The Cavum Septi Pellucidi in Utero

Introduction

My name is Tom Winter.
I'm from the University of Utah,
and we'll be talking about the Cavum Septi Lucidi.

For the next 40 minutes
we're gonna be talking about the Cavum Septi Lucidi in
utero, and we'll divide the talk up into what it is,
talk about embryology and anatomy
because that's fascinating, some of the etymology there.
Then we'll give four reasons why we need to pay attention
to this on our routine prenatal scans.
And we'll talk about some imaging pitfalls.

This is kind of my face.
When people first started talking
to me about the CSP I was completely clueless.

What is the Cavum Septi Lucidi?

So what is it? Basically conceptually, think of the pros.
Cephalon cleaving into two hemispheres
in the telencephalon.
And so you have this fissure, this cavity
between the left and the right sides.
And this cavity gets walled off on the top
by the corpus callosum
and on the bottom by the columns of the fornix.
So this is the cavity we're looking at walled off on the top
by the corpus callosum and on the bottom by the fornix.

In terms of word origin, this is often misused
and I certainly didn't know this stuff,
but I talked to one of the Latin professors on campus.
So we talk about the cavum, which is Latin for cave
or hollow space or hole or cavity.
We talk about the septum, which comes from Latin,
meaning a wall, think of your nasal septum.
And then we talk about lucidum,
which means translucent or clear.
And that's where our word translucent comes from.
And then Latin is infamous for its suffixes
and its variations there.
So it turns out that septum is neuter.
And then you can have a singular form septum,
a plural form septa,
and then the genitive form, which is septi.
And this is the equivalent of the English possessive form 's.
So we say God's son or the dog's bone.
So when we say septi, we're saying it belongs to something.
So literally the correct term,
Cavum Septi Lucidi means the cave
of the clear walls right there, the cave belonging
to the clear walls.

And it's got some different names, the septum lucidum
or the fifth ventricle.
And as we talked about
before, when the corpus callosum forms from the lamina
terminalis, it's intimately associated
with two paired clear walls, the septa pellucida.
And that's why the CSP, the Cavum Septi Pellucidi is
so important because embryologically, it's derived very,
very closely with the corpus callosum
and that's one of its biggest advantages to us.

The space
between these two clear walls is one cavity,
but for bizarre historical reasons
it has two different names.
So anterior rostral to the foramina of Monroe,
it's called the Cavum Septi Lucidi.
And in back, caudal to the vertical plane defined
by the columns of the fornix, it is known as the Cavum Vergae
and Vergae was named after the Italian anatomist in 1850.
And this is also called the sixth
ventricle, although it's not a ventricle.

So here's just a diagram just showing this one cavity there,
but it's called the CSP in front,
it's called the Cavum Vergae.
In back in practice,
people typically just call this whole thing the Cavum Septi
Lucidi, even though that's not quite correct.

So in most individuals this cavity gets obliterated
as these two walls come together and fuse.
So when the two walls come together, we now have
to use the singular term.
So it's septum and then the suffix follows the declension.
So we have the septum lucidum, the clear wall as opposed
to the septa pellucida, the two clear walls.
And this closure of the cavity begins at about 24 weeks.
Gestational age starts in the back
and goes from back to front.
And by term closure's occurred in the back in
almost all patients.
And by three to six months of age the entire CSP is closed
and you have the septum lucidum,
although in a small minority this
remains open until adulthood.

So here's a normal adult brain on the left.
And then on the right we see the CSP
and I got this from the literature
and note that they got the Latin wrong
because this should be the cavum septi lucidi,
not the cavum septum lucidum as they have it listed
and lots of people make this mistake.

So here are normal examples.
You can see between the arrows there
and the axial plane, that's the CSP.
And in the coronal plane, that's the CSP right there
on MRI in the axial plane
and coronal plane,
the arrows denote just this thin collection of fluid
between the frontal horns.
And then here we are postnatally up in the NICU
and you can see the cavity with on top
of it the corpus callosum, the genu
with the corpus callosum up there.
And obviously attenuation patterns change
as you go from immature to mature.

So here's a postnatal scan on a 29 week showing the CSP
between the lateral ventricles, right there on axial
and coronal images and then a term infant.
It's much smaller
and much less obvious getting
with the embryology that we learned there.

You can have a variant as we talked about,
where the entire cavity stays big further than usual.
And so most of the time this is a normal variant.
At the end of the talk, we'll talk about
maybe a pathologic association with this.

Reasons to Pay Attention to the CSP

Reason 1: Guidelines

Now let's talk about why does it matter.

Reason number one are guidelines and
before guidelines went on the web that shows how old I am.
But the ACR in Reston, Virginia, the American College
of Radiology was talking to one of the secretaries
and she said that the number one source of phone calls
that the ACR gets from the community was not from
radiologists, but it was from lawyers asking
for copies of the guidelines.

So the guidelines that are promulgated
by the American College of Obstetricians and Gynecologists,
the American College of Radiology,
the multidisciplinary societies like the American Institute
of Ultrasound in Medicine,
even though they're called guidelines in
practice, they're standards.
So if you don't meet these guidelines,
you're really liable there.

And the guidelines,
this is the most recent version here,
republished last year.
Basically talk about six things in the head and neck.
And it's really three pairs of threes.
Number one are the ventricles, so ventricles
and choroid plexus number two is the posterior fossa,
which is cerebellum and cisterna magna.
And number three is the midline, which is the falx
and the cavum septi lucidi.

So the CSP is required by the guidelines.
So there's one reason why it matters.

Reason 2: Agenesis of the Corpus Callosum

The next reason it matters is let's
take a look at this fetus.
We have this image right here
that I'm gonna give you two different fetuses,
a coronal and an axial.
And your diagnosis here is obviously agenesis
of the corpus callosum.

When we go back to fetus number one,
you can see the preferential dilatation
of the occipital horns of the lateral ventricles.
And this is colpocephaly from Greek meaning hollow head.
And then on number two on the coronal image there,
if there's any Texas Longhorn fans in the audience,
this is the classic hook 'em horns appearance.
Other people call it the Viking helmet, the steer, a lot
of different terms there.
But you have this very unique
appearance on the coronal plane.
And then on the axial plane you have the parallel
configuration to the lateral ventricles there.

So our diagnosis here is obviously a CC.
This is one of the most common CNS abnormalities.
Prevalence varies all over the place,
but it's about 4% of CNS malformations.
It's about half a percent in the general population
and 3% in the developmentally disabled population.
It's associated with major cerebral
or extracerebral malformations including tons of syndromes
and metabolic diseases, up to an 80% risk
of associated brain malformations.

So if you see agenesis, don't stop looking up
to 80% of them have something else.
And don't just look in the brain up
to 60% have something else in the heart, the arms
and legs, the GU system.

So suspect a CC when the cavum is not seen.
And generally you see the cavum on all
three required views of the head.
You see it on the ventricles,
you see it on the trans thalamic, you see it on the
20 degrees suboccipital trans cerebellar view
of the cisterna magna right there.
It shows up on all of them there.

So what are some additional indirect signs of a CC?
We talked about the Greek hollow head
or colpocephaly, which is defined as preferential dilatation
of the occipital horns
of the lateral ventricles right there.
That in absence the CSP are probably the two best signs
look for these parallel lateral ventricles.
The third ventricle which no longer has the roof
of the corpus callosum may be high riding
and you may have a azygos pericallosal artery,
just one artery there instead of the paired ones.
But this is a hard diagnosis to make and it's often missed.

Here's another patient. You can see the preferential
dilatation of the occipital horn right there.
Here you can see the parallel lateral ventricles.
Here you can see the hook horns
appearance on the coronal plane.
And then on the sagittal plane you see the sunburst gyri,
the lack of corpus callosum.

Here's another one, parallel lateral ventricles.
In the coronal plane you see the hook 'em horns.
And then in the sagittal plane we have dysgenesis,
the corpus callosum.
And remember as we talked about,
the corpus callosum starts in front and forms to the back.
So here we have a little bit of the genu and the body,
but none of the splenium.

You can have a high riding third ventricle.
Don't mistake this for the cavum septi lucidi
because again without the corpus callosum,
in about half the cases, the third ventricle will rise up
and mimic the corpus callosum.
But this is too far posterior.

And again we said 80% association
of intracranial abnormalities.
Here we have a posterior fossa cyst.
Here you can see that cyst with the dilated frontal horns.
Here you can see the colpocephaly
and here you can see the colpocephaly,
the preferential dilatation
of the posterior occipital horns versus the frontal.

You can see the Dandy-Walker variant cyst.
And here's just a cine clip showing that big cyst back there
and the preferential dilatation.
And here's the MRI in utero on
that patient the hook him horns appearance on the coronal
and the big cyst in the posterior fossa
for the Dandy-Walker variant.
And on the axial plane you have the parallel lateral
ventricles with preferential dilatation posteriorly.
And on the axial plane you can see the posterior fossa cyst.

Here's patient you see looks kind
of normal except we don't see the cavum
and look for the interdigitating gyri in
the front of the brain.
The gyri shouldn't connect right there.
And then here, this is not the cavum
because it's too far posterior.
This is an interhemispheric cyst.

So now we're starting to get worried.
We have an interhemispheric cyst,
we have the interdigitating gyri up anteriorly,
we have absence of the cavum septi lucidi.
And then on this parasagittal view we have this fluid
density at the back of the orbit.
This is a coloboma that comes from Greek for defect.
So it could be generic,
but in medicine we use coloboma to refer
to a defect of the eyeball there.
And so this is a classic example
of a Aicardi syndrome.

So the initial clinical description was infantile spasms,
agenesis and ocular abnormalities in a female fetus.
The combination of a agenesis
with posterior fossa malformation of any type
and heterotopia suggest a Aicardi,
especially if there's eye abnormalities.
This is X-linked dominant so males don't survive
and it's an important thing to suggest.

So again, interdigitating gyri happens
to the CSP high riding third ventricle.
And here we are on the in utero MRI showing
polymicrogyria and the coloboma.
And then we have a normal Sylvian fissure on the left side.
Absence of the Sylvian fissure on the right side
and on the axial view you see the high riding third
ventricle And you can do fancy 3D things.
This is tomographic ultrasound imaging right here,
just stacking up things there
and that's the interhemispheric cyst.

So the second reason for why the CSP matters is the common
and difficult to diagnose things like agenesis
of the corpus callosum including a Aicardi

Reason 3: Septo-Optic Dysplasia

Third reason.
This is a really characteristic appearance.
Here's axial and coronal. And this is a common pitfall.
People say, oh I see fluid up front,
therefore everything has to be happy.
But remember we need the box of the cavum septi lucidi
between the frontal horns
of the lateral ventricles right there.
And we don't have that.

And here's the postnatal scan on this fetus.
Again, we have fluid anteriorly
but there's no box right there.
And this is classic for one diagnosis here.
And this is septo-optic dysplasia, which is absence
of the CSP hypoplasia of the optic nerves
and a dysfunction of the hypothalamic pituitary axis.
This is associated with all sorts of things,
developmental delay, intellectual impairment, seizures,
growth restriction postnatally there.

And really your only differential
Here would be lobar holoprosencephaly versus this
Versus true isolated absence of the septum pellucidum.
But that's really unusual there.

So as we'll talk about in a little bit,
with septo-optic dysplasia, you tend not
to have ventriculomegaly.
The corpus callosum most but not all times looks normal.
The columns of the fornix are paired, they're not fused.
You have paired anterior cerebral arteries
and you have a falx all the way up.

And this is one where you have to recognize absence
of the CSP 'cause
otherwise you'll never make the diagnosis
of septo-optic dysplasia.
But this box-like appearance here is absolutely
classic for it.
And here's just another case.
We've had two cases of this in
the first four months of the year.

So here you can see in utero ultrasound
and MR showing this squared off appearance.
And here's the postnatal.

So the uncommon and very difficult to diagnose.
You're only gonna make this diagnosis if you recognize
absence of the CSP and that's septo-optic dysplasia.
They should probably include schizencephaly in here,
but we'll see that in a minute.

Reason 4: Review of Midline Abnormalities

And then our fourth and final reason is basically an excuse
to review a whole host of midline abnormalities.
There's a lot of stuff that goes on up here.
Absence of the CSP is not crucial to making this diagnosis
because there's generally so much other stuff going on
that you can make it.
But let's take this opportunity
to review a little neuroanatomy and neuro embryology.

