Untargeted analysis of volatile markers of post-exercise fat oxidation in exhaled breath
Authors: André Homeyer, Júlia Blanka Sziládi, Jan-Philipp Redlich, Jonathan Beauchamp, Y Lan Pham
Organizations: Fraunhofer Institute for Digital Medicine MEVIS, Bremen, Germany · Department of Sensory Analytics and Technologies, Fraunhofer Institute for Process Engineering and Packaging IVV, Freising, Germany
Breath acetone represents a promising non-invasive biomarker for monitoring fat oxidation during exercise. However, its utility is limited by confounding factors, as well as by the fact that significant changes in concentration occur only hours post-exercise, which makes real-time assessment difficult. We performed an untargeted screening for volatile organic compounds (VOCs) that could serve as markers of fat oxidation beyond acetone, and investigated whether breath measurements taken during exercise could predict post-exercise changes in fat oxidation. Nineteen participants completed two 25-min cycling sessions separated by a brief 5-min rest period. VOC emissions were analysed using proton-transfer-reaction time-of-flight mass spectrometry (PTR-TOF-MS) during exercise and after a 90-min recovery period. Blood β-hydroxybutyrate (BOHB) concentrations served as the reference marker for fat oxidation. Among 773 relevant analytical features detected in the PTR-TOF-MS measurements, only four signals exhibited strong correlations with BOHB (ρ≥ 0.82, p = 0.0002)-all attributable to acetone or its isotopologues or fragments. End-of-exercise measurements of these signals enabled accurate prediction of participants with substantial post-exercise BOHB changes (F1 score ≥ 0.83, accuracy = 0.89). Our study did not reveal any novel breath-based biomarkers of fat oxidation, but it confirmed acetone as the key marker. Moreover, our findings suggest that breath acetone measurements during exercise may already enable basic predictions of post-exercise fat oxidation.
Figures & tables
Participants
19
Sex [f/m]
6/13
Age [years]
30.2 ± 7.2
BMI [kg/m 2 ]
23.6 ± 2.7
Height [cm]
175.5 ± 8.2
Weight [kg]
73.3 ± 13.1
Table 1: Cohort demographics and metadata. Values are presented as mean ± standard deviation. BMI = body mass index.
Figure 1: Experimental protocol. Left: Participants completed two 25-min cycling sessions on a cycle ergometer, interspersed by a 5-min rest period. Centre: Exhaled breath was sampled via a buffered end-tidal (BET) sampler. A metronome displayed on a notebook helped maintain a constant pedalling rhythm. Right: PTR-TOF-MS measurements of exhaled breath were recorded throughout both cycling sessions and at follow-up (90 min post-exercise). Blood BOHB concentrations were determined immediately before and after the cycling sessions and at follow-up.
Figure 2: Data pre-processing. The raw PTR-TOF-MS measurements of every participant were condensed into eleven 5-minute steps. For each step, a TOF spectrum was computed by masking the cycles belonging to end-tidal phases and averaging the respective TOF bin values.
Signal ( m/z )
Spearman’s ρ
59.214
0.90 (p = 0.0002)
31.027
0.88 (p = 0.0002)
60.051
0.83 (p = 0.0002)
59.046
0.82 (p = 0.0002)
20.025
0.57 (p = 0.0126)
17.027
-0.55 (p = 0.0160)
Table 2: Untargeted search results. The ten of the 773 relevant signals with the highest absolute Spearman coefficient values at the follow-up time point. Sorted from high to low.
Signal ( m/z )
59.046
59.214
60.051
31.027
0.97
0.99
0.91
59.046
0.98
0.89
59.214
0.91
Table 3: Spearman correlation coefficients between the four signals at the follow-up time point.
Figure 3: Distributions of the logarithmic intensities of the four significant signals at the follow-up measurement. Data comprise the mean end-tidal plateau signal intensities of four exhalations at follow-up for each participant.
Figure 4: Signal changes over time. Top: Distributions of relative changes in signal intensity at m/z 59.046 (left) and m/z 60.051 (right) compared to the start of exercise. The signals represent protonated acetone and its \hphantom{{}^{\text{13}}_{\text{}}}{\vphantom{\text{X}}}{}^{\mathchoice{\hbox to0.0pt{\hss\displaystyle\vphantom{\smash[t]{\text{2}}}\text{13}}}{\hbox to0.0pt{\hss\textstyle\vphantom{\smash[t]{\text{2}}}\text{13}}}{\hbox to0.0pt{\hss\scriptstyle\vphantom{\smash[t]{\text{2}}}\text{13}}}{\hbox to0.0pt{\hss\scriptscriptstyle\vphantom{\smash[t]{\text{2}}}\text{13}}}}\kern 0.0pt\text{C} -isotopologue, respectively. Bottom: Correlation plots of the relative changes in m/z 59.046 and BOHB at the end of training (52.5 min; left) and the follow-up time point (147.5 min; right).
Figure 5: Post-exercise predictability. Top: Spearman correlations between relative changes in signal intensity of the two acetone signals ( m/z 59.046, m/z 60.051) at different time points and relative changes in BOHB at follow-up. Bottom: Predictability of relative changes in BOHB at follow-up based on relative changes in the acetone signals at the end of exercise: absolute error distributions (left) and scatterplot of observed versus predicted values (right).
Aug 30, 2026·Nadejda Drenska, Matthew Lemoine, Gowri Priya Sunkara +3Graph Neural Networks
Department of Mathematics, Louisiana State University, Baton Rouge, LA 70803, USA. · Pennington Biomedical Research Center(PBRC), Baton Rouge, LA 70808, USA.