© 2007 American Thoracic Society doi: 10.1164/rccm.200705-784ED
Host–Pathogen Interactions during COPD ExacerbationsMoving on from Microbiology by Numbers?University of Southampton, Southampton, United Kingdom Despite recent advances in our understanding and treatment of chronic obstructive pulmonary disease (COPD), exacerbations of this condition remain an important clinical phenomenon. Indeed, despite their high incidence and impact on health care resources, the etiology of exacerbations remains incompletely understood. In particular, the role that airway bacterial infection plays during these events remains uncertain. The finding that bacterial pathogens are present in significant concentrations in the airways of patients with COPD at exacerbation (1, 2) cannot be considered definitive evidence of their role in triggering these events. Bacteria are also commonly isolated from the airway in patients with stable COPD in significant concentrations (3), and debate centers on how changes in bacterial flora may contribute to the inflammatory milieu seen at exacerbation and whether the key process is that of spontaneously increasing airway bacterial load or acquisition of new bacterial strains. Currently, antibiotics are used to treat a significant proportion of exacerbations, but their efficacy in controlled trials has been equivocal, with definite benefit confined to patients with specific symptoms (4). Therefore, improving our understanding of how airway bacterial infection may trigger exacerbations or modulate their severity may lead to new therapeutic strategies and improve outcomes. In this issue of the Journal (pp. 356–361), Sethi and colleagues test the hypothesis that rising airway bacterial loads are a significant process in triggering COPD exacerbations (5). Their use of a prospective cohort study is appropriate when attempting to address this issue, as only by comparing airway samples from the same population in the stable state and at subsequent exacerbation can dynamic changes in bacteria be properly assessed. This article represents a further analysis of data taken from their long-running cohort, and has used the same airway sampling and laboratory techniques as earlier studies which suggested that new strain acquisition is a possible mechanism of exacerbation induction (6). The study followed 104 patients with COPD for more than 81 months, with data from 560 exacerbations and 2,449 stable visits analyzed. It is thus well powered to address this complex question of the role of airway bacteria in COPD. When there was no observed change in bacterial strain at exacerbation, airway bacterial loads did not rise or even fell. Only when new strains of Haemophilus influenzae and Moraxella catarrhalis were isolated were higher airway bacterial loads were seen. The authors concluded that rising bacterial load is not an independent mechanism of exacerbation induction and that new strain acquisition is a key process. The validity of these conclusions depends upon the study methodology. The use of conventional microbiological culture techniques to identify and quantify airway bacteria carries inherent flaws. First, a number of respiratory pathogens cannot be cultured at all using conventional techniques, and the role of these organisms at exacerbation may be overlooked (7). Furthermore, the quantification of cultured bacteria depends not only on the concentration of organisms in the original airway sample but also the performance of the particular pathogen in culture. Hence, these techniques can underestimate the relative contribution of more fastidious organisms and may have contributed to previous reports of inter-species differences in bacterial load. While the problems with microbiologic culture methodology are common to many studies of this nature, two features peculiar to the work of Sethi and colleagues may compound the inaccuracies of the quantitative technique used. First, estimated bacterial counts were used for a proportion of samples and may have affected recorded concentrations by up to one log unit in the context of observed changes in load of less than one log at exacerbation. Second, patients attended the study clinic on a routine basis rather than only being seen at the onset of exacerbation symptoms. Therefore, the timing of airway sampling may have varied with respect to exacerbation onset from one event to another. As bacterial load, but not strain, fluctuates significantly with time after exacerbation onset (8), this aspect of study design may have introduced considerable noise into the analysis. The findings of Sethi and coworkers may appear to challenge the relationship between high concentrations of airway pathogens and heightened airway inflammation that was found in studies of patients with stable COPD (9–11). Extrapolation from these findings in stable disease led to hypotheses that a sudden rise in bacterial load may trigger exacerbation. However, the findings of the present study suggest that it is important to consider that distinct processes may exist in the stable state and at exacerbation. A significant new antigenic challenge may be required to trigger an inflammatory change sufficient to cause exacerbation. However, low-grade chronic inflammation driven by bacterial colonization may be an important factor, and indeed therapeutic target, in stable disease progression (11, 12). Therefore, quantitative bacteriology using molecular detection techniques, particularly in stable disease, should continue to be a useful tool in future studies. Spontaneous changes in airway flora are unlikely to occur without additional factors that alter the host–pathogen equilibrium. The importance of viral infection in triggering exacerbations has been established by work in this area (13, 14). More recently, study of the combined effect of bacterial and viral infection at exacerbation has revealed a synergistic effect, both on airway inflammation and clinical severity (15, 16). Viral infection plays an important role in modulating the airway immune and inflammatory profile which may alter the delicate balance between colonizing bacteria and host response. It is therefore disappointing that only bacterial strain change and not virus detection was undertaken in the study by Sethi and colleagues, as it is possible that observed strain changes may themselves be the result of viral infection. The analysis for this study uses the relatively robust technique of generalized estimating equations, which can help to determine the relative importance of load and strain changes at exacerbation compared with the stable state. However, the analysis is limited to this "all or nothing" approach without measurements of clinical or inflammatory severity of exacerbation. While changes in bacterial load may not be the primary trigger to exacerbation, they may play an important role in modulating the nature of the event or indeed the therapeutic response. The present study addresses an important etiological conundrum. While the heterogeneity and complexity of the infective processes modulating the nature of inflammation at COPD exacerbation has not been fully addressed, the authors' main assertion remains a valid one. Simply quantifying airway bacteria will not sufficiently advance our understanding of airway infection at exacerbation. The key mechanisms driving COPD exacerbations will only be delineated by developing a more sophisticated approach to studying the complex host–pathogen interactions that exist in this condition. Such analyses are also likely to result in future therapeutic advances. FOOTNOTES Conflict of Interest Statement: T.M.A.W. has no financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
