SCCM Resource Library-Bulletproof Solutions to Challenges and Controversies in Penetrating Traumatic Brain Injury

Video Transcription

All right, folks, good morning. My name is Chris Kramer. I'm from the University of Chicago. I'm going to be moderating the session this morning  on penetrating traumatic brain injury. We have a pretty full session for you today. So again, that's why we're going to make sure that we start on time here and everything.  So again, we're going to be talking about penetrating traumatic brain injury today. Unfortunately, in the era of gun violence epidemic,  this is something that's unfortunately been seen a lot more commonly in hospitals throughout the nation. And additionally, these are very challenging patients  to treat, and many of them are often young individuals. And we've found that even though a gunshot to the head might seem like it might be an unsurvivable injury,  we've actually had some fairly good outcomes in some patients. And so what we'd like to do is present to you a comprehensive and multidisciplinary panel of experts  here to talk about some of these controversies in these patients. So we're going to start with Dr. Fernando Goldenberg, a good friend of mine from the University of Chicago.  Fernando is an associate professor there and is also the director of the NeuroICU and co-director of our stroke program. So without further ado, Dr. Goldenberg.  Thank you. Thank you, Chris. Thank you to the organizers for the invitation to participate in this symposium.  Can you turn it off for a moment? Yeah, sure.  Okay, I'm going to talk about ballistics and pathophysiology of penetrating traumatic brain injury. Waiting for the slides.  Like in the old times, without slides, drawing something, that'd be fun.  So how, there you go, okay. How many of you take care of patients who suffer a gunshot wound to the head? Oh, pretty much, more than half. Excellent. Wonderful.  And how many of you are neurointensivists? Great. Nurses?  Intensivists? All the rest? Okay. Thank you. Okay. So that's the title of my lecture. And I don't have any disclosures. And what I'm going to talk about today is a little bit  about some global data about mortality. Spend a little bit of time on ballistics, mechanisms of brain injury. I'm going to spend a couple of minutes on vascular injury  in these kind of patients. And I'll present a couple of cases at the end. You all know, and probably some other lectures are going  to this subject again, but there's a crisis in relationship to the firearm relay mortality. And as you can see there, there has been in the last few years a significant increase  in the deaths related to this problem. Probably around 40,000 to 45,000 deaths per year in the last few years happened in the United States  as a consequence of a firearm related event. And there has been an association between the caliber of the bullet and the likelihood of dying.  This is a very interesting study comparing probably the three more common calibers used in the civilian penetrating brain injury. Caliber 22, 9 millimeters, and 45.  And the larger the caliber, the higher the likelihood of patients dying. If you compare a small caliber with medium caliber,  the odds ratio of dying because of a similar injury is more than twice, and more than four and a half times if you compare a small caliber with a 45 millimeters.  And there's some evidence that there's a trend for the criminals in the last few years to use fancier arms, and that includes larger caliber.  So death, one of the mechanisms that account for increased number of deaths in the last few years is also a change in the guns that are being used by criminals.  So penetrating, these are, you know, broad definitions. TB as a consequence of physical insult that exceeds the capacity of the biological organism to tolerate it.  That physical insult initiates a cascade of biochemical and pathological consequences. And penetrating head injury, different from closed TBI,  typically associated with two type of projectiles. The ones that move at high velocities, such as rifle bullets, and the ones that move at low velocities, defined as less  than 120 meters per second, like typical of knives. I'm not going to talk about knives. I'm going to talk about gunshot wound to the head.  So I'm going to spend a couple of minutes about ballistics, because I find that it's quite interesting. And it's a field of mechanics that deals  with the launching, the flight behavior, and the effects of the projectiles. And there are three main parts. The internal ballistics are the study of the flight  of the projectile inside of a firearm. The external ballistics has to be behavior of the bullets from their departure of the firearm until they streak their target.  I'm not going to talk about that. And finally, the terminal ballistics or wound ballistics is the effect of the projectile on the target. I'm going to spend some time on that, too.  So the internal ballistics, this is the internal working of a firearm, as you know. There's an initial event that is pulling the trigger. Then there's a primer that ignites  and generates a very high heat, 2,000 degrees centigrade, inside of the cartridge. That powder in the cartridge burns and produces a hot gas,  which causes pressure to build in the cartridge. And at a certain point, the pressure in the cartridge exceeds the friction that holds the bullet  in the place and begins to accelerate down the bore. And finally, the bullet exits the firearm. It is interesting that something called rifling, that is a phenomenon responsible  for the spin of a bullet, you can see there. It's a bullet that, and that spinning of the bullet that happens on the stars inside of the firearm is one  that gives stability and contributes to the target accuracy. What happens when the bullet hits the target? So the primary way a bullet causes damage is  through the permanent cavity it leaves, right? The hole that is created as the bullet passes through the skin, the bone, and the brain in these circumstances.  This wound channel is the same diameter, approximately, as the bullet or the bullet fragment. And it's a function of the bullet penetration and expansion.  But there's a secondary way that sometimes is way more important that the bullet cause damage that is a temporary cavity.  And what happens is that when the bullet hits a soft tissue, that tissue acts more like a fluid than a solid as it gives way and tries to absorb the bullet's energy.  The bullet doesn't immediately penetrate the tissue. The tissue tries to absorb that energy, creating an impact crater that stretches in and the bullet penetrates.  As the bullet continues its path, it pushes in a violent way the tissue ahead of it, both directly and indirectly, in such a way that the tissue is stretched beyond its elasticity  and therefore is cut and torn, creating a significant amount of damage, and we'll see it in a minute. After the cavity expands to a maximum size,  it starts to collapse under negative pressure and can eventually suck in some debris. In relatively inelastic tissue, like the brain, that results  in a track of injury that often is 10 to 20 times the diameter of the original bullet. So just look at this video. This is a bullet hitting a gel structure similar to the brain.  And as you see, there's a temporary cavity that is being created, and you'll see what happens. Then that temporary cavity starts closing  down under negative pressure and eventually will close down. But those tissues are being destroyed, and therefore their final resultant damage is not the  diameter of the bullet, but it's 10 to 20 times the diameter of the bullet. So the ability to penetrate intracranial space is determined by the energy and the shape of the object,  the angle of approach, and the characteristics of the intervening tissues. The primary injury to the brain is related to the ballistic properties of the projectile  and any secondary projectile, such as debris, bone, metallic fragments. And the size of the temporary cavity is related to velocity, mass, and the shape of the projectile.  One important thing is that the lower velocity projectiles, like an arrow or a knife, do not cause, generally,  they don't cause the secondary damage and temporary cavity. However, high-velocity projectiles do so. And the higher the velocity, like weapons used during war,  create even more damage, right? The higher the velocity, the larger the temporary cavity created by them. The shape of projectile also affects the size  of the temporary cavity through its influence on the yaw. And the yaw is the rotation of the projectile along its long axis. Those projectiles, as you will see in a slide now,  rotate 90 degree in relation to its long axis. And therefore, the path of destruction becomes larger and larger. This also explains why the exit wound is generally larger  than the entrance wound in patients who have perforating injury. So this is just an example of the yawing of the projectile.  