Airway remodeling in horses with mild and moderate asthma

Abstract Background There is a remodeling of the central airways in horses with severe asthma but whether a similar process occurs in horses with the mild or moderate asthma (MMA) is unknown. Objectives To evaluate lesions affecting the central airways of horses with MMA. Animals Twelve horses with MMA and 8 control horses. Methods Case‐control retrospective study of horses classified as MMA affected or controls based on history and bronchoalveolar lavage fluid cytology. Endobronchial biopsies were analyzed using histomorphometry and a semiquantitative histologic scoring system. Results Histomorphometry identified epithelial hyperplasia (47 μm2/μm [34‐57 μm2/μm]; P = .02), a thickened lamina propria (166 μm [73‐336 μm]; P = .04), and smooth muscle fibrosis (42% [33%‐78%]; P = .04) in horses with MMA when compared to controls horses (24 μm2/μm [21‐80 μm2/μm]; 76 μm [36‐176 μm]; and 33% [26%‐52%], respectively). The semiquantitative score results indicated, in horses with MMA, the presence of epithelial hyperplasia (7 of the 12 horses with MMA and only 1 of the 8 control horses had a score of 1/1), and submucosal inflammatory leucocytes in the central airway (11 of the 12 horses with MMA and only 4 of the 8 control horses had a score ≥ 1/2). Conclusions and Clinical Relevance Tissue remodeling of the bronchial lamina propria, epithelium, and smooth muscle was present in horses with MMA.


| INTRODUCTION
Mild or moderate asthma (MMA; also known as inflammatory airway disease) in horses is associated with lower airway obstruction 1,2 and is a common cause of exercise intolerance in performance horses of all ages. 3,4 Eighty percent of racehorses in their first year of training experience lower airway inflammation 5 and airway inflammation is associated with decreased racing speed. 6 The condition could be underdiagnosed because affected horses often do not have clinical signs suggestive of a lower airway disease. 7 In clinical practice, the diagnosis of MMA is based on the presence of compatible clinical signs combined with abnormal bronchoalveolar lavage (BAL) fluid cytology. 8 Laboratory studies reveal abnormal lung function at rest, 9 decreased PaO 2 , 10 and decreased VO 2 Abbreviations: ASM, airway smooth muscle; BAL, bronchoalveolar lavage; HEPS, hematoxylin-eosin-phloxine-saffron; MMA, mild or moderate equine asthma; SA, severe asthma. peak during exercise. 11 Although diagnostic tools are helpful to study the physiological effects of MMA, use is currently limited by the specialized equipment and expertise required. Bronchoalveolar lavage, however, can easily be performed in the field.
Remodeling in the central and peripheral airways of horses with severe asthma (SA) affect the epithelium, the lamina propria, and the bronchial smooth muscle layer. [12][13][14][15][16][17][18][19][20] These structural changes correlate with decreased lung function in horses with SA, and could contribute to the impaired gas exchanged during exercise in MMA. 17,[20][21][22][23] Airway inflammation and obstruction are usually managed therapeutically and airway remodeling in SA is only partially reversed by antigen avoidance, and administration of corticosteroids and bronchodilators. 21,24 Although there is limited information about airway remodeling in horses with MMA, the presence of histological changes might be associated with the severity of the clinical signs. 20,25,26 Considering the presence of lower airway obstruction in horses with asthma of all severities, we postulated that lesions are present in central airways of horses with MMA. Our objective was to assess the histologic alterations in the central airways of horses with MMA using histomorphometry and a semiquantitative scoring system.

| MATERIAL AND METHODS
History, CBC, serum biochemistry, BAL cytology differential count, and endobronchial biopsies from horses with MMA and control horses from the Equine Respiratory Tissue Bank were studied. Characteristics of horses with MMA were previously reported. 27 This exploratory study was retrospective, controlled, and blinded. The experimental protocol was approved by the University of Montreal's Animal Care Committee (Rech-1647) and informed consent was provided by all owners.

| BAL fluid collection and analysis
Collection of BAL fluid was performed as described. 28 The BAL differential count was made on archived samples blindly and in accordance with previously described methods 27 by a board-certified clinical pathologist (C. Grimes).

| Endobronchial biopsies
Endobronchial biopsies were obtained by videoendoscopy after collection of the BAL fluid. Briefly, the bronchoscope was advanced until it wedged to the most distal airway, and a smooth oval disposable biopsy forceps (FB-234U, Olympus, Richmond Hill, Canada; diameter, 2.85 mm) was used to sample sequentially the most caudal to the most cranial carina until reaching the main carina. 13 Samples were fixed in 10% neutralbuffered formalin for 48 to 72 hours before paraffin embedding. 29

