Elsevier

The Lancet

Volume 374, Issue 9691, 29 August–4 September 2009, Pages 712-719
The Lancet

Articles
Budesonide and the risk of pneumonia: a meta-analysis of individual patient data

https://doi.org/10.1016/S0140-6736(09)61250-2Get rights and content

Summary

Background

Concern is continuing about increased risk of pneumonia in patients with chronic obstructive pulmonary disease (COPD) who use inhaled corticosteroids. We aimed to establish the effects of inhaled budesonide on the risk of pneumonia in such patients.

Methods

We pooled patient data from seven large clinical trials of inhaled budesonide (320–1280 μg/day), with or without formoterol, versus control regimen (placebo or formoterol alone) in patients with stable COPD and at least 6 months of follow-up. The primary analysis compared treatment groups for the risk of pneumonia as an adverse event or serious adverse event during the trial or within 15 days of the trial end. Cox proportional hazards regression was used to analyse the data on an intention-to-treat basis. Data were adjusted for patients' age, sex, smoking status, body-mass index, and postbronchodilator percent of predicted forced expiratory volume in 1 s (FEV1).

Findings

We analysed data from 7042 patients, of whom 3801 were on inhaled budesonide and 3241 were on control treatment, with 5212 patient-years of exposure to treatment. We recorded no significant difference between treatment groups for the occurrence of pneumonia as an adverse event (3% [n=122 patients] vs 3% [n=103]; adjusted hazard ratio 1·05, 95% CI 0·81–1·37) or a serious adverse event (1% [n=53] vs 2% [n=50]; 0·92, 0·62–1·35), or for time to pneumonia as an adverse event (log-rank test 0·94) or a serious adverse event (0·61). Increasing age and decreasing percent of predicted FEV1 were the only two variables that were significantly associated with occurrence of pneumonia as an adverse event or a serious adverse event.

Interpretation

Budesonide treatment for 12 months does not increase the risk of pneumonia in patients with COPD during that time and therefore is safe for clinical use in such patients.

Funding

Michael Smith Foundation for Health Research.

Introduction

Inhaled corticosteroids, given with and without longacting β2 agonists, reduce the occurrence of disease exacerbation and improve quality of life for patients with chronic obstructive pulmonary disease (COPD), but they have also been associated with increased risk of pneumonia.1 Results from the largest clinical trial so far showed that inhaled corticosteroids increase the risk of pneumonia by nearly 50%.1 In an observational study of a large health database, Ernst and colleagues2 reported a 70% increase in the risk of hospital admission for pneumonia in patients who used inhaled corticosteroids. Findings from a meta-analysis indicated that on average, inhaled corticosteroids raised the risk of pneumonia by 34%;3 however, results differed widely between studies with some trials reporting increased risk,1, 4 whereas others reported reduced risk.5, 6 The excess risk seemed to be mainly restricted to patients who received doses of inhaled corticosteroids exceeding 1000 μg/day beclometasone or equivalent.3

Although these data are striking, several important limitations have affected this meta-analysis and previous meta-analyses.3, 7, 8 First, the trials included in the meta-analyses were heterogeneous in terms of study drug and duration, which will probably have contributed to the large variability in results. Indeed in one meta-analysis, use of a random-effects model to correct for heterogeneity across studies yielded a confidence interval that included the possibility of no difference.7 Second, none of the previous meta-analyses had access to data about patient characteristics and thus could not adequately adjust for potential confounders such as age, lung function, or other clinical features. Third, the previous analyses largely focused on any pneumonia event documented as an adverse event. However, the diagnostic accuracy of this endpoint is uncertain since many events were not validated with established criteria for pneumonia, including an opacity on chest radiographs.7 Moreover, many such events are not associated with significant morbidity or mortality.9 Conversely, hospital admissions for pneumonia are usually better documented with imaging studies and laboratory investigations, and are associated with substantial morbidity and mortality.10, 11 Since all hospital admissions due to pneumonia will be reported as serious adverse events, pneumonia as a serious adverse event might be a more accurate and clinically relevant endpoint.

