ICU delirium - what to do?

Todd Fraser on 28-07-2010

We're implementing a new sedation-agitation protocol in our unit. Its been a very interesting exercise in evidence based medicine. While delirium, as defined by the CAM-ICU score, appears very common in ICU patients, there are few, if any, placebo controlled trials studying the efficacy of the commonly used agents (midazolam, haloperidol, dexmeditomidine, olanzepine to name but a few). It appears that implementing protocols to address delirium improve other markers such as length of stay, but how much of this is Hawthorne effect? Others compare one drug with another (eg MENDS trial comparing Dexmetatomidine and Lorazepam) and show similar effects - but is it the delirium or the use of a sedative drug that makes the difference? Another issue has been that of hypoactive delirium. While not as common as agitated delirium it appears similarly bad for you. There is almost no evidence of efficacy for any agent in this group. So we seem now to be good at detecting delirium, but we don't look for it enough, and when we do, we don't really know what to do with it. Hmmmmmm.....

6 Comments

Got something to say?

Log in to your account

Forgotten your password?

Todd Fraser wrote 06-04-2011 01:02:59 pm
Check out our new podcast interview with Dr Michael Reade on this subject. Michael's group previously published a comparison of Haloperidol and Dexmeditomidine in ICU delirium, and is now heading up 2 Australasian trials on the topic. Its a great interview, available Monday 6th June 2011

Todd Fraser wrote 04-26-2011 06:10:05 pm
We went to a lot of trouble to develop a protocol for delirium screening and management, utilising the CAM-ICU. Really like the protocol, but no one uses it. Not sure why.

Is CAM-ICU well utilised in other places?

Anthony Tzannes wrote 05-09-2011 12:17:02 am
The what? (I think that answers for where I currently work - tend to do a clinical as opposed to use formally scored assessment)

Something else to look up!

Todd Fraser wrote 05-09-2011 10:32:18 am
Yep, that's what I thought!

I don't have a good feel for this yet - those that propose the use of clinical scores such as CAM-ICU tell us that we will detect delirium much more commonly if we use it, particularly hypoactive delirium. However, there is patchy evidence at best for treatment modalities in ICU delirium, particularly the hypoactive type!

Additionally, CAM-ICU is more labour intensive (at least for me) than I was led to believe. Investing that time needs to be met with reward, which at this point in time, I don't see.

Benjamin Moran wrote 05-10-2011 04:24:09 pm
I wouldn't be so quick to shelve it. I see the usefulness of the CAM-ICU (or the ICDSC), like most things in ICU, as multifactorial. Firstly, to perform the assessment, a sedation holiday is required. This has been shown to improve outcomes and reduce the need for investigations (i.e. CT scanning). Secondly, as you mentioned, a hypoactive delerium may be diagnosed. This may impact on certain treatment modalities for that patient (i.e. not using benzodiazepines for sedation). Other non-pharmacological therapies may also be enforced (presence of family members, attempts to improve sleep-wake cycle, adequate natural lighting, noise-reduction at night (good luck with that one!)). It will also result in the investigation of the aetiology of the delerium that would have previously gone uninvestigated.

Although there are limited therapies available, given the morbidity and mortality associated with delerium, surely it would be better to diagnose it and treat the underlying causes in an attempt to holt progression. Anecdotally, our unit is using dexmedetomidine (you really have to pick your pt!) and quetiepine as the front line for delerium treatment. Certainly, there is less benzodiazepine use in these patients. Whether this has influenced outcome in our unit remains to be seen.

Todd Fraser wrote 05-13-2011 08:33:13 pm
All good points Ben

There does, however, seem to be very little data out there on hypoactive delirium. While we know that the hyperactive kind is bad for you (but don't yet know if treating it improves outcome as such, let alone know HOW to treat it), we know little about the quiet type.

Perhaps, as you say, that once we have a definition for it (using CAM-ICU or some other score), we may be able to get better information. Unfortunately, like all "syndromes" without a hard diagnostic test, this may prove contentious and difficult to get a consensus definition.