LAUSR

Tidal expiratory flow limitation (EFL), which may herald airway closure (AC), is a mechanism of loss of aeration in ARDS. In this prospective, short-term, 2-center, study we measured static and dynamic chest wall (Est,cw and Edyn,cw) and lung (Est,L and Edyn,L) elastance with esophageal pressure, EFL and AC at 5 cmH2O positive end-expiratory pressure (PEEP) in intubated, sedated and paralyzed ARDS patients. For EFL determination we used the atmospheric method and a new device allowing comparison of tidal flow during expiration to PEEP and to atmosphere. AC was validated when airway opening pressure (AOP), assessed from volume-pressure curve, was found greater than PEEP by at least 1 cmH2O. EFL was defined whenever flow does not increase between exhalation to PEEP and to atmosphere over all or part of expiration. Elastance values were expressed as percentage of normal predicted values (%N). Among the 25 patients included, 8 had EFL (32%) and 13 AOP (52%). Between patients with and without EFL Edyn,cw (median (1st-3rd quartiles)) was 70 (16-127) and 102 (70-142) %N (P=0.32) and Edyn,L338 (332-763) and 224 (160-275) %N (P<0.001). The corresponding values for Est,cw and Est,L were 70 (56-88) and 85 (64-103) %N (P=0.35), and 248 (206-348) and 170 (144-195) (P=0.02), respectively. Dynamic EL had an area receiver operating characteristic curve of 0.88 [95% confidence intervals 0.83-0.92] for EFL and 0.74[0.68-0.79] for AOP. Higher Edyn,L was accurate to predict EFL in ARDS patients, AC can occur independently of EFL and both should be assessed concurrently in ARDS patients.