NEONATAL seizures are common! Do we ALWAYS treat them? What are the (million!) diseases that can cause seizures in babies? What is EPILEPSY? How do we work-up the cause for the seizure? Where do we start!!?? Do we always need a spinal tap? How do we manage seizures? What about imaging? What medications do we use? When do we STOP meds? What is the prognosis of an infant with seizures? --------------------------------------------------------------------- Do you want extra Tala Talks education? A neonatal tip a week? Access to LIVE videos? Behind the scene members-only videos? Worksheets of the videos? Do you want to decide which videos we should be publishing next? If so- join our new (low-priced!) membership community! Join here: https://www.youtube.com/channel/UCxv2qGB7HR5lR2D7IdoUAJA/join Help us guide the future of this channel and get involved in our community!
For PDF summaries of the videos, links to relevant articles (+ 50 free Multiple Choice Questions/ Answers and explanations!!), AND a bunch of other stuff, subscribe to our Neonatal Nuggets Newsletter here: https://talatalksnicu.substack.com/subscribe ---------------------------------------------------------------------- Please be sure to give us a like, comment about future topics, and subscribe to this channel to stay up to date on all things NICU! Also, click the notification bell after you subscribe to gain your weekly NICU knowledge as soon as it's loaded! --------------------------------------------------------------------- Dr. Tala is a board-certified neonatologist and has worked in busy level III and IV units for the past 17 years. She has won multiple teaching awards throughout her time as a neonatologist. ---------------------------------------------------------------------- References: Thumbnail from this website: https://www.childneurologyfoundation.org/disorder/infantile-spasms/ https://drdina.ca Epilepsia, Volume: 62, Issue: 3, Pages: 615-628, First published: 01 February 2021, DOI: (10.1111/epi.16815) Pressler RM, Cilio MR, Mizrahi EM, Moshé SL, Nunes ML, Plouin P, Vanhatalo S, Yozawitz E, de Vries LS, Puthenveettil Vinayan K, Triki CC, Wilmshurst JM, Yamamoto H, Zuberi SM. The ILAE classification of seizures and the epilepsies: Modification for seizures in the neonate. Position paper by the ILAE Task Force on Neonatal Seizures. Epilepsia. 2021 Mar;62(3):615-628. doi: 10.1111/epi.16815. Epub 2021 Feb 1. PMID: 33522601. Huang LT, Tain YL, Lai MC, Yang SN. Neonatal seizures: dialogues between clinic and bench. J Formos Med Assoc. 2012 May;111(5):239-44. doi: 10.1016/j.jfma.2011.12.007. Epub 2012 Apr 7. PMID: 22656393. Mosca S, Martins J, Temudo T. Transient benign paroxysmal movement disorders in infancy. Rev Neurol. 2022 Feb 16;74(4):135-140. English, Spanish. doi: 10.33588/rn.7404.2021326. PMID: 35148422; PMCID: PMC11502177. Sandoval Karamian AG, DiGiovine MP, Massey SL. Neonatal Seizures. Pediatr Rev. 2024 Jul 1;45(7):381-393. doi: 10.1542/pir.2023-006016. PMID: 38945992. Santarone ME, Pietrafusa N, Fusco L. Neonatal seizures: When semiology points to etiology. Seizure. 2020 Aug;80:161-165. doi: 10.1016/j.seizure.2020.06.025. Epub 2020 Jun 18. PMID: 32585613. Epilepsia, Volume: 62, Issue: 3, Pages: 615-628, First published: 01 February 2021, DOI: (10.1111/epi.16815) Malone A, Ryan CA, Fitzgerald A, Burgoyne L, Connolly S, Boylan GB. Interobserver agreement in neonatal seizure identification. Epilepsia. 2009 Sep;50(9):2097-101. doi: 10.1111/j.1528-1167.2009.02132.x. Epub 2009 Jun 1. PMID: 19490044. Geneviève Malfilâtre, Luc Mony, Danièle Hasaerts, Patricia Vignolo-Diard, Marie-Dominique Lamblin, Emilie Bourel-Ponchel, Technical recommendations and interpretation guidelines for electroencephalography for premature and full-term newborns, Neurophysiologie Clinique, Volume 51, Issue 1, 2021, Pages 35-60,ISSN 0987-7053, https://doi.org/10.1016/j.neucli.2020.10.005. Puthenveettil, Vinayan & Moshe, Solomon. (2014). Neonatal seizures and epilepsies. International Journal of Epilepsy. 1. 10.1016/j.ijep.2014.08.001. Summarization of neonatal video EEG for seizure and artifact detection - Scientific Figure on ResearchGate. Available from: https://www.researchgate.net/figure/A-Typical-Setup-of-Video-EEG-System_fig2_279851141 [ Yozawitz E. Neonatal Seizures. N Engl J Med. 2023 May 4;388(18):1692-1700. doi: 10.1056/NEJMra2300188. PMID: 37133587.
---------------------------------------------------------------------------- ***TIMESTAMPS*** 00:00 Intro 00:42 What causes seizures? 07:20 Work-up of seizures? 23:55 Manage and treat seizures? 31:!5 Prognosis after seizures ---------------------------------------------------------------------------- *Disclaimer*: This video is intended for educational purposes only and while we strive to give the most accurate information, errors may occur. Subsequently, this video should not be a replacement for medical advice.
AI Generated·May contain errors
Now, because you just watched the first video on seizures, you know what a seizure is and
how we diagnose them.
And hopefully you've been applying your clinical skills in the NICU.
However, that isn't really that helpful unless you know how to figure out what caused the
seizure because most seizures are provoked and how we treat them.
So that's what we're going to be covering today.
Hi, I'm Dr. Tala and I've been a neonatologist for 17 years now.
And today, in our follow-up video, we are going to be discussing what causes the seizures,
how do we find the underlying cause, how do we treat seizures, and what the prognosis
is for infants that actually have had a seizure.
Let's start with one.
What causes seizures?
Well, as you can imagine, anything that messes with a baby's brain has a potential to cause
seizures.
The other important thing that you should realize in neonates, as opposed to older kids
and adults, is that the vast majority of the seizures are provoked.
That most of the time, something is causing the seizure to happen and it is a short occurrence.
And if we get rid of whatever caused it to happen, then hopefully the baby won't have
any more seizures.
This differs from epilepsy.
And remember that epilepsy is when somebody has at least two or more unprovoked seizures
with no obvious underlying cause.
So for example, if a baby has meningitis and then has a seizure, even if they have two
or three seizures, then this doesn't count as epilepsy because the baby is getting seizures
because of the meningitis.
Relatively few of the seizures that are actually diagnosed are idiopathic, which means that
we don't really know why they happen, or considered neonatal epilepsy syndromes.
And this is where there are actual structural or genetic abnormalities or an inborn error
of metabolism that is causing the baby to actually have these seizures.
Obviously, even though it is quite rare as a cause of seizure, it is very important to
pick up on them, not just for treatment, but also for prognosis.
What are the most common causes of seizures?
Well, in term babies, it's HIE or hypoxic ischemic encephalopathy.
And normally these seizures happen six to eight hours after the insult.
So basically six to eight hours after birth, and can happen up to about 24 hours of life.
They can also happen as the baby's being re-warmed if the baby's being cooled.
So that's also another aside.
When the baby has a seizure, can also help you kind of figure out what's causing that
seizure.
In preterm babies, as you can guess, the highest cause for having a seizure would be a bleed,
specifically an IVH.
Overall, though, anything that messes up with the functioning of the brain, whether it's
a mass, a bleed, a metabolic abnormality, anything really can actually end up causing
a seizure.
Look at this pie chart from the New England Journal of Medicine, and you can see like
the relative causes of each etiology causing the seizure.
Obviously these are kind of like broad categories, and it doesn't go into the minute details
of all the tiny different things that can cause a seizure.
And in reality, there's probably a bit of overlapping in the seizure etiology.
So check out this chart from Pediatrics in Review, and you can see all the different
large groups of causes of things that cause seizures.
You can see that the percentages are actually a little bit different from the New England
Journal of Medicine one, and you have to realize that in your unit, there's going to be different
percentages of what's causing seizures.
If you don't have any preemies in your unit, then you are not going to have a lot of IVHs
causing seizures.
Let's cover a few more of those etiologies, just kind of like in a very broad context.
So like we already said, in a premature baby, they're more likely to have an IVH, but an
older baby could also have, or a preemie baby could have a bleed elsewhere as well.
For example, a subdural hematoma if there's trauma at the time of delivery, also a subarachnoid
hemorrhage.
Both of these can end up in seizures in babies.
Perinatal arterial ischemic stroke can happen usually from an embolism.
So like a little clot travels up through the blood vessels and then ends up in an artery
in the brain.
So that area of the brain onwards doesn't get all the blood that it needs.
Often that embolism can come from like the placenta or the umbilical cord, or really
just about anywhere in the body.
If a baby does have a stroke, the most common finding that they actually have is a seizure.
Often it is like a focal seizure as well.
A focal clonic seizure, now that you know those words, and normally it will happen kind
of 24 to 72 hours after birth.
A perinatal stroke should be considered in this situation, especially if there are these
risk factors.
So if the mother has oligohydramnios, choreoamnionitis, premature rupture of the membranes, preeclampsia,
diabetes, or a smoking history, and the infant has risk factors as well, or the infant has
risk factors such as congenital heart disease, an infection, coagulopathy.
Also boys are more likely to have an ischemic stroke than girls.
Since most of the strokes are most likely to happen in the middle cerebral artery, which
also covers the area of the motor cortex, these babies are most likely to present with
like a clonic focal seizure.
So again, pretty easy to pick up.
And if you see it, then you can be pretty certain the baby is having a seizure.
Also acute metabolic disturbances can cause seizures.
So for example, having abnormal electrolytes.
So as you all know, if you have hypoglycemia, if you have hypocalcemia, hyponatremia, or
hypernatremia, so low or high sodium, or hypomagnesemia, these can all cause neonatal seizures.
Realize that even if a baby does have a seizure, if you then correct the metabolic abnormality,
then the seizures should stop.
And these are babies that don't need to be on anti-seizure medications.
And of course, acute symptomatic seizures can also be caused by infections.
So for example, a group B strep meningitis or a carpi simplex virus encephalitis.
And this is obviously one of the important things to think about.
As we've all gotten so much better at preventing and treating these babies, this cause as a
cause of seizures is definitely going down.
And then there are so many brain abnormalities that can end up with seizures.
Whether it's a congenital brain malformation, for example, a schizencephaly or a lisencephaly,
or whether there is a tumor in the brain or a bleed in the brain of uncertain etiology,
whether it's hydrocephalus.
So there are a lot of structural abnormalities that can end up presenting because a baby
does have a seizure.
Two, how do we go about finding the underlying cause for the seizure?
Before we get to the diagnostic algorithm, I just want to talk briefly about the importance
of getting a good family history, birth history, perinatal history, as well as doing a good
physical exam to make sure that you're not missing any other reasons for why the baby
might be having a seizure.
We'll start with a family history.
Obviously, if the mother tells you that she has generations of her family that have had
epilepsy, then there's a high chance that there is a genetic reason for why this baby
might end up having a seizure.
And it really is epilepsy in this situation, as in not an acute reason for provoking the
seizure.
Some seizures, for example, those due to non-ketotic hyperglycemia, an inborn error of metabolism,
may start in utero.
And the mother may report that she had these frequent episodes of like this rhythmic jerking
by the baby.
Sometimes the mothers think that the babies are hiccuping.
And really, after the baby is born, we figure out that the baby is actually having seizures.
I don't want anybody to watch this thinking when their baby is hiccuping that they're
really seizuring.
Babies hiccup a lot, so don't freak out.
I'm just saying it's easy to look the other way.
Now let's move on to the physical exam.
Obviously, you should be doing a really good physical exam because it can also reveal things
for why the baby might be having a seizure.
So if the baby has microcephaly or a really small head, then maybe the baby has cerebral
dysgenesis or has a genetic abnormality, or baby has an infection like a torch infection
that's causing the seizure.
