Noninvasive methods for assessment of airway inflammation in occupational settings


  • Edited by: Jean Bousquet

Santiago Quirce, Hospital La Paz, Department of Allergy, P. Castellana 261, 28046 Madrid Spain. Tel: +34 91 727 70 80 Fax: +34 91 727 70 50 E-mail:


To cite this article: Quirce S, Lemière C, de Blay F, del Pozo V, Gerth Van Wijk R, Maestrelli P, Pauli G, Pignatti P, Raulf-Heimsoth M, Sastre J, Storaas T, Moscato G. Noninvasive methods for assessment of airway inflammation in occupational settings. Allergy 2010; 65: 445–458.


The present document is a consensus statement reached by a panel of experts on noninvasive methods for assessment of airway inflammation in the investigation of occupational respiratory diseases, such as occupational rhinitis, occupational asthma, and nonasthmatic eosinophilic bronchitis. Both the upper and the lower airway inflammation have been reviewed and appraised reinforcing the concept of ‘united airway disease’ in the occupational settings. The most widely used noninvasive methods to assess bronchial inflammation are covered: induced sputum, fractional exhaled nitric oxide (FeNO) concentration, and exhaled breath condensate. Nasal inflammation may be assessed by noninvasive approaches such as nasal cytology and nasal lavage, which provide information on different aspects of inflammatory processes (cellular vs mediators). Key messages and suggestions on the use of noninvasive methods for assessment of airway inflammation in the investigation and diagnosis of occupational airway diseases are issued.

Noninvasive methods for assessing airway inflammation are increasingly used in the investigation and management of asthma and rhinitis. In asthma, the analysis of cells and mediators in induced sputum has been widely applied for studying bronchial inflammation (1). The use of the fractional exhaled nitric oxide (FeNO) concentration as a surrogate marker of eosinophilic airway inflammation has been suggested (2), whereas exhaled breath condensate (EBC) is still under evaluation (3). Nasal inflammation may be assessed by noninvasive methods such as nasal cytology and nasal lavage (NAL), which provide information on different aspects of inflammatory processes (cellular vs mediators) (4).

Noninvasive methods for the evaluation of airway inflammation have also been used in the investigation of occupational respiratory diseases (5–8). However, there is currently no consensus on the role of new means that assess airway inflammation in the evaluation of occupational respiratory disease, such as FeNO and induced sputum (9). There is a need to increase the use and availability of these tests in the investigation of occupational asthma (OA) (10), nonasthmatic eosinophilic bronchitis (NAEB) (11), and occupational rhinitis (12). It should be considered, however, that subjective symptom scores and objective physiological measurements are not necessarily correlated with the severity of airway inflammation. This is probably because of the fact that neural regulation, hyperreactivity, and perception of severity of symptoms play important role in overall severity of the disease.

The main objective of this document, elaborated as a consensus statement, is to issue key messages and consensus suggestions on noninvasive methods for assessment of airway inflammation in the investigation and diagnosis of occupational airway diseases based on existing scientific evidence and the expertise of a panel of physicians coming from different countries. Both upper and lower airway inflammation have been reviewed and appraised reinforcing the concept of ‘united airway disease’ (4) in the occupational settings.

Induced sputum analysis

The aim of sputum induction is to collect an adequate sample of secretions from lower airways in subjects who do not produce sputum spontaneously. Inhalation of isotonic or hypertonic solutions administered by nebulization has been demonstrated to induce a small amount of airway secretion that can be expectorated and analyzed (13–15). Some investigators change concentration during the procedure, starting with 3% and subsequently increasing to 4% and 5%, with a cumulative duration of nebulization of 15–20 min. However, it seems that hypertonic saline 3% is as successful as 3–5% given sequentially (16). There is no difference in the cellular composition of sputum induced with either isotonic or hypertonic saline (17), and different saline concentrations do not affect total and differential cell counts (DCC) in induced sputum (16), eosinophil cationic protein (ECP), or histamine level.

Repeating sputum induction 8–24 h after an initial induction can cause an increase in neutrophil levels in the second sputum sample (18). An interval of 48 h between two inductions gave reproducible cell counts in normal subjects (19). However, there are many studies on OA in which induced sputum is performed 24 h after a baseline induction, showing reproducible results on neutrophil counts.

Pretreatment with a short-acting β2-agonist (e.g. salbutamol) is to be recommended as the standard procedure to prevent excessive bronchoconstriction (13, 15, 20). Salbutamol has no effect on sputum inflammatory cell percentages (16, 20). Methacholine challenge has no influence on sputum inflammatory cell counts (21).

Two methods for processing the expectorate have evolved. The first involves selecting all viscid or denser portions from the expectorated sample with the aid of an inverted microscope or by visual aspect (13, 22). The second approach involves processing the entire expectorate, comprising sputum plus variable amounts of saliva (23). The reproducibility of cell counts has been reported to be lower if squamous cell contamination represents >20% of all recovered cells (24). There is conflicting data as to whether or not DCC differ between the two methods.

It is recommended that sputum be processed as soon as possible or within 2 h to ensure optimum cell counting and staining. Complete homogenization is important and can be achieved by the use of dithiothreitol (DTT) or dithioerythritol (DTE) to break the disulphide bonds in mucin molecules, allowing cells to be released. Once the sputum cells have been obtained, the investigations are directed to obtain DCC, and the most common method used is the stain after cytocentrifugation method. The nonsquamous DCC are expressed as a percentage of the total nonsquamous cells. Several reports have studied induced sputum cell count in healthy adults (25–27). Table 1 summarizes data of DCC in induced sputum from healthy individuals.

Table 1.   Differential cell counts in induced sputum from healthy individuals (mean ± SD)
 % cells
Belda et al. (25)Spanevello et al. (26)Thomas et al. (27)
Neutrophils37.5 ± 20.127.3 ± 13.047.0 ± 7.0
Eosinophils0.4 ± 0.90.6 ± 0.80.3 ± 0.6
Macrophages58.8 ± 21.069.2 ± 13.049.0 ± 25.2
Lymphocytes1.0 ± 1.11.0 ± 1.21.0 ± 1.4
Epithelial cells1.6 ± 3.91.5 ± 1.82.5 ± 3.2

Induced sputum cells can be phenotypically analyzed by flow cytometry using monoclonal antibodies to select the single population of analysis (28, 29). DTT has no effect when sputum sample is processed for flow cytometry (28, 29). Even if present in small amount in sputum, lymphocytes are quite stable cells, and their phenotypic identification with the use of monoclonal antibodies anti-CD3, -CD4, -CD8, and -CD19 is accurate. Furthermore, the lymphocyte subset can be evaluated even on frozen sputum samples (30). Sputum T-cell activation has also been characterized in asthmatic patients through the expression of some membrane markers such as CD25, CD69, and CD103 (31). To evaluate Th1 or Th2 skewing of sputum lymphocytes, the expression of intracellular cytokines has been measured by flow cytometry (32). Mamessier et al. (33) demonstrated an increase in activated sputum T cells producing interferon (IFN)-γ and interleukin (IL)-13 after a specific inhalation challenge (SIC) in subjects with OA confirming the presence of a Th1/Th2 mixed population also in occupational setting. Flow cytometry is useful in determining cells usually present in small amounts in sputum as lymphocytes and basophils (34). However, reference values for sputum cell distribution specifically evaluated by flow cytometry are still lacking. The possibility of applying flow cytometry for the evaluation of airway inflammation in the occupational setting, besides the research field, depends on the identification of specific activation markers, on T cells, B cells, macrophages, eosinophils or neutrophils, useful and significant to highlight an inflammatory response induced after a SIC.

An array of mediators can be measured in induced sputum supernatant by using immunoassays (13, 14, 35). The mediators can reflect different aspects of airway inflammation (35) and remodeling including eosinophil activation (e.g. ECP), mast cell activation (e.g. tryptase), cytokine production (e.g. IL-5), and microvascular leakage (e.g. albumin and fibrinogen). An important issue is the possibility that induction itself or subsequent sputum processing activates airway inflammatory cells (36). In particular, the effect of interference by sulfhydryl-reducing reagents, such as DTE or DTT, is important (37). However, DTT does not appear to interfere with the assays of IL-5, histamine, immunoglobulin A, fibrinogen, albumin, or tryptase, or appreciably to interfere with immunocytochemical staining of granulocyte-macrophage colony stimulating factor, tumor necrosis factor-α, IL-8 or most lymphocyte surface markers; it may slightly decrease staining of EG2 and HLA-DR (35, 38), and ECP data are controversial (35, 39).

Eicosanoids are lipidic mediators whose levels are important in asthmatic status, PGE2 could be measured repeatable, and interference in the cysteinyl-leukotrienes assays by DTT is unlikely because concentrations were not significantly different in sputum treated with and without DTT (40). The use of a protein-rich milieu during the incubation of sputum with DTT has been demonstrated to reduce the detrimental effect of DTT both on cells and on soluble mediators (41).

Key messages:

  •  Reference values for sputum cells detected with flow cytometry are still lacking.
  •  There is a need to identify the relationships between cell activation markers (new or classic ones) and the inflammatory pattern arising from exposure to specific causative agents to benefit from the of use flow cytometry in sputum cell evaluation.

Use of induced sputum and exhaled nitric oxide in the investigation of OA and related conditions

Changes in sputum cell counts during specific inhalation challenges

SIC remain the reference tests for diagnosing OA. However, those tests are sometimes difficult to interpret, especially when the patients are unable to perform reliable spirometric maneuvers. The addition of an objective measure to SIC is therefore likely to improve the diagnosis of OA. Furthermore, analysis of sputum cells may be useful in the investigation of the effects of occupational agents upon experimental exposures because it provides direct information on type, intensity, and time course of airway inflammation. The occurrence of an asthmatic reaction during allergen inhalation challenges or during asthma exacerbations is most of the time accompanied by an increase in sputum eosinophil counts (42). However, occupational agents differ from common inhalants because many of them are chemicals with irritant properties. Inhaling these agents may induce a different type of airway inflammation compared with common inhalants.

Changes in sputum cell counts after exposure to high and low molecular weight agents in subjects with occupational asthma

High molecular weight agents are proteins that usually induce an immediate or a dual asthmatic reaction. Similar to the airway inflammation found after common allergen inhalation challenges, an increase in sputum eosinophils has been observed after SIC to a number of high molecular weight agents: cereals (6), oil seed rape flour (43), Lathyrus sativus flour (44), Lepidoglyphus destructor (45), spores of Pleurotus ostreatus (46), or tampico fibers found in agave leaves (47). The exposure to low molecular weight agents, which are chemicals that often induce delayed asthmatic reactions, can also induce sputum eosinophilia. For example, isocyanates (5), acrylates (48, 49), red cedar (50), exotic woods (51), persulfate (52), manganese (53), or styrene (34) have been shown to induce an increase in sputum eosinophil counts. Exposure to manganese (53) and styrene (34) can also induce an increase in sputum basophils counts. Overexpression of LTC4, relative underproduction of PGE2, and greater eosinophilia in induced sputum was observed in patients with positive SIC to high or low molecular weight occupational agents 24 h after the challenge (54).

A neutrophilic airway inflammation has been reported after exposure to isocyanates (55, 56). The factors that influence the type of airway inflammation induced by the exposure to occupational agents are unclear. The concentration and the length of exposure to these agents may play a role (55). Few studies have looked at the impact of exposure to occupational agents in healthy subjects. Exposure to isocyanates (5), acrylates or flour (6) does not seem to induce eosinophilic or neutrophilic inflammation in healthy subjects, but a neutrophilic inflammatory response has been found in the induced sputum of healthy subjects after short-term exposure to irritant agents such as ozone (57), diesel exhaust (58), and endotoxin (59).

Interpretation of changes in sputum cell counts during specific inhalation challenges

Using sputum DCC during the investigation of OA can improve its diagnosis by bringing an additional objective measure to this investigation. Although the reliability of FEV1 and PC20 can be affected by inadequate spirometric maneuvers, the presence or the absence of airway inflammation cannot be affected by improper maneuvers during sputum collection.

The best timing for the collection of induced sputum with respect to the exposure to occupational agents is likely to be 7–24 h after exposure. Indeed, an increase in sputum eosinophils has been shown to occur 7 h after exposure to occupational agents and persist 24 h after exposure (50). It should be considered that the time course of sputum influx may be different when other cell types are assessed.

The magnitude of increase in sputum eosinophil counts occurring after exposure to occupational agents that should be regarded as clinically significant is not clearly established. An increase in absolute eosinophil counts of 0.26 × 106/ml compared with baseline values yields a sensitivity of 82% and a specificity of 91.7% for predicting a 20% fall in FEV1 (6). A 2% cutoff increase  in sputum eosinophil counts after SIC is considered to be the most discriminant value associated with an asthmatic response (6–8).

The increase in sputum eosinophil counts tends to occur with levels of experimental exposures lower than those necessary to elicit the asthmatic reactions induced by high and low molecular weight agents (60). Vandenplas et al. (61) showed that an increase in sputum eosinophil counts greater than 3% after the first day of exposure during SIC was the most accurate parameter for predicting the development of an asthmatic response on subsequent exposures with a sensitivity of 67% and a specificity of 97%. Therefore, an important increase in the sputum eosinophil counts in the absence of an FEV1 fall should incite pursuing the investigation (9, 11).

The lack of increase in sputum eosinophil counts after exposure to occupational agents should not rule out the diagnosis of OA. Indeed, some subjects can experience a 20% fall in FEV1 without showing sputum eosinophilia (50), whereas others can experience a 20% fall in FEV1 accompanied by an important increase in airway inflammation without airway hyperresponsiveness to methacholine (62, 63). Interfering factors that can modify the sputum cell response should be considered in the interpretation. Treatment with corticosteroids may blunt eosinophil influx, endotoxin contamination may favor sputum neutrophilia, and relatively high levels of exposure to chemicals may produce irritant effects.

In conclusion, sputum cell counts bring an additional objective measure in the investigation by SIC. Further studies are needed to improve the interpretation of the changes in sputum cell counts occurring after exposure to occupational agents.

Changes in induced sputum in response to various work exposures

The investigation of OA can be made by performing SIC in the laboratory or by monitoring the functional and inflammatory changes during periods at and away from the workplace. The majority of studies have described the changes in sputum cell counts observed in response to occupational exposures during the performance of SIC in the laboratory. However, during these tests, the workers are exposed to a single specific substance, which probably differs from the workplace where the workers are exposed to multiple agents often combining irritants and sensitizing properties.

The studies that have assessed the changes in induced sputum at the workplace are scarce. One of the first studies that investigated the changes in sputum cell counts between periods at and away from the workplace studied subjects with OA and asthmatics without OA working in the same environment (7). The workers were mostly exposed to low molecular weight agents. The diagnosis of OA was made if their asthma symptoms were worse at work and if there was either a FEV1 fall ≥20% or a fourfold change or more in PC20 between periods at and away from work. Sputum induction was performed at the end of periods at work and away from work. The subjects with OA had a large increase in sputum ECP when at work, which resolved when they were removed from their workplace. According to the inclusion criteria, the study population comprised subjects with OA who had very large changes in FEV1 and PC20 between periods at and away from work. However, this population differs from the majority of subjects with OA who do not show such large functional changes between periods at and off work, especially if they are treated with inhaled corticosteroids and long-acting beta2 agonists. Therefore, such large changes in sputum eosinophil counts may not be always observed in clinical practice. Another study looked at the changes in sputum cell counts in subjects suspected to have OA because of high and low molecular weight agents, whose diagnosis was subsequently confirmed by SIC (8). When at work, the subjects with OA had a significant increase in sputum eosinophils, whereas the group with negative SIC had higher neutrophil counts compared with the periods away from work. The changes in sputum eosinophil counts were smaller than those observed in the previous study (7). Anees et al. (64) examined the changes in induced sputum in subjects with OA because of low molecular weight agents while working. No comparison with periods away from work was made. Thirty-eight workers were investigated. Only 14 had sputum eosinophils greater than 2.2% when at work. However, the diagnosis of OA was not based upon the same criteria for all subjects. The authors reported that the workers had a sputum neutrophilia (59% of neutrophils). However, the important variability in sputum neutrophil counts makes difficult to consider this sputum neutrophil count as a significant increase from a normal count without having a comparison between periods at and away from work. Indeed, the normal values of neutrophils reported in healthy subjects vary from 27.3 ± 13.0 to 47.7 ± 7.0% (65).

