重症肺炎行有创机械通气患者ICU死亡的多因素分析及风险模型建立

Multivariate analysis of death in the patients with severe pneumonia undergoing invasive mechanical ventilation and the establishment of risk model

目的分析重症肺炎行有创机械通气患者ICU死亡的危险因素,制定风险预警模型。方法回顾性收集2019年3月至2021年3月开封市中心医院EICU收治的138例重症肺炎行有创机械通气患者的25项资料:包括一般资料[年龄、性别、体重指数(BMI)]和入院时相关指标[体温(T)、呼吸频率(R)、心率(HR)、平均动脉压(MAP)、白细胞计数(WBC)、血小板计数(PLT)、血红蛋白(Hb)、血清钾(K^(+))、血清钠(Na^(+))、血清氯(Cl-)、动脉氢离子浓度指数(pH)、氧合指数(OI)、血乳酸(LAC)、D-二聚体、血肌酐(SCr)、总胆红素(TBIL)、白蛋白(ALB)、C-反应蛋白(CRP)、脑尿钠肽(pro-BNP)、降钙素原(PCT)、GCS评分、是否行病原体宏基因组学(mNGS)检测等]。根据患者在ICU治疗最终是否死亡分为存活组和死亡组,比较两组间各项指标的差异,对差异有统计学意义的指标应用二元Logistic回归分析,判断其与患者死亡的相关性,明确独立危险因素并建立死亡风险模型。结果存活组和死亡组BMI、T、Na^(+)、D-二聚体、SCr、PCT、是否行mNGS检测等7项指标差异有统计学意义(P<0.05),进一步多因素Logistic回归分析结果显示,BMI、SCr、是否行mNGS检测是患者死亡结局的独立危险因素,建立风险模型logit P=In(P/1-P)=-2.939+1.313×(BMI≤18.5)+1.725×(SCr>133μmol/L)+1.294×(未行mNGS检测),当P>0.39时患者死亡风险高,敏感度和特异度分别为72.7%和76.1%。结论BMI、SCr、是否行mNGS检测三者联合可较好地预测重症肺炎行有创机械通气患者的ICU预后。

Objective To analyze the risk factors of death in the patients with severe pneumonia complicated with invasive mechanical ventilation,and to construct risk models to guide clinical judgment and decision-making.Methods 25 items of clinical data of 138 patients with severe pneumonia treated in EICU of Kaifeng Central Hospital from March 2019 to March 2021 were retrospectively collected,including age,sex,body mass index(BMI),body temperature(T),respiratory rate(R),heart rate(HR),mean arterial pressure(MAP),white blood cell count(WBC),platelet count(PLT),hemoglobin(Hb),serum potassium(K^(+)),serum sodium(Na^(+)),serum chlorine(Cl-),arterial hydrogen ion concentration(pH),oxygenation index(OI),blood lactate(LAC),D-dimer,serum creatinine(SCr),total bilirubin(TBIL),albumin(ALB),C-reactive protein(CRP),pro-brain natriuretic peptide(pro-BNP),procalcitonin(PCT),GCS score,and whether to conduct pathogenic metagenomic next-generation sequencing(mNGS)detection.Patients were divided into two groups(survival group and death group)based on different outcomes in the ICU,the differences of clinical data in two groups were compared.The binary Logistic regression model was used to analyze the risk factors of patients in order to judge the correlation with the death,to determine the independent risk factors for the prognosis and to establish the risk model.Results BMI,T,Na^(+),D-dimer,SCr,PCT,and whether to perform mNGS or not were related to the prognosis of patients(P<0.05).Further multivariate Logistic regression analysis showed that BMI,SCr,and whether to perform mNGS or not were independent risk factors for the prognosis of patients in ICU.Risk model logit P=In(P/1-P)=-2.939+1.313×(BMI≤18.5)+1.725×(SCr>133μmol/L)+1.294×(without mNGS detection),the risk of death was high when P>0.39 with the sensitivity and specificity of 72.7%and 76.1%.Conclusions The combination of BMI,SCr and whether to perform the mNGS or not can better predict the prognosis of severe pneumonia patients with invasive mechanical ventilation in ICU.