Abstract: Objective To investigate factors affecting clinical endpoint events during antiviral therapy for chronic hepatitis B (CHB) and establishing an accurate prediction system. Methods A total of 200 CHB patients who received antiviral treatment in our hospital from February 2017 to January 2020 our hospital were selected as the subjects. The clinical data were collected. The time of onset, end or end point of the antiviral treatment was the cut-off time. The risk factors of clinical end point events affecting antiviral treatment were analyzed to establish a prognostic model. The clinical value of different systems including with the end stage liver disease model (MELD), MELD-Na score, Fontana-Index score,hild-Turcotte-Pugh (CTP) score and CTP-Cr score was compared. Results CTP-Cr score (7.51±1.61), MELD-Na score (15.07±6.63), ultrasound score (12.21±3.25), Index score (3.14±1.47), MELD score (10.67±3.36), CTP score (6.58±2.14), Na (127.56±10.15), international normalized ratio (INR) (1.34±0.35), prothrombin time (PT) (14.58±2.14), direct bilirubin (DBil) (19.69±28.66), total bilirubin (TBil) (50.88±57.45) were significantly different from those of CTP-Cr in the group with end-point events score (8.96±1.35), MELD-Na score (22.76±4.91), ultrasound score (18.74±4.29), Index score (4.69±1.69), MELD score (16.74±3.07), CTP score (9.14±2.07), Na (117.96±7.45), INR (1.72±0.28), PT (18.57±2.31), DBil (35.47±30.14), TBil (78.96±80.14), the difference was statistically significant (P<0.05); no end-point event group The incidence of SBP (0.63%), hepatorenal syndrome (0.00%), and hepatic encephalopathy (1.25%) were lower than those of the end-point event group (10.00%, 7.50%, 10.00%) (P<0.05); Multivariate analysis showed that the prognostic risk factors with significant differences were mainly INR,WBC,TBil, and ultrasound score (P<0.05). The receiver operator characteristic (ROC) curves of different systems were conducted, and the area under the curve (AUC) was calculated (Index: 0.807, MELD-Na: 0.827, MELD: 0.804, CTP-Cr: 0.800, CTP: 0.791). The AUC of?the new system A and B were 0.854 and 0.834 respectively. There was no significant difference between CTP-Cr, CTP, prognosis system B and prognosis system A (P>0.05). Conclusion INR, WBC, TBil, hepatorenal syndrome, and ultrasound score are independent risk factors that affect the clinical endpoints of antiviral treatment. New system A based on the basis of INR, TBil, hepatorenal syndrome, and ultrasound has a high value for predicting and evaluating clinical endpoint events.

Key words: Antiviral therapy, Chronic hepatitis B, Cox regression analysis, Clinical endpoint events, Prediction system