Increased pulmonary edema with crystalloid compared to colloid resuscitation of shock associated with increased vascular permeability.

Physiologic effects of colloidal (5% albumin, 6% hydroxyethylstarch, 6% dextran-70) and crystalloidal (Ringer's lactate) fluids were examined in rats (six in each fluid group) after infusion of an LD99 of rattlesnake venom, previously shown to produce shock secondary to increased vascular permeability. Venom infusion (iv, 2.0 mg/kg, 30 min) was followed by fluid infusion (iv, 30 min) in quantities sufficient to reverse venom-induced hemoconcentration. Venom infusion decreased mean arterial pressure, increased blood lactate, and increased hematocrit in all animals (p less than .01). Fluid infusion reversed these changes, although six times the volume of crystalloid was required to produce hemodilution comparable to the colloidal fluids (120 ml/kg Ringer's lactate, 20 ml/kg colloids). Although no significant changes in the respiratory parameters were noted after administration of the three colloids, Ringer's lactate produced decreases in PaC2 (107 +/- 6 torr, mean +/- SEM to 72 +/- 7, p less than .05), increases in PaCO2 (30 +/- 4 torr to 55 +/- 5, P less than .001), decreases in plasma colloid osmotic pressure (14.2 +/- 0.8 torr to 6.6 +/- 0.9, p less than .001) at the end of fluid infusion. These changes were associated with significantly increased wet-dry lung ratios (p less than .001) in identically prepared animals, sacrificed after fluid infusion. In spite of the development of pulmonary edema in the crystalloid-treated animals, survival was similar for each group (6/6 for albumin and dextran, 5/6 for hydroxyethylstarch and Ringer's lactate). We thus conclude that both colloidal and crystalloidal fluid resuscitation leads to survival in permeability shock. Resuscitation with crystalloidal fluid, however, requires significantly greater volumes and is associated with the development of pulmonary edema.