The ethics of reverse triage or ventilator throughput.
26/03/20 07:56
There is lots of discussion in ICU circles about reverse triage.
It goes something like this:
“Two people need a ventilator, one is in the emergency department, young, needing intubation, the other is on the ICU, elderly perhaps, not doing so well. If they had both arrived into the ED simultaneously and there was only one ventilator; the ICU doctor would have triaged to treat the young person first. So, given the actual scenario, reverse triage suggests the ICU doctor should go up to ICU and remove the ventilator and adopt palliation for the current patient already in the ICU, to make the ventilator available for the young person in the emergency department who has a greater chance of benefiting.”
It’s attractive in its utilitarian simplicity but misunderstands triage which was designed for simultaneous mass casualties. Only in simultaneous mass casualties is it as simple as ‘one or another’.
It goes back to the typical ICU conundrum of the ‘last bed’. We wrote about this in our paper.
“The current resource limitations of the ‘last bed’ have a potential, rather than definite, ability to limit timely access to the intensive care for another patient (justice). Not infrequently, in analysing the situation using MORAL balance, additional options present themselves…. it might be that, on close examination, the last bed is illusory, and that additional resource can be found to accommodate the patient… However, in other circumstances, the last bed may not be illusory, but absolutely real [and] the facts and outcomes of relevance for the clinical scenario are altered. One can see how this might swing the balance leading to a different decision.”
We have found that the larger an ICU the more illusory the ‘last bed’ is.
The more ventilators you have the greater chance a change in one of your patients, for better or worse, will occur.
The better.
Don’t just focus on the admissions and ventilator numbers – focus on your ventilator throughput. The coming off of ventilation for ICU patients and making this as safe and swift as possible will be just as important as planning for who goes on. Try and be accurate in your assessment.
Who in your critical care is leading the ‘coming off’ group? It is just as important as the who gets admitted group.
The worse.
Let’s look at some simple numbers for worse.
(I appreciate someone cleverer than me could analyse this in more detail).
A tertiary referral 21 bed ICU, 100% occupancy (pretty standard in UK), ICU mortality 15%, mean length of stay for those who die 6 days. Died in ICU is 200 patients per year.
i.e. approximately 1 death every 2 days.
If we were to double the number of ICU beds and fill them with a similar cohort of patients, we might have one death / day.
But international COVID-19 data suggests 50% mortality, mean length of stay for those who die 4 days.
It’s not just one death / day but multiple deaths a day.
And ICUs are gearing up for 3-4 times greater number of ventilated beds.
Sadly, in this pandemic, and experience from Italy and Spain suggests, we are looking at a death every few hours.
A large number of beds mean a larger throughput.
Focus on those getting better, off a ventilator.
Focus on those getting worse, by reassessing their survival chances.
But don’t follow the Swiss guidance – “Patients in the ICU must be assessed regularly (at least every 48 hours) and interprofessionally.”
48 hours!
UK standard is twice a day consultant review – outside of pandemic situations.
Pandemic situations call for greater input, not less.
Here are our suggestions (and we know it is hard!)
Acutely
Assess ventilated ICU patients’ multiple times per day by a senior intensivist, with colleague and MDT support.
It goes something like this:
“Two people need a ventilator, one is in the emergency department, young, needing intubation, the other is on the ICU, elderly perhaps, not doing so well. If they had both arrived into the ED simultaneously and there was only one ventilator; the ICU doctor would have triaged to treat the young person first. So, given the actual scenario, reverse triage suggests the ICU doctor should go up to ICU and remove the ventilator and adopt palliation for the current patient already in the ICU, to make the ventilator available for the young person in the emergency department who has a greater chance of benefiting.”
It’s attractive in its utilitarian simplicity but misunderstands triage which was designed for simultaneous mass casualties. Only in simultaneous mass casualties is it as simple as ‘one or another’.
It goes back to the typical ICU conundrum of the ‘last bed’. We wrote about this in our paper.
“The current resource limitations of the ‘last bed’ have a potential, rather than definite, ability to limit timely access to the intensive care for another patient (justice). Not infrequently, in analysing the situation using MORAL balance, additional options present themselves…. it might be that, on close examination, the last bed is illusory, and that additional resource can be found to accommodate the patient… However, in other circumstances, the last bed may not be illusory, but absolutely real [and] the facts and outcomes of relevance for the clinical scenario are altered. One can see how this might swing the balance leading to a different decision.”
We have found that the larger an ICU the more illusory the ‘last bed’ is.
The more ventilators you have the greater chance a change in one of your patients, for better or worse, will occur.
The better.
Don’t just focus on the admissions and ventilator numbers – focus on your ventilator throughput. The coming off of ventilation for ICU patients and making this as safe and swift as possible will be just as important as planning for who goes on. Try and be accurate in your assessment.
Who in your critical care is leading the ‘coming off’ group? It is just as important as the who gets admitted group.
The worse.
Let’s look at some simple numbers for worse.
(I appreciate someone cleverer than me could analyse this in more detail).
A tertiary referral 21 bed ICU, 100% occupancy (pretty standard in UK), ICU mortality 15%, mean length of stay for those who die 6 days. Died in ICU is 200 patients per year.
i.e. approximately 1 death every 2 days.
If we were to double the number of ICU beds and fill them with a similar cohort of patients, we might have one death / day.
But international COVID-19 data suggests 50% mortality, mean length of stay for those who die 4 days.
It’s not just one death / day but multiple deaths a day.
And ICUs are gearing up for 3-4 times greater number of ventilated beds.
Sadly, in this pandemic, and experience from Italy and Spain suggests, we are looking at a death every few hours.
A large number of beds mean a larger throughput.
Focus on those getting better, off a ventilator.
Focus on those getting worse, by reassessing their survival chances.
But don’t follow the Swiss guidance – “Patients in the ICU must be assessed regularly (at least every 48 hours) and interprofessionally.”
48 hours!
UK standard is twice a day consultant review – outside of pandemic situations.
Pandemic situations call for greater input, not less.
Here are our suggestions (and we know it is hard!)
Acutely
- Stick the tube in (intubate) the young person – get someone to bag them.
- Wait, something is likely to change, for better or worse on the ICU, another option will materialise.
- Make individualised decisions for each patient.
Assess ventilated ICU patients’ multiple times per day by a senior intensivist, with colleague and MDT support.
- Focus on your ventilator throughput. The coming off. For better or worse.
- Duration of need for ventilation will swing the MORAL balance.
- But keep assessing, as accurately as you can, each ventilated patient as an individual.