This week we look at the RINSE trial1, a trial examining the prehospital use of hypothermia in post cardiac arrest patients.
Therapeutic hypothermia has had a bit of a rollercoaster ride over the last few years. It was all the rage following three small trials) which initially suggested a significant benefit from cooling patients to 33C following out of hospital VF cardiac arrest. It’s worth noting that the evidence for non shockable rhythmns was never that good)2.
The in 2014 we got the TTM trial (Targeted temperature management)3 which showed that there was little or no difference between keeping patients at 33C or 36C. Not everyone believed it of course, there is quite a lot of animal data around that suggests that hypothermia ‘should’ work; it just seems that we can’t get it to work in humans.
One criticism of the TTM trial is that patients do not achieve the cooling target very quickly. This has led to a number of pre-hospital services continuing with hypothermia strategies despite TTM. To be honest this is fine as TTM showed no real difference and so one could argue that it does not matter if you do or you don’t (from an evidence perspective at least). Cochrane recently reviewed this and found little evidence for pre-hospital cooling but the evidence was not definitive4 although lab based studies have suggested a potential advantage for intra-arrest cooling 5.
What we need is evidence for or against early cooling. We’ve previously looked at such a trial here at St.Emlyn’s 678 and there are a number of other trials including a small trial by some of the the authors of this paper9 which did not show a survival benefit. This is not just about earlier cooling either as there is a pathophysiological argument that intra-arrest cooling can improve ROSC and thus survival10. So intra-arrest cooling is not the same as post cardiac arrest cooling (as with the TTM trial) as we are looking at different outcomes and resuscitation strategies.
This week the RINSE trial has been released as a pre-publication on the Circulation website.
What kind of trial is this?
It’s a randomised controlled trial which is important as this is a trial of therapy. Observational data on hypothermia suffers from survivor bias and case selection issues and so we really do need something that formalises the decision to give or not give the intervention to the patients. Patients were recruited in blocks of 10 using sealed envelopes kept on ambulances. I’m not a big fan of envelopes…., but I can’t think of a better practical method here.
Tell me about the patients.
This is an Australian study across South Australia, West Australia and Victoria. The trial protocol is published here1111. Patients over 18 were enrolled with a few usual exclusions (pregnant, trauma, or already cold). Interestingly all rhythmns were included despite earlier evidence only really showing evidence for hypothermia for VF/VT. However, you could argue that this is really a different question, does cooling improve ROSC (as well as subsequent survival and neuro benefit) as opposed to in those patients who have already had a ROSC. Also, in the trial protocol they point out that the survival from non-shockable rhythmns is so poor (2%) that it’s worth including this group as otherwise we may never get any information.
What intervention was trialled?
In the intervention arm paramedics infused 2000ml of ice cold saline to patients during cardiac arrest. If that did not get the temperature below 34.5C more saline was infused until the temperature was achieved.
In the control arm patients were cooled when they arrived at hospital.
As with many cardiac resuscitation trials recruitment was challenging with just 1324 patients being recruited from 22,775 patients in cardiac arrest. This does cause some concern as those included patients, or responders may have been different from those excluded.
PAUSE – remember TTM? Well that came out when RINSE was recruiting and that led to a change in hospital management in the control arm. In other words TTM sort of scuppered RINSE and so recruitment was ended early. They got to 48% of the target number of recruited patients. OUCH! A real nightmare for any research team, but let’s not abandon the study, what can we learn?
What are the main findings?
The primary outcome was survival to hospital discharg (Ed – Fair enough, although I prefer neurological status at a later time using something like GOS). They did look at later neuro status and ROSC as secondary outcomes.
1198 patients were recruited with 618 randomised to hypothermia. The key outcome of survival to hospital discharge was actually slightly worse in the cooling group (10.2% vd 11.4% p=0.71). For the secondary outcome of ROSC there was similarly a decrease for the cooling group which was statistically significant (41.2% compared with 50.6%, p=0.03).
The authors postulate several mechanisms why they did not see the advantages suggested by lab based studies in the trial. Decreased myocardial perfusion pressure from volumenoverload and an increase in post ROSC pulmonary oedema are suggested.
It’s also worth noting that many patients did not get the full 2000ml of saline.
So should we abandon cooling?
It’s tricky. The lab data says yes, the real world data says no. The real question is whether it might cause harm and that can be by two mechanisms. Firstly an intervention may be directly harmful in which case you should not do it. Secondly, spending time doing interventions that don’t work distract you from doing stuff that does.
At the moment it seems tricky to justify taking time to cool when you might be better doing something else.
Before you go please don’t forget to…
- Subscribe to the blog (look top right for the link)
- Subscribe to our PODCAST on iTunes
- Follow us on twitter @stemlyns
- See our best pics and photos on Instagram
- PLEASE Like us on Facebook
- Find out more about the St.Emlyn’s team