When can and should we withdraw from continuing resuscitation in the ED?
The standby phone rings: Post cardiac arrest patient, approximately 50. Intubated, BP 80/40, GCS 3. ETA 5 minutes. You meet the crew in resus and get the story. Your fifty-year-old man was seen to collapse on the street and bystander CPR commenced. Paramedics arrived within 8 fminutes and confirmed asystole that converted to PEA (pulseless electrical activity) after adrenaline. They had ably gone through the ALS (Advanced Life Support) algorithms and after around 20 minutes they got a pulse He’s arrested twice since, but responded to adrenaline and CPR on both occasions. Total down time, they reckon, is 50 minutes, give or take. The Police are trying to get in touch with his family.
On examination he’s intubated, his ECG is globally ischaemic but there’s no ST elevation. GCS is indeed 3, and he makes no spontaneous respiratory effort. There are no signs or history to suggest trauma, and he’s warm to touch. Blood gas shows a typical post arrest mixed acidosis, with nothing else shouting at you. As you page ICU and arrange transfer for a CT brain (? massive SAH) you wonder how many times you’ve been here before. Must be two dozen. Past experience tells you where it’s going, and you know it is not going to be a happy ending.
The Intensivist on-call is there pretty sharpish, and looks worried. The Unit is full to bursting, and the patient will need to be transferred out. She begins the negotiations to organise a bed.
An hour passes, the scan shows hypoxic brain injury but nothing else. Blood gases are improving, as is the patient’s ECG. He is maintaining a blood pressure with a low dose noradrenaline infusion. He’s s still not breathing or requiring sedation though.
You’ve spoken with his family. They are stunned. He wouldn’t have wanted to be a vegetable, they say (their words, not yours). They ask you to tell them what’s going to happen. If it’s anything less than a good prognosis they tell you that he would not have wanted to be resuscitated. They tell you that he is a strong man and would never want to be left dependent on others. They know this as the family have spoken about it in the past after a cousin was left with severe brain injuries following a motorcycle accident. They don’t want him to end up the same way.
You reassure them that we have policies to admit all patients to the ITU, and you do. Sadly, although your patient survives they are left in a persistent vegative state. Three months later he’s still on the ICU. At four months you walk into a lift and find the family in there on the way up to ICU. You ask how they are. ‘
‘Awful doctor, I just wish we’d let him go. This is not what he or anyone of us would have wanted. I know you had to do what you did, but I wish you’d just have let him die’
An hour later you’re back in the ED when the standby phone rings. Cardiac arrest, 55 year old male. ROSC. GCS 3. BP 75/30. They’ll be with you in 2 minutes…..
Ed – all cases here are hypothetical used to illustrate the complexities of decision making in the resus room.
Diagnosing death in the ED.
We are here to save lives. I would argue the Emergency Medicine is a very pure expression of a Doctor’s role: we meet a patient, we use all the tools at our disposal to work out what is happening, we do our utmost to make it better. Simple aims (complicated means): simple people. Right? We’re here to save lives.
Sometimes we can’t. Sometimes the damage seems too great, and all we can do is the best we can, to support the patient and wait and see.
Can we ever, though, say “Hold, enough”? Are there any circumstances when we can be sure that we’ve done all we can, that there is no possibility of a satisfactory outcome? Can we ever withdraw life-sustaining treatment in the ED? The case above illustrates a set of circumstances where the data suggests that the prognosis is likely to be poor.
The Academy of Medical Royal Colleges code of practice for the diagnosis and confirmation of death (2008)2 presents a series of recommendations regarding the diagnosis of death and withdrawal of life sustaining treatment. This code considers both the patient in cardio-respiratory arrest, and the apnoeic patient with a beating heart but suspected neurological injury incompatible with life. I will explore this challenging situation with regard to this code of practice, using a series of hypothetical cases to provoke discussion…
Ed – if you are likely to be involved in this sort of decision for real, then you need more than this blog. Read the guidance, discuss with colleagues and make a plan.
Your patient arrives. As before 50 year old OOH CA. Again, intubated without drugs, GCS 3 with no respiratory effort, but maintaining a blood pressure with boluses of IV adrenaline. You get a bad feeling and so does the ICU team on call. CT brain is performed without incident, and you see immediately that there is significant brain injury. There is massive intracerebral blood, with early ischaemic changes visible. You discuss the patient with your Neurosurgical Centre, and they state the situation is unsalvageable.
Is it possible to diagnose death and withdraw ventilation on this gentleman, considering the prolonged downtime and CT findings?
In this case, the answer is probably yes. In circumstances where the brain is demonstrably catastrophically injured, in this case by intracranial haemorrhage, in other circumstances potentially by trauma, it is possible to prognosticate effectively. This patient’s brain is irreversibly damaged. Within hours of the injury the clinicians involved can be sure that he will not survive, and the time is appropriate to discuss organ donation with his family.
Your patient arrives, Mrs. B. Fifty years old and a type II diabetic, Mrs. B has been brought from home, where she has been unwell for weeks. She has been off her legs, eating and drinking poorly, with increasing confusion over the last days. She was found unresponsive and apnoeic this morning by her daughter, and CPR commenced. On arrival of the paramedics it was confirmed she had suffered a PEA arrest. After around 30 minutes circulation was restored following fluids, CPR and multiple adrenalines administered per intra-osseous access. On arrival she is maintaining a pulse, although apnoiec and comatose without sedation.
Initial blood gas shows massive metabolic derangement. Sodium reads 170mmol/L, potassium 8mmol/L, and glucose 60mmol/L. Again she has a marked mixed acidosis.
CT brain shows ischaemic injury but no intracerebral bleeding, and over the next hours she maintains a blood pressure, although GCS remains 3. Her biochemical dysfunction is slowly responding to treatment, but she remains comatose. Is it possible to diagnose death and withdraw ventilation on this lady, considering the prolonged downtime and CT findings?
In this circumstance, the answer is no. As the AMRC Code of Practice notes, prognostication in a patient with a period of cerebral hypoxia following cardiac arrest or hypo perfusion is challenging. Before consideration of withdrawal of life sustaining treatment in such a case brainstem reflex testing should take place, but this should only take place after reversible causes of coma have been excluded. One such potentially reversible cause is severe metabolic derangement.
The code of practice states that in severe hypo or hypernatraemia (Sodium less than 115mmol/L or greater than 160mmol/L), severe hypo or hypomagnesaemia or phosphataemia (<0.5 or >3mmol/L) or massive hypokalaemia (potassium <1mmol/L can present with flaccid quadriplegia) careful normalisation of biochemical parameters must be performed before brainstem testing can take place. Other metabolic causes to be considered include massive hypo or hyperglycaemia, Addisonian crisis and myxoedema coma. In Mrs. B’s case therefore, prognostication must be tempered by the possibility that her hypernatraemia and hyperglycaemia may be contributing to her coma and her care should reflect this.
Red standby- 35 year old man found collapsed in a bar in city centre at 2am on Saturday morning. Immediate bystander CPR was provided and PEA arrest confirmed within minutes, which was unresponsive to ALS provided by the Ambulance service. Downtime by arrival at the ED is 40 minutes, but following 2 rounds of CPR in Resus he regains an output. Police report that he had attended a dance club, and that friends have spoken of use of ketamine and MDMA tonight, along with alcohol. He was found unresponsive in a dark corner of the club, and had been assumed asleep by his friends.
Initial investigations show no gross metabolic disturbance, and CT demonstrates widespread severe ischaemic injury but no catastrophic brain damage. He remains apnoiec and requires inotropic and ventilatory support. Several hours pass and he shows no signs of spontaneous respiration, despite no sedation being administered.
Is it possible to diagnose death and withdraw ventilation considering the prolonged downtime and CT findings?
This is unfortunately an increasingly common occurrence in EDs in the UK. Multiple recreational drug use is a popular pastime, and many resus rooms are filled by severely obtunded clubbers any night of the week. However, as the AMRC recommendations note, where cardiac arrest may have been caused by the respiratory depressant effects of drug taking, it is necessary to ensure these same effects have been excluded as a cause in prolonging the patient’s obtunded state post-arrest. As such, before it is possible to perform brainstem testing a “washing out” period should take place to ensure depressant drugs are not acting to maintain a patient’s coma. It should also be born in mind that when street drugs are involved it is never possible to know exactly what has been taken, and in what quantity (and when). Further, with many novel recreational drugs on the market effects of overdose and interactions are rarely well known. A period of days of supportive care is indicated to ensure that these potential depressant actions are no longer in effect before it is possible to establish brain death, and consider organ donation.