So some basic embryology for us here
you have ventral induction of the neural tube taking place
between weeks five to 10 right there,
There are three primary vesicles. There are four
Secondary vesicles
and the three primary vesicles form the forebrain,
the midbrain and the hind brain.
But since we're in medicine we can't use those terms.
We have to make up fancy Latin terms.
So we have the prosencephalon, the mesencephalon
and the rhombencephalon
and those are just synonyms
for forebrain midbrain hind brain.

And then each of those three primary vesicles
or two of those primary vesicles have two
secondary vesicles each.
So the prosencephalon has the telencephalon,
which is the cerebral hemispheres
and the diencephalon, which is the thalamus.
And then the rhombencephalon has the metencephalon
and the myelencephalon right there.
And those form the cerebellum and pons and the medulla bla.

So holoprosencephaly is a developmental field defect
of cleavage of the prosencephalon.
The prosencephalon the forebrain,
should form two hemispheres in the telencephalon there.
So corpus callosum, agenesis
and SOD are part of the spectrum right there.
And because so much is happening at five to 10 weeks,
all these things are going on at the same time.

As DeMyer said, the face predicts the brain, meaning
that many times you'll have facial abnormalities associated
with disorders of prosencephalic cleavage.
And these include cyclopia, cebocephaly,
hypotelorism, ethmocephaly yada yada yada.

So here's a classic example of a lobar holoprosencephaly
fused thalami monoventricle, same
different fetus right here, a lobar holoprosencephaly.
Here's a different one with semilobar.

So basically there is no absolute demarcation
between any of these entities.
It is a spectrum. There are other classification systems out
there, but most people use DeMyer's classification system
of a lobar, semilobar and alobar.
But just think of it as a spectrum from bad to less bad.

And here's the same fetus there you see the mantle
of cortex in the frontal area on both in utero
ultrasound and mr.
And then here you have the mantle of cortex on top.
Different patient, different patient.

So you're getting cleavage posteriorly but not rostrally.
Here's the same patient showing the unfused mantle
of cortex up front.
And then here's a alobar holoprosencephaly.

And there was a really fascinating paper
by a guy named Paul Pilu that raised a great point.
I don't think it's true all the time,
but it's fun to think about and I think it's a useful,
but it's just not a perfect distinguishing factor.
And he said that in alobar holoprosencephaly,
you've got pretty good splitting from front
to back, almost all of the front but not all of it.
In addition, the thalami, which you should have
down here in the corners, a pointer doesn't work
but they are fused and central.

So here's an example of alobar holoprosencephaly,
we've got good separation in back,
pretty good separation in front,
but the pathologists say as long
as you've got a couple fibers that aren't fused,
that are fused crossing the midline,
it's a holoprosencephaly spectrum there.
And look on the postnatal ultrasound, the postnatal mr.
We have the fused fornix which is suggestive
although not diagnostic of alobar.

And here's another one we don't see the CSP on Doppler
ultrasound we see an azygos anterior cerebral artery, just one
of them coming up the middle right there
we have the fused thalami.

Here we have a suggestion
of rhombencephalosynapsis which is associated
with the holoprosencephaly there,
but see how the folia
of the cerebellum are crossing over the midline.

Here's another alobar right here
with pretty good separation from back to front
but just not perfect.
And you can see on the coronal in utero ultrasound
and MR the fused thalami.

And so we just talked about holoprosencephaly, abnormal cleavage
of the prosencephalon in the fifth week classic subdivision.
The alobar has the flat squared roof fused thalami.
Some people, not everybody,
but some people say
that septo-optic dysplasia is the least severe
form of the holoprosencephaly.

And remember it's associated with lots
and lots of facial abnormalities like cebocephaly means
monkey face there cyclopia superior
to the eyes ethmocephaly.

So we have the axial image which looks really funky,
we're not sure what it is.
There's a blob there. And then on the coronal image there's
the blob sitting in front.
And it's not till we go sagittal
that you can see this is a cyclopia with olfactory tissue in it,
which is superior to the single orbit of anophthalmia.