Another case of sputum neutrophilia was reported in a worker exposed to metal working fluid who had a marked increase in neutrophils when at work (82%), which resolved after periods away from work (56%) (66). The sputum findings were mirrored by corresponding changes in spirometry and PC20 methacholine.

In conclusion, although subjects with OA seem to show predominantly an eosinophilic airway inflammation when at work, neutrophilic inflammation has been also described. The determinants leading to an eosinophilic or a neutrophilic response are poorly understood. Further research is needed to investigate whether or not the type of inflammation observed is related to the OA severity or prognosis.

Key messages:

  •  The majority of subjects with OA show an eosinophilic airway inflammation after exposure to occupational agents during SIC.
  •  An increase in sputum eosinophil counts greater than 3% after SIC often precedes the occurrence of functional changes on subsequente exposures.
  •  An isolated increase in sputum eosinophil count of at least 2% without functional changes should incite to pursue the investigation by increasing the duration of exposure.

Use of FeNO in the investigation of occupational asthma

Nitric oxide (NO) is produced in the respiratory tract by activation of NO synthase in various cell types and is detectable in exhaled air. Fractional concentrations of exhaled NO (FeNO) can be measured online with fixed or portable instruments, or exhaled air can be collected for offline FeNO measurements. Recommendations for standardized procedures of measurement of FeNO have been published by the European Respiratory Society (ERS) and the American Thoracic Society (ATS) (2).

FeNO is elevated in untreated asthma and falls after corticosteroid treatment. Although correlations between FeNO levels and percentages of eosinophils in induced sputum have been consistently demonstrated, in subjects treated with inhaled corticosteroids and in severe asthma, the correlation between FeNO and sputum eosinophils appears to be poor (67). Compared with induced sputum, assessment of FeNO is totally noninvasive, quick, and relatively simple to perform. However, elevated FeNO is not specific for asthma and eosinophilic inflammation because it has been found in other diseases and several conditions may influence exhaled NO (2).

Some studies examined the usefulness of FeNO in the investigation of OA, but with inconsistent results (50, 68–77). No significant changes in FeNO were observed in asthmatic reactions induced by western red cedar (50). Only patients with low basal FeNO showed an elevation of FeNO after a significant bronchoconstriction induced by various occupational agents in the work of Piipari et al. (72). In subjects exposed to isocyanates with respiratory symptoms, those with a positive SIC exhibited a greater increase in FeNO than subjects with negative challenge (73). However, a remarkable proportion (28%) of SIC-negative subjects showed an increase in FeNO >50% of baseline values. Increase in FeNO was observed in the majority of subjects who had an asthmatic reaction after SIC with latex, but also in 35% of subjects with latex-induced rhinitis (68). In contrast, no significant relationship between FeNO and workplace exposure was detected in subjects with self-reported symptoms to latex (77). FeNO in subjects sensitized to lupin in the workplace was not different from that measured in nonsensitized subjects (76). Exposure to laboratory animals tended to increase FeNO in sensitized workers (74, 75). Higher concentrations of FeNO were detected in nonsmoking aluminium potroom workers with asthma-like symptoms, but not in symptomatic smokers (70). A recent study showed that isocyanate-induced asthmatic reactions were associated with a consistent increase in FeNO which was maximal at 48 h postexposure, whereas FeNO did not vary with isocyanate exposure in occupational rhinitis and in nonsensitized subjects (78).

A study carried out in farmers, bakers, and health care workers showed increased FeNO levels only 24 h after SIC, along with a rise in the proportion of eosinophils in induced sputum and in NAL fluid in the cases with diagnosed OA. A significant correlation was found between FeNO level at 24 h after SIC and the percentage of eosinophils in nasal fluids before and 4 and 24 h after SIC, as well as in sputum before and 24 h after SIC in subjects with diagnosed OA (71).

Concentrations of FeNO were shown to decrease in farmers with OA after an educational intervention aiming to decrease their level of exposure to the offending agent (79).

There are some issues that should be considered in the interpretation of the conflicting results obtained by the studies which analyzed FeNO after SIC with occupational agents. One is the insufficient duration of monitoring of patients. Indeed, maximum increase in FeNO occurred 48 h after allergen challenge with D. pteronyssinus in atopic subjects with dual asthmatic responses (80), while the last measurement of FeNO in occupational studies was obtained 20–24 h after challenge (50, 68, 69, 72, 73). Secondly, corticosteroids inhibit the induction of NO synthase, and FeNO falls after treatment with oral or inhaled corticosteroids in subjects with asthma (2). In the studies which included patients on steroid treatment at the time of the test, FeNO response might have been blunted (50, 68, 73). Finally, an increase in NO production in the presence of bronchoconstriction might have been underestimated. Because FeNO is measured by a constant mouth flow, the reduction in the volume of conducting airways as a consequence of brochoconstriction will lead to an increase in airflow within the conducting airways. When velocities are increased, the exhaled gas has less residence time in the airways, and thus less time for the airway epithelium and inflammatory cells to load the bolus of expirate with NO, resulting in lower exhaled NO concentrations (81). This explanation is consistent with the observation that airway constriction induced by histamine challenge is associated briefly with a reduction in exhaled NO levels (82).

Key messages:

  •  Although the measurement of FeNO has some advantages over the analysis of induced sputum in OA, the interpretation of increased FeNO is more difficult than sputum DCC because it is less specific and several confounding factors may influence the results.
  •  Several investigations of FeNO using SIC gave conflicting results. Studies in the workplace/natural setting are limited, and prospective studies are not available.

