探讨脑出血患者气管切开并发肺部感染的预测方法及价值。

选取亳州市人民医院重症医学科自2018年5月至2020年2月收治的142例脑出血气管切开患者为研究对象，按照脑出血气管切开患者并发肺部感染情况，分为发生组和未发生组。分析脑出血气管切开患者肺部感染危险因素，对比2组患者气管切开术后次日血清白蛋白（Alb）、降钙素原（PCT）及D-二聚体（D-D）水平。采用受试者工作特征（ROC）曲线分析血清Alb、PCT、D-D水平及三者联合预测脑出血气管切开并发肺部感染的价值。

本组患者中有33例（23.24%）并发肺部感染，纳入发生组，余109例患者纳入未发生组。发生组脑出血气管切开患者的血清Alb水平[（34.28±7.61）g/L]明显低于未发生组[（48.15±9.27）g/L]，PCT[（0.25±0.06）ng/mL]和D-D水平[（253.16±41.27）μg/L]均明显高于未发生组[（0.17±0.05）ng/mL，（168.41±35.24）μg/L]，差异均具有统计学意义（P<0.05）。慢性阻塞性肺疾病、吸烟史、低蛋白血症、气管切开距离发病时间≥5 d、气管插管、使用呼吸机、应用广谱抗菌药物、进行颅脑手术、血清Alb<40 g/L、PCT≥0.15 ng/L、D-D≥200 μg/L均是影响出血性气管切开患者肺部感染发生的独立危险因素（P<0.05），入院GCS评分是其保护因素（OR=0.551，P=0.023）；ROC分析显示，血清Alb、PCT、D-D水平预测脑出血气管切开患者并发肺部感染的最佳截断点分别为36.29 g/L、0.23 ng/mL、228.05 μg/L；血清Alb、PCT及D-D三者联合预测脑出血气管切开患者并发肺部感染的灵敏度、特异度、曲线下面积（AUC）分别为69.70%、97.25%、0.912，特异度和AUC均高于单独预测（P<0.05），灵敏度与单独预测差异无统计学意义（P>0.05）。

脑出血气管切开患者并发肺部感染风险高，且慢性阻塞性肺疾病、吸烟史、低蛋白血症等与血清Alb<40 g/L、PCT≥0.15 ng/L、D-D≥200 μg/L均是其影响因素，血清Alb、PCT、D-D水平联合可用于预测肺部感染。

To explore the predictive method and value of tracheotomy complicated with pulmonary infection in hemorrhagic stroke.

One hundred and forty-two patients with hemorrhagic stroke undergoing tracheotomy from May 2018 to February 2020 in Critical Care Medicine Department of Bozhou People's Hospital were enrolled as subjects. They were divided into the occurrence group and the non occurrence group according whether pulmonary infection happened or not. The risk factors of pulmonary infection in patients with intracerebral hemorrhage after tracheotomy were analyzed, and the serum albumin (Alb), procalcitonin (PCT) and D-dimer (D-D) levels were compared between the two groups on the next day after tracheotomy. The receiver operating characteristic (ROC) curve was used to analyze the serum Alb, PCT, D-D levels and the value of the combination of the three in predicting pulmonary infection after tracheotomy after intracerebral hemorrhage.

Thirty-three patients (23.24%) in this group were complicated with pulmonary infection, and were included in the occurrence group, while 109 patients in the non-occurrence group. The serum Alb level of patients with hemorrhagic stroke after tracheotomy in the occurrence group was significantly lower than that of the non-occurrence group [(34.28±7.61) g/L vs (48.15±9.27) g/L], and the levels of PCT and D-D were significantly higher than that of the non-occurrence group [(0.25±0.06) ng/mL vs (0.17±0.05) ng/mL; (253.16±41.27) μg/L vs (168.41±35.24) μg/L] (P<0.05). Chronic obstructive pulmonary disease, smoking history, hypoproteinemia, tracheotomy distance and onset time ≥ 5 d, tracheal intubation, ventilator use, broad-spectrum antibiotics, perform craniocerebral surgery, serum ALb<40 g/L, PCT≥0.15 ng/L, D-D≥200 μg/L were independent risk factors of pulmonary infection in patients with hemorrhagic stroke undergoing tracheotomy (P<0.05), and admission GCS score was also an influencing factor (OR=0.551, P=0.023). ROC analysis showed that the optimal cut-off points for serum Alb, PCT, and D-D levels to predict pulmonary infection in patients with hemorrhagic stroke complicated by tracheotomy were 36.29 g/L, 0.23 ng/mL, 228.05 μg/L, respectively. The sensitivity, specificity and area under curve (AUC) of serum ALb, PCT and D-D combination were 69.70%, 97.25% and 0.912 respectively. The specificity and AUC were higher than those of serum ALb, PCT and D-D alone (P<0.05), and the sensitivity was not statistically significant different from those alone (P>0.05).

Patients with intracerebral hemorrhage undergoing tracheotomy have a high risk of pulmonary infection, and chronic obstructive pulmonary disease, smoking history, hypoproteinemia, etc. and serum Alb<40 g/L, PCT≥0.15 ng/L, and D-D≥200 g/L are the influencing factors, and the combination of serum ALb, PCT and D-D levels can be used to predict pulmonary infection.