Poor-quality diets (PQD) are associated with non-communicable diseases^{(Reference Sur and Fay1)}. Furthermore, limited dietary variety may contribute to poor gut microbial diversity, which is associated with cancer risk^{(Reference Appuni, Rubens and Ramamoorthy2)}. Technological advances provide opportunities to explore alternative health biomarkers. Specifically, volatile organic compounds (VOC) in exhaled breath may offer a non-invasive biomarker for nutritional status. This pilot study aimed, as part of a wider pan-European collaboration, to characterise the VOC profiles of adults with poor- quality diets to assess their utility as a biomarker of nutritional status.

Fifty participants were recruited in two sub-groups: i) adults with PQD (<3 portions of fruit and vegetables per day, n 33) and ii) adults with iron deficiency anaemia (ID; n 17). Participants attended the University after a 12 hour fast, with no smoking or chewing gum prior to sampling. Baseline anthropometric measures were taken, and a fasted breath sample was collected according to standard procedures into a Tedlar bag (Zefon Int'l, USA). Participants consumed a standard breakfast and completed online questionnaires evaluating general health and lifestyle. A final breath sample was collected 2 hours post-breakfast. Samples were shipped to researchers at the Biosense Institute (BIOS, Serbia), for extraction and analysis within 3 days of collection, using membrane-inlet mass spectrometry. Reagents used were methanol, ethanol, acetone, isoprene and n- pentane (Sigma-Aldrich, US) all in liquid phase. A favourable ethical opinion was received from the University ethics committee. Parametric analysis was conducted using SPSS (IBM), significance was set at p < 0.05, data are presented as mean ± SD.

Mean BMI was 32.7 ± 13.3 kg/m² for the whole sample, with no significant (NS) difference between the two groups; 30% were male. Fasted acetone levels were observed to be higher in the PQD than in the ID group (NS, 227.0 ± 246.5 vs 191.2 ± 206.4 ppb, p = 0.298). The PQD group presented with lower levels of fasted pentane (NS) (19.8 ± 19.6 vs 26.6 ± 30.5 ppb; p = 0.498). Furthermore, a self-reported urban (as opposed to rural) living environment (82%), was associated with higher fasted acetone (NS, 224.2 ± 246.2 vs 176.1 ± 164.4 ppb; p = 0.290) and ethanol (NS, 386.2 ± 316.9 vs 175.7 ± 104.4 ppb; p = 0.091) concentrations. There were no significant differences observed between the VOC profiles of plant-based (n 8) and omnivorous (n 37) participants.

In conclusion, this pilot study presents novel data on VOC profiles for two population groups at nutritional risk. Despite a lack of significant between-group differences the group as a whole presented with higher fasted pentane than recommended for healthy adults (0-10 ppb;^{(Reference Mürtz3)}). Living environments appeared to impact breath profiles and warrant further investigation in various population groups to progress our understanding of the use of VOC profiles as nutritional biomarkers.