Brain death, and how to diagnose it
Last week we left off acknowledging that, while brain death emerged as a social construct in the wake of advances in critical care and organ transplantation, making the determination of brain death is not so straightforward. In this week’s episode, part 2 of our brain death segment, Drs. Mike Rubenstein and Joshua Levine round off our discussion on brain death by defining it in the present day, and recognizing the limitations of our exam and diagnostic testing.
You would think it would be more straightforward to legally and medically confirm brain death. The 1981 Uniform Determination of Death Act stated that death can be determined using “acceptable medical standards,” but nobody ever said what that was at the time. There was no clear authority or consensus statement that defined brain death. Several papers and arguments had been made in the preceding 20-30 years, as we discussed in last week’s episode, but there was no formal agreement in place. Ultimately, in 1995, the American Academy of Neurology published a consensus definition on the determination of brain death.
Much of the content reviewed in today’s episode comes from the revised version, the 2010 AAN Guidelines, on the determination of brain death in adults. So listening to this episode should be helpful for when you are thinking about whether a patient meets established criteria for brain death. (Note: You should NOT use the information in this episode to determine brain death. Please refer to your local or regional guidelines for that.)
First. You should know the cause of severe brain injury. In the absence of a mechanism of possible brain death, you cannot declare brain death. The patient may have a hidden fentanyl patch, could have severe hepatic encephalopathy, or could have overdosed on phenobarbital–and these may produce a comatose or apparently “brain dead” state. And you might miss this if you rely on the neurologic exam alone. In addition to knowing the cause, the cause must be a known mechanism of irreversible cerebral injury. Just because you know the patient had a subarachnoid hemorrhage and now they are unresponsive, it doesn’t necessarily mean that the hemorrhage was significant enough to cause irreversible dysfunction of the entire brain and brainstem. You’ve got to find out why the SAH caused total CNS dysfunction.
Second. You must exclude known confounders that can mimic brain death:
- Hypothermia (“you must be warm and dead”)
- Extreme electrolyte derangements (especially hypoglycemia, hyponatremia)
- Intoxication (benzodiazepines, opiates, sedatives, paralytic agents)
Third. The physical exam. According to AAN guidelines, only one physician is needed to determine brain death. The exam starts with the assessment of cortical responsiveness. Recognition of pain, or noxious stimulation, is commonly performed in this instance. Nailbed pressure is hardly enough. Unfortunately, to be as assured that the patient is comatose, you will want to inflict the most amount of pain upon the patient in order to determine if the patient is aware of this stimulation. Sternal rub is one maneuver you can perform which inflicts significant discomfort, but this can be limited in patients with severe cervical or lower brainstem injury. I’ve learned to rely on supraorbital pressure, because you don’t need an intact medulla in order to perceive pain in the trigeminal nerve distribution. After you assess for cortical responsiveness, you work your way through the brainstem exam–assessing each cranial nerve you can.
As part of the brainstem assessment, apnea must be confirmed. This is defined as the “absence of a breathing drive.” Typically, the patient is pre-oxygenated for 10+ minutes to a PaO2 of >200mmHg. Then the patient is disconnected from the ventilator (or in some cases, like at my institution, the ventilator is continued but regular, sparse breaths are provided in order to prevent hypoxemia and subsequent cardiac arrest). The diagnosis of apnea is confirmed if (1) there is no respiratory effort, and (2) the PCO2 increases to more than 60mmHg (or for patients with CO2 retention, the PCO2 increases by at least 20mmHg over their baseline). Sometimes patients may actually arrest during the brain death assessment, which can be confusing. The provider is then forced to treat the patient as if he or she is still alive, but without a heartbeat or breathing lungs, and therefore the provider is required to resuscitate this potentially “dead” or “braindead” patient.
Sometimes there are elements of the physical exam, or even behaviors, that may confuse you for conscious behavior or purposeful movements. Often you’ll recognize them as stereotyped movements or actions that are known to be associated with severe brain injury or spinal cord reflexes (things like spontaneous posturing). But some can be more subtle and the family can feel extremely distressed by it, thinking that their loved one is in discomfort or ever so slightly alert. The patient may exhibit ocular microtremor, cyclical dilation and constriction of pupils, slow arm lifting and head turning, spontaneous posturing, small twitches, or even myoclonus following massive anoxic brain injury, and even whole body shivering. But these are not conscious, willful behaviors. They do not indicate the patient has intact cerebral function.
If the apnea test cannot be performed (e.g., the patient is on ECMO, the apnea test is indeterminant, or the patient does not tolerate a trial off ventilation), then ancillary testing may be necessary. The EEG, like the cerebral angiogram, was an ancillary test used to determine whether there was any cortical function. You’ll recall that an isolectric, or unresponsive, EEG was part of Robert Schwab’s original criteria for “death in spite of cardiac action,” and later was part of Guy Alexandre’s criteria for neurologic death, and the Harvard Ad Hoc Committee’s criteria for irreversible coma as well. But experts have argued against using such “confirmatory” testing because they may be falsely negative (meaning the patient has some intact brain function, but clinically they meet brain death criteria) or because they may be confounded by technique or interpretation.
One thing we did not discuss in this week’s episode is the concept of a repeat brain death exam. The current AAN guidelines do not recommend for or against repeat physical assessments to determine death by neurologic criteria. However, many institutions implement policies that require a second examination, often by a second healthcare provider. But then the question arises, how long after the first exam should a second test be conducted? Who should perform the exam? How do you resolve conflicting exam results? Often a repeat exam 6-12 hours later seems to be the case. But each hospital is going to have their own policy. I would recommend you check with your institutional guidelines on how the determination of brain death is made.
Have patients recovered from a “positive” “brain death” assessment? According to the 2010 AAN guidelines, no patient has survived after confirmation of brain death testing. However, according to Dr. Joshua Levine, there have been one or two reports of patients who lived despite the diagnosis of brain death. In fact, if you search the literature, you might come across several cases. But it is unclear to know how well the brain death assessment was, if confounders were excluded, and ancillary testing (if performed) were interpreted correctly. Regardless, these patients who “recovered” after a brain death determination, universally had a poor prognosis.
This week’s episode concludes with some advice from Dr. Mike Rubenstein who discusses his experience in difficult scenarios. This is worth taking a listen.
BrainWaves’ podcasts and online content are intended for medical education purposes only and should not be used for routine clinical decision making. Please consult your institution’s policy and/or ethics counselor on the most appropriate and up-to-date method for determining death by neurologic criteria.
- Wijdicks EF. Brain death. Handbook of clinical neurology. 2013;118:191-203.
- Wijdicks EF, Varelas PN, Gronseth GS, Greer DM and American Academy of N. Evidence-based guideline update: determining brain death in adults: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2010;74:1911-8.