Your patient, Mrs. D, is brought into resus, intubated and being ventilated by hand. She is around 60 years old, and was in Outpatients of a nearby community hospital when she was seen to clutch her chest, turn blue and collapse. There was immediate ALS from the clinic’s staff and after 10 minutes ROSC achieved from PEA arrest. She is currently GCS 4, with spontaneous extensor posturing and agonal breathing against the ventilator. Propofol infusion allows ventilation, and she is prepared for CT brain. Working diagnosis of her cause of collapse is massive PE, and CTPA is agreed. Relatives are being sought.
The trip to scan is uneventful, although she now requires increasing pressor support. CT brain again shows ischaemic injury, however CT chest is more alarming. She does indeed have a saddle PE, but it also shows widespread lung and bone metastases. As she arrives back in Resus her niece arrives in the Department. You sit her down in the relatives’ room and she is sanguine. Her aunt has breast cancer which has not responded to treatment. She is a proud lady and insisted on attending her hospital appointments alone, not wishing to be a burden on her family. Mrs. D’s niece says she would not want to be admitted to ICU.
Is it possible to withdraw resuscitation at this time, in line with the patient’s wishes?
Here the answer is yes. The ARMC code of practice states that, subsequent to the intubation and ventilation of a patient (for example post arrest or to facilitate emergent imaging) information may come to light which will affect further management decisions. This may be related to the acute presentation- as in the case 1, where brain imaging demonstrates catastrophic injury, or may be due to clinicians learning of the patient’s pre-morbid state. For instance, it may be discovered, as here, the patient is being palliated, or suffers from end-stage disease that would mean they would never recover to leave ICU. The code of practice highlights that if this new information shows that Intensive Care management will inevitably be futile, or against the patient’s best interests, then irrespective of brain death withdrawal of ventilator support may be appropriate. In such cases careful discussion with family should take place, and an appropriate place for palliation sought.
So yes, there are occasions where we can say “Hold, enough” for intubated, ventilated patients in the ED, although there are more occasions when we should not. Consideration of the AMRC code of practice highlights that before withdrawal of life sustaining care is possible every reversible cause should be excluded, unless cerebral injury is too great or pre-morbid state precludes ICU treatment. In most cases a period of supportive care and observation is mandated before brainstem testing and withdrawal of resuscitation can be considered.
If confronted with a situation where withdrawal of resuscitation may be considered in the ED also remember the following: Medicine, especially critical care medicine, is a team game. Decisions like these should be taken in concert with other senior (read Consultant) colleagues, and always in discussion with the patient’s family. Further, when considering the management of an incapacitous patient (in England) one must consider the Mental Capacity Act3, taking into account previously expressed wishes, the patient’s values, any Lasting Power of Attorney and so forth. Finally, the clinician in this situation must always explore the possibility of organ or tissue donation, which may provide some solace to relatives, and hope to others.
Finally, the clinician in this situation must always explore the possibility of organ or tissue donation, which may provide some solace to relatives, and hope to others.
In conclusion, as Emergency Physicians we will be faced with situations where the question of withdrawing life sustaining treatment may arise. In such a situation it is vital that we engage with colleagues, family members and act within the AMRC’s code of conduct in order to arrive at the right position ethically and legally. In order to do so it is necessary to have an awareness of this area, and I hope this has offered a way in to consider these issues.
Senior Trainee in Virchester & Alumnus of St.Emlyn’s
- Shakespeare, W. Macbeth; Act 5, scene 8.
- Academy of Medical Royal Colleges (2008); A CODE OF PRACTICE FOR THE DIAGNOSIS AND CONFIRMATION OF DEATH. Accessed online at: http://www.bts.org.uk/Documents/A%20CODE%20OF%20PRACTICE%20FOR%20THE%20DIAGNOSIS%20AND%20CONFIRMATION%20OF%20DEATH.pdf 18th February 2016
- Mental Capacity Act (2005); ISBN 0 10 560905 6
(Ed – all cases here are hypothetical used to illustrate the complexities of decision making in the resus room).
- The mental capacity act
- Difficult conversations about dying in resus
- Family presence for brain stem testing
- Communicating terrible news with Liz Crowe
- Let’s talk about dying.
Before you go please don’t forget to…