The Usefulness of the Sputum Gram Stain in the Diagnosis of Pneumonia

Abstract

The value of the sputum Gram stain in the diagnosis of pneumonia is controversial. Use of the Gram stain to screen sputum samples for contamination before culture is well accepted, but the ability of the Gram stain to predict the microbial cause of pneumonia and guide antibiotic therapy is less certain. Studies in which the criteria for Gram stain interpretation are clearly identified indicate that, when a specimen meets acceptable levels of purulence and salivary contamination and shows a predominant organism, the Gram stain predicts the results of sputum, transtracheal aspirate, and blood cultures with high specificity. Moreover, Gram stains meeting these criteria have been shown to be accurate guides to antibiotic treatment. In intubated patients, the Gram stain of lower respiratory secretions collected by endotracheal aspirate, protected specimen brush, or bronchoalveolar lavage provides immediate and reasonably specific information regarding the presence of pneumonia and the causative agent. However, many patients with pneumonia do not produce an adequate sputum specimen, and not all Gram stains show a predominant organism. Furthermore, Gram stains are often poorly prepared or misinterpreted by inexperienced clinicians. Common errors include interpreting an inadequate or improperly stained specimen, overlooking Gram-negative bacilli (particularly Haemophilus influenzae), and mistakenly identifying all Gram-positive organisms as pneumococci. Thus the sputum Gram stain is helpful in the diagnosis of pneumonia when it is interpreted according to specific criteria and when attention is paid to the avoidance of common pitfalls. In the patients from whom a diagnostic specimen is available, the sputum Gram stain can be an accurate, low-cost guide to effective, narrow-spectrum antibiotic therapy.