Death in 2025: New Definitions, Ethical Questions & Medical Guidelines

“death is not the opposite of life, but a part of it”

Haruki Murakami 

What is death? If you were to ask most members of the public or indeed any member of the healthcare professional team they would most likely respond along the lines of “death is when a person stops breathing and their heart stops beating”. This is something we all learn or sadly experience throughout our lives – when a loved one or an adored childhood pet dies- their heart stops. 

But what actually is death? Is it more than cessation of cardiorespiratory physiology? The Academy of Medical Royal Colleges (AOMRC) in the UK and the international community think so, with a formal definition of death by neurological criteria having been medically accepted since 1979.¹ This is a trickier definition, as it is based on cessation of brainstem function, which can occur in a person who is still warm, well perfused and has a beating heart. The acceptance of death in this situation is often very hard for everyone. Even having worked within the Emergency Department and Intensive Care setting for many years, we still find discomfort standing at the end of a patient’s bed – their heartbeat as regular as a metronome, oxygen saturations reading 100%, the ventilator giving even breaths and yet knowing that we consider them to be dead. If we battle with this feeling, we can only begin to imagine what that must feel like for the family and loved ones sat at the patient’s bed side.

So, this is a blog post about the grim topic of death. It aims to clearly define it, categorise they ways in which it is diagnosed and then highlight recent changes to the diagnosis of death by neurological criteria (DNC), which came into formal existence on January  1^st 2025. This update raises interesting questions, some ethical and some practical. And interesting questions are always worthy of consideration and discussion. 

What is death?

The AOMRC define death as “permanent loss of the capacity for consciousness combined with the permanent loss of the capacity to breathe”. There are 3 processes by which this can come about which allow for the clinician to reach a diagnosis.

Somatic death: although rarely seen in the in-hospital setting this is an important diagnosis. It allows professionals from various emergency services such as midwives, paramedics, and police officers for example to avoid attempting what would be traumatic resuscitation in situations which are futile. The diagnosis is based either on traumatic criteria which include the presence of overwhelming physical trauma or time based with physical evidence that the individual died hours or days prior.

Circulatory criteria: This is the criteria by which clinicians and allied health professionals are most familiar. This diagnosis is made by a trained medical practitioner observing and assessing the patient for a minimum of 5 minutes. During this time the patient has continuous unconsciousness and continuous absence of both breathing and circulation.

Neurological criteria: Diagnosis of death in this way is reserved almost exclusively for patients in the Intensive Care setting, when irreversible cessation of brain function can be confirmed following devastating brain injury (DBI). There are some Intensive care units which will encounter this rarely, but in tertiary neurological intensive care units (such as the one in which we currently work) it may be part of day-to-day practice. 

Death by Neurological Criteria (DNC): a brief history

The first UK guidance on diagnosis of “Brain death” in the UK that we can find reference to is an article in the BMJ from November 1976 where the Medical Royal Colleges statement is published in first iteration. The essential criteria for diagnosis of death based on neurological criteria has actually changed very little since then.

Firstly, the patient must meet the precondition of DBI, defined as known aetiology or mechanism, GCS 3, no observable brainstem reflexes and dependent on mechanical ventilation with apnoea. Initial considerations then include assurance on possible reversible pathology (such as endocrine or electrolyte abnormalities); the recognition that the effects of drugs such as narcotics and neuromuscular blockers should not be influencing the clinical picture; that the patient should not be hypothermic; and that any testing should not be performed until >6 hours following the loss of the last observed brainstem reflex. There is clear subsequent guidance on how to then test individual brainstem reflexes and in particular perform an apnoea test, to confirm the absence of brain stem function. It was accepted from the beginning that tests should be performed more than once, by 2 different clinicians each time and involve at least one medical consultant, to mitigate error and prevent interobserver variability. 

As you might expect, since 1976 the criteria by which neurological death is diagnosed has become more formalised. The AOMRC produced a formal consensus document in 2008 which quickly became the current UK standard. This provided guidance on making the diagnosis of DNC in patients 2 months and older (with separate criteria for neonates, produced in conjunction with the Royal College of Paediatrics and Child Health). This document describes necessary preconditions, acceptable levels of physiological stability, and the specific methods of assessment. A template is also provided to annotate brainstem reflex testing by assessing for absence of pupillary reaction to light, corneal reflex, pain reflex, cough and gag reflexes, and eye movement during caloric testing of the ears. Extensive detail was also provided on the apnoea test which involved first reducing the minute volume to allow PaC02 rise to 6.0kPa or higher and a pH of less than 7.4. Disconnection from the ventilator followed for 5 minutes with observation for spontaneous respiration. For the apnoea test to support the diagnosis of death by neurological criteria, the PaC02 had to rise by at least 0.5kPa during this time with no spontaneous breathing effort observed. 

Testing was always conducted twice, but with the time of death being recorded as the time at which the first set of tests were completed. This always seemed a bit weird and was often challenging to explain to the NOK during the diagnosis of DNC.

Red Flags and ‘additional cautions’

So, death is death. But it can ethically and legally happen to the whole body, or just the brain. And as medical practice advances, so of course does our understanding of the complexities of death. Over the last 10 years, several unusual index cases were reported during practical use of this guideline which gave the community pause and necessitated guideline amendments/updates. These issues were previously described as ‘red flags’ for the diagnosis of DNC, worth highlighting individually:

Aetiology primarily isolated to the posterior fossa or brainstem

What about a brainstem tumour that could be potentially resectable? The AOMRC tackle this with the following statement – In isolated posterior fossa and brainstem pathology without supratentorial involvement we recommend that an MRI is undertaken to more accurately delineate the extent of involvement and damage to the brainstem and other posterior fossa structures before a diagnosis of death using neurological criteria is made.

Patient receiving therapeutic steroids to reduce brain oedema (tumour, abscess, meningitis or trauma).

What about an oedematous brain that has the potential to respond to treatment? The AOMRC tackle this with the following statement: If corticosteroids are being used therapeutically to reduce brain oedema (e.g. tumour, abscess, meningitis or trauma), we recommend that a diagnosis of death using neurological criteria is supported by the addition of an ancillary investigation.

2018 – Death on ECMO

Use of VV and VA ECMO offer specific challenges to DNC regarding pharmacokinetics and apnoea testing. Supplementary guidance was produced by the AOMRC with support from FICM and others on the diagnosis of DNC when the patient is supported with ECMO. There is a strong focus on the use of drug level testing, administration of reversal agents, assessment of neuromuscular blockade, ensuring that gas measurements reflect cerebral PaCO2 and use of ancillary testing. 

2021 – Therapeutic decompressive craniectomy

No-one wants to make a diagnosis of death, which is then reversed by someone else. Unfortunately, this situation appears to have occurred in 2021 in the UK, when a patient who had undergone bilateral therapeutic decompressive craniectomies was diagnosed DNC and then went on to start breathing and regain consciousness. A consensus update was produced rapidly and decompressive craniectomy added to the list of red flag conditions in the AOMRC guideline.

2023 Ancillary testing

In the presence of red flags or challenges to clinical assessment (such as ocular injury precluding eye examination for example), clinical teams would often pursue imaging to support the diagnosis of death. Imaging methods were variable and subjective. A consensus guideline was released in 2023 describing the preferred use of CT angiography as a measure to standardise and provide assurance in this circumstance. A clinical diagnosis of DNC cannot be supported if the CTA demonstrates contrast opacification of any of the vessels specified in the 4-point criteria, as described by Frampas in 2009.²

Death in 2025: The AOMRC update

17 years is probably overdue for a complete update to any guideline, so we don’t think it’s surprising to see a the AOMRC publish again in 2025 on this complex issue. An initial unsettling gut reaction to a change in the way we diagnose death is also not surprising, even to the point of mild clinical panic. If the guidance is changing, then does that mean what stood before was in some way inaccurate or incomplete? Are conversations with families going to get even harder on this? And if we need to continually update the way we diagnose death, then are we actually sure what death is? 

The AOMRC has tried to assuage these concerns with the following statement, “The updated Code of Practice does not represent an alteration in the validity of historic diagnoses of death. However, medicine is constantly evolving so this Code includes a number of important updates based on growing global medical consensus, technological developments and lessons learned from individual clinical cases”.

Many are still uneasy despite this assurance. If the validity of prior testing still stands then what are the lessons learnt from individual clinical cases? And what would bereaved families say if they knew the way in which their relative was previously diagnosed dead, has now been updated? However, having digested the document in full and discussed with colleagues, our initial unease has been well mitigated. The first point being that the courts have always been very willing to listen to the medical community on how death can be and is diagnosed. Retrospective challenges are therefore highly unlikely. Second, updating guidelines in accordance with evidence, case law and lived experience can only ever be a good thing. Third, this update puts the UK on an even international footing as part of the World Brain Death Project.³  

So what are the actual changes?

1. Age categories: the new guidance can be utilised for all ages over 37 weeks gestation. The caveats being that between 37 weeks gestation and 2 years there must be 24 hours before testing takes place, 24 hours in-between the 2 sets of tests and there should be no ancillary testing utilised).
2. Apnoea Test: This is changing significantly, but for all the right reasons. The starting PaC02 has now been lowered to £5.3KPa but there is a requirement for there to be a far bigger rise at the end of the test with a PaC02 ³8.0KPa and pH<7.3. It is expected that this may not be feasible within a 5 minute period and therefore this is now the minimum suggested apnoea test period. This calls in to question whether there are a cohort of patients who will be unable to withstand this duration of apnoea without becoming hypoxic or cardiovascularly unstable. Ultimately, the anxiety is that this change will mean less completed DNC testing, with a subsequent increase in the use of ancillary testing. However, we don’t think there is any clear evidence to support this anxiety. We are always impressed by the apnoeic oxygenation evident in these cases and the relative stability of PaO2 during complete apnoea. The AOMRC highlights that this change does not call in to question the validity of previous tests, but has been adjusted to more closely align with international practice. 
3. Time of death: Time of death will now be formally recorded as the time at which the second set of tests are completed. This is clearer for everyone and allows families to be with their loved one at a clearly stated time of death. If ancillary testing is used then the time of death is the time that those test results are made available to both of the clinicians undertaking the diagnosis of death. 
4. Minimum Temperature of 36 degrees: This is a change from the previous guidance of a minimum temperature of 34 degrees. There is also now guidance to recommend a minimum 24 hour observation period following correction of hypothermia (attaining a core temp of 36 degrees or higher), to ensure absence of reversible causes prior to testing. 
5. Both eyes and ears must be examinable: If this is not possible then the updated guidance now recommends additional ancillary testing.

Death – where do we go from here?

Given the above changes, I think that the anticipation is that a proportion of patients in whom ancillary testing was not required previously will now require it to fulfil the diagnosis of death by neurological criteria. This is a real initial concern, as many of these tests are complicated and can only be undertaken in a specialist neurosciences centre. There are also often delays in obtaining and reporting department neurophysiology assessments such as EEG/evoked potentials and/or specialist imaging such as CT angiography, transcranial Doppler or MR. There is also a potential concern that such a revision will create a cadre of patients who are subjectively, but not categorically dead – those with catastrophic CT angiography findings and absence of the majority of clinical brain stem reflexes, but with some opacification in a single vessel and a blocked ear canal. 

However, these concerns are unfounded as yet, and we will need to implement this guidance and evaluate. In addition, the detailed and sequential evaluation of patients with presumed devastating brain injury and their management at the end of life is a routine aspect of critical care, increasingly informed by evidence and consensus. None of the changes in this document require a major change in practice and the international consensus work should only serve to make this challenging situation more standardisable, reliable and transparent. 

Death is tricky. Neuroprognostication is tricky. Managing uncertainty is tricky. But these aspects all make the job worthwhile, stimulating and validating. And helping families navigate this path can be some of the most rewarding work we do. 

It seems strange to wish people a happy new year after 2500 words on death, but we hope 2025 is prosperous for everyone and that St Emlyns can be of assistance throughout. 

References and further reading

1. Diagnosis of death. Memorandum issued by the honorary secretary of the Conference of Medical Royal Colleges and their Faculties in the United Kingdom on 15 January 1979. Br Med J 1979;1(6159):332. doi: 10.1136/bmj.1.6159.332 [published Online First: 1979/02/03]

2. Frampas E, Videcoq M, de Kerviler E, et al. CT angiography for brain death diagnosis. AJNR Am J Neuroradiol 2009;30(8):1566-70. doi: 10.3174/ajnr.A1614 [published Online First: 2009/05/02]

3. Greer DM, Shemie SD, Lewis A, et al. Determination of Brain Death/Death by Neurologic Criteria: The World Brain Death Project. JAMA 2020;324(11):1078-97. doi: 10.1001/jama.2020.11586 [published Online First: 2020/08/08]

4. AoMRC. 2025 Code of Practice for the diagnosis and confirmation of death. https://www.aomrc.org.uk/2025-code-of-practice-for-the-diagnosis-and-confirmation-of-death/

5. Lewis A. An Update on Brain Death/Death by Neurologic Criteria since the World Brain Death Project. Semin Neurol. 2024 Jun;44(3):236-262. doi: 10.1055/s-0044-1786020. Epub 2024 Apr 15. PMID: 38621707.