Basics in Perinatal Neurosonography - HD

Introduction

Hi, I'm Harris Cohen from Memphis, Tennessee.

I'm the chairman of radiology at the University of Tennessee Health Science Center and I am radiologist and chief at Le Bonheur Children's Hospital.

My lecture today will be on topics in perinatal neurosonography.

In a limited amount of time, we will review some normal fetal central nervous system anatomy.

Go over some current axioms of fetal central nervous system evaluation and diagnosis and note some technical concerns and points affecting imaging analysis.

We'll review some of the more common congenital anomalies.

Central nervous system anomalies are the second most common major anomaly after congenital heart disease.

They may be of devastating clinical consequence.

Knowledge of them are important for parents and obstetricians to make decisions that require accurate and reliable imaging information.

Traditionally, over the last 30 years that's been via ultrasound and over the last 15 years or so, MRI has aided us in diagnosis.

The vast majority of anomalies are already present by the time the process of early brain development is complete, the vast majority of anomalies are already present by the time ultrasound can adequately image affected portions of the brain.

A key to evaluating the abnormal is knowing what normal is.

Normal Fetal CNS Anatomy

This is an example of a fetal brain in which the frontal horns and cavum septum pellucidum are noted.

The black arrow now points to the frontal horn on the downside of this fetus and this arrow points to the cavum septum pellucidum seen more commonly in the premature neonate than the full term neonate, but obviously commonly seen in fetal life.

The arrow now points to the anchor sign, which is at the junction between left and right frontal brain.

When there is significant extra axial fluid, one may sometimes see black area separating the echogenic line into two lines between the frontal brain.

Here's an image of the normal corpus callosum and the cavum septum pellucidum and its posterior extension as a cavum vergae.

In sagittal images of a fetus, one can see the nose and lips here, and this is the front portion of the corpus callosum.

The front, the back, the cavum septum pellucidum extending posteriorly is the cavum vergae, perhaps more easily seen here, a more posterior extension.

The normal temporal brain can look echoic enough to suggest hydrocephalus.

One can see here that the downside brain is echoic here, but not echolucent to scare the reader into considering hydrocephalus.

It's a little more echolucent on this view and it is only the presence of this rectangular echogenic area as evidence of this sylvian fissure that helps us know that there is no hydrocephalus.

Another key point that prevents you from considering hydrocephalus is that the choroid plexus within the lateral ventricle has not drooped or dangled into a fluid containing area.

So despite the fact that it appears echogenic, it is not hydrocephalus.

And this is just another image similar to the one I showed, which highlights the fact that sometimes in evaluating the fetal brain, one may think one has asymmetry or asymmetrical hydrocephalus because one sees dilatation on the downside while the near side, the upside of the head is not seen to have hydrocephalus.

But that in part is because of a near field imaging artifact that fills in the area of the brain that would've been seen as anechoic dilated ventricle.

The example I'm showing you is in a normal brain without hydrocephalus, but again, highlights the echogenic nature of the temporal brain that looks almost like the CSF that surrounds the choroid plexus within the lateral ventricle.

And arrows are pointing to the near side fill in artifact.

This is another image, somewhat older, that uses black and white imaging to show color Doppler highlighting these three vessels within the sylvian fissure, which in a sense prove that this echoic area is not hydrocephalus.

And again, the choroid plexus within the atrium of the lateral ventricle does not droop down into the echoic area.

And again, the normal temporal brain can look echoic enough to suggest hydrocephalus.

I'm pointing out the middle cerebral artery tributaries with black arrows.

Their existence and the fact that the choroid plexus does not dangle into the echoic area can help prove normalcy and indicate a pseudo hydrocephalus so to speak, and not a hydrocephalus.

Another key important normal finding is the shape of the fetal cerebellum.

The shape of the fetal cerebellum is that of a bi lobed peanut or a figure of eight.

The cisterna magna, which sits behind it as an anechoic area, has an AP measurement between two and 10 millimeters.

That is what's seen in normal situation.

One can see the figure of eight or peanut shaped cerebellum with an anechoic area posterior to it.

That is the cisterna magna.

And here are other examples, the cerebellum with the cisterna magna behind it, the cerebellum with the cisterna magna behind it and perhaps an easier ability to see the vermis as a little bit different than the cerebellar hemispheres.

And I often liken this to Mr. Peanut and I drew a hat and some feet hopefully Planters won't get upset.

Ventriculomegaly and Hydrocephalus

Ventriculomegaly and hydrocephalus obviously is a key diagnosis and in very significant cases it's easily seen.

There is significant ventricular enlargement here.

There is significant ventricular enlargement on this image, which is somewhat oblique.

Ventriculomegaly or hydrocephalus often is the presenting sign of central nervous system abnormality.

Ventriculomegaly is probably the more correct term, meaning a dilated ventricle as opposed to hydrocephalus which intimates increased pressure within the system.

Increased cerebral spinal fluid volume in lateral ventricles unrelated to dysgenesis or cerebral atrophy is oftentimes a definition for ventriculomegaly.

It may be the dilated ventricles may be dilated on the basis of excessive cerebral spinal fluid formation.

The choroid plexus produces cerebral spinal fluid and in cases of choroid plexus papillomas, which are not common, one may produce excessive CSF to dilate up the system.

But more often it's due to decreased resorption of fluid.

The fluid normally occurs in the ventricle, part of it goes down the spinal column, the rest of it goes up the brain and both the one, the part that goes down the spinal column and then goes up the brain or the one the part that goes up the brain goes to the arachnoid granulations at the top of the head where they're resorbed.

An inability to resorb it perhaps on the basis of a intraventricular hemorrhage or other causes can cause significant CSF within the ventricular system and with and beyond it obstruction to CSF flow from its site of formation in the choroid plexus to the site of resorption in arachnoid granulations therefore is key to potentially causing ventriculomegaly.

When one sees ventriculomegaly, one has to know that there is a high rate of associated other anomalies, even if one doesn't see that at that time, serial exams may be needed for diagnosis of change in ventricular size over time, the methods of diagnosis of ventriculomegaly can be via gestalt and it can be seeing the choroid dangle within anechoic area.

One of the key aids in making this diagnosis has been the knowledge from the group in San Francisco in the 1980s that the atrium of lateral ventricle is the first area dilate up in ventriculomegaly, that it traditionally is less than 10 millimeters and that its average is 6.2 millimeters which indicates that 10 millimeters and greater would be at least three standard deviations beyond the norm.

Hence greater than 10 millimeters of dilation of the ventricle at the atrium is an indication of hydrocephalus.

The atrium of the lateral ventricle again is the first ventricular area to dilate in hydrocephalus and recognition of this allows early diagnosis.

The normal atrial measurements remain the same throughout pregnancy.

So unlike other measurement tools that had been used prior to this information, a diagnosis can be made without exact knowledge of the mother's last menstrual period.

An important issue and the San Francisco Group's paper was Cardoza et al. Radiology 1988.

So again, the best method for evaluating ventricular size for enlargement is to measure the atrium of the lateral ventricle and here are examples of that.

This is a fetal brain turned in a way that it looks similar to a coronal ultrasound image of a neonate.

I do this in part to tip my hat to the neonatal ultrasound work that helped give us information on how to evaluate fetal brain, but this is an example of a normal coronal fetal brain.

One can see choroid plexus and one sees no CSF surrounding it to intimate any abnormality.

A few millimeters of CSF can be seen in normal situations and we see some echogenic white matter one assesses this fetal brain very similar to the way one would assess the ultrasound, the neonatal brain.

And this is just an image of a neonatal brain similar to that fetal brain positioned in such a way in which one sees echogenic choroid plexus and a small amount of CSF and the anterior aspects of these lateral ventricles.

This is an example of a dangling choroid sign in which choroid is seen to droop into an enormously dilated ventricular system.

One can see a small amount of brain at the periphery proving significant hydrocephalus and going against something that people are concerned about in such situations.

That is to say hydranencephaly.

This is again another example of a dangling choroid in severe hydrocephalus.

This is a neonate whose septum pellucidum on one side is intact and on the other side is not seen.

This is usually on the basis of fenestration, often seen in long-term hydrocephalus.

In this case, we can see some echoes within the dilated frontal horn of this individual and know that that hydrocephalus probably on the basis of intraventricular hemorrhage led to some fenestration of the septum pellucidum.

If one sees fenestration, then one can then say that when one doesn't see portions of the septum pellucidum, it's due to increased pressure.

There are times, however, in which one might not see any septum pellucidum, I believe this is residual septum pellucidum in significantly dilated frontal horns, a dilated third ventricle in dilated atria of the lateral ventricles.

The atria have a little bit of echogenicity within them intimating that there has been hemorrhage in the past in this case again, one would believe that this was due to fenestration of the septum pellucidum.

However, one has to worry about the possibility if one saw no septum pellucidum, one would have to worry about absence of the septum pellucidum, absence of the septum pellucidum may be an abnormality just onto itself, but it may be associated with a number of other problems involving both the optic area and the olfactory area in what is known as the septo-optic dysplasia syndrome.

This is an example of a fetal brain that's dilated in a less than usual case in which there was a significant TORCH infection creating the bright echogenicity around the ventricular margins.

But one does not see septum pellucidum and one sees a dilated third ventricle and one would have to worry in this case about whether there is true absence of the septum pellucidum.

Fenestrations can be seen both in the fetus as well as the neonate.

And again, the image I'm showing you here shows no cavum septum pellucidum, no septum pellucidum, and therefore one is concerned about whether this is due to fenestrated septum pellucidum or absent septum pellucidum.

This is just an example of a fetus who's normal.

You can see an echoic temporal brain, but it's not anechoic and one sees frontal horns, frontal horns and septum pellucidum.

Again, absent septum pellucidum may be an isolated abnormality.

However, it may be associated with septo-optic dysplasia, the septo-optic dysplasia syndrome, which has associated hypothalamic dysfunction, hypothyroidism, as well as problems with vision, smell, coordination and intelligence.

Something I've learned from MRI this year that has added to my ultrasound knowledge has been based upon a series of cases sent to me by obstetricians in which there was an apparent boxed cavum in which no septum pellucidum was seen.

And there was a box for the cavum.

And we've looked at these patients and evaluated them and in most instances made a decision as to what the diagnosis was via ultrasound.

But in two cases, our MRI helped us, and this is an example of one of those cases.

This is a coronal MRI image and one sees somewhat dilated frontal horns, but a line extending with CSF from the frontal horn to the periphery of the brain.

And this is evidence of schizencephaly.

A sagittal image of the same patient shows a pointing of the lateral portion of the lateral ventricle, a pointing of it in the middle.

That again should be an indicator to worry about schizencephaly.

So schizencephaly, when it's very large, wide lip schizencephaly is easily diagnosed.

It is however less well diagnosed of its thin lip.

And in this case the image on your right is an individual with a wide lip schizencephaly on the left, but on the right has a schizencephaly with a much narrower lip and was more difficult to see.

So again, here's a fetus.

And I have an arrow pointing to a small extension off the ventricle that is an indicator of schizencephaly.

Sometimes the more global view of MRI may aid analysis.

And again, that was the MRI image of a neonate who we picked up schizencephaly in only because of the MRI.

Congenital CNS Malformations Classification

Congenital CNS malformations have classification systems.

There are dorsal induction abnormalities that include anencephaly and encephalocele myelomeningocele.

Meningocele Chiari two malformation all occurring at about four weeks gestational age way before anything can be seen via ultrasound.

Chiari regression syndrome purportedly occurs at four to seven weeks.

There are ventral induction disorders such as holoprosencephaly, which develops at five to six weeks, and the Dandy Walker malformation, which occurs at seven to 10 weeks.

Anencephaly

Anencephaly is the absence of cerebral hemispheres.

There is a large defect in or complete absence of skull.

The base of the skull and facial features are said by some to be unaffected, but I think the facial features are not normal.

I think the orbits look a little bit wider than normal.

This is an example of a patient with anencephaly.

The arrow points, the echogenic material seen above the orbits.

There is no calvarium, there was failed closure of the cranial end of the neural tube as in all cases of anencephaly.

The material seen above the orbits is area cerebrovasculosa, also known as angiostrongyloides.

Area cerebrovasculosa.

Encephalocele

Encephalocele intracranial structures protrude through a calvarial defect most often in the United States, a posterior midline defect in some areas in Asia frontal cephaloceles are more common.

Cephaloceles contain brain cranial meningoceles contain meninges only cranial meningoceles.

Cranial meningocele may be associated with other anomalies such as agenesis of the corpus callosum or Dandy Walker Complex Meckel Gruber syndrome is a syndrome in which encephalocele is associated with cystic kidneys and polydactyly.

Whenever one sees a cystic mass posterior to the calvarium, one has to decide whether it's an encephalocele which must have an associated cranial defect, whether it's a myelomeningocele involving the cervical vertebral bodies which require a vertebral defect or whether it's just a cystic mass such as a cystic hygroma.

This is an example.

The arrow is pointing to the calvarium that's intact with this major separation between the calvarium through which this large area of brain extended in an encephalocele.

Is the orbit anteriorly and the large amount of brain seen extending beyond the calvarium is part of the encephalocele.

This is an older image, but it shows an encephalocele extending from the top of this patient's head.

There is a calvarial defect that I pointed out and it's pointed out by those arrows.

And this individual also had an abdomen in which on this axial view, it consists only of kidneys, indicating that the kidneys are very large and they had cystic areas within them.

This was an individual who had large fetal kidneys in combination with an encephalocele who had Meckel Gruber syndrome.

Myelomeningocele

The fetal spine consists of three echogenic structures.

Two of them are the posterior elements which should converge posteriorly, not separate from each other as one goes further to the back and the skin is normally seen posterior to the spine and should be intact.

If it isn't, then you have to worry about a myelomeningocele.

If spinal material is contained, it's myelomeningocele, if only meninges are contained, it's a meningocele.

There is an intact skin line pointed out and on the image to your right, two posterior elements that are echogenic are seen in combination with the anterior element that make up the eventual vertebral body completely bony vertebral body.

But at this point, it's not all bony myelomeningeal analysis requires looking at the vertebral column.

When one sees this very nice image in sagittal display, one is really only looking at two of the three posterior elements.

So it really requires one to look down the vertebral column, transversely axially to make sure there's no defect in any of the vertebral bodies.

This is a sagittal image with the head over here and the buttocks over here in which if one goes down the vertebral column, there is a mass that was a bony abnormality, but a bony abnormality associated with a soft tissue mass seen posterior to the patient's back.

And this was a case of myelomeningocele.

On this image, one can see the posterior divergence of the elements and a soft tissue mass.

Another myelomeningocele.

The two arrows are seen to be divergent as one goes backward.

Sometimes one is fortunate enough to get an image in fetal life that is a coronal image that would allow one to make a diagnosis of myelomeningocele similar to what was made on plain film in neonates, which is seeing as one goes down the vertebral column, a widening of posterior elements.

And that is the case in this fetus who had a myelomeningocele.

Nowadays one can do coronal reconstructions on both 3D reconstructions on both ultrasound and MRI.

In this case, we were able to reconstruct the images of this fetus to show actual vertebral body sticking out of the normally straight spinal column that was associated with myelomeningocele.

It was once considered a miss if one had three vertebral bodies missing and hence if you were the radiologist and you didn't pick up myelomeningocele, you were to blame if there was a three vertebral body long defect.

But sometimes the baby was positioned and you couldn't tell sometimes the sac that extended beyond the back ruptured and sometimes the sac was obscured.

And so in those cases, despite high AFP levels, one might miss a myelomeningocele.

Again, even rigorous imaging looking for posterior mass, looking for divergence of posterior elements or noting a defect in the posterior line and not seeing one is limited by fetal positioning amniotic fluid levels as well as maternal body habitus.

Therefore, we were all very fortunate when again from the San Francisco group, information was developed that said a basic fact that myelomeningoceles are associated almost a hundred percent of the time with the Chiari two malformation.

That was a known piece of information.

The methodology to find this out via fetal ultrasound was taught to us and can be summed up in a sense with the banana sign.

Mr. Peanut. The normal cerebellum will if in a Chiari two malformation the intracranial contents is pulled down into the upper cervical area, it will lose its shape and become banana shaped and the cisterna magna will be effaced and will no longer be anywhere from two to three to 10 millimeters or so, but it'll be less than that indicating that Chiari two malformation occurred.

So again, this is an image that shows a normal peanut shaped cerebellum.

That's normal, that's a banana.

And this is an image in which there are two findings here.

One, the Mr. Peanut is not seen posteriorly, which is on the reader's left, and there also happens to be some flattening of the frontal bones, which we will talk about.

So I'm now gonna show you arrows that point to a banana or a plantain.

You no longer see the cerebellum.

You no longer see this cisterna magna.

Well, you don't see the cerebellum as its normal shape.

And this is a myelomeningocele.

This happens to be an ultrasound showing material from the subtentorial area in the high cervical region.

And this is an MRI showing material seen normally temporally a little bit lower than normal.

Again, a Chiari two malformation.

The lower arrows show the cystic abnormality posterior to the spine, which was a myelomeningocele.

The other sign that I intimated and we'll talk about now is the lemon sign loss of convexity or flattening of the frontal bones of the calvarium seen between 18 and 22 weeks according to Nyberg in the late eighties showed 71% sensitivity and a hundred percent specificity for myelomeningocele.

He had a nine, he indicated a 99.5% negative predictive value, meaning you don't see a lemon sign, there's no myelomeningocele.

Work by Ball and Philly published in 93 indicated that other intracranial and extracranial anomalies may have associated lemon sign.

That's a picture of a lemon.

And here's an example of a lemon sign.

The arrow is pointing to the concavity of bone, which is more sensitive than just a mere flattening of bone.

It has been indicated that mild lemons mere flattening of the frontal bone can be seen in 1.4% of normal fetuses.

Nyberg indicated that the shape of the lemon apparently returns to normal after 24 weeks and that the sign is unreliable after 24 weeks.

We were reviewing some work in my teaching file in downstate a number of years ago and this is an image from the days in which we shot Polaroid film, but one can see a lemon like calvarium and an individual who had a myelomeningocele.

But this, nobody indicated this was a lemon because this was way before it was first reported.

Holoprosencephaly

Holoprosencephaly has various forms.

The worst is a lobar holoprosencephaly.

Less bad is semilobar and less bad than that is lobar holoprosencephaly the classic alobar holoprosencephaly has a single ventricle fused thalami.

The prosencephalon or forebrain fails to develop into two hemispheres, a disorder of diverticulation.

HOX genes fail to activate in the midline, allowing structures normally paired on right and left side to fuse.

This is an example of a coronal image in one in which one is seeing a single ventricle and fused thalami a case of holoprosencephaly.

This is another image of the holoprosencephaly and a pathology specimen of one of the cases we had with holoprosencephaly single ventricle seen.

Holoprosencephaly patients have associated midline facial anomalies.

The coronal image of this patient's face shows the orbits closer together than normal.

This is an example of cyclopia.

There is a single nostril.

This is a normal child with two nostrils seen in the lower left hand corner.

This individual was born and one can see the single nostril that was associated as a facial anomaly with a patient's holoprosencephaly.

And there are many varieties of midline facial anomalies associated with holoprosencephaly.

This is an example just of the great images we can currently take with modern equipment of lips and nostrils of a normal individual.

And here we see a defect in which one sees a defect in the lip and can see the orphan associated nostril defect on the ipsilateral side of a one sided cleft.

Dandy-Walker Complex

The Dandy Walker complex is a spectrum of abnormalities.

