口腔颌面缺损皮瓣修复重建手术术后出血再手术风险预测模型的构建  

Construction of postoperative bleeding reoperation risk prediction model for reconstruction of oral and maxillofacial defects with flaps

作　　者：张丽萍[1] 刘秋玲[1] 黄敏红[1] 陈佳仪 王建广[1] 王友元[1] ZHANG Li-ping;LIU Qiu-ling;HUANG Min-hong;CHEN Jia-yi;WANG Jian-guang;WANG You-yuan(Department of Oral and Maxillofacial Surgery,Sun Yat-sen Memorial Hospital,Sun Yat-sen University,Guangdong Guangzhou 510120,China)

机构地区：[1]中山大学孙逸仙纪念医院口腔颌面外科,广东广州510120

出　　处：《临床口腔医学杂志》2024年第2期79-83,共5页Journal of Clinical Stomatology

摘　　要：目的:探讨口腔颌面部缺损皮瓣修复重建手术术后出血再手术的危险因素,并建立Logistic回归预测模型。方法:回顾分析2021年1~10月在中山大学孙逸仙纪念医院口腔颌面外科住院治疗的口腔颌面部缺损皮瓣修复重建手术的769例患者资料,其中发生术后出血再手术21例。采用单因素和多因素Logistic分析手术后发生出血再手术的独立危险因素,构建风险预测模型,并通过ROC曲线检验模型的预测效果,以多因素Logistic回归中P<0.05的变量构建列线图。结果:多因素Logistic回归(逐步法)分析结果显示,糖尿病、缺损部位、手术方式是否包含颈总动脉结扎(挽救性手术)、伤口积液、伤口感染进入最终模型,其中糖尿病(OR=3.817,95%CI 1.193~11.239)、颌骨部位缺损(OR=4.313,95%CI 1.274~13.544)、伤口积液(OR=9.149,95%CI 2.998~26.911)及伤口感染(OR=5.940,95%CI 1.624~19.220)是术后出血再次手术的独立预测因子(均P<0.05)。本模型的ROC曲线下面积为0.900(95%CI 0.832,0.968),灵敏度为76.2%,特异度为91.2%,列线图C指数为0.896(95%CI 0.829~0.964),临床决策曲线结果显示在0.03~0.21的风险范围内。结论:本研究成功构建的术后出血再次手术预测模型可为口腔颌面头颈外科医护人员提供参考,指导临床采取及时有效的干预措施,为预防出血再手术提供依据。Objective:To investigate the risk factors of bleeding and reoperation after oral and maxillofacial defect flap repair and reconstruction surgery,and to establish a Logistic regression prediction model.Methods:The data of 769 patients with oral and maxillofacial defect flaps who underwent reconstruction and repair in the Department of Oral and Maxillofacial Surgery,Sun Yat-sen Memorial Hospital from January to October 2021 were retrospectively analyzed.Among them,21 patients had postoperative bleeding and reoperation.Univariate and multivariate Logistic analysis were used to analyze the independent risk factors of postoperative bleeding and reoperation,and the risk prediction model was constructed.The prediction effect of the model was tested by ROC curve,and the nomogram was constructed with variables with P<0.05 in multivariate logistic regression.Results:Multifactor logistic regression analysis showed that diabetes mellitus,defect site,surgical method including carotid artery ligation,wound effusion,and wound infection were included in the final model.Diabetes mellitus(OR=3.817,95%CI 1.193~11.239),jaw defect(OR=4.313,95%CI 1.274~13.544),wound effusion(OR=9.149,95%CI 2.998~26.911)and wound infection(OR=5.940,95%CI 1.624~19.220)were independent predictors of reoperation for postoperative bleeding(P<0.05).The area under the ROC curve of this model was 0.900(95%CI 0.832,0.968),the sensitivity was 76.2%,the specificity was 91.2%,and the nomogram C index was 0.896(95%CI 0.829~0.964),the clinical decision curve results showed a risk range of 0.03~0.21.Conclusion:The model constructed in this study has a good prediction effect,which can provide a reference for medical staff of oral and maxillofacial head and neck surgery,guide the clinical implementation of timely and effective intervention measures,and provide a basis for the prevention of bleeding reoperation.

关 键 词：口腔颌面部缺损 皮瓣 修复重建 术后出血 风险预测模型 列线图 

分 类 号：R782.2[医药卫生—口腔医学]