This month’s question concerns extubation of the critical care patient. What strategies (if any) do people put in place to try and minimise extubation failure? What do people feel is an acceptable failure rate for extubation attempts on ICU?

This month’s paper is a nice FOAM summary of the issues  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2760915/


Many thanks for this question,

Rather than a specific strategy I do have a mental checklist that consists of 5 questions in the first instance:

1. Has the illness resolved?

2. Will the patient be able to cooperate with physiotherapy (and therefore is also a measure of appropriate cognition)?

3. Will the airway be an issue without the ETT?

4. Is the respiratory reserve sufficient?

5. Will the heart tolerate an absence of PEEP and tachycardia?

In general, when any of these answers answers are clear cut we tend to either wait and reassess (if we think they may resolve) or perform a tracheostomy.  But when they are more ambiguous, I think the appropriateness depends on the clinical context and the overall spectrum of the answer to each question when combined e.g. I would probably accept a greater degree of agitation for a trial of extubation in the absence of chest problems in those with a resolving head injury, and in our OOHCA would accept a lower GCS in the presence of a strong ventricle. This is just my personal practice and not based on any robust evidence. But it is also not infrequent to have surprises either way, which clearly demonstrates the limitations in the strategies we use.

In terms of acceptable rates for failure, I personally think that defining “failure” as re-intubation within 72hrs is probably unreasonable (unless 71 of those hours have been with a GCS of 8 and constant NIV) and would probably be more interested in examining those patients who were re-intubated within a (albeit arbitrary) 24hrs, as a changeable component. It would be interesting to track our rates of failure and compare with other units (I do also wonder if it may be used in some benchmarking metrics nationally???), but as a throwaway comment I would be concerned if it was greater than 10%  24 hours……


Agree it would be interesting to see what our need for reintubation is, and how quickly it happens. I think 71 hours is very different from 30 minutes

However the danger is that we change behaviour if this is viewed as a marker of quality – you can avoid any failures by not extubating or giving everyone a tracheostomy; a non-judgemental learning review of each one could be informative.


I agree A that a ‘mental’ checklist is a good strategy for maximising the correct selection of patients to extubate. I also look at fluid balance and aim to have patients on the drier side for 24 hours before extubation. Pain control also needs to be addressed.  I also like to do a SBT for a good period of time and look at the RSBI as a guide, but this is obviously only a guide. I think what is also vital is having a good MDT discussion with the physio and nurses about any concerns re: extubation failure. I guess there are so many aspects that should/can be considered to maximise chances of extubation success and why a checklist is a good idea.

In terms of preventing re-intubation following extubation there is some more recent evidence that extubating directly to NIV may be beneficial. I am personally becoming more and more a fan of HFNO2 for the first 24 hours after extubation.


C, I agree with you re HFNO2. Following the Hernandez paper in 2016 (https://www.ncbi.nlm.nih.gov/pubmed/26975498) I thought that this would catch on as a strategy, but I don’t know of any units that routinely extubate to HFNO2.  The question it raises is whether we should (as they did in the study) give HFNO2 to everyone after extubation (expensive, but would get the best results) or triage higher-risk patients to receive it (not too far away from what we already do).

As a quality measure, it doesn’t really tick the necessary boxes, and it would need to be examined with the tracheotomy rate, the number of ventilator days, overall length of stay, VAP rates, etc.
This was my question so thanks to everyone who contributed!
Extubation is a bit of pet subject for me as I think it doesn’t get the attention it deserves in terms of the impact that getting it right or wrong can have on outcomes. To make matters worse there is a paucity of robust evidence out there to guide us. Finally as a triple whammy I think a lot of people confuse extubation with weaning ie you can pass a SBT but still fail a trial of extubation for a variety of reasons….
In terms of how we might improve things I feel that as with all complex multi-factorial problems the answer is also going to be multi-dimensional.
In HDFT we’ve tried to standardise the weaning process and the type of SBT we all do, we’ve brought in a checklist to make people consider +/- address the common factors for a failed extubation before you do it and finally encouraged people to consider HFNO2 for the high risk group post extubation.
So far we’ve reduced our failure rates (and tracheostomy rates interestingly!) and succeeded in “getting it right” at the first attempt more often. That said our failure rate is still 12% so lots more room for improvement….
The main issues for failed extubation remain poor cough, excessive secretions and airway oedema (despite attempts to address these beforehand).
So moving ahead clearly more work has to be done on clarifying whether we can identify and prevent a failed extubation, or whether it is better to simply go straight for a tracheostomy.
I suspect we’ll need “big data” to help us answer that though not a single centre QIP!!!

How many of us do a specific SBT as a test of readiness ? and How many of us use it as a guide for identifying issues around liberation from ventilation?

A major portion of the literature around the world on identifying patients with difficulty getting of ventilation do use SBT as a guide alongside duration of ICU stay  ie  Delayed and failed weaners

Our practise of using pressure support ventilation as a default mode for spontaneous breathing means we always do a trial of SBT  but do we name it ? time it ? repeat it? or use it as a guide?

The lack of standardised practise makes it hard to get any meaningful data to identify patient subsets

Apart from patient related issues there are nursing  Medical organisational factors that seems to determine the practise in each Unit

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s