And so you will understand how the bullet starts spinning. And as it exits the firearm and hits the tissue, eventually it starts yawing along the long axis.  And it accelerates, and you will see within a couple of seconds how now the cavity that is created by the bullet becomes larger and larger.  So the final result of the direct damage is the addition of the permanent cavity and the temporary cavity as explained before. These are some concepts.  Shotguns simultaneously deliver multiple pellets with a large collective mass. In close range, those pellets all remain compact arrangement. At greater distance, they disperse.  You all know that. And so that's the part in relationship to ballistics that is really interesting and justifies the amount of brain damage.  Anyway, when we assess a patient, there are planes of the lesion that we try to determine. And these are the four planes. I'm not going to spend too much time.  There's something called lower plane, something with crossing the C plane. And those patients usually have a bi-hemispheric injury, crossing the X plane or a combination  of crossing the C and an X plane. And another thing that you will see in the literature and you all know is the definition of wound types.  And they define as perforating when there's an entry and exit wound. It completely perforates the skull. Penetrating is there's entry wound, but there's no exit wound.  Ricochet is a penetrating bullet that may bounce in the internal part of the bone and may bounce more than once eventually. And careening is penetrating skulls  and traveling along the periphery of the cortex without entering the brain parenchyma. And tangential is the one that strikes without penetrating the brain.  So a couple of interesting, oh, thank you. A couple of interesting things. We opened a trauma, a level one trauma center. They will be showing some more data in a few minutes.  And in the first year, we admitted 55 patients who had severe penetrating TBI that made it to the neuro ICU. These were the most common cause of assault, 85% or 86% of the patients.  The median GCS was five. And, you know, almost half of them have absent pupillary response bilaterally. But interestingly, the inpatient mortality was about 50%.  And another 50% of the patients survived. Forty percent had a good short-term outcome. We defined that as patients going back home or patients going to an acute rehab institution.  We don't have long-term follow-up. Dr. Kumar, I know, is going to talk about many of these things in a couple of minutes. So I'm not going to spend time on this.  And specific resuscitation and the initial management of these patients. However, one interesting finding in our population, at least, is that out of those, sorry, 55 patients,  42 were tested for tox screen. And 57% of those patients tested positive for some CNS depressing drug. The median GCS of those patients was three.  So keep in mind that the GCS may be a confounding determination initially, especially in patients under some drug intoxication.  Sorry about that. One of the things that are really interesting is that the incidence of intracranial vascular injury in our own experience.  In the last year and a half, we collected 99 patients with severe penetrating TBI. You can see in that diagram. But at the end of the day,  45% of the patients had dedicated craniovascular image, mostly CTA and CT venogram. And the characteristics of arterial injuries are here.  So we looked at 35 patients out of 45, meaning 77% of the patients had some degree of intracranial vascular injury, either arterial or venous.  These are the arteries that were mostly affected. And these are the pathophysiology of those lesions, the development of pseudoaneurysm, occlusion of the artery, transection,  intimal injury, or full dissection. And these are the characteristics of the venous sinus injuries with occlusion, impingement, transection, or some indeterminate  when we weren't able to classify. But these are the sinus that were explored. So the thing to remember, 77% of the patients have some degree of intracranial vascular injury.  Now I'm going to present a couple of cases just for illustration purposes. This is a patient who has an entrance here on the right side, has an exit wound on the left side.  And as you can see, this patient had a perforating kind of injury. It's bihemispheric. It is transventricular. You'll see later those patients in general had bad prognosis  and has an occlusion, a traumatic occlusion of the anteocerebral artery that is clearly manifested here by a distal ACA on the left, a cerebral infarction.  This other patient is a patient who had a perforating also brain injury. It's also bihemispheric. As you can see, this is the entrance. This is the exit.  It is transventricular, as you can see. The ventricle has been involved but also has a brainstem involvement that are in the midbrain. And in general, these patients are the ones  who had a poor prognosis. This one is different. This is a patient who had also a penetrating brain injury, didn't have an exit wound. It's bihemispheric.  But interestingly, despite all the massive destruction in the anterior superior frontal lobes bilateral and the skull destruction, this is pre-anterior  and superior in the frontal lobe. So the final outcome was almost back to normal. So in summary, just to be OK with the time, this is a highly fatal disease.  However, the likelihood of obtaining a decent or sometimes very good outcome is possible. Keep in mind the initial GCS may be confounded by the drugs  that many of these patients test positive admission. Keep in mind the high likelihood of intracranial vascular injury and of course much more research is needed in the future.  Thank you. So now, I'm sorry folks, we're going to save all the questions till the end just to get in the interest of time.  So I'd like to continue the discussion with Dr. Mona Kumar. Dr. Kumar is the Associate Professor of Neurology and also the Director of the NeuroICU  at the University of Pennsylvania. Dr. Kumar.  Good morning.  So as Chris mentioned, I'm a neurointensivist. I work at the University of Pennsylvania. I'll be speaking to you today about managing the shattered skull complex neurosurgical  and neurocritical care problems in penetrating traumatic brain injury. I have no real relevant disclosures. So we're going to spend the next 15 minutes talking largely  about the management of penetrating TBI. And we're going to discuss largely the evidence or the guidelines around penetrating TBI. And spoiler alert, not much has changed  since the published guidelines in 2001. But don't leave, please stay, stay. There's some interesting and relevant literature actually  published by a lot of the people up on this dais largely around ICP monitoring, antibiosis, and surgical intervention. And so we can talk about that over the next few minutes.  So TBI is extraordinarily common. And we know that. 2.5 million ED visits. And that's probably still an underestimate. Penetrating TBI accounts for about 1 to 2% of that.  The problem is it is the most lethal form of TBI. And so even though it's only 1 to 2% of patients, it accounts for over 40% of total TBI deaths.  It has a very high mortality, as Dr. Goldenberg mentioned. Mortality is 90%. And most of that happens within the first 72 hours. It's a disease of young men predominantly.  But its lethality increases with age. And so as I mentioned, we're going to be talking about the guidelines for management of penetrating TBI that were originally published in 2001.  And that represented collaboration between the International and U.S. Brain Injury Associations, as well as the American Academy of Neurologic Surgeons  and the Congress of Neurologic Surgeons. But unfortunately, there's not a lot that has changed over the last 20 years. And so we're going to look to the military  to really provide information about this, acknowledging that military TBI is quite different than civilian TBI. And we can talk more about that.  