Macrocephaly or like a huge head, often that's associated with hydrocephalus, that can be
associated with a structural abnormality or a genetic abnormality.
For example, babies, especially baby boys, because it's X-linked, can be born with an
aqueductal stenosis.
Really, anytime a baby has any abnormal dysmorphic features, then you should be concerned that
something is going on with the brain.
This is especially true when the abnormal features are central.
So for example, if there is something abnormal with the nose or you have a central incisor
or, for example, you have seizures in a cleft palate, I would be very worried that there
is something abnormal with the brain.
Other things that you should be looking at is like hyper or hypotelorism, if the eyes
are too close or too far apart from each other.
So really take a very good look at the baby for dysmorphic features.
Then take a very good look at the baby's skin because abnormalities you find may suggest
a neurocutaneous syndrome.
For example, in neurofibromatosis, you might see cafe au lait patches, which are actually
very difficult to see in the newborn period.
In tuberous sclerosis, you may see ash leaf spots.
Actually, tuberous sclerosis is one of the few neurocutaneous syndromes where the babies
may start having seizures in the neonatal period.
So again, you're looking for ash leaf spots in tuberous sclerosis, or as you would definitely
not miss, you may see a port wine stain in Sturge-Weber syndrome, or you could see a
rash suggestive of incontinentia pigmenti, which honestly I've never seen, but I've had
to study, and it's a very rare disease.
It's X-linked dominant, and it starts off looking like a blistering rash, and then it
devolves into kind of like wart-like lesions.
This can be associated with neurological abnormalities, including seizures.
Obviously, do a good cardiac exam.
If the baby does have congenital heart disease, then the baby is at increased risk of having
a stroke.
So that's the board answer with the congenital heart disease and focal seizures.
And then overall, what does the baby look like?
Does the baby seem to have some sort of encephalopathy, so something actually affecting the underlying
functioning of the brain?
Is the baby sleepy and lethargic?
Is there decreased tone?
Is there increased tone?
What are the reflexes like?
What's the suck like?
What are the pupillary reactions like?
So you can have an overlying feel of whether this is something very specific in the brain
or whether the whole body seems to be affected.
After you take a good history and perform a good physical exam, you can start going
down the diagnostic algorithm.
So for now, what I'm going to follow is the diagnostic algorithm put forward by Dr. Jozewicz
in the New England Journal of Medicine article.
Let's go through this beautiful diagnostic algorithm.
The first thing is, if you think the baby is having a seizure, then do some blood work
immediately.
Obviously, the easiest things to rule out would be some electrolyte abnormalities.
So send a set of electrolytes.
So you're checking the sodium, the calcium, the glucose, the magnesium, making sure that
those aren't the obvious causes for the seizure.
They also recommend just getting a CBC.
Even though it's not a great screening test for an infection, it's not sensitive or specific,
it's very easy to get and it's a bedside.
So get that as well.
And then if you are suspecting a seizure, then try to start setting up an EEG.
Obviously, it is well acknowledged that actually getting the EEG in place does take some time.
Obviously, if the baby had a seizure because of an abnormal electrolyte issue, then you
need to correct that.
Or if you think the baby has an infection, then obviously treat for the infection.
If the baby, like we already said, did have a seizure because of an abnormal electrolyte
issue, it is very unlikely that that baby is going to need an anti-seizure medication.
Just correct the electrolyte issue.
If those previous studies didn't reveal the underlying cause, then we have to broaden
the differential.
So one of the first things that you then get is a blood gas or preferably an arterial blood
gas to check for acidosis.
If the baby does have severe acidosis, then maybe it was just the hypoxemia and the acidosis
that caused the baby to have a seizure, just like in a baby with HIV.
But also it's important to know just how sick the baby is or how acidotic the baby is, because
if the baby is acidotic and the brain already isn't getting enough oxygen, then anything
else that the baby might be going through, like a borderline low sugar or something else,
is a lot more likely to cause a seizure.
Like a lot of things in medicine, if two things are bad, not just one, then you're much more
likely to have that worse outcome.
Then evaluate for infection.
