By Kathryn Radigan, MD, MSc

Attending Physician, Division of Pulmonary and Critical Care, Stroger Hospital of Cook County, Chicago

Dr. Radigan reports no financial relationships relevant to this field of study.

SYNOPSIS: In hospitalized patients who deteriorated from suspected infection, Sepsis-3 septic shock criteria predicted in-hospital mortality better than systemic inflammatory response syndrome-based criteria.

SOURCE: Fernando SM, Reardon PM, Rochwerg B, et al. Sepsis-3 septic shock criteria and associated mortality among infected hospitalized patients assessed by a rapid response team. Chest 2018;154:309-316.

Since sepsis remains the leading cause of in-hospital death, appropriate recognition, prognostication, and treatment are the most important components to improving survival. Rapid response teams (RRTs), groups trained specifically to respond to deteriorating patients, have been a major component in these efforts. Identification of sepsis patients has been based on the presence of the systemic inflammatory response syndrome (SIRS) since 1991. However, the authors of the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) criteria recently proposed using the Sequential Organ Failure Assessment (SOFA) score and the quick Sequential Organ Failure Assessment (qSOFA) score for detection, risk stratification, and prognostication of patients.

To optimize the timing of evaluation, the more simplified qSOFA criteria has been popularized. Two out of the three qSOFA criteria (low blood pressure [systolic ≤ 100 mmHg], higher respiratory rate [≥ 22 breaths/min], and altered mental status [Glasgow Coma Scale score < 15]) with suspicion for infection indicates an elevated risk of death. The authors of Sepsis-3 also identified septic shock as the initiation of vasopressors to maintain a mean arterial pressure of ≥ 65 mmHg with a serum lactate > 2.0 mmol/L after adequate fluid resuscitation.

On the other hand, SIRS-based sepsis criteria include two out of the four following criteria: fever > 38°C or < 36°C, heart rate > 90 beats/minute, respiratory rate > 20 breaths/minute or arterial PaCO2 < 32 mmHg, or abnormal white blood cell count in patients with suspected infection. The authors of the SIRS-based septic shock criteria expanded the definition to include persistent hypotension (systolic blood pressure < 90 mmHg) requiring administration of vasopressors or evidence of perfusion abnormalities (including acidosis, lactic acid, altered mental state, or oliguria).

To compare the prognostic accuracy of the SIRS-based and Sepsis-3 criteria for predicting in-hospital mortality among hospitalized patients with suspected infection and who received an RRT call, Fernando et al prospectively collected registry data from two different academic hospitals within the Ottawa Hospital network system from 2012-2016. Patients who were administered antibiotics with cultures were suspected of infection. If staff drew cultures first, it was necessary to administer antibiotics within 72 hours. If staff administered antibiotics first, it was necessary to send cultures to the lab within 24 hours. The authors excluded patients with incomplete demographic/outcome data or those already with scheduled RRT follow-up.

Results revealed that more of the hospitalized patients met the SIRS-based septic shock criteria (n = 545) compared to the Sepsis-3 septic shock criteria (n = 418). Compared to patients who met the SIRS-based septic shock criteria, patients meeting the Sepsis-3 septic shock criteria demonstrated higher in-hospital mortality (40.9% vs. 33.5%; P < 0.001), ICU admission (99.5% vs. 89.2%; P < 0.001), and discharge rates to long-term care (66.3% vs. 53.7%; P < 0.001).

As for the prediction of in-hospital mortality, the sensitivity was higher for the SIRS criteria compared to qSOFA (91.6% vs. 64.9%), but specificity was higher for qSOFA compared to the SIRS criteria (92.2% vs. 23.6%). For hospitalized patients with deterioration from suspected infection, those who met Sepsis-3 septic shock criteria were at a higher risk of in-hospital mortality compared to those who met the SIRS-based criteria. Therefore, Sepsis-3 may be the preferred method for prognostication, and the SIRS-based criteria may be the preferred method to screen patients for consideration of ICU admission.


