Multiprofessional Critical Care Review: Pediatric (2023)-Neurointensive Care and Monitoring

Neurointensive Care and Monitoring
Video Transcription
Good morning everyone. My name is Dr. Ed Conway. I'm the Chief of Pediatric Critical Care Medicine  and the Chairman of the Department of Pediatrics, Jacobi Medical Center in the Bronx, New York. I'm delighted to be invited today to the 11th presentation for the Pediatric CCM  Board Review Course. I'll be doing parts one and two of neurocritical care. Full disclosure,  I'm a prior question writer for the general PEDS exam as well as for the critical care exam,  and I served on the editorial board of PICU-PREP for eight years. I have nothing to disclose.  The first several slides that I'll be sharing with you are available on the American Board  of Pediatrics testing site. You log in under your name, your password, and find the Pediatric  Critical Care Medicine, the content outline, two of the slides which I'm sharing with you  on this slide. One is the critical care medicine content domains, of which you'll see there are 10  domains. No surprises there as you read through them, as well as, excuse me, universal tasks for  the critical care medicine. Basically, what we're expected to be able to do to learn to be  practitioners as well as to successfully pass the exam. Again, there are no surprises on these  particular slides. I find this slide exceedingly helpful because it tells you how the exam is made  up. If you look at the domains, the 10 domains, it gives you the percentages of the exam weight. Assuming that part A over four hours or part A over two hours has 200 questions,  that's about a minute of question. You should always figure out your time per question. That'll  help you as the exam goes on. All questions are worth the same. Do not leave any blanks at the  end if you run short of time. You're best to guess rather than leaving a blank. This just gives  you an example. If there were 200 questions and 16% are organ function and physiology, that would  be about 32 questions. Some people find this helpful when they're studying how much time to  put in where. One of the tips I'll give you is the areas that we seem to know best we spend the  most time on only because we feel most comfortable. The areas you're weakest on is where you should spend your time.  So  the calendar on the prior slide gives you an idea. I recorded this on April 1st and the exam is on  November 3rd. So as of today, you have 216 days to study. That time will evaporate quite rapidly.  Summer's coming, new jobs moving, new challenges, etc. So again, start to balance your study time  and allow a certain amount of time. The key thing, do as many old questions as you can find from any  source and there are many sources out there. But just to give you an idea on how the exam is  written, critical care practitioners are solicited from the American Board of Pediatrics. We're then  sent a batch of content specs. So for example, the spec for this particular slide would be  recognize the world's first surgical exorcism. So I would write a brief clinical scenario, two or three sentences to introduce the topic. I would then write the question  with the correct answer and three distractors. Approximately 80% of the test takers should get  it correct. So theoretically, you should be able to read the question, close your eyes before you  read the answers and know what the right answer is. It's a good feeling when it happens. As I  study or I tell people to study per topic, the way to break it up is recognition of an entity. You  know how to make the diagnosis, understand the etiology and the pathophysiology of the particular  entity they're asking about, know how to diagnose it, whether it be laboratories, scans, EMGs, EEGs,  et cetera. Know how to manage it and know the complications of the entity and also that may  be iatrogenically induced when caring for the said patient. So assuming, you know, based on at least  my experience with MOC and most exams going forward, there are now four choices rather than five. I can tell you as a question writer, it was always fairly easy to write a question,  come up with the right answer and then usually three other choices. Coming up with the fourth  distractor was usually quite difficult. So here is a sample question. Look at the picture. The  biggest mistake we make when taking examinations is not reading the question thoroughly, properly  or missing a word. So here you see a physician standing at a patient's bedside, literally  standing on the oxygen hose and you'll see the patient appears a little bit blue around the gills  there. So the question is based on this picture, which of the following would not cause cyanosis  in this patient? So would it be a peds emergency medicine doc, peds GI doc, an infectious disease  doctor or your critical care partner? Again, it's difficult. I would be playing some music in the  background here if we were doing this live. However, recording this makes it quite difficult  to multitask. So the correct answer here would be one or A, the peds ER doc. Think about it.  Your partner will be the first one at the bedside telling you what you should have done or what you  should do going forward because they're not on service. Peds ID doc would be a consultant, would  be a reasonable consult for most of our patients and peds GI would gladly come to the bedside.  You're not going to see the peds ER doc once that patient is admitted. So that would be the  correct answer for choice A. So I think the easiest way to think about this, we're going to do the  central nervous system in general, is what are the primary brain injuries? Trauma, which we covered  in another lecture, hemorrhage, ischemia, tumor and infection, which will also be covered in another  session. Secondary brain injuries, increased intracranial pressure, cerebral edema, hypoxia,  hypercarbia, and electrolyte abnormalities to cite just a few. So let's start with the anatomy.  We'll start here with the skin, which we can see here on the infant's scalp. Underneath, we have  bone, ostium. We then have the dura, dura meaning tough mother. Above it, the epidural space. Below  it, the subdural space. We then have the arachnoid membrane. And beneath that is the subarachnoid  space, which is the happening space of the brain. This is where the blood vasculature is located,  as well as the cranial nerves. And then we get down to the pia mater, which is actually on top  of the brain tissue itself and cortex. If you're a visual learner, here's the skull. Here's the  anterior fontanelle. And what we're looking at here is the dura. If we remove the dura, we are  looking at the epidural space, thus the epidural artery, which is the only artery that's not  protected under the dura. And if we remove the dura and we look down, what we see here is the  central sinus, the largest vein in the body. And we have the deep perforating veins. These are the  veins that get torn in non-accidental head trauma as the head's shaken to and fro. And then the  glistening here is we see the arachnoid. And under the arachnoid, we can see here some of the arteries. We can see some of the veins. And we'll come back to those a bit later.  So here, what I'd like to do is review a concept many of you are familiar with, which is the Monroe-Kelly hypothesis. And what it states that's in the black box,  which is sort of the brain and the other components, 80% is brain, 10% is CSF,  and 10% is blood volume. And again, this is then encased by dura and then protected by calvarium.  So the Kelly-Monroe concept, 80% brain, 10% CSF, 10% blood volume states that if one of these  components goes up, the other must go down. So if we look here at the intracranial volume,  if we have a little bit of brain swelling, for example, in a DKA with a little bit of edema,  the body can compensate. The body compensates by opening or draining into the craniospinal axis will displace CSF and some blood volume. If, for example, there's a massive bleed  and the pressure gets to a certain inflection point, you'll see that it takes off exponentially.  Usually a pressure of about 25 is called a plateau wave, is a harbinger of bad things to come. And what they like to test on the exam is concepts. So the question for something like  this might be, what causes you to move from 0.2 to 0.3? So that means there'd have to be  some increase in volume without a compensatory decrease somewhere else. And I'll show that on  the next slide reflected in spring diagrams. But what I put here, this is some data from  the laboratory. And what it shows, a normal ICP in infant is about 10. In an adult, it's about 15.  So if we assume this is normal here, what this shows is that if we do a craniectomy, so we remove the calvarium in the bony plate, we can add a small amount of volume and the brain  will take it, the cranial space. If we remove the dura, the tough inelastic membrane that's on top of the brain when it's being challenged and swelling, this shows that we  can tolerate a lot more volume added until we get to a dangerously high pressure, which is why  an all-out effort for a traumatic brain injury may be to remove the dura and the cranium and  then put some sort of synthetic dura in its place or bovine dura, and in essence, creating  a fontanelle, allowing that space to bulge to keep the pressure normal. So these are concepts  which are frequently tested. So for myself and many of the other aging intensivists out there,  an easy way to relieve some pressure would be to take away about two dozen passwords that we have to store up there, which our hospital system are always challenging us to change.  Having been an engineer in my prior life, we describe everything sort of in springs. So if we think of the Kelly-Monroe hypothesis that this outer dark box here is the dura matter  and the calvarium, and everything's sitting in a happy equilibrium here, 80% blood, I'm sorry, 10% blood, 80% brain tissue, 10% CSF gives a normal infant a pressure of 10.  Here, following a small head injury, we have a little bit of brain tissue that's contused,  and it's sort of, Kelly-Monroe says that maintain the pressure less than 20. If one goes up,  something else has to go down. So what you can see here is the swelling of the tissue with the blood  is forcing it down. So the blood, two thirds of the blood volume sits on the venous side  as compassionate vessels. It will displace down the cranial spinal axis. And then if we look and perhaps the injury isn't recognized, child's not brought to attention rapidly,  we can see that the swelling has increased, the hematoma is massively increased. And as we go  across here, the CSF has been displaced down the cranial spinal axis also. So I encourage you to  think of the brain as a sponge that's wet. And as you start to squeeze that sponge and give it less  room to expand, displacement, just think of the water dripping down is sort of the CSF and the  luxury blood on the venous side going down the cranial spinal axis. So the bigger you are, the bigger your cranial spinal axis, you'll tolerate it better.  