老年重症社区获得性肺炎合并心功能不全机械通气撤机失败的预测与分析  

Prediction and analysis of risk factors related to weaning failure from invasive mechanical ventilation in the elderly with severe community-acquired pneumonia complicated with cardiac dysfunction

作　　者：朱瑶丽[1] 杨嘉雯[1] 李瑶瑶 郑晓滨[2] 谭萃妍 程滔 黎宁君 钟劲[1] 陈华 陈敏英[4] Zhu Yaoli;Yang Jiawen;Li Yaoyao;Zheng Xiaobin;Tan Cuiyan;Cheng Tao;Li Ningjun;Zhong Jin;Chen Hua;Chen Minying(Department of Critical Care Unit,the Fifth Affiliated Hospital of Sun Yat-sen University,Zhuhai 519000,China)

机构地区：[1]中山大学附属第五医院重症医学科,珠海519000 [2]中山大学附属第五医院呼吸内科,珠海519000 [3]中山大学附属第五医院EICU,珠海519000 [4]中山大学附属第一医院重症医学科,广州510080

出　　处：《新医学》2020年第3期205-211,共7页Journal of New Medicine

基　　金：珠海市科技计划项目(20171009E030045)。

摘　　要：目的探索有创机械通气的老年重症社区获得性肺炎(SCAP)合并心功能不全患者撤机失败的影响因素。方法收集进行有创机械通气并符合撤机条件的105例65岁及以上的老年SCAP合并心功能不全患者临床资料,进行自主呼吸试验(SBT),撤机48 h内无需呼吸机支持纳入撤机成功组,不能通过SBT或SBT后撤机48 h内需重新上机纳入撤机失败组,对撤机失败的危险因素进行单因素及多因素Logistic回归分析,通过回归模型构建新的联合预测因子和建立受试者工作特征(ROC)曲线对比联合预测因子与各原始指标ROC曲线下面积(AUC),以约登指数最大值时确定最佳临界值,计算灵敏度、特异度及预测准确率等参数。结果撤机成功49例,撤机失败56例。撤机失败组患COPD者比例、机械通气时间≥14 d者比例、N端脑钠肽激素原(NT-proBNP)、急性生理与慢性健康状况系统评分Ⅱ(APACHEⅡ)及序贯器官衰竭(SOFA)评分高于撤机成功组(P均<0.05);撤机失败组患者CD3^+T淋巴细胞计数、CD3^+CD4^+T淋巴细胞计数、CD3^+CD8^+T淋巴细胞计数均低于撤机成功组患者(P均<0.001)。多因素Logistic回归结果显示,NT-proBNP异常(> 5000 ng/L,OR=0.132,P=0.003)、机械通气时间≥14 d(OR=15.692,P=0.001)、CD3^+CD4^+T淋巴细胞计数(OR=0.995,P=0.013)是影响撤机的主要因素。其中NT-proBNP异常(> 5000 ng/L)和机械通气时间≥14 d为危险因素,CD3^+CD4^+T淋巴细胞计数为保护因素。建立多因素回归预测模型,绘制ROC曲线,3个分类变量拟合的联合预测因子预测撤机失败的AUC高于机械通气时间≥14 d的AUC(0.894 vs. 0.710,P <0.05)。预测因子最佳临界值为0.653时,预测撤机失败的灵敏度为80.4%,特异度为89.8%,预测准确率为70.2%。撤机失败组ICU住院时间和病死率均高于撤机成功组(P均<0.05)。结论 NT-proBNP异常(> 5000 ng/L)和机械通气时间≥14 d是老年SCAP合并心功能不全患者机械通气撤机失败危险因�Objective To predict and analyze the risk factors of weaning failure from invasive mechanical ventilation in the elderly patients diagnosed with severe community-acquired pneumonia(SCAP) complicated with cardiac dysfunction. Methods Clinical data of 105 SCAP patients aged 65 years and above who received invasive mechanical ventilation and met the weaning criteria were collected. Patients who passed the spontaneous breathing trial(SBT) and did not need the ventilator within 48 h were enrolled into the weaning success group. Those who failed to pass the SBT or required the ventilator within 48 h were enrolled into the weaning failure group. The risk factors associated with weaning failure were statistically analyzed by univariate and multivariate Logistic regression analyses. Logistic regression model was formulated to construct the new combined predictive factors and delineate the receiver operating characteristic(ROC) curve. The area under the ROC curve(AUC) for both the new and original factors was statistically compared. The optimal cut-off value was obtained where the Youden index reached the maximum value. The parameters, such as sensitivity, specificity and predictive accuracy, were calculated and statistically compared. Results Fortynine patients were allocated in the weaning success group and 56 in the weaning failure group. In the weaning failure group, the proportion of chronic obstructive pulmonary disease(COPD), the proportion of duration for mechanical ventilation ≥ 14 d, N-Terminal pro-brain natriuretic peptide(NT-proBNP) level, Acute Physiology and Chronic Health Evaluation system Ⅱ (APACHE Ⅱ) and Sequential Organ Failure Assessme nt(SOFA) scores were significantly higher compared with those in the weaning success group(all P < 0.05);The CD3^+ T, CD3^+CD4^+ T and CD3^+CD8^+ T cell counts in the weaning failure group were remarkably lower than those in the weaning success group(all P < 0.001). Multivariate Logistic analysis revealed that abnormal NT-proBNP level(> 5000 ng/L, OR = 0.132, P = 0.0

关 键 词：老年 社区获得性肺炎 撤机 危险因素 预测 

分 类 号：R563.1[医药卫生—呼吸系统] R541[医药卫生—内科学]