To study the rate and regulation of alveolar fluid clearance in acute pneumonia, we created a model of Pseudomonas aeruginosa pneumonia in rats. To measure alveolar liquid and protein clearance, we instilled into the airspaces a 5% bovine albumin solution with 1.5 Ci of 125 I-human albumin, 24 h after intratracheal instillation of bacteria. The concentration of unlabeled and labeled protein in the distal airspaces over 1 h was used as an index of net alveolar fluid clearance. Since there was histologic evidence of alveolar epithelial injury, several methods were used to measure alveolar fluid clearance, including the use of experiments in rats with blood flow and the use of experiments in rats without blood flow, so that movement across the epithelial barrier would be minimized in the latter group. The results with each method were identical. We found that P. aeruginosa pneumonia increased alveolar liquid clearance over 1 h by 48% in studies with blood flow, and by 43% in rats without blood flow, compared with respective controls ( P Ͻ 0.05). In both studies, this increase was inhibited with amiloride. However, propranolol had no inhibitory effect, thus ruling out a catecholamine-dependent mechanism to explain the increase in alveolar fluid clearance. An antitumor necrosis factor-␣ neutralizing antibody, instilled into the lung 5 min before bacteria, prevented the increase in alveolar liquid clearance in rats with pneumonia ( P Ͻ 0.05). Also, TNF ␣ (5 g) instilled in normal rats increased alveolar liquid clearance by 43% over 1 h compared with control rats ( P Ͻ 0.05). In normal rats instilled with TNF ␣ , propranolol had no inhibitory effect. In conclusion, gram-negative pneumonia markedly upregulates net alveolar epithelial fluid clearance, in part by a TNF ␣ -dependent mechanism. This finding provides a novel mechanism for the upregulation of alveolar epithelial sodium and fluid transport from the distal airspaces of the lung. ( J. Clin. Invest. 1997. 99:325-335.)