Interestingly, more patients have died in civilian gunshot wounds in the United States than died in the Korean and Vietnam Wars. And yet a lot of the way we manage patients is based  on military understanding. And so you can see kind of therein lies the rub. However, we've been really fortunate that the Trauma Quality Improvement Program has come together.  And there's a lot of information and repositories from these level one and level two trauma centers that can inform our practice.  So in a nutshell, the 2001 Brain Trauma Foundation or the combined guidelines from the AA and SCNS, is that there were no high quality data to support strong recommendations.  But options included extensive surgical deprediment for wounds, evacuation of intracranial hematomas. They actually deferred ICP management to the first  and second iteration of the brain trauma guidelines for severe TBI. And the two things that were a little bit more prescriptive about were antibiotic and seizure prophylaxis  where they said you should be pretty aggressive with that. And so I think it's helpful to kind of put all of this together in the context of a case.  So this was a patient that I took care of a little while ago. 33-year-old man found shot in West Philadelphia where our level one trauma center is.  He had a King Airway in the field. On admission, his blood pressure was 90 over 60. He was vomiting on arrival to the ED. He had evidence of urinary and fecal incontinence.  And all he had on his trauma survey were two scalp blacks. One looked like a GSW entry wound. His first GCS was a 3T. And importantly, and we'll talk about this in a little bit,  his post-resuscitation GCS was still a 3T. And you can see one cut of the axial head CT here. And the salient parts of this axial head CT are that the injury was bihemispheric.  It was transventricular. And you can see the interventricular hemorrhage there in the center. And multilobar. And so we're going to look at those particular characteristics  and walk that through. But before we can talk about how we approach this patient, we have to really understand survivability of GSWs. And so every day in the United States,  over 300 patients or people are shot. And 100 of them die from their wounds. But two out of three patients actually survive. And so what does that patient population look like?  Of the 100 people who are shot, 61 die from suicide. And unfortunately, the survivability from a suicide attempt is much, much lower. So about one in six survive.  And so you can see here from the CDC, over the last 20 years, there's been a significant increase in the overall number of GSWs and deaths.  But TQIP, so there was a study looking at about 25,000 patients out of the Trauma Quality Improvement Program. And they corroborated this evidence from the CDC  that the incidence of GSWs is increasing. And so in their kind of small subset, which only represents a certain number of trauma programs in the country,  the incidence of penetrating trauma went up. So the blue line you can see from 2010 to 2014 nearly doubles the number of GSWs. But interestingly, the red line shows  that the mortality peaked in 2011, but then it's coming down. So it peaked at 48% and came down to 40%. Now why would that be? It's possible that that's related  to neurosurgical intervention. Because over that same time period, surgical interventions went up as mortality came down. So might that mean that the things that we're doing  in our ICUs are actually making a difference? And so in terms of neurosurgical intervention, which I'll defer to Dr. Horowitz about specifically talking about procedures,  it seems that surgical intervention is associated with lower mortality. Now you could say that it's kind of cherry picking the patients. Those patients we think are survivable  or can survive their injury go for surgery. And those who don't, don't. So I think there is a risk of bias in a lot of this data. However, even if you stratify by GCS, you identify  that patients who do better with surgery among the lowest strata of GCS. So I think that that is suggestive. Whether this impacts quality of life,  there's one Brazilian study that demonstrated that those patients who have favorable outcome had surgery. So 71% of patients who had surgery had a favorable outcome compared to 30%.  Again, the same risk of bias exists with that data. But I do believe that there might be a signal there. And I think the most critical piece of this is  that favorable outcome is possible with surgery in patients with low GCS. All right. So in 2018, there was a review of over 1,700 penetrating TBI patients by a team led by Maragos.  And they came up with a beautiful algorithm looking at predictors of survivability and mortality. And what they basically identified,  there are a couple critical clinical determinants that can help decide what patients are, warrant aggressive management, and which patients don't.  And so the first step in this algorithm is basic ATLS, neuroimaging, and neurosurgical triage. And part of that parcel of early management, I would argue, includes identification  of hypotension and possible treatment. So back in 1992, there was a paper that actually looked at early resuscitation. So admission GCS followed by resuscitation,  followed by a second GCS, to see whether use of that second GCS is more informative than the original. Now that first paper back in 1992 actually concluded  that they didn't think that there was much of a signal there. However, what was interesting was that there was a mortality difference for those patients  who presented with hypotension. More patients died who had shock than those who didn't. And also, when they looked at the GCS again,  the post-resuscitation GCS of 6 to 8, which was a much, much smaller category because of their 150 patients, 123 of them had a GCS of 3T. So the patients in the 6 to 8 and greater  than 9 category were really, really small. So I don't think we can really look at this for a lot of information here, but there was a difference in mortality post-resuscitation.  So patients who were before resuscitation with a GCS of 6 to 8, 70% of them died, whereas 55% died after resuscitation, which suggests that there might be an impact  of resuscitation here. And Deng and colleagues reported also in 2019 that patients who have hypotension have worse outcomes. And there was one interesting paper that looked  at very hyperaggressive fluid resuscitation in patients with penetrating TBI to either optimize clinical outcome and survivability of the injury or to support organ health  for those patients who unfortunately don't survive and go on to become organ donors. All right, so in a nutshell, this is kind of a busy slide,  but basically what this Maragos team did was they looked at mortality risk from these different parameters. And they identified that number one, GCS is going  to be your largest predictor of who does well. And so if you come in and your GCS is 9 to 15, that might warrant aggressive neurosurgical management.  And that could be tier one measures like fluid resuscitation, ICP monitoring, hyperosmolar therapy, and tier two measures, which might include more aggressive surgical intervention.  If your GCS is in that moderate category of 6 to 8, it guides you by using pupillary abnormalities to decide kind of further approach.  So if you have a GCS of 6 to 8 and at least one reactive pupil, aggressive neurosurgical management may be indicated. Conversely, if you have bilateral fixed  and dilated pupils, because the odds ratio of mortality was 67 for patients with fixed and dilated pupils compared to normally reactive pupils,  they say that those patients probably aren't going to do well. If you have patients with a GCS of 3 to 5, they would stratify first by pupillary abnormality  and then by neuroimaging. So if you have reactive pupils, but then you have bihemispheric or transventricular injury, then it may not be warranted to be  as aggressive, although if you had one reactive pupil, it might suggest that neurosurgical management is indicated. So what did we do in my patient's case?  So in my patient's case, the patient had a GCS of 3. The patient had biohemispheric and transventricular injury so that patient would go into that moribund category.  However, we were at a relatively aggressive trauma center and so we decided to place both an EVD because of the amount of intraventricular hemorrhage that the patient had  and a parenchymal monitor with advanced neuromonitoring, including brain oxygen monitoring and cerebral microanalysis to guide therapy.  