Chronic obstructive pulmonary disease (COPD) is a chronic inflammatory condition of the lung caused by an abnormal response to noxious gases and particles, particularly cigarette smoke. There is increasing evidence that bacteria contribute to the pathophysiology of COPD both during stable and exacerbation states, and the use of molecular diagnostic techniques have increased our ability to detect and thus investigate their relationship with clinical features, particularly during periods of stability. The main aim of this thesis was to investigate the hypothesis that airway pathogenic bacteria presence, load and species, as detected by qPCR, in stable COPD drives a differential inflammatory response, contributing to systemic inflammation and altering the natural history of the disease, including exacerbation susceptibility and characteristics, and may be related to defective innate immunity, particularly impaired macrophage phagocytosis.

The work in this thesis has demonstrated that in stable COPD there is an apparent bacterial load threshold, at which airway inflammation is significantly greater than in patients with either low bacterial loads or no pathogenic bacteria. Patients with pathogenic airway bacteria above this load threshold have a shorter time to next exacerbation, suggesting bacterial load is an important modulator of exacerbation susceptibility. H. influenzae has been consistently shown to play an important role in the pathogenesis of COPD, with a species-specific inflammatory response observed not only in the stable state, but also associated with its increased load at exacerbation, independent of other bacterial or viral pathogens, although this inflammatory response was discordant with patient reported outcomes. No species-specific or bacterial load effect was seen in systemic inflammation, except with changes in M. catarrhalis load at exacerbation, and there were generally poor relationships between the same biomarkers measured in both airway and systemic compartments. In addition, bacterial colonisation appears to be influenced by bacterial load at exacerbation although a prior stable sputum sample with pathogenic bacteria detected is a good predictor of subsequent stable state bacterial presence.

Using monocyte-derived macrophages (MDMs) as a model of alveolar macrophages, impaired phagocytosis was investigated as a mechanism for bacterial colonisation. Although decreased MDM phagocytosis to H. influenzae at stable state was associated with higher exacerbation frequency, this could not be explained by the relationship between phagocytosis and bacterial presence, load or species. Furthermore, MDM phagocytosis is a stable phenomenon and cannot provide a biological explanation for changes in colonisation status in stable disease or the heterogeneity of exacerbation aetiology. However, differential activation of MDMs occurs at exacerbations, which may represent different monocyte populations.

The findings from this thesis have important implications in the management of COPD to modifying exacerbation risk. The evidence provided should encourage clinical trials to investigate the use of prophylactic antibiotics specifically for COPD patients with high bacterial loads and those at risk of H. influenzae infection at both stable and exacerbated states, and the future development of novel treatments such as specific anti-cytokine agents or those designed to improve macrophage phagocytosis, thereby improving the clinical outcomes for patients with COPD.