Special considerations on nonasthmatic eosinophilic bronchitis

NAEB was first described by Gibson et al. in 1989 (83), and it is now considered a relatively common cause of chronic cough (84). This disorder is characterized by the presence of eosinophilic airway inflammation, similar to that seen in asthma. However, in contrast to asthma, NAEB is not associated with variable airflow limitation or airway hyperresponsiveness (84). The differences in airway physiology are related to the differences in the localization of mast cells within the airway wall, with airway smooth muscle infiltration occurring in patients with asthma (85). It has also been found that the concentrations of PGE2 in induced sputum are significantly higher in patients with NAEB than in patients with asthma (29).

The usefulness of FeNO measurements in the diagnosis of NAEB has been examined in a few papers. Oh et al. (86) have reported the role of FeNO for the investigation of chronic cough, especially of NAEB. The FeNO and induced sputum eosinophils were significantly higher in the asthma and NAEB groups than those in the other groups. FeNO levels were significantly correlated with induced sputum eosinophils in the asthma and NAEB groups. In the nonasthmatic groups, the sensitivity and specificity of FeNO for detecting NAEB, using 31.7 ppb as the FeNO cutoff point, were 86% and 76%, respectively. The positive and negative predictive values were 47% and 95%, respectively. Berlyne et al. (87) measured FeNO in NAEB patients, irrespective of the use or nonuse of corticosteroid therapy, and reported elevated FeNO levels that were significantly higher than those observed in patients with asthma. Brightling et al. (88) found significantly higher FeNO levels in subjects with NAEB or asthma than in normal controls. Sato et al. (89), however, have reported that patients with NAEB (sensitized to cedar pollen) do not show increased FeNO levels.

NAEB may arise from exposure to occupational agents, and this condition has been labeled occupational eosinophilic bronchitis (90). In fact, NAEB can be regarded as a variant syndrome of OA when it develops as a consequence of work exposures, and work-related changes in sputum eosinophil counts are significant and reproducible (11, 91). Exposure to several occupational allergens or sensitizers has been shown to induce occupational NAEB: acrylates (90), egg lysozyme (11), epoxy resin hardener (92), latex (93), mushroom spores (94), welding fumes, formaldehyde (95), chloramine (96), isocyanates, wheat flour (97), and fungal α-amylase (98). Diagnostic criteria for occupational NAEB have been published (Table 2) (11).

Table 2.   Diagnostic criteria for occupational eosinophilic bronchitis (adapted from ref. 11)
Isolated persistent cough (lasting more than 3 weeks) that worsensat work and improves away from work
Sputum eosinophilia ≥ 3% in sputum
Increases in sputum eosinophils are related to exposure to theoffending agent (either at work or after specific inhalation challengein the laboratory)
Spirometric parameters are normal and are not significantlyaffected by exposure to the offending agent
Absence of bronchial hyperresponsiveness to methacholine both atwork and away from work
Other causes of chronic cough are ruled out

The cough of patients with NAEB usually responds well to treatment with inhaled corticosteroids, but the dose and duration of treatment differ between patients (83, 84). For patients with occupational NAEB, avoidance of the causal allergen or occupational sensitizer is the best treatment (84). The condition can be transient, episodic, or persistent unless treated. Although it has been reported that patients with NAEB may develop asthma or progressive chronic airflow limitation, the available data suggest that the most likely outcome is that NAEB usually turns into a persistent condition (99).

Key messages:

  •  NAEB may arise from exposure to occupational agents, and it is characterized by persistent cough and sputum eosinophilia that are work-related.
  •  FeNO measurement may be useful as part of the initial evaluation for chronic cough, especially for the exclusion of NAEB.
  •  Although the cough and airway eosinophilia usually respond well to inhaled corticosteroids, avoidance of the causative occupational allergen or sensitizer is the best treatment

Exhaled breath condensate

For evaluation of airway inflammation, the collection of EBC is a useful and noninvasive method (3). A wide range of volatile substances in gas phase and nonvolatile compounds from the respiratory tract can be collected in condensed water during the sampling without affecting airway function or inflammation. EBC can be sampled from individuals on multiple occasions, with robust and easy to handle condensing devices, allowing monitoring the time course of an inflammatory response as well as the response to pharmacological therapy and follow-up in longitudinal studies. One indication is epidemiological surveys or screening for work-related lung alterations. Another approach is to use compounds of EBC as a tool to verify symptoms claimed to result of environmental exposure. Therefore, low priced and easy to handle equipment suitable for field studies are preferred.

Recommendations for EBC collection and the potential pitfalls of the technique are summarized in the ATS/ERS Task Force report (100). The fact that a variety of condenser designs are used limits the comparison of the results of different studies, and it has to be considered that the EBC collection device and collecting circumstances are potential confounding factors. Sputum induction seems to modify the levels of biomarker in EBC, therefore it is recommended to collect EBC before the assessment of the inflammatory response caused by the induction of sputum (101).

Measurements of biomarkers of effect in EBC are suggested as a way of exploring adverse effects in the context of air pollution exposure. In asymptomatic welders, e.g., increased concentrations of hydrogen peroxide in EBC were detected independent of the different metal fumes and gases generated according to material and method used for the welding process (102). The inflammatory status of the airways was modulated by the exposure profile, and in welders exposed to cadmium, chromium, iron, lead, and nickel, EBC pH was lower than in welders processing aluminium and iron at the workplace. Ferrazzoni et al. (78) demonstrated that isocyanate-induced asthmatic reactions are not associated with acidification of EBC. Aside investigating local inflammatory effects, EBC in this context might simply be useful in monitoring the exposure to occupational toxic elements. A good correlation between chromium levels and biomarkers of oxidative stress (H2O2, malondialdehyde) could be observed in EBC of otherwise asymptomatic chrome-plating workers (103).