Severe sepsis accounts for almost 10% of all deaths.1 Interestingly, this study highlighted that although the Sepsis-3 criteria may predict in-hospital mortality more accurately, the SIRS-based criteria may be more helpful as a screening tool. This particular study and a recently published meta-analysis affirm that qSOFA has poor sensitivity, but a more reasonable specificity.2 Since SIRS criteria has low specificity with higher sensitivity, it will lead to fewer missed patients but many more false positives. Meanwhile, qSOFA leads to more missed patients but significantly fewer false positives. Although neither qSOFA nor SIRS is an ideal screening tool, the data do not support abandoning the traditional SIRS criteria for the more novel qSOFA criteria. Ideally, it would be best to use these tools together to identify at-risk patients and those with a high likelihood for deterioration.

qSOFA was extrapolated from the full SOFA score to create a mnemonic that can help clinicians remember the components effectively: THAM (tachypnea, hypotension, altered mentation). However, application of these simplified criteria may not be as simple as the mnemonic.3 Depending on the quality and dedication of nursing staff, the accuracy of the Glasgow Coma Scale may be compromised. Freund et al studied the applicability of qSOFA compared to the full SOFA score. Despite its reliance on only three parameters, 14% of recruited patients were excluded from the analysis due to missing values.4 Furthermore, it is known that the Glasgow Coma Scale is used accurately by experienced and highly trained users. Still, inexperienced users make consistent errors, limiting the reliability and accuracy of this measurement.5 Unfortunately, the error rates were highest at the intermediate levels of consciousness, for which the recognition of changes in mental status often is most critical. Although the Fernando et al findings support the continued use of the Glasgow Coma Scale by appropriately trained and seasoned personnel, the results raise doubts about the reliability of the scale when untrained or inexperienced staff use it.

Although there is apprehension with the accuracy of the Glasgow Coma Scale for qSOFA, there remains concern about using respiratory rate as a major criterion for both the SIRS-based and Sepsis-3 criteria. Unlike other vital signs, this is a significant concern, as respiratory rate is the only vital sign that is not measured by a machine and tends to be the most neglected. Although measuring respiratory rate is a simple task, it is rarely completed with this standard in mind. In a different study, Mukkamala et al asked medical students to evaluate patients’ respiratory rates within an hour of nursing staff recording those rates. For respiratory rates > 23 breaths/min, nurses were correct only 15% of the time.6 As the respiratory rate is used for both the SIRS-based and Sepsis-3 criteria, this may be one of the issues that make both criteria less than ideal.

Sepsis-3 criteria may be the favored method for prognostication, whereas SIRS-based criteria may be the preferred method to screen patients for consideration of ICU admission. Future studies are necessary to continue to explore the benefits of qSOFA and potentially reveal a more precise and reliable screening tool. Most importantly, it is paramount to remember that neither set of criteria is diagnostic. Using clinical judgment along with these guides remains the ideal approach.


  1. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303-1310.
  2. Fernando SM, Tran A, Taljaard M, et al. Prognostic accuracy of the quick sequential organ failure assessment for mortality in patients with suspected infection. Ann Intern Med 2018;169:264-265.
  3. Vincent JL, Grimaldi D. Quick sequential organ failure assessment: Big databases vs. intelligent doctors. J Thorac Dis 2016;8:E996-E998.
  4. Freund Y, Lemachatti N, Krastinova E, et al. Prognostic accuracy of Sepsis-3 criteria for in-hospital mortality among patients with suspected infection presenting to the emergency department. JAMA 2017;317:301-308.
  5. Rowley G, Fielding K. Reliability and accuracy of the Glasgow Coma Scale with experienced and inexperienced users. Lancet 1991;337:535-538.
  6. Mukkamala SG, Gennings C, Wenzel RP. R = 20: Bias in the reporting of respiratory rates. Am J Emerg Med 2008;26:237-239.