The last few years have been busy when it comes to sepsis. We have seen the National Confidential Enquiry into Patient Outcome and Death (NCEPOD), new definitions from the Sepsis-3 group, and a set of guidelines from the National Institute of Health and Care Excellence (NICE). Despite all this work, sepsis remains difficult to describe and diagnose. Though showing a trend in the right direction, the response to critical therapies is still not as rapid as we would like it to be for our loved ones.
At least 150 000 patients in the UK are affected by sepsis every year. At the current rate, it is predicted that this number will at least double every 10 years. Sepsis has high mortality rates (up to 40%); it represents a huge burden to the NHS, consuming a significant proportion of intensive care bed-days, and costs an estimated £2.5 billion per annum. We would all agree, that we can and should do better – pre-hospital (PH) practitioners have a huge amount to contribute to decreasing the mortality and morbidity burden from sepsis, and we are likely to see this realised over the next few years.
Definitions:
In February 2016, the International Consensus Definitions for Sepsis Task Force published recommendations for Sepsis-3. Major changes from earlier definitions included dropping the Systemic Inflammatory Response Syndrome (SIRS) criteria and introducing sequential Sepsis-related Organ Failure Assessment (SOFA) scoring. Sequential scoring will be impossible in some environments (including PH care), in which case the group recommends a quick-SOFA (qSOFA). qSOFA supports the diagnosis of sepsis in patients with suspected infection when two of the following criteria are present:
This new sepsis definition has not been prospectively validated and its PH test characteristic is unknown. As such, NICE and the UKST are recommending the National Early Warning Score (NEWS) and the suspicion of infection as a trigger to screen for sepsis.
Clinical suspicion, diagnosis and point-of-care testing:
Considering sepsis in all undifferentiated medical patients is the primary intervention available to pre-hospital practitioners, which will reduce mortality from sepsis. Using electronic observations systems utilising standardised composite scoring systems (e.g. NEWS) with built in trigger alerts will greatly assist with this process.
Distinguishing sepsis from severe infection is unimportant – both conditions will benefit from early, experienced review and timely intervention. Sepsis mimics like acute asthma or pancreatitis should be considered, though those within these diagnostic categories who are truly “sick” will often receive sepsis-like treatments alongside their diagnosis-specific therapy.
Many ambulance services have considered point-of-care testing to help in the early diagnosis of sepsis. Lactate appears useful with a strong correlation between hyperlactaemia and mortality across a range of pathologies, but lacks the specificity in isolation to confirm sepsis. A low/normal lactate may be useful in identifying those with moderate or mild infection for whom a trial of community therapy may be appropriate.
Point-of-care (POC) C-reactive protein (CRP) and pro-calcitonin tests are currently available, but these lack the test characteristics to be useful in isolation. Composite risk stratification using a combination of one or more POC tests combined with physiology may represent a useful rule-in/rule-out; however, it should be subject to full clinical and health-economic analysis prior to implementation.
Oxygen, fluids and antibiotics:
High flow oxygen was previously recommended in sepsis. This is no longer the case. The recommendation now is to aim for “normoxia” where possible, titrating oxygen delivery to what we presume are normal oxygen saturations for the patient. This requires investment in fixed performance oxygen delivery devices such as venturi masks.
Patients with sepsis are often under fluid resuscitated. Disproportionate caution in patients labelled with heart failure exacerbates this, and leads to harm. Delivering rapid boluses of crystalloid (e.g. 0.9% sodium chloride) with frequent (re)assessment is sensible. Evaluations are underway for (relatively) low tech/inexpensive cardiac output monitoring devices which will guide therapy in the emergency and PH environment.
Early antibiotics are seen as the panacea for the early management of those unwell with infection. It is unclear how early is early, but given the variations at point of presentation, the significant time associated with the PH phase and the PH/hospital interface, delivery of antibiotics by PH providers should and is being investigated. Increased use of antibiotics in the PH environment is inevitable – but the devil is in the detail – which antibiotics, to which severity patients, with which conditions – and (of course) at what cost!
Pre-alert and handover:
There is strong evidence that this improves the onward care for patients with sepsis. High acuity patients with suspected sepsis should be pre-alerted. Sepsis, if suspected, should be mentioned by name at handover – this decreases the time to senior review, and if required, antibiotic and fluid resuscitation.
Going forward:
In-hospital sepsis is in stasis. New definitions, guidelines and recommendations will not result in the care revolutions they were intended to inspire. The PH environment is ripe for delivering excellent evidence based sepsis care, and where we don't know the answers, delivering robust trials that will lead us to them.