中国实用外科杂志 ›› 2023, Vol. 43 ›› Issue (05): 566-571.DOI: 10.19538/j.cjps.issn1005-2208.2023.05.17

• 论著 • 上一篇    下一篇

腹腔镜肝切除术中转开放手术危险因素分析及风险预测模型的建立#br#

黄少坚,李芷西,梁汉标,韩泽民,黄    灿,钟    权,贺    卓,王    恺,李川江   

  1. 南方医科大学南方医院普通外科学肝胆胰外科,广东广州 510000
  • 出版日期:2023-05-01 发布日期:2023-05-29

  • Online:2023-05-01 Published:2023-05-29

摘要: 目的    探讨腹腔镜肝切除术中转开放手术危险因素及构建中转开放手术的风险预测模型。方法    回顾性分析2019年6月至2022年5月南方医科大学南方医院肝胆外科收治的938例行腹腔镜肝切除术病人的临床病例信息,依据是否中转开放手术分为腹腔镜组(836例)和中转组(102例),收集所有纳入病人基本信息和临床病例资料,分析两组组间差异及因素,基于多因素分析结果构建中转开放手术列线图预测模型。结果    单因素分析结果显示,在有无腹部手术史、有无血管性介入治疗史、是否首次行肝切除、有无腹腔积液、肝肿瘤最大直径、是否为困难肝段切除、切除范围等,差异具有统计学意义(P<0.05)。多因素分析结果显示,有腹部手术史(OR=1.716,95%CI为1.023-2.878)、非首次行肝切除(OR=3.585,95%CI为1.705-7.538)、肝肿瘤最大直径≥5cm(OR=2.680,95%CI为1.646-4.363)、困难肝段切除(OR=2.953,95%CI为1.755-4.967)、肝段切除范围≥3个肝段(OR=1.901,95%CI为1.099-3.290)为腹腔镜肝切除术中转开放手术的独立危险因素(P均<0.05)。依据多因素分析结果,纳入5个独立危险因素变量,构建腹腔镜肝切除术中转开放手术的列线图模型。模型的C-index为0.746(95%CI为0.692-0.800),Hosmer-Lemeshow拟合优度检验结果P=0.541>0.05,绘制ROC曲线,曲线下面积(AUC)为0.746(95%CI为0.692-0.800,P<0.001)。绘制的临床决策曲线(DCA)结果示在阈值概率范围0.05~0.70内模型具有临床效用。结论    有腹部手术史,非首次行肝切除,肝肿瘤最大直径≥5 cm,困难肝段切除,肝段切除范围≥3个肝段是腹腔镜肝切除术中转开放手术的独立危险因素。基于此构建的列线图模型的校准曲线拟合程度理想,在预测的中转开放手术率与实际的中转开放手术率有较好的一致性,模型具有良好预测能力,但仍有待行外部验证来进一步证实。 

关键词: 肝切除术, 腹腔镜, 危险因素, 中转开放手术, 列线图模型

Abstract: Analysis of risk factors and establishment of risk prediction model for conversion to open surgery in laparoscopic hepatectomy        HUANG Shao-jian, LI Zhi-xi, LIANG Han-biao, et al. Department of Hepatobiliary and Pancreatic Surgery, Nanfang Hospital, Southern Medical University, Guangzhou 510000, China
Corresponding author: Li Chuan-jiang, E-mail: licj76@163.com
Abstract    Objective    To investigate the risk factors of conversion to laparotomy during laparoscopic hepatectomy and establish a risk prediction model for the conversion. Methods    Clinical case information of 938 patients undergoing laparoscopic hepatectomy admitted to the Department of Hepatobiliary Surgery, Nanfang Hospital of Southern Medical University from June 2019 to May 2022 was retrospectively analyzed. The patients were divided into laparoscopic group(836 cases) and conversion group(102 cases) according to whether were transferred to open surgery. Basic information and clinical case data of all included patients were collected. The differences between the two groups and their factors were analyzed, and R language was used to establish and evaluate the prediction model of conversion nomograms based on the results of multi-factor analysis. Results    The results of single-factor analysis showed that: Between the two groups of comparisons, there were statistically significant differences in the history of abdominal surgery, history of vascular interventional therapy, being or being not the first hepatic resection, presence or absence of ascites, the maximum diameter of liver tumor, difficult location hepatectomy and resection scope (P < 0.05). The results of multi-factor analysis showed that: History of abdominal surgery (OR=1.716, 95%CI 1.023-2.878), non-first-time hepatectomy (OR=3.585, 95%CI 1.705-7.538), maximum diameter of liver tumor ≥5cm (OR=2.680,95%CI 1.646-4.363), difficult location hepatectomy (OR=2.953,95%CI 1.755-4.967), and scope of hepatectomy ≥3 liver segments (OR=1.901,95%CI1.099-3.290) were independent risk factors for conversion to laparotomy during laparoscopic hepatectomy (P < 0.05). According to the results of multivariate analysis, including the five independent risk factor variables, the nomogram prediction model of laparoscopic liver resection converted to laparotomy was constructed. The C-index of the model was 0.746 (95%CI 0.692-0.800), the P value of Hosmer and Lemeshow Test was 0.541>0.05, ROC curve was drawn, and the area under the curve(AUC) was 0.746 (95%CI 0.692-0.800, P < 0.001). The results of the decision curve analysis (DCA) showed that the model had clinical utility in the threshold probability range of 0.05-0.70. Conclusion History of abdominal surgery, non-first-time hepatectomy, maximum diameter of liver tumor ≥5cm, resection of liver segments at difficult sites, and resection scope of liver segments ≥3 segments are independent risk factors for conversion to open surgery during laparoscopic hepatectomy. The calibration curve of the nomogram prediction model constructed based on this method has a good degree of fitting, and the predicted conversion rate is in good agreement with the actual conversion rate, and the nomogram prediction model has good prediction ability. However, it still needs to be further verified by external verification

Key words: hepatectomy, laparoscope, risk factors, conversion to open surgery, nomogram model