Exhaled breath condensate Ph, hydrogen peroxide and leukotriene B4 in horses: measurement considerations and application for the investigation of lower airway disease

The measurement of biomarkers in exhaled breath condensate (EBC) is an expanding field of non-invasive respiratory assessment. In humans, several biomarkers, including hydrogen peroxide (H₂O₂), pH and leukotriene B₄ (LTB₄) have proven useful for the detection and monitoring of inflammatory lung diseases, including asthma and chronic obstructive pulmonary disease. While EBC analysis could offer an alternative to conventional invasive methods for the diagnosis of equine lower airway inflammation (LAI), little information is available regarding EBC biomarkers in horses. The aims of this study were to (1) determine the intra- and inter-day consistency of EBC pH, H₂O₂ and LTB₄ measurements in horses, (2) determine whether these biomarkers are influenced by environmental, animal or methodological factors, (3) identify associations between these biomarkers and cytological and endoscopic definitions of LAI and (4) develop a superior tandem mass-spectrometry (MS/MS) method for H₂O₂ measurement.
Biomarker measurement consistency was assessed using EBC samples collected from 10 horses on three occasions (day 1), and at midday on days 2 and 3. Influences of environmental, animal and methodological factors on biomarkers were determined using EBC samples collected from 20 horses (pH; LTB₄) and 10 horses (H₂O₂). Clinical, endoscopic and airway cytological findings from 47 horses were compared with EBC pH, H₂O₂ and LTB₄ measurements by univariate and multivariable analyses. Dichotomous (presence/absence of LAI) and continuous (respiratory fluid differential cell counts) outcome variables were evaluated and environmental and methodological factors were included. Standard H₂O₂ solutions were used to compare flow injection analysis and liquid chromatography (LC) using either fluorescence (FLD) or MS/MS detection with instrument grade water or pooled EBC as matrix for standard preparation. The influence of sample matrix was evaluated by comparison of peak areas and gradients of standard curves.
Intra- and inter-day consistency for both H₂O₂ and pH were adequate (intraclass correlation coefficients ≥0.8). The intra- and inter-day consistency for LTB₄ was poor. H₂O₂ was influenced by ambient temperature (T_A, P=0.004), humidity (P=0.009), time of day (P=0.009) and collection location (P=0.007), while pH was influenced by respiratory rate during EBC collection (P<0.001) and T_A (P<0.001). H₂O₂ concentrations and pH were higher in horses with LAI and were positively associated with bronchoalveolar lavage fluid (BALF) neutrophil percentage (P=0.033 and P=0.01, respectively). BALF mast cell percentage was negatively associated with EBC pH (P=0.041) and BALF eosinophil percentage was positively associated with EBC LTB₄ (P=0.035). The LC-MS/MS method was more sensitive (limit of detection 0.038 µM) than previously published FLD methods. The intra- and inter-day assay coefficients of variation were 1.0-3.2% and 1.7-2.3%, respectively. A significant matrix effect was identified (P<0.001).
The study demonstrated adequate consistency of H₂O₂ and pH, and that both biomarkers were altered by inflammation, suggesting that use of these biomarkers in the diagnosis and monitoring of LAI, may be warranted. However, environmental, animal and methodological factors can influence these biomarkers and should be considered in the interpretation of results. The LC-MS/MS method was 3.7 times more sensitive with superior selectivity and repeatability compared with existing FLD methods. However, the effect of matrix-solution should be considered prior to sample analysis.