The effect of 6% hydroxyethyl starch (hetastarch) and 5% human serum albumin (albumin) infusion on coagulation in 12 patients with septic shock was evaluated. Patients were randomly assigned to receive either albumin (N = 6) or hetastarch (N = 6) infusion over a 24-hour study interval. The prothrombin time (PT), partial thromboplastin time (PTT), and quantitative platelet count (PC) were obtained prior to and following 24 hours of fluid infusion. Hetastarch patients received 4934 +/- 1354 mL and albumin patients received 3067 +/- 256 mL over the study interval. After hetastarch infusion, the PT increased 2.2 +/- 0.7 seconds, PTT increased 20.0 +/- 4.1 seconds (P less than .01), and PC decreased 158 +/- 36 X 10(3)/mm3 (P less than .02). After albumin administration, the PT decreased 1.2 +/- 1.7 seconds, PTT increased 20.5 +/- 10.6 seconds and PC decreased 100 +/- 34 X 10(3)/mm3. There were no significant differences in the changes in PT, PTT, or PC between the groups. The authors conclude that hetastarch infusion does not result in increased bleeding in patients with septic shock compared with albumin infusion, despite moderate effects on the hemostatic coagulation profile.