应用强迫振荡技术优化慢性阻塞性肺疾病患者经鼻罩无创通气压力水平的探讨

Clinical application of the forced oscillation technique for titration of pressure support levels in chronic obstructive pulmonary disease patients during nasal noninvasive positive pressure ventilation

目的探讨利用强迫振荡阻抗和电抗方法优化COPD患者经鼻罩无创通气压力水平的合理性。方法研究对象为2007年6月至2008年2月在广州呼吸疾病研究所住院的8例COPD合并2型呼吸衰竭患者，分别给予不同压力水平的双水平气道正压通气和连续气道正压通气（CPAP），采用5Hz强迫振荡信号检测各呼吸周期的吸气相阻抗，以食道测压法测定的吸气相肺阻力为参照，比较两者的一致性和相关性；计算CPAP分别为4、8和12cmH2O（1cm H2O＝0．098kPa）时各呼吸周期的呼气相与吸气相电抗之差（△Xrs），分析各呼吸周期的跨肺压与流量的关系，将上述各呼吸周期划分为有或无呼气流速受限（EFL），比较CPAP水平、△Xrs值和显示为EFL的呼吸周期数之间的数量关系，计算检测EFL呼吸周期敏感度和特异度均最高的△Xrs阈值。正态分布资料均数的比较采用配对t检验，非正态分布资料采用非参数检验，相关分析采用直线相关分析方法。结果8例COPD患者的吸气相阻抗和吸气相肺阻力的均值分别为（6．5±1．6）cmH2O·s^-1·L^-1和（9±5）cmH2O·s^-1·L^-1（相关系数为0．64，P〈0．01），两者差值的中位数为-2．7（-5．7-0．7）cmH2O·S^-1·L^-1，一致性范围在（-10．6～4．9）cmH2O·s^-1·L^-1，吸气相肺阻力=-1．62＋1．69吸气相阻抗（F=109．6，P〈0．01）。CPAP水平增高，△Xrs显著降低，显示为EFL的呼吸周期数显著减少，区分呼吸周期有或无EFL的△xrs阈值为1．83cm H2O·s^-1·L^-1，敏感度为94％，特异度为97％。结论5Hz强迫振荡吸气相阻抗可用于预测吸气相肺阻力，参照吸气相肺阻力设置无创通气的吸气相正压水平，能有效克服增高的肺阻力；△Xrs反映EFL程度，逐步增高CPAP或呼气末正压的水平，使△Xrs≤1．83cmH2O·s^-1·L^-1，可消除绝大部分呼吸周期的EFL。

Objective To study the use of oscillatory resistance （Rrs） and reactance （Xrs） to guide the settings of pressure level of ventilators in patients with severe chronic obstructive pulmonary disease （COPD） during noninvasive positive pressure ventilation （NPPV）. Methods Five Hz oscillatory inspiratory Rrs （Rs,in） at different NPPV press.ure levels in 8 patients were calculated and compared to inspiratory pulmonary resistance （RL,in） measured by means of oesophageal manometry. The difference between inspiratory and expiratory Xrs （△Xrs） at different CPAP levels were also measured for each breathing cycle, which were subsequently analyzed and classified as flow-limited （EFL） or non-EFL by means of Mead-Whittenberger method. Then, the relationship among CPAP levels, the percentage of EFL breathing cycles at different CPAP levels and △Xrs were analyzed and the threshold value of △Xrs with maximum sensitivity and specificity to detect EFL were calculated. Results The mean values of Rrs,in and RL,in were （6.5±1.6） and （9±5） cm H2O·s^-1·L^-1 （1 cm H2O=0.098 kPa）, respectively,r =0.64, P〈0.01, but the agreement between these two resistances were poor and with a median of - 2.7（-5.7 -0. 7）cm H2O·s^-1· L^-1, and the limits of agreement ranged from -10.6 cm H2O·s^-1·L^-1 to 4.9 cm H2O·s^-1·L^-1. The linear regression equation of Rrs,in to RL,in was RL,in= -1.62±1.69 Rrs,in, F= 109.6, P〈0. 01. Increase in CPAP level resulted in decrease in △Xrs and in numbers of EFL breathing cycles. The mean value of △Xrs in EFL breathing cycles was significantly higher than that in NFL, and the threshold value of △Xrs for detecting EFL was 1.83 cm H2O·s^-1·L^-1 , with a sensitivity of 94% and specificity of 97% respectively. Conclusions Although slightly underestimated, Rrs,in is helpful to guide the settings of inspiratory positive airway pressure level, which will appropriately overcome the elevated pulmonary resistance, and the threshold value of △Xrs, of end-expiratory pressure, which will eliminate the majority of EFL breathing cycles.