BACKGROUND: Variable ventilation recruits alveoli in atelectatic lungs, but it is unknown how it compares with conventional recruitment manoeuvres.

OBJECTIVES: To test whether mechanical ventilation with variable tidal volumes and conventional recruitment manoeuvres have comparable effects on lung function.

DESIGN: Randomised crossover study.

SETTING: University hospital research facility.

ANIMALS: Eleven juvenile mechanically ventilated pigs with atelectasis created by saline lung lavage.

INTERVENTIONS: Lung recruitment was performed using two strategies, both with an individualised optimal positive-end expiratory pressure (PEEP) associated with the best respiratory system elastance during a decremental PEEP trial: conventional recruitment manoeuvres (stepwise increase of PEEP) in pressure-controlled mode) followed by 50 min of volume-controlled ventilation (VCV) with constant tidal volume, and variable ventilation, consisting of 50 min of VCV with random variation in tidal volume.

MAIN OUTCOME MEASURES: Before and 50 min after each recruitment manoeuvre strategy, lung aeration was assessed by computed tomography, and relative lung perfusion and ventilation (0% = dorsal, 100% = ventral) were determined by electrical impedance tomography.

RESULTS: After 50 min, variable ventilation and stepwise recruitment manoeuvres decreased the relative mass of poorly and nonaerated lung tissue (percent lung mass: 35.3 ± 6.2 versus 34.2 ± 6.6, P = 0.303); reduced poorly aerated lung mass compared with baseline (-3.5 ± 4.0%, P = 0.016, and -5.2 ± 2.8%, P < 0.001, respectively), and reduced nonaerated lung mass compared with baseline (-7.2 ± 2.5%, P < 0.001; and -4.7 ± 2.8%, P < 0.001 respectively), while the distribution of relative perfusion was barely affected (variable ventilation: -0.8 ± 1.1%, P = 0.044; stepwise recruitment manoeuvres: -0.4 ± 0.9%, P = 0.167). Compared with baseline, variable ventilation and stepwise recruitment manoeuvres increased PaO2 (172 ± 85mmHg, P = 0.001; and 213 ± 73 mmHg, P < 0.001, respectively), reduced PaCO2 (-9.6 ± 8.1 mmHg, P = 0.003; and -6.7 ± 4.6 mmHg, P < 0.001, respectively), and decreased elastance (-11.4 ± 6.3 cmH2O, P < 0.001; and -14.1 ± 3.3 cmH2O, P < 0.001, respectively). Mean arterial pressure decreased during stepwise recruitment manoeuvres (-24 ± 8 mmHg, P = 0.006), but not variable ventilation.

CONCLUSION: In this model of lung atelectasis, variable ventilation and stepwise recruitment manoeuvres effectively recruited lungs, but only variable ventilation did not adversely affect haemodynamics.

TRIAL REGISTRATION: This study was registered and approved by Landesdirektion Dresden, Germany (DD24-5131/354/64).