Background: Breath analysis by secondary electro-spray ionization mass spectrometry (SESI-HRMS) can identify VOCs distinguishing asthmatic children from healthy controls. However, the method requires real-time exhalation maneuvers that are challenging for preschool children. This study aimed to evaluate a remote breath collection method suitable for younger children.

Methods: We included 59 children aged 3-6 years (29 with recurrent wheezing (RW) and 30 healthy controls). Breath collection was conducted using 0.7-liter Nalophan bags equipped with a valve and mouthpiece. Children were instructed to exhale into the bags, and the contents were transferred within 15 minutes to SESI-HRMS. Detection of previously identified asthma markers was performed on the acquired mass spectra with a mass tolerance of 0.002 Da, followed by statistical analysis.

Results: Of 375 m/z features linked with asthma in school children, 128 were detectable in preschoolers with the remote method. A total of 20 features were significant in the new population (FDR adjusted p<0.05). Predictive performance using supervised machine learning revealed a ROC AUC of 0.78 (95% CI: 0.66 - 0.89) for distinguishing RW from healthy controls. Representatives of lysine degradation, aldehydes and the chemical families of fatty acid metabolites and monosaccharides were rediscovered among the significant predictors in preschoolers.

Discussion: This study is the first to evaluate a transition from real-time to remote breath collection for asthma screening in preschoolers using SESI-HRMS. Despite feature loss, significant VOCs associated with asthma were identified, indicating the potential of breath analysis for early asthma detection in young children.