Cochrane Clinical Answers - Fulltext - Is there randomized controlled trial evidence to support the use of prolonged antibiotics in people with non-cystic fibrosis bronchiectasis?

Question:Is there randomized controlled trial evidence to support the use of prolonged antibiotics in people with non-cystic fibrosis bronchiectasis?

Clinical Answer:

Continuous antibiotics appeared to reduce exacerbations and improve treatment response during exacerbations, at the expense of increased antibiotic resistance and possibly increased adverse events. Judicious and selective use of continuous antibiotics in patients with non-cystic fibrosis bronchiectasis is warranted.

There was moderate-quality evidence that continuous antibiotics reduced exacerbations, with 271 per 1000 people (95% CI 126 to 385) with antibiotics compared with 546 per 1000 people with placebo having experienced exacerbations. However, rates of hospitalization were similar between groups.

Two RCTs with 110 participants showed that clinical response rates appeared higher in patients on continuous antibiotics, with on average 153 per 1000 people taking antibiotics having failed to respond clinically during exacerbations compared with 378 per 1000 people taking placebo.

However, more patients on continuous antibiotics seemed to have experienced adverse events, including diarrhea, dyspnea, chest pain/palpitation and abdominal pain. The estimates were mostly underpowered, which precluded any clear conclusion on these outcomes.

Moderate-quality evidence showed that more patients on continuous antibiotics developed antibiotic resistance (148 per 1000 people (95% CI 56 to 334) on continuous antibiotics compared with 47 per 1000 on placebo) and low-quality evidence suggested that patients in both groups experienced similar quality of life.

Full outcome data is detailed below:

1.Continuous antibiotics versus placebo

  • Population, Intervention, Comparator

    Population:

    Participants (mostly adults with mean age of 59 years) with 148 children under 18 years of age (12.8%) diagnosed with non-cystic fibrosis bronchiectasis confirmed by computed tomography scan in all (14 trials) or some (1 trial; others clinical diagnosis only) patients or bronchogram/chest radiography (1 trial). Baseline lung function was reported variably across studies (studies that reported it showed moderate and severe disease in adults and mild disease in children)

    Intervention:

    Antibiotics (and route of administration) assessed in the studies were tobramycin (2 trials; nebulized), ceftazidime (1 trial; nebulized), amoxycillin (1 trial; oral), roxithromycin (2 trials; oral), penicillin (1 trial; oral), oxytetracycline (1 trial; oral), erythromycin (3 trials; oral), gentamicin (1 trial; nebulized) and azithromycin (4 trials; oral). Duration of treatment ranged from 4 to 83 weeks

    Comparator:

    Placebo (13 trials) or usual medical care (details not reported; 2 trials)

  • OUTCOME 1.1: Exacerbations (follow-up 1.5 to 24 months)

    Quality of the evidence:

    The reviewers performed a GRADE assessment of the quality of evidence for this outcome at this time point and stated that the evidence was moderate quality. See Summary of findings from Cochrane review

    Narrative result:

    13 RCTs with 654 participants found that fewer people receiving prolonged antibiotics had exacerbations compared with placebo.

    Relative effect or mean difference:

    There was a statistically significant difference between groups, in favor of antibiotics (OR 0.31, 95% CI 0.19 to 0.52).

    Forest plot from Cochrane Review

    Absolute effect:

    271 per 1000 people (95% CI 126 to 385) with antibiotics compared with 546 per 1000 people with placebo.

    Reference:
    Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
    • OUTCOME 1.2: Hospitalizations (follow-up 1.5 to 24 months)

      Quality of the evidence:

      The reviewers performed a GRADE assessment of the quality of evidence for this outcome at this time point and stated that the evidence was moderate quality. See Summary of findings from Cochrane review

      Narrative result:

      Seven RCTs with 643 participants found no statistically significant difference between groups, but given the low event rate, the analysis would have been too underpowered to detect a clinically meaningful difference if present.

