Abstract
Because of difficulties in accurately determining an etiologic diagnosis, the ideal treatment for acute community-acquired lower respiratory tract infections remains to be established. Suggested regimens, in the absence of persuasive findings from gram-stained preparations of sputum, are best guesses made on the basis of clinical and epidemiological data. Results from randomized controlled trials are only partially helpful since most trials require the isolation of a potential pathogen, a condition that excludes as many as half of all potential study participants. Nonetheless, initial therapy for patients producing phlegm with abundant polymorphonuclear leukocytes should be chosen on the basis of the predominant organisms seen on a stained sputum smear. Patients who are admitted to the hospital from home, >50 years of age, and not producing phlegm can be safely treated with amoxicillin or a second-generation cephalosporin. Atypical pneumonia in younger patients is best treated with erythromycin or tetracycline. Patients admitted from nursing homes, who frequently fail to respond to an initial treatment course in the absence of a diagnostic sputum sample, should receive empirical therapy with an agent that is also active against aerobic gram-negative rods.

This publication has 0 references indexed in Scilit: