Dr. Stey rightly underscores the evidence for the role of the mucolytic agent N-acetylcysteine in the outpatient management of stable COPD and in prevention of acute exacerbations of COPD. However, our review focused on the management of acute exacerbations in the emergency department and in the hospital setting. Preventive measures and management of stable COPD were excluded from the review because they did not pertain to our specific research question. Drs. Niederman, Sethi, and Martinez favor the use of newer, broad-spectrum agents in the management of severe acute exacerbations of COPD. They base their recommendation on their view that such exacerbations are bacterial in origin (1), on the increasing rates of antibiotic resistance, and on studies that were excluded from our review because of methodologic flaws (retrospective design and no stratification for disease severity [2]; open-label design and a control group that included the use of broad-spectrum antibiotics [3]). The reasoning behind our recommendation is as follows. First, it is not clear that all acute exacerbations of COPD have a bacterial cause (4). Second, antibiotic trials in acute exacerbations of COPD have shown no benefit or small benefit for severe exacerbations. These trials are often criticized because they are mostly from the 1960s and 1970s and used "old" narrow-spectrum antibiotics. It is argued that with today's rates of antibiotic resistance, particularly in Streptococcus pneumoniae and Haemophilus influenzae, these trials do not reflect the current practice environment. However, although at the time of those trials the bacteria were generally sensitive to narrow-spectrum agents, only a small benefit was seen in patients with more severe exacerbations (5). We agree that patients with COPD are at higher risk for carrying resistant bacteria, depending on the frequency of hospitalizations and on disease severity, but there is currently no evidence to support the use of broad-spectrum antibiotics over narrow-spectrum ones. Thus, as frequently happens, we have come to a different conclusion when interpreting the evidence. We maintain that the increase in antibiotic-resistant organisms emphasizes the need to restrict the use of antibiotics to agents with proven efficacy in cases where they will make a significant difference in outcome. We assert that there is currently insufficient evidence to justify the use of broad-spectrum agents in acute exacerbations of COPD. Our hope is that these guidelines will decrease the unnecessary use of broad-spectrum antibiotics and stimulate further research comparing older and newer agents.