Predictive value of combined detection of prothrombin time-international normalized ratio-to-albumin ratio, model for end-stage liver disease-Na and high-density lipoproteincholesterol in prognosis of patients with acute-on-chronic liver failure

Objective To explore the predictive value of combined detection of prothrombin time-international normalized ratio-to-albumin ratio(PTAR), model for end-stage liver disease-Na (MELD-Na) and high-density lipoproteincholesterol(HDL-C) in prognosis of patients with acute-on-chronic liver failure (ACLF).

Methods The clinical data of 154 patients with ACLF were retrospectively analyzed and divided into survival group (95 cases) and death group (59 cases) according to the prognosis on the 90th day after diagnosis. The clinical baseline indexes at admission were compared between the two groups. The correlations among PTAR, HDL-C and MELD-Na were analyzed. The prognostic factors of ACLF patients were analyzed. Receiver operating characteristic (ROC) curves of PTAR, HDL-C, MELD-Na and their combination prediction(PTAR+MELD-Na+HDL-C) models were drawn to evaluate their prognostic efficacy in ACLF patients. According to the truncation value (0.530) of the combined model, patients were divided into group A (< 0.530, n=47) and group B (≥0.530, n=107). The survival rates of the two groups were compared by Kaplan-Meier curve analysis.

Results There were significant differences in age, total bilirubin (TBil), albumin (ALB), creatinine (Cr), prothrombin time (PT), serum sodium (Na+), white blood cell (WBC), platelet (PLT), PTAR, HDL-C and MELD-Na between two groups (P < 0.05). PTAR was positively correlated with MELD-Na (r=0.407, P < 0.001); PTAR and MELD-Na were negatively correlated with HDL-C (r=-0.316, -0303, P < 0.001). Age, TBil, PT, PTAR and MELD-Na were independent influencing factors for the prognosis of ACLF patients (P < 0.05), and HDL-C was an effective predictor of the prognosis of ACLF patients. The area under curves (AUC) of PTAR, HDL-C and MELD-Na were 0.722, 0.717 and 0.738, respectively; the AUC of the combined prediction model was 0.801(95%CI, 0.722 to 0.881, P < 0.001), the sensitivity was 0.627, the specificity was 0.916, the Jorden index was 0.543, and the cutoff value was 0.530. Kaplan-Meier curve analysis showed that the cumulative survival rate in the group A was significantly higher than that in the group B (P < 0.001).

Conclusion PTAR, HDL-C and MELD-Na are all independent influencing factors for the short-term prognosis of ACLF patients, and the combined prediction model of the above three factors has a good prognostic value for ACLF patients.