Thread by @WhistlingDixie4

🧠 Heading here this morning for joint @ERC_resus session ‘How do I predict neurological outcome after cardiac arrest” at #LIVES2022

This is a core part of my day to day and frequent conversation I have with the junior team so interested to listen …

So far the audience have been asked four times to move closer to the front. Nobody moves.

The first q is ‘how do you do it’.

Sandroni says Neuron specific enolase (NSE) for 3 days, clinical exam from day 3, EEG starting early…used later, CT is repeated in patients with prolonged unconsciousness and MRI in those that aren’t waking up..

They have protocol on unit

Hoedemaekers and Taccone also speak about collecting EEG and NSE in days before they use it

Taccone says they don’t used MRI before day 5 and don’t repeat CT head routinely (outside of looking for cause of arrest on arrival)

The talk both the absolute levels of NSE are primary concern but also they look whether it’s increasing

❓Audience member asked about NSE and unclear evidence.

(They mention a systematic review. I think it might be this one)

pubmed.ncbi.nlm.nih.gov/34663643/

Taccone affirms there are better ones, but we’re limited by availability.

Sandroni says that Nuerofilament light (NFL) is the future but it has specific complication characteristics and establishing a threshold isn’t straightforward

Hoedemaekers says you need to know the weakness and strengths of your biomarker

She mentions the caution of haemolysis concurrent with NSE

(I haven’t heard that before, but found this pubmed.ncbi.nlm.nih.gov/30758520/)

❓Someone asks a questions about PCI/Cath lab but they’re told they’re in the wrong session

(Just as well they’re on the app 😳)

Horn wants to talk about whether EEG under sedation are reliable

✅Hoedemaekers says if you’re getting a malignant or highly pattern then yes it still is reliable.

❗️But it is vital to know at what point time in the post arrest period they were taken

Hoedemaekers is asked to what a malignant EEG is…she talks about ‘flat (and some other things)

Sandroni says ‘suppression plus minus burst’

Oddo says involve neurology

(For me - issue is report signposting what I need normal/malignant/highly malignant SE or myoclonus)

He says they do EEGs during sedation holds, so it’s at a nadir

🔅Sandroni says that the robustness of EEG is actually demonstrated most at 12 hours

❓Audience members asks something along the lines of why do we give sedative when we’re not doing hypothermia now

Answer given is to apply the evidence of TTM2 then you need to apply that protocol and we don’t know yet if early withdrawal of sedative is ‘ok’

Hoedemaekers says actually if patients are ‘trying’ to wake and we’re expecting good outcome. We wake them up

Sandroni says moderate sedation is part of the standard neuroprotective approach and he wouldn’t let this go in the first 24 hours

❓Audience from 🇦🇺 says they don’t use NSE. Is it actually required when you give *enough* time and repeated clinical exam.

He gets it won’t allow ‘early’ prediction but it will allow prediction

Oddo (chair) rephrased the question and asks what should you prioritise

Sandroni says we often overlook the importance of clinical exam. He said we know now that person with myoclonus for examine can be ‘awake’ but unable to demonstrate it

He says he has no special preference for specific tools

Horn pushes him to pick - what then for that patient with myoclonus?
what tool makes a difference?

He says EEG

Benign and malignant myoclonus will have different patterns

Hoedemaekers says to remember that clinical exam is great BUT false +ve for motor score is high and may lead to over pessimism.

Taccone says something to the effect a systematic protocolised approach for all is what he prefers

is it really necessary to have continuous EEG

(Am interested - We don’t have it)

Taccone said If you want to know what the brain is doing, most need intermittent EEG, which is as effective as continuous, but MORE than once a day

But he does it continuously because he can

So two of panel have continuous EEG. Sandroni doesn’t.

Sandroni says he does one EEG ~12 hours because benign EEG when done early is best at predicting good outcome

❗️but malignant patterns NOT as predictive of bad outcome, early

A 🇬🇧 doctor says they tend to do the EEG >/= 3 days+ into stay …. can we talk a bit more about how use this early EEG that’s been talked about

Horn says the sensitivity of EEG decrease over time but the specificity stays the same

He said people misinterpret the algorithm and feel they need to ‘wait’ for 3 days to get the info and that’s not the case

(I have to say I’ve not realised this and certainly haven’t really heard it done <24 hrs. Makes note to reread ERC paper linked to image below)

A doctor from 🏴󠁧󠁢󠁳󠁣󠁴󠁿 next (don’t @ me over the flag. The last dr said 🇬🇧. This one said 🏴󠁧󠁢󠁳󠁣󠁴󠁿)

He says the hard cases are indeterminate M3-4…EEG not highly malignant or benign …what time scales?

He doesn’t have NSE

Sandroni said we need more tools to make the algorithm more sensitive … Hoedemaekers said it is about time but maybe just combining everything is the best way

Maybe future is about using specific combinations to look at specific parts of the pathways

Taccone says we also have to live with grey zone

(Nobody has mentioned the patients goals of treatments and values yet and I’m surprised because when you know that I think it’s sometimes less hard)

Wrap up comes.

There’s lots of unanswered questions on the app…

This has definitely been the session with the most audience questions that I’ve been in

If you’ve not read the 2022 @ERC_Research they’re in @yourICM and here pubmed.ncbi.nlm.nih.gov/33765189/

Ends
#LIVES2022