However, each chart had a record of the patients’ symptoms, physical examination, and therapy given. The data were reviewed for the initial visit before reviewing the records of subsequent visits in an attempt to minimize bias. In many circumstances, the clinician was not aware of the patient’s PFTs.
Furthermore, some important factors, which could help in deciding the therapy choice, were not assessed in our study (eg, bacteriologic assessment of sputum, number of prior exacerbations). The lack of microbiologic data and in vitro susceptibility tests limits our ability to confirm our hypothesis about resistance contributing to the higher relapse rates of those treated with amoxicillin.
One further possible limitation of this study is that some of the patients with persistent or worsening symptoms (relapses) could have been treated at other institutions, but this is not likely because most of our patients receive all of their health care at the Veterans Medical Center or one of its satellite clinics. Furthermore, no intervention was performed to verify patient compliance with the prescribed medications. However, it is unlikely that lack of compliance with amoxicillin would have been a significant factor associated with increased recurrence.
Patients with documented COPD, even with mild symptoms at presentation, benefit from antibiotic therapy. However, the choice of antibiotic is important (because resistant organisms are increasing and are likely contributing to treatment failures) and should probably be based on the resistance profile to antibiotics in the institution where the patient is being treated. We feel that this retrospective study raises many questions about the current treatment recommendations for AECB, and supports the need for a prospective controlled trial to answer questions about the most appropriate use of antibiotics in these patients.