A comparison of methods to identify open-lung PEEP

Abstract

Purpose: Many methods exist in the literature for identifying PEEP to set in ARDS patients following a lung recruitment maneuver (RM). We compared ten published parameters for setting PEEP following a RM. Methods: Lung injury was induced by bilateral lung lavage in 14 female Dorset sheep, yielding a PaO 2 100-150 mmHg at FIO2 1.0 and PEEP 5 cmH 2O. A quasi-static P-V curve was then performed using the supersyringe method; PEEP was set to 20 cmH2O and a RM performed with pressure control ventilation (inspiratory pressure set to 40-50 cmH 2O), until PaO2 + PaCO2 > 400 mmHg. Following the RM, a decremental PEEP trial was performed. The PEEP was decreased in 1 cmH2O steps every 5 min until 15 cmH2O was reached. Parameters measured during the decremental PEEP trial were compared with parameters obtained from the P-V curve. Results: For setting PEEP, maximum dynamic tidal respiratory compliance, maximum PaO2, maximum PaO 2 + PaCO2, and minimum shunt calculated during the decremental PEEP trial, and the lower Pflex and point of maximal compliance increase on the inflation limb of the P-V curve (Pmci,i) were statistically indistinguishable. The PEEP value obtained using the deflation upper Pflex and the point of maximal compliance decrease on the deflation limb were significantly higher, and the true inflection point on the inflation limb and minimum PaCO2 were significantly lower than the other variables. Conclusion: In this animal model of ARDS, dynamic tidal respiratory compliance, maximum PaO2, maximum PaO2 + PaCO2, minimum shunt, inflation lower Pflex and Pmci,i yield similar values for PEEP following a recruitment maneuver. © 2009 Springer-Verlag.