Abstract: Background　The effectiveness of mechanical ventilation by position in patients with acute respiratory distress syndrome（ARDS） has been explored in some studies，but the therapeutic effect differences between lateral position ventilation （LPV） and prone position ventilation （PPV），and between patients with ARDS originating from pulmonary disease （ARDSp） and ARDS originating from extrapulmonary disease （ARDSexp） have not been clarified. Objective　To compare the therapeutic effects of LPV and PPV in patients with ARDSp and ARDSexp，and to explore their application strategy and value. Methods　85 patients with ARDS （including ARDSp and ARDSexp patients） who received treatment （without experience of changing ventilation position in this episode prior to this study） in the Intensive Care Unit，Affiliated Hospital of North Sichuan Medical College during September 2019 to May 2020 were enrolled，and divided into four groups for comparing LPV and PPV：ARDSp+ LPV group （n=21），ARDSexp+ LPV group （n=21），ARDSp+ PPV group （n=22） and ARDSexp+ PPV group （n=21）.Indices monitored before，and 0.5，1，and 2 hours after changing ventilation position were also collected，including：（1）baseline data such as age，sex，BMI，temperature，heart rate，and APACHEⅡ （2） arterial blood gas parameters：partial pressure of oxygen（PaO2），partial pressure of carbon dioxide （PCO2），arterial oxygen saturation（SaO2） and PaO2/FiO2 ratio；（3） respiratory mechanical parameters：peak airway pressure （Ppeak），rapid shallow breathing index （RSBI），plateau pressure （Pplat） and static pulmonary compliance （Cst）；（4） ICU sonography results：left ventricular ejection fraction （LVEF），inferior vena cava diameter （IVC-D），left ventricular derived cardiac output （CO），tricuspid annular plane systolic excursion （TAPSE），mitral annular plane systolic excursion （MAPSE），ratio of peak E-wave velocity/ peak e' velocity （E/e'），mean arterial pressure （MAP） and lung ultrasound （LUS） score. （5） cardiac biomarker：serum NT-proBNP；（6） prognosis：28-day mortality and incidence of complications. Results　PaO2/FiO2 was influenced significantly by the interaction effect of treatment time and ventilation position （P<0.05）. It was also impacted obviously by the main effect of treatment time（P<0.05），as well as by that of ventilation position （P<0.05）. SaO2 was affected notably by the interaction effect of treatment time and ventilation position（P<0.05）. It was also influenced considerably by the main effect of treatment time （P<0.05），as well as by that of ventilation position （P<0.05）. PaO2 was affected remarkably by the interaction effect of treatment time and ventilation position （P<0.05）. It was also affected greatly by the main effect of treatment time（P<0.05），and by that of ventilation position （P<0.05）. The PaO2/FiO2 and PaO2 in all groups were much better after changing the supine position ventilation（SPV） to LPV or PPV for 0.5，1 and 2 hours （P<0.05）. The PaO2/FiO2 and PaO2 at 0.5，1 and 2 hours after changing SPV to LPV were much higher in ARDSexp+ LPV group than in ARDSp+ LPV group （P<0.05）. The PaO2/FiO2 and PaO2 at 0.5，1 and 2 hours after changing SPV to PPV were much higher in ARDSexp+ PPV group than in ARDSp+ PPV group （P<0.05）. The PaO2/FiO2 and PaO2 at 0.5，1 and 2 hours after postural changes in ARDSexp+ PPV group were higher than those in ARDSexp+ LPV group （P<0.05）. The PaO2/FiO2 and PaO2 at 0.5，1 and 2 hours after postural changes in ARDSp+ PPV group were higher than those in ARDSp+ LPV group （P<0.05）. The SaO2 in all groups increased significantly after changing the SPV to LPV or PPV for 2 hours （P<0.05）. LVEF was influenced by the interaction effect of treatment time and ventilation position （P<0.05）. It was also impacted obviously by the main effect of treatment time（P<0.05），as well as by that of ventilation position（P<0.05）. The LVEF in all groups increased significantly after changing the SPV to LPV or PPV for 1 and 2 hours （P<0.05）. The mean LUS score was influenced significantly by the interaction effect of treatment time and ventilation position （P<0.05）. It was also impacted obviously by the main effect of treatment time（P<0.05），and by that of ventilation position （P<0.001）. The mean LUS score in all groups reduced significantly after changing the SPV to LPV or PPV for 1 and 2 hours （P<0.05）. The mean LUS score was negatively correlated with PaO2/FiO2 during SPV（r=-0.439，P<0.001）. And the correlation between them remained negative after SPV was changed to LPV or PPV for 1 （r=-0.348，P=0.001） and 2 hours（r=-0.327，P=0.002）. No obvious complications were found during the treatment. There was no significant difference in 28-day mortality among the four groups （χ2=1.333，P>0.05）. Conclusion　Changing ventilation posture rapidly and accurately according to patient condition for improving therapeutic effectiveness may be a future research hotspot in ARDS. ARDSexp patients gained more benefits from changing ventilation posture than ARDSp patients. Although PPV was slightly better than LPV in improving lung ventilation，LPV should be considered first when treating patients with complex conditions.