At this point, you're doing definitely more than just sending a CBC.
You're probably sending cultures, probably a blood culture, a urine culture, depending
on how old the baby is.
And then at this point, you're probably also doing some sort of screening for torch infections.
So for toxoplasmosis, for rubella, CMV, herpes simplex virus, HIV, hepatitis, syphilis, the
whole nine yards, all of these can cause seizures.
You should be performing a lumbar puncture or a spinal tap.
I was actually taught that literally any time a baby has a seizure, unless there is a very
clear obvious reason for it, so a preemie baby has like a massive head bleed or a baby
has a sugar of five for like three days, then we should be doing a spinal tap.
Remember that if you do a spinal tap, then make sure that you try to keep extra CSF.
So keep an extra bottle with extra fluid in it.
Because if we don't have an infectious reason or the tests that you send don't actually
give us the answer, then neurology may come back and ask for all these like metabolic
labs that they want to send on the CSF to make weird diagnoses of inborn errors of metabolism.
Another caveat here, if you think that the baby is having seizures because of an infection,
but the baby is too unstable to actually tap, do not stop from treating before you get the
tap.
Obviously, anytime that you're worried that a baby is really sick from an infection, you
want to get those antibiotics or those antivirals in as quickly as possible.
In this algorithm, it actually says at this point to perform genetic testing.
This is a bit interesting to me because it is kind of quite early on.
Other references suggest actually sending genetic testing if the initial evaluations
come back negative and the imaging and everything is negative, or if there's a strong family
history of having epilepsy.
There are genetic panels that you can send, so you can send the specific study for epilepsy
or the different genes that are involved in possible epilepsy syndromes.
Just be sure that if you are sending that panel at this point, you really should have
a neurologist or a geneticist involved just to make sure that you're sending the right
one.
Obviously, at some point, we need to be doing neuroimaging.
If a baby has a seizure, even if you think it is from meningitis, then you should be
getting some sort of neuroimaging to make sure that nothing else is going on structurally
or the baby doesn't have a bleed or something else abnormal as well.
As you all know, the ultrasonography or an ultrasound of the brain is always kind of
like our first line step, and we can do it because the babies have the anterior fontanelle,
so we have a window into their brain.
They are quick.
They are easy to use.
We can do it at bedside so we don't have to transport a sick kid, and we can get the results
very, very quickly.
And because of the ease of access, it is recommended at this step just to get the ultrasound first.
And ultrasounds are still really good at detecting bleeds, especially the larger bleeds.
They might miss some tiny ones, maybe would miss a tiny subarachnoid hemorrhage or a tiny
subdural hemorrhage.
They are not as good at getting stuff in the posterior aspect of the brain, and you may
miss some AVMs or vein of gallen malformations and stuff like that.
But overall, as an initial screening test to look at the brain, they're fantastic.
The other slight caveat that you have to realize is that ultrasounds are user-dependent, so
the pictures are only as good as the images that are obtained by the ultrasound technician.
As part of this step, we should be monitoring the EEG.
Obviously, at this point, we don't have an obvious reason for the seizures.
We're still trying to figure out what's going on.
We're still worried about the seizures.
We really should be monitoring the electrical activity of the brain at this point.
And like we already said, this EEG should be monitored continuously, even as we start
the baby on the antiepileptic medication, because sometimes the electrical activity,
the electrical seizures continue, even if the baby clinically doesn't look like it's
having a seizure any longer.
So make sure that you're continuing to monitor the baby on the EEG.
We should also be monitoring the baby on an EKG or the electrocardiogram, just monitoring
the baby's heart rate and the SATs and everything.
If a neonate is seizing, hopefully you've already moved them to the NICU and you are
actually putting them on a monitor and watching that.
Sometimes very rarely, arrhythmias can actually cause seizures.
But really why this is more important is that we can see the autonomic signs of seizures
on the EKG and make sure that the baby is stable, even if the baby is having seizures.
So yes, definitely be monitoring these babies.
Obviously, at this point, if you found out what's causing the seizure, then you should
be trying to actually treat that cause if you can.
