Background: The usefulness of sputum Gram stain and culture in guiding microbiological diagnosis of community-acquired pneumonia (CAP) is controversial. We evaluated the role of sputum examination at a university teaching hospital.
Material/Methods: Three hundred forty-seven adult patients with CAP were enrolled in this study. Before administering antibiotic therapy, sputum was collected and its quality evaluated. Samples were gram stained and those of good quality were assessed for a predominant morphotype.
Results: Sputum samples were obtained from 216 patients (62%), and of these 124 (57%) samples were good quality and 80 (65%) showed a predominant morphotype. Sputum culture yielded a causative organism in 70 (88%) of the 80 samples with a predominant morphotype. In the cases of patients who had received previous antibiotic treatment, a good quality samples showing a predominant morphotype and positive culture was less frequently obtained from than from those who had not (p<0.0001). The sensitivity and specificity of the gram-positive diplococci identification in the sputum culture of S. pneumoniae were 68.2% and 93.8%, respectively, and the sensitivity and specificity of the gram-negative coccobacilli identification in the sputum culture of H. influenzae were 76.2% and 100%, respectively.
Conclusions: Gram stain of sputum samples was useful in guiding microbiological diagnosis of CAP in 23% of patients. The Gram stain and culture of sputum samples obtained from patients who have received antibiotic treatment was unreliable. The presence of gram-positive diplococci and gram-negative coccobacilli was highly specific for the culture of S. pneumoniae and H. influenzae, respectively.

Keywords: Hospitalization, Streptococcus pneumoniae - isolation & purification, Sputum - microbiology, Process Assessment (Health Care), Pneumonia, Pneumococcal - microbiology, Phenazines, Gentian Violet, Community-Acquired Infections - microbiology, Colony Count, Microbial, Aged, 80 and over, Adolescent