Decreasing hydrostatic pressure does not uniformly decrease high-pressure pulmonary edema.

Pulmonary artery wedge pressure (PAWP) of 30 mm Hg with left atrial balloon inflation for 1 1/2 hours produced pulmonary edema in eight dogs. PAWP was then decreased to 10 mm Hg for two hours, and shunt, lung water (extravascular thermal volume, or ETV, by thermal dye), and perfusion distribution (radiomicrosphere technique) were measured and compared with four other dogs (group 1) whose PAWP was maintained at 10 mm Hg. The eight dogs with PAWP of 30 mm Hg for 1 1/2 hours were retrospectively subdivided into two groups of four based on ETV (group 2, double baseline ETV; group 3, triple baseline ETV). Baseline ETV and shunt were similar for all groups and remained unchanged for group 1. At 1 1/2 hours, 2 hours (1/2 hour after decreasing PAWP), 2 1/2 hours, and 3 1/2 hours, respectively, ETV were: 13.9 +/- 1.9, 12.8 +/- 2.0, 9.3 +/- 1.5, and 8.5 +/- 1.0 ml/kg in group 2; and 21.9 +/- 2.1, 22.7 +/- 2.2, 22.5 +/- 2.0, and 22.2 +/- 2.0 ml/kg in group 3. A more variable rate of edema formation was detected in eight additional dogs, but failure to resolve higher levels of edema after decreasing PAWP was also demonstrated in this group. Edema was greatest in lower lobes and decreased lobar perfusion. Shunt was higher in group 3 than in group 2 at 1 1/2 hours and decreased in group 2 but not in group 3 at 3 1/2 hours. Changes in colloid osmotic pressure may account for the differences in edema formation and resolution, but our data suggest that, independent of the rate of edema formation, a decrease in vascular exchange surface area at higher levels of edema may inhibit edema resolution when PAWP is decreased.