Abstract
For several years there is much emphasis in breath analysis for the diagnosis of lung and especially infectious diseases.
It has been known for a long time that various bacteria produce characteristic smells which are caused by specific volatile organic compounds (VOCs) released as products of bacterial metabolism during growth. In the present in- vitro study, differential ion mobility spectrometry (DMS) was used as a fast, sensitive method for the identification of bacterial VOCs by bacterial cultures.
The aim of this study was to assess, whether discrimination between different bacteria by comparison of the VOC-pattern is possible.
The headspace over the cultures was examined by DMS 2, 4, 6 and 24h after incubation. For the analysis, clinical isolates of E. coli, P. aeruginosa, methicillin susceptible S. aureus und methicillin resistant S. aureus, 10 strains each, and non-inoculated culture media were used as a control. The evaluation of the VOC-pattern was based on a peak detection and cluster analysis. Changes in VOC concentration, depending on the time of incubation, were found in the headspace of the bacterial cultures. Based on VOC profiles, a significant differentiation could already be seen after two hours (Mann-Whitney U-Test; α ≤ 0,05). The generated classifiers showed discrimination between the four bacteria after 2, 4, 6 and 24h with a sensitivity of around 80% and a specificity of between 90% and 100%.
As a proof-of-concept, this study has shown encouraging results, suggesting that DMS may be a cost- effective tool for a fast identification of bacteria in cultures and possibly in exhaled breath of patients.
- © 2014 ERS