Alfred researchers validate new sepsis criteria – with a catch

Sepsis is life threatening overreaction of the body to infection and it remains a common cause of death and disability worldwide.  It is one of the commonest reasons for someone to become critically ill and require treatment in an intensive care unit (ICU).  Until recently, sepsis had been defined as infection plus features of the systemic inflammatory response syndrome (SIRS), such as tachycardia, hypotension or tachypnea. In addition – the SIRS criteria have been used to gauge the severity of sepsis. But this definition is vague, making both clinical management of, and research on, patients with sepsis difficult.

In an effort to combat these problems, the European Society of Intensive Care Medicine (ESICM) and the Society of Critical Care Medicine (SCCM) convened the “Sepsis 3” taskforce to generate a new data-driven definition of sepsis.  They found that the Sequential Organ Failure Assessment (SOFA) score, instead of SIRS, was more accurate in identifying those at greatest risk of dying from infection.  As such, sepsis can now be more accurately defined as infection plus new organ dysfunction.   At the same time, “qSOFA” (Quick SOFA) was devised – a method of quickly screening patients for sepsis.  The 3 components of qSOFA are Hypotension, Altered mental status and Tachypnoea (HAT).  However these recommendations were derived from a relatively small cohort of critically ill patients.

Today in JAMA, researchers from The Alfred ICU, the Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcome and Resource Evaluation, and the School of Public Health and Preventive Medicine at Monash University, published the results of a large study which aims to validate the Sepsis 3 definition in an Australasian cohort.  One of the authors, A/Prof Andrew Udy Explains; “It is imperative that this new definition is tested in other healthcare settings, to assess its validity and applicability outside of North America.  The Australian and New Zealand healthcare systems are unique.”

 In a large Australasian cohort of patients admitted to ICU, The Alfred researchers confirmed that SOFA is a valid method of diagnosing sepsis.  However, they also showed that qSOFA isn’t that useful, in that it is less accurate in identifying patients who are likely to die.  The researchers used the ANZICS Adult Patient Database to look at all adult patients admitted to Australasian ICUs between 2000 and 2015 with infection (or suspected infection).  They then calculated SOFA and qSOFA scores on these patients, and compared them to the old SIRS criteria, to see if they were any better at identifying patients with the greatest risk of dying.  They used the same criteria as Sepsis 3 (an increase in 2 points in SOFA or a qSOFA of 2 or more) and compared this to the old definition (2 or more SIRS criteria) to see which was better at predicting mortality.

By reviewing a huge population of over 184,000 patients, the authors of the study found that the mean age of the patients was 63 years, the most common cause of sepsis was bacterial pneumonia (18%) and that the overall mortality from sepsis over the 15 year study period was 19%.

The researchers found that an increase of 2 or more in the SOFA score was better than SIRS at predicting mortality. In patients with a SOFA score of 2 or more, mortality was 20%, compared with 4% for those with a SOFA of zero.  For the clinician at the bedside, it emphasizes the significance of identifying organ dysfunction with infection, as this is robustly associated with a higher mortality risk for an individual patient.  This may also allow for better planning of treatment strategies and more informed discussions with patients and families.” Says Udy, putting their results into a clinical context.  This holds true for all patient groups, old or young, previously healthy or chronically ill” He adds.

qSOFA, on the other hand, wasn’t that much better than chance at predicting mortality in this cohort of patients – meaning that it would perform very poorly as a sepsis screening tool in the ICU. The researchers conclude that it shouldn’t be used to screen for or diagnose sepsis in this setting. Udy again; “Don’t rely on SIRS or qSOFA to diagnose sepsis within the ICU.”

Raith EP, Udy AA, Bailey M, McGloughlin S, MacIsaac C, Bellomo R, Pilcher DV, for the Australian and New Zealand Intensive Care Society (ANZICS) Centre for Outcomes and Resource Evaluation (CORE). Prognostic Accuracy of the SOFA Score, SIRS Criteria, and qSOFA Score for In-Hospital Mortality Among Adults With Suspected Infection Admitted to the Intensive Care Unit. JAMA. 2017;317(3):290-300. doi:10.1001/jama.2016.20328

2 comments to “Alfred researchers validate new sepsis criteria – with a catch”
  1. Just curious as to why capnography changes associated with infection wouldn’t be strongly looked at as a diagnostic agent. Wouldn’t metabolic changes show up faster than cascading organ failure??

    • Hi,

      I’m not sure anyone has used capnography as a screening tool or a diagnostic tool for sepsis. I suspect it would be neither sensitive nor specific enough. Infection on it’s own wouldn’t result in any noticeable change in ETCO2, and while sepsis may (for example a fall in ETCO2 as a result of respiratory compensation for metabolic acidosis) this could be caused by many other things. Furthermore, this paper shows that changes in vital signs (RR, BP etc) as measured by SIRS/qSOFA may not be adequate for diagnosing sepsis. Changes in ETCO2 in an awake patient would be very likely to be accompanied by changes in RR. So for the moment it looks like this is the best we’ve got!


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.