OBJECTIVES: To examine the potentially modifiable drivers that injure and heal the "baby lung" of acute respiratory distress syndrome and describe a rational clinical approach to favor benefit. DATA SOURCES: Published experimental studies and clinical papers that address varied aspects of ventilator-induced lung injury pathogenesis and its consequences. STUDY SELECTION: Published information relevant to the novel hypothesis of progressive lung vulnerability and to the biophysical responses of lung injury and repair. DATA EXTRACTION: None. DATA SYNTHESIS: In acute respiratory distress syndrome, the reduced size and capacity for gas exchange of the functioning "baby lung" imply loss of ventilatory capability that dwindles in proportion to severity of lung injury. Concentrating the entire ventilation workload and increasing perfusion to these already overtaxed units accentuates their potential for progressive injury. Unlike static airspace pressures, which, in theory, apply universally to aerated structures of all dimensions, the components of tidal inflation that relate to power (which include frequency and flow) progressively intensify their tissue-stressing effects on parenchyma and microvasculature as the ventilated compartment shrinks further, especially during the first phase of the evolving injury. This "ventilator-induced lung injury vortex" of the shrinking baby lung is opposed by reactive, adaptive, and reparative processes. In this context, relatively little attention has been paid to the evolving interactions between lung injury and response and to the timing of interventions that worsen, limit or reverse a potentially accelerating ventilator-induced lung injury process. Although universal and modifiable drivers hold the potential to progressively injure the functional lung units of acute respiratory distress syndrome in a positive feedback cycle, measures can be taken to interrupt that process and encourage growth and healing of the "baby lung" of severe acute respiratory distress syndrome.