Lung-Protective Ventilation and ARDS Mortality: Part I


Editor's note: Three important studies of ventilation strategy in the acute respiratory distress syndrome (ARDS) appeared in the same recent issue of the New England Journal of Medicine. One of these studies (Weg JG, et al. N Engl J Med 1998;338(6):341-436) was discussed last month. Here, Associate Editor Robert M. Kacmarek summarizes and provides an editorial critique of the other two articles.

Synopsis: Patients with ARDS randomized to a lung protective management strategy had a lower 28-day mortality than patients ventilated with high airway pressures and PEEP levels determined based on oxygenation targets.

Source: Amato MBP, et al. N Engl J Med 1998;338: 347-354.

Based on animal experiments and data from human case series, Amato and associates in Sao Paulo, Brazil, performed a randomized, prospective, controlled trial to evaluate the effect of a protective ventilatory strategy on mortality in ARDS. In the treatment group (n = 29, age 33 ± 13 years, number of extra-pulmonary organ failures 2.6 ± 1.3, and lung injury score 3.4 ± 0.4), patients were ventilated with a lung protective strategy. A static pressure-volume curve was performed to determine initial PEEP level (2 cm H2O above the lower inflection point, or 16 cm H2O if no inflection point was observed) and maximum peak alveolar (end inspiratory plateau) pressure was limited to the upper deflection point on the pressure-volume curve. All patients were ventilated in pressure control or support ventilation. Permissive hypercapnia was allowed, and the target PaO2 was 80 mmHg. Whenever the ventilator was disconnected, a continuous positive airway pressure of 35-40 cm H2O was applied for 40 seconds as a lung recruitment maneuver.

In the control group (n = 24, age 36 ± 14, number of extra-pulmonary organ failures 2.7 ± 1.5, and lung injury score 3.2 ± 0.4), patients were ventilated in volume-control/assist-control at a tidal volume of 12 mL/kg, with rate set to keep PaCO2 between 35 and 38 mmHg. PEEP and FIO2 were adjusted by an algorithm to keep FIO2 less than 0.6 and PaO2 approximately 80 mmHg. All other aspects of care between groups were similar.

After 28 days, 11 patients (38%) in the treatment group and 17 (71%) in the control group had died. Weaning rates (66% vs 29%) and clinical barotrauma rates (7% vs 42%) were significantly better in the treatment group than in the control group. However, hospital survival rates (55% treatment vs 29% control) were not statistically different. Initial PEEP levels differed greatly between groups (16.3 ± 0.7 vs 6.9 ± 0.8 cm H2O in the lung-protective and control groups, respectively). By the second day after enrollment, plateau pressures (23.9 ± 0.7 vs 37.8 ± 1.2 cm H2O), peak airway pressures 24.0 ± 0.7 vs 45.5 ± 1.5 cm H2O), tidal volume (387 ± 7 vs 738 ± 17 mL), PaCO2 (50.8 ± 1.1 vs 35.0 ± 0.7 mmHg), and PaO2/FIO2 (239 ± 6 vs 146 ± 7 mmHg) also differed significantly between the treatment groups.