Xi Ying-Long, Li Bin, Pu Lun-Qing, Ma Tao, Luo Xing-Peng
Southern Central Hospital of Yunnan Province, Yunnan, China.
The First People's Hospital of Longquanyi District, Sichuan, China.
Medicine (Baltimore). 2025 Jul 18;104(29):e43397. doi: 10.1097/MD.0000000000043397.
This study explores the risk factors of postoperative infection of intertrochanteric fracture of the femur to construct and validate the prediction model, so as to provide a basis for clinical early intervention. Clinical data of patients with intertrochanteric fracture of the femur treated in our hospital from January 2020 to December 2024 were retrospectively collected. Independent risk factors were screened by univariate and multivariate logistic regression analyses, and a nomogram was constructed based on the regression model. The predictive efficacy was assessed by calculating the area under the curve using the receiver operating characteristic curve, and the model fit was verified by the Hosmer-Lemeshow goodness-of-fit test. In addition, the Bootstrap method (repeated sampling 1000 times) combined with 10-fold cross-validation was utilized to enhance the stability of the model, while calibration curves were plotted and decision curves were analyzed to assess the predictive accuracy and clinical utility value of the model. Five hundred and sixty-four patients were finally included, of which 36 cases developed postoperative infections (infection rate 6.38%). Results of univariate and multivariate logistic regression analyses showed that intensive care unit admission (odds ratio [ OR[, 6.283; 95% confidence interval [CI], 2.233-17.463), age > 75 years (OR, 2.793; 95% CI, 1.193-6.513), urinary catheter time > 24 h (OR, 3.563; 95% CI, 1.223-8.543), surgery time > 2 h (OR, 3.330; 95% CI, 1.200-7.880), and use of steroids (OR, 3.010; 95% CI, 1.150-7.400) were independent risk factors for postoperative infection. Postoperative infection of intertrochanteric femoral fracture is influenced by multiple factors, and the prediction model constructed in this study showed high predictive accuracy and clinical utility, which can provide strong support for early identification and intervention of patients at high risk of postoperative infection.
本研究探讨股骨粗隆间骨折术后感染的危险因素，构建并验证预测模型，为临床早期干预提供依据。回顾性收集2020年1月至2024年12月在我院接受治疗的股骨粗隆间骨折患者的临床资料。通过单因素和多因素logistic回归分析筛选独立危险因素，并基于回归模型构建列线图。采用受试者工作特征曲线计算曲线下面积评估预测效能，通过Hosmer-Lemeshow拟合优度检验验证模型拟合情况。此外，利用Bootstrap法（重复抽样1000次）结合10倍交叉验证提高模型稳定性，绘制校准曲线并分析决策曲线以评估模型的预测准确性和临床实用价值。最终纳入564例患者，其中36例发生术后感染（感染率6.38%）。单因素和多因素logistic回归分析结果显示，入住重症监护病房（比值比[OR]，6.283；95%置信区间[CI]，2.233 - 17.463）、年龄>75岁（OR，2.793；95%CI，1.193 - 6.513）、导尿时间>24小时（OR，3.563；95%CI，1.223 - 8.543）、手术时间>2小时（OR，3.330；95%CI，1.200 - 7.880）以及使用类固醇（OR，3.010；95%CI，1.150 - 7.400）是术后感染的独立危险因素。股骨粗隆间骨折术后感染受多种因素影响，本研究构建的预测模型具有较高的预测准确性和临床实用性，可为术后感染高危患者的早期识别和干预提供有力支持。
