Background and aim: Acute circulatory failure (ACF) is associated with high mortality rates in critically ill cirrhotic patients. Only a few accurate scoring models exist specific to critically ill cirrhotic patients with acute circulatory failure (CICCF) for mortality risk assessment. The aim was to develop and evaluate a novel model specific to CICCF.
Patients and methods: This study collected and analyzed the data on CICCF from the Multiparameter Intelligent Monitoring in Intensive Care-III database. The acute circulatory failure-chronic liver failure-sequential organ failure assessment (ACF-CLIF-SOFA) score was derived by Cox's proportional hazards regression. Performance analysis of ACF-CLIF-SOFA against CLIF-SOFA and model for end-stage liver disease systems was completed using area under the receiver operating characteristic curve.
Results: ACF-CLIF-SOFA identified six independent factors: mean arterial pressure [hazard ratio (HR)=0.984, 95% confidence interval (CI): 0.978-0.990, P<0.001], vasopressin (HR=1.548, 95% CI: 1.273-1.883, P<0.001), temperature (HR=0.764, 95% CI: 0.694-0.840, P<0.001), bilirubin (HR=1.031, 95% CI: 1.022-1.041, P<0.001), lactate (HR=1.113, 95% CI: 1.084-1.142, P<0.001), and urine output (HR=0.854, 95% CI: 0.767-0.951, P=0.004). ACF-CLIF-SOFA showed a better predictive performance than CLIF-SOFA and model for end-stage liver disease in terms of predicting mortality (0.769 vs. 0.729 vs. 0.713 at 30 days, 0.757 vs. 0.707 vs. 0.698 at 90 days, 0.733 vs. 0.685 vs. 0.691 at 1 year, respectively, all P<0.05).
Conclusion: ACF-CLIF-SOFA, as the first model specific to CICCF, enables a more accurate prediction at 30-day, 90-day, and 1-year follow-up periods than other existing scoring systems.