Exhaled particles and small airways
Originally, studies on exhaled droplets explored properties of airborne transmission of infectious diseases. More recently, the interest focuses on properties of exhaled droplets as biomarkers, enabled by the development of technical equipment and methods for chemical analysis. Because exhaled droplets contain nonvolatile substances, particles is the physical designation. This review aims to outline the development in the area of exhaled particles, particularly regarding biomarkers and the connection with small airways, i e airways with an internal diameter < 2 mm.
Generation mechanisms, sites of origin, number concentrations of exhaled particles and the content of nonvolatile substances are studied. Exhaled particles range in diameter from 0.01 and 1000 μm depending on generation mechanism and site of origin. Airway reopening is one scientifically substantiated particle generation mechanism. During deep expirations, small airways close and the reopening process produces minute particles. When exhaled, these particles have a diameter of < 4 μm. A size discriminating sampling of particles < 4 μm and determination of the size distribution, allows exhaled particle mass to be estimated. The median mass is represented by particles in the size range of 0.7 to 1.0 μm. Half an hour of repeated deep expirations result in samples in the order of nanogram to microgram. The source of these samples is the respiratory tract ling fluid of small airways and consists of lipids and proteins, similarly to surfactant. Early clinical studies of e g chronic obstructive pulmonary disease and asthma, reported altered particle formation and particle composition.
The physical properties and content of exhaled particles generated by the airway reopening mechanism offers an exciting noninvasive way to obtain samples from the respiratory tract lining fluid of small airways. The biomarker potential is only at the beginning to be explored.
KeywordsExhaled particles Small airways Airway closure Airway opening Surfactant Proteomics SP-A Albumin DPPC POPC
Exhaled air is an aerosol containing endogenously generated droplets. These droplets contain water and nonvolatile material, and “particles” is therefore the physical designation, even though they are liquid droplets. Studies of exhaled particles originally aimed to understand the transmission of airborne infections. More recently, however, interest has extended to include a search for biomarkers of pathology in the airways. The close proximity to pathological processes in the airways makes exhaled particles an attractive option for clinical investigation. Knowledge of the site of origin and mechanisms of generation of exhaled particles constitutes an important basis for exploring associated biomarkers.
Along with exhaled particles, volatile- and semi-volatile substances may carry important biomarkers such as exhaled nitric oxide [1, 2], e g regarding the effect of the injury on small airways due to the mechanical stress following cyclic opening and closure among patients with chronic obstructive pulmonary disease (COPD) . Exhaled Breath Condensate and certain physiological methods also contribute to the diagnosis of small airway disease [4, 5, 6]. The present review, however, focuses on endogenously produced exhaled particles originating in the respiratory tract lining fluid (RTLF) along the airways including the pharynx and mouth. It is limited to studies presenting the count number and size distribution of exhaled particles, therefore a wide range of important studies on small airways have been omitted. We have highlighted key studies reporting the increasing appreciation of the origin and characteristics of exhaled particles. We also present available information on the use of exhaled particles from small airways as biomarkers.
It is very difficult to directly determine size of small droplets (diameter < 10 μm) floating in air. In practice, however, a measured property that depends on particle size is commonly used to indirectly estimate the size. Additional file 1, “Technical and methodological considerations,” outlines the various sizing methods employed in the studies reported here. Presumably, particles generated in situ and then exhaled are liquid spheres. Aqueous droplets will equilibrate with the water vapor in the surrounding air. It follows that their size depends on the surrounding air temperature and humidity as well as the particles’ composition. Equilibration is a rapid process (< 1 s) for small droplets, but may be confounded by the presence of a surfactant layer covering the droplet’s surface slowing evaporation or condensation.
Collection of exhaled particles
Chemical analysis requires sampling exhaled particles. The design of the sampling equipment will inevitably affect the size range of the collected sample. For examples, long tubing, parts not at roughly 35 °C, and sharp turns will contribute to losses of particles, particularly those that are relatively large.
