动态肺部超声评分对重症新型冠状病毒肺炎呼吸治疗的指导

Guiding value of dynamic lung ultrasound score in the respiratory support treatment of severe COVID-19

目的探讨肺部超声评分(LUS)动态改变对重症新型冠状病毒肺炎(COVID-19)患者呼吸支持治疗的指导价值。方法对在ICU住院治疗的29例重症COVID-19患者实施LUS,收集同期临床资料,根据起始呼吸治疗结局分为成功组和失败组,分析LUS和临床特征及LUS动态改变对呼吸支持策略的预测价值。结果29例重症COVID-19患者,男性23例(79.31%),年龄(60.9±12.44)岁,病程(11.00±4.06)d,氧合指数(184.55±61.98)mmHg,LUS(18.34±4.51)分。共行118次LUS,含经鼻高流量氧疗(HFNC)63次,无创正压通气(NPPV)6次,有创正压通气(IPPV)22次,俯卧位通气(PPV)5次,体外膜氧合(ECMO)22次。其中HFNC组LUS明显低于其他呼吸治疗组,ECMO组LUS显著高于IPPV组(P<0.05)。23例无创呼吸治疗成功14例,失败9例,成功组LUS由(18.14±4.44)分下降至(15.07±3.93)分,差异有统计学意义(P<0.05),失败组LUS由(18.44±3.64)分上升至(21.56±3.32)分,差异有统计学意义(P<0.05),治疗前两组LUS无统计学差异(P>0.05)。以LUS≥20.5分预测IPPV的灵敏度为77.6%、特异度为73.9%,LUS≥23.5分预测PPV或ECMO的敏感度为85.2%、特异度为83.5%。结论LUS动态改变趋势是预测呼吸治疗能否成功的有效指标,可为重症COVID-19患者呼吸支持介入时机和手段提供依据。

Objective To explore the guiding value of dynamic changes of lung ultrasound score(LUS)for respiratory support therapy in patients with severe coronavirus disease 2019(COVID-19).Methods LUS were performed on 29 patients with severe COVID-19 admitted to ICU.The corresponding clinical data of patients were collected.LUS and clinical characteristics of patients with different respiratory support therapy were analyzed,and the patients were divided into success group and failure group according to the outcome of initial respiratory therapy.The dynamic changes of LUS and its predictive value for respiratory support strategies were analyzed.Results Twenty-nine patients were enrolled,including 23 males(79.31%),age(60.9±12.44)years old,course of disease(11.00±4.06)days,oxygenation index(184.55±61.98)mmHg,LUS(18.34±4.51)points.A total of 118 times of LUS were performed in 29 patients,including 63 times of nasal high-flow oxygen therapy(HFNC),6 times of non-invasive positive pressure ventilation(NPPV),22 times of invasive positive pressure ventilation(IPPV),5 times of IPPV combined with prone position ventilation(PPV),and 22 times of IPPV combined with extracorporeal membrane oxygenation(ECMO).LUS of HFNC group was significantly lower than that of other groups,and LUS of ECMO group was significantly higher than IPPV group(P<0.05).14 of 23 cases non-invasive respiratory treatment successed and 9 cases failed.After treatment,LUS decreased significantly in the successful group(18.14±4.44 vs 15.07±3.93,P<0.05),and increased significantly in the fail-ure group(18.44±3.6421.56±3.32,P<0.05),there was no statistical difference in LUS between the two groups before treat-ment(P>0.05).Analysis of the area under the ROC curve(AUC)revealed that,LUS≥20.5 predicted a sensitivity of 77.6% and a specificity of 73.9%for IPPV in patients with severe COVID-19,and LUS≥23.5 predicted a sensitivity of 85.2% and a specificity of 83.5%for further PPV or ECMO respiratory support.Conclusion Dynamic LUS can predict the respiratory treatment style and provide an important evidence for the adjustment of respiratory treatment strategy in severe COVID-19 patients.