临床外科杂志 ›› 2024, Vol. 32 ›› Issue (6): 644-647.doi: 10.3969/j.issn.1005-6483.2024.06.024

• 论著 • 上一篇    下一篇

创伤性胫骨平台骨折术后感染的危险因素分析及Nomogram预测模型构建

张伟 韦昌辉 卢俊浩 彭晨健 柏茂盛 王军   

  1. 210000 南京市中医院骨伤科
  • 收稿日期:2023-05-11 出版日期:2024-06-20 发布日期:2024-06-20
  • 通讯作者: 王军,Email:wangjun198607@163.com

Analysis of risk factors for postoperative infection in traumatic tibial plateau fractures and constr uction of a Nomogram prediction model

ZHANG Wei,WEI Changhui,LU Junhao,PENG Chenjian,BAI Maosheng,WANG Jun   

  1. Department of Orthopedics and Traumatology,Traditional Chinese Medicine Hospital,Nanjing 210000,China
  • Received:2023-05-11 Online:2024-06-20 Published:2024-06-20

摘要: 目的 分析创伤性胫骨平台骨折术后感染的危险因素,并构建Nomogram预测模型。方法 2019年10月~2021年8月行手术治疗的创伤性胫骨平台骨折病人148例,根据术后是否出现感染分为感染组(20例)和未感染组(128例)。比较两组病人的一般资料;采用单因素分析及多因素Logistic回归分析影响创伤性胫骨平台骨折术后感染的危险因素;采用ROC分析有统计学意义的连续性变量的预测价值;采用内部数据验证Nomogram模型预测效能。结果 两组病人年龄、性别等一般资料比较,差异无统计学意义(P>0.05),与未感染组相比,感染组病人糖尿病、开放型骨折类型、骨筋膜室综合征占比较高,手术时间和住院时间较长(P<0.05);糖尿病(是)、骨折类型(开放型)、骨筋膜室综合征(有)、手术时间(>3小时)是影响创伤性胫骨平台骨折术后感染的危险因素;手术时间和住院时间的AUC为0.792、0.651;最佳截断值为3小时、13天(P<0.05);Nomogram模型预测创伤性胫骨平台骨折术后感染风险的C-index 为0.744(0.651~0.807);模型预测创伤性胫骨平台骨折术后感染风险的阈值>0.09。结论 糖尿病(是)、骨折类型(开放型)、骨筋膜室综合征(有)、手术时间(>3小时)是影响创伤性胫骨平台骨折术后感染的危险因素,且基于以上变量构建的Nomogram模型有较好的预测价值。

关键词: 创伤性;胫骨平台骨折;术后感染;Nomogram预测模型

Abstract: Objective To analyse the risk factors for postoperative infection in traumatic tibial plateau fractures and to construct a Nomogram prediction model.Methods One hundred and forty-eight patients with traumatic tibial plateau fractures who underwent surgery in our hospital from October 2019 to August 2021 were selected for the study,and were divided into an infected group (n=20) and an uninfected group (n=128) according to whether they developed infection after surgery.The general data of the two groups were compared;the predictive value of statistically significant continuous variables was analysed using the ROC experiment;the risk factors affecting postoperative infection in traumatic tibial plateau fractures were analysed using the logistic regression experiment;and the clinical efficacy of the Nomogram model was verified using internal data.Results In the comparison of general data such as age and gender between the two groups,the differences were not statistically significant(P>0.05).Compared with the uninfected group,patients in the infected group had a higher percentage of diabetes mellitus,open fracture type,and osteofascial compartment syndrome,and longer operative time and hospital stay(P<0.05);diabetes (yes),fracture type (open),osteofascial compartment syndrome (yes),and operative time (>3 h) were risk factors affecting postoperative infection in traumatic tibial plateau fractures.The AUCs for operative time and hospital stay were not 0.792 and 0.651;the optimal stage values were not 3 h and 13 d(P<0.05);the Nomogram model predicted a C-index of 0.744 (0.651-0.807) for the risk of postoperative infection in traumatic tibial plateau fractures.The model predicted the risk of infection after traumatic tibial plateau fracture at a threshold of > 0.09.Conclusion Diabetes mellitus (yes),fracture type (open),osteofascial compartment syndrome (yes),and operative time (>3 h) were risk factors affecting postoperative infection in traumatic tibial plateau fractures,and the Nomogram model constructed based on the above variables had good predictive value.

Key words: traumatic;tibial plateau fracture;postoperative infection;Nomogram prediction model

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!