An interesting conversation erupted last night on Twitter (which resulted in me being awake way past my bedtime, i.e. 8pm) regarding the management of a specific issue in resuscitation from cardiac arrest: the combative dead person.
In recent times, following the emphasis that has been placed on good quality CPR both bystander and rescuer, many of us have been noticing this somewhat spooky phenomenon more often. Trying on Twitter to nut out how to manage these patients was tough, so I would love some feedback: have you seen this happening and what do you do about it?
In patients who have arrested in public where bystander CPR is commenced immediately, it is increasingly common for us to see patients who present in a non-perfusing rhythm (usually VF it seems) but who, thanks to excellent CPR, still have cerebral perfusion. This results in patients who start doing rather unexpected and unwelcome things whilst we jump up and down on their chest. I personally have seen patients reaching for the hands of the person doing compressions, reaching towards their head as I tried to manipulate their airway, and looking around, seemingly aware of what is happening. Many other paramedics and medical practitioners have reported the same issues, although there is scant mention in the literature. Nevertheless, anecdote or not, this misbehaving is common enough now that my protocols, and the protocols of colleagues in New Zealand, specifically mention the phenomenon.
There are a few issues that this raises. First of all is the difficulties that come with having the patient moving around during resuscitation. Usually it is easy to physically restrain the arms, which will at least stop them from grabbing your hands while you do CPR. It does make life tricky when trying to place an IV however, and if the IV is placed in the cubital fossa it is then prone to kinking off unless hands are tied down. However the bigger issue that I have noted in several cases is difficulty managing the airway and ventilating. These patients often seem to have strong jaw tone which makes placement of an airway impossible, and their respiratory effort, whilst inadequate on it’s own, interferes with ventilation.
The other issue that is a bit trickier to get to grips with, is the possibility of these patients having awareness during their arrest. It is very spooky to have a patient looking around whilst in VF and many providers find it more than a little disconcerting. I do not know if this represents true awareness, but I would certainly rather they definitely had no awareness!
So what do we do about it? There seems to be two different approaches in the EMS protocols that I am aware of, with many of us prefering a third option. I have no firm answers as I’m still exploring the various issues myself.
The approach favoured in my protocols is to give small boluses (1-2mg) of midazolam during the arrest if gag reflexes prevent intubation, but they don’t address these patients who are localising to pain and have strong jaw tone. Using a NMBA is strongly frowned upon. The rationale for this approach is not given. The polar opposite of this is the New Zealand approach which calls for the administration of rocuronium only, with the rationale that the patient is not aware, so does not need sedation, and sedation “carries some risk in this setting”
I am trying to wrap my head around both of these concepts, so excuse me while I think out loud for a bit.
Sedation only: In my experience sedation alone has not been enough to allow me to successfully manage the patient, in particular to allow me to place an airway and ventilate. As noted above, there is some opposition to using sedation due the risk it carries, and further to this there was the line of thought that these patients don’t remember their arrest, are not truely aware, and therefore sedation is unecessary.
Whilst I don’t think sedation alone is enough, I’m not sure how I feel about either of these objections to using sedation. First of all, the risk: typically when we use sedation (especially something like midazolam) we are worried about negatively impacting on perfusion, primarily through vasodilation. Obviously the patient in cardiac arrest has about as poor perfusion as you can possibly have, and CPR only restores a small fraction of that. So perhaps concern over negatively impacting haemodynamics is warranted. On the other hand, we are pouring large (excessive?) doses of adrenaline into these patients at the same time specifically to cause vasoconstriction. So would a little bit of midazolam really matter? I suspect it wouldn’t, but I don’t really know. Perhaps (as in most things) ketamine would be a better option, but I don’t think that there is any reason to avoid midazolam.
The second objection is a bit more complex and perhaps bothersome, not from a practical management point of view, but rather an ethical standpoint. It seems to me that these patients are aware during CPR. I don’t know if they are or not, but it certainly seems that way as they localise to pain and look around. This makes me uncomfortable.
However the argument runs that if the patient does not remember the event, then there is no problem.
But I don’t know if it is as simple as that. I don’t think that not remembering suffering is the same as not suffering in the first place, but it is a tricky ethical conundrum. In palliative care for example, relief of pain and suffering is of paramount importance, and clearly those patients are not going to recall anything at the end of their treatment!
Of course there is also the distinct possibility that provision of sedation is actually for the benefit of the paramedic; some pyschological first aid for us to make us feel as though we are doing something good for a patient.
Paralysis only: I’m pretty sure I don’t like this idea much at all. Obviously in terms of stopping movement and allowing best intubating conditions, paralysis is king. I have no problem with this at all, but I do feel uncomfortable in providing long term paralysis without sedation in any patient. It may be that they will not remember anything, but again, I don’t think that is the same as not suffering in the first place. I could perhaps be comfortable with short term paralysis with suxamethonium only in this situation (sux and an apology) but then there may be issues with using sux in the first place. If the arrest was caused by hyperkalaemia then there is perhaps potential to exacerbate this by using sux. With that said, we really have no way of knowing whether this is the case in the field. We may take a guess if the patient is a dialysis patient for example, but we won’t really know. Again, we also come back to the issue of whether it will really make a difference anyway?
If the patient has hyper-k severe enough to have caused arrest, will the rise in K+ from sux be enough to be clinically significant? I don’t know, so if someone does, please tell me!
So. Heaps of writing, nothing decided.
I think where I stand at the moment is: Sedation only hasn’t seemed adequate to me in the past, but I don’t think there are harms associated with using it. I don’t know which agent is best; it’s probably ketamine, but at the moment I’m ok with cautious use of midazolam.
Paralysis only seems to be inhumane, especially if long term paralysis (rocuronium, pancuronium or vecuronium) is used. Short term (suxamethonium) paralysis without sedation may be ok to secure an airway, but it doesn’t deal with any ongoing movement/biting etc and there may theoretically be issues with potassium.
Which leaves me at the option I have used with success in the past: RSI with low doses of sedation and sux (as I don’t have roc yet), then ongoing sedation with whatever is authorised (midazolam for me) if required. Although this approach is not officially sanctioned by the powers that be, I have never run into issues when using it. I take care with my paperwork to explain the circumstances and my decision making and (as ever) this is enough.
Right! Please let me know if you have seen this phenomenon yourself and what you do to address the issues. Do you have protocol or is it up to the individual to problem solve?
Things I still don’t know are: Is sedation bad in these patients? Is suxamethonium bad in these patients? And most importantly, is not remembering bad things happening the same as bad things not happening?
Please get in touch with your thoughts.