胆囊结石患者并发急性胆源性胰腺炎的影响因素分析及列线图预测模型构建
作者:
通讯作者:
作者单位:

江苏省太仓市第一人民医院 肝胆外科,江苏 太仓 215400

作者简介:

陆颖超,江苏省太仓市第一人民医院副主任医师,主要从事肝胆外科方面的研究。

基金项目:


Analysis of influencing factors for acute biliary pancreatitis in patients with cholecystolithiasis and construction of nomogram prediction model
Author:
Affiliation:

Department of Hepatobiliary Surgery, the First People's Hospital of Taicang, Taicang, Jiangsu 215400, 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:

    目的 由胆囊结石诱发的急性胆源性胰腺炎(ABP)起病急,进展迅速,严重时危及生命。然而目前对于胆囊结石诱发ABP的机制及危险因素并非十分明确,且缺乏预测手段。因此,本研究探讨胆囊结石患者并发ABP的相关危险因素,并构建ABP发生风险的预测模型。方法 选取江苏省太仓市第一人民医院2018年1月—2021年3月期间收治的503例因腹痛入院并确诊为胆囊结石的患者为研究对象,收集患者临床资料、实验室指标以及ABP发生的情况。用单因素与多因素分析筛选ABP发生的危险因素,用ROC曲线分析各因素预测ABP的曲线下面积(AUC)与最佳截断值,构建列线图预测模型量化患者风险,并用校准曲线及决策曲线分析评估其临床预测效能。结果 503例胆囊结石患者中,119例(23.66%)并发ABP。与无ABP的患者比较,发生ABP的患者的APACHE Ⅱ评分、胆囊大小异常比例、多发胆囊结石比例、胆总管结石比例、血清淀粉酶(AMS)、C-反应蛋白(CRP)、降钙素原(PCT)以及中性粒细胞和淋巴细胞计数比值(NLR)均升高,而胆囊壁厚度减低(均P<0.05);ROC曲线分析结果显示,APACHE Ⅱ评分、胆囊壁厚度、AMS、CRP、PCT、NLR的AUC分别为0.681、0.769、0.886、0.734、0.869、0.822,最佳截断值分别为13.89、1.89 mm、382.10 U/L、18.69 mg/L、5.76 μg/L、3.05;多因素Logistic回归分析显示,胆囊壁厚度(<1.89 mm)、多发胆囊结石、AMS(≥382.10 U/L)、CRP(≥18.69 mg/L)、PCT(≥3.68 g/dL)及NLR(≥3.05)是胆囊结石患者并发ABP发生的独立危险因素(均P<0.05);根据上述独立影响因素构建的列线图的C指数为0.691(95% CI=0.661~0.735),风险阈值0.14,并且列线图模型的临床净收益显著高于任何单个指标预测结果。结论 胆囊壁厚度、多发胆囊结石、AMS、CRP、PCT以及NLR为胆囊结石患者并发ABP密切相关,基于以上因素构建的列线图模型对胆囊结石患者ABP发生的早期识别与预警有一定的临床价值。

    Abstract:

    Background and Aims Acute biliary pancreatitis (ABP) secondary to gallbladder stones has a rapid onset and swift progression, which can be life threatening in severe cases. However, the mechanism and risk factors for ABP induced by gallbladder stones are not entirely clear at present. Therefore, this study was conducted to investigate the risk factors for ABP in patients with cholecystolithiasis, and to construct a predictive model for the risk of ABP.Methods A total of 503 patients admitted for abdominal pain and diagnosed as cholecystolithiasis from January 2018 to March 2021 were enrolled as study subjects. The general clinical data, laboratory data and the occurrence of ABP of the patients were gathered. The risk factors for ABP were screened by univariate and multivariate analyses. The area under curve (AUC) and best cut-off value of each risk factor were determined by ROC curve analysis. A nomogram predictive model was constructed to quantify patient risk, and its clinical predictive ability was assessed by calibration curve and decision curve analyses.Results Among the 503 patients with cholecystolithiasis, 119 cases (23.66%) developed ABP. In patients with ABP compared with those without ABP, the APACHE Ⅱ score, proportion of cases with abnormal gallbladder size, proportion of cases with multiple gallbladder stones, proportion of cases with common bile duct stones, amylase (AMS), C-reactive protein (CRP), procalcitonin (PCT) and neutrophil to lymphocyte ratio (NLR) were increased (P<0.05), while the gallbladder wall thickness was decreased significantly (all P<0.05). Results of ROC curve analysis showed that the AUC values for APACHE Ⅱ score, gallbladder wall thickness, AMS, CRP, PCT and NLR were 0.681, 0.769, 0.886, 0.734, 0.869 and 0.822, and the best cut-off values were 13.89, 1.89 mm, 382.10 U/L, 18.69 mg/L, 5.76 μg/L and 3.05, respectively. Multivariate Logistic regression analysis showed that gallbladder wall thickness (<1.89 mm), multiple gallstones, AMS (≥382.10 U/L), CRP (≥18.69 mg/L), PCT (≥3.68 g/dL) and NLR (≥3.05) were independent risk factors for the occurrence of ABP in patients with cholecystolithiasis (all P<0.05). For the nomogram constructed by integrating the independent risk factors, the C-index was 0.691 (95% CI=0.661-0.735), and risk threshold was 0.14, and the clinical net benefit of the nomogram model was significantly higher than that predicted by any single variable.Conclusion Gallbladder wall thickness, multiple gallstones, AMS, CRP, PCT, and NLR are factors closely related to the occurrence of ABP in patients with gallbladder stones. The nomogram model constructed based on these factors has certain clinical value for early identification and warning of ABP in patients with gallbladder stones.

    表 3 胆囊结石患者并发ABP风险的多因素Logistic回归分析Table 3 Multivariate Logistic regression analysis of risk factors for occurrence of ABP in patients with gallbladder stones
    表 2 ROC曲线分析结果Table 2 Results of ROC curve analyses
    图1 ROC曲线分析Fig.1 ROC curve analyses
    图2 预测胆囊结石患者并发ABP高风险的列线图模型Fig.2 Nomogram model for predicting the occurrence of ABP in patients with gallbladder stones
    图3 列线图模型预测能力的校准曲线Fig.3 Calibration curve analysis of the predictive ability of the nomogram model
    图4 决策曲线分析Fig.4 Decision curve analysis
    参考文献
    相似文献
    引证文献
引用本文

陆颖超,黄锦山,徐红星,沈丹枫.胆囊结石患者并发急性胆源性胰腺炎的影响因素分析及列线图预测模型构建[J].中国普通外科杂志,2023,32(8):1199-1207.
DOI:10.7659/j. issn.1005-6947.2023.08.008

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:2022-07-14
  • 最后修改日期:2023-01-11
  • 录用日期:
  • 在线发布日期: 2023-11-03