慢加急性肝衰竭合并肺部感染危险因素及预后分析

李晶晶.慢加急性肝衰竭合并肺部感染危险因素及预后分析[J].内科急危重症杂志,2023,29(5):386-392

扫码阅读全文

DOI：10.11768/nkjwzzzz20230507

英文关键词:

基金项目:

作者	单位	E-mail
李晶晶	华中科技大学同济医学院附属同济医院	xpzhao@tjh.tjmu.edu.cn

摘要点击次数: 408

全文下载次数: 524

中文摘要:

摘要目的：探讨慢加急性肝衰竭(ACLF)患者发生肺部感染的危险因素及其对预后的影响，并建立预后预测模型。方法：回顾性分析310例ACLF患者的临床资料，通过Logistic回归分析ACLF患者发生肺部感染的危险因素，并利用Nomogram方法建立预后预测模型。结果：ACLF患者感染发生率65.5%，其中肺部感染发生率为38.7%。ACLF合并肺部感染患者较无感染及肺外感染患者的短期病死率更高。Logistic回归多因素分析显示肺部感染的独立危险因素包括侵入性操作、使用激素治疗及高敏C反应蛋白；而年龄、总胆红素、国际标准化比率（INR）以及合适的抗生素治疗是影响肺部感染患者30d病死率的独立危险因素。纳入相应的危险因素，建立名为TAIST的Nomogram预测模型，其受试者工作特征曲线下面积为0.844(95%CI 0.741～0.946)，具有比其他预后模型更高的判别性能。结论：ACLF合并肺部感染患者预后差，TAIST模型有助于早期识别其危险因素并优化治疗策略。

英文摘要:

Abstract Objective: To explore the risk factors and prognostic characteristics of acute-on-chronic liver failure (ACLF) patients with pulmonary infection and establish a Nomogram for predicting 30-day mortality. Method:Clinical data of 310 ACLF patients were analyzed retrospectively. The potential risk factors of developing pulmonary infection were studied by logistic regression and a prognostic Nomogram was established to predict the 30-day mortality of ACLF patients with pulmonary infection. Result: The incidence of infections was 65.5% and of which lung infection was 38.7%. Higher short-term mortality was observed in patients with pulmonary infection than in those without infections or with infections at other sites. The independent risk factors of pulmonary infection included invasive procedures, glucocorticoid treatment and hsCRP by multivariate logistic regression analyses. In addition, age, total bilirubin, INR and appropriate empirical antibiotic strategies were shown to be independent risk factors for 30-day mortality of ACLF patients with pulmonary infection. Subsequently, a Nomogram named TAIST was constructed including these independent risk factors. The Nomogram showed a higher discriminative performance than other prognostic models, with an area under the receiver operating characteristic curve (AUC) of 0.845 (95%CI: 0.744-0.945). Conclusion: ACLF patients combined with pulmonary infection had poor prognosis. TAIST model may be helpful for early recognition of risk factors and optimized treatment strategy.