The Cavum Septi Pellucidi in Utero - HD

Introduction

My name's Tom Winter.
I'm from the University of Utah,
and we'll be speaking about the Cavum Septi Lucidi in utero.

We're gonna be talking about the Cavum Septi Lucidi in utero.
I have no relevant disclosures,
and we'll divide the talk up into several sections.
We'll talk about what it is first, embryology,
anatomy and etymology.
And then most importantly,
we'll talk about why does it matter
and give four reasons why we need to care about it.
And then we'll finish up with some imaging pitfalls.

If you had asked me 20 years ago about the CSP,
this would've been my response there.
And I can't imagine how many cases I missed over the years
because I didn't think about this.
There's a great quote that my partner, Paula Woodward,
uses all the time to the residents that has been attributed
to the author of Sherlock Holmes.
And that's, you see what you look for
and you look for what you know.
And before I knew about the CSP
or thought about it, I never saw it
and I never saw its absence.
So hopefully this lecture will help us with that problem.

What is the CSP?

Remember that in embryology you have two cerebral
hemispheres, the cephalon, and they split.
And so think of the CSP as
that longitudinal cerebral fissure
between the two hemispheres, which becomes walled off on top
by the corpus callosum and walled off below by the fornix.
So this is the CSP right
here, the etymology.

Etymology

I worked with a Latin professor on campus
to figure this out, but Cavum is Latin
for basically a cave or a hollow space.
Septum is Latin for a wall
or enclosure like the nasal septum.
And then there are ses in Latin, septum is neuter singular,
SEPTA's plural, and septi is generative
and this is very equivalent to an English, the apostrophe
s like the dog's bone or God's son.
And then lucidum is Latin for translucent or clear.
So when we're saying Cavum Septi Lucidi,
literally we're saying the cave
that belongs to the clear walls.
So it actually makes some sense there.

In the past it's been called the septum lucidum
or the fifth ventricle.
That's kind of a bad term
because it doesn't communicate directly with the CSF,
but embryologically it's important
because around 12 weeks gestational age,
the corpus callosum starts
to develop from the lamina terminalis as a bundle of fibers
that connects the two hemispheres.
Remember that a commissure is the connection
between the two hemispheres.
And depending how you count it, there are six or seven,
but by far and away the biggest is the corpus callosum.
And this development of the corpus callosal commissure is
intimately associated with the development
of the Septa Lucida, the two paired cleared membranes.
And why that's important, we'll talk about in a little bit.

To make it even more confusing,
there's one physical space between the two Septa Lucida,
the two clear membranes,
but for whatever silly reason, it has two different names
in front it's called the Cavum Septi Lucidi.
And in the back it's called the Cavum Vergae or Varga.
And the dividing point is the foramen of Monro.
Vergae came from Vergae's ventricle
and it was the cavity described
by the Italian anatomist in 1851, which is known
as the six ventricle, which is an equally silly term.

So here's a diagram showing the single physical space,
but in front it has one name, the CSP,
and in back it has another name.
The Cavum Vergae.
And then there's a whole bunch of other fluid collections.
Here's the VE tum.

So in most individuals,
the cavum is closed early in childhood
and the two septa are fused.
So this cave starts to get smaller and smaller.
As the two clear walls come together,
the two clear walls fuse.
So now we use the Latin singular
because there's one membrane.
So we use the singular suffix
and it's now known as the septum lucidum.
This closure
of the Cavum Vergae begins at about six months of age
and progresses from back to front, like a lot
of things do in the brain by term closures occurred
posteriorly in 97%.
So there's generally only a true CSP at birth.
And by three to six months the CSP is closed.
And 85% of us,
although in a minority, you get a small
residua till adulthood.

So this is normal. We have the cavum sep
or we have the septum pellucidum,
just the clear membrane in no space.
And then here's the cavum septi pellucidi in the
literature, I got this from the web often uses a misnomer.
So this should be the cavum septi pellucidi,
not the cavum septum lucidum because this means singular.
And there's obviously two clear walls there.

So here's some normal anatomy. Here are the two clear walls.
Here's the cave in the coronal image.
Here are the two clear walls.
And here's the cave in utero.
MRI two clear walls cave in between
coronal two clear walls cave in between.
In a postnatal ultrasound we see the corpus callosum,
we see the two clear walls.
And here's the cave. Here's a premature infant.
And so it's much bigger because the cave
and Vergae is still making up the bulk of this.
So it goes all the way back.
This is not a high riding third ventricle
or an interhemispheric cyst.
This is just A-C-S-P-A Vergae.
And here's a normal term infant postnatally showing a much
smaller CSP in front and no cavum Vergae.
And again, this is a normal variant here in a later
gestational age fetus showing the fluid
going all the way back.

So is this the CSP right here we have two walls,
but notice that there's a line going down the middle,
something we didn't see before.
So this was a superb article
by Peter Callan in the UCSF group talking
about this pitfall.
We want to take our image of the CSP further, cephalic right here.
If we take it further caudal, we're gonna pick up the,
columns of the fornix.
Now remember the fornix is a bunch of white matter tracks
that is, takes the output from the hippocampus
and it's very, very important in laying down memory.
And if you're too far down, you will pick these two fornices
right next to each other.
So this is where we want to take the image.
If we mistakenly take it too far down, we'll get an image
of these two fornices.

Let's talk about the etymology, the fornix before we finish up right here with this section.
So the fornix means a vault like or arch structure,
and it comes from Latin arch vault or vaulted chamber.
And given how my mind works, I was kind of curious
why the term fornix is used for the fornix in the brain
and also the English word fornication.
And it turns out that in ancient Rome prostitutes more
or less hung out around the arches
surrounding the Colosseum.
And in Latin, the act
of carrying on an illicit sexual relationship came
to be called going under the arches or fornication.
So here's a picture of the Colosseum
showing the arches there.
And this is pretty much the only thing
that my residents remember from this talk.
But here's two more examples of the columns of the fornix.

Now this isn't for sure there's some
disagreement, but I like this theory.
It explains it saying that when you have the two,
paired fornices like this and in the midline is the interface
and where they're co-opted, we get
that white line down the middle.
So the CSP has a line on either side like this does,
but it shouldn't have the white line down the middle.
Here's normal white line's on either side,
but the white line of the midline is not present.
So we're gonna come back to that, but that's a pitfall.

Why Does It Matter?

Reason 1: Guidelines

Now let's go ahead and talk about why it matters.
The first reason.
We all know
that the guidelines are really standards even though we've
tried to call them guidelines.
The lawyers use these as standards
and there's various iterations of them.
This is the most recent that I found published
last summer.
But this is ACOG, A-C-R-S-R-U-A-I-U-M out there.
And in their guidelines for an obstetric exam,
they list six things that we need to do in the head.
If you're a lumper like I am, it's really three things.
So the cerebellum and the cistern
magna is one thing, the choroid plexus
and the lateral ventricles is another thing in the midline
falx and the cavum septi lucidi is another thing.
So that's reason number one why it matters.
It's required by the so-called guidelines
which are actually standards.

Reason 2: Agenesis of the Corpus Callosum

Okay, now let's get onto something more interesting.
Here's one fetus axial view through the ventricles.
Here's fetus number two, a coronal view through the head.
And here's fetus number three
and axial view through the head.
Both of these are obviously MRs.
What's your diagnosis in all three of these fetuses?
And the answer of course is agenesis of the corpus callosum
or hypogenesis of the corpus callosum.
This is one of the most common CNS anomalies out there.
Prevalence varies depending on pretest probability,
but up to 5% in the developmentally disabled population.
In the reference that I used, it was associated
with major cerebral
or extra cerebral malformations including 50 different human
congenital syndromes.
I heard Ruth give one
of her always superb lectures a couple months ago in her
reference used 2000 syndromes.
But bottom line, agenesis
or hypogenesis of the corpus callosum is
associated with a lot of stuff.

One thing to remember about the brain,
this comes from another great paper from the UCSF group,
but if you see one thing, keep looking, in the case
of agenesis, there's up to an 80% risk
of associated brain anomalies including dandy walker,
gyral anomalies, heterotopia, et cetera,
and not to be forgotten up to a 60% risk of things outside
of the brain heart disease.
MSK gu. So bottom line, if you see agenesis, there's a lot
of bad stuff in the brain
and a lot of bad stuff away from the brain.

Prenatal diagnosis of agenesis is tough.
We miss this a lot as a society there.
So the easiest way
to pick it up is when you don't see the CSP on the axial
trans thalamic view.
In practice it's generally big enough
that you'll often see it on the lateral ventricular view
and even the cisterna magna view
because remember that's a 20 degree angled,
suboccipital pragmatic plane.
So that comes up higher.

Now there are a whole bunch of additional indirect signs
that we've all been taught.
The colpocephaly, the preferential dilatation
of the occipital horns increased separation
of the hemispheres with the body
of the lateral ventricles parallel to each other
and shifted laterally in abnormal third ventricle,
which extends upward between the lateral ventricles
and a half of cases, the so-called high riding third
ventricle in an abnormal course
of the pericallosal artery.
The problem is, as much as we study
for these on tests in practice, people miss these.

So look at this fetus.
We have preferential dilatation of the occipital horns,
parallel lateral ventricles.
On the coronal view,
you have this various names if you're from Texas,
the hook 'em horns appearance.
If you're a Vikings fan,
it looks like the horns on a Viking helmet.
And then look on the sagittal view,
there is complete agenesis of the corpus callosum.
Here's another one, parallel lateral ventricles
hook 'em horns appearance on the coronal view.
And then on the sagittal view, we have hypogenesis.
Remember that the corpus callosum starts at the genu
and then moves posteriorly towards the rostrum
and then the body and the splenium up here.
So all we have is the genu pitfall,
high riding third ventricle.

So we talked about that you don't have the corpus callosum
as the roof between that cavity that's formed
by the fusion of the two fissures right there,
the splitting of the two fissures.
And so if you don't have the corpus callosum as a roof,
the third ventricle floats high just like a helium balloon.
And for the purist in the audience, there's a lot
of discussion about high riding third ventricle inner
hemispheric cyst.
But we're just gonna use the conventional terms here
and notice that we have preferential dilatation
of the occipital horns of the lateral ventricles
associated with lots of things as we talked about.

Here's a classic dandy walker malformation.
We heard a lot about this yesterday.
Dandy walker malformation. And then we have fluid up here.
But do you see the clear walls?
No colpocephaly with marked
ventricular magaly.
And then notice that even though we have fluid in front,
we don't have the clear walls.
And here's the video clip showing the,
dandy walker variant, whatever they're can call it nowadays.
And nothing that's a true CSP anteriorly.
Here's the mr showing the hook
and horn's appearance on the coronal
and then agenesis of the corpus callosum
with dandy walker malformation
parallel lateral ventricles,
and then verian agenesis, splitting
of the cerebral hemispheres in the big cyst.

Here's another fetus.
Kind of scary because the cistern
of Magna looks great, but remember how we said
that you can see the cavum septi lucidi on
that oftentimes, but on this one
we're not seeing it up here.
Here's the high riding third ventricle,
not the cavum, the cavus in front.
And then notice that we have T two hyperintensity
posterior to the globe.
So what's your diagnosis here?
This is for the syns
or the genetics counselors in the audience.
But this is a Aicardi syndrome, which is,
initial description was infantile spasm,
agenesis ocular abnormalities.
But in a female fetus, if you see agenesis
with a posterior fossette malformation, cortical dysplasia
and eye abnormalities, this was a boma.
Think of a Aicardi, it's X-linked,
it's lethal in men, so you don't see it there.
And postnatal retinal exam confirms the diagnosis.
So again, high riding. Third, no cavum there.
Female fetus, parallel lateral ventricles,
cooma here a Aicardi syndrome.

Here's another fetus.
Notice how we don't have a sylvan fissure on this side.
The high riding third ventricle
or the interhemispheric cyst.
And then here's the interhemispheric cyst
and multiple planes from a,
3D ultrasound showing no true cavum in front.

So that's the second reason why we do this.
It's the common
and difficult to diagnose agenesis of the corpus callosum,
including weird things like Aicardi.
We miss this as a society all the time.
There's all the secondary signs.
But if you just think absence of the CSP,
the first nine things on your differential are gonna be
agenesis or hypogenesis, the corpus callosum.

Reason 3: Septo-Optic Dysplasia

Okay, let's move on to another reason here.
We have fluid in front, axial
and coronal, but what's missing?
We don't see the two walls, the two septa lucida,
they're not present.
Here's this fetus postnatally. We don't see the two walls.
So what's your diagnosis in this case?
This is septo-optic dysplasia.
This is absence of the cavum hypoplasia of the optic nerves
and various types of hypothalamic pituitary dysfunction
that lead to all sorts of problems there.
It's interesting, they can actually diagnose this in utero
by testing hormones in mom's urine
because the fetus should be producing hormones
that don't get produced if you have hypothalamic
pituitary dysfunction.
And you can measure that in mom if you're a lumper.
This is the most mild form of holoprosencephaly.
If you're a splitter, it's a second, second diagnosis.
When I was reading for this in other papers, I was struck by
how much heterogeneity there is in the literature,
how much argument, all the stuff
that we learned in medical school
or residency where they put these diagnoses into neat
pigeonholes really doesn't work.
But the only way you're ever gonna make this diagnosis
prenatally is to see absence of the cavum.
There are no other secondary signs.
We can't visualize the optic nerves or the hypothalamus
or the thalamus with the required degree
of anatomic resolution.
So again, there's fluid up here,
but where are the clear walls?
They're absent. This was septo-optic dysplasia.

Here's another fetus.
There's fluid upfront, but there's no clear walls.
No clear walls. No septa pellucida on the axial Mr
and on the postnatal scan, again, no clear walls.
So this is the third reason why we want to look for the CSP.
We're looking for the uncommon
and very difficult to diagnose
entities like septo-optic dysplasia.

Reason 4: Midline Abnormalities

And in the last 12 minutes right here, let's talk about
a whole host of midline abnormalities.
We don't need absence of the CSP to make this diagnosis
because there's a lot more going on,
but it's a nice way to review a vehicle, if you will,
to review midline anatomy.
Now Anne reviewed some of this yesterday,
but basically ventral induction takes place
during weeks five to 10.
There are three primary vesicles, the forebrain,
the midbrain, and the hind brain.
But since we're in medicine, we have to have fancy,
Greek and Latin derived roots for these.
So we have the prosencephalon, the mesencephalon,
and the rhombencephalon, same thing as forebrain, mid-brain,
hind brain, and then four secondary vesicles,
the telencephalon, which is the cerebral hemispheres
and the diencephalon, which is the thalamus
and the hypothalamus.
For the purposes of this talk
where we're talking about the midline,
all we care about is the prosencephalon.

So we've got a paper coming out in a couple months talking
about the full spectrum of holoprosencephaly.
What I learned in medical school was the DeMyer
classification, alobar, semilobar and lobar.
There are other classifications out there,
but this is the main one.
Then in the 1990s, Dr.
Kovich, the guru of brain anatomy, came up
with a fourth variant called the Middle Interhemispheric,
variant or syntelencephaly.
But that's all part
of this spectrum here there's a more
severe part of the spectrum.
Over here we have anencephaly then aprosencephaly absence
of the forebrain and acephaly, absence
of the cerebral hemispheres.
Then there are less severe forms.
Then there are several of those.
And then there's two specific forms called
minimal and microform.
So we're just gonna focus on the
DeMyer spectrum right here.

But this is classic alobar.
We heard Phyllis talk, give some beautiful examples
of it in the first trimester, the fused thalami,
the monoventricle, the solitary cortex there.
Here's a really pretty view on a sagittal mr,
the so-called bawling cup appearance
where you have the residual cortex here
and a huge amount of fluid in the back.
Then there's a spectrum from alobar
to semilobar to lobar.
There's no absolute delineation there.
But notice that we have fused cortex across the front
of the brain and on the mr the fused
cortex across the front of the brain.
Then in lobar holoprosencephaly, the cleavage
continues successfully from posterior to
rostral right there.
So we do have some separation up high,
but if you go down towards the cribriform plate,
there's fusion there.
And what is this?
This is really classic learn to look for this.
This is the fused fornices coming back
to those ever present fornices there and Dr.
Palou from Italy wrote a nice paper talking about
how if you see this, it's pathognomonic for alobar prosencephaly.
My partner Ann Kennedy, had a nice paper saying,
that's a great rule, but there are some exceptions.
And they went over the exceptions.
But bottom line, if you see the single cord here,
think fused fornices.
And the first thing you should think
of is lobar holoprosencephaly,
but realize that it's also seen in other things.
Obviously we don't need absence of the CSP
to make the diagnosis right here.
There's so much more going on.

And then here's what was described in the 1990s by Kovich.
Here's an in-utero scan showing cleavage in
front, cleavage in back.
But on this coronal autopsy that we did for the parents,
there's continuity of the cortex across the vertex
of the head up here.
So this is called the middle interhemispheric variant.
Here's a case that we had from last year
where you can see on the axial images cleavage in front
and back, cleavage in front and back.
But on the coronal images, we go from cleavage in front
to complete continuity across the midline here
to cleavage in back
and clinically this is equivalent
to semilobar holoprosencephaly in terms of outcomes.

Next schizencephaly.
I learned that this was a vascular insult of the MCA, kind
of the more, accepted term now is a neuronal migrational
anomaly, but it's complex.
There are two types, the open lip
and the closed lip, which are basically just on
the spectrum there.
This is a hard diagnosis to make in the second trimester.
So non visualization of the CSP may be your only clue.
Wesley Lee had a nice paper talking about
how absent cavum is noted in 70% of cases of schizencephaly
right there, particularly when it's bilateral.
As we said before, like many things in the brain, the,
there are lots of other associations, big ventricles,
agenesis, polymicrogyria,
heterotopia if you like, etymology.
This comes from the Greek schizo,
which means I tear, split or cut.
This is where we get our schizophrenia diagnosis
and cephale, which is brain.
So a tear or split in the brain.
And here you see on the ultrasound right here,
here's the mr showing the open lip type
notice on this one we don't see the septa lucida.
So just because you see fluid, don't call it,
here's the open lip cleft here
and notice the reverberation artifact anteriorly.
So you think, ah, it's just unilateral.
But we miss a lot of bilateral schizencephaly
because of this artifact.
In this case though, on the mr,
it was truly just unilateral.

Next hydranencephaly.
Basically the whole top of the brain just turns to mush.
The classic explanation is a huge infarct coming from
bilateral internal carotid artery occlusion.
More recent data suggests
that it might be viral in many cases.
They have a great model in cows for this.
And in monkeys, if you ligate the internal jugular veins
bilaterally, you get the same thing.
So combination, viral
or vascular there obviously non visualization
of the CSP is an expected finding.
We're not gonna require that.
But notice that we have midline structures.
So this is not alobar holoprosencephaly.
We have a little bit of brain.
And what happens is that you get a little bit
of brain in the temporal lobes
because you have collateral circulation from the basilar,
vertebral, basilar, circulation right there.
But basically nothing above that.
But we still have separation. So this is not alobar.

Basilar encephalocele and in Kovich when he described,
talked about absence of the, septum right there.
Two of his three patients with basilar, encephaloceles
had, agenesis.
So there's fluid, but there's no clear walls.
Here's the classic basilar encephalocele
and the western world.
It's occipital. If you live in Singapore,
it's frontal or occipital.
Here's the pretty 3D sagittal showing it.
Obviously prognosis depends on
how much brain is out right there, but no CSP.

And then finally, it's not real sexy,
but it's common chronic,
severe hydrocephalus from whatever cause It could be
aqueduct stenosis, it could be Chiari two,
it could be anything there.
But in this case, the CSP formed
and it was normal embryologically,
but then the severe hydrocephalus just tore it like the
sheets of a, old fashioned sailing ship in a gale right there.
It just gets fenestrated and ripped.
And this is actually one of the more common causes,
just because hydrocephalus is common.
So in this case, which was aqueduct stenosis,
we have big lateral ventricles.
The pressure from the big ventricles just tears
or fenestrations the septa pellucida.

So that was our fourth reason, an excuse
to review a whole host
of midline abnormalities right there from holopros to schizencephaly
to hydranencephaly, basilar encephalocele
and chronic severe hydrocephalus.

Imaging Pitfalls

This is usually what the residents say
to me when I talk too much.
So let's just finish with, four pitfalls here,
and we'll go through these one by one
pitfall number one from Peter Callan's paper.
If you take your plane of section too far, inferiorly,
you're getting the paired columns of the fornices
with the coaptation of the fornices in the middle,
making this white line.
This hypoechoic area is not fluid,
it's actually the fiber tracts in the fornix.
And again, if you're a purist,
there's some debate about this,
but this is the most accepted explanation.

Next, don't confuse the high riding third ventricle
or a prominent interhemispheric fissure for the cavum.
The cavum should be further rostral here.
This is in the midline. You have no corpus callosum.
The third ventricle floats up like a helium balloon
number next, normal variant, Cavum Vergae.
And so this is not a high riding third ventricle.
This is not an inter hemispheric cyst.
This is the normal cavum.
It's just a little bit bigger than normal.
Most people say it's a normal variant.

And number four, to quote one of the gurus,
Roy Philly, don't make it up.
There are no clear walls here.
So we have fluid in front,
but we don't have the clear walls.
So don't just say, oh, there's fluid up there.
But this was a case of septo-optic dysplasia
and we saw other cases of schizencephaly.

Conclusion

So why does it matter reviewing the four, reasons.
Number one, we have to do it by the guidelines.
Remember, those are defacto standards.
Number two, and most importantly, the common
and difficult to diagnose.
This is one of the easiest ways to pick up agenesis
of the corpus callosum.
Number three, uncommon
and essentially impossible to diagnose
entities like septo-optic dysplasia.
We've had four of these in the last two years.
And then number four, it's an excuse to review a whole host
of midline abnormalities.
And we went through some cool embryology there.

So to review the talk, we talked about
what it was from embryology anatomy and etymology.
We talked about why it matters, the four main reasons.
And we talked about imaging pitfalls.
I want to thank Professor Tescano, with all the Latin,
any mistakes in here are mine, not hers.
And if you like the talks, go
to the NASA astronomy picture of the day.
If you like the pictures and they have some
gorgeous images up there.
Thank you very much.