Abstract
OBJECTIVE: To describe the use of sputum Gram stain and culture in patients with community-acquired pneumonia (CAP) and to determine the factors that are associated with obtaining sputum for culture. TYPE OF STUDY: Prospective observational cohort study of patients hospitalized for treatment of CAP at four medical institutions in three geographic locations. MAIN MEASUREMENTS: Results of Gram stain and culture of sputum; comparison of patients who had sputum processed for culture within 24 h of admission with those who did not have such a specimen processed during the first week of hospitalization; and the results of investigator assignment of etiology of pneumonia according to predefined criteria. RESULTS: Four hundred and seventy-eight of 1339 (36%) patients had a sputum specimen processed for culture within 24 h of admission. Patients who had a sputum specimen processed within 24 h of admission were more likely to be hospitalized at the Boston site (odds ratio [OR] 20.6) or Pittsburgh sites (3.4) and to have current sputum production, chronic obstructive lung disease and moderate or large amount of sputum. Female sex (0.4), neutropenia (0.05), and do not resuscitate status (0.36) were important predictors of failure to have a sputum processed for culture. The rate of Streptococcus pneumoniae isolation was highest in Boston, 53 of 269 (19.3 %) patients (P<0.001) compared with the other sites; Moraxella catarrhalis was isolated only at the Boston site. Sputum culture results served as the basis for the assignment of an etiological diagnosis of the pneumonia by investigators in 67% of 397 patients. CONCLUSIONS: Sputum is not processed for culture in the majority of patients with CAP. The factors that determine whether sputum is processed for culture within 24 h of admission are site of care and a variety of patient factors. Common respiratory pathogens when present in sputum culture tend to be used to assign an etiological diagnosis. A positive sputum culture result appears not to result in a more favourable outcome.