ABSTRACT
BACKGROUND & AIMS: An algorithm including Sepsis-3 criteria and quick Sequential Organ Failure Assessment (qSOFA) was recently proposed to predict severity of infection in cirrhosis. However, its applicability among patients without a baseline SOFA available for Sepsis-3 definition is unknown. We sought to investigate the applicability and prognostic value of qSOFA and Sepsis-3 criteria in patients with cirrhosis hospitalised for bacterial infections, without pre-hospitalisation SOFA. METHODS: In this cohort study, 164 patients were followed up to 30 days. Data collection, including the prognostic models, was performed at admission and at day-3. RESULTS: All patients fulfilled Sepsis-3 criteria (admission SOFA ≥ 2) and, therefore, admission Sepsis-3 was not included in further analysis. Admission qSOFA was an independent predictor of survival (HR = 2.271, P = 0.015). For patients initially classified as high risk by qSOFA, Chronic Liver Failure - Sequential Organ Failure Assessment (CLIF-SOFA) was the only prognostic predictor. Among patients initially classified as low risk by qSOFA, the following parameters evaluated at day-3 were independent predictors of survival: qSOFA, acute-on-chronic liver failure, and Child-Pugh classification. Although not independently related to survival, Sepsis-3 criteria at day-3 was associated with lower 30-day survival in Kaplan-Meier analysis (66% vs 85%, P = 0.008). However, prognosis was better predicted by day-3 qSOFA, with 30-day Kaplan-Meier survival probability of 88% when qSOFA < 2 and 24% among those with qSOFA ≥ 2. CONCLUSION: Sepsis-3 criteria evaluated at admission are very limited in infected patients with cirrhosis without baseline SOFA. qSOFA was independently related to survival and appears to be a valuable tool for determining severity of infection and to follow patients initially classified as low risk.
