with the stenosis of the aqueduct of Sylvius lasting explanation, with minor modifications, is still widely cited
for over a year (Fig. 6). These clinical results and the pre- by studies investigating hydrocephalus development.
78 D. Orešković et al. / Neuroscience 354 (2017) 69–87
Fig. 6. Schematic representation the blockade of the aqueducts of Sylvius by plastic cannula in a cat. (A) Collection of the CSF from the brain
ventricles at physiological CSF pressure. (B) Measuring the CSF pressure in isolated brain ventricles. (C) Collection of the CSF from brain ventricles
at physiological CSF pressure during imitation of choroid plexus function by infusion of artificial CSF (13.00 mL/min).
Since the CSF is produced by the CPs pushing against
ICP (Pollay et al., 1983; Davson et al., 1987), hydro-
cephalus or over-accumulation of the CSF is caused by
medical conditions that block its normal circulation or
absorption (Rekate, 2008).
Let us briefly return to Dandy’s experiment (1919),
which very realistically showed that CP was essential for
hydrocephalus development and offered a suggestion
for hydrocephalus therapy. Namely, when the foramina
of Monro was obstructed, the ventricle with the CP,
because of its nature as a biological pump, dilated, while
the contralateral ventricle without the CP, collapsed
(Fig. 2). From this observation, Dandy concluded the fol-
lowing: ‘‘the only absolute proof that cerebrospinal fluid is
formed from the choroid plexus. At the same time, it is
proven that the ependyma lining the ventricles is not con-
cerned in the production of cerebrospinal fluid.” Since
Dandy was a neurosurgeon, for him it was logical to treat
hydrocephalus by choroid plexectomy. For years, plexec-
tomy was the most widespread type of infantile hydro-
cephalus treatment in the United States. However, over
decades it became clear that CP cauterization or bilateral
extirpation invariably failed to benefit the patients. There-
fore, they were abandoned as a means of hydrocephalus
treatment, and today it has only historical significance
(Milhorat, 1974, 1976). The problem with this treatment
is not only that the results obtained in a canine model
were applied to humans, but also that the results them-
selves were not sufficiently tested. The inability of choroid
plexectomy to cure hydrocephalus is evidence enough
that the CPs are not the main source of active CSF forma-
tion and that CP is not a powerful biological pump.
After choroid plexectomy was proved to be an
inadequate method for hydrocephalus treatment,
endoscopic methods have emerged as a new treatment
modality (Pople and Ettles, 1995; Wellons et al., 2002;
Enchev and Oi, 2008). Although they yielded somewhat
better results, the CP coagulation did not significantly
reduce the ventricular size and only a third of patients
achieved lasting improvement without the need for CSF
shunts (Pople and Ettles, 1995). In the study by Griffith
and Jamjoom (1990) the shunting was required in almost
half of the patients one week to thirteen months after the
CP coagulation. The best example is the case of hydro-
cephalus with idiopathic CSF overproduction described
by Trevisi et al. (2014), in which dual shunting was
required in spite of bilateral endoscopic CP coagulation.
All these cases demonstrate that the hydrocephalus
development can still occur even when the CPs are
removed.
A recent publication, in which the cause of
hydrocephalus has been explained by CP hyperplasia,
also showed no better results (Hallaert et al., 2012). Out
of sixteen patients who underwent plexectomy or CP
coagulation, only four achieved improvement, one had
incomplete medical documentation, whereas eleven
patients required the VP shunt re-installation.
But although the new endoscopic approaches to CPs
failed to provide satisfactory results, they nevertheless
achieved some success (Griffith and Jamjoom, 1990;
Pople and Ettles, 1995; Hallaert et al., 2012). The ques-
tion is whether this success could be attributed to the
CPs removal. Let us analyze a case report (Hallaert
et al., 2012) in which successful endoscopic coagulation
of hyperplastic CPs was performed to treat hydro-
cephalus, and the obtained result was presented as direct
evidence that the CP secreted CSF (Spector et al., 2015),
i.e. that CP is a biological pump. The endoscopic coagu-
lation of both lateral ventricles’ CPs was performed by an
occipital approach on a 3-year-old girl. Throughout the
procedure, the ventricles were rinsed at body temperature
with Ringer solution to prevent them from collapsing,
remove the debris from the tissue and to control a small
bleeding (Hallaert et al., 2012). After surgical treatment,
the CSF was drained through the external ventricular
catheter for 14 days. In accordance with the common
neurosurgical practice, the patient was probably also trea-
ted with antibiotics, although this is not mentioned in the
 D. Orešković et al. / Neuroscience 354 (2017) 69–87
article. Therefore, it is obvious that additional procedures
used during treatment, other than coagulation of CPs,
could have improved the patient’s condition. Namely,
recently it has been proposed that the increased osmolal-
ity of CSF led to ventricular dilatation and hydrocephalus
development (Klarica et al., 1996, 2013; Krishnamurthy
et al., 2009, 2012, 2014). In such case, rinsing of ventri-
cles with Ringer solution, several days of drainage of
CSF and administration of antibiotics could be what
helped the patient. Since there is no information on the
patient’s CSF osmolality (Hallaert et al., 2012), it is not
possible to assume that the mentioned procedure
improved the patient’s condition, and this cannot be said
for the coagulation of the CPs itself. Anyhow, to take into
account a single successful case report (Hallaert et al.,
2012) of CP coagulation as evidence confirming the CP
as the site of CSF secretion (Spector et al., 2015) is not
scientifically justified, especially if we know that in other
eleven analyzed patients with hyperplastic CPs, plexec-
tomy or CP coagulation were unsuccessful in hydro-
cephalus treatment (see above). We could even claim
that these 11 patients could serve as evidence of pre-
cisely the opposite, that CP are not the site of CSF secre-
tion. Therefore, case report observations in humans
should not be quoted as direct evidence for any hypothe-
ses. Only when hypotheses are based on experiments
should a case report be presented as a complement to
experimental results. We must always be aware that in
the case of neurosurgical removal or coagulation of the
CPs (as the hydrocephalus treatment), the procedure
never includes only plexectomy, but also a number of
other clinical interventions.
This can be observed even better when plexectomy is
used to treat hydrocephalus caused by CP tumors. CP
tumors are rare intraventricular neoplasms of
neuroectodermal origin, ranging from benign CP
papillomas (CPP) to malignant CP carcinomas (CPC).
They may be created by epithelial cells of the CP. It is
believed that such pathology leads to increased CSF
secretion (over-secretion; Kahn and Lutros, 1952;
Milhorat et al., 1976b), probably associated with the
obstruction of CSF pathways, thus causing increased
ICP and hydrocephalus. CP tumors are large highly vas-
cularized structures, so hypertrophy of the choroid arter-
ies is a common angiographic feature. The CP tumor is
a site where the blood–brain barrier integrity is disrupted
and the ventricular cavity freely exposed to the harmful
substances. Namely, the substances from blood can
freely enter the CSF, changing its physiological composi-
tion, and together with tumor metabolism products, alter-
ing the CSF homeostasis. Also, the growth of tumor alone
could significantly influence ICP, perfusion of the brain
and movement (pulsation) of the CSF, further affecting
the CSF homeostasis. The usual treatment procedure is
total surgical excision of the CP with a tumor (plexec-
tomy), with or without adjuvant chemo and/or radiother-
apy depending on the patient (Pencalet et al., 1998).
Plexextomy is performed to stop the CSF overproduction,
resulting in normalization of ICP and ultimately in hydro-
cephalus withdrawal. Although surgical excisions are
often successful in terms of tumor treatment, they are
79
not as successful when it comes to hydrocephalus treat-
ment (Milhorat et al., 1976a; Jooma and Grant, 1983;
Ivan et al., 1986) making CSF shunt implantation often
necessary (Raimondi and Gutierrez, 1975; Velasco-
Siles and Raimondi, 1982). For instance, 18 of 23 patients
following surgical treatment required permanent shunting
(Lena et al., 1990). If hydrocephalus was caused by CSF
over-secretion, it should disappear after the removal of
the source of CSF over-secretion (the CP affected by
tumor). Since hydrocephalus can persist after plexectomy
its development obviously cannot be explained in such a
simple way. Hence, the curative effect of plexectomy is
not achieved by removing the over-secreting CP, but by
removing the CP with the tumor as the main cause of
pathological change (see above) in the CSF system.
These results show that we still do not know enough
about hydrocephalus development (Orešković and
Klarica, 2011; Krishnamurthy et al., 2014; Klarica et al.,
2015) and that the role of the CP in hydrocephalus pathol-
ogy is not so important because hydrocephalus can
develop in the same intensity with or without the CPs
(Milhorat, 1969, 1974; Radoš et al., 2014a), and the
removal of the CPs in hydrocephalic patients usually does
not result in healing. Also a recent article (Orešković and
Klarica, 2011) has shown that there are no exact scientific
proofs for the circulatory hypothesis of hydrocephalus
(CP as a pump) and the classic hypothesis of CSF phys-
iology, but just the opposite, that the existence of hydro-
cephalus is the proof that these hypotheses are
groundless.
THE ROLE OF THE CHOROID PLEXUS IN THE
CSF HOMEOSTASIS
CSF homeostasis is the property of the CSF system in
which the variables are regulated in such a way that
internal conditions remain under physiological
conditions. According to the classic hypothesis, the CPs
should have considerable importance in the CSF
homeostasis. Of all the variables that affect the
homeostasis, we will analyze some aspects in which the
CP affects the CSF volume, biochemical composition of
CSF and CSF pressure. As was previously mentioned,
we analyzed only the experimental results obtained in
animals that were comparable to clinical results in
humans. Although the use of animal results requires
caution, avoiding the use of animal results in explaining
human physiology is not quite scientifically correct.
Namely, no mammalian CSF physiology is considerably
different from human CSF physiology, and the greatest
part of the information related to human CSF physiology
has been obtained by experimental animal models.
The choroid plexus and regulation of CSF volume
The CP as the generator of CSF formation/secretion
(Davson et al., 1987; Brown et al., 2004; Spector et al.,
2015) and by its active nature as a biological pump should
account for all of the CSF volume in the entire CSF sys-
tem. However, CSF production is not a regulated or
self-regulated process, which means the CP is a not reg-
ulating or self-regulating organ. Despite decades of
80 D. Orešković et al. / Neuroscience 354 (2017) 69–87
research, the physiological process of the regulation of
CSF secretion by the CPs is still unknown. In accordance
with the classic view, the regulation of the CP function by
CSF pressure is not in physiological range (Heisey et al.,
1962; Rubin et al., 1966; Cutler et al., 1968; Sklar et al.,
1980; Pollay et al., 1983) and the obtained results suggest
that ICP should be considerably or chronically increased
to decrease CSF secretion (Silverberg et al., 2002;
Damkier et al., 2013). Also, it has not been confirmed that
the autonomic nervous system regulates CSF secretion,
reflecting the technical limitations associated with quanti-
fying CSF secretion (Lindvall and Owman, 1981). Further-
more, a cutback in CSF absorption would not result in a
cutback in CSF secretion, as expected if constant physio-
logical CSF volume is to be retained. The CSF secretion
would, more or less, continue at the same rate regardless
of the CSF absorption. Additionally, one of the most
important evolutionary characteristics of organs is the
ability to modify their function in response to changes in
the organism. The function of an organ depends on inter-
nal and external factors (conditions that specifically affect
the work of individual organs, sleeping, physical activity,
psychological state, coldness, heat etc). In each of these
situations, organ activity is permanently altered and
adapted to new situations to maintain homeostasis. How-
ever, the CP will continue to secrete the CSF at
unchanged physiological rate even if it resulted in
increased CSF pressure, hydrocephalus or a life-
threatening condition. It is difficult to imagine that evolu-
tion would develop an organ that could by its primary
physiological function (secretion) pose a potential threat
to one’s life. This shows how difficult it is to support the
thesis that the CPs are the main regulators of the CSF
volume. Anyhow, the CPs, which are located in the brain’s
ventricles, should, according to the classic hypothesis, be
equally important for internal and external CSF volume,
although this does not seem to be the case in both ani-
mals and humans.
Studies in some sharks, in which there is no
communication between internal and external CSF
showed that in spite of the presence of the CP in
isolated brain ventricles, there was no ventricular
dilatation in physiological conditions under stable and
constant presence of the CSF in the subarachnoid
space (Kappers, 1958; Oppelt et al., 1964). Furthermore,
there is a report on a female patient, who despite an
obstruction of the aqueduct of Sylvius and complete
absence of CSF movement through the aqueduct did
not develop hydrocephalus for over 5 years of follow up
(see Is the choroid plexus really CSF pump?) (Radoš
et al., 2014a). Today, a year later the obstruction still per-
sists and there are no signs of hydrocephalus develop-
ment (Fig. 5) and no clinical symptoms such as
headache, nausea, vomiting, ataxia, dementia, etc. In
light of the presumption that the CPs represent the CSF
pumps, the absence of hydrocephalus and other symp-
toms is inexplicable. It is hard to explain CSF homeosta-
sis and normal CSF volume with a constant CSF turnover
in front and behind the obstruction site in light of classic
hypothesis in this patient (Fig. 5) as well as in the patient
with the stenosis of the aqueduct of Sylvius lasting for
over a year (Fig. 6). Namely, in the latter patient there
was a considerably larger CSF volume behind the
obstruction than in the ventricles (approximately 10:1),
which should be sustained only by one CP located in
the IV ventricle. Also, even more difficult is to offer a rea-
sonable explanation in sharks. How can there exist the
turnover of CSF and constant external volume of CSF
when there is no open communication between internal
and external CSF system, i.e. how can basic daily physi-
ological function be maintained in external CSF without
the presence of the CPs?
Furthermore, the CSF is formed in some lower
vertebrates that do not have the CPs in the ventricular
cavities (Kappers, 1958; Oppelt et al., 1964; Cserr,
1971), as well as within the neural tube of fetal pigs
(Weed, 1917) and humans (Kappers, 1958) even before
the appearance of the CPs anlage (Bueno et al., 2014).
Therefore, during the entire lifetime in some species, or
only during embryonic/fetal development in others, the
CSF is normally produced although the CPs (as a main
place of CSF secretion) do not exist. A very similar case
has recently been reported in a patient with hydranen-
cephaly and macrocephaly (Fig. 7; Radoš et al., 2014b).