Key words: Respiratory distress syndrome, adult；Pulmonary ventilation；Lateral position ventilation；Prone position ventilation；Ultrasonography；Treatment outcome

摘要： 背景　国内外关于急性呼吸窘迫综合征（ARDS）患者的不同体位机械通气治疗效果有一定的研究，但并未阐明侧卧位通气（LPV）与俯卧位通气（PPV），以及肺内源性ARDS（ARDSp）和肺外源性ARDS（ARDSexp）患者不同体位机械通气的治疗效果差异。目的　比较不同病因ARDS患者LPV和PPV治疗效果，探讨其应用策略和价值。方法　纳入川北医学院附属医院重症医学科2019年9月—2020年5月收治的85例ARDS患者，本次发病期间尚未进行过体位变化通气治疗。根据病因及通气方法分为ARDSp+LPV组（n=21），ARDSexp+LPV组（n=21），ARDSp+PPV组（n=22），ARDSexp+PPV组（n=21）。监测患者体位变化前和体位变化后0.5 h、1 h、2 h的各项指标变化，包括：（1）基线资料：年龄、性别、体质指数、体温、心率、急性生理与慢性健康评估系统（APACHEⅡ）评分。（2）血气分析监测：动脉血氧分压（PaO2）、动脉血二氧化碳分压（PaCO2）、动脉血氧饱和度（SaO2）和氧合指数（PaO2/FiO2）。（3）呼吸力学监测：气道峰压（Ppeak）、呼吸浅快指数（RSBI）、气道平台压（Pplat）和静态肺顺应性（Cst）。（4）重症超声监测左心室射血分数（LVEF）、下腔静脉直径（IVC-D）、左心输出量（CO）、右房室瓣环平面收缩位移（TAPSE）、左房室瓣环平面收缩位移（MAPSE）、左心室舒张早期E峰与左房室侧壁瓣环舒张期e'比值（E/e'）和平均动脉压（MAP）。重症肺部超声监测患者肺超声（LUS）评分。（5）心功能指标监测：氨基末端脑钠肽前体（NT-proBNP）。（6）预后情况：患者28 d死亡率和并发症发生率。结果　PaO2/FiO2在时间和方法上存在交互作用（P<0.05），时间在PaO2/FiO2上主效应显著（P<0.05），方法在PaO2/FIO2上主效应显著（P<0.05）。SaO2在时间和方法上存在交互作用（P<0.05），时间在SaO2上主效应显著（P<0.05），方法在SaO2上主效应显著（P<0.05）。PaO2在时间和方法上存在交互作用（P<0.05），时间在PaO2上主效应显著（P<0.05），方法在PaO2上主效应显著（P<0.05）。4组ARDS患者体位变化0.5 h、1 h和2 h时的PaO2/FiO2和PaO2高于仰卧位通气（SPV）时（P<0.05）。ARDSexp+LPV组患者体位变化0.5 h、1 h和2 h时的PaO2/FiO2和PaO2高于ARDSp+LPV组（P<0.05），ARDSexp+PPV组患者体位变化0.5 h、1 h和2 h时的PaO2/FiO2和PaO2高于ARDSp+PPV组（P<0.05）。ARDSexp+PPV组患者体位变化0.5 h、1 h和2 h时的PaO2/FiO2和PaO2高于ARDSexp+LPV组（P<0.05），ARDSp+PPV组患者体位变化0.5 h、1 h和2 h时的PaO2/FiO2和PaO2高于ARDSp+LPV组（P<0.05）。4组患者体位变化2 h时的SaO2高于SPV时（P<0.05）。LVEF在时间和方法上存在交互作用（P<0.05），时间在LVEF上主效应显著（P<0.05），方法在LVEF上主效应显著（P<0.05）。4组ARDS患者体位变化1 h和2 h时的LVEF高于SPV时（P<0.05）。平均LUS评分在时间和方法上存在交互作用（P<0.05），时间在平均LUS评分上主效应显著（P<0.05），方法在平均LUS评分上主效应显著（P<0.05）。4组ARDS患者体位变化1 h和2 h时的平均LUS评分高于SPV时（P<0.05）。体位变化前和变化后平均LUS评分与PaO2/FiO2均呈负相关（SPV：r=-0.439，P<0.001；体位变化1 h：r=-0.348，P=0.001；体位变化2 h：r=-0.327，P=0.002）。治疗期间未发现明显并发症，4组28 d死亡率比较，差异无统计学意义（χ2=1.333，P>0.05）。结论　ARDS患者的快速精准化体位变化通气治疗是未来的热点之一。ARDSexp患者体位变化通气治疗改善优于ARDSp患者。PPV对肺通气改善稍优于LPV，但复杂环境下应首先考虑LPV。

关键词: 呼吸窘迫综合征, 成人；肺通气；侧卧位通气；俯卧位通气；超声检查；治疗结果