[Clinical study of the setting of positive end expiratory pressure in patients with acute cardiogenic pulmonary edema during mechanical ventilation].

OBJECTIVE: To compare the effects on hemodynamics and lung mechanics when different positive end expiratory pressure (PEEP) levels were used in acute cardiac pulmonary edema. METHODS: Thirty-nine patients with respiratory failure and treated with mechanical ventilation were divided into two groups according to cardiac index (CI). The changes of hemodynamics [including cardiac output (CO), CI, pulmonary capillary blood flow (PCBF), central venous pressure (CVP), surround vascular resistance (SVR)], lung mechanical variables [intrinsic positive end expiratory pressure (PEEPi), peak inspiratory pressure (Ppeak), mean of airway pressure (Pmean), minute volume (MV), alveolar tidal volume (Vtalv)], percutaneous saturation of oxygen (SpO(2)) and blood pressure (BP) were determined with when selected different PEEP levels under the bi-level positive airway pressure (BIPAP) mode in normal cardiac function group (n=18, CI> or =2.0 L.min(-1).m(-2)) and poor cardiac function group (n=20, CI<2.0 L.min(-1)xm(-2)). RESULTS: In normal cardiac function group, PEEP has no effect on hemodynamics when varied from 0-13 cm H(2)O (1 cm H(2)O=0.098 kPa), and Ppeak and PEEPi increased with the elevation of PEEP. Resistance of airway (R) fell when PEEP was increased. However, in poor cardiac function group SVR, CO, CI showed curvilinear changes with an increase in PEEP, and CO, CI were high when PEEP changed from 0 to 7 cm H(2)O, but dropped markedly when PEEP was increased in 13 cm H(2)O, while the changes of SVR was the reverse of CO and CI. The effects on lung mechanics varied relatively smaller range when PEEP was set between 5-7 cm H(2)O, and the burden of breathing was slight. CONCLUSION: The mode of artificial ventilation should be adjusted according to the changes in hemodynamics and lung mechanics. PEEP should be individualized, and 5-7 cm H(2)O (normally below 10 cm H(2)O) is suitable.