重型颅脑损伤气管切开患者再发肺部感染的影响因素分析及列线图预测模型构建

Analysis of Influencing Factors of Recurrent Pulmonary Infection in Patients with Severe Craniocerebral Injury Undergoing Tracheotomy and Construction of Nomogram Prediction Model

目的分析重型颅脑损伤气管切开患者再发肺部感染的影响因素分析并构建列线图预测模型.方法回顾性分析2020年12月—2022年4月滨州医学院附属医院康复医学科重症康复单元收治的118例重型颅脑损伤气管切开患者的临床资料,依据是否再发肺部感染分为两组,分别是感染组和非感染组.采用单因素分析对重型颅脑损伤气管切开患者再发肺部感染的影响因素进行分析,再将单因素分析中差异有统计学意义的因素纳入多因素Logistic回归分析,筛选出重型颅脑损伤气管切开患者再发肺部感染的独立影响因素,之后据此构建列线图预测模型,并对预测模型的效能进行验证.结果118例重型颅脑损伤气管切开患者中再发肺部感染患者61例,再发肺部感染率为51.69%.多因素Logistic回归分析结果显示,低血清白蛋白、高D-二聚体、抗生素联合应用是重型颅脑损伤气管切开患者再发肺部感染的独立危险因素(P<0.05),康复治疗方案中含体外膈肌起搏器治疗可降低再发肺部感染的风险(P<0.05).对列线图预测模型进行分析得出,其拟合优度良好(P=0.380),受试者工作特征曲线的曲线下面积为0.915.结论低血清白蛋白、高D-二聚体、抗生素联合应用是重型颅脑损伤气管切开患者再发肺部感染的独立危险因素,体外膈肌起搏器治疗是重型颅脑损伤气管切开患者再发肺部感染的保护因素,在多因素Logistic回归分析基础上建立列线图预测模型对再发肺部感染具有良好的预测效能.

Objective To analyze the influencing factors of recurrent pulmonary infection in patients with severe craniocerebral injury undergoing tracheotomy and to construct a nomogram prediction model.Methods The clinical data of 118 patients with severe craniocerebral injury undergoing tracheotomy in the Intensive Rehabilitation Unit of the Department of Rehabilitation Medicine of Binzhou Medical University Hospital from December 2020 to April 2022 were retrospectively analyzed,and they were divided into two groups based on whether they had a recurring lung infection,an infected group and a non-infected group,respectively.The factors influencing the recurrence of pulmonary infection in patients with severe craniocerebral injury undergoing tracheotomy were analyzed using univariate analysis,the factors with statistically significant differences in the univariate analysis were then included in the multi-factor logistic regression analysis,to screen out the independent influencing factors of recurrent pulmonary infection in patients with severe craniocerebral injury undergoing tracheotomy,and then the nomogram prediction model was constructed accordingly,and the efficacy of the nomogram prediction model was validated.Results There were 61 patients with recurrent pulmonary infection among 118 patients with severe craniocerebral injury undergoing tracheotomy,and the rate of recurrent pulmonary infection was 51.69%.Multi-factor logistic regression analysis showed that low serum albumin,high D-dimer,and antibiotic combination were independent risk factors for recurrent pulmonary infection in patients with heavy craniocerebral injury undergoing tracheotomy(P<0.05),rehabilitation regimens containing extracorporeal diaphragm pacemaker therapy reduced the risk of recurrent pulmonary infections(P<0.05).The nomogram prediction model was analyzed,and the goodness of fit was good(P=0.380),and the area under curve of receiver operating characteristic curve was 0.915.Conclusion Low serum albumin,high D-dimer,and antibiotic combination were independent risk factors for recurrent pulmonary infection in patients with heavy craniocerebral injury undergoing tracheotomy,rehabilitation regimens containing extracorporeal diaphragm pacemaker therapy is a protective factor for recurrent pulmonary infection in patients with severe craniocerebral injury undergoing tracheotomy,the nomogram prediction model based on multivariate Logistic regression analysis has good predictive efficiency for recurrent pulmonary infection.