To the Editors:
We read with great interest the letter by Liu et al. 1 and fully agree that exhaled breath condensate (EBC) collection devices significantly affect EBC biomarker levels, and that absolute values from different devices are not directly comparable 1, 2. We also agree that one single (currently available) collection device may not be ideal for all applications and that, in future studies, each marker may need to be tested by a variety of devices in order to determine the optimal collection apparatus 2. In our opinion, some additional considerations of EBC methodology should be made. We believe that these may change our current view on the variability associated with the measurement of exhaled markers in EBC. In turn, this altered consciousness may provide a tentative way out of current methodological controversies and, by consequence, it may offer new perspectives towards the standardisation of EBC methodology 3.
First, the levels of nitrogen oxides and total protein measured by Liu et al. 1 in EBC were expressed as nmol or μg “per 10 min”, and may suggest an influence of the duration of EBC collection on the concentration of these biomarkers in EBC. In fact, we also believe that such an influence exists; however, the question is whether standardisation for time results in the best reproducibility. So far, no study has answered this, whereas some studies have standardised for EBC volume or a dilution factor 4. Alternatively, measurements can be standardised for the volume of exhaled air, which resulted in significant correlations between exhaled nitrite and leukotriene B4 in the study of Knobloch et al. 5 in calves. In the data reported by Liu et al. 1, EBC was collected during a fixed period of 10 min in each tidally breathing subject, by each of the different collection systems. Therefore, it seems logical to report the level of biomarkers in the resulting condensate in relation to this period of time. However, this bypasses an important methodological EBC collection issue, as it does not take into account that each of these devices are open-ended systems. In an open-ended collection system, the available condensation surface will be saturated after a certain period of time and thus, before and beyond this point in time, noncondensed exhaled breath may be lost through the open end of the collection device. By consequence, exhaled respiratory droplets with their possible solutes (or biomarkers) can also be lost. For this reason, there is in fact an influence of the duration of collection on the concentration of biomarkers in EBC, although we should consider this as an influence of the condensation saturation point of each device for each single subject. As confirmation of this theoretical consideration, we found in several studies that the volume of collected EBC within a certain period of time is highly variable, which makes standardisation for time not optimal, in our opinion. Therefore, we believe, this issue may only be solved through use of a closed collection system that enables “global condensation”, i.e. the condensation of the total exhaled breath volume within a certain period of time 6. As it is now evident from several studies that collection devices can influence EBC marker concentration, the time has come to unravel which factors are mainly responsible for this effect: dimensions of the device, condensation surface, temperature of the condenser tube, closed or open design, and so on 1, 2, 7–9.
Secondly, we agree with Liu et al. 1 that it is essential for EBC methodology to have reproducibility data on (all) biomarkers in (all) collection devices. Moreover, we even suggest that it is essential to have reproducibility data on each biomarker, both per specific assay, as well as per collection device. Furthermore, we should be careful when interpreting data from only six healthy subjects. Likewise, total protein measurements, a rough value for the total protein content in a biofluid, were performed using an assay with a lower limit of detection of 4 μg·mL−1, and thus one should be equally careful when interpreting median values of 6.6±3.3 μg total protein 1. Furthermore, and with respect to the variability of markers in EBC, it may be that the reproducibility of markers can be improved by standardising for exhaled volume, time or a dilution factor 10. In fact, we urgently need more studies to address this important topic.
Thirdly, regarding pH in EBC, we would like to refer to the study of Prieto et al. 9 who demonstrated higher values of deaerated and nondeaerated pH using the EcoScreen compared with the Rtube.
Taken together, it is evident from several studies that exhaled breath condensate collection devices have an influence on concentrations of biomarkers in exhaled breath condensate 1, 2, 7–9. However, the key questions remain: which device characteristics are mainly responsible for this effect? And what are the optimal device characteristics? Moreover, it is important to study the influence of different designs of exhaled breath condensate collection devices on biomarker reproducibility for each group of markers (cytokines, nitrogen oxides, markers of oxidative stress, leukotrienes and so on) in sufficiently large groups of subjects, with or without disease. We believe that this “climate change” may offer new perspectives towards a better standardisation of exhaled breath condensate measurements.
Statement of interest
None declared.
- © ERS Journals Ltd