Abstract

Severe pulmonary edema sometimes develops despite normal pulmonary capillary wedge pressure (Ppw). The equation describing net transvascular flux of lung liquid predicts decreased edema when hydrostatic pressure is reduced or when colloid osmotic pressure is increased in the pulmonary vessels. We tested these predictions in a model of pulmonary capillary leak produced in 35 dogs by intravenous oleic acid. 1 h later, the dogs were divided into five equal groups and treated for 4 h in different ways: (a) not treated, to serve as the control group (Ppw = 11.1 mm Hg); (b) given albumin to increase colloid osmotic pressure by 5 mm Hg (Ppw = 10.6 mm Hg); (c) ventilated with 10 cm H2O positive end-expiratory pressure (Peep) (transmural Ppw = 10.4 mm Hg); (d) phlebotomized to reduce Ppw to 6 mm Hg; (e) infused with nitroprusside, which also reduced Ppw to 6 mm Hg. Phlebotomy and nitroprusside reduced the edema in excised lungs by 50% (P< 0.001), but Peep and albumin did not affect the edema. Pulmonary shunt decreased on Peep and increased on nitroprusside, and lung compliance was not different among the treatment groups, demonstrating that these variables are poor indicators of changes in edema. Cardiac output decreased during the treatment period in all but the nitroprusside group, where Ppw decreased and cardiac output did not. We conclude that canine oleic acid pulmonary edema is reduced by small reductions in hydrostatic pressure, but not by increased colloid osmotic pressure, because the vascular permeability to liquid and protein is increased. These results suggest that low pressure pulmonary edema may be reduced by seeking the lowest Ppw consistent with adequate cardiac output enhanced by vasoactive agents like nitroprusside. Further, colloid infusions and Peep are not helpful in reducing edema, so they may be used in the lowest amount that provides adequate circulating volume and arterial O2 saturation on nontoxic inspired O2. Until these therapeutic principles receive adequate clinical trial, they provide a rationale for carefully monitored cardiovascular manipulation in treating patients with pulmonary capillary leak.

Authors

R. M. Prewitt, J. McCarthy, L. D. H. Wood

×

Other pages: