Its the start of winter here in NHS land. Twitter and the mainstream media is full of stories of over-run departments, patients everywhere and staff working flat to the point of breaking. But don’t let this distract you from the latest in EBM. Or from the fact we all know we work in one of the greatest jobs in medicine. At the front door, identifying the sick amongst a sea of not sick, risk stratifying, seeing and treating. All the stuff we got into this game for.
I’ve decided to make a few changes to the way the blog runs, with (hopefully) more frequent posts with slightly shorter reviews. Let me know what you think….
One of the big pushes in UK EM is the early recognition and treatment of sepsis.
There are systems in place to help us all spot it.
We all know about the ‘sepsis 6’…..
Fluid that’s important right….?
But how much?
A few years ago I remember listening to Paul Marik talking about fluid in sepsis. This was around the time the FEAST trial came out showing benefit of a restricted fluid resuscitation in kids with sepsis. (be mindful that whilst this trial was well conducted there are issues around external validity, some of which we’ll reiterate here)
This paper caught my eye this week via a quick-review on EMlitofNote – check out their review here.
Title of Paper: Effect of an Early Resuscitation Protocol on In-hospital Mortality Among Adults With Sepsis and Hypotension.
Published: JAMA 2017
Population: Adults aged >18 with sepsis and hypotension. (At a Zambian University Hospital)
Intervention: Protocol driven fluid resuscitation of 2 litres in an hour, followed by a further 2 litres in 4 hours, with protocolised administration of IV Dopamine.
Comparitor: Clinician directed IV fluid, PRBC and Vasopressors.
Outcome: Primary – In hospital mortality.
Now protocolised sepsis care has been studied in 3 large multi centre trials in the not too distant pass with the PROMISE, PROCESS and ARISE trials showing no benefit to protocol delivered care (in line with the original Rivers work). Check out St-Emlyns review of the PROMISE trial and then continue to explore this fascinating area of the medical literature!
Now this wasn’t strictly protocol and goal driven care. This was ‘everyone gets 4 litres of crystalloid’ and then if MAP remained at <65mmHg after 2 litres then Dopamine was started at 10 μg/kg/min. So a similar but not the same.
The control group got a combination of treatment guided by the clinician. 40% got no fluid bolus at all. Nearly all had some maintenance fluid over the following 24 hour period.
What did they find?
A difference in hospital mortality of 15.1% in favour of the control group. An absolute risk increase of 15.1%. Or a Number Needed to Harm of 6.6 (95% CI 3.5-50 – based on the limits expressed in the absolute difference). Now these are wide limits. This was a small study. It had 80% power to detect a 20% difference and thats going to affect the accuracy of the results.
What about the patients?.
So this is a BIG limitation if you’re considering applying the findings here to your patients.
Firstly inclusion criteria were reasonable. Adults with sepsis and hypotension (the used 2x SIRS criteria plus SBP <90mmHg or MAP <65mmHg). I was interested to see they excluded (along with a long list of other exclusions) those with oxygen saturations of <90% and a respiratory rate >40. I cant quite piece together why to be honest. Lots of my unwell sepsis patients have pneumonia, and hypoxia.
Patients were similar at baseline, but young (mean age of 39ish). I think of the vast majority of our sepsis patients with low BP. They’re usually more elderly than that.
They also had a 90% (ish) HIV rate amongst the included patients. I’ve not seen a HIV positive patient this year (seriously we have a ridiculously low HIV rate in Sheffield!).
The mean Hb across the groups was 7.8g/DL (ahhhhhh the simple pleasure of a decimal point in a Hb reading we are now deprived of here….it would be 78 in ‘new money’)
The other stark difference was the lack of ICU availability to those in the study. 99.5% of the patients were cared for on a ward without the availability of mechanical ventilation due to resource limitations in the hospital. That’s not a criticism of the care the patients got, nor a criticism of the system, but it is a different type of care than that offered to seriously unwell sepsis patients in the UK (and a lot of the English speaking world).
In this cohort of patients the combination of 4 litres of crystalloid in 5 hours, plus dopamine as the vasopressor of choice may increase in hospital mortality with an estimated NNH of 6.6. For me in the UK this paper is interesting but is unlikely to change my current practice (which, granted, is already NOT aligned to the intervention arm here), however I think that given there are millions of patients similar to these across the world it is likely to be significant for many many Emergency Physicians and Medical staff across the globe.
Thanks for taking the time to read the blog.
Keep fighting the good fight
Winter is coming……