[Prognosis analysis of multi-indicator combined with sequential organ failure assessment in patients with sepsis]

To explore the prognostic value of early multiple detection indicators in combination with sequential organ failure assessment (SOFA) in sepsis patients.

A retrospective analysis was conducted. Patients with sepsis admitted to the department of critical care medicine of Huanggang Central Hospital of Yangtze University from May 2020 to May 2022 were selected as the research subjects. Coagulation indicators, inflammatory factors, blood routine, liver and kidney function, and blood gas analysis were collected at admission. Organ dysfunction was assessed based on the SOFA score within 24 hours after admission. Patients were divided into a survival group and a death group according to the outcome of 28 days in ICU. Differences in the above indicators between the two groups were compared. Multifactorial Logistic regression analysis was used to analyze prognostic factors of 28-day mortality in sepsis patients. Receiver operator characteristic curve (ROC curve) was drawn to analyze the predictive performance of various indicators, the SOFA score, and the combine model for the 28-day outcome in patients with sepsis.

A total of 101 patients with sepsis were enrolled, 56 patients survived and 45 patients died. Compared to the survival group, patients in the death group were older, the proportion of patients with septic shock was larger, the SOFA score, and the proportion of pulmonary infection were higher, the prothrombin time (PT) and activated partial thromboplastin time (APTT) were significantly prolonged, the prothrombin activity (PTA) was significantly shortened, and antithrombin (AT) was significantly decreased, the levels of hypersensitivity C-reactive protein (hs-CRP), blood urea nitrogen (BUN), total bilirubin (TBil), and lactic acid (Lac) were significantly increased, while the platelet count (PLT) was significantly decreased. Multifactorial Logistic regression analysis showed that pulmonary infection [odds ratio (OR) = 0.010, 95% confidence interval (95%CI) was 0.001-0.164, P = 0.001], AT (OR = 0.944, 95%CI was 0.910-0.978, P = 0.002), hs-CRP (OR = 1.008, 95%CI was 1.001-1.015, P = 0.017), Lac (OR = 1.619, 95%CI was 1.195-2.193, P = 0.002), and SOFA score (OR = 1.363, 95%CI was 1.076-1.727, P = 0.010) were independent prognostic factors for 28-day mortality in patients. A combined model was constructed using pulmonary infection, AT, hs-CRP, Lac, and SOFA score. ROC curve analysis showed that the area under the ROC curve (AUC) for the combine model in predicting sepsis prognosis was 0.936 (95%CI was 0.869-0.975, P < 0.001), which was higher in value compared to single indicators (AUC of AT, hs-CRP, Lac, and SOFA score were 0.775, 0.666, 0.802, 0.796, respectively, all P < 0.05).

The predictive ability of the SOFA score for sepsis patient outcomes is limited. The combine model combining infection site, AT, hs-CRP, and Lac shows better predictive ability.