BACKGROUND: We dissected total power into its primary components to resolve its relative contributions to tissue damage (VILI). We hypothesized that driving power or elastic (dynamic) power offers more precise VILI risk indicators than raw total power. The relative correlations of these three measures of power with VILI-induced histologic changes and injury biomarkers were determined using a rodent model of acute respiratory distress syndrome (ARDS). Herein, we have significantly extended the scope of our previous research. METHODS: Data analyses were performed in male Wistar rats that received endotoxin intratracheally to induce ARDS. After 24 h, they were randomized to 1 h of volume-controlled ventilation with low V(T) = 6 ml/kg and different PEEP levels (3, 5.5, 7.5, 9.5, and 11 cmH(2)O). Applied levels of driving power, dynamic power inclusive of PEEP, and total power were correlated with VILI indicators [lung histology and biological markers associated with inflammation (interleukin-6), alveolar stretch (amphiregulin), and epithelial (club cell protein (CC)-16) and endothelial (intercellular adhesion molecule-1) cell damage in lung tissue]. RESULTS: Driving power was higher at PEEP-11 than other PEEP levels. Dynamic power and total power increased progressively from PEEP-5.5 and PEEP-7.5, respectively, to PEEP-11. Driving power, dynamic power, and total power each correlated with the majority of VILI indicators. However, when correlations were performed from PEEP-3 to PEEP-9.5, no relationships were observed between driving power and VILI indicators, whereas dynamic power and total power remained well correlated with CC-16 expression, alveolar collapse, and lung hyperinflation. CONCLUSIONS: In this mild-moderate ARDS model, dynamic power, not driving power alone, emerged as the key promoter of VILI. Moreover, hazards from driving power were conditioned by the requirement to pass a tidal stress threshold. When estimating VILI hazard from repeated mechanical strains, PEEP must not be disregarded as a major target for modification.