Mechanical ventilation with plateau pressure lower than 35 cm H2O and high positive end-expiratory pressure (PEEP) has been recommended as lung protective strategy. Ten patients with ARDS (five from pulmonary [p] and five from extrapulmonary [exp] origin), underwent 2 h of lung protective strategy, 1 h of lung protective strategy with three consecutive sighs/min at 45 cm H2O plateau pressure, and 1 h of lung protective strategy. Total minute ventilation, PEEP (14.0 +/- 2.2 cm H2O), inspiratory oxygen fraction, and mean airway pressure were kept constant. After 1 h of sigh we found that: (1) PaO2 increased (from 92.8 +/- 18.6 to 137.6 +/- 23.9 mm Hg, p < 0.01), venous admixture and PaCO2 decreased (from 38 +/- 12 to 28 +/- 14%, p < 0.01; and from 52.7 +/- 19.4 to 49.1 +/- 18.4 mm Hg, p < 0.05, respectively); (2) end-expiratory lung volume increased (from 1.49 +/- 0.58 to 1.91 +/- 0.67 L, p < 0.01), and was significantly correlated with the oxygenation (r = 0.82, p < 0.01) and lung elastance (r = 0.76, p < 0.01) improvement. Sigh was more effective in ARDSexp than in ARDSp. After 1 h of sigh interruption, all the physiologic variables returned to baseline. The derecruitment was correlated with PaCO2 (r = 0.86, p < 0.01). We conclude that: (1) lung protective strategy alone at the PEEP level used in this study may not provide full lung recruitment and best oxygenation; (2) application of sigh during lung protective strategy may improve recruitment and oxygenation.