Previous measurements of acetone concentrations in the exhaled breath of healthy individuals and the small amount of comparable data for individuals suffering from diabetes are briefly reviewed as a prelude to the presentation of new data on the sporadic and wide variations of breath acetone that occur in ostensibly healthy individuals. Data are also presented which show that following a ketogenic diet taken by eight healthy individuals their breath acetone concentrations increased up to five times over the subsequent 6 h. Similarly, the breath acetone increased six and nine times when a low carbohydrate diet was taken by two volunteers and remained high for the several days for which the diet was continued. These new data, together with the previous data, clearly indicate that diet and natural intra-individual biological and diurnal variability result in wide variations in breath acetone concentration. This places an uncertainty in the use of breath acetone alone to monitor blood glucose and glycaemic control, except and unless the individual acts as their own control and is cognizant of the need for dietary control.

In the light of the exciting recent developments in the detection of Pseudomonas aeruginosa in the breath of cystic fibrosis patients by measuring exhaled HCN, an in vitro study has been conducted to identify and quantify the volatile compounds emitted into the gas phase by other respiratory pathogens. Selected ion flow tube mass spectrometry (SIFT-MS) was used to investigate clinical isolates of Staphylococcus aureus (SA), Streptococcus pneumoniae (SP), and Haemophilus influenzae (HI). Six volatile compounds, mainly alcohols, ketones and aldehydes, were found to be elevated in the headspace of SA cultures and eight were elevated in the SP cultures. It is clear from the mass spectra that a number of other compounds were present at low levels that, as yet, cannot be identified by SIFT-MS alone. Only indole and ethanol were somewhat elevated in the headspace of some of the HI cultures. Principal component analyses (PCA) indicated that the SA cultures clearly separated into two distinct groups in terms of their volatile compound emissions. This may relate to genetic or phenotypic differences. It is postulated that the very efficient production of ethanol and acetaldehyde by SA and SP may provide indicators of airways infection by these bacteria if the levels of these compounds are elevated in exhaled breath above those levels expected due to their normal endogenously produced levels. Plans are in place to check this postulate by breath analysis studies involving patients with pulmonary infections with these organisms.

Sampling of cell culture headspace into SIFT-MS for direct analysis of volatile compounds.

This detailed study has provided the kinetics data, in particular the product ion distributions, for the reactions of a number of volatile aldehydes, which allows their analyses by SIFT-MS in humid air, including exhaled breath, food emanations and other biogenic media.