The impactor is a sampling device that allows the collection of a size discriminated samples from an aerosol. Details are presented in Additional file 1.
Chemical analysis of collected exhaled particles
The great challenge analytically is the extremely low amounts of collected analytes, in the range of picogram (pg) per liter of exhaled air. Electron microscope and X-ray dispersive analysis or surface mass spectrometry, e.g. time-of-flight secondary ion mass spectrometry (TOF-SIMS) can analyze the deposited particles directly [7, 8]. Exhaled particles collected by impaction need to be appropriately desorbed. For proteins, immunological methods have dominated so far, but mass spectrometric methods have emerged for proteins as well as for lipids [9, 10]. Proteomics analysis has been able to quantify over 200 proteins by combining DNA-markers with PCR amplification of small amounts of particles (in the order of hundred ng) .
More than 70 years ago Duguid  aimed to assess the mechanisms of airborne transmission of infection from the mouth and throat. Five participants performed different breathing maneuvers, including normal mouth breathing, counting softly and loudly from 1 to 100, and performing various cough maneuvers. Immediately before these maneuvers, he had applied bacteria to the mucous membranes of the throat and nose. In a separate session, he applied a dye to the surfaces of the mouth, front teeth, lips, and tip of the tongue. Exhaled particles ended up either on a bacterial growth medium or on a glass slide for particle counting using a microscope.
Comment: Studies were on particles generated in the upper airways only. There was no information on the particles < 20 μm exhaled during normal mouth breathing. Particles > 20 μm were indeed exhaled during all other breathing activities, except for normal breathing.
About 20 years later Loudon and Roberts  aimed to determine the numbers and sizes of exhaled particles using a sampling technique that allowed comparisons of the frequency distribution of all particles > 1 μm. Three participants in two experiments performed a series of 15 coughs into a box and in two other experiments counted loudly from 1 to 100 into the box. Before each experiment, the participant swabbed the inside of his mouth with dye. After each experiment, the box was closed and particles settled on paper slips over 30 min. Settled particles were counted, as were the remaining airborne particles, which were deposited on a Millipore filter in the exit port of the box.
Comments: Studies were limited to particles generated in the mouth and there is no information on exhaled particles during normal breathing.
In 1997 Papineni and Rosenthal  presented results on droplets in exhaled breath obtained by two methods: (a) a real-time analysis by an optical particle counter (OPC) and (b) analysis of dried droplet residues by electron microscopy. The mouth was without dye, and consequently the site of origin of the exhaled particles was not necessarily the mouth region. The OPC and associated software presented particle sizes in six channels between 0.3 and 2.5 μm. Nose breathing, mouth breathing, coughing, and talking were studied in five healthy participants using the OPC and electron microscopic analysis of the mouth breathing particles was conducted with three of the participants.
Results according to the OPC method showed that mouth breathing resulted in 12.5 particles/L for diameters < 1 μm and 1.9 particles/L for diameters > 1 μm. Coughing resulted in 83.2 particles/L for diameters < 1 μm and 13.4 particles/L for diameters > 1 μm.
Comments: Normal breathing does indeed exhale submicron as well as larger particles. The site of origin and mechanisms of generation are still unknown; however, X-ray dispersive analysis of the residue of one particle revealed contents of potassium, calcium, and chloride, consistent with RTLF origin.
In 2004 Edwards et al.  investigated the ability to transiently diminish the number of exhaled particles by administering nebulized aerosols to human participants. Particles were measured by an OPC providing counts in six bins between 0.09 – > 0.5 μm. Eleven healthy participants were investigated on three visits separated by at least a week in a crossover placebo-controlled design. An aerosol was inhaled on the two first visits, either isotonic saline with a surface tension of 72 dyne/cm or a surfactant simulant consisting of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) or 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol with a surface tension of 42 dyne/cm. No aerosol treatment was conducted on the third visit. Participants wore nose clamps and breathed large tidal volumes of close to 1 L, inhaling particle-free air. Exhaled particles were measured for 2 min immediately before and 5 min, 30 min, 1 h, 2 h, and 6 h after inhalation.