So here's another sample question number two. A newly diagnosed seven-year-old with lupus is being treated with pulse steroids, presents with a sudden onset of headache,  visual changes, and several generalized brief tonic-clonic seizures, each of which lasting less  than five minutes. Her vital signs are a heart rate of 120, respiratory rate of 16. Her blood  pressure is 138 over 95. Again, a seven-year-old. Her MRI is shown below. Which of the following  statements is most correct about this entity? So take a look at the slide of the MRI. And now let's  take a look at the question. Which of the following statements is most correct about this entity?  Long-term use of antiepileptics will be required. Blood pressure should be rapidly lowered.  Long-acting blood pressure meds will be required. It's caused by dysregulation of cerebral blood flow. So the correct answer is D or 4, dysregulation of cerebral blood flow. What  this patient, this young lady, a seven-year-old with lupus, is demonstrating is what's called  PRES, posterior reversible encephalopathy syndrome. It consists of headaches, visual complaints, and seizures, subcortical edema without infarction, which is that white area in the  occipital area on the MRI. It's associated with particular drugs, cyclosporine and tacrolimus, or two of the more common, excuse me, is preferential involvement of the parietal  and the occipital lobes. And the hypothesis is that there's poor sympathetic innervation of the posterior circulation. So you don't want to acutely lower the blood pressure.  You don't want to use anything long-acting because once you bring the pressure down, you want to be able to sort of maintain it. And three, this patient, these kids do get better  over time. And here's a follow-up MRI of the same patient several weeks later. You'll notice  early on the occipital edema that shows up quite nicely here. You can see other areas that light up  a tad. And here you can see it's almost resolved and the child was back to totally normal function.  Another frequently tested concept is that of cerebral edema. So the key thing is to remember  that the brain has tight junctions and these are endothelial cells. So the vasculature is the  tight junction. If you saw a question, what's it comprised of? And there's different types of cerebral edema. So vasogenic edema, where there's leakage through the tight junction,  an example of this will be the current concept of cerebral edema seen in diabetic patients.  And then there's cytotoxic edema, where the endothelial cells may be destroyed and or the neurons as well and involve cell death and swelling in the entire area. Carbon monoxide  might be a good way to think of cytotoxin edema. And this afternoon, New York Governor George  Pataki highlighted the state's security plans for the July 4th holiday weekend. We will have  everything from biological chemical detection teams deployed in strategic areas of the state  to bomb dogs, radiological detectors, the most advanced technology, thousands of personnel.  Among those thousands, the deployment of 2,000 National Guard members and another 2,000 on standby. He's having a seizure.  We want them to be on alert, communicating with our intelligence.  Most importantly is the Indian Point nuclear power plant. More people live within 50 miles  of the facility than any other nuclear plant in the country. Of course, we're in continuous  communications with the federal government and the federal intelligence agency, so we're really  kept up to speed on what is going on around us and what the potential is. Westchester County  police have been reviewing their July 4th security plans every day for the last few weeks. Additional officers at the bomb squad will monitor the chemical damage.  On top of that, we'll have patients having a seizure.  Just flip it, turn it, turn it. This? Yeah. This way? Yeah.  Press the button.  Give me another one. No. Did you press? A few. Okay. Yeah, go ahead. Okay. You okay? Huh? What?  Thanks.  It just turned and I was like, oh no.  We're supposed to start tonight. Tonight. Tonight, yeah.  How long did it start? No, it wasn't that long. It wasn't that long. It wasn't that long. He's supposed to deliver? Yeah. Yeah. Yeah, it wasn't, but it was pretty violent. Okay.  Hi! He'll fall asleep, he won't come. He usually just knocks off. Especially if it was violent, he's probably dead tired. Probably dead tired.  It's a long time, isn't it? Yeah.  So, following the seizure witnessed in the preceding slide, when should an anti-epileptic be administered? Immediately, as soon as you walk into that room and the patient sees it.  Should you wait five minutes? Should you wait ten minutes? Or this type of seizure doesn't require an AED. Again, in the past, music would be playing in the background here,  perhaps the Jeopardy song again.  And so, again, remember, you probably have a minute, and this would be a freebie, I call it, that you can answer quickly and save time for some of the more challenging  questions that you will be faced with. The correct answer is after five minutes.  The more recent management recommendations for status epilepticus includes sort of a recently defined operational definition, where one witnesses a generalized tonic-clonic seizure,  lasting more than five minutes, or two consecutive seizures where the patient doesn't wake up between. Normally, generalized tonic-clonic seizures rarely last more than five  minutes. That's in about 70% of patients. However, we have no good markers or indicators of those patients and the type of seizure that will continue longer.  So it's unlikely if it doesn't stop by five minutes that it will. And the longer it continues, the seizure, the harder it's going to be to stop pharmacologically,  and there's a greater degree of neuronal damage. And the cells that are involved in epilepsy, it's the gray matter of the brain.  Other types of epilepsies and definitions one should be familiar with  include one called refractory status epilepticus. And that's a persistence of seizure activity, despite appropriate medical and anti-epileptic drugs.  These patients have continued motor presentations, automatisms, et cetera, of the seizures. They can last greater than 60 minutes,  or some use the definition having failed maximum dosing of two separate epileptic drugs, anti-epileptic drugs.  The incidence in adults is about 9% with a mortality of 39%. However, we don't know the true incidence of RSE in pediatric patients.  We do know that it has a high morbidity and mortality of 25 to 30%. And RSE is usually associated with encephalitis, i.e. some of the more significant, severe,  and rarer of the viral encephalitis, which will be covered in a separate talk. Another entity to be familiar with is one non-convulsive status epilepticus, NCSE.  The actual incidence in pediatrics is unknown. However, one needs to consider it after a patient has been successfully treated for a  full-blown status epilepticus, as witnessed several slides ago. These patients may appear sleepy, lethargic, altered mentation,  or subtle or absent motor movements. It's an EEG diagnosis. If a patient doesn't wake up after 30 to 60 minutes following a seizure from  the metabolic hyperactivity and or the effects of the drugs we use to stop the seizures, one should consider non-convulsive status and consider ordering an EEG.  I always smile when folks say, oh, we have to get a video EEG.  In NCSE, there's nothing to see. The patients are lying there with no motor activity. However, when you do an EEG, it may look like the next slide I'm going to share.  Here, one can actively see that there's active seizure activity going on in the lower half of this EEG. And the patient shown on the prior slide was just lying there with no motor  activity witnessed on the video.  Common etiologies of status epilepticus include fever, most commonly, medication change or not taking it or medication interaction.  Many of the AEDs are cleared in the liver and other drugs that may induce or  suppress P450 enzymes can have a contributing role there. Unknown is about 9%. Metabolic, always think hypoglycemia.  Hyponatremia. And lastly, hypocalcemia as potential causes.  Congenital brain abnormalities, anoxic injury, trauma, vascular infection, tumor, and drugs are a small percentage of the etiologies that can cause status  epilepticus.  Well, there are billions of neurons located in the human brain. There are many, there are two major classes that we concern ourselves with treating status  epilepticus. And one is the inhibitory system, which uses GABA as a neurotransmitter, as an inhibitory neuron telling the excitatory neurons to basically chill.  And then we have excitatory neurons that use glutamate as an excitatory  neurotransmitter. And then we have astrocytes, which actually have been found recently to play a role. So it's sort of a yin and a yang of balance in the CNS.  Many of the drugs that we use work either to enhance the inhibitory or to  block the excitatory neurons. And the neurons all speak to each other. So it's a cacophony of noise, neurotransmitter release.  There's hundreds of potential neurotransmitters in the human brain that can play a role here. One of the simplest ways to think about the yang and yin is sort of GABA  versus glutamate, the ones that keep you inhibitory. And then we have the neurotransmitters that are excitatory. And again, whether we can block one or enhance the other,  pretend for good neuro anti-epileptic drugs to stop neurotransmitters.  So the basic physiology of normal neuronal firing, is here's a cell that we want to either excitatory and tell it to fire or  inhibitory. So this would be a GABA cell, a neuron, and this would be a glutamate excitatory. And basically what happens as the resting potential lowers and action  potential in the cell fires, basically what benzodiazepines do and other drugs that work on GABA, GABA-like, they open the chloride up.  Chloride being negative pours into the cell, hyperpolarizes the cell. Therefore the cell cannot be depolarized. So by making the cell more negative with the chloride going in.  So many of the AEDs that are out there work in similar matters, blocking either sodium, potassium, or perhaps calcium channels.  This is basically a summary chart of early physiologic changes in seizing and  late, and this line is about 30 minutes. So early on is very hyperadnergic. Blood pressure goes up, heart rate goes up. PO2 goes down, they get hypoxemic.  PCO2 can go up because they're seizing the respiratory muscles and the diaphragm aren't working. They become hypercarbic. If that continues, that can increase into cranial pressure.  Acidosis can make things worse. Adult patients more common than kids. We'll see arrhythmias, renal failure, hyperkalemia in the older patients again. But what happens early on is,  is a massive increase with the sympathetic release, endogenous epinephrine release, norepinephrine, dopamine, vasoconstriction to perfuse the brain and cerebral blood flow increases  magnitudes higher than normal. And as it gets later, that increase in CBF decreases. However, the brain is still in need of that increased flow,  oxygen, glucose, a lot of proteins getting delivered, and you still have this elevated metabolic rate. That can lead to ischemia. That's the operational definition.  Recognize a seizure early on and treating it sort of at a five minute window. Again, for first time exam takers, the primary exam likes to test your knowledge of physiology.  So here's a way of using a respiratory equation to figure out what's going on in a seizure. So again, you're probably familiar with the concepts.  