Assessment of mortality by qSOFA in patients with sepsis outside ICU: A post hoc subgroup analysis by the Japanese Association for Acute Medicine Sepsis Registry Study Group.

Quick sequential organ failure assessment (qSOFA) was proposed in the new sepsis definition (Sepsis-3). Although qSOFA was created to identify patients with suspected infection and likely to have poor outcomes, the clinical utility of qSOFA to screen sepsis has not been fully evaluated. We investigated the number of patients diagnosed as having severe sepsis who could not be identified by the qSOFA criteria and what clinical signs could complement the qSOFA score. This retrospective analysis of a multicenter prospective registry included adult patients with severe sepsis diagnosed outside the intensive care unit (ICU) by conventional criteria proposed in 2003. We conducted receiver operating characteristic (ROC) analyses to assess the predictive value for in-hospital mortality and compared clinical characteristics between survivors and non-survivors with qSOFA score ≤ 1 point (qSOFA-negative). Among 387 eligible patients, 63 (16.3%) patients were categorized as qSOFA-negative, and 10 (15.9%) of these patients died. The area under the ROC curve for the qSOFA score was 0.615, which was superior to that for the systemic inflammatory response syndrome score (0.531, P = 0.019) but inferior to that for the SOFA score (0.702, P = 0.005). Multivariate logistic regression analysis showed that hypothermia might be associated with poor outcome independently of qSOFA criteria. Our findings suggested that qSOFA had a suboptimal level of predictive value outside the ICU and could not identify 16.3% of patients who were once actually diagnosed with sepsis. Hypothermia might be associated with an increased risk of death that cannot be identified by qSOFA.