胰源性门静脉高压胃静脉破裂出血危险因素分析及列线图预测模型构建
作者:
通讯作者:
作者单位:

1.湖南中医药大学第一附属医院 肝胆胰与疝外科,湖南 长沙 410007;2.湖南省人民医院/湖南师范大学第一附属医院 胰脾外科,湖南 长沙 410005

作者简介:

刘鹏,湖南中医药大学第一附属医院主治医师,主要从事肝胆胰方面的研究。

基金项目:


Analysis of risk factors for gastric venous rupture bleeding caused by pancreatic portal hypertension and construction of nomogram predictive model
Author:
Affiliation:

1.Department of Hepatopancreatobiliary and Hernia Surgery, the First Hospital of Hunan University of Chinese Medicine, Changsha 410007, China;2.Department of Pancreatic and Splenic Surgery, Hunan Provincial People's Hospital/the First Affiliated Hospital of Hunan Normal University, Changsha 410005, China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 音频文件
  • |
  • 视频文件
    摘要:

    背景与目的 由于近年来胰腺相关疾病的发病率和诊断率逐渐增加,胰源性门静脉高压(PPH)及其并发症的发病率及诊断率也随之增加。PPH可出现腹痛、脾功能亢进、消化道出血等临床表现,胃静脉破裂出血是PPH最严重的并发症,可导致失血性休克,甚至死亡。而目前尚无公认的标准来预测PPH并发胃静脉破裂出血的风险。因此,本研究探讨PPH胃静脉破裂出血的相关危险因素并建立风险预测模型,为临床该病的防治提供参考。方法 以湖南中医药大学第一附属医院和湖南省人民医院自2012年1月—2022年1月收治的176例PPH患者为研究对象,根据是否并发胃静脉破裂出血分为出血组(24例)和未出血组(152例)。收集两组患者的一般资料:性别、年龄、病因、高血压、糖尿病,实验室结果:脾功能亢进、凝血酶原时间(PT)、凝血酶时间(TT)、纤维蛋白原(FIB)、活化部分凝血酶原时间(APTT)与影像学指标:胃左静脉(LGV)内径、胃网膜静脉(GEV)内径、GEV内径/LGV内径比值。采用单因素与多因素Logistic回归分析筛选出PPH并发胃静脉破裂出血的影响因素,建立列线图风险预测模型,并通过受试者工作特征(ROC)曲线、校准曲线和决策曲线评估模型的预测效能、一致性和临床价值。结果 单因素分析显示,糖尿病、脾功能亢进、PT、FIB、LGV内径、GEV内径/LGV内径比值与PPH并发胃静脉破裂出血有关(均P<0.05);多因素Logistic回归显示,糖尿病(OR=0.144,95% CI=0.031~0.675,P=0.014)、LGV内径(OR=3.129,95% CI=1.608~6.090,P=0.001)是PPH并发胃静脉破裂出血的独立危险因素;而FIB(OR=0.580,95% CI=0.348~0.966,P=0.037)、GEV内径/LGV内径比值(OR=0.024,95% CI=0.001~0.696,P=0.030)是PPH并发胃静脉破裂出血的独立保护性因素。纳入以上因素与有临床意义的GEV内径构建PPH胃静脉破裂出血的列线图模型,该模型的ROC曲线下面积为0.954,校准曲线与标准曲线具有很高的拟合度,决策曲线分析显示,当风险阈值在2%~85%之间时,能得到很好的净获益值。结论 糖尿病、LGV内径,FIB、GEV内径/LGV内径比值与PPH并发胃静脉破裂出血密切相关,所建立的列线图风险预测模型可有效识别PPH并发胃静脉破裂出血的高风险患者,可为临床决策提供参考。

    Abstract:

    Background and Aims The incidence and diagnosis rate of pancreatic-related diseases have been increasing in recent years, leading to a corresponding rise in the incidence and diagnosis rate of pancreatic portal hypertension (PPH) and its complications. PPH can manifest with symptoms such as abdominal pain, splenomegaly, and gastrointestinal bleeding. Gastric venous rupture bleeding is the most severe complication of PPH, which can result in hemorrhagic shock and even death. Currently, there is no recognized standard for predicting the risk of gastric venous rupture bleeding in patients with PPH. Therefore, this study explored the relevant risk factors for gastric variceal rupture/bleeding in PPH patients and established a risk prediction model to provide a reference for preventing and treating this condition in clinical practice.Methods A total of 176 patients with PPH admitted to the First Affiliated Hospital of Hunan University of Chinese Medicine and Hunan Provincial People's Hospital from January 2012 to January 2022 were included in this study. Patients were divided into bleeding group (24 cases) and non-bleeding group (152 cases) based on whether they developed gastric venous rupture bleeding. General information including sex age, etiology, hypertension, and diabetes, laboratory results such as splenomegaly, prothrombin time (PT), thrombin time (TT), fibrinogen (FIB), activated partial thromboplastin time (APTT), and imaging indicators including the inner diameter of the left gastric vein (LGV), the inner diameter of the gastroesophageal vein (GEV), and the GEV inner diameter /LGV inner diameter ratio were collected for both groups. Univariate and multivariate Logistic regression analyses were conducted to identify the influencing factors for PPH complicated with gastric variceal rupture/bleeding, a nomogram risk prediction model was established, and then the predictive efficacy, consistency, and clinical value of the model were evaluated through receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis.Results Univariate analysis showed that diabetes, splenomegaly, PT, FIB, LGV inner diameter, and GEV inner diameter/LGV inner diameter ratio were associated with PPH complicated with gastric variceal rupture/bleeding (all P<0.05). Multivariate Logistic regression analysis revealed that diabetes (OR=0.144, 95% CI=0.031-0.675, P=0.014) and LGV diameter (OR=3.129, 95% CI=1.608-6.090, P=0.001) were independent risk factors for PPH complicated with gastric venous rupture bleeding, while FIB (OR=0.580, 95% CI=0.348-0.966, P=0.037) and GEV/LGV ratio (OR=0.024, 95% CI=0.001-0.696, P=0.030) were independent protective factors. A nomogram model for predicting the risk of gastric venous rupture bleeding in PPH incorporating these factors and clinically significant GEV inner diameter demonstrated an area under the ROC curve of 0.954. The calibration curve and decision curve analysis showed high fitting degree and net benefit values when the risk threshold was between 2% and 85%.Conclusion Diabetes, LGV inner diameter, FIB, and GEV inner diameter/LGV inner diameter ratio are closely related to PPH complicated with gastric venous rupture bleeding. The established nomogram risk prediction model can effectively identify high-risk patients for gastric venous rupture bleeding in PPH and provide a reference for clinical decision-making.

    表 2 差异指标的二元Logistic回归分析Table 2 Binary Logistic regression analysis of the different variables
    图1 PPH并发胃静脉破裂出血列线图预测模型Fig.1 Nomogram predictive model for PPH complicated with gastric venous rupture bleeding
    图2 列线图模型预测效能评价 A:ROC曲线;B:校准曲线;C:决策曲线Fig.2 Evaluation of predictive efficacy of the nomogram model A: ROC curve; B: Calibration curve; C: Decision curve
    图1 PPH并发胃静脉破裂出血列线图预测模型Fig.1 Nomogram predictive model for PPH complicated with gastric venous rupture bleeding
    图2 列线图模型预测效能评价 A:ROC曲线;B:校准曲线;C:决策曲线Fig.2 Evaluation of predictive efficacy of the nomogram model A: ROC curve; B: Calibration curve; C: Decision curve
    表 1 患者临床资料Table 1 Clinical data of patients
    参考文献
    相似文献
    引证文献
引用本文

刘鹏,向兴华,李文斌,李国光,朱朝庚.胰源性门静脉高压胃静脉破裂出血危险因素分析及列线图预测模型构建[J].中国普通外科杂志,2024,33(3):400-407.
DOI:10.7659/j. issn.1005-6947.2024.03.011

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2023-09-11
  • 最后修改日期:2024-03-18
  • 录用日期:
  • 在线发布日期: 2024-04-10