Background: In the new Sepsis-3 definition, sepsis is defined as "life-threatening organ dysfunction due to a dysregulated host response to infection." We tested the predictive validity of the systematic inflammatory response syndrome (SIRS) criteria in patients in the Sepsis-3 cohort.
Methods: Among 1243 electronic health records from 1 January to 31 December 2015 at Sichuan University West China Hospital, we identified patients with sepsis and septic shock according to the Sepsis-3 definition and divided them into 2 subsets: SIRS-positive and SIRS-negative. We compared their characteristics and outcomes as well as the predictive validity of the SIRS criteria for in-hospital mortality.
Results: Of the 1243 patients, 631 were enrolled. Among these, 538 (85.3%) patients had SIRS-positive sepsis or septic shock, 168 (31.2%) of whom died, and 93 (14.7%) had SIRS-negative sepsis or septic shock, 20 (21.5%) of whom died (p = 0.06). Over a 1-year period, these groups had similar characteristics and changes in mortality. Among patients of the Sepsis-3 cohort admitted to the intensive care unit, the predictive validity for in-hospital mortality was lower for the SIRS criteria (area under the receiver operating characteristic curve [AUROC], 0.53; 95% confidence interval [95% CI], 0.49-0.57) than for the sequential (sepsis-related) organ failure assessment (SOFA) criteria (AUROC, 0.70; 95% CI, 0.66-0.74; p ≤ 0.01 for both). The SIRS score had poor predictive validity for the risk of in-hospital mortality.
Conclusions: In this cohort study of the new Sepsis-3 definition, we found that the SIRS criteria are weaker than the SOFA criteria with respect to their predictive efficacy for in-hospital death.
Keywords: Critically ill patients; Intensive care unit; Sepsis; Sequential (sepsis-related) organ failure assessment; Systematic inflammatory response syndrome.