But what are the data for ICP monitoring in these patients? So surprisingly, there are none. So from 2001, basically the guidelines state that there are insufficient evidence  to support ICP monitoring. However, if you're unable to assess the neurologic exam accurately, you need to evacuate,  you think you need to evacuate a mass lesion, but you're not sure or studies, CT scans suggest high ICP, it may be reasonable to monitor ICP.  They actually defer to the first and second iteration of the severe head injury guidelines. And so if we do the same, what does the Brain Trauma Foundation fourth edition say  about ICP monitoring? They again say that there's insufficient data to make recommendations and that, but that ICP monitoring can be recommended to reduce  in-hospital and two-week mortality and that it may be considered in patients who seem salvageable. I'll point you to the CBIC guidelines. So there's two publications that have come  out with the last six months. It was the Seattle Brain Injury Conference. And there they actually are a little more prescriptive than the Brain Trauma Foundation guidelines  because they recognize that although the data aren't there to make these recommendations, how do kind of experts in the field have a consensus around management?  And so there's an ICP guideline and a more advanced neuromonitoring guideline. If you haven't seen those, they may be helpful.  What's interesting is that the ICP monitoring data were confirmed by the increased utilization of EVD in the SCARUPA paper looking at the cohort of patients from 2010 to 2014.  EVD placement increased 480%. So whether that's related to the improvement in mortality that we saw or that patients are getting more aggressive  treatment I think is a consideration. However, what do you think? What's better, an intraparenchymal monitor or an EVD? This is where it gets complicated.  So there's one Chinese study from 2015 that looked at EVD and intraparenchymal monitors in non-penetrating severe TBI. And what they found was that patients  who had intraparenchymal monitors had more refractory high ICP and lower survival than patients who had an EVD. Conversely, Randy Chestnut looked at the COBRIT database  which was a brain injury treatment trial using sitacholine and they found using randomized control trial data, although a retrospective analysis, that patients  with EVDs actually fared worse. And why might that be? So I don't know, but could it be because of infection? We know that these patients have a high risk of infection.  CNS infection after penetrating trauma has been kind of the focus really of medical therapy. And that's because in the military, these patients have gaping wounds.  They're in the field and there's a long delay to bring them to medical therapy. As Dr. Goldenberg mentioned, in civilian gunshot wounds, because of the, like, lower velocity  of these missiles, it can drag in hair, clothing into the intracranial cavity and serve as an itis for infection.  We know that CSF leaks, either rhinorrhea or otorrhea, are associated with increased infection. So the paradigm has always been to be very aggressive about antibiotic management.  And that's what the initial guidelines said. However, since then, studies have looked at civilian TBI and found an incidence of 25% of infection.  Now that infection rate is still really high. It's not quite as high as the 60% that they quote in the military, but it's still very high.  Interestingly, though, the rate of infection did not seem to matter whether patients got prophylactic antibiotics or not.  As a matter of fact, more patients who received prophylactic antibiotics had infection than did not.  And they identified in this study some risk factors for infection, which let me just cut to the chase. The Cochrane Review recommends against prophylactic antibiotics.  The IDSA, the Infectious Disease Society of America, recommends against prophylactic antibiotics.  Most importantly, Deb Stein, who's up here, she recommends against prophylactic antibiotics. And that's because it turned out when she did, when they presented a paper at East  last year of 17 TBI centers with 763 patients, it turned out that the incidence of CNS infection was actually only around 7%.  And those risk factors that were identified before were not actually there. So I think that we have to rethink the approach to infection.  And so really, if your patient's getting a craniotomy, give prophylactic, you know, peri-op prophylactic antibiotics as you would normally.  If a patient's getting an EVD, go with an NCS, the Neurocritical Care Society guidelines for EVD management, one dose of antibiotics during insertion, preferably use an antibiotic  impregnated catheter if possible. Don't prophylact your patients routinely. And if you have a persistent leak, maybe a more urgent surgery is intended.  Seizure prophylaxis, the skinny is, early seizures are significant. However, late seizures are well-controlled and not very much different from severe TBI.  And having your neurosurgeon kind of muck around in there to get retained fragments out, probably not helpful for seizure control.  Dr. Goldenberg already talked about vascular complications. I'd be remiss also with Deb Stein up here if I didn't talk about coagulopathy, which  is definitely worse in patients with penetrating TBI than with severe TBI. Unfortunately, we don't really have any treatments.  Ophthalmic injury is very important, especially in those patients who have suicide because of the trajectory of the bullet. They often get more ocular injury than not.  This was my patient. So that patient that I presented at the beginning actually regained consciousness on hospital  day 18, started following commands, had bilateral upper extremity motor movements that were purposeful.  But the most debilitating injury that he had at the time of discharge was that his tongue was swollen.  Now, we have a case series of these patients, and this is not because the bullet trajectory was anywhere near the mouth. We have actually no idea.  We think maybe it has something to do with early seizure, although we're not quite sure.  So I think thinking about head injury in terms of all the different contents of things within the head is important.  But most importantly, he left and he came back for his cranioplasty, joking around and talking with his mom.  And so I think that even all of those negative predictors that are out there in the right patient can make a huge difference.  So in summary, not much more is known about penetrating TBI over the last 20 minutes or 20 years. I'm clearly over my time. But the incidence of penetrating TBI continues to rise.  I think the most important thing that we can do is prevent these injuries. But mortality may be decreasing, largely due to interventions from our neurosurgical colleagues  who you'll be hearing from soon. Consider ICP monitoring. And I think watch this space, this intraparenchymal monitoring versus EVD.  I think a much more judicious use of antibiotics is intended, and seizure monitoring tends not to be that much different. Thank you very much.  And now I'd like to introduce, in addition to our program, Dr. Pelleg Horowitz, who's also an assistant professor of neurosurgery at the University of Chicago and was formerly  the interim director of our neurotrauma program. He'll be talking about neurosurgical management of patients with penetrated traumatic brain injury. Thanks, Chris.  So I'm going to skip through most of the background, which there's a fascinating history of neurosurgical trauma, which you can read up on in your own time, and jump to the present  moment where we opened our trauma center, our level one trauma center, on the south side of Chicago nearly two years ago today.  In the first year, we treated 3,000 adult trauma activations. And the interesting aspect of our population is that our penetrating injury rate is much  higher than the national average. We have 40% penetrating injury compared to 10%, 20% in most level one trauma centers, even at major institutions and cities.  What you are not looking at is an aerial nighttime view of Chicago. The yellow dots represent gunshot incidents with an injury, and the red dots are gunshot  incidents with a fatality over the last six years. Focusing in on our local neighborhood, the blue dot is our institution.  Over just 2018, the year where we opened our trauma center, these are the injuries. The gray dots are injuries on other years.  I'm not going to talk about all the other sharp implements that people injure their brains with. We'll focus on the gunshot wounds.  