      Relative effect or mean difference:

      There was no statistically significant difference between groups (OR 0.40, 95% CI 0.14 to 1.11).

      Forest plot from Cochrane Review

      Absolute effect:

      37 per 1000 people (95% CI 13 to 96) with antibiotics compared with 87 per 1000 people with placebo.

      Reference:
      Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
      • OUTCOME 1.3: Response rates (follow-up 6 weeks to 12 months)

        Risk of bias of studies:

        The reviewers did not perform a GRADE assessment of the quality of the evidence. Both studies failed to report adequate allocation concealment or random sequence generation, and one had unclear numbers of withdrawals; both blinded outcome assessors.

        Narrative result:

        Two RCTs with 110 participants found that fewer people failed to show a response (using physician assessment of diary cards or overall medical condition) with antibiotics compared with placebo.

        Relative effect or mean difference:

        There was a statistically significant difference between groups, in favor of placebo (Peto OR 0.30, 95% CI 0.14 to 0.63).

        Forest plot from Cochrane Review

        Absolute effect:

        153 per 1000 people (95% CI 79 to 276) failed to respond with antibiotics compared with 378 per 1000 people with placebo.

        Reference:
        Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
        • OUTCOME 1.4: Adverse events (follow-up 1.5 months to 12 months)

          Narrative result:

          Withdrawals due to intolerable adverse events, and the number of people with rash, increased cough, nausea, fatigue, hemoptysis, fever, increased sputum, headache, bronchospasm, vomiting and acute sinusitis were similar in both groups, although some of the analyses would have been too underpowered to detect clinically meaningful differences if present. The number of people experiencing diarrhea, dyspnea, chest pain/palpitations and abdominal pain were higher with antibiotics. Click below for full details.

          Reference:
          Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
          • Subgroup analysis 1.4.1: Adverse events - [subgroup: Diarrhea]
            Risk of bias of studies:

            The reviewers did not perform a GRADE assessment of the quality of the evidence. In this analysis 1/3 (33%) of the studies failed to report adequate allocation concealment, all reported adequate blinding of participants/carers/outcome assessors and 2/3 (66%) had unclear numbers of withdrawals.

            Narrative result:

            Three RCTs with 231 participants found that more people receiving antibiotics had diarrhea compared with placebo.

            Relative effect or mean difference:

            There was a statistically significant difference between groups, in favor of placebo (Peto OR 3.33, 95% CI 1.50 to 7.37).

            Forest plot from Cochrane Review

            Absolute effect:

            164 per 1000 people (95% CI 81 to 303) with antibiotics compared with 56 per 1000 people with placebo.

            Reference:
            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
          • Subgroup analysis 1.4.2: Adverse events - [subgroup: Dyspnea]
            Risk of bias of studies:

            The reviewers did not perform a GRADE assessment of the quality of the evidence. The study failed to report adequate allocation concealment and random sequence generation processes, reported adequate blinding of participants/carers/outcome assessors and had unclear numbers of withdrawals.

            Narrative result:

            One RCT with 74 participants found that more people receiving antibiotics had dyspnea compared with placebo.

            Relative effect or mean difference:

            There was a statistically significant difference between groups, in favor of placebo (Peto OR 4.41, 95% CI 1.43 to 13.61).

            Forest plot from Cochrane Review

            Absolute effect:

            280 per 1000 people (95% CI 112 to 546) with antibiotics compared with 81 per 1000 people with placebo.

            Reference:
            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
          • Subgroup analysis 1.4.3: Adverse events - [subgroup: Chest pain/palpitations]
            Risk of bias of studies:

            The reviewers did not perform a GRADE assessment of the quality of the evidence. In this analysis 2/3 (66%) of the studies failed to report adequate allocation concealment and had unclear numbers of withdrawals; all reported adequate blinding of participants/carers/outcome assessors.