Nevertheless, there is an urgent need for the standardization of the collection technique as well as the assessment and the evaluation of the already existing biomarkers (in parallel with the assessment of new mediators in different disease states) to establish normal values of biomarkers. Such research followed by cross-day variation studies, longitudinal follow-ups, and clinical trials using EBC biomarker analysis may lead to the introduction of EBC in clinical practice and into the routine use for the diagnosis of work-related respiratory diseases.

Key messages:

  •  EBC collection is a noninvasive and repeatable tool, and the EBC analysis can reflect oxidative stress, acidification, and inflammation in the airways.
  •  EBC analysis may be useful in occupational studies on a group level (using the same method) and in individuals when serving as their own controls.
  •  Based on methodological limitations, lack of standardization and difficulities in the interpretation of the data (e.g. calculation of several confounding factors), EBC collection and analysis are foremost research tools and not yet suitable for the clinical diagnosis setting.

Assessment of nasal inflammation

The verification of nasal inflammation is a key aspect of diagnosing rhinitis, but how to do it objectively in the individual is still a difficult task, in occupational rhinitis, as in other types of rhinitis (12). Symptom scoring is used in nasal provocation testing (104). Indirectly, the effect of nasal inflammation may be assessed by measurement of nasal patency (105). However, subjective symptom scores and nasal patency measurements are not necessarily correlated with severity of nasal inflammation; so, they may be related to disease severity independently of inflammation.

Nasal mucosal blood flow may be investigated by laser Doppler technique or Xenon wash-out methods (106). In this section, the main focus will be on nasal NO (nNO), NAL methods, and inflammatory markers, but also nasal cytology will be appraised.

Nasal nitric oxide and occupational rhinitis

In patients with allergic rhinitis, increased nNO levels have been demonstrated (107). Enhanced inductible NOS (iNOS) expression within the nasal epithelium may generate these levels. Enhanced iNos expression is considered as a consequence of persistent nasal inflammation. Paranasal sinuses—more than nasal epithelium—are the source of high levels of nNO. In sinuses, NO is continuously produced at levels of 25–30 ppm (108).

If nNO can be considered as a valid marker for nasal inflammation, assessment of NO might be helpful in characterizing patients with work-related rhinitis. Patients with work-related rhinitis might be distinguished from colleagues without disease. Ideally, fluctuations in NO should be associated with variations in exposure to occupational allergens of other stimuli. Finally, NO might be a tool to assess the outcome of a nasal challenge test or a provocation at the workplace. However, before accepting measurement of nNO as a tool to evaluate work-related disorders of the upper airways, nNO measurement should be validated in well-characterized conditions such as allergic rhinitis.

Standard operation procedures have been established to measure NO in both upper and lower airways (2). Normal levels of nNO may range from 400 to 900 ppb. High levels may reflect nasal inflammation, whereas low levels may be seen in conditions such as nasal blockage, polyps, and cystic fibrosis. Levels below 105 ppb may be predictive for primary ciliary discinesia (108).

The level of nNO measured in allergic rhinitis may be increased by nasal inflammation. However, nasal blockage and secretions will occlude the ostia of the paranasal sinuses thereby lowering nNO levels. These opposite phenomena may explain why in several studies nNO is increased in allergic rhinitis (109), whereas in other studies, no differences in nNO levels have been seen between patients with allergic rhinitis and healthy subjects (110). Also, the high background level of nNO may mask small fluctuations or increase. The effect of nasal blockage also explains why nNO decreases during the acute phase of the nasal challenge (111).

Only a few studies focus on determination of nNO in occupational allergy. In one study among laboratory workers, it was shown that exhaled NO was raised in those with laboratory animal allergy symptoms compared with asymptomatic subjects (75). A trend of increased NO by allergic status was observed; asymptomatic, to early laboratory animal allergy, to asthma. Symptomatic subjects also had raised nNO vs asymptomatic subjects (mean difference 378 ppb, < 0.05) (75). In two studies, nNo was measured in NAL instead of exhaled air. In a study among paper-mill workers, there was no statistically significant relationship between nNO concentration and nitrate in NAL fluid or nasal symptoms (112). In another study, the association between the development of rhinitis reactions during workplace-related challenge tests with latex and nasal allergic inflammation was studied. The NO derivative concentrations in NAL fluid were significantly increased 6 h postchallenge compared with the prechallenge values (113).

In conclusion, a few studies suggest that measurement of nNO might be helpful in the diagnostic workup of occupational allergy. However, at this stage, determination of nNO in patients with rhinitis is an experimental technique hampered by the opposite effects arising from nasal blockage and the level of nasal inflammation. Not withstanding, nNO measurements could be used as objective (intraindividual) measure of nasal inflammation in selected patients with occupational rhinitis after excluding those with rhinosinusitis, nasal polyps, ciliary dyskinesia, and other factors that may increase or decrease nNO levels.

Nasal lavage methods

NAL has been extensively used in experimental/laboratory research to elucidate the luminal cell recruitment, cell activation, and plasma protein extravasation in the nasal mucosa, not only under natural challenge conditions, but also when using a wide range of different stimuli. The method has been less used under clinical and epidemiologic circumstances, and the sources of variability and the repeatability of the findings are poorly substantiated under field conditions (114).

