Objective: To study the effect of positive end-expiratory pressure (PEEP) on right ventricular hemodynamics and ejection fraction in patients with chronic obstructive pulmonary disease and positive alveolar pressure throughout expiration by dynamic hyperinflation (auto-PEEP).
Design: Open, prospective, controlled trial.
Setting: General intensive care unit of a community hospital.
Patients: Ten patients sedated and paralyzed with an acute exacerbation of chronic obstructive pulmonary disease undergoing mechanical ventilation.
Interventions: Insertion of a pulmonary artery catheter modified with a rapid response thermistor and a radial arterial catheter. PEEP was then increased from 0 (PEEP 0) to auto-PEEP level (PEEP = auto-PEEP) and 5 cm H2O above that (PEEP = auto-PEEP +5).
Measurements: At each level of PEEP, airway pressures, flow and volume, hemodynamic variables (including right ventricular ejection fraction by thermodilution technique), and blood gas analyses were recorded.
Main results: The mean auto-PEEP was 6.6 +/- 2.8 cm H2O and the total PEEP reached was 12.2 +/- 2.4 cm H2O. The degree of lung inflation induced by PEEP averaged 145 +/- 87 mL with PEEP = auto-PEEP and 495 +/- 133 mL with PEEP = auto-PEEP + 5. The PEEP = auto-PEEP caused a right ventricular end-diastolic pressure increase, but there was no other significant hemodynamic change. With PEEP = auto-PEEP + 5, there was a significant increase in intravascular pressures; this amount of PEEP reduced cardiac output (from 4.40 +/- 1.38 L/min at PEEP 0 to 4.13 +/- 1.48 L/min; p < .05). The cardiac output reduction induced by PEEP = auto-PEEP + 5 was > 10% in only five cases and this group of patients had significantly lower right ventricular volumes than the group with less cardiac output variation (right ventricular end-diastolic volume: 64 +/- 9 vs. 96 +/- 26 mL/m2; right ventricular end-systolic volume: 38 +/- 6 vs. 65 +/- 21 mL/m2; p < .05) without significant difference in the other variables that were measured. Neither right ventricular ejection fraction nor right ventricle volumes changed as PEEP increased, but there were marked interpatient differences and also pronounced changes in volume between stages in individual patients.
Conclusions: In the study conditions, PEEP application up to values approaching auto-PEEP did not result in the impairment of right ventricular hemodynamics, while higher levels reduced cardiac output in selected patients.