We studied the effect of iv administration of biodegradable macromolecules on microvascular permeability after ischemia-reperfusion injury in a rat gastrocnemius model. After 2 h of tourniquet ischemia of the rats' hind limb, groups of animals were given iv lactated Ringer's solution (RL), serum albumin 5%, or varying MW fractions of biodegradable macromolecules of hydroxyethyl starch (HES), glycogen, and dextran. At the conclusion of the 24-h reperfusion period, the rat gastrocnemius muscles were collected. Water and K+ differences between the ischemic and control muscles were compared. Rats given a 100,000 to 300,000-dalton fraction of HES had significantly decreased water content (5.1 +/- 3.4%) when compared to rats receiving RL (8.3 +/- 2.2, p less than .01), less than 100,000 dalton HES (8.3 +/- 3.2, p less than .05), less than 300,000 glycogen (7.9 +/- 2.5, p less than .01), or dextran 150,000 (8.3 +/- 1.5, p less than .05). Rats given 100,000 to 300,000-dalton HES also had significantly higher ischemic muscle K+ content as compared to the nontourniquet control (difference 14.2 +/- 9.7 mEq/g) than rats receiving any of the other solutions (range 32.5 to 39.3) except the 300,000 to 1,000,000-dalton fraction of HES. Regression analysis comparison of K+ difference to the histologic evaluation of the muscles on the criteria of polymorphonuclear infiltration and interstitial edema (0, best; 3, worst) had a Pearson correlation coefficient of r = .73. Reduction of abnormally increased microvascular permeability may be accomplished by the iv use of appropriate sized biodegradable macromolecules.