Two main methods are being used; a ‘head-back’ with the palate closing off the nasopharynx, or a ‘head-forward’ method, or modifications of these (115, 116). Both methods are very well tolerated. The first (head-back) is quickly performed, relying on the co-operation between the investigator and the subject, and where the crucial point is whether the subject manages to close off the nasopharynx (117). The latter method with the use of the nasal pool device has the advantage that a group may be investigated at the same time under guidance of the investigator, but where some will have problems keeping the pressure on the nasal pool container throughout the 5 min. Modifications of a ‘head-forward’/nasal pool technique have replaced the compressible container with a syringe connected to a rubber tube or Foley catheter and equipped with an inflatable balloon serving as a nasal adapter (118). There are great differences in dwelling times of the fluid in the nasal cavities in all the different methods used.

In an experimental model, Belda et al. (119) compared the two methods ‘head-back’/Naclerio and ‘head-forward’/Greiff-Grünberg. The modified Greiff-Grünberg method gave higher and more repeatable total cell counts and, in subjects with rhinitis, more reproducible ECP levels compared with the Naclerio method. Both methods were able to discriminate between healthy and rhinitics.

The NAL fluid once collected has to be kept cold. After measuring volume/weight, the lavage fluid is centrifuged between 400 and 1000 g for 5–20 min. The sediment or cell pellet may be used for cytology, see later. The supernatant is stored as aliquots of 0.5–1.0 ml at −20 to −70 °C until assay is performed (114).

Nasal lavage and inflammatory markers

Many different mediators, cytokines, and chemokines have been measured in NAL studies in occupational settings (Table 3). Which one of the many inflammatory markers to use will partly depend on the purpose of the investigation; research, nasal provocation testing, detecting health effects of occupational exposures, or monitoring the effect of interventions.

Table 3.   Inflammatory markers used in nasal lavage in occupational settings
Authors (ref.)Inflammatory markersStudy designOccupation/subjectsExposureNumber of subjects
  1. α2M, α2-macroglobuline; ECP, eosinophil cationic protein; PMN, polymorphonuclear cells; TAME, N-alpha-tosyl-L-arginine-methyl esterase; MPO, myeloperoxidase; NO, nitric oxide derivative; sICAM, soluble intercellular adhesion molecule-1; TNF-α, tumor necrosis factor; PGE2/PGF/PGD2, prostaglandins; LTB4/peptide LT, Leukotriens; GSH, reduced glutathione; GSSG, oxidized glutathione; IFN-γ, interferon-γ.

Nielsen et al., 1994 (134)Tryptase, ECP, TAMECase–controlWeapon industry?Acid anhydrides43 + 27
Hauser et al., 1995 (126)PMN cell countsField studyBoilermakers, utility workersFuel oil ash (Vanadium)37 (49)
Åhman et al., 1995 (135)Albumin, tryptase, ECP, cellsCase–controlIndustrial arts teachersWood dust24 + 24
Gorski et al., 1998 (125)ECP, tryptase, albumin, total proteinCase–control challengeBakersFlour100 + 40
Raulf-Heimsoth et al., 2000 (113)ECP, NO derivative, IL-5, IL-8, tryptase, sICAM, cellsCase–control challengeHealth care workers and physiciansLatex32 + 6
Hellgren et al., 2001 (136)IL-8Case–controlSoft paper millersSoft paper dust37 + 36
Palczynski et al., 2001 (137)Tryptase, ECP, albuminCase–control challengeHealth care workersGlutaralde-hyde21 + 10
Lund et al., 2002 (138)ECP, MPO, IL-6, IL-8, TNF-α, PGE2,PGF, PGD2, LTB4, peptide LT, GSH, GSSG, uric acid, total proteinExperimental challengeHealthy volunteersHydrogen Fluoride10
Littorin et al., 2002 (139)Albumin, ECP, MPO, cellsCase–controlAutomobile factoryIsocyanates (heating polyurethane)38 + 9
Larsson et al., 2002 (140)IL-6, IL-8, cellsCase–controlSwine farmingDust aerosols, endotoxin7 + 9
Palczynski et al., 2003 (141)Tryptase, ECP, albuminCase–controlHealth care workersChloramine13 + 6
Priha et al., 2004 (142)IL-6, IL-8, TNF-α, IFN-γCase–controlFurniture makersMedium-density fiber/wood-dust45 + 15
Tuomainen et al., 2006 (143)IL-4, IL-6, IL-12, TNF-α, NOExperimental challengeAsthmatics/NonasthmaticsDegraded PVC products10 (5 + 5)
Raulf-Heimsoth et al., 2007 (144)IL-1β, IL-5, IL-6, IL-8, TNF-α, NO, albumin, total proteinCross-sectional cross-shift case–controlAsphalt workersBitumen74 + 49
Storaas et al., 2007 (120)α2M, ECPCross-sectional fieldBakersFlour183
Diab et al., 2008 (145)AlbuminCase–control challengeHairdressersPersulfate29 + 12
Bakke et al., 2008 (146)ECP, MPO, lysozyme, albuminCross-sectional fieldUniversity staffIndoor air173
Gaughan et al., 2008 (147)ECP, MPO, albuminCohort follow-upWildland firefightersAldehydes, respirable particles, CO58
Holmstrøm et al., 2008 (148)ECP, MPO, tryptase, albuminCase–controlFarmers (milk, grain producers, swine) 53 + 15
Krakowiak et al., 2008 (149)IL-18Case–control challengeBakersFlour19 + 9

ECP is one of the most studied markers of inflammation, also in terms of repeatability (see later), and may serve as a general marker of mucosal inflammation, both in processes of eosinophil and neutrophil activation, and regardless of whether the subjects are atopics or nonatopics (114, 120, 121). A key feature of mucosal inflammation is the exudation of plasma proteins such as albumin, α2-macroglobulin, and others, which can be monitored by analysis of plasma proteins in nasal lavages (122). The measurement of a marker of this process should provide a single integrated measurement of inflammation that reflects the underlying tissue processes (114).

