Antimicrobial-induced endotoxaemia in patients with sepsis in the field of acute pyelonephritis.

BACKGROUND: In vitro results have shown that antimicrobial agents may induce the Gram-negative bacteria to release endotoxins (LPS), which in turn, could trigger the secretion of cytokines from monocytes. AIMS: To compare the effect of cefuroxime, netilmicin or ciprofloxacin on serum levels of LPS and tumour necrosis factor-alpha (TNFalpha). METHODS: Seventy-four patients with acute pyelonephritis caused by Gram-negative bacteria and signs of sepsis were randomly assigned to receive one of three intravenous regimens of cefuroxime, netilmicin or ciprofloxacin. Blood samples were collected before therapy and at specified time intervals for 96 hours after the initiation of treatment for the determination of serum levels of LPS and of TNFalpha. RESULTS: Patients treated with cefuroxime presented an early peak of LPS and of TNFalpha in serum two hours after the initiation of treatment compared to the other study groups. After that time interval, concentrations of LPS and TNFalpha were similar in all the study groups. Fever accompanied by endotoxaemia was still detected for 48 hours after the start of therapy in 36, 37.5 and 36% of patients treated with cefuroxime, netilmicin and ciprofloxacin respectively. The corresponding figures for these agents at 72 hours were 28, 12.5 and 24%, respective and 12, 4.2 and 4% at 96 hours (P value not significant). CONCLUSIONS: With the exception of an early peak in the serum levels of LPS and TNFalpha in patients treated with cefuroxime, no significant difference could be detected amongst the study groups as far as their effect on serum levels of LPS and TNFalpha were concerned. This suggests that these three antimicrobial agents may be administered safely at the early stages of sepsis.

Lipid peroxidation by Pseudomonas aeruginosa in the pathogenesis of nosocomial sepsis.

BACKGROUND: To study whether Pseudomonas aeruginosa may directly trigger peroxidation of polyunsaturated fatty acids, since lipid peroxidation is a mechanism involved in the pathogenesis of sepsis. METHODS: Gamma-linolenic acid (GLA) was administered intravenously at a dose of 25mg/kg in an infusion time of 10 minutes to seven male rabbits. Blood samples were collected from the hepatic veins and from the carotid artery at regular time intervals. One clinical isolate was ex vivo incubated with the serum derived from the latter samples and concentrations of malondialdehyde (MDA) were determined during incubation in the growth medium by the thiobarbiturate assay. RESULTS: Elevated concentrations of MDA compared to their basal levels were found over the first three hours of incubation in the presence of samples collected 30 to 60 minutes after the end of the infusion of GLA. After infusion of GLA concentrations of arachidonic acid in the serum increased to concentrations comparable to those detected in sepsis. CONCLUSION: Direct triggering of lipid peroxidation by nosocomial isolates might be proposed as a pathogenetic mechanism of sepsis.