Establishment of a nomogram model for predicting bleeding risk based on endoscopic ultrasound and clinical data among patients with liver cirrhosis

Abstract

Objective: To study the predictive value for esophageal variceal bleeding (EVB) after esophageal variceal ligation (EVL) based on clinical data and endoscopic ultrasound examination among patients with liver cirrhosis. Methods: Routine endoscopy and endoscopic ultrasound were performed on 400 preventive EVL cases who met the enrollment criteria, and their clinical indicators, microscopic manifestations, diameter and number of peripheral collateral veins, para-esophageal veins, perforating veins, and so on were monitored. Multivariate Cox proportional hazard regression analysis was performed to determine the relevant factors. Nomogram predictive model was established based on the independent prognostic factors. Nomogram internal validation was carried out with C-index to check the prediction accuracy, calibration curve evaluation consistency, and clinical benefit evaluation for decision-making. Results: Age, gender, etiology, peri-ECV number, para-ECV diameter, portal vein diameter, and azygous vein diameter were the risk factors for disease progression (P < 0.05), and they were all included in the establishment of nomogram. The C-index obtained by internal validation was 0.864, and the area under the receiver operating characteristic curve (AUC) of the predicting bleeding progression model was 0.994 (P < 0.001), suggesting that the disease had high predictive value and the calibration curve validity had consistency. Conclusion: Nomogram predictive model established based on the clinical data and endoscopic ultrasound examination is a preventive and therapeutic intervention for liver cirrhotic patients with esophageal varices, which can effectively improve the therapeutic effects of this population, reduce the EVs occurrence, and improve their quality of life. In addition, it also has an outstanding favorable performance. Therefore, it has certain guiding significance for the judgment of clinical treatment.