Histamine is rapidly metabolized by histaminases and N-methyl transferase, thus the mast cell degranulation products tryptase or the prostaglandin PGD2 are recommended as markers of mast cell activation (123). The most used marker of neutrophil activation is the myeloperoxidase (124).

Nasal and mucosal responsiveness/reactivity

Nasal hyperresponsiveness may be seen as one of the characteristics of nasal inflammation. The method of monitoring by nasal lavages the ability of histamine challenge to produce plasma exudation gives the opportunity to verify nasal exudative hyperresponsiveness, as demonstrated in bakers (120, 125).

Repeatability and validity of nasal lavage

The response after histamine or allergen challenge has been found reproducible when measured as N-alpha-tosyl-L-arginine-methyl esterase (TAME) esterase, albumin, or ECP in NAL (114). The intrasubject variance compared with the intersubject variance in ECP levels in NAL has been shown to differ whether the study subjects are children, healthy volunteers or with rhinitis, short-term or follow-up over a longer period, and with gender (119, 125). Some authors have proposed cutoff values defining a significant change in eosinophil counts in NAL (117, 125). The short-term reproducibility of cell counts in NAL has revealed satisfactory intrasubject variance (125, 126).

There have been attempts to define norm values for nasal mucosal ECP, and other methods of collecting mucosal outputs than NAL have been advocated. The problem especially with the dilution effect in NAL has been addressed (127), which may be solved by adding an exogenous dilution marker to the washing fluid, such as inulin, radioactive albumin, and lithium chloride. Other factors may also affect levels of cellular markers and mediators in NAL, like steroid treatment for eosinophil activation markers, or other treatments and environmental conditions (127) that are beyond the scope of this paper.

Nasal lavage used in field studies

NAL has most often been used in an experimental design in research purposes, or as part of the response parameters in provocation testing. The NAL method is far less applied as a means of objectively monitoring nasal disease in ‘natural’ exposure settings. Douwes et al. (128) did a field study with pre and postshift NAL (‘head-back’) in workers of a compost plant visited on two occasions 1 year apart. In a field study in Bergen, Norway, the exposure and NAL (‘head-forward’) data suggested a dose-response relation between exposure to flour dust and plasma exudation/ECP levels in the nasal mucosa (120). Other of the few studies performed at the work place include the study by Granstrand et al. (129) on wood-surface coating industry workers using a ‘head-back’ method. It is difficult to find studies comparing the levels of indices of inflammation at work, and when the subjects have had a longer period away from work.

In conclusion, NAL is well tolerated, rather simple, and rapid to perform. NAL is a useful method for the detection of nasal inflammation in occupational settings where comparison can be made using test subjects as their own controls and may be used to confirm the diagnosis of occupational rhinitis in challenge testing. The lack of norm values and standardization makes the method less useful to single out the diseased individual from a group. On a group level, NAL may be useful monitoring effects of exposure and interventions and is an excellent research tool.

Nasal cytology

Several techniques have been elaborated to harvest nasal cells. When doing NAL, the cell pellet portion may be used, and at least eosinophils are readily identified (114, 115). Secretions can be blown onto paper or plastic wrap, and then placed on a glass slide. The processing of nasal blown secretions to solubilize mucus, as performed for sputum, has recently been demonstrated to allow reproducible data, at least for nasal eosinophils (130). The disadvantages with these techniques are that the cells originate only from the secretions, and do not necessarily reflect the present patophysiologic processes in the epithelium, and some subjects will have problems with blowing the nose effectively enough. With a plastic curette scraping the surface in the middle-third of the inferior turbinate, a nasal specimen of both the secretions and the surface epithelium may be easily obtained (131). Nasal brushing utilizes a small plastic-coated steel wire brush with nylon bristles, and is placed between the septum and the inferior turbinate, and rotated while being removed (132). Both scraping and brushing methods have the advantage of adequacy of specimen but may cause a slight irritation.

Meltzer in the review by Howarth et al. (114) has detailed the most used processing methods and has provided guides to the grading and interpreting of the nasal cytograms. A study by Raulf-Heimsoth et al. (113) demonstrates the possibility of combining the NAL (for mediator analysis) and brushing (for cytograms) methods in an occupational setting, each method revealing different aspects of the inflammatory process. The sole use of nasal brushing has been used in a 5 years follow-up of Swiss customs officers exposed to diesel engine emission (133).

Pignatti et al. (130) have recently shown that the evaluation of nasal blown secretions in occupational setting might be useful in monitoring eosinophilic inflammation after a specific nasal provocation test. These authors suggest a 4% and/or 1 × 104 eosinophils/ml cutoff for a significant postchallenge eosinophil increase.

Key messages:

  •  NAL is a useful method in occupational settings on a group level and in the individual when the subject acts as his/her own control.
  •  Nasal cytology and NAL may both be used as a clinical tool to objectively measure nasal inflammation, and the reliability of both methods depends on the sampling technique and sample analysis (cell counts vs cellular marker measurements).
  •  nNO is foremost a research tool, and as yet not suitable in the clinical setting.