Abstract
Background: The elucidation of volatile organic compounds specifically produced by microorganisms may assist in developing a fast and accurate methodology to determine pulmonary bacterial infections. Development of this methodology might ultimately lead to the identification of bacterial species in breath.
Methods: Over 300 bacterial headspace samples from 4 different micro organisms were analyzed by gas chromatography-mass spectrometry to identify relevant VOCs, and compose profiles of VOCs that are specific for any of the micro organisms Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus and Klebsiella pneumoniae. Differently abundant VOCs were determined and classification models based on support vector machines (SVM) were build to allow classification of the samples.
Findings: We were able to identify a large number of compounds that show significantly different availability in bacterial cultures compared to medium and in bacterial cultures cross-compared. We identified compounds demonstrating highly significant differences between the four E. coli strains and between the two S. aureus isolates: methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive Staphylococcus aureus (MSSA). SVM models were able to classify the micro organisms with very high degrees of sensitivity and specificity based on 6 VOCs from headspace.
Interpretation: We demonstrated that identification of the studied micro organisms is possible based on a few compounds measured in headspace of cultures. It provides a fast, non-invasive, and sensitive technique as a potential diagnostic approach in medical microbiology.
- © 2011 ERS