The results showed that without aerosol treatment, particle number concentration varied among participants between 1 and 10,000/L exhaled air and also varied considerably within participants between the six measurements during each visit. The authors subdivided the results into high (n = 6) and low (n = 5) particle producers and found that saline delivery resulted in a statistically significant drop of particle emission among high producers and a tendency to increase emission among low producers. Administration of the surfactant simulant amplified particle emission by a factor of about five! No effects on particle size distribution were observed following administration of the saline or the surfactant simulant and the predominant particle size was 0.15–0.2 μm. The diminished particle emission after saline administration among high particle producers was explained by a presumed shift towards large particles outside the OPC’s range resulting in a substantial fraction of particles to deposit in the airways.
Comments: Subdivision into high and low particle producers is probably misleading. Recent studies showed that exhaled particles are distributed approximately log normally with no sign of two size modes [16, 17]. Considering all participants, there was no significant reduction of exhaled particles within the size range of the OPC. Administration of a surfactant simulant to the participants substantially increased particle emission, indicating that surface tension is important. The surfactant simulant with a surface tension of about 42 dyn/cm may in fact increase the surface tension of small airways, particularly at low lung volumes when the surface tension is normally close to zero. Increased surface tension increases particle production [18, 19]. Conclusions from the cough machine results are relevant to coughing and forced exhalation, but not to human tidal breathing. The cough machine generates particles by the burst of air destabilizing the mucus/air interface by shear forces to form submicron droplets.
Watanabe et al.  studied in vitro effects, particularly of isotonic sodium chloride, on the propensity of RTLF to form small droplets of various aerosolized formulations and widely varying surface tensions and viscoelastic properties. The main experiments were performed on the cough machine used by Edwards et al. , which was altered by reducing the applied air pressure from about 126 kPa to about 21 kPa to simulate a less violent breathing maneuver. The model measured particle production caused by simulated breathing over the mucus mimetic trachea after the various aerosolized formulations were applied. Particle production was assessed by an OPC covering particle sizes between 0.09 and > 0.5 μm.
Comments: The simulated breathing maneuver corresponds to a rather forceful expiration. Under these circumstances, the trachea model reveals a new mechanism: the RTLF/air interface may be stabilized by gelation of the mucus by salt water, making RTLF less prone to disintegrate into very small particles.
Exhaled particles during various breathing maneuvers
Exhaled particle number concentration (n/L)
in through the nose and out through the mouth
Comments: In agreement with Papineni and Rosenthal  the results show that normal mouth breathing generates particles but no generation mechanism is suggested. During vocalization, however, vibrating vocal cords and air passage through adducted epiglottis almost certainly produces exhaled particles.
Chao et al.  measured the droplet size distribution in close proximity to the mouth opening during coughing and speaking. A sample of 11 healthy volunteers under 30 years of age were asked to count loudly and slowly 10 times from 1 to 100. After a break, they coughed 50 times with lips closed before each cough.
Comment: This study is the first to measure the size interval from about 2 μm to 2000 μm with the same measuring system and with an experimental set-up optimized to measure particles unaffected by evaporation/condensation during speaking and coughing. The counts of particles in the largest size classes were very low but represented almost all the volume and mass. It is worth bearing in mind that the mass of one particle with a diameter of 150 μm corresponds to almost 6.6 million particles with a diameter of 0.8 μm assuming similar density and spherical shapes.
