Steamed hay and alfalfa pellets for the management of severe equine asthma

Abstract

Background: Steaming hay significantly reduces respirable particles and provides a palatable alternative to dry hay for horses with asthma, but there are few prospective studies demonstrating the clinical efficacy of steamed hay.

Objectives: To compare the efficacy of alfalfa pellets and steamed hay in improving lung function and inflammation of horses with severe asthma (SEA).

Study design: Controlled crossover study.

Methods: Ten horses with SEA were enrolled and nine completed the study. Horses were housed indoors and fed hay. Once in exacerbation, they were fed pellets and steamed hay for 4 weeks, in a crossover design. Weighted clinical scores and lung function were recorded weekly. Bronchoalveolar lavage fluid (BALF) cytology and mucus scores were recorded before and after each diet.

Results: Based on linear mixed model (LMM) analysis, weighted clinical scores significantly improved over time (p < 0.001, no diet effect or time-diet interactions). With pellets, weighted clinical scores decreased from (median (interquartile range)) 13 (5.5) to 2 (1.5), while with steamed hay, they decreased from 10 (9.5) to 6 (8.5). With pellets, lung resistance decreased significantly from a baseline of (mean (SD)) 2.62 (0.77) cmH₂O/L/s to 1.17 (0.66), 0.79 (0.54), 0.70 (0.20), 0.62 (0.18) on Weeks 1-4, respectively (LMM with post hoc tests, p < 0.001). With steamed hay, the resistance decreased significantly from a baseline of 2.34 (0.93) cmH₂O/L/s to 1.38 (0.49) and 1.51 (0.66) on Weeks 1 and 2, respectively (p < 0.03). Neutrophils BALF decreased significantly with both diets (pellets: 40.2 (24.4) to 20.1 (11.0) %; steamed hay 30.9 (20.2) to 25.7 (17.6) %; LMM, p = 0.006).

Main limitations: A small number of horses in a research setting. Dust was not measured in the stalls.

Conclusions: Clinical scores, lung function and BALF inflammation of horses with SEA improved with steamed hay and pellets, but the effect on lung function was more pronounced with pellets.

Keywords: dust; environment; heaves; horse; mucus; recurrent airway obstruction.

FIGURE 1

Experimental timeline. Lung function, bronchoalveolar lavage (BAL), mucus scores and weighted clinical scores were realised in the stable where the horses were housed. Mucus scores were based on videos recorded during BALF collection and anonymised for blinded scoring by a veterinarian. Weighted clinical scores were performed by a veterinarian blinded to the diet. At the end of each period of 4 weeks, the presence of residual bronchospasm was evaluated with the administration of hyoscine butylbromide. For two horses, the initial exacerbation period lasted 6 and 8 weeks instead of 4.

FIGURE 2

Weighted clinical scores. Blinded 23‐point clinical scores during exacerbation (Exa), and after 1–4 weeks with pellets (open circles) and steamed hay (black circles). There was a significant time effect (p < 0.001) and scores are significantly lower than during exacerbation on Weeks 2, 3 and 4 (*p < 0.001). There was no significant effect of diet and no time‐diet interaction. Results are expressed as median and interquartile range (IQR).

FIGURE 3

Lung function. (A) Difference in transpulmonary pressure (ΔP _L), (B) lung resistance and (C) lung elastance during exacerbation (Exa), and after 1–4 weeks with pellets (open circles) and steamed hay (black circles). There was a significant time effect (p < 0.001) and significant time‐diet interactions (p ≤ 0.03). *Significant difference from baseline for both diets (p ≤ 0.02), ^#Significant difference from baseline for pellets only (p < 0.009) and ^†significantly difference between diets (p ≤ 0.02). Results are expressed as mean ± SD. The grey area represents normal values.

FIGURE 4

Residual airway obstruction after 4 weeks. (A) The difference in transpulmonary pressure (ΔP _L), (B) lung resistance and (C) lung elastance before (open circles) and 10 min after (black circles) the administration of hyoscine butylbromide (HB). ‘Before’ data are the same as in Week 4. Each horse is represented individually, and the grey area represents normal values. There was a significant diet effect for R _L (p = 0.049) and time‐diet interactions for ∆P _L and R _L (p ≤ 0.015). The percentage of reduction of ∆P _L and E _L was significant larger for steamed hay than pellets (p ≤ 0.001). * indicates a significant difference between groups before administration and ^# indicates a greater percentage of reduction post‐bronchodilation with steamed hay compared to pellets.

FIGURE 5

Bronchoalveolar lavage fluid (BALF) neutrophils and tracheal mucus scores. (A) Blinded neutrophil differential cell counts when horses were in exacerbation (open circles) and after 4 weeks with pellets and steamed hay. There was a significant time effect, with higher BALF neutrophils during exacerbation (p = 0.006), but no diet effect or time‐diet interaction. The grey area represents normal values. (B) Blinded 5‐point tracheal mucus score during exacerbation (Exa) and after 4 weeks with pellets (open circles) and steamed hay (black circles). There was a significant time effect (p < 0.001) and significant time‐diet interaction (p = 0.005). * indicates significantly difference from exacerbation and ^# indicates significantly difference from steamed hay (p < 0.001). Each horse is represented individually.

FIGURE 6

Concentration of respirable particles before and after steaming. The concentrations of respirable particles (Resp), which include particles of 4 μm or less, are measured by shaking 2 kg of hay for 3 min before and after the steaming process. There was a significant effect of steaming (before–after), of the type of hay (standard versus dusty) and a significant steaming‐type of hay interaction (p ≤ 0.008). * indicates a significant difference between the type of hay (p < 0.0001 and 0.015 for dry and steamed, respectively) and ^# indicates a significant difference compared to pre‐steaming values (p < 0.0001). Results are expressed as mean ± SD.