ARDS is patchy even when the initial insult to the lung is diffuse, but the spatial distribution of injury and potential mediators have not been fully explored. Increased vascular permeability is central to the pathogenesis of ARDS, and regional differences may contribute to heterogeneous lung injury. Our first hypothesis is that rates of protein leak vary throughout the lung during endotoxemia and have a non-random spatial distribution. We will measure rates of albumin leak into the extravascular space in 1.0 cm3 lung regions in endotoxemic pigs to characterize their spatial distribution, and validate leak rates as measures of local histologic injury. The observed spatial distribution of leak rates will suggest mediators of protein leak that share a similar spatial distribution. Our second hypothesis is that differences in regional pulmonary blood flow (Q) cause local differences in vascular permeability and protein leak. We will make high resolution measurements of Q, albumin leak rate, and plasma volume (a surrogate for capillary surface area [CSA]) to assess the correlation between regional Q and albumin leak rates, and determine whether this correlation is explained by effects of Q on vascular permeability or capillary surface area. Our third hypothesis is that variability in regional Q is increased during endotoxemia, and redistribution of regional Q contributes to ventilation/perfusion (V/Q) inequality rather than resulting from hypoxic pulmonary vasoconstriction. We will make high resolution measurements of regional V and Q before and during endotoxemia and quantify changes in their variability, spatial distribution, and V/Q matching. Characterizing regional differences in protein leak is an important first step in understanding the heterogeneity of lung injury and potential mediators of non-cardiogenic pulmonary edema. If high regional blood flows or associated microvascular pressures increase vascular permeability during endotoxemia, strategies that decrease pulmonary Q or its heterogeneity may prevent pulmonary vascular injury in patients with sepsis-induced lung injury.