And when we first started seeing these patients come in, we were looking to the literature to guide us on what to do, what patients to try to save.  It's an embarrassment that in comparison to a similar epidemiologic problem of gliobastoma, there's next to no literature on this information. And I'll skip again.  There's really no guidelines to guide what we do in surgery. The next best level of evidence we have is just ask a bunch of neurosurgeons what you  would do, and there's no consensus. We know from a couple of studies in the mid-range GCS population that if you operate, you will save lives.  And even in the lower GCS groups, you can save lives and influence outcomes. So here, Glasgow outcome scale of four is a reasonably good outcome.  Not working, but relatively independent. We've learned at our institution that if you're going to do something, it needs to be quick, it needs to be efficient.  And they have to have specific goals about what you're going to do for the surgery for the patient.  So we try to get our patients into the operating room within 30 minutes of deciding to operate. We've been pretty good about that over the first year.  We try to keep the case times down. Just get in, do damage control, fix what needs to be fixed, and get the patient to your intensivist to help kind of fix the rest of the problems.  And then the goals of surgery. So you've heard about decompressive hemocrinectomy, ICP monitoring. So those are our main tools for treating increased ICP.  And so the ICP monitoring and EVDs, obviously, we do those mostly bedside. We don't do those in the OR.  But if the ICP remains refractory, or if we see something on the scan that's concerning for future swelling, we'll take them for surgery for decompression.  So this is, I'll go through a number of cases. This is a patient who was shot while driving, hit a tree, had a seizure, and was brought in GCS 5T.  He improved to GCS 8T with appropriate management of his seizure and with some mannitol. Because of the swelling, the crowding of the basal cisterns, we decided we're going to  skip the ICP monitor. We'll just take him for a decompression. And we did that. We placed a contralateral ICP monitor. And he did quite well.  He eventually came back for his cranioplasty. By the time he left the initial hospital stay, he was talking, following commands on his right side, left paretic.  And he's interested in going back to work in some limited capacity as a police officer.  This is another patient who came in talking, was shot in the posterior fossa, rapidly declined, became bradycardic, got intubated.  And so the decompression we do for that is a posterior fossa decompression, similar to strokes in that area. And an AVD was placed as well.  This patient was also discharged home and eventually came back with another gunshot wound three months later.  The guidelines have a suggestion that if there is a mass lesion at the discretion of the providers, you could go in to try and remove that. Mass lesion here means blood, usually.  So epidural, subdural, intraparenchymal hematoma, trying to control active bleeding. Rarely, we chase after a large skull fragment that's depressed.  So here, there was a patient who came in with a reasonable GCS, but had a hematoma that was expanding and was taken to the operating room. That was evacuated.  He also had a hemicraniectomy and an EVD placed. And he's living at home. He has some headaches, but he's doing fine. This part of the surgery is a little bit more challenging.  So debridement and wound closure, obviously, by definition, penetrating brain injury, there's an open wound. There's a connection between the brain and the outside world.  There's devitalized tissue because of all that heat and energy that goes through the scalp as the bullet's going in.  Sometimes, it just is as simple as stapling or suturing that small entry wound at the bedside. And that can be done.  Sometimes, it requires more complex wound closure techniques, rotating some tissue underneath, getting plastic surgery involved to cover big defects.  We were seeing so many of these cases that we had to develop an open brain protocol for our ICU nursing team to help with families, cope with that, and help reduce the risk of  infection. And then what I personally find one of the more interesting aspects of this is putting the pieces all back together, the reconstruction.  There are obviously a couple of reasons for this. One is cosmetic. One is to prevent spinal fluid leaks.  This is an example of a 48-woman shot in the head, arrived with a borderline GCS of 6T. Between the CAT scan and the CTA that was done on the same scanner, she had expansion  of the left frontal hematoma, the side that was not where most of the bullet trajectory was. And we just went straight to the operating room.  And you can see on the 3D reconstruction the amount of damage that the skull sustains, the out-fracturing, the changes that that undergoes.  To fix this, we took off all the bone fragments on the way and harvested a pericranial flap for reconstruction, evacuated the hematoma, controlled the bleeding source, put all the  skull back together, put it back on, and placed an EVD monitor just to check pressure post-op. And she's done very well.  This is another example that wasn't managed as well up front. It was a very far forward bifrontal injury.  We felt this patient was going to survive, so we just placed an ICP monitor, treated him medically. Four weeks later, he left for rehab. He was following commands, ambulatory.  But then he came back a month later with a CSF leak, pneumocephalus, and intracranial abscess. So this involves something called cranial nasal separation, getting rid of the frontal  sinus or whatever other sinus is involved and separating it from the brain. You have to patch the dura. You have to bring healthy tissue over that patch.  And you should consider diverting CSF with an EVD or a lumbar drain as things heal. So this is the positioning. It's a large bicoronal incision.  We separate the inner layer of the scalp, the pericranium, because that's a good, healthy vascularized tissue.  And we slip this over the dura and remove the frontal sinus so that that pericranium is sitting there instead. And that effectively isolates the nose from the brain.  And he was walking and talking and came back to clinic with good cosmetic result as well. This was a similar patient, also managed medically.  There was a gunshot from the right medial canthus across the ethmoid sinuses and then out through the back of the left temporal lobe.  Initially managed with an ICP monitor and eventually following commands. But then he developed a CSF leak and a worsening encephalocele down the ethmoid sinuses.  So this is what you're watching is after debriding that brain, we get hemostasis with some materials. And then we bring in an artificial dural patch.  And the really tricky part of that is actually sewing it in so that it's watertight, so that there's not going to be CSF leakage. And that's what that looks like.  And then the pericranium graft, which has been harvested on the way in, is tucked in over that area and then folded back over the remaining dura to prevent a leak.  So that's the result. The yellow arrows show on the sagittal where the pericranial graft is coming in, going all the way to the back of the defect, and coming up the front.  And he is also brushing his teeth with his good hand before going to rehab. You need to be aware that there can be some surprises. We talked about some of the vascular injuries.  I wasn't as attentive to this on this particular case. We saw a big subdural, midline shift, and a temporal hematoma.  This went straight to the operating room for a decompressive hemicraniectomy and anterotemporal lobectomy.  In the process, I identified the defect in the skull base right around the petrous carotid and had a big gush of red blood.  So we took him to IR, embolized the carotid, and took him back for a number of other surgeries. And with his permission, we were showing the video here.  Seven surgeries later, he's doing pretty well. So we've had a lot of good successes with select management of surgical patients with TBI.  We have a pretty high volume of these cases, and we're learning quickly in the absence of evidence.  We can see about somewhere between one and three penetrating cranial injuries a week. And overall, just the intervention needs to be timely, it needs to be planned appropriately  for the problems that you're seeing at surgery. And thank you.  