CHIBA score: a novel model for predicting 3-month mortality in a cohort of Decompensated Liver Disease (DCLD)

Abstract Background Decompensated liver disease (DCLD) has high mortality, and its prediction is important to prognosticate prioritize patients for transplantation. MELD, MELD variants, CTP were widely tested mortality with few drawbacks. The aim of the study propose a new prognostic model DCLD which better than existing scores. Materials methods Retrospective 321 enrolled. Patient relatives telephonically contacted regarding date death, at 3 months was assessed. Logistic regression done, coefficient beta independent variables found out, CHIBA score proposed. = creatinine × 0.6 + HE 0.4+ INR 0.8 bilirubin 0.125 ascites 1.2) where C stands creatinine, H hepatic encephalopathy, I INR, B bilirubin, A ascites. Results AUROC 0.793 (at cutoff > 5.5, it sensitivity 66% specificity 76%) compared MELD-Na 0.735 (cutoff 25, 65%, 72%); 0.727 17 80.37% 55.14%); I-MELD 0.72; MESO index UKELD 0.686. For validation, 214 selected, in validation cohort 0.77. At 60% 77%. Conclusion superior variants predicting 3-month validated an external cohort. It can be calculated bedside as simple no logarithmic it.