Abstract: Objective Endoscopic ultrasonography (EUS) was used to predict the recurrence of esophageal varices after esophageal variceal ligation (EVL) in patients with hepatitis B liver cirrhosis (LC), so as to effectively evaluate the prognosis of the patients. Methods A total of 116 LC patients treated with EVL from January 2016 to September 2019 were reviewed, including 38 patients with recurrence (29 males and 9 females) and 78 patients without recurrence (52 males and 26 females). T-test was used for measurement data, chi-square test was used for counting data, and Logistic regression analysis was used to explore the influencing factors of EVL recurrence. Results The Child-Pugh scores in EVL recurrent group and non-recurrent group were (8.5±3.2) points and (6.5±3.0) points respectively, and the difference was statistically significant (t=8.980, P<0.05). The EV degrees of F2 and F3 in recurrent group were 20 cases (52.6%) and 18 cases (44.4%), respectively, while those in non-recurrent group were 63 cases (80.8%) and 15 cases (19.2%) respectively, and the difference was statistically significant (χ²=-4.244, P<0.05). The diameters of EV, para-esophageal collateral vein and communicating vein in recurrent group were (10.1±2.5) mm, (9.1±2.4) mm and (3.5±0.6) mm, respectively, compared with those in non-recurrent group [(8.2±2.3) mm, (7.0±1.7) mm and (3.0±0.5) mm], the difference was statistically significant (t=10.548, 10.246, 4.405, P<0.05). There was no significant difference in splenomegaly, PLT and the number of communicating branches between the two groups. The above statistically different indicators were included in multivariate analysis. The results showed that Child-Pugh score, the diameter of para-esophageal collateral vein and the maximum diameter of communicating vein were independent risk factors for the recurrence of EVL, while the degree of EV, gastric varices and EV diameter were not independent risk factors for the recurrence of EVL. The sensitivity, specificity and accuracy of EUS in the diagnosis of EVL recurrence were 81.6% (31/38), 92.3% (72/78) and 88.8% (103/116), respectively. Conclusion EUS could accurately predict the recurrence of LC patients after EVL. For patients with poor Child-Pugh score, abnormal diameter of para-esophageal collateral vein and maximum diameter of communicating vein, it is necessary to pay attention and formulate reasonable treatment strategies.

Key words: Endoscopic ultrasonography, Esophageal variceal ligation, Hepatitis B cirrhosis, Para-esophageal collateral vein