Airway reopening hypothesis
At about the same time as the study discussed above, several independent groups dealt with the notion that one important mechanism for particle generation is the reopening of closed airways [18, 22, 23, 24] – as previously posited by Edwards et al. . The fact that small peripheral airways normally close following a deep expiration was originally shown by Milic-Emili and coworkers in 1966–1968 [25, 26, 27] and elegantly confirmed by Burger and Macklem  and Engel et al. . In upright position, the apical parts of the lungs are more expanded than basal parts due to the weight of the lungs. During an expiration to low lung volumes, the basal airways collapse with the airway walls pasted together by RTLF and reopen again on inspiration. There is a simple single-breath test to determine the volume at which extensive airway closure (the closing volume) begins . To the best of our knowledge, the precise location of airway closure along the airway tree is not known in humans but is generally considered to be in the small airways. In dogs, airway closure appears to take place in airways with an internal diameter of 0.4–0.6 mm . Some airways may close at higher lung volumes than indicated by the closing volume , and massive airway closure may occur during tidal breathing  at low lung volumes (low functional residual capacity) as in people who are obese or whose closing volumes are increased by a disease such as COPD . Then there is a risk of mechanical injury of the small airways due to the cyclic closing and reopening .
Inspiring a normal breath volume via the nose and exhaling via the mouth.
Inspiring a normal breath volume via the mouth over a 3-s period, followed immediately by a 1-s full deep exhalation.
Rapid inspiration of a normal breath volume via the mouth, followed by holding the breath for 2, 3, 5, or 10 s and full deep exhalation over 3-s;
Inspiring a normal breath volume via the mouth over a 3-s period, followed immediately by a 3-s full deep exhalation.
Comments: Effects of deep exhalation confirm the airway-reopening hypothesis. Breath holding causes time-dependent preferential settling of larger particles in the airways, thereby preventing their exhalation.
A Hannover research group presented two parallel studies in 2010 [18, 35]. Schwarz et al.  measured exhaled particles, flow rates, and tidal volumes online during single breaths in 21 healthy participants aged 21 to 63 years. Spirometry and lung volumes were obtained. Particle concentrations and size distributions were measured online in a temperature-regulated box at 37 °C using a condensation nuclei counter and a laser spectrometer. Six diameter intervals were found ranging from 0.1 to > 5 μm. The protocol involved varying tidal volumes between 20 and 80% of the forced vital capacity. Tidal volumes < 0.7 L were disregarded because response times of the online measuring devices were too slow. One test assessed intra-participant variability through repeated breathing maneuvers after 2 h rest on the same day and during a second visit within 2 months.
Comments: The airway reopening hypothesis was challenged by effects of deep exhalation and the hypothesis was strengthened. The observed inter-individual variation was large, but was assessed by correlations.
The parallel study by Haslbeck et al.  investigated particle formation by rupture of surfactant films using computations in a fluid dynamics model. A simplified instrument comprising a biconcave cylinder < 0.5 mm in diameter modeled the small airway structure. A liquid film was applied with uniform thickness in the middle of a cylinder and blocked the cylinder passage. The model described the thinned circular film before rupture and the associated drop formation. The critical thickness of film rupture was 0.2 μm, allowing for computations of film rupture and drop formation as a function of the parameter’s surface tension (0.1–20 dyn/cm), viscosity, and density. The model did not consider the movements of the wall and the drop in pressure across the film. Particle emission was measured in 16 healthy participants in the same way as in the study by Schwarz et al. .
Comments: The results of the computational model of fluid dynamics simulating small airway opening are consistent with the airway reopening hypothesis. The effect of surface tension was later confirmed .
Comments: The airway reopening hypothesis is further strengthened. The breathing maneuvers described show that lung volumes where airway closure (and reopening) prevails generate the vast majority of the exhaled particles.
Comments: Tidal breathing emits a mode of extremely small particles, whose mass is negligible. The widely different size distributions resulting from the two breathing maneuvers and the lack of correlation between them suggest different sites of origin.
Johnson et al.  extended previous work [21, 23] and integrated results from the APS assessments of particles with diameters mainly from 0.7 to 20 μm and a droplet deposition analysis (DDA) covering diameter > 20 μm thus spanning a wide range of particle sizes. Their equipment was essentially the wind tunnel set-up described previously . Fifteen healthy participants < 35 years of age participated in the APS studies. Eight were included in the DDA after an oral rinse containing a food dye. As the number of exhaled large droplets was very low, participants had to cough 50 times to produce an adequate number of droplets of each size. The APS counts were corrected for evaporative and dilution effects. Combining DDA results and results from the APS after transformation onto a common scale produced a composite size distribution. Only average results were presented from all individuals due to very large inter- and intra-individual variation. The analysis applied the mixture model assuming log normal distributions.