OK, last, but certainly not least, we have Deb Stein, who's a professor and director of the trauma program at UCSF. Thank you.  In the interest of time, I will actually skip through some of my slides just to not be repetitive and to also give us some opportunity to have some time for questions.  I have no relevant disclosures related to this talk. Like everybody else, I'm going to spend my time talking about gunshot wounds, obviously,  perforating injuries from a variety of different sharp objects can occur, thermometer sticking out of the brain was my favorite. I don't have a good picture of it, unfortunately.  Intracranial temperature was fine, by the way. And as people have talked about, we know this is a huge public health problem. Again, I'll skip through this pretty quickly.  But I do want to just highlight what Dr. Kumar mentioned as well, is that the death rate from suicide is higher than the death rate from homicide.  Obviously, it doesn't make homicides not important and not very relevant. But recognizing that when we talk about it as a public health problem, suicidality is  really a larger problem and one that is vastly undertreated. So I just wanted my soapbox about that for a quick moment.  We do know that penetrating TBIs are the most lethal of all firearm injuries. These patients typically will die very quickly, as opposed to the trimodal distribution of  injury death that we're very used to, in which we have a significant percentage of patients that survive their initial injury but then subsequently die of complications.  These patients typically will die early on, either at the scene or within the first hour or two of injury.  And these are mortality rates from a copulation just from a couple of years ago, but some of these studies are pretty old.  And we know that the mortality for our perforating or transhemispheric gunshot wounds is upwards of over 60% in adults and about a third of children.  So children typically do do better. And can we predict who is going to do well and who is going to do poorly? And what does prognosis look like?  And I think we want to spend a moment just thinking about why we prognosticate, why. So one reason is obviously so we can inform patients' families.  And I think that's probably at least as important as any other reason to do this. But the other reason is, and we've heard a lot about resources and whether we want to  expend resources on patients who we think have non-survivable or non-recoverable injuries. And it's kind of an interesting and ethical dilemma in some cases.  But I think thinking about why we prognosticate is actually as important as doing it. So what's the difference between these two patients?  Well, I'll tell you the difference is the guy on the left-hand side died, and the guy on the right-hand side left the hospital on hospital day 22, who now works with the Trauma  Survivors Network and is a violence intervention counselor. These patients presented very similarly. Their injuries look very similar.  And they did have pretty similar physiologic parameters. So what is it about that allows us to prognosticate that the patient on that right-hand side  is going to do well, and the patient on the left-hand side is going to do poorly? And the short answer is there isn't a whole lot.  We have heard Franjo give us a very nice talk on what is the significance of the ballistics, and the ballistics and the wounding weapon are what are going to determine the energy  and this transfer of energy to the brain. We heard about that, and that's certainly going to be highly relevant.  So the factors that are associated with mortality are going to be your velocity, the intracranial  path and location, not surprisingly to anybody, the size and type of the penetrating object, as we heard, and as well, as a trauma surgeon, other associated injuries.  So the injury to the brain is often the most significant, but not always, and something to keep in mind. So can we prognosticate for penetrating TBI?  Many of you in this room are probably familiar with these two databases, the CRASH and the IMPACT databases.  These are very large, well-validated algorithms that are used to predict outcome following brain injury.  Unfortunately, as you're also probably aware, these specifically excluded patients with penetrating brain injury.  So these are great models to use for your blunt brain-injured patient, but not necessarily for your penetrating brain-injured patient.  Older studies, as Dr. Kumar pointed out really nicely, the post-resuscitation GCS is the most reliable and significant clinical predictor.  The other ones that are listed there will surprise nobody in this room, obviously, nonreactive  pupils, hypotension, respiratory arrest, coagulopathy, suicidality, advanced age, and then the trajectory of the wound.  And mortality in these older studies approaches 100% in patients who present with poor neurologic status.  And as we heard, that risk is absolutely potentially modifiable. The question is, who are we going to intervene on and provide surgical or high-quality neurocritical  care interventions on? This is a study that Bijan Arabi, one of my former collaborators and colleagues from the  University of Maryland, published with a 24-month follow-up, which is a pretty good follow-up for any trauma series.  And basically, they looked at the immunovariate analysis predictors of outcome. And they found that, obviously, admission GCS, the missile trajectory, whether or not  the patient had surgery, pupillary response to light, the patency of the basal cisterns,  age, and whether or not the patient had intraventricular BUD were predictive of outcome in univariate analysis.  And when he did his multivariate analysis, you can see, not surprisingly, the GCS, the patency of the basal cisterns were highly significant.  And when you excluded the GCS, because of exactly what we all know, many of these patients come in with their GCS very confounded by intoxicants, is that missile trajectory and  pupillary response to light were highly predictive of outcome. This is data that Susan Muschlegel put together.  That was a compilation of data from the Baltimore Center, as well as her center in Massachusetts.  413 patients, factors that were associated with survival, again, no big surprise, anything on this list.  And it was able to actually establish an ROC value of 0.96 in this patient population for survival following brain injury.  And she coined the term the spin score, which is the higher the score, the better you're going to do.  And this is borne out in an additional multicenter validation study that they did, and that was just published this past year.  And of note is that the admission INR, or the presence or absence of coagulopathy, was highly significant.  And in those patients who did not have their coagulation parameters measured, they were unable to validate the score.  So just something to keep in mind when you see these patients early on. Again, in the interest of time, I'll go through this very quickly.  But we have to think about it's not just about lived or died. Alex Vladko reminds me every time he reviews one of my papers. He says to me, there is a fate worse than death.  And that is absolutely, obviously true for patients who have severe neurologic injury. And so we always want to think about functional outcome.  Again, even the CRASH database, it'll give us a risk of unfavorable functional outcome at six months. That's going to be measured by your Glasgow outcome score.  I don't know about you, but I wouldn't know for my kid, am I OK with a lower moderate disability but not an upper severe disability? What does that actually mean?  What does it mean to patients? What does it mean to their families? I could tell you very little. Some of the scores are a little bit better.  The rating score gets a little bit more granular, but really not much. And I think what we want to be thinking about as we go forward with this, and I would highly  encourage everybody in this room to think about how to look at functional outcome following brain injury or following neurologic injury overall, but particularly in a field that's  so poorly studied as penetrating traumatic brain injury, is what we really want to be looking at is those patient-reported outcomes.  What's important to the patient, as opposed to what do we think is important about their brushing their teeth, or they're playing golf, or what do we think?  This is just one example of some of those outcome measures, the province measures, how they encourage you to go and look at them.  