There is some sort of precedent for admission to ICU, in that guidance for prognostication after cardiac arrest has made us far more likely to admit all patients post ROSC (assuming the reason for the arrest is treatable). I write ‘some sort of’ precedent because, although both patient groups have potentially suffered cerebral damage, there is a difference in that the head injured patient has a CT showing gross structural damage, whereas post ROSC there is nothing so tangible. Perhaps if CT were able to show hypoxia associated changes instantly (which it can’t), a better comparison would be the head injured patient with a patient post-ROSC with gross ischaemia on CT.
The million-dollar question raised however is: Is the chance of survival with good function after devastating brain injury such that admission to ICU is in a patients’ best interests? (Ignoring that for any individual patient with any disease, this chance is unquantifiable with any accuracy, and we have no real concept of what threshold for admission is appropriate). It’s true there are certain disease states we have been unduly pessimistic about, leading to self-fulfilling prophecies; maybe this is one of them? As good an example as any is COPD; this paper was a ‘game-changer’ for many clinicians. Even ventilation with fibrotic lung disease isn’t always un-survivable - how we identify the patient with reversible features is difficult, but should we therefore offer ventilation to all? (This is a really good review). Does this guidance facilitate access to critical care for a patient group with a lower chance of favorable outcome than we would usually feel comfortable admitting?
We face these difficult admission decisions every day, taking into account cost as well as possible benefit. For the patient with end stage fibrosis the potential cost is a prolonged death with a high treatment burden. With the head injured patient, the potential cost is that but also something else; it is risking survival with poor neurological outcome. From the viewpoint of my own values and beliefs (which may be completely different to yours or those of my patient), this may be the worst possible outcome. The difficulty is of course that we have no idea what the actual cost for our individual patient will be.
These injuries are occasionally termed ‘un-survivable’, and in many cases that will be true – indeed the given definition of catastrophic brain injury is one that is a direct threat to life. One potential advantage of admission is that once death is recognised to be inevitable, I think an ICU can provide excellent end of life care. In America, 58% of people dying in hospital do so in an ICU (data from 2004), and whilst I don’t think a default of ‘doing everything until nothing more can be done’, or using critical care as a preferred place for palliative care is appropriate, this benefit can’t be denied. In any case, if we maintain that the intention is to identify recovery this can never be more than a secondary benefit so wouldn’t alter a decision whether to admit.
The elephant in the room is that admission to ICU will increase the likelihood of organ donation by non-survivors*, allowing time and discussion away from a busy emergency department. Some patients may in fact suffer brain death during this period of observation, which again is ‘favourable’ from a donation perspective. The statement is clear in that a desire or ability to donate should not be influential, but I think there will always be a perception that this conflict exists. Perhaps the best solution is to be open about it; to not let donation guide management but to not be apologetic about this benefit either (I don’t think the authors help themselves here by the number of times it is mentioned, almost as a running theme).
Overall I welcome this statement; it provides some clarity in a difficult area. I do think we need to be careful about how we apply it, so that our intentions are clear and that we don’t offer false hope. I think the key to this is honesty with yourself as to why you are admitting the patient, but perhaps even more importantly with the patient’s family. Telling a family that the situation suggests real and severe brain damage, and that there is a high chance of death but that we can’t be sure at this point in time is honest. Where this guidance shouldn’t be used is as justification to admit for palliation or to increase organ retrieval.
If a patient ultimately has a ‘good death’ and part of that death is fulfilling a wish to donate, after being given every reasonable opportunity to demonstrate neurological recovery, that to me is surely a good thing.
I've heard that certain areas of the UK have adopted this guidance as standard practice; hopefully we'll see their experiences published before long.
Finally, as a bit of a postscript, something about this does continue to confuse me however. If we can’t prognosticate early, why do we know that neurosurgery wouldn’t make a difference - or put another way, on what evidence is the decision to not operate so certain? There’s an earlier post on this site “bilateral fixed dilated pupils – should we transfer” that asks a similar question.
*(COI - I'm the lead for organ donation).