To address these limitations, we pooled data for patient characteristics and results from seven large clinical trials of inhaled budesonide to establish the effects of treatment on the risk of any adverse and serious adverse events of pneumonia in patients with COPD.

Section snippets

Data sources, search strategy, and study selection

US and AT searched Medline, EmBase, and an internal AstraZeneca product database of published studies, Planet, for randomised controlled trials of inhaled budesonide, with or without the longacting β2 agonist formoterol, compared with a control regimen (placebo or formoterol alone) for patients with COPD. We searched using the seach term “COPD” in combination with “budesonide”, “Pulmicort”, “budesonide and formoterol”, “Symbicort”, “placebo”, “formoterol”, “Oxis”, or “Foradil”. No language or

Results

Table 1 shows the baseline characteristics and follow-up of participants in the seven trials. We analysed data from more than 7000 participants who had more than 5000 patient-years of exposure. Table 2 shows the overall baseline characteristics of participants, which were included in the primary analysis. Mean age of participants was 61·6 years (SD 9·7), and mean postbronchodilator FEV1 was 45·5% (20·2) of predicted. Overall, 22% (n=1523 patients) of the cohort were in the Global initiative for

Discussion

In our study we have shown that budesonide, with or without a longacting β2 agonist, is not associated with increased risk of pneumonia reported as an adverse event or a serious adverse event. Adjustments for potential confounders such as age, baseline lung function, and smoking status did not affect overall results related to both treatment selection and outcome. Overall, the 1-year risk of pneumonia as an adverse event was low in both treatment groups at about 3%, which is consistent with

References (31)

  • JA Wedzicha et al.

    The prevention of chronic obstructive pulmonary disease exacerbations by salmeterol/fluticasone propionate or tiotropium bromide

    Am J Respir Crit Care Med

    (2008)
  • PS Burge et al.

    Randomised, double blind, placebo controlled study of fluticasone propionate in patients with moderate to severe chronic obstructive pulmonary disease: the ISOLDE trial

    BMJ

    (2000)
  • LJ Nannini et al.

    Combined corticosteroid and long-acting beta-agonist in one inhaler versus long-acting beta-agonists for chronic obstructive pulmonary disease

    Cochrane Database Syst Rev

    (2007)
  • S Singh et al.

    Long-term use of inhaled corticosteroids and the risk of pneumonia in chronic obstructive pulmonary disease: a meta-analysis

    Arch Intern Med

    (2009)
  • BG Feagan et al.

    Treatment and outcomes of community-acquired pneumonia at Canadian hospitals

    CMAJ

    (2000)
  • Cited by (172)

    • GETting to know the many causes and faces of COPD

      2022, The Lancet Respiratory Medicine
    • Comparative Effectiveness and Safety of Different Types of Inhaled Long-Acting β<inf>2</inf>-Agonist Plus Inhaled Long-Acting Muscarinic Antagonist vs Inhaled Long-Acting β<inf>2</inf>-Agonist Plus Inhaled Corticosteroid Fixed-Dose Combinations in COPD A Propensity Score-Inverse Probability of Treatment Weighting Cohort Study

      2021, Chest
      Citation Excerpt :

      For example, budesonide was observed to stay in the lungs longer than fluticasone, possibly because of the differences in physicochemical properties between the two ICSs; accordingly, budesonide was suspected to exert an antiinflammatory effect longer than fluticasone.43-45 In addition, our findings regarding SAL/FP having a higher risk of pneumonia than the other two LABA/ICS combinations were supported by previous literature that reported an ICS intraclass difference in the risk of pneumonia44,46 and by studies that revealed an increased risk of pneumonia confined to fluticasone-containing, but not budesonide-containing, LABA/ICS regimens.47-49 Another large cohort study also found that FF/BUD therapy was more effective in reducing exacerbations and was safer in decreasing pneumonia risks than SAL/FP therapy.50

    • Chronic cough in the elderly

      2019, Pulmonary Pharmacology and Therapeutics
    View all citing articles on Scopus
    View full text