Are Antibiotics Effective Treatment for Acute Bronchitis? A Meta-Analysis

Journal of Family Practice, Dec, 1998 by John J. Smucny, Lorne A. Becker, Richard H. Glazier, Warren McIsaac

A Meta-Analysis

BACKGROUND. Most clinicians prescribe antibiotics for acute bronchitis in spite of recommendations against this practice. Because the results of individual clinical trials have been mixed, we conducted a meta-analysis to determine whether antibiotics are effective treatment for acute bronchitis.

METHODS. We conducted a comprehensive search to identify all trials in which patients who had a diagnosis of acute bronchitis were randomly assigned to treatment with an antibiotic or placebo. Patient-oriented outcomes of importance that were reported in at least 3 studies were quantitatively summarized.

RESULTS. Nine studies met the inclusion criteria, but not all trials provided data for each outcome. Patients given antibiotics were less likely to have a cough (relative risk [RR] = 0.69; 95% confidence interval [CI], 0.49 - 0.98) and be considered unimproved (RR = 0.51; 95% CI, 0.30 - 0.88) at a follow-up visit; but they were not less likely to have a productive cough (RR = 0.79; 95% CI, 0.60 - 1.03), activity limitations (RR = 0.59; 95% CI, 0.24 - 1.44), or feel ill (RR = 0.70; 95% CI, 0.31 - 1.58). Antibiotic-treated patients had a slightly shorter duration of productive cough (weighted mean difference [WMD] = -0.56 days; 95% CI, -1.09 to -0.04), but not of overall cough (WMD = -0.94; 95% CI, -2.08 to 0.21) or activity limitations (WMD = -0.49; 95% CI, -1.07 to 0.10). Patients treated with antibiotics did not report significantly more adverse effects (RR = 1.47; 95% CI, 0.82 - 2.65).

CONCLUSIONS. Antibiotics may be modestly effective for a minority of patients with acute bronchitis. It is not clear which patient subgroups might benefit, and the failure of some studies to report negative findings may have resulted in overestimates of the benefits of antibiotics. Antibiotics are not necessary for every patient with acute bronchitis.

KEY WORDS. Acute bronchitis; antibiotics; treatment; meta-analysis. (J Fam Pract 1998; 47:453-460)

Antibiotic treatment for acute bronchitis is controversial. Recent studies show the majority of patients who visit a physician for acute bronchitis (65% to 80%) are prescribed antibiotics.[1-3] Experts, however, generally state that antibiotics are not indicated for acute bronchitis in patients who do not have chronic pulmonary disease,[4,5] and recent editorials have exhorted physicians to stop prescribing antibiotics for this condition.[6-7] These recommendations reflect a view that acute bronchitis is a self-limited illness caused mainly by viral pathogens. For the individual patient, unwarranted antibiotic treatment leads to unnecessary costs, as well as risks of adverse effects and future infection with resistant bacteria.[8,9] The increasing prevalence of resistant bacteria is thought to be related to the over-prescription of antibiotics, and national efforts are under way to decrease the inappropriate use of antibiotics.[10]

Clinical trials of the effectiveness of antibiotics in treating acute bronchitis have had mixed results. Two descriptive reviews of these trials concluded that antibiotic therapy is generally not supported by the literature because a majority of the trials did not demonstrate antibiotics to be effective.[11,12] Neither review combined data from published studies into a quantitative summary or meta-analysis. Both noted that some trials did demonstrate benefits from antibiotic treatment and therefore suggested that further research was needed. Since these reviews were published, additional trials have indeed been reported.

The studies that have been published thus far have had rather small sample sizes, which can lead to type II errors. A major rationale for conducting a meta-analysis is that the power of the analysis is greater than that of smaller individual studies. Furthermore, as long as individual studies have similar design and methodological quality, a meta-analysis can provide a more accurate measure of the overall effectiveness of an intervention than qualitative reviews that tally how many studies either show or do not show a benefit.[13] A meta-analysis also provides a quantitative summary estimate of treatment effect. A recently published meta-analysis of the effectiveness of antibiotics in patients who present with an acute cough did not find significant differences in the risks of productive cough or overall unimprovement after 7 to 11 days of therapy, or any differences in the risks of adverse effects.[14] This study did not examine the overall duration of symptoms or other potentially important outcomes.

We conducted this quantitative overview of all randomized controlled trials of antibiotic treatment for acute bronchitis to answer the question: Are antibiotics effective treatment for acute bronchitis? We specifically wanted to determine whether antibiotics: (1) lead to improved outcomes of clinical importance in otherwise healthy patients with acute bronchitis and, if so, what the magnitude of the treatment effects are; and (2) lead to increased risk of adverse effects that may outweigh any benefits in symptomatic relief.