Key messages

• Prophylactic antibiotics probably reduce the risk of exacerbations, but they do not decrease hospitalizations or mortality in patients with chronic obstructive pulmonary disease.
• The risk of adverse effects of prophylactic antibiotics in patients with chronic obstructive pulmonary disease is not clear because the certainty of the evidence is very low.

What is the evidence.
See evidence matrix  in Epistemonikos later

We found five systematic reviews [1],[2],[3],[4],[5], considering 10 primary studies [6],[7],[8],[9],[10],[11],[12],[13],[14],[15], including eight randomized controlled trials [6],[7],[8],[9],[10],[12],[13],[14].

This table and the summary in general are based on the latter. 

What types of patients were included

All studies included patients with FEV1/FVC <70% with moderate to very severe disease.

Three studies included patients with a history of exacerbations in the previous months [6],[10],[13].

What types of interventions were included

The antibiotic used was a macrolide in seven studies; azithromycin [6],[8],[10], clarithromycin [7] and erythromycin [9],[12],[14]; and a quinolone in only one study (moxifloxacin) [13].

Prophylactic antibiotics were administered continuously in five studies [6],[7],[9],[12],[14] and intermittently or pulsed in the other three [8],[10],[13].

Concomitant therapy was the following: inhaled corticosteroids alone in two studies [7],[12], theophylline plus inhaled anticholinergics in one study [14], inhaled corticosteroids plus inhaled anticholinergics and beta2 agonists in one study [6], inhaled corticosteroids plus inhaled beta2 agonists, inhaled anticholinergics and theophylline in one study [9], and concomitant therapy was not reported in three studies [8],[10],[13].

Seven studies used a placebo for comparison [6],[7],[9],[12],[13],[14].

What types of outcomes
were measured

Different systematic reviews performed meta-analysis of the following outcomes:

• exacerbations
• mortality
• hospitalizations
• quality of life
• time to first exacerbation
• adverse effects

To whom this evidence does and does not apply

• This evidence applies to patients with moderate to severe chronic obstructive pulmonary disease. It is reasonable to extrapolate to patients at an earlier stage, for whom the expected benefit should be lower or non-existent.

• The outcomes selected for the summary of findings are those considered critical for decision-making by the authors of this article.

• While there is a benefit on exacerbations, this does not translate into changes in hospitalizations or mortality, at least in the average patient.
• Moreover, there is uncertainty about adverse effects, based on studies summarized here, but it is reasonable to expect these would arise. The associated costs should also be considered, and the effects on antibiotic resistance at the population level.
• However, it is possible that in patients with frequent exacerbations, the benefit might be greater than risks and costs of this intervention.

• It is likely that most patients will prefer not to take antibiotics regularly in order to avoid having to occasionally take them in case of an exacerbation that would not impact on hospitalizations or mortality.
• Clinicians would probably incorporate the population factor of antibiotic resistance to the decision, so they might not be inclined to support this intervention.
• Some patients or their clinicians might be in favor of this intervention in case of frequent exacerbations, a history of severe exacerbations, or other factors that can make them put high value on the small expected benefit.

• The cost of the intervention is probably not a determining factor in the clinical decision.
• It is not possible to adequately estimate the cost/benefit of this intervention due to the uncertainty about the costs of adverse effects, at both individual and population level. 

Differences between this summary and other sources

• Our summary is consistent with the systematic reviews identified.
• Our summary is in agreement with the main clinical guidelines, which differ slightly between them. The GOLD guideline states there is no role for prophylactic antibiotics, while the joint guideline of the American College of Chest Physicians and the Canadian Thoracic Society suggests their use in patients with moderate to severe disease who have a history of exacerbations in the past year despite optimal therapy [16],[17].

• The probability that future evidence change what we know about the benefits of prophylactic antibiotics in chronic obstructive pulmonary disease is very low, because of the certainty of the evidence.
• There is a high probability that future evidence give us better information about adverse effects, such as the ongoing Cochrane review that will synthesize the adverse effects of macrolides [18].