作者：徐娅静，戚洪亮，丰陈

单位：

1.230031安徽省芜湖市，中国人民解放军联勤保障部队第901医院麻醉科　 2.211299江苏省南京市溧水区人民医院麻醉科 通信作者：丰陈，E-mail：10224961@qq.com

单位（英文）：

1.Department of Anesthesiology, 901 Hospital of Joint Logistics Support Force of Chinese People's Liberation Army, Wuhu 230031, China 2.Department of Anesthesiology, People's Hospital of Lishui District, Nanjing, Nanjing 211299, China Corresponding author: FENG Chen, E-mail: 10224961@qq.com

关键词：

呼吸,人工; 机械通气; 重症监护病房; 成年人; 撤机失败; 综合肺指数; 预测;

关键词（英文）：

Respiration, artificial; Mechanical ventilations; Intensive care units; Adult; Withdrawal failure; Integratedpulmonary index; Forecasting

中图分类号：

DOI：

10.12114/j.issn.1008-5971.2022.00.330

基金项目：

南京市卫生科技发展专项（YKK20180）；江苏大学临床医学科技发展基金项目（JLY20180218）

摘要：

目的 探讨综合肺指数（IPI）对成人ICU机械通气患者撤机失败的预测价值。方法 选取2021年5月至2022年5月于中国人民解放军联勤保障部队第901医院ICU住院，因各种原因导致呼吸衰竭并行机械通气的患者109例，根据撤机结果将患者分为撤机成功组（n=80）和撤机失败组（n=29）。比较撤机成功组与撤机失败组的临床资料。采用多因素Logistic回归分析探讨成人ICU机械通气患者撤机失败的影响因素。采用ROC曲线评估IPI、入ICU时急性生理与慢性健康评价系统Ⅱ（APACHE Ⅱ）评分及两者联合对成人ICU机械通气患者撤机失败的预测价值。根据IPI将患者分为IPI<4分组（n=35）和IPI≥4分组（n=74），比较IPI<4分组和IPI≥4分组治疗时间及气管切开率。结果撤机成功组入ICU时APACHE Ⅱ评分低于撤机失败组，呼吸频率（RR）、脉率（PR）慢于撤机失败组，血氧饱和度（Pet CO2）、IPI高于撤机失败组（P<0.05）。多因素Logistic回归分析结果显示，入ICU时APACHE Ⅱ评分、IPI为成人ICU机械通气患者撤机失败的独立影响因素（P<0.05）。ROC曲线分析结果显示，入ICU时APACHE Ⅱ评分和IPI预测成人ICU机械通气患者撤机失败的AUC分别为0.760[95%CI(0.652,0.869)]和0.875[95%CI(0.802,0.947)]，最佳截断值分别为17分、4分，灵敏度分别为0.750、0.756，特异度分别为0.750、0.854，两者联合预测成人ICU机械通气患者撤机失败的AUC为0.891[95%CI(0.821,0.961)]。IPI<4分组和IPI≥4分组总住院时间比较，差异无统计学意义（P>0.05）;IPI<4分组机械通气时间、ICU住院时间长于IPI≥4分组，气管切开率高于IPI≥4分组（P<0.05）。结论IPI对成人ICU机械通气患者撤机失败有一定预测效能，且联合入ICU时APACHE Ⅱ评分可提高其预测效能。

英文摘要：

【Abstract】　Objective To investigate the predictive value of integrated pulmonary index (IPI) on withdrawal failure in adult patients with mechanical ventilation in ICU. Methods A total of 109 patients admitted to ICU of 901 Hospital of Joint Logistics Support Force of Chinese People's Liberation Army from May 2021 to May 2022 and treated with mechanical ventilation for respiratory failure due to various reasons were selected. According to the withdrawal results, patients were divided into weaning success group (n=80) and weaning failure group (n=29) . The clinical data of the weaning success group and the weaning failure group were compared. Multivariate Logistic regression analysis was used to investigate the influencing factors of withdrawal failure of adult patients with mechanical ventilation in ICU. ROC curve was used to evaluate the predictive value of IPI, acute physiology and chronic health evaluation Ⅱ (APACHEⅡ) scores on admission to ICU and their combination for withdrawal failure in adult patients with mechanical ventilation in ICU. According to IPI, the patients were divided into IPI < 4 group (n=35) and IPI ≥ 4 group (n=74) , and the treatment time and tracheotomy rate of IPI < 4 group and IPI ≥ 4 group were compared. Results The APACHEⅡ score on admission to ICU of the weaning success group was lower than that of the weaning failure group, respiration rate (RR) and pulse rate (PR) were slower than those of the weaning failure group, and end-tidal carbon dioxide partial pressure (PetCO2) and IPI were higher than those of the weaning failure group (P < 0.05) . The results of multivariate Logistic regression analysis showed that APACHEⅡscore on admission to ICU and IPI were independent influencing factors of withdrawal failure of adult patients with mechanical ventilation in ICU (P < 0.05) . ROC curve analysis results showed that the AUC of APACHE Ⅱscore on admission to ICU and IPI for predicting withdrawal failure of adult patients with mechanical ventilation in ICU was 0.760 [95%CI (0.652, 0.869) ] and 0.875 [95%CI (0.802, 0.947) ] , the optimal cut-off values were 17 points and 4 points, the sensitivity was 0.750 and 0.756, specificity was 0.750 and 0.854, respectively. The AUC of their combination for predicting withdrawal failure of adult patients with mechanical ventilation in ICU was 0.891 [95%CI (0.821, 0.961) ] . There was no significant difference in total hospital stay between IPI < 4 group and IPI ≥ 4 group (P > 0.05) . The mechanical ventilation time and ICU stay time in IPI < 4 group were longer than those in IPI ≥ 4 group, and the tracheotomy rate was higher than that in IPI ≥ 4 group (P < 0.05) . Conclusion IPI has a certain ability to predict withdrawal failure of adult patients with mechanical ventilation in ICU, and the combination of APACHE Ⅱ score on admission to ICU can improve its predictive efficiency.

参考文献：