Prospective studies suggest that 10 to 25 percent of cases of community-acquired pneumonia are caused by atypical pathogens such as Legionella pneumophila, Chlamydia species and Mycoplasma pneumoniae. Guidelines from the American Thoracic Society for the management of community-acquired pneumonia, however, do not recommend routine testing for atypical pathogens. The recommended empiric therapy for hospitalized patients who are not severely ill includes a second- or third-generation cephalosporin or a beta lactam–beta lactamase inhibitor with or without a macrolide. To evaluate the role of atypical pathogens and macrolide antibiotics in community-acquired pneumonia, Mundy and colleagues conducted a retrospective study of patients hospitalized because of community-acquired pneumonia between 1990 and 1991.

Routine sputum and blood cultures were obtained in all patients. Patients were also evaluated for the presence of atypical pathogens, including L. pneumophila, Chlamydia pneumoniae, Chlamydia psittaci and M. pneumoniae. The empiric antibiotic treatment was reviewed to determine which agents were used and if the selection had any impact on the outcome. Of the 385 patients with community-acquired pneumonia, 31 (8 percent) had pneumonia caused by atypical pathogens. Sixteen of the atypical cases also had a second common pathogen. In most cases of pneumonia caused by atypical pathogens, the pathogen was identified after patients were discharged from the hospital. On the basis of the American Thoracic Society guidelines, 260 patients were classified as severely ill, but only 57 were admitted to the intensive care unit. Only four of the patients with an atypical pathogen received a seven-day course of erythromycin or tetracycline. No patient in whom an atypical pathogen was detected died. None of the patients received macrolide therapy.

The authors conclude that early establishment of the causative agents for community-acquired pneumonia is problematic. Only 8 percent of the patients in this study had atypical pathogens, and most were identified after the patients were discharged. The fact that no macrolides were used suggests that routine administration of these antibiotics in all patients with community-acquired pneumonia is not necessary. The authors note that patients who are seriously ill and require admission to an intensive care unit should receive a macrolide antibiotic with one of the other recommended antibiotics, in accordance with the American Thoracic Society guidelines.