A Nationwide Comparison Between Sepsis-2 and Sepsis-3 Definition in Japan

• Published in: Journal of intensive care medicine 2020-12, Vol.35 (12), p.1389-1395
• Main Authors: Takauji, Shuhei, Hayakawa, Mineji, Fujita, Satoshi
• Format: Article
• Language: English
• Subjects: Hospital Mortality; Humans; Intensive Care Units; Japan - epidemiology; Sepsis - classification; Sepsis - diagnosis; Sepsis - mortality; Shock, Septic - diagnosis; Shock, Septic - mortality
• ISSN: 0885-0666; 1525-1489
• DOI: 10.1177/0885066618823151

Background: Currently, it remains controversial whether the Sepsis-3 definition provides the most appropriate criteria for clinical use. The purpose of this study was to compare between the Sepsis-2 and Sepsis-3 definitions using Japan’s nationwide registry. Methods: Data were obtained from a multicenter registry conducted at 42 intensive care units (ICUs) throughout Japan, in which patients received treatment for severe sepsis or septic shock between January 2011 and December 2013. Results: A total of 2797 patients diagnosed using the Sepsis-2 criteria were included in the present study. These patients were categorized into “Severe sepsis” (n = 1154) and “Sepsis-2 shock” (n = 1643) groups. Among the “Sepsis-2 shock” group, patients who did not meet the Sepsis-3 criteria for septic shock were categorized into the “Sepsis-2 shock-only” (n = 448, 27.3%) group, while patients who met the Sepsis-3 criteria for septic shock were categorized into “Sepsis-3 shock (n = 1195, 72.7%)” group. The ICU mortality in the “Sepsis-3 shock” group, “Sepsis-2 shock-only” group, and “Severe sepsis” group was 28.5%, 10.9%, and 14.1%, respectively. We observed no significant difference between the “Severe sepsis” and “Sepsis-2 shock-only” groups in terms of in-hospital survival (P = .098), while the “Sepsis-3 shock” group had the highest in-hospital mortality rate (P < .001). In a multivariate logistic regression analysis, liver insufficiency and immunocompromised status were independent prognostic factors in the “Sepsis-2 shock-only” group. In contrast, chronic heart disease and chronic hemodialysis were independent prognostic factors in the “Sepsis-3 shock” group. Conclusions: The ICU mortality of the “Sepsis-2 shock-only” group was significantly low. Besides septic shock diagnosed by the Sepsis-3 definition selects patients with more severe cases of sepsis among the “Sepsis-2 shock” group.