Plasma colloid oncotic pressure-pulmonary artery occlusion pressure gradient. A poor predictor of pulmonary edema in surgical intensive care unit patients

T. D. Rafferty, R. Ljungquist, L. Firestone, A. Curtis, C. Raven, S. Hui and P. G. Barash

Because Starling's equation contains four factors that theoretically influence fluid movement across the pulmonary capillary bed, we prospectively examined the relationship between the plasma colloid oncotic pressure (PCOP) minus the mean pulmonary artery occlusion pressure (PAOP) (the only two presently available clinically for measurement) and pulmonary edema determined in terms of percent venous admixture (Qs/Qt) and roentgenographically in 17 consecutive patients in a surgical intensive care unit to determine whether this PCOP-PAOP gradient could accurately predict the presence of pulmonary edema. The PCOP-PAOP gradient proved to be a poor predictor of pulmonary edema determined by these means. We believe this resulted from the inherent insensitivity of the PCOP-PAOP gradient as an estimate of the net intravascular filtration pressure, and Qs/Qt and roentgenograms as measurements of lung edema, as well as the multiplicity of variables involved in the genesis of pulmonary edema in the clinical situation.