Survival rates in ARDS with conventional ventilation using high oxygen fractions and low PEEP levels have been reported to be less than 10%. In three prospective evaluations of ARDS in the 1980s, mortality rates remained greater than 60%. Early studies using high-level PEEP therapy in severe ARDS by Douglas, Downs, Kirby, and Civetta showed improved survival rates with ranges between 60% and 80%. In 1979 Gallagher reviewed 59 patients with ARDS who were treated with PEEP greater than 15 cm H2O titrated to improve FRC by achieving an intrapulmonary shunt fraction of 15%. The overall survival was 65%, with only 5% of the patients dying secondary to respiratory failure. In the more recent study by Miller in trauma patients and later by DiRusso in a variety of surgical patients, the overall mortality rate for those patients receiving PEEP greater than 15 cm H2O was 20% to 30%. Of the 14 patients who died, only seven (10% of the total) succumbed to respiratory failure. The remaining patients died from the primary underlying disease with normal oxygenation or after significant weaning from high PEEP levels. By using a goal-oriented approach to the management of patients with severe ARDS, we have found that high-level PEEP therapy was effective in lowering the intrapulmonary shunt and improving the SaO2 at acceptable levels of inspired oxygen. All of these patients were ventilated with traditional high tidal volumes (10 to 15 mL/kg) and therefore exhibited high peak inspiratory airway pressures. This support method did not seem to cause lung injury or an excessive amount of barotrauma in these patients, but in fact, was associated with a lower mortality rate (30%) than reported in other studies of patients with lesser degrees of lung oxygenation dysfunction and extrapulmonary organ system dysfunction. Currently available information indicates that increases in mean airway pressure (induced with PEEP or other modes of ventilatory support to restore losses in FRC that occur during ARDS) and limiting exposure to toxic concentrations of oxygen minimize ventilator-induced secondary lung injury and maximize chances for survival. Arbitrary limitations of peak inspiratory or end-expiratory airway pressure or mandatory tidal volume in patients with severe ARDS seem to be unfounded. Failure to achieve adequate physiologic end-points in these patients may increase morbidity and mortality rates.