Methods: We analysed the effects of complement depletion and of C5a inhibition on haemodynamic parameters, oxygen delivery (DO2), oxygen consumption (VO2), oxygen extraction ratio (OER) and blood lactate levels after live bacteria infusion in pigs.
Results: In the first series of experiments, animals were decomplemented by cobra venom factor (CVF, 125 micrograms kg-1) and challenged with 1.3 x 10(9) Escherichia coli kg-1. In a second series, animals were treated with neutralizing anti-C5a monoclonal antibodies (mAb) T13/9 before infusion of an increased E. coli dosage (1 x 10(10) E. coli kg-1). Administration of Gram-negative bacteria resulted in hypotension, tachycardia, pulmonary hypertension and decreased cardiac output typical for severe sepsis. These alterations were more pronounced in animals challenged with a higher bacteria concentration (1 x 10(10) E. coli kg-1, n = 5) than with a lower dosage (1.3 x 10(9) E. coli kg-1, n = 4). Complement depletion by CVF injection 24 h before E. coli infusion (n = 4), or anti-C5a mAb T13/9 administration (n = 4) had no effect on the changes in haemodynamic parameters and in DO2 associated with E. coli challenge. Application of either 1.3 x 10(9) or 1 x 10(10) E. coli kg-1 resulted in a marked decrease in VO2 and an increase in blood lactate levels, whereas the OER did not change throughout the experiment. In contrast, pretreatment with CVF 24 h before low-dose E. coli (1.3 x 10(9) kg-1) administration resulted in a significant increase in VO2 (P < 0.05) and in OER (P < 0.05) compared with untreated septic animals (n = 4). No hyperlactaemia occurred in complement-depleted septic animals compared with complement-sufficient animals (P < 0.05). Animals challenged with a high E. coli dose (1 x 10(1) kg-1) and treated with anti-C5a mAbs showed a pronounced increase in VO2 and OER (P < 0.05) accompanied by an attenuated increase in lactate levels (P < 0.05) compared with untreated septic animals.
Conclusion: The results demonstrate an improved oxygen use after complement depletion in this model of severe Gram-negative sepsis. Furthermore, a similar effect was seen after specifically neutralizing C5a by mAbs, indicating a role of C5a in the underlying mechanism.