| Histopathologic analysis
Biopsies were stained with alcian blue and hematoxylin-eosin-phloxinesaffron (HEPS) and with modified Russell-Movat's pentachrome as previously described. 30 Twenty-four biopsies in horses with MMA (2 biopsies/horse) and 18 biopsies of control horses (from 2 to 4 biopsies/ horse) reaching the inclusion criteria were stained by modified Russell-Movat's pentachrome and were scored based on quality (1-5/5). 13 The quality score evaluates the orientation, architecture, structure preservation, and presence of airway smooth muscle (ASM) in the biopsies. A score of 1/5 represents a poor-quality tissue where the continuity between epithelium, extracellular matrix (ECM), and smooth muscle is lost. A score of 2/5 represents a well-oriented tissue, but its architecture is not completely preserved. A score of 3/5 represents a welloriented tissue, and its architecture is preserved at least in 50% of the biopsy. A score of 4/5 represents an optimal tissue orientation for the biopsy with of loss of continuity between tissues. Finally, a score of 5/5 represents an optimal tissue orientation for the biopsy, the tissue architecture is perfectly conserved, and the parenchymal borders of the smooth muscle layer are clearly identifiable. Biopsies with a score ≥ 3/5 were studied.
Twenty-six biopsies from MMA and control horses were analyzed.
Twelve horses with MMA were included in the study (biopsies from 11 horses with Movat staining and 1 horse with HEPS staining).
For the 11 horses with Movat staining, 14 biopsies with a quality score ≥ 3/5 were included (1 biopsy/horse; n = 8 and 2 biopsies/ horse; n = 3). In addition, for 2 asthmatic horses 1 biopsy with a lower quality score (2/5; good tissue orientation, tissue architecture not completely preserved) but acceptable to perform measurement were also included.
Histologic features assessed included: biopsy total area, smooth muscle (total and relative area), fibrosis within the smooth muscle, epithelium (total and relative area), goblet cells in the surface epithelium, airway extracellular matrix (total and relative area), and elastic fiber in the airway extracellular matrix: They were measured using point counting (200 points/tissue) in Movat stained biopsies using newCAST software (Visiopharm [version 6.5.2.2303], Hoersholm, Denmark). 31 Basement membrane thickness, distance between basement membrane and muscle (evaluating lamina propria thickness), and epithelium area corrected by basement membrane length (evaluating epithelium hyperplasia) were measured on perpendicularly sliced Movat-stained biopsies using ImageJ software (ImageJ In 1 asthmatic horse, the histomorphometry analysis was performed on an HEPS stained biopsy; fibrosis within the smooth muscle, goblet cells in the surface epithelium, and elastic fibers in the airway extracellular matrix could not be evaluated. Biopsies were evaluated by a board-certified anatomic pathologist (P. Hélie) blinded to horse identity using a semiquantitative histologic score previously described for horses with SA. 14 The pathologist attributed 0, 1, or 2 points in accordance with 10 predefined criteria.
The extracellular matrix was evaluated for the thickness of the basement membrane (/1), submucosal inflammatory cells (/2), and mucous glands (/1). Finally, smooth muscle was evaluated for the presence of fibrosis (/2), mucous glands (/1), and smooth muscle ending visible (/1). The score was read on Movat-stained biopsies for 18 horses (11 MMA and 7 control horses). For 1 horse with MMA, the score was read on a HEPS-and alcian blue-stained biopsy. For 1 control horse, the score was read on a Movat-stained biopsy completed with HEPS-and alcian blue-stained biopsies. When 2 biopsies per horse were available, each parameter of the semiquantitative score was read on the tissue with the highest quality. Therefore, only 1 semiquantitative score (/14) per horse was obtained.

| Statistical analysis
The Mann-Whitney U test was used to evaluate difference between MMA and control groups regarding BAL cytology differential counts, histomorphometric measurements. A Spearman correlation was used to evaluate the relationship between the histomorphometric measurements and the duration of disease, age, or BAL cytology differential counts, as well as the correlation between the histomorphometric measurements within the same biopsy. However, only BAL fluid neutrophils were significantly increased in horses with MMA (P = .003; Figure 1) when compared with controls.
The The smooth muscle surface area in central and peripheral airways is increased in horses with SA during exacerbation when compared to the remission state 21,24 and to control horses. 16 Airway smooth muscle is central to airway hypersensitivity in humans, and this is likely to be true in horses. 32 In the current study, however, the percentage of smooth muscle in the endobronchial biopsies was similar in both groups of horses. A comparable finding in horses with SA is reported. 13 A proposed explanation is that the thickening of the lamina propria might have resulted in incomplete sampling of the smooth muscle layer. 13,16 This limits the value of endobronchial biopsies when assessing ASM via endobronchial biopsies. Evaluation of other indices of smooth muscle remodeling, such as myocytes density, cell proliferation, and apoptosis in endobronchial biopsies from horses with MMA would be feasible using these tissues, as they are found to be altered in horses with SA and do not require full thickness ASM sampling. 16 A prospective study of horses with MMA using endobronchial ultrasound could also be considered as a noninvasive alternative to evaluating ASM remodeling, including progression of the disease over time, and after treatment. 12 Bronchial smooth muscle fibrosis was significantly increased in horses with MMA compared to control horses. In horses with SA, the smooth muscle fibrosis is increased in peripheral and central airways. 24 Although we started with a larger cohort, several horses did not meet inclusion criteria. A higher number of cases might have allowed more complete characterization of the remodeling features of MMA, and the presence of different remodeling phenotypes. Analyzing a higher number of biopsies by horses would have been preferable, but this was not possible in this client owned cohort. In a previous study, 24 5