This pilot study demonstrated that VOCs can be detected and measured non-invasively from breath samples of preterm and full-term infants. In addition, the study showed that full-term infants express different concentrations of VOCs than preterm infants.

Measuring the VOCs from infants’ exhaled breath samples non-invasively is novel. VOCs have been traditionally measured in blood, urine and stool. However, there are limitations related to sampling availability, possibility of cross contamination, and the burden of sample’s processing [10,11,12,13,14,15]. Non-invasive breath sample has several advantages, such as being always available, could be collected at any time, multiple times a day, potential for real-time results, and avoid multiple blood sampling or obtaining urine or stool. Collecting fecal samples from newborn infants has some difficulties due to the lack of regular samples available when infants, especially preterm infants and especially if they are on parenteral nutrition, do not necessarily defecate every day. Therefore, multiple attempts were made to measure VOCs from exhaled breath. Subjects in these attempts were intubated and supported with invasive mechanical ventilation. A single-center study examined tracheal aspirates from intubated preterm infants using a sensor of electronic nose system [16]. However, one of the limitations of this device is the lack of sensitivity; and when simultaneously measuring multiple compounds, detection limits could be challenging [17]. In another study, VOCs were measured in intubated infants using the High-Performance Liquid Chromatography (HPLC) by nano-HPLC coupled to high-resolution Mass Spectroscopy (MS) [18]. Although in adults, multiple studies measured VOCs successfully in breath non-invasively, we could only find a single recent study that measured them in preterm infants non-invasively using the Gas Chromatography–Mass Spectroscopy (GC–MS) device [19]. The current study measured VOCs non-invasively in preterm and full-term infants. This study used the SIFT-MS for analysis. SIFT-MS is relatively newer compared to the GC–MS. SIFT-MS allows real-time absolute quantification of several trace gases simultaneously, even when an abundance of atmospheric gas is present [20] and that might be the reason that, in this study, the device equally detected VOCs in samples of respiratory circuit and of ambient air of the incubator. Of note, GC–MS is relatively expensive and requires highly trained operators [17].

In the current study, we intended to measure VOCs in preterm infants <30 weeks of GA, because this specific preterm population is liable to complications of prematurity and tends to adopt alternative physiological processes and metabolic pathways that might differ significantly from full-term infants. We also included a control group of full-term infants who were ≥37 weeks of GA to validate the detection and measuring processes. There was a significant difference between both groups for certain compounds, specifically, the 2-propanol, acetaldehyde, acetone, acetonitrile, benzene, ethanol, isoprene, pentane, 3-methylhexane, 2-nonene, ethane, triethylamine, and trimethylamine. All these compounds might have potential sources within the body. Isoprene is a marker of cholesterol synthesis, pentane -produced during lipid peroxidation- and ethane increase after tissue injury, ethanol is generated from bacterial metabolism, acetaldehyde is formed after the oxidation of ethanol, methane -produced from gut flora- is an indicator of carbohydrate malabsorption, acetone is produced when the body utilizes fat rather than glucose for energy, 2-propanol is produced from acetone’s reduction, hydrogen sulfide is a by-product of bacterial metabolism in the mouth, benzene and acrylonitrile usually have exogenous sources, and trimethylamine is produced by gut flora [6, 21, 22].

This study has several strengths, it showed the feasibility of detection and measuring VOCs in newborn infants from the exhaled breath using non-invasive method. Furthermore, to our knowledge, this is the first study that showed that full-term infants express different concentrations of VOCs from preterm infants. In addition, the investigator who processed the samples was blinded, therefore there was no operator bias. Nevertheless, given the pilot nature of this study, the sample size was small; that could impose a limitation when comparing groups as the study was not powered for such comparison. However, the study fulfilled its feasibility aim. Although this study was meant to test feasibility of detecting VOCs in non-intubated babies only, we recommend including controls in future studies for comparing infants with vs without breathing tubes. In addition, the following step in future studies would be to ensure reproducibility of this new technique.

In conclusion, it is feasible to measure VOCs in breath samples of non-intubated preterm and full-term infants non-invasively. Full-term infants express different concentrations of VOCs than preterm infants. Further studies are needed to examine the utility of measuring VOCs to identify and monitor neonatal diseases and predict their outcomes.