So if it is hyperglycemia or if it is herpes encephalitis, then make sure that you do start
acyclovir, whatever it is that you think is causing it, as well as actually treating the
seizures.
If the baby is still having seizures and you still don't know what caused it, then we have
to start going down more diagnostic pathways into even more rare stuff that could be happening.
One of the next steps is to perform an MRI.
I would argue that really any time a baby has a seizure, you're probably going to end
up with an MRI, even if you have an etiology for it, even if you see a bleed on an ultrasound
As you all know, an MRI can give much more details of the structure of the brain, the
extent of the bleed, the amount of hypoxic ischemic damage that occurred, lots of other
things that could be missed on an ultrasound.
So really most of the time that a baby is having a seizure, we would want to get an
MRI, but especially if we don't have a reason for it.
As you all know, generally we're taking babies to radiology to get an MRI.
There are a few portable MRI machines.
So generally the baby has to be stable enough to actually transport them to MRI.
By the way, if you are suspecting a stroke based on clinical features, so for example,
like we said, we're seeing that focal clonic jerking or the focal clonic seizures, then
not just an MRI would be recommended, but an MRA and MRV, which is magnetic resonance
angiography and venography.
So that's where they're actually putting the dye into the arteries and the veins.
And you can get a really good idea about whether there was a stroke or whether there was an
embolus or something in those areas.
So consider that if you think that the baby has a stroke or if the MRI actually is indicative
that the baby did have a stroke.
If the baby is still having seizures and you've already tried the conventional anti-seizure
medications and nothing seems to be working, then at this point it's recommended to give
trials of pyridoxine, pyridoxal phosphate, folinic acid, and biotin to correct uncommon
vitamin responsive epilepsies.
Again, very rare, but if you're having these unresponsive seizures, then you really should
be trying this.
Continuing down the algorithm, then realize that this baby should be on EEG monitoring
until the baby is seizure free.
So at least for 12 to 24 hours.
You should also be considering inborn errors of metabolism at this point.
Like I said, we're getting into the more and more rare causes of the seizures.
So if you haven't already sent it, then send an ammonia, a lactate, make sure obviously
that the glucose is okay, the acid is okay.
At this point also consider sending the urine organic acids and serum amino acids.
Then I'm sure the neurologist will also want some weird metabolites in the CSF, but at
least you should be sending glucose and glycine in the CSF.
And that's where that extra CSF from the tubes comes in.
Other things that you should consider if you haven't done it yet, again, we said screening
for infection again, with the emphasis on torch infections and other weird infections
that you may be missing.
Was anybody in the family sick?
Again, we're reaching like rarer, rarer stages here at this point.
If you haven't sent any genetic studies, then at this point you should definitely be sending
some epileptic genetic panels, or you could do a microarray if that's what you have in
your unit or the whole exome sequencing.
Again, I recommend that if you are actually sending genetic testing, then get a neurologist
or a geneticist involved just so that you do end up sending the right thing.
Other things that you could do is get an ophthalmologist to come and look at the eyes, see if that
gives any clues, whether it's a torch infection or anything abnormal with the retina or anything
else, some weird inborn errors in metabolism can present that way.
And another thing that you could do is actually look at the placental pathology.
That can sometimes give clues like with a torch infection or some sort of embolism or
something like that.
In all, lots of diagnostic steps.
Realize you shouldn't just be throwing the sink at the baby immediately when the baby
first has a seizure, especially if it was just because of hyperglycemia.
So it's important to go step-by-step as you're also trying to stop the seizures from happening.
We don't want the baby to be in status just while you're waiting to send genetic panels
or whatever.
Three, how do we manage and treat seizures?
Well, like every emergency in medicine, we have to take care of the ABCs.
So take care of airway, breathing and circulation for the babies first.
Make sure that the baby is stable and breathing.
Then, like we said, the beginning of the algorithm is to immediately send labs.
Make sure that we're not missing obvious reversible causes of the seizures, the electrolyte disturbances.
Send a CBC as well and start antibiotics if that's what you need to do.
