Just a quickie this week as I am so busy with planning the SMACC party, the St.Emlyn’s LIVE conference and the Teaching CooP course in Manchester. I’m also off to #BadEMFest18 in a few weeks and need to start answering some emails from my good friends down there so we can organise our sessions and workshops (come to the Major Incident one – it’s going to be fab – and Craig I promise I’m ready).
Anyway, back to some science.
Back in 2016 Niall blogged on a paper in the Annals of Emergency medicine on the use of Isopropyl swabs to reduce nausea in the ED 1. Back then in an RCT of 84 patients the data looked fairly good. We advocated the use of swab sniffing whilst you were getting your stuff together to give a ‘real’ antiemetic and thought that it was something that could be done at triage or in the prehospital environment. It’s quick, easy and cheap (a rare combination), and more than that it sometimes works!
However, we should always be skeptical of small trials in a single centre with convenience sampling. They tend to overestimate the effects of an intervention and so we are always looking out for confirmatory studies. This week we have one.
April et al have published a similar study2 and although it’s still a convenience sample in a single centre 3 , it does give us a few more insights into the therapy. The abstract is below, but as always we strongly recommend you read the full paper before you make up your mind on this.
It’s different to the Beadle study in 2016. They did a three arm study with some interesting combinations.
- Group 1. Inhaled isopropyl alcohol and 4mg oral ondansetron.
- Group 2. Inhaled isopropyl alcohol and oral placebo
- Group 3. Inhaled Saline and 4mg oral ondansetron.
That’s an interesting combination of therapies but sort of fits with what we might want to know about whether isopropyl alcohol has an additional effect with ondansetron (which is my go to antiemetic in the ED).
They collected data using a VAS at 10,20,30, and 60 minutes and then hourly until discharge.
Tell me about the patients.
We always need to know if they are similar to our own and pretty much the answer is yes. These are adult patients with a primary complaint that includes nausea and vomiting. There does not seem to be anything especially exciting about them, but it is a single centre and as we all know patient populations do differ between institutions and especially in healthcare economies that have payment for service models.
The baseline groups were fairly similar, but there were some gender bias concerns. However, the problem here is about numbers. With only 40ish patients in the groups you need to have huge differences for them to be foud with basic stats testing. I always worry about small samples in studies like this. The potential underlying causes of nausea and vomiting are wide and you would only need a few patients in one arm to really skew the results.
What did they find?
Firstly the alcohol swabs seem to work as they did in the Beadle study. However, it is a bit weird on the effectiveness of Ondansetron which I’ve found to be a pretty good drug. In terms of the three groups the reduction in nausea on an 100mm scale was as follows.
- Group 1. Inhaled isopropyl alcohol and 4mg oral ondansetron. – 30mm
- Group 2. Inhaled isopropyl alcohol and oral placebo – 32mm
- Group 3. Inhaled Saline and 4mg oral ondansetron. – 9mm
That makes it look as though alcohol is by far the best way to reduce nausea and so we should all be doing it right? Well let’s think about this….
Are there any concerns with the study?
We’ve already mentioned convenience sampling and single centre, and for the UK a different(ish) population. We also know that it’s hard to blind patients with the alcohol as the whole point is that it smells! This is also known to those who are helping record the data and deal with the patients. There is lots of potential bias in there.
The patient group also had some interesting exclusions such as patients with an IV in, but I’m OK with that as if you have IV access then just give an IV antiemetic.
We don’t have a great breakdown of the underlying cause against the three treatment arms. There is some breakdown in the paper by how long each patient stayed in the ED, but by then we are talking about very small numbers in each of the sub analyses and I’m I think it’s courageous to draw too many conclusions there.
There is also the question of Ondansetron, which in this study really does not seem to work at all. I’m unclear why as that’s not my understanding of the literature or my personal experience. I don’t have a good feel from the paper as to why this is.
So where does that leave us?
Well, we have two recent RCTs that suggest sniffing alcohol swabs works. They both have similar flaws, but they are consistent. Until we know better it seems like a reasonable tool for the ED clinician who wants to reduce nausea quickly.
A second question would be whether this study is robust enough to suggest that we should swap Ondansetron for alcohol swabs. In that regard my answer is no. These are really interesting data, but it’s not robust enough to go that far (and in the authors defence they neither designed this study to do that, nor do they suggest it).
So. Get sniffing. Personally, I’m off to get a whiff of a rather nice Albarino I’ve got chilling in the fridge.
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
- Like us on Facebook
- Find out more about the St.Emlyn’s team