食管癌根治性切除术后吻合口瘘发生危险因素分析及预测模型的构建

Analysis of risk factors and construction of prediction model of occurrence of anastomotic leakage after radical resection of esophageal cancer

目的:探讨食管癌根治性切除术后吻合口瘘发生的危险因素,并构建列线预测模型以预测吻合口瘘发生风险。方法:回顾性分析行食管癌根治性切除术的102例患者的临床资料,包括性别、年龄等一般资料和手术时间、吻合方式、食管替代器官等围术期指标;按术后是否发生吻合口瘘将患者分为吻合口瘘组(n=19)和非吻合口瘘组(n=83)。采用单因素及Logsitic回归模型分析相关指标与食管癌根治性切除患者术后发生吻合口瘘的关系,并依据筛选出的独立影响因素构建列线图预测模型;采用受试者工作特征曲线(ROC)分析预测模型的预测能力。结果:单因素分析结果显示,美国麻醉医师协会(ASA)分级、胸部手术史、呼吸系统合并症、手术时间、吻合方式、术后呼吸系统并发症发生情况、术后切口愈合情况等均与术后发生吻合口瘘相关(P<0.05)。Logistic回归分析显示,胸部手术史(OR=2.368)、呼吸系统合并症(OR=3.108)、单层手工吻合(OR=2.457)、术后呼吸系统并发症(OR=2.956)、术后切口愈合不良(OR=3.842)均是食管癌根治性切除术后患者发生吻合口瘘的独立影响因素(P<0.05)。ROC曲线分析显示,根据独立影响因素构建列线图预测模型的曲线下面积(AUC)为0.802。结论:胸部手术史、呼吸系统合并症、单层手工吻合、术后呼吸系统并发症、术后切口愈合不良均是食管癌根治性切除患者术后发生吻合口瘘的独立危险因素,基于以上危险因素构建的列线图预测模型对术后吻合口漏发生风险的预测价值较高。

Objective:To investigate the risk factors of occurrence of anastomotic leakage after radical resection of esophageal cancer,and to construct a nomographic prediction model to predict the risk of anastomotic leakage.Methods:A retrospective analysis was performed on the clinical data of 102 patients who underwent radical resection of esophageal cancer,including general data of gender,age and perioperative indicators of surgical time,anastomosis method and esophageal replacement organs.Patients were divided into the anastomotic leakage group(n=19)and the non-anastomotic leakage group(n=83)according to whether anastomotic leakage occurred after surgery.Univariate analysis and Logsitic regression model analysis were used to analyze the relationship between relevant indicators and occurrence of anastomotic leakage in patients after radical resection of esophageal cancer,and nomographic prediction model was constructed based on the selected independent influencing factors,and receiver operating characteristic curve(ROC)was used to analyze the predictive ability of the prediction model.Results:Univariate analysis showed that ASA grade,history of chest surgery,respiratory comorbidities,surgical time,anastomosis method,occurrence of postoperative respiratory complications and postoperative incision healing were all related to the occurrence of postoperative anastomotic leakage(P<0.05).Logistic regression analysis showed that the history of chest surgery(OR=2.368),respiratory comorbidities(OR=3.108),single layer manual anastomosis(OR=2.457),postoperative respiratory complications(OR=2.956)and poor healing of postoperative incision(OR=3.842)were independent influencing factors for the occurrence of anastomotic leakage in patients after radical resection of esophageal cancer(P<0.05).ROC curve analysis showed that the area under the curve(AUC)of the nomographic prediction model constructed based on independent influencing factors was 0.802.Conclusion:The history of chest surgery,respiratory comorbidities,single-layer manual anastomosis,postoperative respiratory complications,and poor healing of postoperative incision are all independent risk factors for the occurrence of anastomotic leakage in patients after radical resection of esophageal cancer.The nomographic prediction model constructed based on the above risk factors has a good predictive ability in predicting the risk of postoperative anastomotic leakage.