Objective To explore the risk factors of esophageal varices in patients with hepatitis B cirrhosis, and establish a nomogram model to predict the risk of esophageal varices(EVs).

Methods The clinical data of patients with hepatitis B cirrhosis were with a retrospective cohort study analyzed.A total of 241 patients with hepatitis B cirrhosis were recruited, and randomly divided into the training cohort(169 cases) and validation cohort(72 cases) in a 7∶3 ratio.Univariate analysis and multivariate logistic regression analysis were used to analyze the risk factors of EVs.The nomogram to predict the probability of EVs occurrence in hepatitis B cirrhosis patients was constructed.The ROC curve was used to evaluate the accuracy of the nomogram in predicting the presence of hepatitis B cirrhosis.Decision curve analysis was adopted to determine the clinical benefits of nomogram.

Results The results showed that the platelet distribution width decreased, and the peak velocity of portal vein slowing and spleen length increasing were the independent risk factors of EVs in hepatitis B cirrhosis patients(P < 0.05 to P < 0.01).The nomogram model was successfully constructed.Whether it was the training cohort or the validation cohort, the calibration curve of nomogram was close to the standard curve.The nomogram was well calibrated, and the results for predicting the risk of EVs occurrence were accurate and reliable.The results of DCA showed that in both the training cohort and validation cohort patients, the nomogram DCA curves were above the zero net benefit line, which indicated that the nomogram could achieve clinical benefits.

Conclusions A nomogram model based on platelet distribution width, peak velocity portal vein of spleen length is constructed and verified, which can non-invasively predict the risk of EVs in patients with hepatitis B cirrhosis.