Classic Dandy Walker malformation occurs 30% of the time there's aplasia or hypoplasia cerebellar vermis there's a large posterior fossa cyst communicating with a fourth ventricle through a defect in the vermis.

There's increased volume to the subtentorial area.

Uplifting of the tentorium hydrocephalus is frequently seen but not universal and extra central nervous system abnormalities are common.

Oftentimes one can learn that one is to worry about the Dandy Walker by seeing cisterna magna greater than 10 millimeters.

Dandy Walker variant as part of this Dandy Walker complex seen 15% of the time fluid-filled space communicates with the fourth ventricle, but the cyst is smaller.

There's milder vermian dysplasia, there's less hydrocephalus.

Posterior fossa is not enlarged.

And mega cisterna magna purportedly occurs in 54% of all the patients within the Dandy Walker complex in which all you see is a cisterna magna greater than 10 millimeters.

Despite the lesser anatomical abnormalities of the variant, they still carry a risk for aneuploidy.

And in a 92 article by Estroff, she indicated that 47% of patients with these milder abnormalities had extra CNS abnormalities and 29% of them had abnormal karyotypes.

Again, a view of the cerebellum and the cisterna magna posterior to it and a view of a very large cystic area posterior to the cerebellum.

And in absence of the vermis allowing the cystic mass posterior to extend into the fourth ventricle arrow pointing now to one cerebellar hemisphere, the enlarged cisterna magna communicates with the fourth ventricle through the area of vermian dysplasia, a Dandy Walker cyst.

And here are some postnatal images.

On your left is a coronal image of the brain showing a prominent cystic area.

I think it's a smiling lion sign.

And an area that's cystic between the two cerebellar hemispheres.

On the other hand, to your right is a posterior fontanel view in which the transducer is posterior.

And the first thing it sees is the cystic area posterior to cerebellum.

And one is seeing the cerebellar hemisphere and the other cerebellar hemisphere, but one see CSF extending out to area of the fourth ventricle.

This is just a sagittal view showing the tentorium as high posterior is to the reader's left.

To the viewer's left. And this is a more modern MRI showing a very high tentorium and a dysgenetic cerebellum in a patient with a Dandy Walker malformation.

Agenesis of the Corpus Callosum

Agenesis of the corpus callosum.

Another anomaly. I'm showing you an MRI of a normal corpus callosum in a child.

In cases in which the corpus is absent, there'll be wide separation of the frontal horns.

Some people call this Texas Longhorn sign.

There's a high riding third ventricle.

Sometimes the horns, frontal horns look like they're coming out of the third ventricle and that is called a Viking's helmet sign.

But they're all different names. These signs and what one mostly has is the corpus callosum is absent.

One no longer has the two ventricles in relationship to each other and they run parallel to each other sometimes with large white fiber tracks between them as in this case of Probst bundles, more apparent.

I'm gonna show you the post pro.

This is the paralleling lateral ventricles and the P stands for Probst bundles. These are a little more apparent than I usually see.

This is just an example of a high riding third ventricle with two frontal horns extending from it or apparently extending from it, which is the bull's head or Texas Longhorn sign of Viking's helmet sign.

These are just examples of a neonatal ultrasound and a neonatal MRI showing paralleling lateral ventricles.

One can see some colpocephaly, which is non-specific dilatation of the posterior aspect of the lateral ventricle compared to the anterior aspect.

One sees that on the ultrasound on both sides and on the MRI on the patient's left.

It has an association at times with agenesis of corpus callosum, but it is non-specific finding here are parallel white CSF filled ventricles on the mesial side of the brain in patients with agenesis of the corpus callosum, rather than having gyri and sulci conform to where the corpus callosum would be.

The gyri and sulci side point toward the area of the third ventricle or massa intermedia, as in the image you see on your right gyri that intersect the third ventricle is a common finding, the sunburst sign of agenesis of the corpus callosum.

Summary

In summary, in a limited time, we reviewed some normal anatomy.

We noted a few technical concerns to consider.

We reviewed a few key anomalies.

The key point, learn from your cases, learn from prenatal and postnatal, follow-up and learn from other modalities.

The end.