Thirty years ago Staphylococcus aureus and not Pseudomonas aeruginosawas considered to be the most important lung pathogen in cystic fibrosis.1 Those who believed that P aeruginosa was a pathogen in cystic fibrosis thought that various virulence factors such as exotoxin A, exoenzyme S, elastase, alkaline protease, phospholipase C, lipopolysaccharide and phenazine pigments were responsible for the lung tissue damage by drawing parallels with acute P aeruginosa infections in patients with burns or leukaemia.2 Only acute exacerbations, frequently caused by a virus,3 were therefore treated with antibiotics, although invasive disseminatingP aeruginosa infection including bacteraemia was never found in cystic fibrosis.4 However, a very pronounced antibody response to P aeruginosaantigens, including its virulence factors, was detected in patients with cystic fibrosis and the pathogenesis of the lung tissue damage was subsequently found to be caused by immune complex mediated inflammation dominated by polymorphonuclear leucocytes releasing proteolytic enzymes.5-7 Since the annual mortality of cystic fibrosis patients with chronic P aeruginosa infection in the Danish Cystic Fibrosis Centre increased to nearly 20% in 1974, a comprehensive therapeutic approach was started to try to reduce the inflammation by (1) reducing the antigenic load by treating the patients with intravenous antibiotics regularly for two weeks every three months (maintenance therapy = chronic suppressive therapy), (2) reducing the antibody titres by plasmapheresis, and (3) the use of nebulised corticosteroids. The use of nebulised steroids was not successful at that time, probably because the dose of steroid used was too small8; a recent study with a larger dose was found to be effective in reducing inflammation.9 Attempts to reduce the antibody titres by plasmapheresis were not successful as it was only possible to reduce the titre of anti-pseudomonas antibodies by 50–80% for a period of a few weeks (unpublished results). However, Szaff et al 10 found reduction of the antigenic load with intravenous antibiotics to be successful in 58 patients with cystic fibrosis (2.9 courses/year in 1976–80, approximately every three months) compared with 51 historical controls (one course/year in 1971–75 against acute exacerbations). This study included all patients with chronic P aeruginosa infection defined as an increase in the number of precipitating antibodies to these bacteria.5 A follow up study in 198511 showed an increase from a five year survival of 54% to a 10 year survival of 90% from the onset of chronic P aeruginosa infection and a decrease in the annual mortality from 10–20% to 1–2%. The addition of nebulised colistin, prevention of cross infection in the clinic, and early aggressive treatment of the initialP aeruginosa infection further improved survival.12

In this issue of Thorax Elbornet al 13 report a prospective randomised multicentre study in which they compared elective and symptomatic treatment with intravenous antibiotics of cystic fibrosis patients infected with P aeruginosa. No benefit of the elective approach was found. This is hardly surprising since the difference in the amount of intravenous antibiotic used in the two groups of patients was only 45%, 24%, and 33% in the one, two, and three year periods of the study, whereas the difference in each year during the five year maintenance treatment period reported by Szaff et al 10 in 1983 was 190%. The bacteriological effect obtained by Szaffet al 10 was higher than that achieved in current studies, with 35–36% being free ofP aeruginosa at the end of the treatment period, a few for up to three months.10 This probably reflects the higher bacteriological efficacy of treating the chronic infection early14 and the lower level of resistance 20–25 years ago.15 Furthermore, whereas none of the patients in the study by Szaff et al received nebulised antibiotics, these were given to 40% of the symptomatic patients and 25% of the elective group in the study by Elbornet al, further decreasing the difference between the two arms of the study.13 Another major difference between the two studies is the early treatment approach used by Szaff et al. All new chronically infected cystic fibrosis patients were treated regularly from the onset of the infection during the maintenance treatment period, since onset of infection before puberty was found to be associated with a poor prognosis,10 and the major benefit on the survival of the patients was maintenance of lung function in the younger patients as confirmed by Elborn et al.13

Several reports have shown the benefit on lung function of the treatment of P aeruginosa infection in patients with cystic fibrosis16 ,17 but, although the proteolytic activity in the lungs decreases during treatment, it is still significant between courses.17 The addition of daily nebulised colistin18 to the maintenance regime12 or the use of four weekly cycles of on/off nebulised tobramycin19 has further improved the maintenance of lung function in these patients, but a subsequent analysis of the placebo group in the study by Ramseyet al 19 showed that treatment of exacerbations only did not arrest the progressive decline in lung function in patients with cystic fibrosis.20 An important conclusion of the study by Elborn et al 13 is the suggestion that many patients with advanced disease need 3–4 annual courses of antibiotics for respiratory exacerbations. However, therapeutic findings from other studies indicate that the decline in lung function continues to take place between courses but can be diminished by the intensive use of nebulised antibiotics and steroids.9 ,18 ,20 The major side effects of the intensive use of antibiotics in cystic fibrosis are the development of resistance, allergy to β-lactam antibiotics, possible ototoxicity and renal toxicity caused by aminoglycosides (although this has not yet been a significant problem), cost, and compliance of patients, as also reported by Elbornet al.13 New efficient anti-pseudomonas antibiotics and new treatment strategies are therefore needed for patients with cystic fibrosis.21-24