To explore whether routinely administered colloids and crystalloids influence the hemostatic system, we studied 60 patients undergoing knee replacement surgery during randomized intravascular fluid administration using 6% hydroxyethyl starch 200/0.5 (HES) or 4% modified gelatin (GEL) in addition to a basal infusion of lactated Ringer's solution (RL), or exclusively RL. In addition to routine coagulation tests, measurements of coagulation factors were performed. Also, functional measurements of the in vitro bleeding time by use of the platelet function analyzer (PFA-100 and ROTEG analysis (ROTEG(R); extrinsically and intrinsically [Ex; In] activated measurements of clotting time, CT [s]; clot formation time, CFT [s]; clot strength, A20 [mm]; fibrinogen component of the clot, FibA20 [mm]; and maximal clot elasticity) were used. Time dependency of variables was analyzed with a repeated-measures analysis of variance (all groups pooled); differences between groups were detected by comparing the calculated area under the curve (AUC(A-D)). For all variables, except ExCT, ExCFT, and InCFT, a significant time dependency was demonstrated, indicating that impaired platelet-mediated hemostasis and clot formation occurred with IV administration of fluids. Total clot strength, fibrinogen part, and clot elasticity decreased significantly more in the colloid groups than in the RL group (InA20: HES, -13.0 mm; GEL, -11.5 mm; RL, -1.3 mm; P = 0.042; FibA20: HES, -10.5 mm; GEL, -6.0 mm; RL, -1.3 mm: P < 0.0001; MCE: HES, -48; GEL, -35; RL, -15.8; P < 0.0001). The decrease in fibronectin concentrations was significantly smaller with GEL as compared with HES, whereas a weak trend toward a larger decrease in fibrinogen concentrations was observed with both colloids. Results show that colloid administration reduces final clot strength more than does RL alone, which also exhibited effects, albeit minor, on the coagulation system. The reduction in total clot strength was due to impaired fibrinogen polymerization, resulting in a decreased fibrinogen part of the clot and reduced clot elasticity.
Implications: Our data suggest that during deliberate colloid administration, critically impaired fibrinogen polymerization and reduced fibrinogen concentrations might be reached earlier than expected. Therefore, maintaining fibrinogen concentrations seems essential when continuing blood loss is bridged by colloid infusion until transfusion triggers are reached, especially in patients already exhibiting borderline fibrinogen levels at baseline.