外科理论与实践

外科理论与实践 ›› 2025, Vol. 30 ›› Issue (05): 400-408.doi: 10.16139/j.1007-9610.2025.05.05

• 论著 • 上一篇    下一篇

肥胖病人腹腔镜胃袖状切除术后急性高血压的预测模型构建

王玥a, 郭军威b,*, 袁航b, 杜磊a, 贾许杨a, 卜乐b(), 卢列盛()   

  1. 同济大学附属第十人民医院 a.代谢外科； b.肥胖症诊疗中心，上海 200072
  • 收稿日期:2025-05-18 出版日期:2025-09-25 发布日期:2025-12-09
  • 通讯作者: 卢列盛，E-mail: luliesheng1980@163.com;
    卜乐，E-mail: geyingjun@hotmail.com
  • 作者简介:第一联系人：

    *：共同第一作者。

Construction of prediction model for acute hypertension following laparoscopic sleeve gastrectomy in obese patients

WANG Yuea, GUO Junweib,*, YUAN Hangb, DU Leia, JIA Xuyanga, BU Leb(), Lu Liesheng()   

  1. Department of Metabolic Surgery; b. Obesity Diagnosis and Treatment Center, Tenth People’s Hospital of Tongji University, Shanghai 200072, China
  • Received:2025-05-18 Online:2025-09-25 Published:2025-12-09

RichHTML

0

PDF (PC)

17

摘要：

目的： 探讨肥胖病人腹腔镜胃袖状切除术（LSG）后发生急性术后高血压（APH）的高危因素，并建立预测模型。方法： 回顾性收集2021年8月至2023年12月期间在本院代谢外科接受LSG的肥胖病人资料，分析其临床数据和实验室指标。采用Logistic-LASSO回归筛选APH的独立危险因素，并基于此构建列线图预测模型。通过受试者工作特征(ROC)曲线、Bootstrap重采样、校准曲线、Hosmer-Lemeshow（H-L）检验、决策曲线分析（DCA）和临床影响曲线（CIC）评估模型的预测效能和临床适用性。结果： APH发生率为55.90%。体质量指数（BMI）、血小板计数、球蛋白、尿酸、血钠、纤维蛋白原、空腹血糖以及术前基线舒张压等具有潜在的预测价值。其中BMI（OR=1.066, 95%CI: 1.003~1.137, P=0.046）、血小板计数（OR=0.994, 95%CI: 0.998~0.999, P=0.027）、纤维蛋白原（OR=1.943, 95%CI: 1.128~3.479, P=0.02）和术前基线舒张压（OR=0.953, 95%CI: 0.918~0.985, P=0.006）为独立高危因素。列线图模型曲线下面积(AUC)为0.783（95% CI: 0.711~0.855），灵敏度0.817，特异度0.689。Bootstrap重抽样AUC为0.776（95%CI: 0.702~0.849）。H-L检验P>0.05。校准曲线显示模型拟合良好。DCA和CIC均显示模型具有较好的筛查效率。结论： BMI、血小板计数、纤维蛋白原和术前基线舒张压是LSG后发生APH的独立高危因素，构建的列线图模型具有良好的预测效能和临床实用性，有助于早期筛查和预防LSG病人发生APH。

关键词: 腹腔镜胃袖状切除术, 急性高血压, 危险因素, 列线图

Abstract:

Objective To investigate the high-risk factors associated with acute postoperative hypertension (APH) following laparoscopic sleeve gastrectomy(LSG) in obese patients and to establish a predictive model. Methods A retrospective analysis was conducted on clinical data and laboratory parameters of obese patients who underwent LSG at Department of Metabolic Surgery in our hospital from August 2021 to December 2023. Logistic-LASSO regression analysis was used to identify independent risk factors for APH. A nomogram predictive model was developed based on these factors. The predictive performance and clinical utility of the model were assessed using the receiver operating characteristic (ROC) curve, Bootstrap resampling, calibration curve, Hosmer-Lemeshow (H-L) test, decision curve analysis (DCA), and clinical impact curve (CIC). Results The incidence of APH was 55.90%. Body mass index (BMI), platelet count, globulin, uric acid, sodium, fibrinogen, fasting blood glucose, and preoperative diastolic pressure had potential predictive value. Among them, BMI (OR=1.066, 95% CI: 1.003-1.137, P=0.046), platelet count (OR=0.994, 95% CI: 0.998-0.999, P=0.027), fibrinogen (OR=1.943, 95% CI: 1.128-3.479, P=0.02), and preoperative diastolic blood pressure (OR=0.953, 95% CI: 0.918-0.985, P = 0.006) were identified as independent high-risk factors. The area under the curve (AUC) of the nomogram was 0.783 (95% CI: 0.711-0.855), with a sensitivity of 0.817 and a specificity of 0.689. The AUC based on Bootstrap resampling was 0.776 (95% CI: 0.702-0.849). The H-L test yielded P>0.05, and the calibration curve showed good model fit. Both DCA and CIC demonstrated favorable screening efficiency. Conclusions BMI, platelet count, fibrinogen, and preoperative diastolic blood pressure are independent high-risk factors for APH following LSG. The developed nomogram model exhibits good predictive performance and clinical applicability, providing a valuable tool for early screening and prevention of APH in LSG patients.

Key words: Laparoscopic sleeve gastrectomy (LSG), Acute hypertension, Risk factor, Nomogram

中图分类号: 

Baidu
map