And here are the fused thalami and the monoventricle.
And there was an echogenic intracardiac focus.
So as we all know, this constellation of findings
and holoprosencephaly in general's associated with trisomy 13.

Here's an unusual one
where in utero we see good separation up front,
good separation up back.
But on this postnatal autopsy image,
we see continuity across the midline.
So this is syntelencephaly also known
as the middle interhemispheric variant
of holoprosencephaly.
It was described by Barkovich in 1993.
It's in the spectrum of holoprosencephaly.
But here what happens, instead
of having your fusion in front, which is classical,
what happens in holoprosencephaly here you get cleavage in front,
cleavage in back, but fusion over the vertex.

And so again, syntelencephaly
or the middle interhemispheric variant,
what your diagnosis here here are axial
in utero ultrasounds and MRIs.
And note that this is not a normal cavum at the big era.
We have fluid up there
but there's no box between the lateral horns
of the frontal ventricles.
And this is syntelencephaly.

And I was taught that this is a vascular insult
but I think we've learned more since then.
It's probably disorder of neuronal migration.
There are two types, there's open lip, there's closed lip,
there's unilateral, there's bilateral,
it's probably again all of a spectrum right there.
And even recently,
this was a nice paper in JUM last October, it's tough
to diagnose with ultrasound in the second trimester
to see the cleft of schizencephaly,
but there's an absent cavum in 70%
of these affected fetuses.

So we want to go ahead and recognize the absent cavum
and then start finding out what it's associated with.
And this schizencephaly is associated with big vents, agenesis,
polymicrogyria and gray matter migrational abnormalities.

So here you see the triangular appearance
of the schizencephaly on the ultrasound and the MR
and the absent cavum up front.
Make sure that you don't confuse this with the Sylvian
fissure or the lateral fissure
or sulcus there that's a little more square whereas in schizencephaly
the defect is more triangular.

What your diagnosis here Then we went
and did an in utero mr.
And this is pretty classic note
that we have a preserved falx.
So this is classic for hydranencephaly which is replacement
of most of the cerebral hemispheres with fluid.
You spare the brainstem cephalic structures there
because those are provided.
Remember the hindbrain is far down.
So it gets posterior fossa circulation and the inferior
and medial aspects of the frontal lobes
and the temporal lobes are often partially spared
because you get some collateral circulation there.

This can be created in an animal model
by occluding the internal carotid
arteries bilaterally there.
So we think it's a vascular insult,
but an overwhelming neonatal
or antenatal TORCH infection may give it too.
Obviously you don't need absence of the CSP
to make this diagnosis because the associated findings are
so horrendous.

So this is hydranencephaly.

Here we see what's this posteriorly?
This is a posterior occipital encephalocele.
Then in front that is not the CSP, that's the fused thalami.
And then here we have a sagittal view showing
the encephalocele.
We have colpocephaly or Greek hollow head on the axial view.
And Barkovich in his original description of
absent CSP described basilar encephaloceles
as being associated with this.

Now obviously if we lived in Singapore,
maybe you'd have more tal or frontal encephaloceles.
I don't know if the association holds up there or not.

And then the final thing that we're gonna talk about
is the association of hydrocephalus with absent CSP.

So here we have big time hydrocephalus right there, lateral
and third ventricles, but we don't see the cavum And
this is probably a relatively common cause
of absent CSP because the hydro just gets so big
and so bad that it just shreds the cavum
and you get little wisps remaining right there.
So probably one of the more common causes for CSP just due
to bad hydro from whatever cause.
And again here this one,
this particular cause was aqueduct stenosis.

Persistent CSP and Variants

Okay, what is this? Is this normal or is this abnormal?
In coronal and axial?
You can get measurements in the
handout I gave you in the paper.
There are lots of references as there is
for everything else in obstetric ultrasound about a normal
value of the CSP is a function of gestational age.
So you can look those up.

Most of the time this is a normal variant,
but I took this from the Harvard website here.
They're talking about in adults who have a persistent CSP,
maybe being associated
with neuro affective disorders like schizophrenia.
So again, something to think about right there.
Although most of the time this is probably normal,

Imaging Pitfalls

This is generally what the residents say
to me when I talk too much and they're bored.
So we're just gonna go ahead and talk about some pitfalls
and go ahead and sum up.

