The influence of colloidal plasma substitutes on the release of tumor necrosis factor alpha (TNF-alpha) in human whole blood in vitro

Objective: To investigate the influence of colloidal plasma substitutes on human cytokine network, especially of tumor necrosis factor alpha (TNF-alpha), in vitro. Design: Heparinized whole blood samples from 8 healthy volunteers were divided and set on various concentrations of artificial colloidal plasma substitutes (native = 0 mg/ml; 5 mg/ml; 15 mg/ml). As colloids were used hydroxyethyl starch 200/0,5 (HES), dextran 60 (DX), urea-linked gelatin (GEL) and human albumin solution (HA). After incubation (24 h; 5% CO2; 37 degrees C; with and without concomitant stimulation of blood cells with phytohemagglutinin [PHA]) measurement of TNF-alpha release was performed via ELISA by the method described by Gallati. For the statistical evaluation a repeated measures analysis of variance was used. Results: Basic level of TNF-alpha was found between 226 and 273 pg/ml (0 mg of each colloid/ml), stimulation with PHA without any colloid increased the TNF-alpha level about fourfold (1,066-1,260 pg/ml; 0 mg of each colloid/ml). At 5 mg/ml and 15 mg/ml without PHA all 3 artificial colloids rose the level of TNF-alpha (up to 50%). Under concomitant stimulation each colloid induced additional TNF-alpha release in comparison to PHA alone. The changes elicited by DX and GEL were statistically significant (p<0.001 and p=0.005, respectively) in contrast to those induced by human albumin solution or HES. Conclusion: In relation to TNF-alpha plasma substitutes are not inert substances as perhaps suspected. The questions whether the observed effects exist in vivo, how far other cytokines are influenced and the question about the clinical importance are subject of ongoing studies.