The goal of treatment of neonatal seizures is to reduce the seizure burden, but also
to try to prevent any poor neurodevelopmental outcomes that we would expect if loads of
seizures went untreated.
Neuronal apoptosis, which is where the neurons, the cells in the brain, undergo a pre-programmed
cell death, and poor neurodevelopmental outcomes have been associated with untreated seizures,
both in animal as well as human models.
So we know that having prolonged untreated seizures is bad.
Having said that, there is no high-quality data from randomized trials to suggest that
actually treating seizures actually helps neurodevelopmental outcomes.
And you can imagine just how difficult of a study that is to do.
I don't know if I would be willing to randomize into a study where my baby just didn't get
an anti-epileptic medication just because we were trying to figure out whether it would
affect neurodevelopmental outcome or not.
And like we talked about in the previous video, we shouldn't be treating anything that even
closely resembles a seizure.
We really should only be treating the seizures with EEG correlation.
So like if we actually see the seizures on the EEG, unless it is like a clonic seizure
that we would expect to see in a stroke or something.
So we're not treating every abnormal movement in a baby, even if it looks like a seizure.
But there isn't really a lot of data on exactly how or when we should be treating the babies
or which medications we should use.
And honestly, a lot of the medications that we do end up using and are actually recommended
to be used are actually off-label.
Then we need to be worried about the actual effects from the anti-seizure medications
on the brain.
So it has been shown in studies that phenobarbital may affect the development of the brain, may
actually induce apoptosis by itself, and may just affect neurodevelopmental outcome.
So this is not something that we're just giving very casually.
Having said that, it's still felt that the burden of untreated seizures is a lot worse
on the developing brain than actually giving the anti-seizure medication.
So right now, if there is a documented seizure on the EEG, we should be treating it.
The New England Journal article that we keep referencing actually suggests that we should
speak to the family about treating seizures, that it's not just such an obvious decision
to be made, that we should be weighing the risks and the benefits of giving babies anti-seizure
medications and kind of consenting them.
So which medications?
The ILAE recommends that phenobarbital should be our starting medication to stop seizures,
no matter the cause of the seizure.
So whether it was HIE or a stroke or IVH, we should be loading babies with phenobarbital
first.
The recommended loading dose is 20 milligrams per kilogram.
And then if you need to give like another little dose, a bolus loading dose, you can
give another 10 to 20 milligrams per kilogram.
Realize that if you're giving like above 40 milligrams per kilogram in a short period
of time, that's when you can really start affecting the baby's respiratory status and
the blood pressure.
You can end up with respiratory depression and hypertension, and it can really throw
off the baby's neurological exam.
They can become just really out of it and lethargic just from the medication.
We went through this phase of sometimes using Keppra or Levetiracetam, I never really used
that word, instead of starting with phenobarb.
Keppra does seem to have fewer side effects, even if it's given in higher quantities.
And a study from a few years ago showed that maybe babies given Keppra had maybe improved
neurodevelopmental outcomes versus those that were given phenobarb.
However, a randomized study, so this was Levetiracetam versus phenobarb for neonatal seizures, the
which was a small to be randomized control trial.
It showed that phenobarbital was more effective than Keppra at 24 hours for the treatment
of neonatal seizures, which means that the phenobarb was much better at stopping the
seizures.
And that's exactly our goal, to actually stop the seizures from happening.
Generally, if the phenobarbital has still not managed to stop the seizures from happening,
then we'll give another medication.
It's really uncertain which medication we should be giving next.
Often, we will give Keppra about 60 milligrams per kilogram or a little bit less than that
as a bolus, or we'll give phenytoin, which is we give it at a 20 milligram per kilogram
bolus.
We could also be giving a midazolam, which is a benzodiazepine or a lidocaine as well
to try to stop the seizures.
Again, not a lot of clear evidence which medication we use as a second line drug.
This table from the New England Journal article goes over the different medications and the
doses that we could use after the phenobarbital.
Again, notice that phenobarbital is the only one that's actually FDA approved to treat
neonatal seizures.
And by the way, all the references are below.
The other really interesting thing about phenobarbital is that the volume of distribution of phenobarbital
is about one liter per kilo.
