The release of endotoxin, TNF and IL-6 during the antibiotic treatment of experimental Gram-negative sepsis

To evaluate the role of different antibiotics in the release of endotoxin and the production of tumor necrosis factor-α (TNF) and interleukin 6 (IL-6) during the treatment of experimental Escherichia coli septical peritonitis, we obtained serial blood samples from septic rats treated with placebo, ceftazidime, aztreonam or imipenem. We also studied the effect of taurolidine, given alone or in combination with aztreonam, on the release of endotoxin and IL-6. Despite decreasing levels of viable counts after treatment with ceftazidime, aztreonam or imipenem, levels of free endotoxin increased in all animals. We did not notice any significant differences in the extent of plasma endotoxin release between the different treatment groups. However, we did find significant differences in the IL-6 production between the different treatment groups. After 2 h of treatment, IL-6 levels had increased in all animals with the highest levels in the imipenem treated animals, whereafter IL-6 levels decreased again in the rats treated with imipenem or ceftazidime, while in the rats treated with placebo or aztreonam IL-6 levels further increased. This increase in IL-6 levels was associated with acute mortality. In all antibiotic treated animals TNF levels significantly decreased during therapy. After 2 h of treatment TNF levels were the highest in the imipenem treated rats. The high levels of TNF and IL-6 at t = 2 in the imipenem group were thought to be the result of early bacterial lysis, while the late increase in IL-6 levels in the aztreonam treated animals was thought to be the result of the formation of long bacterial filaments io the abdominal cavity. In the present study, treatment with taurolidine could not prevent or inhibit the release of endotoxin or IL-6, but taurolidine, alone or in combination with aztreonam, unexpectedly caused a dramatic increase in IL-6 levels which was associated with an increased acute mortality. We conclude that antibiotics can cause the release of endotoxin in spite of decreasing levels of bacteremia in vivo. It is suggested that circumstances in which antibiotic-induced filamentation occurs are also conditions that yield excessive (local) LPS release. Our data also suggest that there is a lack of relationship between plasma free endotoxin levels and mortality and that the most important inflammatory compartment was the abdominal cavity in this model