Analysis of influencing factors on the occurrence of esophageal and gastric varices in liver cirrhosis and construction of a column chart prediction model
10.3760/cma.j.cn431274-20240331-00555
- VernacularTitle:肝硬化食管胃静脉曲张发生的影响因素分析及列线图预测模型构建
- Author:
Xiaojuan LI
1
;
Xinxin XU
;
Qingqing FANG
;
Wei CHEN
;
Ying CHEN
Author Information
1. 复旦大学附属闵行医院消化内科，上海 201199
- Keywords:
Liver cirrhosis;
Esophageal and gastric varices;
Risk factors
- From:
Journal of Chinese Physician
2024;26(4):503-508
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the influencing factors of esophageal and gastric varices (EGV) in patients with liver cirrhosis, and to construct a column chart prediction model for EGV occurrence.Methods:A cross-sectional study was conducted to select patients ( n=127) who underwent gastroscopy at Minhang Hospital, Fudan University from January 2022 to April 2023 due to liver cirrhosis. Patients were divided into EGV group ( n=74) and non EGV group ( n=53) based on the presence of EGV. Clinical and auxiliary examination data of the two groups were compared. Univariate and multivariate logistic regression analysis were used to explore the independent influencing factors of EGV occurrence, and a column chart prediction model was further constructed to evaluate its predictive efficacy. Results:Multivariate logistic regression analysis showed that splenomegaly ( OR=28.38, 95% CI: 6.30-127.89, P<0.001), schistosomiasis infection ( OR=0.05, 95% CI: 0.01-0.28, P=0.001), and fatty liver ( OR=0.09, 95% CI: 0.02-0.33, P<0.001) were independent influencing factors for the occurrence of EGV in patients with liver cirrhosis. Based on this model, a column chart was constructed to predict the probability of EGV occurrence in patients with liver cirrhosis. The column chart model showed good consistency and discrimination (area under the curve was 0.927, 95% CI: 0.883-0.972), and was internally validated using the Bootstrap method, showing an average absolute error of 0.029. Conclusions:Spleen enlargement, schistosomiasis infection, and fatty liver are independent influencing factors for the occurrence of EGV in patients with liver cirrhosis. The Nomogram column chart prediction model for the occurrence of EGV in patients with liver cirrhosis has good predictive performance and can provide scientific basis for non-invasive risk screening and prevention strategies for EGV in liver cirrhosis.