But as we think about outcome, and we think about studying outcome, we really do want to think about it from the patient's perspective, not necessarily from our perspective.  I will mention organ donation very briefly. It is a really highly relevant thing. For those of you who are not from the United States, in the United States you have to specifically  state that you are going to be an organ donor, as opposed to in many countries in Europe where you are an organ donor unless you specifically decline.  This has significant implications when these patients come in. They require a lot of resources.  If a patient is designated, or if you feel that the patient may potentially be a candidate  for organ donation, we typically will do everything we can to preserve somatic life in that situation  because we know that brain injury, and specifically penetrating brain injury, is a significant source of organs for our patients who need them.  Lastly, I want to talk for a minute about decision making. We have these models. Okay, given this is blunt trauma, not penetrating, but it spits out a number of 83% poor functional  outcome risk. Well, which is my kid? Is my kid that 83% or is my kid the 17%? And I think that leads us to when we talk to families and we talk to people about prognosis  is that we want to be very careful of the language that we use because we don't ever know for sure. This is a really nice piece of work.  Again, this is done in blunt trauma, but it looks at the concept of threshold values rather than absolutes or percentages.  So for example, longer duration of coma is associated with worst outcome. The threshold value that might be associated with that would be severe disability is unlikely  when a coma lasts less than two weeks, as opposed to a good recovery is unlikely when a coma lasts longer than four weeks.  And just think about it in the way that we communicate with our patients' families. I use the expression best medical guess because we don't know. We don't ever know.  I think more likely and less likely is a really helpful phrase. I would be surprised if he or she wakes up and does well. I think that's also a potentially useful phrase.  And then my favorite one of all is when the family turns to me and says, what would you do if it was your family member? And I say, it's not about what I would do.  It's about what he or she would want for himself. And I think that's a really important thing to keep in mind.  This is about surrogate decision making is about what the patient would want, but recognizing that there is intrinsic conflict, because what mom wants is not necessarily what the  patient would want. And we have to help walk our families through that. I will close with that to allow for some time for questions. Thank you all very much.  Thank you to everyone for listening, and a big thank you to the speakers. We do have some time for a few questions.  There's a session at 11, but you can go ahead and probably stay for about 10, 50 or so. So please go ahead and step up. Just wanted to get the panel's comment on something. I agree.  I would say that if the patient is able to answer outcomes and expectations, quite frankly, they're doing pretty well. But in terms of outcomes, again, we focus a lot on mortality.  And that's clearly not the best outcome. Glasgow outcome score of 3, probably not the best outcome. But should we demand in studies going forward that they use reasonable outcomes?  I mean, because again, as you said, there's a fate worse than death sometimes, probably. Yes. That was simple. And I think the NIH is finally getting that.  I think actually Jeff Manley has done a lot of work to this end about the Glasgow outcome score and how it actually is a very poor predictor of how the patient is actually doing.  It's a very interesting dichotomy. How do you get patient-reported outcomes from patients who can't report? And so there is actually a field, a whole field of science around that.  But the short answer is yes. The NIH, their insistence on GOSC as an outcome measure for neurologic injury is not patient-centered,  first of all, and second of all, is not going to help us, lead us to better outcomes, in my opinion. I completely agree.  I think we struggle more with who we should take to the operating room than what we should do there. And it's something we're always considering and thinking about in that position.  If we had that information, that data, we would certainly use it. Glasgow outcome scale is not the best measure, but unfortunately in our population it's hard  to even get that follow-up. Those people get lost very easily. Awesome talk. I just had a question.  At your institutions, I do neurocritical care, and we also hopefully select patients that would benefit.  But I notice there's a lot of variability with surgical and medical collaboration. And I know it's hard, you know, depending on what neurosurgeon's on call or surgeon,  is at your institutions, is there a collective agreement that, like for example, in our stroke  patients, most surgeons will take, if we ask, with a large sort of evolving stroke, if everything kind of lines up, they'll go.  But in a case where you're asking your surgical colleague, say, if they personally feel this patient's not going to do well, I know it's hard to protocolize it.  But what do you guys do? Like you said, it has to be fast. You can't wait days. Yeah, I agree. And I think probably my intensivist could speak better to our group in general.  I will say that there is quite a bit of variability, even within our own group, of what we will do and who we would take. For me personally, it's a real-time discussion.  I just call up the intensivist as soon as I hear about the patient, and we agree on what to do. So there's not much discrepancy, I think, for me, but perhaps with my colleagues you  could speak to. I agree. I don't have much to add, but I think that one of the beauties is that our trauma center  is a new one and was created from scratch with multiple people coming from different backgrounds and training.  And that, interestingly, allowed to have a very nice collaborative effort where neurosurgery and neurocritical care are called right away in the trauma bay, right after the patient  arrives. And many times, we are all together making a decision collectively within the first few minutes. At the bedside.  I think I can speak to this also from our perspective at Penn. I'll tell you, when we had a level one trauma center and we had one kind of large group  of neurosurgeons at our main campus hospital, it was harder to kind of make sure that that kind of practice variation across surgeons and intensivists was maintained.  And when we moved our level one trauma center to Penn Presbyterian, what that really allowed for was a streamlining of kind of neurosurgical and neurocritical care approach.  And so the surgeons who did the lion's share of trauma call became from a faculty of 20 down to a faculty of five. And that really helped standardize practice.  And I think that iterative review of your trauma cases, if you kind of treat it like you would stroke or other subpopulations where you say, you know, we had this many cerebellar  hemorrhages and this many went to the OR and they actually did better. I think if you show surgeons in your own hands that patients are doing better when they intervene,  they are absolutely willing to do that more often. And I think that at least that helped us really standardize our protocols and practice and kind of move us in a better direction.  I'll say just one thing about that. There isn't a lot of data to guide what we do, but there's a lot of data that says if  we do it in a guideline-based or protocolized fashion, patients do better. So I think it's less important what we're actually doing as long as we're doing it consistently.  And so even if it's just expert opinion, consensus-based guideline development, minimizing variability  is really going to be the thing that's ultimately going to probably the short-term result in better outcomes.  And one last thing, just to add to that very quickly, you know, we've also established a monthly multidisciplinary neurotrauma conference where we get together with the  trauma surgeons, the neurosurgeons, and then us neurointensivists. That's why it's nice if we can keep learning together as the process goes along and try  to unify our approaches for patients. So unfortunately, I think that we're out of time. But feel free if you have any individual questions to come up here and we'd be happy to talk  to you. Thank you.