This study presented a thirty-year old female with an
extremely large intracranial space and hydrocephalus
mainly filled with the CSF (the CSF occupied approxi-
mately 95% of the cranial cavity) and with only small
areas of sustained parenchyma. Since in this patient
there were no CPs inside the supratentorial region, and
she had free communication inside the CSF system, the
classic hypothesis cannot explain the development of so
large hydrocephalus and macrocephaly. It is also difficult
to explain how such a large CSF volume can be renewed
every day and how the CSF homeostasis can be main-
tained for over 30 years in the absence of the CPs.
Equally difficult to explain is the clinical finding of CP cal-
cification (Modic et al., 1980). Namely, although the calci-
fied CP was not histopathologically analyzed, calcification
is expected to diminish active CSF formation (secretion),
especially when we know which complex physiological
processes take place in the CP during secretion
(Fig. 3). However, the fact that the volume (turnover) of
the CSF in the patent was not reduced (Modic et al.,
1980), certainly fails to speak in favor of the CP as the bio-
logical pump and a regulator of constant CSF volume.
The choroid plexus and biochemical composition of
the CSF
As the main source of the CSF, the CPs should be
responsible for physiological composition of the CSF
constituents (substances). Since the CSF is formed
inside the brain ventricles (by CPs) and absorbed on the
surface of the brain, its constituting elements should not
differ significantly inside the CSF system. However,
there is no universal CSF in terms of fluid composition,
since it depends on the site from where it was taken.
The ventricular CSF is different from the lumbar CSF,
and fluid that is taken directly from the CP is quite
different from mixed ventricular CSF. Therefore, there is
really not a single CSF, and the CSF composition
reflects the composition of the brain extracellular fluid
 D. Orešković et al. / Neuroscience 354 (2017) 69–87 81

Fig. 7. (A): Almost complete malacia of the cerebral hemispheres and spared falx and meninges across the supratentorial space at a transversal T2
slice. (B) Preserved parts of the brain parenchyma in the right frontal lobe, mediobasal left temporal lobe, diencephalon, and brainstem at a coronal
T2 slice. (C) Visible remnants of the occipital lobe, diencephalon, brainstem and cerebellum at a sagittal T2 slice. All figures show marked thickening
of the neurocranial bones. Reproduced with permission of the publisher from Radoš et al. (2014a).

(Katzenelborgen, 1935; Bering, 1966; Davson, 1967;
Bulat and Klarica, 2011). It is also well known that con-
centration gradient for different substances exist along
the CSF system (Garelis and Sourkes, 1973; Sjostrom
et al., 1975; Ruckebusch and Sutra, 1984; Degrell and
Nagy, 1990; Orešković et al., 1995). In other words, the
biochemical composition of CSF is different depending
on the CSF site, and this difference is constant. It means
that the exchange of water and other substances between
CSF and the surrounding tissue (see; Is the choroid
plexus really a CSF pump?), takes part everywhere in
the CSF system (Bulat and Klarica, 2011). It is well known
that CSF communicates freely with the brain extracellular
fluid and that substances can diffuse from the CSF into
the CNS under concentration gradient where they are
removed mainly by active transport into the capillaries
(Zmajević et al., 2002). Recently, so-called glymphatic
pathways were proposed for substances distribution
through the CNS parenchyma (Iliff et al., 2012; Jessen
et al., 2015; Ratner et al., 2017). It was suggested that
substance distribution is supported by arterial pulsations,
and slow vasomotion together with respiration drove the
substances into the perivascular space, in the direction
from arteries to the venous system. The reverse process,
from the spinal tissue into the spinal CSF, was also pre-
sented and some substances in the lumbar CSF were
clearly shown to be derived mostly or exclusively from
the spinal cord (Bulat, 1977; Bulat and Živković, 1978).
Such substance exchange between the CNS and CSF
could explain why CSF composition depends on the place
of CSF collection and why a concentration gradient of
substances could exist along the CSF system. Further-
more, the chemical composition of the CSF collected from
the ventricles in dogs (Bering, 1966), the cisterna magna
in rhesus monkeys (Milhorat, 1969; Milhorat et al., 1971;
Hammock and Milhorat, 1973;) and the lumbar CSF in
humans (Milhorat, 1976) was not altered after bilateral
choroid plexectomy. Recently Hladky and Barrand in their
extensive analysis of BBB and CP role in transport of sub-
stances have concluded (2016): ‘‘Although the BBB is
vital for effective brain function, it is less clear that the
CPs are needed in the adult. Thus wile secretion by the
CPs does account for most of the net fluid entry into the
brain, a suitable arrangement of transporters at the BBB
could easily produce a similar secretion”, which is incom-
patible with traditional view, but is completely in accor-
dance with our hypothesis.
Hence, if the CSF composition depends on the site
from which it was taken, if there is a concentration
gradient for different substances along the CSF system
and if the chemical composition of CSF is not altered
with or without CPs, it is obvious that the CP cannot be
the exclusive and dominant place of CSF exchange and
the main source of biochemical composition of the CSF.
The choroid plexus and CSF pressure
After having listed all these arguments, further discussion
about the CP as biological pump seems almost
unnecessary. However, since one of the important roles
of the CP should be the establishment of CSF
hydrostatic pressure gradient, necessary for the CSF
circulation, the effect of the CP on ICP/CSF pressure
should be analyzed. It is true that no study has
discussed the CP as a pump that creates hydrostatic
gradients, but since it is believed that in hypertensive
hydrocephalus the CP can cause high ICP/CSF
pressure, it seems logical that it can also create
hydrostatic gradients. The unidirectional movement
(circulation) of the CSF cannot happen without the
existence of hydrostatic pressure gradient with the
highest value at the site of CSF formation and the
lowest at the site of CSF absorption. No flow, especially
not the CSF flow, can exist if without a difference in
hydrostatic pressure, which should be permanently
present. The key question is whether such hydrostatic
pressure relationship exists in the CSF system and if it
is a consequence of metabolic activity of CP.
82 D. Orešković et al. / Neuroscience 354 (2017) 69–87
As was elaborated above, the CP should be the main
regulator of ICP/CSF pressure, which is best manifested
by the equation of Marmarou et al. (1978) (see Introduc-
tion). The ICP value will depend on the force of CSF
secretion by the CP and on the resistance to this force
during CSF circulation and absorption. This means that
the physiological secretion of the CSF will generate higher
CSF pressure on higher resistance to CSF circulation and
absorption.
Having shown which factors, according to the classic
hypothesis, affect the occurrence of CSF pressure, it is
necessary to consider whether and what kind of
hydrostatic pressure gradient exists in the CSF system.
