PulmCrit Blogitorial - Why the new brain death guidelines are unacceptable

Hello fellow nerds, let's take a look at the new American Association of Neurology (AAN) brain death guidelines. Instead of a tweetorial, I'll jot my thoughts here, so they're accessible to everyone.

Disclaimer: the following discussion applies only to adult medicine.

Most of these guidelines are consistent with the 2020 World Brain Death Project, so they're no big surprise. I will focus on some newer or controversial points. The bold text is what the guidelines say; the regular text is my prattling on about my opinions.

Recommendation: In patients whose core body temperature has been ≦35.5C, clinicians should wait a minimum of 24 hours after the patient has been rewarmed to >36C before evaluating for brain death (Level B recommendation).

I'm afraid that this is going to cause a lot of confusion.

35.5C is only a very mild degree of hypothermia. If a patient's temperature transiently falls below 35.5C and is immediately increased above 36C, these guidelines seem to suggest that brain death evaluation must be delayed for >24 hours. This is arbitrary and not evidence-based.

Severe accidental hypothermia can obviously affect the neurological examination. Likewise, targeted temperature management to 32C can affect the neurological examination. However, I don't think a transient temperature dip below 35.5C will have a substantive effect on the neurological examination.

Recommendation: Maintain a SBP >100 mm and MAP >75 mm. If an individual has a baseline blood pressure that varies significantly from their age-based normal range, clinicians should target an SBP and MAP appropriate for the known chronic baseline of that individual patient.

This makes sense. Cerebral blood flow is determined as (MAP – ICP). Theoretically, a transient fall in blood pressure could lead to a transient drop in cerebral perfusion.

Specific laboratory targets are recommended to exclude confounding due to metabolic derangements. These include the following:

Ammonia >75 uM.
Blood urea nitrogen >75 mg/dL.
Calcium outside of 7-11 mg/dL, or iCal outside of 1-1.3 mM.
Glucose outside of 70-300 mg/dL.
Magnesium outside of 1.5-4 mg/dL.
Potassium outside of 3-6 mM.
Sodium outside of 130-160 mM.
pH outside of 7.3-7.5
Total T4 outside of 3-30 mg/dL
Free T4 outside of 0.4-5 ng/dL.

Yikes, this is going to cause a lot of problems. For example, ICU patients very often have an ionized calcium level outside of 1-1.3 mM (a super narrow window!). For any folks who think that an ionized calcium of 0.9 mM is a big deal, please educate yourselves about calcium in critical illness here.

I'm a little mystified about why exactly the guidelines felt it was necessary to specify these values. Is there any evidence that the prior guidelines were inadequately specific? Are there case reports of patients with an ionized calcium of 0.9 mM who were incorrectly declared brain dead? I doubt it.

Anyway, if these targets cannot be met then the patient can still be declared brain dead with the use of an ancillary test. So, in practice, this will probably just lead to an increase in ancillary testing among patients with weird thyroid numbers, uremia, or persistent dysboxemia.

Intoxication needs to be excluded through the use of toxicology screening (urine and blood), measurement of an alcohol level (if clinically indicated), and allowing five half-lives to pass for any CNS-depressing medication or intoxicant.

This is, without doubt, the worst and most disappointing part of the guidelines.

Debate about this has been going on for a long time. I discussed the topic in more depth previously, but to summarize here:

Urine toxicology is a horrible test.
The concept of waiting 5 half-lives is fine for a therapeutic medication that we gave to the patient. However, this concept falls on its face when it comes to unknown drug overdoses. In the context of large intoxications, metabolism often slows down (toxicokinetics differ from normal pharmacokinetics). Also some drugs have a ridiculously long half-life (e.g., five half-lives for diazepam metabolites is 18 days!).
The ongoing emergence of more potent drugs of abuse (e.g., carfentanil, fentanyl) will exacerbate this problem in the future.

The American College of Medical Toxicologists created a position statement explaining why this approaches to intoxication in brain death is problematic and requires revision. This is an extremely important statement, so I'm reproducing it in full here:

It's very disappointing that the AAN chose to ignore this position statement.

Recommendation: Clinicians must not use ancillary tests to assist in the diagnosis of brain death in the setting of hypothermia or high levels of sedating medications.

This doesn't really make sense to me.

Hypothermia or sedation may reduce cerebral blood flow a wee bit, but they don't decrease it to zero. Therefore, neither hypothermia nor sedation will interfere with the validity of flow-based ancillary studies (e.g., cerebral scintigraphy).

This is listed as a Level A recommendation… but where is the evidence? In the data supplement, the authors correctly state that planar radionuclide perfusion scintigraphy has 100% specificity. I can't find any evidence cited in the guideline that a flow scan would be invalidated by recent hypothermia or sedation.

EEG is no longer recommended as an ancillary test

This makes sense, because EEG may easily be altered by sedating medications. The use of EEG as an ancillary test was always sketchy.

Bottom line?

There's no room for error in guidelines on brain death. Any mistake threatens to undermine public trust in the brain death declaration process. Anything less than perfection is unacceptable.

Thus, these guidelines are unacceptable. The American College of Medical Toxicology issued a position statement warning against the use of drug screens and clearance kinetics when determining brain death. Nonetheless, the AAN guidelines completely ignored this statement (without even including a single toxicologist in their writing group).

By creating a set of guidelines that is inconsistent with recommendations from toxicology experts, the AAN has created a situation where individual physicians and hospitals will be uncomfortably torn between conflicting guidelines. This may inject further confusion and stress into a process that is already extremely challenging.

Going further:

New American Academy of Neurology guidelines are here.
The American College of Medical Toxicology statement on brain death is here.
IBCC chapter on brain death is here. I've updated it so it's consistent with the latest guidelines (with the exception of the intoxication bit, which is based on the American College of Medical Toxicology statement).

Opening image: Photo by Adrien Converse on Unsplash