Diagnostic value of the Sepsis-3 standard for patients with sepsis: a multi-center prospective, observational clinical study / 中华危重病急救医学

ABSTRACT

Objective:To explore the value of Sepsis-3 standard in diagnosis of patients with sepsis.Methods:Patients who were infected or suspected of infection in intensive care unit (ICU) of six hospitals in Jiangsu Province from September 2017 to August 2018 were enrolled. They were divided into four groups group A was in accordance with Sepsis-1 and Sepsis-3, group B only met the Sepsis-1 standard, group C only met the Sepsis-3 standard, and both Sepsis-1 and Sepsis-3 standard did not match in group D. The age, gender, underlying disease, diagnosis and source of infection, vital signs within 24 hours of ICU, systemic inflammatory response syndrome (SIRS) score, sequential organ failure assessment (SOFA) score, acute physiology and chronic health evaluation Ⅱ(APACHEⅡ) score, quick sequential organ failure assessment (qSOFA) score, the length of ICU stay, total hospitalization time, 28-day mortality rate, etc. were recorded. The above collected data were compared and analyzed in groups, and the receiver operating characteristic (ROC) curves of each scoring standard were drawn and calculated. The area under the ROC curve (AUC), and the Youden index of each score was calculated to predict the optimal cut-off value of 28-day mortality in patients with sepsis and its corresponding sensitivity and specificity. Results:A total of 527 patients with infection or suspected infection were enrolled in the study, including 324 patients in group A, 113 patients in group B, 22 patients in group C, 68 patients in group D, and 28-day mortality were 38.9%, 17.7%, 31.8%, and 11.8%, respectively, and there was statistically significant difference among four groups ( P < 0.05). The SIRS scores of the A, B, C, D groups were 3 (1), 2 (1), 1 (0), 1 (0), APACHEⅡ scores were 17 (10), 11 (10), 15 (8), 12 (8), qSOFA score were 2 (1), 1 (1), 1 (1), 1 (2), SOFA scores were 8 (6), 1 (0), 7 (4), 1 (0), respectively, there were statistically significant differences among four group (all P < 0.05). Values of SOFA, qSOFA and SIRS scores were evaluated by ROC to predict the value of 28-day mortality. The results showed that AUC and 95% confidence interval of SOFA score was superior to qSOFA score and SIRS score [0.71 (0.66-0.76) vs. 0.59 (0.55-0.64), 0.57 (0.51-0.62), both P < 0.01]. According to the Youden index, the best cut-off values for the 28-day mortality of SOFA, qSOFA and SIRS scores for sepsis were 7, 2 and 2, respectively, and the sensitivity was 69.4%, 60.1%, 53.6%, the specificity was 61.8%, 76.2%, 51.1%, respectively. Conclusions:The Sepsis-3 standard is superior to the Sepsis-1 standard in the diagnosis and prediction of 28-day mortality in patients with sepsis. qSOFA can be used as an early tool for rapid screening of patients with high-risk sepsis in the ICU bedside.