The content you’re about to read or listen to is at least two years old, which means evidence and guidelines may have changed since it was originally published. This content item won’t be edited but there will be a newer version published if warranted. Check the new publications and curriculum map for updates


Authors: Becky Maxwell, Chris Connolly / Codes: IP3, SLO3 / Published: 14/09/2016

There has been a lot of talk about sepsis over the last year, in fact I heard someone recently saying they had “sepsis” fatigue, there have been so many debates in the FOAMed world this year – starting with SMACCChicago last year, to changes in definitions etc. that we run the risk of people becoming confused and bewildered about what is actually going on the world of sepsis.

NICE guidelines 2016 have been produced to tackle the morbidity and mortality associated with this condition. The algorithms that NICE have produced look a bit complicated and don’t fit easily on the back of a card for you to carry in your pocket. They have already been summarised nicely by Simon Laing in his new podcast “The Resus Room’ so we don’t want to go over all of these again – there’s a lot of parameters etc. to get your head around – for us what we want to do is discuss the practical implications of the Guidelines and what they actually mean.


Life threatening organ dysfunction due to (dysregulated) host response to infection.

Septic shock is persisting hypotension requiring vasopressors to maintain a mean arterial pressure (MAP) of 65 mmHg or more and having a serum lactate level of greater than 2 mmol/l despite adequate volume resuscitation

In our opinion this definitions are easier the idea is to separate sepsis from infection i.e. those who need oral abx or a short course of IV then can go home versus those that will get really sick (old “severe sepsis”)

SO SIRS are out – probably a good thing as were shown actually to have low specificity i.e. quiet a few people had these and weren’t septic

Initially this resulted in qSOFA earlier this year. To remind us that was: Hypotension (<100)/altered mental status/tachypnoea (>22) – note these definitions have not been adopted by NICE – huge debate over them when they came out. There is a nice blog post on St Emlyns summarising these

NICE – want you to think about – could this person have Sepsis??  

Identification of risk factors:

Good to make you think about individual groups and sepsis

IVDU – a sick group of patients in whom it is  hard to apply ‘standard’ age or disease based screening tools, it always find it best to assume they have the worst version of a disease. They’re a difficult group of patients to manage, get a good history from, get reliable observations, IV access, convince to stay in their hospital bed etc.

Some of the categories are fairly ‘obvious’ such as those on imunno-suppressants or those who are immune-suppressed by their diseases eg DM, splenectomy etc.

NICE have included a good reminder that women who have recently had a baby, or a miscarriage are also at particularly high risk, again remembering to take histories on gestational disease such as DM

Risks stratifying:

High risk: again seems fairly sensible – those who are hypoxic, tachycardic or hypotensive with skin changes of poor perfusion = high risk of death. Interesting those deemed medium risk – are those ‘acute change in functional status’ I read that as those who are ‘off legs’.

In kids as always theres a list of respiratory and heart rate changes that make you medium or high risk and I think these are a useful broad tool but it is difficult  to imagine the inter-observer reliability for a counted respiratory rate in a poorly 5 year old and as with all of these markers need taking into account the whole picture in front of you.

Massively pragmatic statement ‘DO NOT USE A PERSONS TEMPERATURE AS A SOLE PREDICTOR OF SEPSIS’. Remember to ask about history of fever and rigors. Remember those who will not mount a temp -old/immunosuppressed/v young.

What NICE are doing with this risk stratification is sensible – picking up the patients before they “fall off their perch” so you can make clinicians no matter where they are based (ward/EDetc.) aware of the patient get an early review and prevent deterioration. It is really important for the nurses to be able to identify those at high risk and flag them up in the busy ED as well – the various boxes in this guideline make this difficult for the Triage nurse  not practical to have this stuck up in Triage and have he nursing staff go through it. Is it worthwhile using NEWS in this scenario? The getting the Traige nurse to flag up appropriate patients to medical practitioners – it may mean you get  patients flagged who don’t have sepsis but what it allows us to do is prioritise and prescribe early antibiotics.


NICE recommends that anyone with suspected sepsis and one high risk feature – do the bloods, give the antibiotics and discuss with a consultant. Now I don’t know about you but I would not expect to be rung at 0300 to be told theres someone with sepsis and fast AF in my resus room. It doesn’t seem to give a time frame on that discussion but I agree they need timely senior review.

Referral of all those with a lactate of >4 or a BP of <90 to critical care. This needs clarification –  are we talking single BP readings less than 90. Or <90 despite good going fluid resuscitation. Again caveats – liver disease patients who probably have a high lactate anyway (you remember the Cori cycle right?) I think this is the same with metformin and renal failure.

Whilst on the subject of lactate it may be worth reminding ourselves that lactate increases are NOT due to anaerobic metabolism or low oxygen delivery.  It is largely driven by endogenous adrenaline stimulating aerobic glycolysis via beta-2 adrenergic receptors. Debate on whether it is a better measure of sepsis rather than MAP or urine output. Listen to Marik’s talk on this from SMACC he does a much better job of explaining this than I can!

There is also a recommendation of a physical consultant review if lactate doesn’t clear by 20% in 1 hour, BP persistently <90, reduced level of consciousness – dies this mean all those who have a GCS of 14 because they’re a bit delirious? Again no clear guidance on the time frame for this review. If this review is to be timely then this is surely a call of arms for us all to be on full shift!


Fluid wise its what we expect. Give a crystalloid bolus, in adults 500mls, and kids 20mls/kg. (refresh on the RCEM NICE IVF in kids podcast – not that we are biased!) and then repeat if no response. If no response to a second bolus then get a consultant review. Not clear what they mean by response, for me it’s a whole patient picture.

The statement to “consider IV fluids” if Lactate < 2 is a great step forward – sepsis is primarily not a volume-depleted state and recent evidence demonstrates that most septic patients are poorly responsive to fluids. Almost all of the administered fluid ends up in the tissues, resulting in severe oedema in vital organs increasing the risk of organ dysfunction. This is a step forward in the right direction that not all patients need lots of fluid…


No mention of vasopressors which is a little bit disappointing in the Guideline. Would have been nice to have some guidance on this especially with the Guideline advocating early ITU referrals. Lots of stuff out there which would have been good to have guidance on…

How early should we start vasopressors?

Good post on PUlmCrit on this, stressing that hypotension in sepsis in multifactorial – “including venodilation, arterial dilation, intravascular volume depletion” etc. giving fluid addresses the volume depletion but ignores the rest – in order to have what they describe as a “balanced approach” you need to add in a vasopressor and taper it to the MAP, response to fluids. Worth thinking about – need ITU colleagues on board and would be nice to have some guidance on this.

Overall, a good guideline which has given us lots to think about. ED and ITUs now just need to sit down and agree on how to implement this best in individual departments.