自主呼吸频率变异对ICU患者临床结局的影响

Effect of spontaneous respiratory rate variability on clinical outcomes of the patients in intensive care unit

目的探究自主呼吸频率变异对ICU患者临床结局的影响。方法采用回顾性队列研究,从重症监护合作研究数据库(eICU-CRD)中提取首次入ICU且有入ICU后首个24 h内自主呼吸频率记录的成年患者,提取患者的年龄、性别、种族、首次急性生理与慢性健康状况评价Ⅳ、疾病诊断、药物使用及自主呼吸频率记录。根据患者入ICU后首个24 h内自主呼吸频率变异系数(coefficient of variation,CV),将患者分为低呼吸频率变异组(CV≤20%)和高呼吸频率变异组(CV>20%)。卡方检验比较两组患者ICU病死率及医院病死率。采用Logistic回归模型分析呼吸频率变异与临床结局的相关性,并进一步在重症监护数据库(MIMIC-Ⅱ)中进行外部验证。结果本研究共纳入40229例患者,低呼吸频率变异组ICU病死率(2.7%vs.2.4%,P=0.043)和医院病死率(5.6%vs.5.1%,P=0.030)均比高呼吸频率变异组高。Logistic回归分析显示,低呼吸频率变异是ICU病死率(OR=1.211,95%CI 1.051~1.396)及医院病死率(OR=1.130,95%CI 1.023~1.248)增加的独立危险因素。外部验证结果显示,低呼吸频率变异组发生ICU及医院死亡事件的风险分别是高呼吸频率变异组的1.329倍(95%CI 1.085~1.627)和1.267倍(95%CI 1.080~1.487)。结论在ICU患者中,低呼吸频率变异与其发生ICU死亡和医院死亡有关。

Objective To investigate the effect of respiratory rate variability on clinical outcomes of the patients in intensive care units(ICU).Methods Adult patients with spontaneous respiratory rate records during the first 24 hours after ICU admission ectracted from eICU collaborative research database were enrolled by using retrospective cohort studies.Extracted information included age,gender,race,first acute physiological and chronic health evaluationⅣscore,disease diagnosis,medication,and recording of spontaneous respiratory rate.Patients were divided into a low respiratory rate variability group[the coefficient of variation(CV)≤20%]and a high respiratory rate variability group(CV>20%)based on CV of spontaneous respiratory rate within the first 24 hour after ICU admission.The ICU mortality and in-hospital mortality were compared between two groups by using chi square test.The correlation between respiratory rate variability and clinical outcomes was analyzed by a Logistic regression model,and external validation was performed in the patients from Medical Information Mart for Intensive CareⅡdatabase(MIMIC-Ⅱ).Results There were 40229 cases included in this study.ICU mortality and in-hospital mortality were significantly higher in low respiratory rate variability group than in high respiratory rate variability group(2.7%vs.2.4%,P=0.043;5.6%vs.5.1%,P=0.030).Logistic regression analysis showed that compared with high respiratory rate variability,the patients with low respiratory rate variability had a higher ICU mortality(OR=1.211,95%CI 1.051-1.396)and higher in-hospital mortality(OR=1.130,95%CI 1.023-1.248)after adjusting for covariates.Furthermore,external validation revealed that the patients in low respiratory rate variability group had an OR of 1.329 for ICU mortality(95%CI 1.085-1.627)and OR of 1.267 for in-hospital mortality(95%CI 1.080-1.487).Conclusions A low respiratory rate variability may be associated with higher ICU mortality and in-hospital mortality in ICU patients.