This study was undertaken to determine the effects of superimposing incremental levels of positive end-expiratory pressure (PEEP) during partial liquid ventilation (PLV) on gas exchange, respiratory mechanics and morphological changes in experimental acute lung injury (ALI). In a prospective trial, six pigs weighing 30+/-5 kg (mean+/-SD) were tracheotomized, submitted to pressure-controlled mechanical ventilation (pc-CMV) and depleted of surfactant by lung lavage. Animals were then mechanically ventilated with three levels of PEEP: 0.5, 1.0 and 1.5 kPa. PLV was then initiated by intratracheal instillation of 30 mL x kg(-1) perfluorocarbon, followed by pc-CMV with PEEP 0.5, 1.0 and 1.5 kPa. Computed tomography (CT)-based analyses of lung volumes and density were obtained after lung lavage, in PLV and during the combined application of PLV and PEEP. Simultaneously, haemodynamics, gas exchange, dynamic compliance (Cdyn) and dynamic resistance (Rdyn) were determined. Statistical analysis was performed using multivariate analyses of variance for repeated measures (p<0.05). In ALI and before PLV, the application of PEEP significantly reduced cardiac output and intrapulmonary shunt. Arterial oxygen tension (Pa,O2) was increased from 6.9 kPa (52 (42, 54) mmHg) (median, (25th and 75th percentile)) to 8.6 kPa (65 (52, 133) mmHg) (PEEP 1.0 kPa) and 15.6 kPa (117 (90, 195) mmHg) (PEEP 15 kPa) (p<0.05). The lung volume obtained by CT increased, CT density was reduced (p<0.05), Cdyn tended to increase and Rdyn to decrease (nonsignificant). PLV increased arterial carbon dioxide tension and reduced pH (p<0.05). CT lung volume and lung density were increased (p<0.05). Superimposing PEEP on PLV increased Pa,O2 from 9.3 kPa (70 (52,124) mmHg) (PEEP 0.5 kPa) to 12.9 kPa (97 (55, 233) mmHg) (PEEP 1.0 kPa) and 403 kPa (303 (64, 426) mmHg) (PEEP 1.5 kPa) (p<0.05), but had no significant effect on CT lung volume and density. It was concluded that in experimental lung injury, positive end-expiratory pressure provided alveolar recruitment. The combined application of positive end-expiratory pressure and partial liquid ventilation significantly augmented oxygenation and might eventually allow either a reduction in the volumes of perfluorocarbons required, or a reduction in positive end-expiratory pressure necessary to maintain pulmonary gas exchange in acute lung injury.