Rationale: Spontaneous breathing during mechanical ventilation increases transpulmonary pressure and VT, and worsens lung injury. Intuitively, controlling VT and transpulmonary pressure might limit injury caused by added spontaneous effort. Objectives: To test the hypothesis that, during spontaneous effort in injured lungs, limitation of VT and transpulmonary pressure by volume-controlled ventilation results in less injurious patterns of inflation. Methods: Dynamic computed tomography was used to determine patterns of regional inflation in rabbits with injured lungs during volume-controlled or pressure-controlled ventilation. Transpulmonary pressure was estimated by using esophageal balloon manometry [PL(es)] with and without spontaneous effort. Local dependent lung stress was estimated as the swing (inspiratory change) in transpulmonary pressure measured by intrapleural manometry in dependent lung and was compared with the swing in PL(es). Electrical impedance tomography was performed to evaluate the inflation pattern in a larger animal (pig) and in a patient with acute respiratory distress syndrome. Measurements and Main Results: Spontaneous breathing in injured lungs increased PL(es) during pressure-controlled (but not volume-controlled) ventilation, but the pattern of dependent lung inflation was the same in both modes. In volume-controlled ventilation, spontaneous effort caused greater inflation and tidal recruitment of dorsal regions (greater than twofold) compared with during muscle paralysis, despite the same VT and PL(es). This was caused by higher local dependent lung stress (measured by intrapleural manometry). In injured lungs, esophageal manometry underestimated local dependent pleural pressure changes during spontaneous effort. Conclusions: Limitation of VT and PL(es) by volume-controlled ventilation could not eliminate harm caused by spontaneous breathing unless the level of spontaneous effort was lowered and local dependent lung stress was reduced.