Do Low Tidal Volumes Prevent Lung Injury in ARDS?

Abstract & Commentary

Synopsis: This multicenter randomized study of mechanically ventilated patients with ARDS found that a ventilator strategy that reduces tidal volumes to less than 10 mL/kg and limits end-inspiratory (plateau) pressure to 25 cm H2O does not improve morbidity or mortality compared to current conventional ventilator management strategies.

Source: Brochard L, et al. Am J Respir Crit Care Med 1998;158:1831-1838.

Brochard and colleagues recruited patients with acute respiratory distress syndrome (ARDS) who had only one organ system involved (single organ failure) and who required mechanical ventilation. Brochard et al randomized the patients either to a low tidal volume (< 10 mL/kg) and low plateau pressure (< 25 cm H2O) ventilator strategy or to a standard management strategy with tidal volume of more than 10 mL/kg (not to exceed 15 mL/kg) and a plateau pressure less than 35 cm H2O. A "PEEP trial" was used to establish the optimal PEEP that improved oxygenation (as measured by PaO2/FiO2) without reducing cardiac output; this was designed to standardize PEEP in both groups. After 116 patients had been enrolled, the study was terminated after an interim analysis performed on the first 100 patients made it clear that enrolling more patients would not show the anticipated change in mortality (from 50 to 30%).

Brochard et al were successful in their efforts to randomize individuals to different management strategies. The ranges of tidal volumes and plateau pressures (from 1-14 days) for the low tidal volume group were 7.1-7.6 mL/kg and 25.7-24.5 cm of water, respectively. The ranges of tidal volumes and plateau pressures for the standard management group were significantly higher (10.3-9.9 mL/kg and 31.7-33.6 cm of water, respectively). There were no significant differences between management strategies in PEEP levels, PaO2, FiO2, incidence of barotrauma, or total mortality at 60 days. PaCO2 was significantly higher in the low tidal volume group as compared to the standard tidal volume group (49.5-53.9 mmHg vs 41.3-44.7 mmHg) and there was a trend toward increasing use of paralytic agents in the low tidal volume group (P = 0.12).


Convincing animal data have suggested that limiting maximal transalveolar pressure to less than 30-35 cm H2O would reduce the risk of ventilator-associated lung injury and overt barotrauma. Limiting tidal volume and end-inspiratory (plateau) pressure reduces transalveolar pressure and has been demonstrated to improve outcomes in small randomized and nonrandomized trials. This current multicenter randomized trial now complements two other recent studies of limited pressure and tidal volume ventilation in patients with ARDS (Amato MBP, et al. N Engl J Med 1998;338:347-354; Stewart TE, et al. N Engl J Med 1998;338:355-361).

With the exception of the study by Amato et al, all studies of low tidal volume ventilation for ARDS reported to date have failed to demonstrate an improvement in mortality or a reduction in barotrauma by limiting tidal volume to less than 8 mL/kg or PPLAT to less than 25-30 cm H2O. In all these studies, the tidal volumes in the control group ranged from 10-12 mL/kg and the PPLAT to less than 35 cm H2O. In effect, the control groups were maintained with low tidal volumes and PPLAT as compared to the historical use of tidal volumes of 15 mL/kg. It appears that maintenance of a PPLAT less than 35 cm H2O is protective, as the rates of pneumothorax were relatively low, at approximately 13%, and the mortality rates ranged from 30-50%, depending on the severity of organ disease. It is also apparent that further reductions in tidal volume and PPLAT confer no added protection. An ongoing multicenter randomized trial of low vs high tidal volume in the United States has recruited more than 800 patients and hopefully will soon be able to answer this question definitively (personal communication: Gordon D. Bernard, MD).

To clarify association of tidal volume and barotrauma in mechanically ventilated patients with ARDS, the multinational trial of aerosolized synthetic surfactant in patients with ARDS was retrospectively evaluated by Weg and colleagues (Weg J, et al. N Engl J Med 1998;338:342-347). Weg et al reported a 6.9% incidence of pneumothorax in 725 patients and found no association between the presence of air leaks or pneumothorax and any tidal volume or pressure examined. The average tidal volume was 11.6 mL/kg and mean airway pressure was 24 cm H2O. Unfortunately, they did not have PPLAT measurements to evaluate. This study suggests that pneumothorax is relatively uncommon in patients with ARDS undergoing conventional mechanical ventilation and may be related to the severity of the lung injury rather than high tidal volume ventilation. However, the historically used high tidal volumes of 15 mL/kg were not evaluated.

The only study to demonstrate an improved mortality with low tidal volume, low plateau pressure ventilation was conducted in Brazil (Amato MBP, et al. N Engl J Med 1998;338:347-354). This study used a number of ventilator manipulations: 1) limiting the inspiratory driving pressure to less than 20 cm H2O above the level of positive end-expiratory pressure (PEEP) by using low tidal volumes and pressure-control ventilation; 2) employing "lung recruiting maneuvers" after suctioning or at any time the endotracheal tube is disconnected from the ventilator (brief application of high pressure, 35-40 cm H2O, for 40 seconds); 3) allowing PaCO2 to rise (permissive hypercapnea); and 4) using PEEP set at 2 cm H2O above the lower inflection point on the pressure-volume curve (PFLEX). The mean PFLEX was approximately 15 cm H2O in this study. Patients were randomized either to this strategy or to a conventional strategy consisting of use of the lowest possible PEEP, tidal volumes of 12 mL/kg, and maintenance of a normal PaCO2. Using this strategy, Amato et al reported a 38% mortality in the protective-ventilation arm of the study and a 71% mortality in the conventional arm. Rates of barotrauma were 7% and 42%, respectively.

There are a number of concerning aspects of this study that require further examination. In particular, the mortality and rates of barotrauma in the control arm were extremely high. In fact, in the two other large randomized trials of low tidal volume vs. conventional tidal volume ventilation (Stewart et al; Brochard et al), the mortality and barotrauma rates in the control arms were much lower. In fact, the mortality in Amato et al’s treatment group was the same as in the control groups of these other studies. Therefore, these results remain suspect and the role of high PEEP and recruiting maneuvers must be further studied.

We can safely make two conclusions. First, limiting plateau pressures to less than 35 cm H2O and tidal volumes to 10-12 mL/kg results in low rates of barotrauma and expected mortality rates for ARDS. Second, further reductions in plateau pressure and tidal volume appear not to improve outcome. Strategies that increase PEEP above the lower inflection point of the pressure-volume curve deserve further study.