Just be careful that you read the question carefully in the wording, because here you'll just see that an increase in PCO2 by seizures,  there's a decreased respiratory drive because we're giving medications to may stop the respiratory drive. You can have an increased VDDT if there's an aspiration.  You can have an increase in VDDT if there's an aspiration. You can have an increase in VDDT if there's an aspiration. So there's many ways to test the concept.  And again, you're probably familiar with it. Take a deep breath and read through the questions carefully. One of the things that one has to be cautious of as well,  there's a lot going on in patients with status epilepticus is to remember the most common causes fever to treat the patient's fever, to bring the fever down.  Hyperpyrexia is a very common cause of fever. It's a very common cause of fever.  And it can increase the risk of excitotoxic neuronal necrosis, because as more of these inhibitory cells become dysfunctional,  the excitatory may get sort of start firing more ad hoc and may make bad worse. So the neurologic injuries seen following a post-traumatic stress disorder are the most common.  And again, there's a lot going on in patients with status epilepticus is to remember the most common causes fever to treat the patient's fever,  the excitatory may get sort of start firing more ad hoc and may make bad worse. And again, the neurologic injuries seen following a prolonged seizure,  it may result from the seizure itself. There is data suggesting elevated serum, neuron-specific enolase. So that's released from the neurons.  And then if we look at areas of the brain that can injure the hippocampus, cerebellum, Purkinje cells, thalamus, amygdala, cortical neuronal processes,  that's the area when we'd see a seizure, it's usually the gray cells that are firing. And what do they all have in common? They all have excitatory amino acid receptor activity.  So again, the concept of the yin and the yang of the inhibitory cells not working and the excitatory cells becoming recruited, explain it,  explain the gray area and explain the damage.  So this is a famous cartoon from a New England Journal article published about a decade ago. And basically anything that happens bad, whether it's ischemia or hypopyrexia,  it all leads to an increase in intracellular calcium. So any question with damage and are looking for a bad guy, calcium is your probable answer.  So I put this chart together, not for memorization or for the dosage. They don't usually test dosage on exams because we can look them up and recommendations change frequently,  but they'll consider first line drugs are your benzodiazepines. Here you've got lorazepam, midazolam, and diaz.  And then we talk about secondary drugs and recent data suggests that it's safe to use Keppra in pediatric patients. My go-to second drug used to be phosphatidylone,  but the data is pretty good looking, at least in the adult world, comparing benzos with second,  second line drugs, which used to be phenytoin, phenobarbital, valproic acid, et cetera. But just, I wanted to show that Keppra has got rapid onset. By and large,  one of the safest drugs that's been shown in pediatric patients is lorazepam. Here's the dose. The onset's two or three minutes. It has an active metabolite and lasts six hours.  My neurology colleagues hate it. I said, but I usually pick lorazepam because I know by the time I get them, it's going to take several hours for them to reply. So again,  for first-time exam takers, the examiners love to test your concepts physiology. They'll ask you to devise blood flow to an artificial organ.  They'll want to test the fifth concept in a myriad of ways. And one of the things is a perfect anti-epileptic, if you had to design it, you'd want it to be GABAergic.  So it opens the chloride channels, hyperpolarizes the cells. So they stop fighting and firing. The NMDA antagonist, stop the excitatory neurotransmitters.  You'd like it to be fast acting. You'd like it to cross the blood vein barrier. You'd like it to have a short half-life.  You'd like it to be neuroprotective and you'd like it to be a favorable risk profile with low to no side effects.  And it's like spouse or a significant other. The perfect one doesn't exist, but we all seem to find it and make things work.  It looks as if Keppra may be one of these particular drugs, although we're not really sure how it works. We know it binds to a vesicle and we know that it decreases  neurotransmitter release, but we don't know very much more in pediatric patients.  One of the content specs is the recognition of EEGs. So on this particular slide, I'm showing you three. One is a patient that was in for continuous EEG monitoring here with a  basically normal EEG rapidly going into status epilepticus, multiple spikes and waves throughout all the waveforms.  The EEG in the middle is a patient with the phone is ringing and they're starting to sit up and reach for the telephone. So a lot of motor movement.  The third EEG is electrocerebral silence. So you've got a flat EEG, a little bit of activity. So in the old days when we use pentobarcoma or currently we use  midazolam drips, we may titrate them to, to electrocerebral silence, which is what you're seeing on this slide. And again, I put this slide up because I want to show you the  And again, I put this slide in just as a reminder that drugs interact with each other. And there are dozens and dozens of anti-epileptics out there. No one seemed to appear weekly.  So just some common antibiotics that can increase seizures and sort of the way they work. And what they do is they block the chloride channel. If you can't get the chloride in,  you can't make the cell hyperpolarized and more negative. It's going to be more vulnerable or susceptible to fire.  So as intensivists, we're quite attuned to being closely watching our ins and outs. So we have to be careful of electrolyte abnormalities.  The most unforgiving insult to the brain would be hypoglycemia, but we also have to watch for hyponatremia, hypocalcemia, hypercalcemia, hypomagnesemia.  And just remember that anything that makes you alkalotic increases the excitability of the brain and acidosis decreases the excitability of the brain.  And that's the way to think of the various diets that one uses for epilepsy.  The ketogenic diet makes you acidotic. The other thing just of note here is because I don't think it's covered in  any other talk is one should be able to recognize the difference between SIADH and cerebral salt wasting.  So just when you get a chance, take a peek at those. It may be in a different chapter, but I'm not aware.  So I'm a native New Yorker with mandibular tachycardia and looking at the time, my clock here, I have to speed up a little bit to cover all the information I want to get to you.  Stroke as you saw in the preceding slide is defined as a prolonged or permanent dysfunction of brain activity, secondary to an interruption of normal vascular flow.  There are two main classifications. One thinks about stroke, which is an unusual entity to see in pediatric patients.  However, those of us that do critical care do see these not infrequently. So when we think about strokes, we have to think about whether it's an ischemic stroke or a hemorrhagic  stroke.  What this graphic shows is that ischemic and hemorrhagic strokes are 50-50 in pediatric patients. However, in adult patients, it's 80 to about 20%.  Most adult strokes are ischemic, very few are hemorrhagic. And that's important when we consider management of the stroke patient. So again, coming back to my  concept of recognition and pathophysiology recognition, in an ischemic stroke, it's usually  a clot blocks flow of brain to a particular area. Some areas are more vulnerable than others.  As we see kids post bypass, the lenticular striates that go off at a 90 degree angle,  up and into the thalamus and the motor area, we see post-op problems with those particular kids.  In a hemorrhagic stroke, if somebody happens to have a spontaneous bleed inside the CNS, whether it's due to a drug effect and or an abnormal vessel, we'll talk about AVMs a little  bit later, et cetera. So again, to further subclassify strokes, we have a hemorrhagic stroke and you've got an ischemic stroke caused by an emboli. And so we think of ischemic,  and when we think of ischemic strokes, we also have to think about, is it an emboli or is it  a thrombosis? And is it an artery or is it a vein inside the CNS itself? And again, to help with  the stroke classification, think about emia of ischemia, cardiac disease, very common in kids  under two, right to left shunt, cyanotic babies, hematologic, primary vasculitis, drugs, metabolic,  migraines can play a role in ischemia, trauma, and lipid abnormalities. And we think of a hemorrhagic  stroke, the classic we think of as a premature intraventricular hemorrhage in the germinal matrix  of our neonatal patients, hypertension, AVMs, aneurysms, bleeding diathesis, tumors, drugs, ischemic transformation. We have an ischemic stroke and then the patient for some reason,  either not recognized, properly managed, and then we get a transformation into a hemorrhagic stroke, which limits sort of at times what we can do to intervene for these kids.  One of the content specs for the CNS section is to recognize an ischemic CAT scan, so here is one,  and a hemorrhagic CAT scan, differentiate those two, and I just want to use this specific pathology  slide to introduce the concept of watershed infarct, and what happens then is a susceptible  area of brain, which may be supplied by two large major arterial systems, the feeder vessels and the  smaller vessels going out may cause a problem, and that area of injured brain isn't getting the  requisite perfusion that it needs. And in pediatric patients, always think about right  to left shunts, part of the history, maybe the patient was adopted or came from another country,  was brought into the United States, has a history of turning blue, etc., so there can be your hemorrhagic stroke right there leading to an ischemic stroke.  The more and more common entities or patients we'll see are patients with sickle cell disease,  and although many of them are screened to prevent stroke or on hypertransfusion protocols, we still see new diagnoses or missed opportunities, so sickle cell patients,  CNS complications, anywhere from 6 to 34 percent, and it depends on the degree of sickle cell hemoglobin that's present. It can be an intra-arterial embolization, and then they  get intravascular sickling, so the vasovasorum with a blood vessel, a part of the endothelium  that supplies the blood vessel, its blood, is where this will occur. You can see it in patients  with renal disease and hypertension. They can stroke, and you can get basilar artery atasia.  The more common place in the sickle is areas of the brain that are perfused by the anterior  communicating artery and the MCA border zone, and we'll come back to that in the subsequent slide,  and so when a patient does suffer a stroke, the determinants of damage  is the degree of flow decrement, volume of tissue involved, duration of ischemia, and is there cerebral edema there as well. It's also essential for a pediatric intensivist  to recognize the major areas of the brain perfused by the major arteries, so for example, the anterior cerebral artery, middle cerebral division, anterior deep branches, posterior  circulation. You will see this concept tested in one manner or another,  and again, sometimes test-taking skills may play more of a role than the knowledge one has,  so if I'm challenged with a large hemorrhage trying to identify which is the likely artery, since the MCA covers the biggest area, you sort of get more bang for your bucks there,  and just remember that the thalamic arteries take off from the middle cerebral artery.  