Abstract
Previous studies have shown inconsistent prognostic accuracy for mortality with both quick sequential organ failure assessment (qSOFA) and the systemic inflammatory response syndrome (SIRS) criteria. We aimed to validate the accuracy of qSOFA and the SIRS criteria for predicting in-hospital mortality in patients with suspected infection in the emergency department. A prospective study was conducted including participants with suspected infection who were hospitalised or died in 34 emergency departments in Japan. Prognostic accuracy of qSOFA and SIRS criteria for in-hospital mortality was assessed by the area under the receiver operating characteristic (AUROC) curve. Of the 1060 participants, 402 (37.9%) and 915 (86.3%) had qSOFA ≥ 2 and SIRS criteria ≥ 2 (given thresholds), respectively, and there were 157 (14.8%) in-hospital deaths. Greater accuracy for in-hospital mortality was shown with qSOFA than with the SIRS criteria (AUROC: 0.64 versus 0.52, difference + 0.13, 95% CI [+ 0.07, + 0.18]). Sensitivity and specificity for predicting in-hospital mortality at the given thresholds were 0.55 and 0.65 based on qSOFA and 0.88 and 0.14 based on SIRS criteria, respectively. To predict in-hospital mortality in patients visiting to the emergency department with suspected infection, qSOFA was demonstrated to be modestly more accurate than the SIRS criteria albeit insufficiently sensitive.Clinical Trial Registration: The study was pre-registered in the University Hospital Medical Information Network Clinical Trials Registry (UMIN000027258).