Kinetics of circulating TNF-alpha and TNF soluble receptors following surgery in a clinical model of sepsis.

The interrelationship between cytokines and their natural antagonists in patients with systemic sepsis are incompletely understood. We have followed the changes in serum levels of TNF-alpha and the two soluble receptors (TNF-sr) in a clinical model of post-operative sepsis. Serial blood samples were taken in patients undergoing percutaneous nephrolithotomy (PCNL) starting pre-operatively and continuing for 24 h thereafter. The levels of TNF-alpha and TNF-sr were raised in patients who became clinically septic and correlated well with the severity of sepsis (using the APACHE III score). In septic patients there was no difference in the pattern of changes in the two types of receptor (TNF-sr55 and TNF-sr75). However, in non-septic patients TNF-sr75 was higher in those with endotoxaemia than those without. This difference was not observed with TNF-sr55 which suggests a different mechanism of release or degree of sensitivity for the two soluble receptors. Regardless of severity of illness, the levels of all three molecules (TNF-alpha and the two receptors) appeared to start rising at about the same time point. The peak TNF-alpha level was reached earlier (2-4 h) than that of the two TNF-sr (4-8 h). The relative rise in TNF-alpha was greater than that of the soluble receptors and this difference was even more marked in those with more severe sepsis. The relationship between peak TNF-alpha and peak TNF-sr was non-linear and the concentration of each TNF-sr appeared to plateau at the higher levels of TNF-alpha. This suggests the exhaustion of a limited pool or saturation of the rate of release. Taken together, these results suggest sepsis develops because of delayed and insufficient secretion of TNF-sr compared with TNF-alpha.