中国实用外科杂志

中国实用外科杂志 ›› 2025, Vol. 45 ›› Issue (04): 443-449.DOI: 10.19538/j.cjps.issn1005-2208.2025.04.11

• 论著 • 上一篇    下一篇

急性闭塞性肠系膜缺血临床特征分析及肠坏死风险预测模型构建

撖毓敏1,2，姜笑梅1，张贵祥1，黄理宾1，冯    丽3，黄    斌4，张    蜀3，刘    聪2,5，杨    烈1   

  1. 1四川大学华西医院普外科 胃肠外科，四川成都 610041；2四川大学疾病分子网络前沿科学中心  基因组稳定性实验室，四川成都 610041；3四川大学华西医院急诊科，四川成都 610041；4四川大学华西医院血管外科，四川成都 610041；5四川大学华西第二医院 出生缺陷及妇女儿童相关疾病教育部重点实验室，四川成都 610041
  • 出版日期:2025-04-01 发布日期:2025-04-30

  • Online:2025-04-01 Published:2025-04-30

摘要： 目的    分析急性闭塞性肠系膜缺血（AOMI）病人临床转归的影响因素，构建并验证肠坏死临床预测模型。方法    回顾性分析2020年1月至2023年12月四川大学华西医院普外科收治的236例AOMI病人临床资料，其中闭塞性动脉肠系膜缺血（AAMI）108例，肠系膜静脉血栓形成（MVT）128例。分析AOMI病例的临床特征，通过Logistic回归分析转归差的危险因素；进一步筛选其中明确肠坏死状态的病人，对比分析其临床特征，构建肠坏死预测模型，并进行内部验证。结果    多因素分析结果显示，器质性心脏病（OR=5.030，95%CI  1.675-15.123）、门静脉高压（OR=16.557，95%CI  1.094-249.326）、白细胞计数≥10×10⁹/L（OR=9.670，95%CI  2.193-42.668）、血小板计数<100×10⁹/L（OR=9.122，95%CI  2.235-37.262）、总胆红素≥17.1 μmol/L（OR=3.175，95%CI  1.148-8.786）及AAMI（OR=38.862，95%CI  2.005-752.530）是转归差的独立危险因素（均P<0.05）。基于肠腔积液（OR=3.466，95%CI 1.468-8.175）、肠系膜肿胀（OR=2.373，95%CI  1.008-5.582）、肠管壁肿胀（OR=1.742，95%CI  0.738-4.093）、白细胞计数（OR=2.509，95%CI  0.950-6.587）、D-二聚体（OR=2.165，95%CI  0.952-4.976）、白蛋白（OR=0.449，95%CI  0.163-1.248）以及淋巴细胞百分比（OR=0.883，95%CI  0.335-2.308）建立肠坏死预测模型，训练集的受试者工作特征曲线下面积（AUC）为0.808（95%CI  0.742-0.891），敏感度为0.742，特异度为0.781；验证集AUC为0.793（95%CI 0.664-0.913），敏感度为0.668，特异度为0.734。组间校准曲线显示良好一致性。结论    基于影像学特征和实验室指标多维度临床数据构建的肠坏死预测模型具有较高准确性和临床实用性，可为AOMI病人早期干预、个体化治疗及急诊流程优化提供量化依据。 

关键词: 急性闭塞性肠系膜缺血, 肠坏死, 临床特征, 影响因素, 预测模型

Abstract: To analyze the factors influencing clinical outcomes in patients with acute occlusive mesenteric ischemia (AOMI) and to develop and validate a clinical prediction model for intestinal necrosis. Methods    A retrospective analysis was conducted on clinical data from 236 AOMI patients admitted to West China Hospital of Sichuan University between January 2020 and December 2023, including 108 cases of acute arterial mesenteric ischemia (AAMI) and 128 cases of mesenteric venous thrombosis (MVT). The clinical characteristics of AOMI cases were analyzed, and logistic regression was used to identify risk factors for poor outcomes. Further, patients with confirmed intestinal necrosis status were selected to compare clinical features, construct a prediction model, and perform internal validation. Results  Multivariate analysis revealed that organic heart disease (OR=5.030, 95%CI 1.675-15.123), portal hypertension (OR=16.557, 95%CI 1.094-249.326), WBC count ≥10×10⁹/L (OR=9.670, 95%CI 2.193-42.668), platelet count <100×10⁹/L (OR=9.122, 95%CI 2.235-37.262), total bilirubin ≥17.1 µmol/L (OR=3.175, 95%CI 1.148-8.786), and AAMI (OR=38.862, 95%CI 2.005-752.530) were independent risk factors for poor outcomes (all P<0.05). The intestinal necrosis prediction model integrated imaging features and laboratory indicators, including intestinal cavity effusion (OR=3.466, 95%CI 1.468-8.175), mesenteric swelling (OR=2.373, 95%CI 1.008-5.582), intestinal wall swelling (OR=1.742, 95%CI 0.738-4.093), white blood cell count (OR=2.509, 95%CI 0.950-6.587), D-dimer (OR=2.165, 95%CI 0.952-4.976), albumin (OR=0.449, 95%CI 0.163-1.248), and lymphocyte percentage (OR=0.883, 95%CI 0.335-2.308). The training set demonstrated an AUC of 0.808 (95%CI 0.742-0.891, sensitivity 0.742, specificity 0.781), while the validation set showed an AUC of 0.793 (95%CI 0.664-0.913, sensitivity 0.668, specificity 0.734), with calibration curves indicating good consistency between groups. Conclusion  The prediction model for intestinal necrosis, constructed based on multidimensional clinical data including imaging features and laboratory indicators, exhibits high accuracy and clinical utility. It provides a quantitative foundation for early intervention, personalized treatment, and optimization of emergency protocols in AOMI patients.

Key words: acute occlusive mesenteric ischemia, intestinal necrosis, clinical features, influencing factors, prediction model