Risk Assessment and Predictive Model Establishment of Clinically Significant Portal Hypertension in Patients With Primary Biliary Cholangitis
10.3969/j.issn.1008-7125.2024.03.001
- VernacularTitle:原发性胆汁性胆管炎患者临床显著性门静脉高压的危险因素分析和预测模型建立
- Author:
Yao LI
1
;
Yuyang LIU
;
Jun QIAN
;
Xiong MA
;
Qixia WANG
Author Information
1. 上海交通大学医学院附属仁济医院消化内科 上海市消化疾病研究所(200001)
- Keywords:
Primary Biliary Cholangitis;
Clinically Significant Portal Hypertension;
Predictive Model;
Nomogram
- From:
Chinese Journal of Gastroenterology
2024;29(3):129-134
- CountryChina
- Language:Chinese
-
Abstract:
Background:Patients with primary biliary cholangitis(PBC)may develop gastroesophageal varices at an early stage.It is of great significance to prevent variceal bleeding for the long-term outcome of PBC.Aims:To assess the risk factors for clinically significant portal hypertension(CSPH)among PBC patients and develop a predictive model.Methods:The clinical data of PBC patients admitted from January 2018 to January 2022 in the Division of Gastroenterology and Hepatology,Renji Hospital,Shanghai Jiao Tong University School of Medicine were retrospectively collected and analyzed.Patients were divided into CSPH and non-CSPH groups according to the diagnostic criteria of CSPH.Multivariate Logistic regression and Lasso regression were used to identify the risk factors and construct CSPH predictive model which was presented as nomogram and internally validated.Results:Of the 458 subjects enrolled in this study,140 cases were in CSPH group while 318 cases in non-CSPH group.Multivariate analysis identified that male gender(OR=2.89),positive for anti-centromere antibody(ACA;OR=2.18)and anti-gp210 antibody(OR=1.75),high baseline total bilirubin(OR=1.01)and IgA(OR=1.27),and low baseline albumin(OR=0.91)and platelet count(OR=0.98)were independent risk factors for CSPH.These variables were selected to form the predictive model and nomogram ultimately.Favorable stability and predictive performance of the model were confirmed by ROC curve analysis(AUC=0.891)and bootstrap method(C-index=0.891).Conclusions:The predictive model constructed in this study has certain reference value for risk stratification of CSPH among PBC patients.Intensive follow-up is recommended for ACA-positive PBC patients in clinical practice for avoidance of gastroesophageal variceal bleeding.
