呼吸机自动测量气道阻断压与标准测量方法的一致性

目的 验证呼吸机自动测量的气道阻断压(airway occlusion pressure,P0.1)与标准测量方法的一致性,以及不同触发形式对自动测量结果的影响。方法 选择采用非阻断自动测量P0.1的呼吸机,分别进行模拟肺和临床试验。调节全自动自主呼吸模拟肺,模拟具有不同呼吸努力和呼吸力学特征的48种状态,应用待检测呼吸机,以压力支持通气模式为模拟肺进行通气。临床研究部分纳入15例接受压力支持通气的成年患者。两部分研究均随机交叉应用压力和流量触发,采集呼吸机自动监测的P0.1后,采用标准呼气末气道阻断法进行P0.1测量,作为标准参考值。采用Bland-Altman检验分析P0.1呼吸机监测值与标准参考值的一致性,计算残差(呼吸机监测值-标准参考值)和95%一致性区间,比较不同触发形式对残差的影响。结果 在模拟肺验证中,P0.1呼吸机监测值与标准参考值之间的残差(95%一致性区间)在压力和流量触发下分别为0.04(－0.63~0.70)和－0.54(－1.44~0.36)cmH₂O(1 cmH₂O=0.098 kPa),两者相比差异具有统计学意义(P<0.001)。临床观察获得了相似的结果,压力和流量触发时的残差(95%一致性区间)分别为－0.11(－0.73~0.52)和－0.54(－1.50~0.59)cmH₂O,差异无统计学意义(P>0.05),表明呼吸机监测值与标准参考值间存在一致性。结论 压力触发时,呼吸机非阻断法自动测量的P0.1与标准方法具有良好的一致性。而流量触发明显低估P0.1,建议自动监测时切换到压力触发。

Accuracy of automatic measurement of airway occlusion pressure by ventilator: bench and clinical study

Objective To investigate the agreement of the airway occlusion pressure (P0.1) measured automatically by the ventilator with the standard measurement method, and the influence of different types of trigger on the automatic measurement. Methods Ventilator with automatic measurement of P0.1 using the unoccluded fitting method was selected. Bench and clinical studies were conducted in a lung simulator and mechanically ventilated patients. A total of 48 conditions were adjusted to simulate different statuses of respiratory effort and mechanics. The tested ventilator was used to deliver the pressure support ventilation to the simulator. The clinical study included 15 adult patients undergoing pressure support ventilation. In both bench and clinical studies, the pressure and flow trigger were used as random crossover. After collecting the P0.1 automatically monitored by the ventilator, the standard end-expiratory airway occlusion was performed to measure the P0.1 as the standard reference value. The Bland-Altman test was used to analyze the agreement between the ventilator monitoring value and the standard reference value, and the bias (ventilator monitoring value-standard reference value) and 95% limit of agreement were calculated. Bias in P0.1 measurement of the two triggering methods was compared. Results In the simulated lung validation, the bias (95% limit of agreement) in P0.1 measurement during pressure and flow trigger was 0.04 (－0.63-0.70) and －0.54 ( －1.44－0.36) cmH₂O, respectively, which was statistically significant (P<0.001). Similar results were obtained in clinical studies, with the bias (95% limit of agreement) at pressure and flow triggering of －0.11 (－0.73-0.52) and －0.54 (－1.50-0.59) cmH₂O, respectively, which was also statistically significant. Conclusion During pressure trigger, an excellent agreement was found in P0.1 measured by the ventilator the standard method. However, the ventilator measurement significantly underestimates P0.1 during flow trigger. It is recommended to switch to pressure trigger during automatic monitoring of P0.1 by ventilator.