Abstract
Objective: To explore the prognostic value of model for end-stage liver disease (MELD) combined with arterial blood lactate (Lac) in admitted patients with hepatitis B virus-associated acute- on-chronic liver failure (HBV-ACLF). Methods: Clinical data of 97 cases with hepatitis B virus-associated acute- on-chronic liver failure (HBV-ACLF) admitted to the First Affiliated Hospital of Suzhou University between March 2016 and March 2018 was retrospectively analyzed. Age, gender, complications, MELD score, lactic acid (Lac), total bilirubin (TBil), creatinine (Cr), serum albumin (Alb), high-sensitivity C-reactive protein (CRP), white blood cell count (WBC), platelet count (PLT), hematocrit (Hct), quantification of HBV DNA and HBsAg, and organ support treatment (artificial liver support system, renal replacement therapy and mechanical ventilation ) were documented after admission. The primary endpoint of treatment was death due to ineffective medical treatment during hospitalization, abandonment of medical treatment due to deterioration of the health condition, and switch to liver transplantation for patients with poor medical treatment. The risk factors for primary endpoint of treatment were analyzed by binary logistic regression. Hosmer-Lemeshow test was used to evaluate the goodness of fit for the scoring system, and the ROC to predict the prognosis of MELD-Lac. Results: Ninety-seven cases with HBV-ACLF were included, 56 cases had good prognosis, and 41 cases had bad prognosis (including two cases with poor medical treatment and liver transplantation). The overall improvement rate was 57.7%. MELD score and Lac value in treated group was significantly lower than non-treated group. Bivariable and multivariable logistic regression analysis showed that the MELD score [odds ratio (OR = 1.806)], and Lac score [odds ratio (OR = 1.820)] was the risk factor for hospitalization and mortality in patients with liver failure (P < 0.05). The area under the ROC curve (AUC) and the 95% confidence interval (95% CI) of prognostic patients with MELD-Lac were significantly better than Lac and MELD scores [0.923 (0.84 to 1.00) vs. 0.804 (0.067 to 0.942) and 0.864 (0.75). 0.977)], P < 0.05. When the MELD-Lac Youden index was set at 0.746, the optimal threshold was 18.36, and the sensitivity and specificity were 91.3% and 83.3%, respectively. Conclusion: MELD-Lac score has a high prognostic value in HBV-ACLF patients.