Evidence review: efficacy

No published randomised controlled trials (RCTs) have investigated nebulised or inhaled colistimethate sodium for treating non-cystic fibrosis bronchiectasis.

Four case series were identified that reported using nebulised colistimethate sodium in people with non-cystic fibrosis bronchiectasis and bronchial colonisation with Pseudomonas aeruginosa. Three of the case series (White et al. 2012, n=30; Dhar et al. 2010, n=19; and Steinfort and Steinfort 2007, n=18) used historical self-controls to compare outcomes before and after treatment with nebulised colistimethate sodium. The fourth case series (Berlana et al. 2011, n=81) compared post-treatment outcomes in people who had received nebulised colistimethate sodium alone, tobramycin alone or both.

White et al. (2012)

To assess the outcomes of treatment, White et al. (2012) retrospectively identified all people (n=30) with non-cystic fibrosis bronchiectasis who had received 'Pseudomonas eradication therapy' at the Royal United Hospital, Bath, between January 2004 and January 2010. The mean age of the patients was 62.2 years. The cause of bronchiectasis was earlier infection in 13 people, connective tissue disease in 3 people, and unknown in 14 people. Associated conditions included chronic obstructive pulmonary disease (COPD, 6 people), asthma (5 people) and allergic bronchopulmonary aspergillosis (3 people).

The decision to start treatment usually followed 2 sputum cultures that were positive for P. aeruginosa (range 1 to 9, median time from first culture to treatment 7.2 months). The Pseudomonas eradication protocol was:

  • intravenous regimen: gentamicin (4 mg/kg) plus ceftazidime (2 g 3 times daily) for 2 weeks, followed by nebulised colistimethate sodium (2 million units twice daily) for 3 months, with or without oral ciprofloxacin (500 mg twice daily) for 3 months.

  • oral regimen: ciprofloxacin (500 mg twice daily) for 3 months plus nebulised colistimethate sodium (2 million units twice daily) for 3 months.

Twelve patients received intravenous antibiotics, 5 received oral ciprofloxacin, and 13 received intravenous antibiotics followed by oral ciprofloxacin. Intravenous antibiotics other than those in the protocol were used at the clinician's discretion in 8 people, mainly because of drug allergy or intolerance. Twenty-five patients also received nebulised colistimethate sodium; 5 people did not receive it because of intolerance (n=3) or because of a clinical decision (n=2).

Eradication of Pseudomonas was defined as a negative culture from the next sputum sample, and recurrence was defined as the first positive culture after treatment. Other outcomes included the number of exacerbations (defined as episodes of increased symptoms of bronchiectasis needing oral or intravenous antibiotics) and hospital admissions for intravenous antibiotics, and lung function (forced expiratory volume in 1 second [FEV1]). Median follow-up was 26.4 months. The results were as follows:

  • Pseudomonas eradication. Pseudomonas was initially eradicated from the sputum of 24 out of 30 patients (80%). Of these, 13 (54%) remained Pseudomonas-free at their latest follow-up (median 14.3 months), and 11 (46%) experienced a recurrence of Pseudomonas (median time to re-infection 6.2 months).

  • Exacerbations. Treatment statistically significantly reduced the mean number of exacerbations needing antibiotics from 3.93 per year before treatment to 2.09 after treatment (p<0.002).

  • Hospital admissions. Treatment did not statistically significantly reduce the mean number of admissions (0.39 per year before treatment and 0.29 per year after treatment, no significant difference).

  • Lung function.Treatment also had no effect on lung function, with mean percentage predicted FEV1 62.1% before treatment and 64.1% after treatment (no significant difference).

Dhar et al. (2010)

Dhar et al. (2010) retrospectively assessed non-cystic fibrosis bronchiectasis and colonisation with P. aeruginosa in people (n=19) who had received a minimum of 6 months of treatment with nebulised colistimethate sodium (1−2 million units twice daily) at the Northumbria NHS Trust. People who had received concomitant macrolide antibiotic treatment for more than 4 weeks were excluded. The mean age of the patients was 66 years, 15 had known causes for their bronchiectasis, although these were not reported.

Outcomes included the number of hospital admissions, exacerbations that needed rescue antibiotics and had positive sputum cultures for P. aeruginosa, and lung function (FEV1). The mean length of data collection before starting colistimethate sodium treatment was 23.6 months and the mean duration of treatment was 21.2 months (range 6–39 months). The results were as follows:

  • Pseudomonas sputum positivity. Colistimethate sodium statistically significantly reduced the mean number of positive samples from 4.2 per year to 0.5 per year (p<0.001).

  • Exacerbations. Colistimethate sodium statistically significantly reduced the mean number of exacerbations from 7.8 per year to 2.7 per year (p<0.001).

  • Hospital admissions. Colistimethate sodium statistically significantly reduced the mean number of admissions from 3.0 per year to 0.95 per year (p=0.002).

  • Lung function.Colistimethate sodium had no effect on lung function: mean FEV1 was 1.13 litres before treatment and 1.14 litres after treatment (no significant difference reported).

Steinfort and Steinfort (2007)

Steinfort and Steinfort (2007) was an Australian case series that prospectively examined the effects of long-term nebulised colistimethate sodium (30 mg daily in 2 ml of saline or salbutamol solution, estimated to equate to around 375,000 units) in 18 people with either confirmed bronchiectasis or severe COPD (FEV1 less than 40% of predicted). All patients had repeated sputum cultures that were positive for multidrug-resistant gram negative bacteria, frequent exacerbations or exacerbations needing hospital admission, poor symptom control, and production of a high volume of sputum between exacerbations. Of the 18 patients (mean age 69 years), 14 had bronchiectasis of unknown cause and 4 had severe COPD or chronic infective bronchitis; 14 had P. aeruginosa isolated from sputum.

Outcomes included measures of lung function, hospital admissions and quality of life (measured on a non-validated visual analogue scale). Lung function was measured on a mean of 9.4 occasions over a period ranging from 2 to 11 years before starting colistimethate sodium treatment. Once treatment started, patients were assessed regularly about every 6 months, over a period of 6–116 months. The results were as follows:

  • Lung function. Three people (16.6%) were reported to show improvement in FEV1. Overall, nebulised colistimethate sodium statistically significantly reduced the mean annual decline in FEV1 (from 104 ml per year before treatment to 44 ml per year after treatment, p=0.035) and forced vital capacity (FVC: from 110 ml per year before treatment to 48 ml per year after treatment, p=0.033). However, it had no significant effect on FEV1 (1070 ml before treatment and 1020 ml after treatment, p=0.400) or FVC (2.0 litres before treatment and 1.9 litres after treatment, p=0.295).

  • Hospital admissions. Colistimethate sodium treatment did not significantly reduce the number of hospital admissions (1.1 per year before treatment and 0.84 per year after treatment, p=0.493).

  • Quality of life. Treatment statistically significantly improved patient-reported quality of life (from 3.6 before treatment to 6.2 after treatment, p=0.001).

Berlana et al. (2011)

Berlana et al. (2011) was a Spanish case series that prospectively compared outcomes in 81 outpatients with non-cystic fibrosis bronchiectasis or COPD who received long-term (at least 12 weeks) nebulised treatment for P. aeruginosa bronchial colonisation between January 2004 and December 2008. Bronchial colonisation was confirmed by 3 positive sputum samples within a 6 month period. Patients were required to be symptom-free at entry.

The 81 patients received 97 courses of antibiotic treatment: 31 received nebulised colistimethate sodium (1 to 2 million units twice daily), 16 received colistimethate sodium plus tobramycin (an aminoglycoside antibiotic, which is also licensed for treating P. aeruginosa colonisation in people with cystic fibrosis), and 50 received tobramycin alone. The choice of treatment depended on clinical judgement. Supplementary use of oral antibiotics was permitted.

The mean age of the patients was 62.5 years. Eighty-six per cent of the courses of inhaled antibiotics (n=83) were primarily for bronchiectasis and the remainder were for COPD. There was no significant difference between the baseline characteristics of those receiving colistimethate sodium or tobramycin plus colistimethate sodium, compared with those receiving tobramycin alone.

Patients were reviewed every 6 months. Mean follow-up was 578 days. The primary outcomes were frequency and duration of hospitalisations for exacerbations. Secondary outcomes included antibiotic use during admission, changes in respiratory function, achievement of sustained P. aeruginosa eradication, mortality, emergence of bacterial resistance and adverse events.

There were no significant differences between colistimethate sodium alone and tobramycin alone in the mean number of hospital admissions, duration of hospital stay or duration of antibiotic use.

Compared with tobramycin alone, the combination of colistimethate sodium and tobramycin was associated with a significant reduction in hospital admissions (0.4 compared with 1.9 per person per year, p<0.05), duration of hospital stay (2.8 days compared with 22.5 days per person per year, p<0.05), and duration of antibiotic use (2.0 days compared with 14.8 days per person per year, p<0.05).

Reliable lung function measures were available for only 32% of the courses of treatment (31/97). No significant differences were found in the mean change per year in lung function tests between the treatment groups.

Of 29 courses of inhaled colistimethate sodium during which sputum culture was performed, P. aeruginosa was eradicated in 17% (n=5), colonisation was reduced in 21% (n=6), culture remained negative in 7% (n=2) and there was no response in 55% (n=16). There was no significant difference in eradication rates between the treatment groups.

Unpublished randomised controlled trial

An RCT, Inhaled Promixin in the treatment of non-cystic fibrosis bronchiectasis, has been completed but not yet published. This double-blind, multicentre RCT included 144 people with non-cystic fibrosis bronchiectasis who had had at least 2 sputum samples positive for P. aeruginosa within the previous 12 months and had completed treatment for an exacerbation of bronchiectasis within 21 days of screening. They were randomised to receive nebulised colistimethate sodium (1 million units per ml) or saline twice daily for up to 6 months. The primary outcome was the number of days from baseline (or first treatment dose) until the first exacerbation. Secondary outcomes included change in sputum mass and culture, change in FEV1, quality of life, adherence and adverse events. Without a fully published version of this study, it is not possible to critically appraise it to assess how it informs practice in this area.