So four pitfalls
and we'll go over these in a little more detail incorrectly.
Attributing the columns, the fornix
to the CSP confusing a high riding third ventricle
or interhemispheric fissure for the CSP,
the normal variant cavum vergae
and noting fluid between the lateral margins
of the frontal horns and calling that the normal CSP.

So again, from Peter's paper we want to remember
that if you're too far inferior, the paired columns
of the fornix can give this appearance.
But we have the white line down the middle,
that is not the CSP
and the corpus callosum can be completely gone
despite this appearance.

Now look at this right here.
Is this the CSP, between the white arrows, you see
what looks like a cavity,
but look at that white line in the middle
of the fetus's head going straight through the cavity.
Remember the cavity shouldn't have anything going
through it right there.
And so what happened here, this is from a superb paper
by Peter Callen and his group,
but basically if you
take your imaging plane too far inferior,
you'll get into the columns of the fornix right there.
And that's abnormal. Whereas we should be imaging more superior
and be in the CSP there.

And this is important because the fornix doesn't have the
relationship to the corpus callosum that the CSP does.
So just because you see the fornix doesn't mean
that the corpus callosum is normal.

And let's also use this nice diagram from Peter's paper just
to talk about the nomenclature of the corpus callosum.
Most authors divide this into four parts, the rostrum,
which comes from the front of the Carthaginian warships,
who the Romans defeated there.
And so it means beak right there or front.
So that's the pointy area.
Then you have the genu, which is in the front
of the corpus callosum, the body, the obvious body,
and then the splenium all the way in the back
of the corpus callosum above the cerebellum.
And then some authors subdivide into a fifth
division, the isthmus, which separates the body
and the splenium right there.

And in general the corpus callosum forms from front to back,
although there is an exception to that.
The rostrum, that little beak-like structure inferiorly
and posteriorly forms later with the splenium.
So it goes genu, then body, then splenium and rostrum.

So again, if you scan too far caudally
as we see in this diagram from Peter's paper there you're
gonna get the columns of the fornix and they're solid,
but they're hypoechoic.
So you may call those the CSP.

So what is the fornix?
Well, fornix is a vault like or arched structure.
And it comes from Latin meaning arch
or vault or vaulted chamber.

And here's the only point of this entire talk
that my residents remember,
but I got curious,
where does the word fornication come from when we're using
fornix to describe these neurologic structures?
And it turned out that in ancient Rome, the ladies
of the night would congregate in
or underneath the vaulted arches surrounding the Colosseum.
So the act of carrying out an illicit sexual relationship
was called going under the arches or fornication.

So remember that's where the word fornix comes from.
That's how we get the derivation of fornication
as you can see in this slide from the Colosseum here.

So here are two more examples.
These are from Peter's paper on columns of the fornix.
And note in both of these we have a white line
going down the middle of it.
And that white line is probably the interface
between the two fornices.
And this is normal. We're now scanning more superior
and we have this nice little box of fluid right there.

Pitfall number two, the high riding third ventricle
or interhemispheric cyst.
Remember the CSP is
between the frontal horns of the lateral ventricles.

Number three, this normal variant cavum vergae
going all the way back.

And number four, seeing this box-like structure,
calling it the CSP.
But remember we don't have that box within the box.
Where is this paired? Septa pellucida. They're absent.

So to quote Roy Filly, don't make it up.
This is septo-optic dysplasia
and this is schizencephaly right here.

Summary

So why does it matter? Four reasons.
Number one, it's required by the guidelines.
Number two, it helps you with the common
and difficult to diagnose agenesis of the corpus callosum.
Number three, it helps you with the uncommon
and very difficult to diagnose classically, SOD,
maybe schizencephaly.
And number four, we can use this as an excuse
to review a whole host of prosencephalic
cleavage disorders.

As we just did, we talked about the
four pitfalls right there.
And so to sum up, we talked about what it is, embryology,
anatomy and etymology,
four reasons why it matters in imaging pitfalls.

And I'd like to thank my friend who's a Latin professor
for educating me about this.
All mistakes are mine, not hers.
And the NASA website for the gorgeous,
pictures from the Hubble telescope and elsewhere there.
Thank you very much.