MUMultiple speciesBmoderateT1hypothesis only
What counts is not visible dust but the respirable fraction in the horse's breathing zone: figures from trials on hay, steaming and haylage.
Executive summary
Equine asthma is a non-contagious inflammation of the lower airways, sustained by inhaled particles from hay, bedding and stable air. It is assessed through the cell profile of the bronchoalveolar lavage. In a randomised trial, steamed hay and haylage both lowered dust levels in the breathing zone, yet the lavage cell profile improved within six weeks only on haylage. How much air quality achieves on its own has not yet been cleanly separated from the effect of medication.
Key points
Equine asthma denotes a non-contagious, recurrent inflammation of the lower airways in the horse. The term covers two forms: the mild to moderate form, formerly listed as inflammatory airway disease, and the severe form, formerly called heaves or recurrent airway obstruction. A 2016 consensus statement put this classification in order. A 2022 scoping review then examined what the case definition actually rests on: a hierarchy of coughing, loss of performance, mucus in the trachea, inflammatory cell profile of the lung lavage and measurable impairment of lung function.
How common the disease is depends less on the region than on the measurement method. In a teaching herd kept at pasture all year round and with no history of respiratory problems, 10% of the horses met the criteria for the severe form in summer and 60% those for the mild form; in winter the figures were 4.3% and 87%. What is remarkable is not the level but the movement: in 61.1% of the animals the cell profile changed between the two seasons.
Why some horses in the same stable fall ill and others do not is partly hereditary. A segregation analysis in the offspring of two affected Warmblood stallions and in 401 Swiss Warmbloods supported a mixed inheritance model with one major gene involved; in one stallion line it behaved dominantly, in the other recessively. Expression remained tied to hay exposure throughout. Predisposition and air work together, not against each other.
Because only very fine particles reach the small airways. In a study of 64 racehorses from eight yards, inhalable and respirable dust as well as endotoxin and beta-glucan were measured simultaneously in the breathing zone. Only the respirable fraction was associated with the proportion of neutrophils in the lung lavage, not the coarser one.
The same work links the inflammation to performance. Of 98 lung lavages taken after races, 78 showed the picture of mild asthma. For every percentage point more mast cells, the performance index was 2.9 points lower, and for every percentage point more neutrophils, 1.4 points lower. Beta-glucan, a component of fungal cell walls, was associated only with the mast cells, dust only with the neutrophils. Two kinds of irritant, two cell profiles.
How strongly the ambient air acts was tested by a randomised, double-blinded trial in twelve polo horses with smoke-induced mild asthma. When the mean fine particulate load of the ambient air fell from about 36 to about 7 micrograms per cubic metre, maximal oxygen uptake rose by an average of 13.2%. There was no difference between the corticosteroid group and the saline group. Two caveats remain: before the final exercise test all the horses were additionally given an inhaled bronchodilator, and there was no comparison group that stayed in the polluted air. The trial separates the drug from the air, but not the air from the passage of time.
The dust in the breathing zone rarely comes from a single source. A prospective survey of 731 horses presented for examination found a positive fungal culture in 55% of the respiratory samples. Horses with visible fungal elements in the tracheal secretions had roughly twice the likelihood of airway inflammation. Straw bedding went with roughly twice the likelihood, dry hay with roughly 2.7 times. These are observational data: they show associations, not a proven chain of causation.
Sequencing of the viable bacterial community of four batches of hay compared dry, soaked and high-temperature steamed hay. Steaming lowered the microbial load most, without narrowing species diversity. Soaking reduced diversity and let the proportion of Gram-negative bacteria rise. Both methods cut back organisms that have been linked to respiratory disease.
A feeding trial in six horses fills in the practical side. After 60 minutes of steaming at 100 degrees, typical moulds were no longer detectable; after 15 minutes of soaking they stood at 50 colony-forming units per gram. When soaked hay was stored, bacteria, moulds and yeasts rose again. Both treated hays were eaten more slowly than untreated hay, and the steamed hay was also chewed markedly more intensively.
The largest randomised comparison on this question ran over six weeks in 43 Thoroughbreds in race training. Dry hay produced around 0.081 milligrams of respirable dust per cubic metre in the breathing zone, steamed hay 0.056 and haylage 0.053. On the cell profile the paths diverged: only in the haylage group did the proportion of neutrophils in the lung lavage fall, from 5.1% to 3.0%, while in the hay group it stood at 6.3%. At the same time the fatty acid profile in the blood shifted towards the inflammation-resolving side, again only on haylage.