Video Summary

The speakers discussed the topic of penetrating traumatic brain injury (TBI) and the challenges in managing these cases. They highlighted that TBI from gun violence is becoming more common and is particularly prevalent among young individuals. Despite the severity of gunshot wounds to the head, some patients have shown good outcomes. The speakers presented a comprehensive panel of experts to discuss controversies and challenges in the treatment of these patients.<br /><br />The ballistics and pathophysiology of penetrating TBI were discussed. It was emphasized that the caliber and velocity of the bullet determine the extent of damage. The temporary cavity created by the bullet can cause significant destruction of the tissue, resulting in a track of injury that is much larger than the bullet itself. The speakers also highlighted the high incidence of intracranial vascular injury in penetrating TBI cases.<br /><br />Surgical intervention was discussed as an option in managing penetrating TBI. The speakers emphasized the importance of timely and efficient surgery to achieve good outcomes. Factors such as the trajectory of the bullet, pupillary response, and the presence of mass lesions were identified as predictive factors of outcomes. The speakers also discussed the challenges in prognosticating for penetrating TBI and the need to consider patient-reported outcomes in assessing functional outcomes.<br /><br />In conclusion, the speakers highlighted the complexity and challenges in managing penetrating TBI cases. They emphasized the need for multidisciplinary collaboration and the importance of individualized approaches in these cases. Further research and the development of guidelines are needed to improve outcomes for patients with penetrating TBI.

Asset Subtitle

Trauma, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Explain the mechanisms by which missiles cause primary and secondary brain injury and formulate an individualized and prioritized approach to the initial management and stabilization of patients with penetrating traumatic brain injury
Describe the complex peri- and postoperative issues encountered in managing patients with penetrating traumatic brain injury
Describe the limitations in the literature regarding the management of these issues
Choose the most appropriate treatment strategy within this context
Illustrate the potential pitfalls in prognosis

Bullet to the Brain: Ballistics, Pathophysiology, and Initial Management of Penetrating Traumatic Brain Injury

Fernando D. Goldenberg

Managing the Shattered Skull: Complex Neurosurgical Problems in Penetrating Traumatic Brain Injury

Monisha A. Kumar

Neurosurgical Management in Penetrating Brain Injury

Peleg M. Horowitz

Prognosis in Penetrating Traumatic Brain Injury: Is There a Magic Bullet to See Into the Crystal Ball?

Deborah Stein

Meta Tag

Content Type Presentation

Knowledge Area Trauma

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Traumatic Brain Injury TBI

Tag Neurotrauma

Year 2020

Keywords

penetrating traumatic brain injury

TBI

gun violence

good outcomes

ballistics

intracranial vascular injury

surgical intervention

predictive factors

patient-reported outcomes

multidisciplinary collaboration

	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.