Feb 222017

Sepsis is sneaky. Physicians, nurses, and epidemiologists struggle to accurately identify patients with sepsis in the emergency department, hospital ward, and in data sets. The so-called SIRS criteria were abandoned as insensitive and nonspecific in the most recent iteration of sepsis care. Sepsis is instead now defined as "life-threatening organ dysfunction caused by a dysregulated host response to infection."

It's an elegant description, but how do we detect sepsis in real clinical situations? There is simply no gold standard for the diagnosis of sepsis. Roughly half of patients who seem to obviously have severe sepsis nevertheless have negative bacteriologic cultures. Lacking a reliable gold standard test, it's impossible to confidently evaluate the precision any test or set of diagnostic criteria for sepsis.

qSOFA: A New Tool to Screen for Sepsis

Instead of using SIRS criteria, physicians are now advised to identify organ dysfunction from sepsis using a scoring system.

The Sequential Organ Failure Score (SOFA) is officially advised for this purpose, but sepsis task force members recognized this cumbersome tool is unlikely to be used consistently by busy physicians. The task force therefore also recommends the wonderfully simple qSOFA score:

  • 1 point each for encephalopathy (Glasgow Coma Scale <15), tachypnea (respiratory rate ≥ 22/min), or systolic blood pressure under 100 mm Hg.
  • Patients with qSOFA ≥2 are at risk for sepsis.

"At risk for sepsis?" What does that even mean? Can we just get a diagnosis here? Again, absent a gold standard, test performance characteristics cannot be calculated for any sepsis diagnostic tool. Instead, researchers have studied the tools' empiric ability to predict outcomes among hospitalized patients with high likelihood of infections.

In large data sets analyzed retrospectively, patients who had qSOFA scores ≥2 in the emergency department or medical ward who also had suspected infection (i.e., got antibiotics or had positive culture results), were much more likely to have bad outcomes (death or ICU stays >3 days) than patients with qSOFA = 1. qSOFA predicted these outcomes as well as the cumbersome SOFA or LODS systems.

What Does the Surviving Sepsis Campaign Say about qSOFA?

The Surviving Sepsis Campaign twists its tongue a bit around qSOFA. The SSC endorses qSOFA, kind of, saying

Patients with sepsis (formerly called severe sepsis) should still be identified by the same organ dysfunction criteria (including lactate level greater than 2 mmol/L). Organ dysfunction may also be identified in the future using the quick Sepsis-Related Organ Failure Assessment (qSOFA) ... [emphasis added]

and validates qSOFA as

a tool for identifying patients at risk of sepsis with a higher risk of hospital death or prolonged intensive care unit (ICU) stay both inside and outside critical care units...

But also says

qSOFA does not define sepsis.

But didn't you just say sepsis is organ dysfunction caused by a dysregulated host response to infection? And that qSOFA is a reasonable way to identify organ dysfunction?

If this sounds confusing or circular, don't blame the SSC (or PulmCCM). When there's no gold standard to diagnose a condition with variable manifestations, its definition inevitably becomes some combination of criteria; in this case, organ dysfunction (however determined) and a strong-enough suspicion of infection.

SSC adds that its new sepsis definitions "recommend using a change in baseline of the total SOFA score [not qSOFA] of two or more points to represent organ dysfunction."

qSOFA's Predictive Ability Seems As Good as Anything Else

A new study published in JAMA adds support for the use of qSOFA as a quick, easy, and useful screening tool for patients at risk of dying from sepsis. 879 mostly elderly patients in Europe were followed after presenting to hospitals with suspected infection (about half were diagnosed with respiratory infections).

Eight percent died in the hospital, and qSOFA helped predict who: hospital mortality was 3% for those with qSOFA of 1, but 24% for those with a qSOFA ≥2. Interestingly, adding lactate to the model did not improve the predictions. Data were incomplete on 14% of patients, limiting the conclusions' strength.

The apparent simplicity of qSOFA belies the complexity and protean presentations of sepsis. But its use should be an improvement over the SIRS criteria, which consistently led physicians to underdiagnose sepsis, exaggerating the negative predictive value of the absence of fever, tachycardia, or leukocytosis. Many patients with sepsis lack these findings.

qSOFA doesn't seem to work well in the ICU, where the full SOFA score may be more accurate. This has been observed before, and a new study in Australia and New Zealand, also in JAMA, echoed the finding in a data analysis of ~185,000 ICU admissions. SOFA itself did not perform well, with area under the curve of 0.75 (qSOFA was 0.60) to accurately predict hospital mortality.

Read more:

Prognostic Accuracy of Sepsis-3 Criteria for In-Hospital Mortality Among Patients With Suspected Infection Presenting to the Emergency Department. JAMA. 2017;317(3):301-308.

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.

Get our weekly email update, and explore our library of practice updates and review articles.

PulmCCM is an independent publication not affiliated with or endorsed by any organization, society or journal referenced on the website. (Terms of Use | Privacy Policy)


Simple qSOFA score predicts sepsis as well as anything else