Published in:
01-12-2020 | Septicemia | Research
External validity of Adult Sepsis Event’s simplified eSOFA criteria: a retrospective analysis of patients with confirmed infection in China
Published in: Annals of Intensive Care | Issue 1/2020
Login to get accessAbstract
Background
The US Centers for Disease Control and Prevention (CDC) recently released simplified eSOFA organ dysfunction criteria of Adult Sepsis Event for sepsis surveillance in the US. Our study aimed to compare the prevalence, characteristics, and outcomes of sepsis patients identified by eSOFA criteria versus Sequential Organ Failure Assessment (SOFA) Score (Sepsis-3) and assess the external validity of eSOFA criteria in China.
Methods
We conducted a retrospective cohort study of adult residents of Yuetan Subdistrict, Beijing, China, who were hospitalized from July 1, 2012 to June 30, 2014. Among patients with infection, sepsis was identified if there was a concurrent rise in SOFA score by 2 or more points (Sepsis-3) or the presence of 1 or more eSOFA criteria: vasopressor initiation, mechanical ventilation initiation, doubling in creatinine, doubling in bilirubin to 2.0 mg/dL or above, 50% or greater decrease in platelet count to less than 100 cells/μL, or lactate equal to or above 2.0 mmol/L. Areas under the receiver operating characteristic curves (AUROCs) for in-hospital mortality were compared between sepsis patients detected by the two criteria, adjusting for baseline characteristics.
Results
Of 1716 hospitalized patients with infection, 935 (54.5%) met Sepsis-3 criteria, 573 (33.4%) met eSOFA criteria, while 475 (27.7%) met both criteria. Demographic and clinical characteristics of sepsis patients meeting Sepsis-3 or eSOFA criteria were similar. In-hospital mortality was higher with eSOFA criteria versus Sepsis-3 (46.6% vs. 32.0%, p < 0.001). eSOFA criteria had high PPV (82.9%), but low sensitivity (50.8%) for the diagnosis of Sepsis-3. Patients meeting both criteria had the highest in-hospital mortality rate (52.8%, all p < 0.001), while patients who only met eSOFA criteria had higher mortality rate than those meeting Sepsis-3 alone (16.3% vs. 10.4%, p = 0.097). The predicted probability for in-hospital mortality was higher with eSOFA criteria versus Sepsis-3 (AUROC 0.830 vs. 0.795, p = 0.001) adjusting for baseline characteristics.
Conclusions
The CDC Adult Sepsis Event’s eSOFA criteria identify a smaller, more severely ill cohort of sepsis patients with similar demographic and clinical characteristics as the more complex Sepsis-3 SOFA score. These results suggest similar performance of eSOFA criteria across diverse populations, with low sensitivity and high specificity for the diagnosis of Sepsis-3.