Podcast 78 – Increased Intra-Cranial Pressure (ICP) and Herniation, aka Brain Code

Today we are going to discuss increased intracranial pressure (ICP) and herniation. This is the first of 13 lectures to go with the upcoming Emergency Neurological Life Support Course that I co-chair.

Screen for Increased Intracranial Pressure

Matt & Mike’s Ultrasound Podcast on Ocular Ultrasound

Tier 0

Head of Bed Up

Temp Normal

PaCO2 35-38 mm Hg

Control Pain/Sedate if Intubated

Tier 1

Osmotic Therapy

Mannitol

Hypertonic Saline

Sodium Bicarb

Tier 2

Propofol (or Phenobarb) Drip titrated to take patient to low levels of sedation scales (5-200 mcg/kg/min)

CPP Optimization

Tier 3

Decompressive Craniectomy

Induced Hypothermia

Pentobarb

Moderate Hyperventialtion (I reserve for patients who are herniating)

Want more Tier 3 therapies?

Get Thomas Scalea’s Lecture on TBI/ICP

  • Already a EMCrit CME Member, go to the CME Learning Site.
  • Want to become a member? Go to the CME Signup Page.
  • Are you in-training in a medical profession, go to the EMCrit In-Training Page.

Now, on to the podcast…

Comments

  1. says

    Dear Scott,

    Thanks for the great podcast.

    Only a few notes-

    I would bring CPP optimisation into tier 0. We all know hypotension and reduced CPP in TBI is bad.
    Normal PaO2 in tier 0. Hypoxia and more recently hyperoxia ( http://archsurg.jamanetwork.com/article.aspx?articleid=1216545) are known to worsen outcomes in TBI.
    While optimising PaCO2 and PaO2 limit Pawp and PEEP with NMBA to limit ICP surges.
    Agree on position being vital. Head up and neutral, but also loosen those ett ties and c-collars.
    I’m with you on avoiding hypotonic solutions in TBI, but I was not aware of any good evidence of CSL/Hartman’s/ringers (osmo~280) being detrimental or 0.9% NaCL being better. Used in many trauma centres and neurosurgical theatres I’ve worked in.
    You got on to ICP monitors and often we would put in Licox catheters which measure PbtO2 as well, and you can consider targetting brain tissue O2 partial pressure >15mmHg. Maybe more of an ICU thing, but if monitor is placed in the ED and you know how to manipulate brain tissue oxygenation, it would certainly come before proceeding to a bifrontal decompressive craniectomy.
    Good sedation is key as you said. Reduces cerebral metabolic O2 demand but also vital for seizure control. Subclinical seizures in TBI can be a real issue for a small proportion. I must admit bolusing sedation and analgesia is one if the first things i do in the raised ICP setting. Targeting burst suppression on a simplified EEG or using BIS has been talked about in the ED and is frequently used in ICUs. Would be very interested to hear your opinion on their future outside of the ICU.

    Kindest regards,

    Peter

  2. says

    Peter,
    Great comments, thanks for writing in.

    CPP in tier 0
    So I think I mentioned CPP maintenance is in tier 0 (generally >55-60 mm Hg) is Tier 0. CPP optimization is actually pushing the CPP beyond these levels, which in some patients will lower ICP, but also has the potential for harm (fluid overload, decreased oxygen delivery from unnecessary increases in afterload, etc.) So to be absolutely clear, CPP<60 should be immediately treated in any patient with potential for increased ICP; push the CPP to see what it does to ICP only with a good amount of knowledge of the trade-offs, an ICP monitor in place, and perhaps, PbrO2 monitors in place.

    Normal PaO2
    Absolutely agree, avoid hypoxemia and hyperoxia at all costs. Not truly an ICP issue, hence why the topic chairs didn’t discuss it. It is mentioned throughout ENLS under the relevant topics. But you can’t make the point too often; it is critical.

    NMBA
    Can’t say I agree with you on this one. We want to be able to examine our neurocrit patients, I don’t want to box their lungs by killing spont breathing, and PEEP when titrated to oxygen requirements has no or negligible effect on ICP. I have a whole lecture on this I gave at Columbia’s Neurocrit conference 1 month ago; happy to send it if you are interested.

    LR
    Given the complete black hole of fluid choice in any condition, you are right that there are no outcome studies on the question. What we know is LR will push the Na to 130; which is not where I want it. Now I must admit I am baffled what leads them to calc LR at 275 with a Na of 130, so I put that question out to the ED tox folks; we’ll see what they come up with.

    PbrO2
    We can’t really say PbrO2 monitors are a treatment for high ICP, they are more of an excuse to ignore ICP elevations.

    Seizures
    Yes, great point about preventing seizures with sedation choice as well. One reason I am not crazy about the new move to dexmedetomidine in this patient group.

    Bolus Sedation
    yes, I must admit to using this myself at times. I am leery of releasing this to the folks out there as a solution lest, we see the big BP swings that come with a less than fully considered big slug of propofol.

    take off CSF
    One other thing I did not put in the cast, b/c it can be dangerous and ED folks should talk to neurosurg to acutally do it rather than doing it themselves: that is taking a few mls of CSF off by lowering the level of the EVD drain for a few seconds.

  3. says

    Great points and thanks for the response.

    CPP individualisation is an interesting topic and really relates to whether cerebrovascular autoregulation is maintained or not. Higher CPP tend to suit those where autoregulation is maintained (which is not as uncommon as people think in TBI, can supply references) as well as maybe hypertensives. Patients with impaired autoregulation may have better outcomes with lower CPPs.With the increasing use of cerebrovascular pressure reactivity indices and TCD, individualised CPP will probably eventually become a standard of care.

    The extremes of PaO2, mainly hypoxia, will increase CBF and hence cerebral blood volume and ICP from my understanding. But also we’re talking extremes of PaO2, <7kPa would be required to significantly alter CBF and ICP.

    The NMBA issue I feel reflects a slight difference in practice between the US and UK. In the UK an accurate pre-intubation neuro assessment along with the CT scan is sufficient most of the time to allow a neurosurgeons/EM/intensivists to decide on appropriate intervention. If we have intubated for neuroprotection reasons it would be unusual for us to then go on and do serial neuro assessments in the acute setting, especially with ICP bolt insitu. After surgery and period of ICP stability on the ICU is a different matter. Given most bolus dose NMBA can be reversed in 30mins I don't see it being a huge problem either way. Spontaneous ventilation is not a significant focus as the major issue in the immediate management is controlling the PaCO2 and avoiding ventilator dyssynchrony and the corresponding spikes in ICP. NMBA also have a role in mild hypothermia associated with shivering and the dramatic increase in CMRO2. Ultimately I'm not suggesting routine deep neuromuscular blockade, however, I haven't come across any compelling evidence/witnessed practice that suggest avoiding them routinely is beneficial.

    Completely agree on the PEEP issue. The PEEP should be titrated according to the patient's associated injuries and oxygenation requirements. No more than is needed, but at the same time I feel the effect of PEEP on the ICP is often exaggerated and not really clinically relevant.

    I would not have any reservations about using CSL/Hartman's over 0.9%NaCL in TBI. I've not read any good literature or heard anything from neurointensivists of late to suggest otherwise. I feel the use of 0.9%NaCL in TBI is more of a historical notion rather then because of a good evidence base or safety profile in TBI.

    While I completely agree PbtO2 management does not directly effect the ICP, it is routinely used in many neuro/trauma centres as part of stratified ICP management algorithms. Higher ICPs (within reason) can be tolerated if an appropriate PbtO2 (>15mmHg) can be targeted and achieved. By tolerating this higher ICP you would avoid committing to the next tier of less well proven therapeutic options -> barb coma, targeted temperature mangement or decompressive craniectomy.

    Agree on bolusing sedation not being a fools game, but a well judged propfol bolus like you said often goes a long way. Midazolam often used instead of or in conjunction with propofol on the NICUs where I’ve worked. Better CVS profile but less effect on CMRO2. Completely agree on the use of an EVD to lower ICP. Like you said though not for the inexperienced and the neurosurgeons tend to have strong opinions on that and will only relinquish control to those they trust. Thanks again Scott and I look forward to your response.

    Kindest regards,
    Peter

    PS there will be an update on ICP management in current opinions in anesthesiology in October which I coauthored and can pass on to you if you’re interested. Will speak to the journal about permission

    • says

      Peter,

      first, I would love to see your article, if you have a unfettered copy, I’d love to place it on the blog for all.

      I agree re: PbrO2; that is exactly what I meant. Higher than acceptable ICPs, but good PbrO2, maybe hold off on the more aggressive therapies. I guess it becomes philosophical whether to include this in a primer on ICP treatment for non-neurointensivists, but it is a great point to make.

      I think the additional monitoring modalities will be nice for CPP optimization, but even now you can plot each sequential MAP increase and the resulting ICP in a few minutes at the bedside if you have push-dose epi and an a-line/ICP monitor.

  4. says

    Peter,

    Had to spend a while thinking this through, but it was v. good b/c I just garnered a whole new vision on Osm. Just as IV bag description of pH is useless, IV bag description of Osm is irrelevant to actual clinical effects.

    LR is 275 in the bag. As soon as you put it into the patient, lactate will be metabolized, most of the K will go intracellular. Net effect of LR will be to drive down sodium, in effect a slightly hypotonic solution.

    This all gelled when you consider D5W; mOSM/L is 250, but we essentially consider it to be free water–why? Because dextrose is glucose and gets taken up by cells leaving essentially free water.

    So question is whether your neurosurgeons/neurointensivists believe in pushing the sodium or not. We shoot for 150 pretty early in the patient’s course. If your folks don’t then LR would make sense; if they do then it would be counter-productive.

  5. says

    Great and thought provoking stuff.

    Certainly there is a difference in solution osmolalities but while you have accounted for the distribution of the osmotically active solutes you must also consider the solvent distribution across ECF/ICF and renal handling of solutes. A greater proportion of solvent stays in the plasma with 0.9%NaCl and along with renal Na&Cl handling means (from what I understand) there is little difference in plasma sodium between the two. http://www.ncbi.nlm.nih.gov/pubmed/12519083/?

    But either way I think my point is not so much CSL is better then 0.9%NaCl but simply it is not so polarised anymore that TBI=0.9%NaCl and CSL is bad.

    Thanks for the discussion.

    Peter

  6. says

    Hi Scott
    Just wanted to get your take on the diagnostic characteristics of ONSD in he ED setting.
    Recent metanalysis by Dubourg: http://www.ncbi.nlm.nih.gov/pubmed/21505900
    Sensitivity 90%, Specificity – 85%. 4 small trials. This is OK, but imperfect when you consider the degree of badness you might be missing in that 10%.

    My take is that we can use ONSD to “rule in” and crank up the interventions / trip to CT. However, it still seems foolhardy to use it for a “rule out” in all but the lowest risk scenarios.
    Maybe serial ONSD exams will be a future substitute for the “neuro Obs” we order on the middle of the road risk patients whom we are wanting to watch until sober, or the sun comes up?

    Love to hear your take on it.

    Also would really love to see a couple of guys (M&M…) do some work on a 1000 healthy volunteers to try and establish a normal curve of ONSD for us all to use. We need to know where the cut-off lays between 5 and 6 mm?

    Casey

    • says

      So here is a post with a dozen articles. But let’s pretend that Dubourg’s tiny sample is correct, with those test characteristics, you have a LR – of 0.12. Not going to find too many tests in medicine better than that.

      Further, I would never use OcUS to rule out an intracranial bleed, you use it r/o the need for immediate ICP treatment. When positive, it is an indication to HAUL ASS to imaging. If it turns out to be a false positive, all you have wasted is some intern stretcher pushing time.

  7. Josh Guttman says

    Hi Scott,

    I am a PGY-4 and when I rotated through our institutions neuroICU, One of the neurointensivists told me that he was unhappy when patients with obviously elevated ICP were intubated in the ED. He felt that we were likely not giving enough sedation/analgesia at the time in our hurry to get the tube in and were likely hurting patients by futzing around with the airway as well as possibly inadequate sedation post-intubation, all causing massive increases in ICP. He demonstrated this by going up to one of our intubated patients with an EVD and shaking the ETT just a little bit and the patient’s ICP shot up from around 15 to 60. His point was he felt that the harm caused by our interventions increasing the ICP were worse than the possibility of aspiration. He had no hard data to prove his point, but argued more on a physiologic basis.

    If you were not going to talk about these issues on a future podcast, I would love to hear your thoughts,

    Josh

    • says

      Doing a clean neuro-stable intubation in a patient with high ICP requires a bunch of knowledge, experience and planning. There is nothing special about it being performed by any individual specialty. Whoever does it needs to be excellent both before, during, and after. His complaint should be that these intubations are not uniformly being performed well. It should not be a complaint that they are being performed in the ED. Until medicine understands that specialty guarantees very little in terms of competence, we will continue to deal with these issues in a way that is not helpful to patient care.

