Complement-mediated neutrophil dysfunction in critical illness

• Main Author: Conway Morris, Andrew
• Published: University of Edinburgh 2011
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.735469

Critical illness, constituting an acute illness or injury resulting in organ dysfunction and failure, is associated with a profound, systemic activation of the immune system and inflammation-mediated organ damage. However, critically ill patients suffer a high rate of nosocomial infection with secondary sepsis being a common cause of death. This high prevalence of secondary infections argues for the influence of an immune suppression that may, at first glance, appear paradoxical in light of the pro-inflammatory nature of critical illness. Although immune cell hypo-function has been noted in clinical and experimental critical illness, the mediators of these effects remain poorly defined. In this thesis, neutrophil function was examined in the context of clinically suspected ventilator-associated pneumonia, a common and lethal intensive care acquired infection (ICU-AI), This demonstrated impaired bactericidal functions (phagocytosis and reactive oxygen species production) in neutrophils from both the peripheral and pulmonary compartments; however there was ample evidence of coexistent neutrophil activation (both cell surface markers and soluble mediators) and inflammation. An investigation of possible mediators of neutrophil dysfunction revealed a major role for C5a, the pro-inflammatory anaphylotoxin derived from complement C5. Recombinant C5a applied to healthy donor neutrophils was able to drive the defect in phagocytosis by phospho-inositol 3 kinase delta-mediated inhibition of RhoA and subsequent down regulation of actin polymerisation. The defects in RhoA and actin function were reversible with granulocyte-macrophage colony stimulating factor (GM-CSF) applied ex vivo, restoring phagocytic function to normal. Similar defects in RhoA and actin, and effective treatment with GMCSF, were found in neutrophils from critically ill patients. In a second cohort of critically ill adults recruited prior to developing any ICU-AI, C5a-mediated neutrophil dysfunction was an independent predictor of acquiring nosocomial infection, being associated with a 5.4 fold increased risk (95% Confidence interval 1.4-21.0). The same cohort of patients also displayed two other features of immune suppression, namely monocyte deactivation and elevated proportions of regulatory T-cells that were also associated with increased risk of infection (relative risk of infection 3 (95%CI1.3-6.9) and 2.4 (95%CI1.3-4.2) respectively). These measures acted additively with C5a-mediated dysfunction, those with no immune impairment having a zero rate of nosocomial infection with cumulative increases in impairments being associated with a progressive increase in risk of infection (p=0.0004 by Chi squared for trend). In conclusion, critical illness is characterised by a complex inflammatory state with features of simultaneous hyper- and hypoactivation. This remarkable duality is illustrated by the ability of a proinflammatory molecule, C5a, to drive neutrophil dysfunction, with this dysfunction being associated with a serious adverse event-nosocomial infection.