Data also suggest that antibiotic treatment during exacerbations might favorably impact subsequent exacerbations.10 A retrospective study of 18,928 Dutch patients with AECOPD compared outcomes among patients who had received antibiotics (most frequently doxycycline or a penicillin) as part of their therapy to those who did not. The authors demonstrated that the median time to the next exacerbation was significantly longer in the patients receiving antibiotics.10 Further, both mortality and overall risk of developing a subsequent exacerbation were significantly decreased in the antibiotic group, with median follow-up of approximately two years.
Indications for Antibiotics
Clinical symptoms. A landmark study by Anthonisen and colleagues set forth three clinical criteria that have formed the basis for treating AECOPD with antibiotics in subsequent studies and in clinical practice.11 Often referred to as the “cardinal symptoms” of AECOPD, these include increased dyspnea, sputum volume, and sputum purulence. In this study, 173 outpatients with COPD were randomized to a 10-day course of antibiotics or placebo at onset of an exacerbation and followed clinically. The authors found that antibiotic-treated patients were significantly more likely than the placebo group to achieve treatment success, defined as resolution of all exacerbated symptoms within 21 days (68.1% vs. 55.0%, P<0.01).
Importantly, treated patients were also significantly less likely to experience clinical deterioration after 72 hours (9.9% vs. 18.9%, P<0.05). Patients with Type I exacerbations, characterized by all three cardinal symptoms, were most likely to benefit from antibiotic therapy, followed by patients with Type II exacerbations, in whom only two of the symptoms were present. Subsequent studies have suggested that sputum purulence correlates well with the presence of acute bacterial infection and therefore may be a reliable clinical indicator of patients who are likely to benefit from antibiotic therapy.12
Laboratory data. While sputum purulence is associated with bacterial infection, sputum culture is less reliable, as pathogenic bacteria are commonly isolated from patients with both AECOPD and stable COPD. In fact, the prevalence of bacterial colonization in moderate to severe COPD might be as high as 50%.13 Therefore, a positive bacterial sputum culture, in the absence of purulence or other signs of infection, is not recommended as the sole basis for which to prescribe antibiotics.
Serum biomarkers, most notably C-reactive protein (CRP) and procalcitonin, have been studied as a newer approach to identify patients who might benefit from antibiotic therapy for AECOPD. Studies have demonstrated increased CRP levels during AECOPD, particularly in patients with purulent sputum and positive bacterial sputum cultures.12 Procalcitonin is preferentially elevated in bacterial infections.
One randomized, placebo-controlled trial in hospitalized patients with AECOPD demonstrated a significant reduction in antibiotic usage based on low procalcitonin levels, without negatively impacting clinical success rate, hospital mortality, subsequent antibiotic needs, or time to next exacerbation.14 However, due to inconsistent evidence, use of these markers to guide antibiotic administration in AECOPD has not yet been definitively established.14,15 Additionally, these laboratory results are often not available at the point of care, potentially limiting their utility in the decision to initiate antibiotics.
Severity of illness. Severity of illness is an important factor in the decision to treat AECOPD with antibiotics. Patients with advanced, underlying airway obstruction, as measured by FEV1, are more likely to have a bacterial cause of AECOPD.16 Additionally, baseline clinical characteristics including advanced age and comorbid conditions, particularly cardiovascular disease and diabetes, increase the risk of severe exacerbations.17
One meta-analysis of placebo-controlled trials found that patients with severe exacerbations were likely to benefit from antibiotic therapy, while patients with mild or moderate exacerbations had no reduction in treatment failure or mortality rates.18 Patients presenting with acute respiratory failure necessitating intensive care and/or ventilator support (noninvasive or invasive) have also been shown to benefit from antibiotics.19