骨盆骨折患者在机器人辅助经皮通道螺钉固定术后感染的危险因素分析及预测模型研究OA

Objective:To investigate the risk factors for surgical site infection after robot-assisted percutaneous screw fixation in patients with pelvic fractures and to construct a nomogram prediction model.Methods:97 patients who underwent robot-assisted percutaneous screw fixation at the First Affiliated Hospital of Air Force Medical University from October 2020 to March 2023 were enrolled.They were divided into the infection group(n=41)and the non-infection group(n=56)based on the occurrence of postoperative infection.Multivariate Logistic regression analysis was used to identify risk factors for postoperative infection and to construct a predictive model.The predictive accuracy and clinical utility of the model were evaluated.Results:A total of 41 strains of pathogens were isolated from patients in the infection group,primarily Staphylococcus aureus,Escherichia coli,Staphylococcus epidermidis,and Pseudomonas aeruginosa.Multivariate analysis showed that BMI,injury severity score(ISS),concomitant traumatic brain injury,acetabular fracture,and abdominal injury were risk factors for postoperative infection in patients with pelvic fractures(P<0.05).A predictive model was constructed based on these risk factors,and the area under the receiver operating characteristic curve was 0.976(95%CI:0.945-1.000).The calibration curve was close to the ideal curve,indicating good discriminative ability and clinical efficacy of the model.Further decision curve analysis showed that when the risk threshold predicted by the model was>0.05,it could provide a net clinical benefit.Conclusion:BMI,ISS score,concomitant traumatic brain injury,acetabular fracture,and abdominal injury are risk factors for postoperative infection in patients with pelvic fractures.The nomogram model constructed based on these factors can provide an intuitive visualization tool for early clinical assessment of infection risk after pelvic fracture surgery.