Positive end-expiratory pressure-induced functional recruitment in patients with acute respiratory distress syndrome

OBJECTIVE: In acute respiratory distress syndrome, alveolar recruitment improves gas exchange only if perfusion of the recruited alveolar units is adequate. To evaluate functional recruitment induced by positive end-expiratory pressure, we assessed pulmonary conductance for gas exchange based on lung diffusion for carbon monoxide and its components, including pulmonary capillary blood volume. DESIGN: Prospective, randomized, crossover study. SETTING: Medical intensive care unit of a university hospital. PATIENTS: Sixteen patients with lung injury/acute respiratory distress syndrome as well as eight control patients under invasive ventilation and eight healthy volunteers. INTERVENTIONS: Mechanical ventilation with two levels of positive end-expiratory pressure (5 and 15 cm H2O). MEASUREMENTS AND MAIN RESULTS: Lung diffusion for carbon monoxide and lung volumes, arterial blood gas analysis, and pressure-volume curves. In patients with acute respiratory distress syndrome, high positive end-expiratory pressure induced a 23% mean lung diffusion for carbon monoxide increase (4.4 +/- 1.7 mm Hg . min vs. 3.6 +/- 1.4 mL . mm Hg . min). In control patients and in healthy volunteers, lung diffusion for carbon monoxide values were (median [interquartile range]) 5.5 [3.8-8.0] mm Hg . min and 19.6 [15.1-20.6] mL . mm Hg . min, respectively. Among patients with acute respiratory distress syndrome, eight showed a >20% lung diffusion for carbon monoxide increase (responders) when increasing positive end-expiratory pressure. In the other eight, lung diffusion for carbon monoxide decreased or showed a <5% increase (nonresponders) with high positive end-expiratory pressure. Compared with nonresponders, responders at low positive end-expiratory pressure had smaller lungs with higher capillary blood volume-to-lung-volume ratio, higher values of the lower inflection point, and significantly greater increases in pulmonary capillary blood volume with high positive end-expiratory pressure. High positive end-expiratory pressure increased PaO2/Fio2 only in the responders. CONCLUSIONS: The functional response to positive end-expiratory pressure in patients with acute lung injury/acute respiratory distress syndrome seems better when the lungs are smaller and with a higher capillary blood-volume-to-lung-volume ratio. Lung diffusion for carbon monoxide measurement supplies additional information about functional lung recruitment, which is not synonymous with mechanical recruitment.