Chronic obstructive pulmonary disease (COPD) affects 14 million people in the United States and results in 90,000 deaths annually. Exacerbations of COPD typically are defined as an increase in dyspnea, cough and production of purulent sputum. The most common reasons for acute exacerbations of COPD are respiratory tract infections and noncompliance with medical therapy. The role of infection and the effectiveness of antibiotics as causes have been matters of controversy because many patients with stable COPD are chronically colonized with bacteria such as Moraxella catarrhalis, Streptococcus pneumoniae and Haemophilus species. Several studies have proposed guidelines for management based on various risk factors. Adams and colleagues retrospectively reviewed medical records of patients with COPD to identify factors that contribute to relapse within two weeks after an acute exacerbation of COPD.
Medical records of patients discharged from the emergency department with a diagnosis of COPD over an 18-month period were reviewed. Inclusion criteria were documented pulmonary function studies within the past three years that showed a forced expiratory volume (FEV1) of up to 80 percent of predicted and an FEV1 to forced vital capacity (FVC) ratio of up to 75 percent. Patients with pneumonia or those who required hospitalization were excluded from the study. Exacerbations were classified as severe (type I), moderate (type II) or mild (type III), based on specific criteria. Type I episodes were characterized by the presence of all of the following: (1) shortness of breath; (2) increased sputum production; and (3) purulent sputum. Type II and type III episodes were defined as the presence of two or one of the three criteria, respectively. FEV1 was greater than 50 percent in type I episodes, between 35 and 49 percent in type II episodes and less than 35 percent in type III episodes. Demographic data, clinical findings and comorbidities were recorded as well. A relapse was defined as a return visit to the emergency department within 14 days of initial presentation. This interval was chosen based on prior studies indicating that patients were at risk of persistent symptoms for two weeks after an initial exacerbation of COPD.
A total of 1,754 patient visits to the emergency department met the study criteria. After exclusion criteria were applied, 362 patient visits were analyzed; of these, 79 visits were followed by relapse, and 283 were not. Mean patient age was 66 years, 99 percent of the patients were men and all had more than a 40-pack-year history of smoking. A comparison between groups revealed that, on average, 38 percent of the patient visits were type I, 49 percent were type II and 13 percent were type III. The risk of relapse was unrelated to types of comorbidities. However, patients who actively smoked, took long-term steroid therapy or were oxygen dependent had a higher risk of relapse, although the differences between groups were not statistically significant. Analysis of the severity of the exacerbations revealed that 43 percent were type I, 33 percent were type II and 24 percent were type III. Ninety-five percent of patients with type I exacerbations were given antibiotics, compared with only 40 percent of patients with type III exacerbations. Overall, antibiotics were given for 270 patient visits; of these, only 19 percent had a relapse within two weeks of initial presentation. Relapses occurred in 32 percent of patient visits when antibiotics were not given. Treatment with amoxicillin (54 percent of patient visits) was associated with a higher rate of relapse than no antibiotic treatment. Collectively, the use of amoxicillin increased the risk of relapse (odds ratio: 3.37), and the use of other antibiotics decreased the rate of relapse (odds ratio: 0.28).
In analyzing the various aspects of this study, the authors concluded that relapse was not related to the severity of baseline disease (FEV1) or initial acute presentation. The only predictive factor for relapse was the initial use of antibiotics. Patients treated with antibiotics had a lower relapse rate within the specified time period, unless they were given amoxicillin. One reason for the latter finding is that there may be a high prevalence of resistant organisms in this population. Therefore, the authors emphasize that the resistance profile of the institution at which the patient is being treated should be considered when selecting antibiotics.