  8. Kirsty Challen says

    Hey Scott

    Writing from the UK where we don’t have a huge tradition of ED crit care and the drug availability is a bit different – what are your thoughts on thiopentone as opposed to propofol for the sedation? I remember it being hypothesised that it provided better suppression of brain metabolic demands but not sure whether there’s anything to support that?

    Kirsty

  9. Dave Barounis says

    Hey Scott,

    Not sure if you are going to address this in later podcasts, but how do you go about managing blood pressure in your patients with severely hypertensive bleeds? We know hematoma expansion is associated with poor outcomes in patients with ICH, and from the INTERACT trial it appears that BP control can reduce hematoma expansion (without evidence of a patient orientated benefit). The feasibility study from the ATACH trial was done and showed we could reduce the BP, but there appeared (granted in a very small pilot study) to be some non-statistically significant safety concerns. Anyways, before the ATATCH II trial data is available what are you doing about BP, drug choice, goal MAP or SBP in patients with ICH.

    Thanks!

    -Dave

  10. Brad Efune says

    What are your thoughts on the hypotensive patient and using propofol as the sedative of choice. I was wondering if it’s still worth using over other sedatives that may not cause as much hypotension when you’re beginning to need pressors to maintain a sufficient MAP to keep the CPP at goal. Basically, does propofol + a pressor have more benefits than other sedatives that may not require a pressor?

    • says

      Propofol is nice b/c if you go too far, you can turn it off. In a hypotensive patient all of the pressors will drop BP from loss of endogenous catechol. So it becomes a style issue. No reason you couldn’t use midazolam for these patients, which is just a bit more hemodynamically stable.

  11. Eric Schwam says

    In regard to your recommendation to do optic nerve u/s to screen for increased ICP, why not just look at the fundus for the presence of venous pulsations? It is quick and easy once learned, and requires no special equipment. If venous pulsations are present, ICP is less than 190 cm H2O and thus is not elevated. Absence of venous pulsations suggests that ICP is higher than that, although 10% of normals may have it. Also helps distinguish papilledema from pseudopapilledema. J Neurol Neurosurg Psychiatry 2003;74:7–9

    • says

      If indeed you find it quick and easy, go for it. Most do not. When I have my panoptic and the patient gaze is midline, I agree not so bad. Trying it with the wall opthalmascopes is a no go. The ultrasound exam is truly quick and easy and has far more literature to validate it.

  12. Mike Jasumback says

    ok, i am clearly way behind but…….

    Where’s ketamine? As I have to often sedate the head pt with borderline BP, might not ketamine be the answer?

    Mike

  13. Patrick Burkhardt says

    Hi everyone-
    as a new listener, I would like to ask the community about another non-surgical intervention in this context: Reversal of anticoagulation in anticoagulation-associated intracranial hemorrhage (aaICH), which is a condition I see quite often. We routinely get these patients´ INR back to normal within the first hour with Prothrombin Complex or FFP.
    In my opinion, the evidence for doing this is rather small, and there seems to be a trend towards the idea that it doesn´t help at all. (Dowlatshahi et al in Stroke some months ago)
    With the advent of the new oral (and unantagonisable) anticoagulants, what would you do with a patient on Dabigatran/Rivaroxaban and ICB? Give PCCs anyway? Dialyse them? Give factor VIIa? Primum non nocere and save the money?
    I´d be happy to get some opinions.
    Thanks, Patrick

    • says

      Patrick,

      Welcome aboard. Can’t say I agree with your contention on reversal of warfarin. I’ve yet to meet a dissenter on the reversal of that drug, despite the absence of a RCT. If you read the discussion of Dowlatshahi’s paper (PMID: 22556194) you will see that his cohort, which didn’t even have very good reversal protocols or success had a 20% lower mortality than non-reversed cohorts.

      As to dabigatran, you may want to see the page on this blog for that one, but the short answer is nobody knows. Rivaroxaban will reverse with PCCs.

  14. Peter says

    Are their specific criteria and preferred pharmacology for managing both hypotension and hypertension in TBI patients?

Trackbacks

  1. […] Podcast 78 – Increased Intra-Cranial Pressure (ICP) and Herniation, aka Brain Code — The first of Scott’s podcasts on topics from the soon to be released ENLS (emergency neurological life support) course. Having worked on the acute non-traumatic weakness topic with my LITFL buddies Oli Flower and Mike Cadogan I know that this ENLS project is awesome. […]

Speak Your Mind (Along with your name, job, and affiliation)