Sometimes it's better to be lucky than good, so just here to emphasize one of the basic tenants that we always discuss on rounds in brain injured patients is the penumbra,  and penumbra means umbrella, and you've got the area of injury that's one and done. However, our goal with neurocritical care providing ABCs is to prevent the extension of this area  that was already damaged. Thus, that's our ultimate goal in neuro ICU, and again, I just wanted to share and reiterate the concept of bad things happening in the brain. It's usually  because calcium goes in, and this is some of the bad things that happen and how quickly they can  happen. One should be familiar with that. Just a brief mention of arteriovenous malformations.  They account for approximately 30 to 50 percent of pediatric hemorrhagic strokes. Peak symptoms  are in the second and the fourth decade of life. 20 percent of patients become symptomatic before  the age of 15, and they usually present with seizures. They involve the cerebral hemispheres, and usually it's the middle cerebral artery is identified as the culprit.  Many of us are pictorial learners, so I put this in here. So here's the arterialization, and you can see the veins which aren't made for high pressure, and there's no intervening  capillaries, and that's why this develops. The risk of re-bleeding in an AVM is about  three percent per year if nothing is done on the first blush. Here's a representative CAT scan of a  arteriovenous malformation. You can see it's a rather large hemorrhage, and you know from  preceding slides, this area will be supplied by the MCA, and just to show you a pathologic specimen  in the area of location, you can see that there's this massive malformation. It looks like spaghetti  in a pile. Basically, arteries and venules, there's no capillaries, don't properly develop in between to block the very high arterial pressure from the very low venous pressure.  So the venous system gets arterialized. It's not made to deal with the high pressure, and thus at susceptible break points can lead to the hemorrhage. Just to mention,  for completeness, intracranial aneurysms are exceedingly rare in pediatric patients. They account for only one or two percent of all aneurysms found. There's about 500 cases in the  literature to date. Infections, in particular, they're reported post-varicella, account for about  10 percent of these aneurysms, and when they rupture, they cause a subarachnoid hemorrhage, and I refer you back to the anatomy slides discussed much earlier in this talk.  Just to note, intracranial aneurysms usually develop at branch points in the arteries and  the vasculature, which are a bit weaker, thus explaining the susceptibility for these aneurysms  under the high flow to rupture. This is one of my hospital's artist rendition of a sacroaneurysm  rupturing and the blood being released into the subarachnoid space that we discussed earlier,  the happening part of the CNS. And here's a pathologic picture of a ruptured subarachnoid hemorrhage. The dura matter, this is the dura, has been removed, as has the subarachnoid,  the arachnoid membrane. So we're looking at the subarachnoid space where the blood is just  diffusely, and here's the subarachnoid space. And again, CSF flows through there, so sometimes  small subarachnoids that you see on day one may not appear on day two or three because they've  been flushed out by the cerebrospinal fluid. And again, one of the post-op, or the post-presentation  complications of a ruptured or a subarachnoid hemorrhage in general, regardless of the etiology,  is that recurrent vasospasm from active metabolic products released by platelets and every other  sort of cell in the body can cause this intense vasoconstriction to other vessels that perfuse  other areas of the brain, making post-subarachnoid hemorrhage complication more difficult.  So now as we near the end of our talk, here's another question. An 18-year-old male is recovering  from a ruptured cerebral aneurysm that was clipped two days earlier, complains of a severe headache.  Vital signs are stable, and the external ventricular drain put out 20 cc's in the past  four hours. Which of the following is the most likely cause of this headache? Is it hydrocephalus,  hyponatremia, hypernatremia, or vasospasm? So again, the key here is this complication is basically asking what's the most common complication seen two days following this event.  So this is a timeline to demonstrate the correct answer, which is hydrocephalus. The EVD drained  the minimal amount, 20 cc's over four hours. We make approximately 10 to 15 cc's of CSF an hour,  so it should have drained more, and it's not. And the subarachnoid blood can actually cause problems. And this is just showing you that hyponatremia can occur, but it's much later,  day seven. Vasospasm can occur, and most commonly we see it day seven to ten. And re-bleeding is  always a high risk, but not one of the choices. Another common topic that's tested is cerebral  vein thrombosis. Again, I refer you back to one of the early anatomy slides when we sort of  pictorially went through the levels of the brain, and we saw the major, the sagittal sinus, and the  bridging veins going off. So here's just the anatomy. So the next slides I put in here are  basically just for sort of self-study and worth your taking a look at. So here's cerebral vein  thrombosis, and common causes are infections, otitis, mastoiditis, and sinusitis. And then you  can see it, dehydration, especially in younger children. So reading the scenario is important as  well. And again, you may be asked to recognize the entity from a CT. So if you see here, I doubt  there'll be an arrow on your exam CT, but you see there's a very small thrombus located here, and also here you can see the main cerebral sagittal sinus with a large thrombosis.  This is a CAT scan. You never want to be the interesting patient, but here's a CAT scan that demonstrates all of the various types of intracranial hemorrhage that will be covered  more thoroughly in the trauma topic. But what you can see here is that there's a sort of a brain  contusion. Subarachnoid blood sort of fills the sulci and gyri. Remember, wherever CSF can flow,  if there's blood in there, a subarachnoid hemorrhage can spread as well. You can see the patient has a  typical lens-shaped epidural hematoma, and here you'll see a subdural hematoma represented here.  Never good to be the good teaching case. So here's our last sample question, and it's a  rather lengthy vignette. Hopefully you won't have any quite this long on the exam, but in essence,  two-month-old was brought to the ED. Fussiness, increased sleeping, and poor feeding. He was well  till three days ago, was taking less formula, and had to be awakened to feed. Physical exam is  difficult to console. Temperature's 36.8, heart rate 160, respiratory rate 30. Again, a two-month-old.  The anterior fontanelle is full. The pupils are four millimeters and reactive. Of the following,  what is the most likely cause of the CAT scan findings? So the challenge for you is to look  at this CAT scan and then look at the choices, and the choices are in an AVM, encephalitis, non-accidental head injury, or von Willebrand deficiency. And again, here's your classic  subdural hemorrhage. So of these choices, an AVM would be a large bleed somewhere, usually by the  MCA perfusion. You don't see that. Kid doesn't sound like an encephalitic picture. Von Willebrand  deficiency, if he did have a bleed, it would be multiple, multifoci. It wouldn't be local  without a history of trauma. So every pediatric exam loves non-accidental head trauma questions. So you're almost guaranteed to see one in some shape or form.  So if you see the word babysitter on pediatric exam, what else do we think about non-accidental  head injury? Well, you can have papilledema. You can have retinal hemorrhages, and we can also  think about rib fractures, usually posterior, because the spine is used as an axis by the perpetrator. And we can have metaphyseal bone injuries in the event the child is swung or  grabbed by an extremity, and the metaphysis tears, the classic bucket handle fracture.  Pediatric brain tumors always think mechanical, that it's an obstruction, but they also make good test questions for neuro exams and focality and testing the cranial nerves.  So I strongly suggest that you do that. There are some craniopharyngioma, you think DI immediately,  posterior fossa tumors, you think of vomiting and obstruction, and you're not that you're  going to have to know specifics about the tumor, but just the complications that they can cause.  And again, you're going to have to recognize things on MRIs. So here you'll see the obstruction,  massive tumor blocking the third ventricle. You can see the massive hydrocephalus here.  And then the second is, here's a posterior fossa tumor squashing down on the aqueduct of silvias, which is blocking the CSF, making it down the cranial spinal axis and back up here.  And here, I just want to remind folks that the only proven data on the utilization of steroids  in pediatric head-injured patients of some sort is those with peritumoral edema. Everywhere else,  we don't use them. So if you're challenged that way, and you've got a patient with a lot of  peritumoral edema, this is a kid with a cliodel wafer chemotherapeutic agent, and they're causing  a lot of edema, this patient responded to steroids. There's no role for steroids in TBI.  In the event the patient has a shunt, the problem is always the shunt until proven otherwise. It's usually the proximal obstruction here, the part that goes in the ventricle,  because it sits in there. It may get plugged by chloroid plexus, or as my neurosurgeons call it,  brain debris. There's a one-way valve that prevents fluid from going back into the head.  If you look at a CSF shunt, there's actually arrows on it. Next time you get the opportunity, blow one up on their shunt series, and you'll see it. And then it drains distally here.  So again, any kind of obstruction can cause a problem. The most common is proximal. The most common is infection, and it's common bugs. It's usually staph epidemidis,  that sort of contaminates the shunt on insertion. The average age of a shunt, they can last about three to five years. The neurosurgeons I work with call it the  neurosurgical annuity. So I wish you well on the exam, and I hope this brief overview is helpful,  helps you focus to study. But again, failing to motivate the swim team with a pep talk, Coase-Davis uses Plan B. Be sure you do have a plan B, because it's going to help you.  And what I did is I have a series of about 10 more slides or 15 that are for you to look at  at your own time and to review. They're common concepts. Ocular function, they love asking  questions on focality on neuro exams. Also, I will review what the slides are there. Just take  a few minutes on your own and look at them. And good luck, and thank you for listening.
	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.