A smaller pilot trial in seven Standardbreds in training points the same way, with wider gaps: 0.02 against 0.06 milligrams of respirable dust per cubic metre and 69 against 160 picograms of beta-glucan. Neutrophils fell on haylage from 2.2% to 0.7%, while the hay group stood at 4.0% after six weeks. Seven horses is few; the direction agrees with the larger trial.
| Forage type | Respirable dust in mg/m³ | Neutrophils after six weeks | Trial |
|---|---|---|---|
| Dry hay | 0.081 | 6.3% | 43 Thoroughbreds, randomised |
| Steamed hay | 0.056 | no fall reported | 43 Thoroughbreds, randomised |
| Haylage | 0.053 | 3.0%, baseline 5.1 | 43 Thoroughbreds, randomised |
| Dry hay (alfalfa) | 0.06 | 4.0% | 7 Standardbreds, pilot trial |
| Haylage (grass and alfalfa) | 0.02 | 0.7%, baseline 2.2 | 7 Standardbreds, pilot trial |
In the sick horse the picture becomes less tidy. In a crossover trial in ten horses with severe asthma, nine of which completed the study, clinical signs, lung function and inflammation improved on alfalfa pellets as well as on steamed hay. Airway resistance fell on pellets over four weeks from 2.62 to 0.62 centimetres of water per litre per second, on steamed hay only temporarily, from 2.34 to 1.38 in the first week and 1.51 in the second. Neutrophils fell on pellets from 40.2% to 20.1%, on steamed hay from 30.9% to 25.7%; at the end they still stood above five per cent in every horse. The clinical score improved over time, without the two forages differing statistically from each other.
A second crossover trial by the same working group tested steamed against dry hay in horses in remission and found no difference. Airway resistance rose over four weeks in both arms, and neutrophils climbed from 6.7% to 13.1% on dry hay and from 5.6% to 10.5% on steamed hay. The researchers attribute this to an unexpectedly mild deterioration and note that dust in the stalls was not measured. Steaming lowers particles measurably; that it prevents an exacerbation is not thereby demonstrated.
In a bronchoalveolar lavage, saline is flushed into a section of the lung through an endoscope or a tube and then drawn back out. What is counted afterwards is which cells are floating in it: neutrophil granulocytes, mast cells, eosinophils and macrophages. The classification of the asthma type is derived from these proportions.
The 2022 scoping review searched 2,275 hits and evaluated 44 studies with 6,092 horses in total, a median of 74 animals per study. A meta-analysis was not possible because measurement and reporting methods were too varied. In almost half of the papers, the informative value of the sample and the blinding were rated as inadequate. The most stable relationship was the one between coughing and tracheal mucus; the least robust, of all things, the one between lavage cell profile and lung function.
A single lavage is also a snapshot. In the pasture herd mentioned, the cell profile changed between summer and winter in 61.1% of the horses. And a randomised trial in thirteen horses showed that an injected corticosteroid can shift the cytokines without altering the cell profile: in the asthmatic animals the activity of the cytokine interleukin-17 was raised around fivefold and that of interleukin-10 lowered about two and a half times; the drug lowered individual cytokines, but the cell composition remained unchanged. As long as the exposure remains, the inflammation remains.
Effective drugs do exist. In a European multicentre study of 224 horses with severe asthma, 73.4% of the animals treated with inhaled ciclesonide reached the predefined treatment success after ten days, against 43.2% on sham treatment. The 43% is the genuinely instructive figure: even without an active substance, almost half of the animals improved within ten days. In racehorses with moderate asthma, a trial in just over twenty animals found clinical signs and mast cell proportion improving on the same substance, even without any change in husbandry.
The route of administration matters too. In a trial in twelve horses with severe asthma, orally administered dexamethasone improved airway resistance while the same amount nebulised did not; the adrenal axis was suppressed in both groups. And a corticosteroid aimed at an entirely different organ has an effect as well: after triamcinolone was injected into both hocks, lung function improved within a week in a trial in ten horses with severe asthma and remained better for up to three weeks. A joint treatment can temporarily mask a lung disease.
What drugs do not achieve is shown by the long-term comparison. Over twelve months, eleven horses with heaves were managed either with inhaled fluticasone or with consistent allergen avoidance alone. Lung function normalised in both groups, faster on the corticosteroid; the inflammation was better controlled on allergen avoidance. Airway smooth muscle mass, that is the structural remodelling of the bronchi, fell by around 30% in both groups. A five-month trial with added azithromycin did lower neutrophils more strongly, but lung function and muscle mass changed to the same degree.
Newer approaches do not shift this order either. In a randomised comparison in twenty horses, a single intrabronchial dose of the animal's own adipose-derived cells did not achieve proof of equivalence with dexamethasone, but did improve the clinical signs and individual inflammatory cytokines, with an effect lasting over a year. The underlying question remains exposure.
Severe equine asthma is used in comparative research as a naturally occurring model of the human asthma form in which neutrophils predominate. Review articles name the shared features: variable airway narrowing, coughing, hyperresponsive airways, mucus overproduction, structural remodelling of the bronchi and a persistently activated innate immune response. Farmers and horses also share the same irritant, namely organic dust from hay and stable.
The differences matter just as much. Human asthma mostly runs through a type 2 immune response with eosinophils, whereas the severe form in the horse is predominantly neutrophilic. In humans there is also smoking, indoor climate and the option of changing occupation, none of which the horse has. And the horse lives its whole life in a single body of air shaped by people. What is measured in the stable is therefore not transferable to a human home.
