Antibiotics for Exacerbations of COPD
Does routine use of antibiotics in patients with an exacerbation of chronic obstructive pulmonary disease (COPD) improve clinical outcomes?
There is consistent evidence from a small number of studies that antibiotics significantly reduce mortality (number needed to treat [NNT] = 8) and treatment failure (NNT = 3) in more severely ill patients with COPD who are hospitalized because of an exacerbation. Two smaller studies found no benefit of antibiotics for patients with an exacerbation of COPD who were treated in the community setting.
Many acute exacerbations of COPD are caused by bacterial infection, but many are caused by viruses or are noninfectious. Although some studies have found the routine use of antibiotics beneficial in patients with COPD exacerbations, others have failed to find a benefit. Ram and colleagues identified 11 studies of patients with COPD who experienced an exacerbation and who were randomly assigned to receive an antibiotic or placebo. An exacerbation was defined as one or more of the following: increased sputum purulence or volume, increased shortness of breath, increased wheeze, chest tightness, or fluid retention. Only two of the studies included outpatients. Study quality was high.
Three studies involving a total of 298 hospital inpatients and using an objective definition of COPD found mortality rates of 3 percent in the antibiotic group and 15 percent in the placebo group (relative risk [RR] = 0.24; 95% confidence interval [CI], 0.1 to 0.57; NNT = 8). Results between studies were consistent and remained so when a fourth study conducted in an intensive care unit was added.
Another important outcome is treatment failure, generally defined as failure to improve, deterioration, or death during the study period. Six studies with a total of 705 patients reported this outcome, but there was significant variation (heterogeneity) between studies, with some showing improvement in patients taking antibiotics and others showing no benefit. After exploring the data for patterns, the authors found that antibiotics consistently reduced the likelihood of treatment failure in four studies with a total of 321 hospitalized patients (28 percent with treatment versus 58 percent with placebo [RR = 0.47; 95% CI, 0.36 to 0.62; NNT = 3]). However, the same was not true for the two outpatient studies with a total of 384 patients. In these studies, antibiotics did not reduce the likelihood of treatment failure (40 percent with antibiotics versus 35 percent with placebo [RR = 1.14; 95% CI, 0.88 to 1.48]). The antibiotics used in the studies were amoxicillin/clavulanate (Augmentin), trimethoprim/sulfamethoxazole (Bactrim, Septra), and amoxicillin.