门静脉高压的无创指标及预测模型研究  

Study on noninvasive index and prediction model of portal hypertention

作　　者：何雨倩 马德强 曾伊凡 王传敏 陈悦 HE Yu-qian;MA De-qiang;ZENG Yi-fan(Department of Infectious Diseases,Taihe Hosipital,Shiyan Hubei,442000,China)

机构地区：[1]十堰市太和医院(湖北医药学院附属医院)感染科,湖北十堰442000

出　　处：《中西医结合肝病杂志》2023年第7期598-602,共5页Chinese Journal of Integrated Traditional and Western Medicine on Liver Diseases

摘　　要：目的:探讨肝静脉压力梯度(HVPG)的无创预测指标,并构建精准预测HVPG的无创预测模型。方法:选取2019年9月至2021年12月于十堰市太和医院感染科住院并行HVPG的肝硬化患者135例,收集一般临床资料、WBC、Hb、PLT、ALT、AST、总胆红素(TBil)、白蛋白(Alb)、国际标准化比值(INR)、门静脉宽度(PV)、胆囊壁厚度(GBWT)、脾直径(SD)及Fibroscan测量的肝硬度(LS)结果。比较HVPG在不同Child-pugh分级中的差异。根据HVPG结果将患者分为3组,分别为5~12 mmHg组、12~20 mmHg组、≥20 mmHg组。分析无创指标在不同HVPG分组中的差异,筛选有差异的指标进行多因素分析并构建HVPG的无创预测模型。通过受试者工作特征曲线(ROC)评价无创预测模型对HVPG的预测价值,并绘制列线图。结果:不同Child-pugh等级中,HVPG随着肝功能受损程度的加重而加重(P<0.05)。随着HVPG的升高,年龄、AST、TBil、INR、SD、LS呈升高趋势,而WBC、PLT、Hb、Alb呈下降趋势。多因素分析显示,PLT、Alb、LS是肝硬化患者HVPG加重的独立危险因素,构建了BPL评分模型:0.067×LS-0.051×PLT-0.122×Alb。对该模型行ROC曲线分析,BPL模型预测HVPG≥12 mmHg和HVPG≥20 mmHg的AUC分别为0.95、0.87,cutoff值分别为-6.50、-4.31,灵敏度、特异度分别为92.70%、88.46%和89.10%、73.75%。结论:无创指标PLT、Alb、LS是肝硬化患者HVPG升高的独立危险因素,基于上述指标构建的BPL评分预测模型可精准预测HVPG。Objective:To investigate the noninvasive prediction index of the hepatic venous pressure gradient(HVPG)and construct a noninvasive prediction model for accurately predicting HVPG.Methods:From September 2019 to December 2021,135 patients with cirrhosis who were hospitalized in the Department of Infectious Diseases of Taihe Hospital in Shiyan City and had HVPG were screened,and the general clinical data,WBC,Hb,PLT,ALT,AST,total bilirubin(TBil),albumin(Alb),international normalized ratio(INR),portal vein width(PV),spleen diameter(SD)and liver stiffness(LS)measured by Fibroscan were collected.Compare the differences in HVPG across different Child-pugh ratings.According to the HVPG results,the patients were divided into three groups,namely 5~12 mmHg group,12~20 mmHg group,and≥20 mmHg group.Analyze the differences of noninvasive indicators in different HVPG groupings,and screening of different indicators for multi-factor analysis and construction of a non-invasive predictive model for HVPG.The predictive value of the non-invasive prediction model for HVPG was evaluated by ROC curve,and the nomogram was drawn.Results:Among the different Child-pugh grades,HVPG worsened with the increase in the degree of liver function impairment(P<0.05).With the increase of HVPG,age,AST,TBIL,INR,SD,and LS showed an upward trend,while WBC,PLT,Hb,and ALB showed a downward trend(P<0.05).The results of multivariate analysis showed that PLT,Alb,and LS were independent risk factors for HVPG exacerbation in patients with cirrhosis,and the BPL score model:0.067×LS-0.051×PLT-0.122×ALB was constructed.Based on the ROC curve analysis of the model,the BPL model predicted that the AUC of HVPG≥12 mmHg and HVPG≥20 mmHg were 0.95、0.87,the cutoff value was-6.50、-4.31,and the sensitivity and specificity were 92.70%、88.46%and 89.10%、73.75%.The BPL model predicted that the AUC of HVPG≥20 mmHg was 0.87,the cutoff value was-4.31,and the sensitivity and specificity were 89.10%and 73.75%.Conclusion:Noninvasive indicators PLT,ALB,LS are indepe

关 键 词：门静脉高压 无创预测 肝硬化 肝静脉压力梯度 

分 类 号：R575.1[医药卫生—消化系统]