So let me tell you why this is helpful.
Whatever dose we give the baby, so say we give 20 milligrams per kilo of a phenobarbital
bolus, the level in blood of the phenobarbital will be about 20 micrograms per ml.
And so we know that a therapeutic dose for a baby is somewhere between 10 and 40 micrograms
per ml.
So we know that if we've given a 20 mg bolus, 20 mg per kg bolus, and then another 20 mg
per kg bolus, if we check that level in the baby's blood, most likely it will be very
close to 40 micrograms per ml, which is the level that we're looking for.
And when do we stop treatment?
This is such an age-old question.
And if you ask 10 neonatologists, 10 neurologists, you're going to get different replies.
Some people say as soon as the seizures are controlled, maybe 24 hours after the seizures
are controlled.
That's extremely aggressive.
Other people say that before the baby leaves the NICU, we can try to come off the anti-seizure
medication.
Or more traditionally, we've kept them on the anti-seizure medication for a few months.
Four, let's talk about the prognosis of neonatal seizures.
I'm sure, as you've already gathered, that the prognosis will really depend on what's
causing the seizure in the first place, how long the seizure lasted untreated, and as
well whether the seizure was responsive to an anti-seizure medication or not.
Obviously, if you have a horrible brain abnormality, then the seizure is just a very small part
of contributing to that poor neurodevelopmental outcome.
As you can probably guess, the ones with the worst prognosis are the ones with a genetic
or metabolic abnormality that kind of have this epilepsy syndrome that are basically
going to exist for life.
They're unprovoked, and it's something that's innately abnormal with the baby's brain
and with the baby's genes.
And so developmentally, they easily have the worst outcome.
It has been well documented, like we've already said, that untreated seizures can cause
neuronal apoptosis as well as hippocampal sclerosis.
And so that in itself can obviously worsen the outcome, but it's difficult to tease
out exactly how much of the bad outcome is from the untreated seizures or from the seizures
versus the original brain injury.
Data from several case series show that if babies have a long length of seizures in a
short period of time, then that will worsen outcome.
For example, we definitely don't want babies to be in status epilepticus.
So we don't want babies to have seizures for five minutes or longer.
That's the definition of status.
Also, another set of studies showed that if babies have more than 12 to 13 minutes of
seizures within an hour period, then their prognosis is a lot worse as well.
So we don't want babies to be like continuously seizing.
But there is good news from the other end of the spectrum.
And studies show that maybe an isolated provoked seizure, whether it goes away or it's treated
with an anti-seizure medication, in animals at least, does not seem to worsen the developmental
outcome.
In fact, animal studies suggest that maybe the immature brain is less likely to be damaged
or injured from a seizure than a more mature brain.
There are also certain EEG patterns, so background EEG patterns.
So what the brain looks like when the baby is not having a seizure on the EEG that can
be more predictive or prognosis.
As you can imagine, if the background EEG pattern is normal, then that's more likely
to end up with a completely good outcome and good prognosis.
If the background activity has background burst suppression or ictal activity, then
these are associated with a worse prognosis.
In one study with HIE, it was found that if the babies had a normal or slightly abnormal
background EEG, then they were most likely going to end up with a normal development.
Another good prognostic sign for HIE and seizures is if the seizures are completely controlled
by day of life two, then we can also be optimistic about a good developmental outcome.
Last thing I want to say about this, and that is that the background activity of the EEG
especially can be really affected by the state that the baby is in.
So literally whether the baby is awake or not, which they can figure out on the EEG,
but also just if the baby is being cooled or the baby's on a lot of medication,
especially anti-seizure medication or other narcotics or benzodiazepines or whatever,
that in itself can affect the background activity of the EEG.
So sometimes we have to wait until the baby is kind of like in a more normal state or
at least completely warmed up before we can trust that background activity, the EEG.
And that brings us to the end of our seizure videos.
I really do hope that you learned something.
If you're interested in more neonatal education, then please subscribe to this channel and
like this video if you did like it.
And if you are interested in telling us which videos you want to see next, then please think
about joining our lovely community and our channel membership.
Again, thank you so much for being here.