Changes in PH in exhaled breath condensate after specific bronchial challenge test in patients with chronic hypersensitivity pneumonitis: a prospective study

The present study demonstrates that a drop in EBC pH of 0.3 or higher after SIC has a sensitivity of 30 % (CI: 12.8 to 54.3 %) and a specificity of 100 % (CI: 65 5–100 %) for diagnosis of HP due to exposure to molds. In the case of HP due to bird proteins, a fall in EBC pH of 0.3 or more after SIC had a sensitivity of 23.8 % (CI: 12.6 to 39.8 %) and a specificity of 66.6 % (35.4 to 88.7 %). This study is the first to analyse the diagnostic performance of the EBC pH in the context of antigen exposure through SIC, in patients with HP due to exposure to bird’s antigens or to molds.

These results may be helpful for establishing a firm diagnosis of HP. In acute forms of the condition the clinical criteria currently in use are sufficient [ 2], but establishing a firm diagnosis in chronic or subacute forms is much more difficult. It has recently been shown that up to 50 % of cases of idiopathic pulmonary fibrosis may be evolving forms of HP that have not been properly diagnosed [ 25] and it is in this context where the SIC may be most useful [ 26].

The SIC has been used to diagnose since the 1960s, but there is still no consensus regarding the variables on which the test’s interpretation should be based. This situation may influence its sensitivity and specificity. In a study with 29 patients basically using clinical criteria such as the onset of symptoms and signs mimicking influenza, Hendrick et al. [ 5] reported a sensitivity of up to 85 % and a specificity of 95 %. Also basing positivity on clinical symptoms, Ohtani et al. [ 7] found no false results in a study of 11 patients with bird fancier's lung. However, other authors using criteria based on falls in the values obtained in lung function tests reported sensitivities ranging from 82 to 92 % and specificities between 76 and 100 % [ 4, 6].

Establishing a positive diagnosis from the appearance of symptoms during the test necessitates a high level of exposure to the antigen, a requirement which may of course have a negative effect on the test’s safety. The use of criteria based on lung function studies to determine diagnosis may allow lower levels of antigen exposure. For example, maximum exposure in Ohtani et al.’s [ 7] study was 680 μg of avian protein, whereas Morell et al. [ 4] used a maximum dose of 200 μg of protein. However, this lower antigen exposure during SIC may yield false negatives: evaluating 113 patients, of whom 88 were diagnosed with HP due to different agents, Muñoz et al [ 26] recently found a sensitivity of 73 % and a specificity of 84 %, and 24 patients finally diagnosed with HP had negative SIC. The results of this study allow us to hypothesize that the measurement of EBC pH during SIC may reduce the number of false negatives and thus improve the test’s diagnostic accuracy. In this regard, three patients in the present study diagnosed with HP but with a negative SIC, experienced a decline of EBC pH greater than 0.3 units after the SIC.

EBC pH measurement is a recently introduced tool which may be useful in assessing various respiratory diseases [ 27]. Several studies have shown that pH values may fall in non-controlled asthma [ 28] or in the context of respiratory infections in patients with bronchiectasis, COPD or cystic fibrosis [ 29]. However there is less experience in the context of interstitial diseases: higher levels have been found in individuals with pulmonary fibrosis [ 30], and lower levels in patients with asbestosis [ 31]. EBC pH is the result of a balance between various buffer systems and the production and release of acids and bases in the airways [ 32]. In healthy individuals, EBC is determined to a significant extent by the NH ₄, HCO ₃ and CO ₂ produced during breathing [ 33], with the most acidic pH being found in the alveolar lumen in the proximal airway. Inflammatory processes trigger a range of mechanisms which produce acidification of these more proximal airways as a possible innate defense mechanism [ 34]. These mechanisms are basically the production and excretion of superoxide ions and protons by the respiratory epithelial cells, the inhibition of glutaminase activity in epithelial cells, and finally the recruitment of macrophages and neutrophils whose lysis in the context of inflammation raises the acidity level of the environment [ 35, 36]. In fact the acidification produced by neutrophil recruitment may well explain some of the findings of the present study. In this sense, previous studies have shown that EBC pH values are decreased during asthma exacerbations but they are not related with spirometric values [ 37]. Furthermore Kostikas et al [ 38] demonstrated that EBC pH levels are negatively correlated with the number of eosinophils in sputum and positive correlated with neutrophilic airway inflammation.

Although the presence of lymphocytic inflammation in the alveoli is characteristic in HP, in the bronchi it has been shown that there may be neutrophilic inflammation, especially in cases in which molds are the causative agent [ 12, 14]. Our group has also recently confirmed that individuals with HP present significantly increased levels of neutrophils in induced sputum following the challenge with fungal agents [ 14]; this may explain the decrease in pH found in the present study after the challenge test in patients with HP caused by molds. Taken together, these findings also allow us to hypothesize that the mechanism of action of the HP may differ depending on the causative agent.

This study has a number of limitations, some of them deriving from factors that may have influenced the pH values recorded. The most important of these factors is smoking: lower levels of pH in EBC have been reported in healthy individuals exposed to tobacco smoke than in non-exposed subjects [ 39, 40]. In our study, however, this is unlikely to have been a determinant, as we did not find significant differences in smoking habits in our four study groups. Another limitation is the small number of patients ultimately included. Future studies are needed, with larger numbers of patients, to confirm the sensitivity and specificity of EBC pH for the diagnosis of HP. Moreover, the environmental exposure to specific antigens before SIC is difficult to demonstrate, especially in patients exposed to molds. If experimental exposure to antigen can change pH of EBC, natural exposition can also influence it, so we can not rule out that this environmental exposure might influence the results. Finally, another possible limitation is the fact that EBC was recorded 24 h after the antigen challenge, which may have affected pH levels observed. As yet, the variability of EBC pH after SIC has not been assessed either in occupational asthma or in HP [ 41]. In any case, the decision to record EBC after 24 h was based on the protocol established by different groups in the context of occupational asthma, in which markers of inflammation are analysed using noninvasive methods in order to help establish the positivity of SIC [ 42].