One finding, however, fits strikingly well. An updated systematic review of occupational asthma evaluated 26 before-and-after studies. Complete removal from exposure improved symptoms and lung function values; merely reducing exposure improved the symptoms but not the measured values. Exactly this pattern shows up with steamed hay. This parallel is a hypothesis about a shared mechanism, not evidence that results from the stable would transfer to the workplace.
On the human side the environmental evidence is broader but not more certain. A 2024 European guideline paper rates it as probable that short-term rises in fine particulate and nitrogen dioxide levels increase hospital admissions and emergency consultations for asthma. That measures to reduce outdoor air pollution lower attacks, by contrast, counts as only weakly supported. The direction is the same across species; the certainty of the statement is not.
No prescriptions can be derived from the evidence, but questions to ask can. The most important concerns the measuring point: what the most informative studies assessed was the air in the horse's breathing zone, not the look of the bale. Hay that barely dusts when held up to the light can still be a burden in the breathing zone, and conversely, treating the hay changes nothing about bedding, the neighbouring stall and ventilation. That the neighbouring stall counts has been measured: when one stable in a shared air space switched from straw and hay to shavings and haylage, dust exposure fell in the adjacent stall as well.
Original analysis
| Measure | Dust in the breathing zone | Microbial content of the forage | Nutrients | Storage and intake | Testing in the diseased horse |
|---|---|---|---|---|---|
| Dry hay | highest measured value, 0.081 mg/m³ | baseline | unchanged | storable | comparison arm in several trials |
| Soaking, 15 minutes | markedly lowered in a real-time measurement from 2007, without an absolute value comparable to the newer figures | moulds not eliminated, 50 CFU per gram; proportion of Gram-negative bacteria rises | soluble components pass into the water | not storable, microbial counts rise within hours; slower intake | none of the controlled studies evaluated here |
| Steaming, 60 minutes at 100 degrees | 0.056 mg/m³ | moulds no longer detectable, strongest microbial reduction without loss of species diversity | precaecal digestibility of crude protein from 56% to 35% | more stable than soaked hay; slower intake, more intensive chewing | two crossover trials with inconsistent results |
| Haylage | 0.053 mg/m³, 0.02 in the pilot trial | not tested in the studies evaluated | fatty acid profile shifts towards the inflammation-resolving side | conservation risks of its own, not tested here | randomised trial in 43 racehorses |
| Alfalfa pellets as a forage replacement | not measured | not tested | forage structure is lost, the ration has to be replaced | not tested | crossover trial with the largest fall in airway resistance |
No. Equine asthma is a reaction of the airways to inhaled particles, not an infection. Several horses in one stable can fall ill at the same time because they breathe the same air, not because they infect one another. Infections can exist alongside it and complicate the assessment; making that distinction is the task of the veterinary examination. A suspected infection changes nothing about the recurring factor of air quality.
Soaking lowers the dust but does not solve everything. A real-time measurement in the breathing zone from 2007 showed that soaking before feeding markedly lowers the concentration of respirable dust, with very long soaking bringing no additional gain. Microbiologically, however, soaking performs worse than steaming: typical moulds remain detectable, the proportion of Gram-negative bacteria rises, and the hygienic improvement is lost within hours of storage. In the more recent comparative trials that also measured the lung lavage cell profile, soaking was not represented as an arm of its own. What it means for the inflammation is therefore open.
No. In a herd kept at pasture all year round with no known respiratory problems, 87% of the horses met the criteria for the mild form in winter and 60% in summer. There is also a form that appears precisely in summer on pasture. Pasture typically lowers exposure to hay dust, but it does not replace an assessment of the individual case by the veterinary surgeon.
Because the inflammation is local. The classification rests on the ratio of the cells in the flushed fluid, which is not reflected in a blood count. At the same time the lavage is not a perfect measure: the 2022 scoping review found the relationship between cell profile and lung function to be, of all the relationships examined, the least robust. And the cell profile changes between the seasons in a substantial proportion of horses.
Not the same, but a related one. The severe form in the horse is used as a naturally occurring model of the human asthma form in which neutrophils predominate. Human asthma, by contrast, mostly runs through eosinophils. What they share is the irritant organic dust and the structural remodelling of the bronchi. A treatment success in one species proves nothing for the other, in either direction.
Function can largely recover, the structural remodelling only partly. In a trial in eleven horses over twelve months, lung function normalised both on consistent allergen avoidance and on inhaled corticosteroid, and airway smooth muscle mass fell by around 30% in both groups. A remainder therefore stays measurable. Whether this gain holds when exposure returns has not been established.
Equine Health
Around a quarter of colic cases seen by vets turned critical: which findings set them apart, which risk factors recur, and what the evidence shows.
HOHorseBmoderate Open access
Comparative Health
Radiocarbon data suggest the Achilles tendon core forms in growth and is barely renewed thereafter: what that means for healing, scar and the equine t
MUMultiple speciesBmoderateT2limited target data Open access
Comparative Health
Serum amyloid A rose about 227-fold in an equine experiment, haptoglobin only 1.1-fold. How inflammation starts, how it ends, what CRP and SAA measure
MUMultiple speciesBmoderateT2limited target data Open access
Once a month: new syntheses, checked studies, and what has changed in the state of the evidence.