This study was designed to compare the influence of various plasma sub
stitutes, administered for preoperative hemodilution, on blood rheolog
y. We studied 40 patients, ASA grade I, who underwent elective facial
reconstructive surgery and received 4% albumin (n = 10), 3.5% dextran
40 (n = 10), gelatin (n = 10), or hydroxyethyl starch (HES) (n = 10).
Ten patients, undergoing the same surgical procedure without hemodilut
ion, were chosen as controls. After hemodilution, hematocrit was decre
ased approximately 30%. Fibrinogen decreased in all tested groups exce
pt in the gelatin group. Plasma viscosity decreased with albumin (1.13
+/- 0.05 to 1.06 +/- 0.03 mPa.s; P < 0.01) and increased with HES (1.
15 +/- 0.04 to 1.22 +/- 0.05 mPa.s; P < 0.01). At a high shear rate, t
he blood viscosity decreased in all groups. In contrast, at a low shea
r rate and at 40% corrected hematocrit, the blood viscosity decreased
in the albumin (15.9 +/- 1.9 to 13.1 +/- 2.1 mPa.s; P < 0.01) and the
dextran 40 (16.9 +/- 2.9 to 12.8 +/- 2.5 mPa.s; P < 0.01) groups and w
as unchanged in the gelatin and the HES groups. Erythrocyte aggregatio
n (measured with primary aggregation time) was markedly decreased in t
he albumin (3.27 +/- 1.74 to 7.03 +/- 2.95 s; P < 0.01) and in the dex
tran 40 (2.72 +/- 0.58 to 6.24 +/-2.55 s; P < 0.001) groups, unchanged
with HES, and increased with gelatin (2.41 +/- 0.90 to 1.55 +/- 0.33
s). These findings suggest that albumin and dextran 40 may be the plas
ma substitutes of choice for preoperative hemodilution when this techn
ique aims to improve rheologic conditions.