Answers are grounded in this paper and cite their sources.
Breath acetone concentrations were measured in 141 subjects (aged 19-91 years, mean = 59.11 years, standard deviation = 12.99 years), male and female, undergoing an oral glucose tolerance test (OGTT), having been referred to clinic on suspicion of type 2 diabetes. Breath samples were measured using an ion-molecule-reaction mass spectrometer, at the commencement of the OGTT, and after 1 and 2 h. Subjects were asked to observe the normal routine before and during the OGTT, which includes an overnight fast and ingestion of 75 g glucose at the beginning of the routine. Several groups of diagnosis were identified: type 2 diabetes mellitus positive (T2DM), n = 22; impaired glucose intolerance (IGT), n = 33; impaired fasting glucose, n = 14; and reactive hypoglycaemia, n = 5. The subjects with no diagnosis (i.e. normoglycaemia) were used as a control group, n = 67. Distributions of breath acetone are presented for the different groups. There was no evidence of a direct relationship between blood glucose (BG) and acetone measurements at any time during the study (0 h: p = 0.4482; 1 h: p = 0.6854; and 2 h: p = 0.1858). Nor were there significant differences between the measurements of breath acetone for the control group and the T2DM group (0 h: p = 0.1759; 1 h: p = 0.4521; and 2 h: p = 0.7343). However, the ratio of breath acetone at 1 h to the initial breath acetone was found to be significantly different for the T2DM group compared to both the control and IGT groups (p = 0.0189 and 0.011, respectively). The T2DM group was also found to be different in terms of ratio of breath acetone after 1 h to that at 2 h during the OGTT. And was distinctive in that it showed a significant dependence upon the level of BG at 2 h (p = 0.0146). We conclude that single measurements of the concentrations of breath acetone cannot be used as a potential screening diagnostic for T2DM diabetes in this cohort, but monitoring the evolution of breath acetone could open a non-invasive window to aid in the diagnosis of metabolic conditions.
No takes yet. Share an insight, caveat, or question.
Andrews et al. (2018) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: