The measurement of antibodies against pathogens is mainly done for diagnostic purposes in order to detect the aetiology of infections or to confirm that protection following vaccination is present against a certain disease.

In some diseases such as chronic Pseudomonas aeruginosa lung infection in patients with cystic fibrosis antibodies induced by the infection obviously do not offer any protection against the infection. On the contrary, high and rapidly increasing levels of antibodies are correlated with a poor prognosis.1 The reason for this is that immune elimination is not mediated by the antibody response because the alginate producing bacteria which are so characteristic of this disease grow as biofilms in the bronchioles.2 The biofilm mode of growth is an ancient survival strategy developed by bacteria living in the environment to exploit the micro-milieu of nutrients on surfaces and to avoid being killed by amoeba and toxic chemicals.2 As a consequence, antigen-antibody reactions take place around the surface of the P aeruginosa biofilm in the airways of patients with cystic fibrosis, eventually leading to chronic inflammation dominated by polymorphonuclear leucocytes. The process of frustrated phagocytosis around the biofilm leads to liberation of proteolytic enzymes such as elastase and oxygen radicals which gradually destroy the tissue of the lungs in patients with cystic fibrosis.2 The antibody response againstP aeruginosa in cystic fibrosis is therefore both a marker of chronicity of the infection and a marker of the inflammation and tissue damage. Measurement of antibodies againstP aeruginosa in patients with cystic fibrosis is therefore done in some centres to distinguish between intermittent P aeruginosa colonisation, which does not give rise to a significant antibody response, and chronic P aeruginosa infection characterised by a significant antibody response.1 ,3 The predictive values of both a positive and a negative antibody test to discriminate between intermittent colonisation and chronic infection is about 90%.4

This has important therapeutic consequences as it is possible to prevent chronic P aeruginosa infection in 80% of patients by early aggressive antibiotic treatment of the intermittent colonisation.5 Likewise, with aggressive suppressive antibiotic treatment of chronic P aeruginosa infection it is possible to maintain lung function for many years and to improve the prognosis of patients with cystic fibrosis, although the infection is virtually never eradicated.6 Generation of the knowledge behind these results has only been possible because patients with cystic fibrosis are followed and treated in large specialised centres. It is therefore possible to obtain blood samples for antibody measurements before and during the onset and course of the intermittent and chronicP aeruginosa infection and to correlate the results with the clinical status and laboratory findings of the patients in longitudinal studies.

This is obviously seldom the case with other diseases. One is therefore forced to rely on the results of transversal studies and to search for any pathogenetic, clinical, or immunological findings which may show an analogy with the chronic P aeruginosainfection in cystic fibrosis. Such hypothesis generating observations may then initiate clinical trials to investigate whether the principles of treatment of infections in cystic fibrosis may also be of benefit to other groups of patients. More than 30 years ago Burns and May7-9 initiated such a series of studies in patients with cystic fibrosis and in adult patients with other chronic bronchial disorders. Similar studies were started in other countries.10-15 The results were concordant in cystic fibrosis patients as described above and also showed that a few non-cystic fibrosis patients with P aeruginosa lung infection had high levels of anti-P aeruginosa antibodies and were harbouring mucoid strains.16 ,17 However, none of these authors continued the research in the non-cystic fibrosis patients and it is therefore not known whether the information gathered from the studies in cystic fibrosis can be extrapolated to non-cystic fibrosis patients.

In this issue of Thorax Caballeroet al 18 have used the Western blot technique to detect IgG antibodies to outer membrane antigens ofP aeruginosa in patients with bronchiectasis who were classified as having no P aeruginosa, alternating P aeruginosaand other bacteria, or continuously colonised withP aeruginosa. They found that the antibody response distinguished between these groups of patients at the level of approximately 75% sensitivity and specificity. Their results are therefore a valuable extension of the previously published results in non-cystic fibrosis patients but await further clinical studies as outlined above.