《Journal of Oral and Maxillofacial Surgery》 ›› 2023, Vol. 33 ›› Issue (6): 385-390. doi: 10.12439/kqhm.1005-4979.2023.06.005

Previous Articles     Next Articles

Construction and validation of risk prediction model for postoperative infection in OSCC patients after radical resection

ZHANG Jin1,2,3(), CHENG Zheng1,2,3, YAN Xiaojing4, LEI Bing1,2,3()   

  1. 1 College of Stomatology, Xi'an Jiaotong University, Key Laboratory of Shaanxi Province Craniofacial Precision Medicine Research, Xi'an 710004
    2 College of Stomatology, Xi'an Jiaotong University, Clinical Research Center of Shaanxi Province for Dental and Maxillofacial Disease, Xi'an 710004
    3 Department of General Dentistry, College of Stomatology, Xi'an Jiaotong University, Xi'an 710004
    4 Shandong Provincial Third Hospital, Jinan 250031, China
  • Received:2023-02-14 Accepted:2023-04-26 Online:2023-12-28 Published:2023-12-26

口腔鳞状细胞癌患者根治性切除术后感染的风险预测模型构建与验证

张瑾1,2,3(), 程政1,2,3, 闫小静4, 雷冰1,2,3()   

  1. 1 西安交通大学口腔医院,陕西省颅颌面精准医学研究重点实验室,西安 710004
    2 西安交通大学口腔医院,陕西省牙颌疾病临床研究中心,西安 710004
    3 西安交通大学口腔医院综合科,西安 710004
    4 山东省立第三医院,济南 250031
  • 通讯作者: 雷冰,主管护师. E-mail:362968122@qq.com
  • 作者简介:
    张瑾,住院医师. E-mail:

Abstract:

Objective: To analyze the risk factors of postoperative infection in patients with oral squamous cell carcinoma (OSCC) after radical resection, to establish an infection risk prediction model, and to verify the model externally to explore its predictive value. Methods: A total of 568 patients with OSCC after radical resection in the Department of Oral and Maxillofacial Surgery of Xi'an Jiaotong University Stomatological Hospital from January 2017 to December 2022 were retrospectively collected. They were randomly divided into the modeling group (386 cases) and the validation group (182 cases), and their clinical data were collected. The multiple factor Logistic regression model was used to analyze the influencing factors of infection in patients, and R software was used to build a line graph prediction model for internal and external verification. Results: Multivariate Logistic regression analysis showed that age ≥60 years, American Society of Anesthesiologists (ASA) grade ≥Ⅱ, diabetes mellitus, tracheotomy, and operation time >260 min were independent risk factors of infection after radical resection in OSCC patients (P<0.05). The results of the Hosmer-Lemeshow test were χ2=8.529, P=0.384, which suggested that the model had good discrimination and accuracy. The area under the curve (AUC) of receiver operating characteristic (ROC) were 0.805 and 0.794 in the modeling group and the validation group respectively. The calibration curve showed that the model has good discrimination and accuracy. Conclusion: The nomogram constructed by age, ASA grade, diabetes mellitus, tracheotomy and time length of surgery can accurately predict the risk of infection in patients with OSCC after radical resection, with high clinical application value.

Key words: oral squamous cell carcinoma, radical resection, infected, risk factors, prediction model

摘要:

目的:分析口腔鳞状细胞癌(oral squamous cell carcinoma,OSCC)患者根治性切除术后的感染危险因素,建立感染风险预测模型,并对该模型进行外部验证,探讨其预测价值。方法:回顾性收集568例2017年1月—2022年12月西安交通大学口腔医院口腔颌面外科收治的OSCC根治性切除术后的患者,将患者随机分为建模组(386例)和验证组(182例),收集其临床资料。采用多因素Logistic回归模型构建术后感染的影响因素,并应用R软件建立列线图预测模型,进行内部和外部验证。结果:多因素Logistic回归分析结果显示,年龄≥60岁、美国麻醉医师协会(American Society of Anesthesiologists,ASA)分级≥Ⅱ级、合并糖尿病、气管切开和手术时长>260 min是OSCC患者根治性切除术后感染的独立危险因素(P<0.05)。经Hosmer-Lemeshow拟合度检验结果显示,χ2=8.529,P=0.384,模型拟合的准确度好。受试者工作特征(receiver operating characteristic,ROC)曲线显示该模型在建模组与验证组预测OSCC患者根治性切除术后感染风险的曲线下面积(area under the curve,AUC)分别为0.805和0.794。校准曲线提示该模型区分度、准确度较好。结论:年龄、ASA分级、糖尿病、气管切开和手术时长所构建的列线图能够较准确地预测OSCC患者根治性切除术后感染的风险,临床应用价值较高。

关键词: 口腔鳞状细胞癌, 根治性切除术, 感染, 危险因素, 预测模型

CLC Number: