PURPOSE: Staphylococcus aureus (SA) is a rare cause of pneumonia, however the guidelines for management of pneumonia result in frequent use of therapy against SA. This approach exposes the patient to the risks associated with the use of these drugs and the possibility of emergence of resistant strains with time. We investigated whether the Gram stain of respiratory tract samples can be used to predict the growth of SA.
METHODS: We examined the Gram stains of tracheal aspirates and expectorated sputa submitted to the microbiology laboratory of our tertiary care hospital over 3 consecutive months while blinded to the laboratory staff Gram stain interpretation and culture results.
RESULTS: A total of 126 samples were analyzed, 30 tracheal aspirates and 96 expectorated sputa. In the expectorated sputa group, 26 of 96 samples grew SA (27%) and in the tracheal aspirate group, 8 of 30 samples grew SA (27%). The sensitivity, specificity and negative predictive value of the Gram stain for SA in the expectorated sputa group was 40.7%, 50.7% and 69.2%. The sensitivity, specificity and negative predictive value of the Gram stain for SA in the tracheal aspirate group was 71.4%, 77.3% and 89.5%.
CONCLUSION: Expectorated sputum is not useful in identifying patients in whom therapy for SA can be withheld. However, tracheal aspirate performed much better with a high negative predictive value (89.5%) for identifying the absence of SA. However, this finding will need to be confirmed with a larger study.
CLINICAL IMPLICATIONS: Withholding therapy for SA in intubated patients without Gram stain evidence of SA in tracheal aspirate is an approach that may be useful to limit the use of unnecessary antibiotics. However, given our small sample size, additional studies will be required.
DISCLOSURE: Sodienye Tetenta, None.