            Narrative result:

            Three RCTs with 274 participants found that more people receiving antibiotics had chest pain/palpitations compared with placebo.

            Relative effect or mean difference:

            There was a statistically significant difference between groups, in favor of placebo (Peto OR 5.21, 95% CI 1.35 to 20.14).

            Forest plot from Cochrane Review

            Absolute effect:

            We could not calculate absolute effects because of low event rates.

            Reference:
            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
          • Subgroup analysis 1.4.4: Adverse events - [subgroup: Abdominal pain]
            Risk of bias of studies:

            The reviewers did not perform a GRADE assessment of the quality of the evidence. In this analysis both studies reported adequate allocation concealment and blinding of participants/carers/outcome assessors, but had unclear numbers of withdrawals.

            Narrative result:

            Two RCTs with 195 participants found that more people receiving antibiotics had abdominal pain compared with placebo.

            Relative effect or mean difference:

            There was a statistically significant difference between groups, in favor of placebo (Peto OR 5.10, 95% CI 1.38 to 18.86).

            Forest plot from Cochrane Review

            Absolute effect:

            We could not calculate absolute effects because of low event rates.

            Reference:
            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
        • OUTCOME 1.5: Antibiotic resistance (follow-up 1.5 to 24 months)

          Quality of the evidence:

          The reviewers performed a GRADE assessment of the quality of evidence for this outcome at this time point and stated that the evidence was moderate quality. See Summary of findings from Cochrane review

          Narrative result:

          Six RCTs with 431 participants found that more people taking antibiotics had antibiotic resistance strains detected compared with placebo. One RCT not included in the meta-analysis detected no resistant strains in either group at the end of 12 months follow-up.

          Relative effect or mean difference:

          There was a statistically significant difference between groups, in favor of placebo (OR 3.48, 95% CI 1.20 to 10.07).

          Forest plot from Cochrane Review

          Absolute effect:

          148 per 1000 people (95% CI 56 to 334) with antibiotics compared with 47 per 1000 people with placebo.

          Reference:
          Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]
          • OUTCOME 1.6: Quality of life (follow-up 6 to 16 months)

            Quality of the evidence:

            The reviewers performed a GRADE assessment of the quality of evidence for this outcome at this time point and stated that the evidence was low quality. See Summary of findings from Cochrane review

            Narrative result:

            Five RCTs with 315 participants found no statistically significant difference between groups, but the analysis would have been too underpowered to detect a clinically meaningful difference if present. This outcome was assessed with the St George Respiratory Questionnaire (scale from 0 to 100; lower is better).

            Relative effect or mean difference:

            There was no statistically significant difference between groups (mean difference -2.75, 95% CI -7.08 to 1.57).

            Forest plot from Cochrane Review

            Reference:
            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]

            Additional Information:

            DOI

            10.1002/cca.950

            Publication Dates

            1. Published Online: 30 OCT 2015

            CCA derived from

            Hnin K, Nguyen C, Carson KV, Evans DJ, Greenstone M, Smith BJ. Prolonged antibiotics for non-cystic fibrosis bronchiectasis in children and adults. Cochrane Database of Systematic Reviews 2015, Issue 8. Art. No.: CD001392. DOI: 10.1002/14651858.CD001392.pub3. [Review search date: February 2014]

            How to Cite

            Is there randomized controlled trial evidence to support the use of prolonged antibiotics in people with non-cystic fibrosis bronchiectasis? Nai Ming Lai (MBBS, MRCP, MRCPCH) (on behalf of Cochrane Clinical Answers Editors). Cochrane Clinical Answers 2012. DOI: 10.1002/cca.950.

            Further Information

            • CCA Associate editor: Nai Ming Lai (MBBS, MRCP, MRCPCH), Associate Professor in Paediatrics, Taylor's University, Selangor, Malaysia.
            • CCA Editor: Karen Pettersen. Correspondence to kpettersen@wiley.com.