It was clearly shown that the ICP/CSF pressure
depends on the body position (Magnaes, 1976a,b;
Klarica et al., 2014). We will analyze the two most impor-
tant positions in which people spend most of their lifetime;
the horizontal and vertical position. The normal CSF pres-
sure in the horizontal recumbent position is around 15 cm
H2O, with the same pressure values inside the spinal sub-
arachnoid space and the cranium (Fig. 8A; P1, P2; P3)
(Magnaes, 1976,b; Davson et al., 1987). Thus, in the hor-
izontal position there is no hydrostatic pressure gradient
that would be necessary for CSF circulation inside the
CSF system. This means that the CSF could not be able
to circulate when people are in this position (e.g. sleeping
at night), that is for a third of their life. In the vertical posi-
tion (standing or sitting; Fig. 8B, C) ICP/CSF pressure in
the cranium is subatmospheric (negative;
P3 = 10 cm H2O) (Bradly, 1970; Fox et al., 1973;
Portnoy et al., 1973; McCullough and Fox, 1974; Welch,
1980; Chapman et al., 1990; Poca et al., 2006; Klarica
et al., 2014), in the upper cervical region or in the foramen
magnum it is similar to atmospheric (zero level;
P2 = 0 cm H2O) and in the lumbar region it is supra-
atmospheric (highly positive; P1 = +60 cm H2O) and cor-
responds to the distance between the cisterna magna
(CM) and the measuring site (Loman, 1934;
Masserman, 1934; Loman et al., 1935; Von Storch
et al., 1937; O’Connell, 1943; Magnaes, 1976a,b). Hence,
while in the horizontal body position the hydrostatic pres-
sure gradient is not present (Fig. 8A), in vertical body
position it is present, but it has an opposite value to the
expected gradients, which should allow the CSF circula-
tion (from the brain ventricles via CM to the lumbal
space). Although it would be expected that the pressure
was the highest in the brain ventricles where the CPs
secrete the CSF, lower in the CM and finally the lowest
in the lumbar region, it is actually the highest in the lumbar
region, lower in CM and the lowest in the brain’s ventricles
(Fig. 8B). Obviously, no hydrostatic pressure gradient
exists in the CSF system, which confirms the hypothesis
of unidirectional circulation from the proposed site of
CSF formation (brain ventricles) along the CSF system
to the CSF absorption site at the brain surface. This also
means that the existing hydrostatic gradient in the CSF
system (Fig. 8B) has nothing to do with the CPs, because
hydrostatic pressure is low and sub-atmospheric (nega-
tive; Fig. 8B, C) at the place where CPs as biological
pump should secrete the CSF (ventricles) and is the high-
est at the opposite site, where there are no CPs (lumbar
space) (Fig. 8B). This all suggests that the CPs are not
involved in ICP/CSF pressure regulation and that the
CSF pressure does not depend on the resistance to
CSF flow and resistance to CSF absorption, as was pro-
posed by Marmarou et al. (1978). Recently a new hypoth-
esis based on experimental and clinical data explained
the origin of such hydrostatic pressure gradient and regu-
lation of ICP/CSF pressure (Klarica et al., 2014). It states
that the CSF movement and composition of substance
within the CSF are not affected by pressure gradients
but that the CSF is carried bi-directionally along the
CSF space affected by the laws of fluid mechanics and
systolic-diastolic pulsations in the blood circulation. In
short, the participation of the CPs in the CSF homeostasis
is not as important as it is shown in the literature.
CONCLUSIONS
The actual role of the CPs in CSF physiology has so far
not been sufficiently explored. It could be said for
certain that the explanation of this role presented in the
textbooks, atlases and review articles is still unclear and
mainly overemphasized. This article showed that the
CPs are not biological pumps and the main sites of CSF
secretion; that do not participate in the regulation of
ICP/CSF pressure; that do not create hydrostatic
pressure gradient in the CSF system and that this
gradient is not permanent (disappeared in the horizontal
position); and that they do not generate the imagined
unidirectional CSF circulation, hydrocephalus
development and increased ICP/CSF pressure. We
strongly believe that the CP is not of crucial importance
for a living organism since no health deterioration was
noticed in patients in whom they were removed (see
above). It is scientific fact that the CPs have active
metabolic nature and that they are highly vascularized
organs immersed into CSF in the brain ventricles
(Damkier et al., 2013; Hladky and Barrand, 2014;
Spector et al., 2015), which supports the idea that their
activity is not one-way traffic from the blood to the CSF
(Hassin, 1924; Di Chiro, 1964; Milhorat et al., 1970).
Instead, the substance exchange takes place in both
directions: into the CSF from the blood and out of the
CSF into the blood. If blood vessels are immersed into
the CSF (as is a case by blood vessels in CP), a fast
and considerable substance exchange could be achieved
between the blood and the CSF, thus maintaining the bio-
chemical balance of the CSF, which should be important
for normal physiological CNS functioning. Also in Fig. 3A
we can see that the CPs are supplied by both anterior
and posterior sets of choroidal arteries and with extremely
rich venous network: choroidal arteries are responsible for
delivering substances to the CP and the venous network
is responsible for removing these substances from the
CP. This observation is consistent with the previously
published data in hydrocephalic children, where it was
suggested that, at least in hydrocephalus, the CP has a
considerable absorption role (Milhorat et al., 1970). Fur-
thermore, it was shown by histopathological facts that
the CP under some pathological conditions certainly rep-
resents an organ of CSF absorption (Hassin, 1924).
 D. Orešković et al. / Neuroscience 354 (2017) 69–87 83

Fig. 8. The scheme of CSF pressure measurement in different parts of the CSF system in different head and body positions. (A) CSF pressure in
the horizontal position of the head and body (right recumbent position) is equal at each recording point (lumbar subarachnoid space, cisterna
magna, intracranial space – lateral ventricle; P1 = P2 = P3), and its value corresponds to the distance between the medial interspinal line and the
parieto-cortical subarachnoid space (h). Thus, CSF pressure in that position can be calculated using the equation for hydrostatic pressure:
P = Pa + qgh, where P is CSF pressure, Pa is atmospheric pressure, q is density, and g is gravitational field acceleration. If the referent pressure
(Pa) is chosen to be zero, CSF pressure is qgh. (B) In the vertical (sitting) position CSF pressure in the cisterna magna (P2) is equal to
atmospheric pressure (Pa). In the lumbar subarachnoid space CSF pressure (P1) is positive (P1 = Pa + qgh1), and its value corresponds to the
distance (CSF hydrostatic column) (h1) between the recording point and the level of referent pressure (Pa). CSF pressure in the cranial CSF system
is subatmospheric (negative or less than atmospheric pressure), and its value corresponds to the distance (h3) between the level of referent
pressure (Pa) and the recording point inside the cranium. (C) Scheme of our hypothesis which explains the appearance of subatmospheric CSF
pressure in the cranial cavity in a vertical position, with unchanged intracranial fluid volume. The scheme of a plastic tube, filled with fluid and open at
the lower end, is presented on right side. Fluid pressure inside the plastic tube at the recording point is subatmospheric, and its value (cm H2O)
corresponds to the distance (h0 3) between the recording point and the open end (P’ = Pa  qgh0 3). Therefore, inside this kind of space
subatmospheric (negative) pressure appears without the fluid volume changes which is in accordance to the law of fluid mechanics. Consequently,
the CSF inside the cranium should also behave according to the same law (left side of the scheme). Thus, the negative value of the hydrostatic CSF
pressure in the cranial cavity does not depend on the shape of the volume, but only on the distance (h3) between the foramen magnum and the
measurement point (P3 = Pa  qgh3) (for details see Klarica et al., 2014).

The above-mentioned results and controversies
connected with the CPs do not fit into the classic
hypothesis, but can be explained by the Bulat-Klarica-Or
ešković hypothesis of CSF physiology, which offers a
different perspective and explanations for the
mechanisms that are at work in CSF physiology and
pathophysiology (see Introduction). In conclusion, we
believe that if we want to achieve better results in the
treatment of diseases related to the CSF it is important
to leave the traditional framework of thinking in CSF
physiology behind.
Acknowledgments—This work has been supported by the Min-
istry of Science, Education and Sport of the Republic of Croatia
(Projects: 1. Serotonergic modulation of obesity: cross-talk
between regulatory molecules and pathways. No. IP-2014-09-
7827; and 2. Pathophysiology of cerebrospinal fluid and intracra-
nial pressure. No. 108-1080231-0023).
84 D. Orešković et al. / Neuroscience 354 (2017) 69–87
REFERENCES
Abbott NJ (2004) Evidence for bulk flow of brain interstitial fluid:
significance for physiology and pathology. Neurochem Int
45:545–552.
Agre P (2006) The aquaporin water channels. Proc Am Thorac Soc
3:5–13.
Bergsneider M (2001) Evolving concepts of cerebrospinal fluid
physiology. Neurosurg Clin N Am 12:631–638.
Bering Jr EA (1952) Water exchange of central nervous system and
cerebrospinal fluid. J Neurosurg 9:275–287.
Bering Jr EA (1955) Choroid plexus and arterial pulsation of
cerebrospinal fluid. Demonstration of the choroid plexuses as a
cerebrospinal fluid pump. AMA Arch Neurol Psychiatry
73:165–172.
Bering Jr EA (1962) Circulation of cerebrospinal fluid. Demonstration
of the choroid plexuses as the generator of the force flow of fluid
and ventricular enlargement. J Neurosurg 19:405–413.
Bering Jr EA (1966) Pathophysiology of hydrocephalus. In: Shulman
K, editor. Workshop on hydrocephalus. Philadelphia: Univer. of
Pennsylvania Press. p. 9–28.
Bradbury MWB (1981) Lymphatics and central nervous system.
Trends In Neurosc 4:100–101.
Bradly KC (1970) Cerebrospinal fluid pressure. J Neurol Neurosurg
Psychiat 33:387–397.
Brierly JF, Field EJ (1948) The connections of the cerebrospinal fluid
space with the lymphatic system. J Anat 82:153–166.
Brinke T, Stopa E, Morrison J, Klinge P (2014) A new look at
cerebrospinal fluid circulation. Fluid Barriers CNS 11:10.
Brodbelt A, Stoodley M (2007) CSF pathways: a review. Br J
Neurosurg 21:510–520.
Brown PD, Davies SL, Speake T, Millar ID (2004) Molecular
mechanisms of cerebrospinal fluid production. Neuroscience
129:957–970.
Bueno D, Parvas M, Garcia-Fernandez J (2014) The embryonic
blood-cerebrospinal fluid barrier function before the formation of
the fetal choroid plexus: role in cerebrospinal fluid formation and
homeostasis. Croat Med J 55:306–316.
Buishas J, Gould IG, Linninger AA (2014) A computational model of
cerebrospinal fluid production and reabsorption driven by Starling
forces. Croat Med J 55:481–497.
Bulat M (1977) On the cerebral origin of 5-hydroxyindoleacetic acid in
the lumbar cerebrospinal fluid. Brain Res 122:388–391.
Bulat M, Klarica M (2011) Recent insights into a new hydrodynamics
of the cerebrospinal fluid. Brain Res Rev 65:99–112.
Bulat M, Živković B (1978) Exchange of 5-hydroxyindoleacetic acid
between the spinal cord and lumbar cerebrospinal fluid. J Physiol
275:191–197.
Chapman PH, Cosman ER, Arnold MA (1990) The relationship
between ventricular fluid pressure and body position in normal
subjects and subjects with shunts: a telemetric study. Neurosurg
26:181–189.
Clark RG, Milhorat TH, Stanley WC, DiChiro G (1971) Experimental
Pantopaque ventriculography. J Neurosur 34:387–395.
Cserr HF (1971) Physiology of the choroid plexus. Physiol Rev
51:273–311.
Cserr HF (1988) Role of secretion and bulk flow of brain interstitial
fluid in brain volume regulation. Ann N Y Acad Sci 529:9–20.
Cushing H (1914) Studies on the cerebro-spinal fluid: I Introduction. J
Med Res 31:1–19.
Cushing H (1925) The third circulation and its channels (Cameron
Lecture). Lancet 2:851–857.
Cutler RWP, Page L, Galichich J, Waters GV (1968) Formation and
absorption of cerebrospinal fluid in man. Brain 91:707–720.
Damkier HH, Brown PD, Praetorius J (2013) Cerebrospinal fluid
secretion by the choroid plexus. Physiol Rev 93:1847–1892.
Dandy WE (1919) Experimental hydrocephalus. Ann Surg
70:129–142.
Dandy WE, Blackfan KD (1914) Internal hydrocephalus:
experimental, clinical and pathological study. Am J Dis Child
8:406–482.
Davson H (1967) Physiology of the cerebrospinal
fluid. London: Churchill.
Davson H, Welch K, Segal MB (1987) Physiology and
pathophysiology of the cerebrospinal fluid. Edinburgh: Churchill-
Livingstone.
de Rougemont J, Ames 3rd A, Nesbett FB, Hofmann HF (1960) Fluid
formed by choroid plexus; a technique for its collection and
comparison of its electrolyte composition with serum and cisternal
fluids. J Neurophysiol 23:485–495.
Degrell I, Nagy E (1990) Concentration gradients for HVA, 5-HIAA,
ascorbic acid and uric acid in cerebrospinal fluid. Biol Psychiatry
15:891–896.
Di Chiro G (1964) Movement of the cerebrospinal fluid in human
beings. Nature 204:290–291.
Dolman D, Drndarski S, Abbot NJ, Rattray M (2005) Induction of
aquaporin 1 but not aquaporin 4 messenger RNA in rat primary
brain microvessels endothelial cells in culture. J Neurochem
93:825–833.
Enchev Y, Oi S (2008) Historical trends of neuroendoscopic surgical
techniques in the treatment of hydrocephalus. Neurosurg Rev
31:249–262.
Flexner IB, Winters H (1932) The rate of formation of cerebrospinal
fluid in etherized cats. Am J Physiol 101:697–710.
Fox JL, McCullough DC, Green RC (1973) Effect of cerebrospinal
fluid shunts on intracranial pressure and on cerebrospinal fluid
dynamics. J Neurol Neurosurg Psychiat 36:302–312.
Garelis E, Sourkes TL (1973) Sites of origin in the central nervous
system of monoamine metabolites measured in human
cerebrospinal fluid. J Neurol Neurosurg Psychiatry 36:625–629.
Greitz D (1993) Cerebrospinal fluid circulation and associated
intracranial dynamics. A radiographic investigation using MR
imaging and radionuclide cisternography (Ph D.
thesis). Stockholm: Karolinska University. p. 80.
Greitz D, Wirestam R, Franck A, Nordell B, Thomsen C, Stahlberg E
(1992) Pulsatile brain movement and associated hydrodynamics
studied by magnetic resonance phase imaging. The Monro_Kellie
doctrine revised. Neuroradiology 34:370–380.
Griffith HJ, Jamjoom AB (1990) The treatment of childhood
hydrocephalus by choroid plexus coagulation and artificial
cerebrospinal fluid perfusion. Br J Neurosur 4:95–100.
Guleke N (1930) Über die Entstehung des Hydrocephalus internus.
Arch Klin Chir 162:533–550.
Hakvoort A, Hasellbach M, Wegner J, Hoheisel D, Galla HJ (1998)
The polarity of choroid plexus epithelia cells in vitro is improved in
serum-free medium. J Neurochem 71:1141–1150.