Normal and deep tidal breathing resulted in the first mode with a count median diameter of 0.8 μm interpreted to have been caused by the airway reopening mechanism.
Speaking, unmodulated vocalization, and coughing resulted in the second mode with a count median diameter of about 1 μm interpreted to have been caused by vocal cord vibrations and aerosolization in the laryngeal region.
Speaking and coughing also resulted the third mode with DDA stain dots with a count median diameter of about 200 μm interpreted to have been produced in the presence of saliva, i.e., between the epiglottis and the lips.
Comments: In addition to previously identified size modes, there was a mode of large particles. This mode relates to particle generation in the upper respiratory tract, including the oral cavity.
Comments: One ought to consider the time-dependent generation of particles in small airways and the deposition of particles in the alveoli and airways when interpreting results or designing breathing maneuvers.
Comments: Unexplained large inter-subject variability remains.
Chemical evidence of origin
Chemical analysis of exhaled particles can shed light on the origin and mechanisms involved in formation process. Papineni and Rosenthal  used electron microscopy and an X-ray dispersive technique for the elemental analysis of droplet residues. They found significant content of potassium, calcium, and chlorine, all abundant in body fluids. Almstrand et al.  analyzed impacted particles using the surface-active mass spectrometry, TOF-SIMS. The analysis showed strong signals from phospholipids in all samples from four healthy participants. The identified phospholipid compound groups, such as phosphatidylcholine, phophatidylglycerol, and phosphatidylinositol, are known constituents of surfactants from analyses of bronchoalveolar lavage.
Comments: This procedure facilitate a normalization of the results and provides an estimate of the concentration in the RTLF of small airways.
The FRC reference maneuver: tidal breathing, inspiration to TLC, and exhalation into the measuring equipment. These results served as baseline values of particle formation.
The forced exhalation maneuver differed from the FRC reference maneuver only by the high expiratory flow rate, intended to result in exhaled particles produced during the exhalation.
A cough maneuver was included, known to generate a high amount of particles.
The airway reopening maneuver: expiration to RV before inspiration to TLC and exhalation into the equipment. This breathing maneuver induced high amounts of particles generated by the airway reopening mechanism.
Comments: Central airways as well as small airways generate exhaled particles but with different compositions.
Comments: It is confirmed that exhaled particles during tidal breathing include relatively large particles that dominates the exhaled mass and that is not associated to airway reopening.
Particles in exhaled breath – A potential biomarker of small airway disease?
Chemical analysis of exhaled particles provides huge possibilities to explore the biomarkers of small airway diseases, and we are probably just beginning to utilize this new biological matrix to its full extent.
Particle emission among patients with COPD appears unclear [39, 45]. Schwarz et al.  reported that there were no differences in particle number concentrations between healthy nonsmokers (n = 16) and COPD patients (n = 28). However, the COPD patients presented in their Fig. 2 clearly emit less particles than the healthy non-smokers do. Lärstad et al.  reported substantially reduced particle emission in COPD patients (n = 13) compared with healthy participants (n = 12). Despite some ambiguity, we consider the results to indicate that COPD patients exhale fewer particles than healthy participants. One reason may be that hyperinflation in the COPD patients prevents their ability to expire to low lung volumes. When healthy participants exhale to CP rather than to RV, their particle emissions were about one third of that at RV . Another reason may be that terminal bronchioles are destroyed in COPD,  resulting in fewer small airways to close and open. Furthermore, available airways may be injured due to the mechanical stress of cyclic closing and opening [3, 33], possibly affecting particle composition and production.
Comments: Proteomics from small airways offers an exciting potential to obtain a fingerprint from small airways.
Comments: Results consistent with the airway reopening hypothesis.
Patients with bronchiolitis obliterans syndrome (BOS) after lung transplants showed lower SP-A particle concentrations than the BOS free group of lung transplants .
Analysis by TOF-SIMS has shown that phospholipids of smokers are more protonated and sodiated than those of non-smokers . As the particles were produced by the airway opening mechanism, the results indicate the effects of smoking on the RTLF of small airways. Smokers also were studied by Schwarz et al.  but no effects on particle emission were found.
Various mechanisms generate particles appearing in exhaled breath. The sites of origin differ, depending on the breathing maneuver applied. The process of reopening small airways is one scientifically substantiated particle generation mechanism. Analyzing the content of exhaled particles as generated by the airway reopening mechanism, offers an exciting noninvasive way to obtain samples of RTLF from small airways. Results from a few early and small clinical studies on COPD, asthma and BOS indicate associations with altered particle formation and particle composition. In the future, analysis of exhaled particles may provide a “fingerprint” of small airways revealing important biomarkers.
The ChAMP (Centre for Allergy Research Highlights Asthma Markers of Phenotype) consortium which is funded by the Swedish Foundation for Strategic Research, the Karolinska Institutet, AstraZeneca & Science for Life Laboratory Joint Research Collaboration, and the Vårdal Foundation.
The Swedish Heart Lung Foundation.
Availability of data and materials
All data generated or analyzed during this study are included in this published article and the appropriate references.
BB wrote the first draft. EL wrote the Technical Considerations. PL, GL and A-CO contributed substantially to the final manuscript.
Ethics approval and consent to participate
Appropriate ethic authority approved the protocols of the referred recent publications.
Consent for publication
The authors are shareholders of PEXA® AB (www.PEXA.se) and A-C Olin is a board member.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 1.Horvath I, Barnes PJ, Loukides S, Sterk PJ, Hogman M, Olin AC, Amann A, Antus B, Baraldi E, Bikov A, et al. A European Respiratory Society technical standard: exhaled biomarkers in lung disease. Eur Respir J. 2017;49.Google Scholar
- 5.Olofson JBB, Bergman B, Ullman A, Svärdsudd K. Prediction of COPD and related events improvs by combining spirometry and the single breath nitrogen test. COPD: J Chron Obstruct Pulmon Dis. 2018 In press.Google Scholar
- 11.Ostling J, Van Geest M, Viklund E, Mirgorodskaya E, Olin A-C. Exploring particles from exhaled air as a new source for protein biomarkers from the airways. Eur Respir J 2017;50:Suppl 61, OA321.Google Scholar
- 17.Chao CYH, Wan MP, Morawska L, Johnson GR, Ristovski ZD, Hargreaves ZD, Mengersen K, Corbett S, Lif Y, Xie X, Katoshevski D. Characterization of expiration air jets and droplet size distributions immediately at the mouth opening. J Aerosol Sci. 2009;40:122–33.Google Scholar
- 25.Dollfuss RE, Milic-Emili T, Bates DV. Regional ventilation of the lung studied with boluses of 133 xenon. Resp Physiol. 1967;2.Google Scholar
- 36.Holmgren H, Ljungström E, Almstrand A-C, Bake B, Olin A-C. Size Distribution of Exhaled Particles in the Range from 0.01 to 2.0 μm; 2010.Google Scholar
- 41.Larsson P, Larstad M, Bake B, Hammar O, Bredberg A, Almstrand AC, Mirgorodskaya E, Olin AC. Exhaled particles as markers of small airway inflammation in subjects with asthma. Clin Physiol Funct Imaging. 2017;37:489–97.Google Scholar
- 44.Ljungkvist Göran US, Åsa T, Karina S, Björn B, Per L, Ann-Charlotte A, Emilia V, Oscar H, Sören S, Olof B, Anna-Carin O. Two techniques to sample non-volatiles in breath – exemplified by methadone. J Breath Res. 2017; In press.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.