Hallaert GG, Vanhauwaert DJ, Logghe K, Van de Brocke C, Baret E,
Van Roost D, Caemaert J (2012) Endoscopic coagulation of
choroid plexus hyperplasia. J Neurosurg Pediatrics 9:169–177.
Hammock MK, Milhorat TH (1973) Recent studies on the formation of
cerebrospinal fluid. Dev Med Child Neurol 15(Suppl. 29):27–34.
Hassin GB (1924) Notes on the nature and origin of the cerebrospinal
fluid. J Nerv Ment Dis 59:113–121.
Hassin GB, Oldberg E, Tinsley M (1937) Changes in the brain in
plexectomized dogs. With comments on the cerebrospinal fluid.
Arch Neurol Psychiat 38:1224–1239.
Heisey SR, Held D, Pappenheimer JR (1962) Bulk flow and diffusion
in the cerebrospinal fluid system of the goat. Am J Physiol
203:775–781.
Hladky SB, Barrand MA (2014) Mechanisms of fluid movement into,
through and out of the brain; evaluation of the evidence. Fluids
Barriers CNS 11:26.
Huber VJ, Tsujita M, Nakada T (2012) Aquaporin in drug discovery
and pharmacotherapy. Mol Aspects Med 33:691–703.
Igarashi H, Tsujita M, Huber VJ, Kwee IL, Nakada T (2013) Inhibition
of aquaporin-4 significantly increases regional cerebral blood flow.
NeuroRiport 24:324–328.
Iliff JJ, Wang M, Liao Y, Plogg BA, Peng W, Gundersen GA,
Benveniste H, Vates GE, Deane R, Goldman SA, Nagelhus EA,
Nedergaard M (2012) A paravascular pathway facilitates CSF
flow through the brain parenchyma and clearance of interstitial
solutes, including amyloid b. Sci Transl Med 4:147ra111.
 D. Orešković et al. / Neuroscience 354 (2017) 69–87
Ivan LP, Martin DJ, Mallya KB, Schneider E (1986) Choroid plexus
papilloma ina 4-month-old child: A case report. J Child Neurol
1:53–55.
Jessen NA, Munk ASF, Lundgaard I, Nedegaard M (2015) The
glymphatic system: A biginner’s guide. Neurochem Res
40:2583–2599.
Johanson CE (2008) Choroid plexus-CSF circulatory dynamics:
Impact on brain growth, metabolism and repair. In: Totowa CP,
editor. Neuroscience in medicine. New Jersey: The Humana
Press. p. 173–200.
Johanson CE, Duncan JAIII, Klinge PM, Brinker T, Stopa EG,
Silveberg GD (2008) Multiplicity of cerebrospinal fluid functions:
new challenges in health and disease. Cerebrospinal Fluid Res
5:10. http://dx.doi.org/10.1186/1743-8454-5-10.
Johnston M, Zakharov A, Koh L, Armstrong D (2005) Subarachnoid
injection of Microfil reveals connections between cerebrospinal
fluid and nasal lymphatic in the non-human primate. Neuropathol
Appl Neurobiol 31:632–640.
Jooma R, Grant DN (1983) Third ventricle choroid plexus papillomas.
Child’s Brain 10:242–250.
Kahn EA, Lutros JT (1952) Hydrocephalus from overproduction of
cerebrospinal fluid (and experience with other papilloma of
choroid plexus). J Neurosurg 9:59–67.
Kanaka C, Ohno K, Okabe A, Kuryama K, Itoh T, Fukuda A, Sato K
(2001) The differential expression patterns of messenger RNAs
encoding K-Cl cotransporters (KCC1,2) and Na-K-2Cl
cotransporter (NKCC1) in the rat nervous system. Neurosci
104:933–946.
Kappers JA (1958) Structural and functional changes in the
telenchepalic choroid plexus during human ontogenesis. In:
Wolstenholme GEW, O’Connor GM, editors. Ciba Found Symp
Cerebrospinal Fluid. Boston: Little Brown. p. 3–31.
Katzenelborgen S (1935) The cerebrospinal fluid and its relation to
the blood. Baltimore: Johns Hopkins Press.
Klarica M, Jukić T, Miše B, Kudelić N, Radoš M, Orešković D (2015)
Experimental spinal stenosis in cats: new insight in mechanisms
of hydrocephalus development. Brain Pathol. 101111/bpa.12337.
Klarica M, Miše B, Vladić A, Radoš M, Orešković D (2013)
‘‘Compensated hyperosmolarity” of cerebrospinal fluid and the
development of hydrocephalus. Neuroscience 248:278–289.
Klarica M, Mitrović M, Orešković D, Jeren T, Gmajnički B, Bulat M
(1996) Osmolarity of cerebrospinal fluid during experimental
meningitis and post mortem. Period Biol 98:41–43.
Klarica M, Orešković D (2014) Enigma of cerebrospinal fluid
dynamics. Croat Med J 55:287–290.
Klarica M, Orešković D, Božić B, Vukić M, Butković V, Bulat M (2009)
New experimental model of acute aqueductal blockade in cats:
Effects on cerebrospinal fluid pressure and the size of brain
ventricles. Neuroscience 158:1397–1405.
Klarica M, Radoš M, Erceg G, Petošić A, Jurjević I, Orešković D
(2014) The influence of body position on cerebrospinal fluid
pressure gradient and movement in cats with normal and impaired
craniospinal communication. PLoS One 9(4):e95229. 1371/
journal.pone. 0095229.
Koh L, Zakharov A, Nagra G, Armstrong D, Friendship R, Johnston M
(2006) Development of cerebrospinal fluid absorption sites in the
pig and rat: connections between the subarachnoid space and
lymphatic vessels in the olfactory turbinates. Anat Embryol
211:335–344.
Krishnamurthy S, Li J, Schultz L, Jenrow KA (2012) Increased CSF
osmolarity reversibly induces hydrocephalus in the normal rat
brain. Fluids Barriers CNS. http://dx.doi.org/10.1186/2045-8118-
9-13.
Krishnamurthy S, Li J, Schultz L, McAllister JPII (2009) Intraventricular
infusion of hyperosmolar dextran induces hydrocephalus: a novel
animal model of hydrocephalus. Cerebrospinal Fluid Res. http://dx.
doi.org/10.1186/1743-8454-6-16.
Krishnamurthy S, Tichnenor MD, Satish AG, Lehmann DB (2014) A
proposed role for efflux transporters in the pathogenesis of
hydrocephalus. Croat Med J 55:366–376.
85
Lena G, Genitori L, Molina J, Legatte JRS, Choux M (1990) Choroid
plexus tumors in children. Review of 24 cases. Acta Neuroch
106:68–72.
Li H, Tornberg J, Kaila K, Airaksinen MS, Rivera C (2002) Patterns of
cation-cloride cotransporter expression during embryonic rodent
CNS development. Eur J Neurosci 16:2358–2370.
Lindvall M, Owman C (1981) Autonomic nerves in the mammalian
choroid plexus and their influence on the formation of
cerebrospinal fluid. J Cereb Blood Flow Metab 1:245–266.
Linninger AA, Tangen K, Hsu C-Y, Frim D (2016) Cerebrospinal fluid
mechanics and its coupling to cerebrovascular dynamics. Annu
Rev Fluid Mechan 48:219–257.
Loman J (1934) Components of cerebrospinal fluid pressure as
affected by changes in posture. Arch Neurol Psychiat
31:679–683.
Loman J, Myerson A, Goldman D (1935) Effects of alterations in
posture on the cerebrospinal fluid pressure. Arch Neurol Psychiat
33:1279–1295.
Lun PM, Monuki ES, Lehtinen MK (2015) Development and functions
of the choroid plexus – cerebrospinal fluid system. Nat Rev
Neurosci 16:445–457.
Lyons MK, Meyer FB (1990) Cerebrospinal fluid physiology and the
management of increased intracranial pressure. Mayo Clin Proc
65:684–707.
Magnaes B (1976a) Body position and cerebrospinal fluid pressure.
Part 1: Clinical studies on the effect of rapid postural changes. J
Neurosurg 44:687–697.
Magnaes B (1976b) Body position and cerebrospinal fluid pressure.
Part 2: Clinical studies on the orthostatic pressure and the
hydrostatic indifferent point. J Neurosurg 44:698–705.
Marmarou A, Shulman K, Rosende RM (1978) A nonlinear analysis of
the cerebrospinal fluid system and intracranial pressure
dynamics. J Neurosurg 48:332–344.
Mase M, Hayashi E, Yamada H, Oshima N, Aoyama K, Hibino S, Ito
Y, Iida A, Miyati T, Mori E (2016) Water turnover in brain and
ventricles in normal volunteers and patients with idiopathic NPH:
dynamic PET study using H15 2 O. Abstract Book of 16th
international symposium on intracranial pressure and
neuromonitoring, Cambridge, USA, O58:119-120.
Masserman JH (1934) Cerebrospinal hydrodynamics. 4 Clinical
experimental studies. Arch Neurol Pshyatry (Lond)
32:523–553.
McCullough DC, Fox JL (1974) Negative inttracranial pressure
hydrocephalus in adults with shunts and its relationship to
production of subdural hematoma. J Neurosurg 40:372–375.
Mehemed TM, Fuchimi Y, Okada T, Yamamoto A, Kanagaski M, Kido
A, Fujimoto K, Sakashita N, Togashi K (2014) Dynamic oxygen-
enhanced MRI of cerebrospinal fluid. PLoS One 9(6):e100723.
http://dx.doi.org/10.1371/journal.pone.0100723.
Milhorat TH (1969) Choroid plexus and cerebrospinal fluid
production. Science 166:1514–1516.
Milhorat TH (1974) Failure of choroid plexectomy as treatment for
hydrocephalus. Surg Ginecol Obstet 139:505–508.
Milhorat TH (1975) The third circulation revisited. J Neurosurg
42:628–645.
Milhorat TH (1976) Structure and function of the choroid plexsus and
other sites of cerebrospinal fluid formation. Int Rev Cytol
47:225–288.
Milhorat TH, Hammock MK, Chien T, Davis DA (1976a) Normal rate
of cerebrospinal fluid formation five years after bilateral choroid
plexectomy. J Neurosurg 44:735–739.
Milhorat TH, Hammock MK, Davis DA, Fenstermacher JD (1976b)
Choroid plexus papilloma. I. Proof of cerebrospinal fluid
overproduction. Childs Brain 2:273–289.
Milhorat TH, Hammock MK, Fenstermacher JD, Rall DP, Levin VA
(1971) Cerebrospinal fluid production by the choroid plexus and
brain. Science 173:330–332.
Milhorat TH, Mosher MB, Hammock MK, Murphy CF (1970) Evidence
for choroid-plexus absorption in hydrocephalus. N Engl J Med
283:286–289.
86 D. Orešković et al. / Neuroscience 354 (2017) 69–87
Milhorat TH (1987) Physiology of the cerebrospinal fluid. In:
Cerebrospinal fluid and the brain edemas. New York,
Boston: Neuroscience Society of New York. p. 39–73.
Modic TM, Weinstein MA, Rothner AD, Erenberg G, Duchesneau
PM, Kaufman B (1980) Calcification of the choroid plexus
visualized by computed tomography. Radiology 135:369–372.
Netsky MG, Shuangshoti S (1970) Studies on the choroid plexus.
Neurosci Res 3:131–173.
Nielsen S, Smith BL, Christensen EI, Agre P (1993) Distribution of the
aquaporin CHIP in secretory and resorptive epithelia and capillary
endothelia. Proc Natl Acad Sci U S A 90:7275–7279.
O’Connell JEA (1943) The vascular factor in intracranial pressure and
maintenance of cerebrospinal fluid circulation. Brain 66:204–228.
Oppelt WW, Patlak CS, Zubrod CG, Rall DP (1964) Ventricular fluid
production rates and turnover in Elasmobranchii. Comp Biochem
Physiol 12:171–177.
Orešković D, Klarica M (2010) The formation of cerebrospinal fluid:
nearly a hundred years of interpretations and misinterpretations.
Brain Res Rev 64:241–262.
Orešković D, Klarica M (2011) Development of hydrocephalus and
classical hypothesis of cerebrospinal fluid hydrodynamics: facts
and illusions. Prog Neurobiol 94:238–258.
Orešković D, Klarica M (2014a) Measurement of cerebrospinal fluid
formation and absorption by ventriculo-cisternal perfusion: what is
really measured? Croat Med J 55:317–327.
Orešković D, Klarica M (2014b) A new look at cerebrospinal fluid
movement. Fluids Barriers CNS 11:16.
Orešković D, Klarica M (2015) The controversy on choroid plexus
function in cerebrospinal fluid production in humans: how long
different views could be neglected? Croat Med J 56:306–310.
Orešković D, Klarica M, Lupret V, Vukić M (2000) The character of the
cerebrospinal fluid production. Neurosci Res Commun 26:69–76.
Orešković D, Klarica M, Vukić M (2001) Does the secretion and
circulation of the cerebrospinal fluid really exist? Med Hypotheses
56:622–624.
Orešković D, Klarica M, Vukić M (2002) The formation and circulation
of cerebrospinal fluid inside the cat brain ventricles: a fact or an
illusion? Neurosci Lett 327:103–106.
Orešković D, Radoš M, Klarica M (2016) Cerebrospinal fluid secretion
by the choroid plexus? Physiol Rev 96:1661–1662.
Orešković D, Sanković M, Frobe A, Klarica M (1995) Physiological
characteristics of some monoamine metabolites in cat
cerebrospinal fluid. Croat Chem Acta 68:511–520.
Oshio K, Watanabe H, Song Y, Verkman AS, Manley GT (2005)
Reduced cerebrospinal fluid production and intracranial pressure
in mice lacking choroid plexus water channel Aquaporin-I. FASEB
J 19:76–78.
Pencalet P, Sainte-Rose C, Lellouch-Tubiana A, Kalifa C, Brunelle F,
Sgouros S, Meyer P, Cinalli G, Zerah M, Pierre-Kahn A, Renier D
(1998) Papillomas and carcinomas of the choroid plexus in
children. J Neurosurg 88:521–528.
Plum F, Siesjö BK (1975) Recent advances in CSF physiology.
Anesthesiology 42:708–730.
Poca MA, Sahuquillo J, Topczewski T, Lastra R, Font ML, Corral E
(2006) Posture-induced changes in intracranial pressure: a
comparative study in patients with or without a cerebrospinal
fluid block at the craniovertebral junction. Neurosurg 58:899–905.
Pollay M (1975) Formation of cerebrospinal fluid. J Neurosurg
42:665–673.
Pollay M (1977) Review of spinal fluid physiology: Production and
absorption in relation to pressure. Clin Neurosurg 24:254–269.
Pollay M (2010) The function and structure of the cerebrospinal fluid
outflow system. Cerebrospinal Fluid Res 7:9.
Pollay M, Curl F (1967) Secretion of cerebrospinal fluid by the
ventricular ependyma of the rabbit. Am J Physiol 213:1031–1038.
Pollay M, Hisey B, Reynolds E (1985) Choroid plexus Na+/K+
activated Adenosine Triphosphatase and cerebrospinal fluid
formation. Neurosurg 17:768–772.
Pollay M, Stevensen A, Estrada E, Kaplan R (1972) Extracorporeal
perfusion of choroid plexus. J Appl Physiol 32:612–617.
Pollay M, Stevens A, Roberts PA (1983) Alteration in choroid plexus
blood flow and cerebrospinal fluid formation by increased
ventricular pressure. In: Wood JH, editor. Neurobiology of
cerebrospinal fluid, 2. New York: Plenum Press. p. 687–695.
Pople IK, Ettles D (1995) The role of endoscopic choroid plexus
coagulation in the management of hydrocephalus. Neurosurg
36:698–701.
Portnoy HD, Schulte R, Fox JL, Croissant P, Tripp L (1973) Anti-
siphon and reversible occlusion valves for shunting in
hydrocephalus and preventing post-shunt subdural hematomas.
J Neurosurg 38:729–738.
Radoš M, Klarica M, Mučić-Pucić B, Nikić I, Raguž M, Galowski V,
Mandić D, Orešković D (2014a) Volumetric analysis of brain
parenchyma in a patient with hydranencephaly and macrocephaly
– case report. Croat Med J 55:388–393.
Radoš M, Orešković D, Radoš M, Jurjević I, Klarica M (2014b) Long
lasting near-obstruction stenosis of mesencephalic aqueduct
without development of hydrocephalus – case report. Croat Med
J 55:394–398.
Raimondi AJ, Gutierrez FA (1975) Diagnosis and surgical treatment
of choroid plexus papillomas. Child’s Brain 1:81–115.
Ratner V, Yi G, Lee H, Nedergaard M, Benveniste H, Tannenbaum A
(2017) Cerebrospinal fluid and interstitial fluid motion via the
glymphatic pathway modeled by optimal mass transport.
NeuroImage 152:530–537.
Rekate LH (2008) The definition and classification of hydrocephalus:
a personal recommendation to stimulate debate. Cerebrospinal
Fluid Research. http://dx.doi.org/10.1186/1743-8454-5-2.
Rubin RC, Henderson ES, Ommaya AK, Walker MD, Rall DP (1966)
The production of cerebrospinal fluid in man and its modification
by acetazolamide. J Neurosurg 25:430–436.
Ruckebusch M, Sutra JF (1984) On the significance of monoamines
and their metabolites in the cerebrospinal fluid of the sheep. J
Physiol 348:457–469.
Sakka L, Coll G, Chazal J (2011) Anatomy and physiology of
cerebrospinal fluid. Eur Ann Otorhinolaryngol Head Neck Dis
128:309–316.
Sato O, Asai T, Amano Y, Hara M, Tsugane R, Yagi M (1972)
Extraventricular origin of cerebrospinal fluid: formation rate
qualitatively measured in the subarachnoid space of dogs. J
Neurosurg 36:276–282.
Sato O, Bering Jr EA, Yagi M, Tsugane R, Hara M, Amano Y, Asai T
(1975) Bulk flow in the cerebrospinal fluid system of the dog. Acta
Neurol Scand 51:1–11.
Segal MB, Pollay M (1977) The secretion of cerebrospinal fluid. Exp
Eye Res (Suppl) 25:128–147.
Silverberg GD, Huhn S, Jaffe RA, Chang SD, Saul T, Heit G, Von
Essen A, Rubenstein E (2002) Downregulation of cerebrospinal
fluid production in patients with chronic hydrocephalus. J
Neurosurg 97:1271–1275.
Sjostrom R, Ekstedt J, Anggard E (1975) Concentration gradients of
monoamine metabolites in human cerebrospinal fluid. J Neurol
Neurosurg Psychiatry 38:666–668.
Sklar FH, Reisch J, Elashvili I, Smith T, Long DM (1980) Effects on
pressure on cerebrospinal fluid formation: nonsteady-state
measurement in dogs. Am J Physiol 239:R277–R284.
Speake T, Freeman LJ, Brown PD (2003) Expression of aquaporin 1
and aquaporin 4 water channels in rat choroid plexus. Biochem
Biophys Acta 1609:80–86.
Spector R, Johanson CE (1989) The mammalian choroid plexus. Sci
Am 261:48–52c.
Spector R, Keep RF, Snodgrass RS, Smith QR, Johanson CE (2015)
A balanced view of choroid plexus structure and function: Focus
on adult humans. Exp Neurol 267:78–86.
Taketomo T, Saito A (1965) Experimental studies on cerebrospinal
fluid flow. Neurology 15:578–586.
Trevisi G, Frassanito P, Di Rocco C (2014) Idiopatic cerebrospinal
fluid overproduction: case-based review of the pathophysiological
mechanism implied in the cerebrospinal fluid production. Croat
Med J 55:377–387.
 D. Orešković et al. / Neuroscience 354 (2017) 69–87 87

Velasco-Siles JM, Raimondi AJ (1982) Choroid plexus papilloma.
Pediatric neurosurgery. Surgery of the developing nervous
system. New York: Grune and Stratton. p. 451–460.
Von Storch TJC, Carmichael EA, Banks TE (1937) Factors producing
lumbar cerebrospinal fluid pressure in man in the errect posture.
Arch Neurol Psychiat 38:1158–1175.
Weed LH (1914) The dual source of CSF. J Med Res 26:93–113.
Weed LH (1917) The development of the cerebrospinal spaces in pig
and in man. Contrib Embryol 5:1–116.
Weed LH (1935) Forces concerned in the absorption of the
cerebrospinal fluid. Am J Physiol 114:40–45.
Welch K (1963) Secretion of cerebrospinal fluid by choroid plexus of
the rabbit. Am J Physiol 205:617–624.
Welch K (1975) The principles of physiology of the cerebrospinal fluid
in relation to hydrocephalus including normal pressure
hydrocephalus. In: Friedlander WJ, editor. Advances in
Neurology. New York: Raven Press. p. 247–332.
Welch K (1980) The intracranial pressure in infants. J Neurosurg
52:693–699.
Wellons 3rd JC, Tubbs RS, Leveque JC, Blount JP, Oakes WJ (2002)
Choroid plexectomy reduces neurosurgical intervention in
patients with hydranencephaly. Pediatr Neurosurg 36:148–152.
Yamada S (2014) Cerebrospinal physiology: visualization of
cerebrospinal fluid dynamics using the magnetic resonance
imaging Time-Spatial Inversion Pulse method. Croat Med J
55:337–346.
Zmajević M, Klarica M, Varda R, Kudelić N, Bulat M (2002)
Elimination of phenolsulfonphthalein from the cerebrospinal fluid
via capillaries in central nervous system in cats by active
transport. Neurosci Lett 321:123–125.

(Received 4 October 2016, Accepted 19 April 2017)

(Available online 27 April 2017)

View publication stats