ETIOLOGY

Hemolytic anemia of newborn horse and mule foals, calves, and piglets occurs because of immune-mediated (antibody-dependent cellular cytotoxicity or type II hypersensitivity) destruction of the neonate’s red blood cells by antibodies acquired from the dam. The specific antibodies are present in the colostrum, absorbed by the neonate, and cause lysis and/or agglutination of red blood cells.

The disease is associated with the natural occurrence of inherited blood groups and only occurs if the dam is exposed to red blood cell antigens that she does not possess. In response to such exposure the dam produces antibodies directed against the foreign red blood cell epitopes. If the neonate possesses a blood type against which the dam has developed antibodies and depending on the red blood cell factor involved, red blood cell destruction can occur. The newborn acquires the red blood cell types that are foreign to the dam through inheritance from the sire. Disease occurs after birth because the fetus is not exposed to the antibodies in utero because of the epitheliochorial placentation in mares and sows and the syndesmochorial placentation of cattle. Antibodies in serum are secreted into the colostrum in the peripartum period and are subsequently ingested and absorbed by the newborn. Mares that have anti-red blood cell factor antibodies in their serum almost invariably have the same antibody in their colostrum.¹ The serum concentration of anti-Aa and Qa antibodies is highest in the last 3 months of gestation and peaks about 1 week after parturition. For neonatal isoerythrolysis to occur:

The disease occurs naturally in foals and piglets but is usually iatrogenic in calves. Exposure of the dam to foreign red cell epitopes may occur at parturition or during gestation as a result of placental lesions although most incompatible pregnancies do not result in sensitization of the mare.

Mares may also be exposed by transfusion of incompatible blood. While whole blood transfusions to mares or fillies are unusual, plasma transfusions are increasingly being used to treat failure or partial failure of transfer of passive immunity to foals. Because plasma usually contains some red blood cells, transfusion of plasma from donors possessing blood group antigens that the foal does not could immunize the filly against these factors with the potential for disease when the foal matures and gives birth. However, most commercial plasma products are harvested from donors that do not have the blood group antigens identified as being problematic.

32 blood group antigens are recognized in horses:²

Neonatal isoerythrolysis is attributable in over 90% of cases to antibodies directed against either Aa or Qa antigens, although disease due to antibodies against Ab, Dc, Da, Db, Ka, Qb, Qc, Qrs, Pa, and Ua is reported.^2,3 The presence in the mare of antibodies against Ca decreases the probability that she will develop anti-Aa antibodies.⁴

15 blood groups have been identified in pigs and the disease is recorded as occurring spontaneously associated with antibodies to the E group antigens. Historically, the main occurrence of neonatal isoerythrolysis in pigs was manmade and related to repeated vaccination against hog cholera using the pooled blood, inactivated by addition of crystal violet, of affected pigs.

The disease has also occurred in calves whose dams had been vaccinated against babesiosis or anaplasmosis using a vaccine containing bovine blood.⁵ As a result of vaccination the dam develops lytic antibodies against sire antigens, usually of the A and F blood groups, and antibodies in colostrum cause acute hemolytic anemia in the calves.

EPIDEMIOLOGY

PATHOGENESIS

The interaction between the antibody and the red cells of the newborn is followed by hemolysis with resultant anemia, hemoglobinuria, and jaundice. Following ingestion and absorption into the systemic circulation, antibodies bind to red blood cell membranes. Parts of the cell membrane of the antibody coated cells are removed from the circulation, probably by the spleen and associated reticulo-endothelial tissues, and affected cells are eventually lysed and release hemoglobin into the circulation. The affected animal develops normovolemic anemia and, if the destruction of red blood cells is sufficient, develops anemic hypoxia and dies. The reaction between red blood cells and antibodies occurs sufficiently quickly that the bone marrow is unable to compensate immediately for the loss of red blood cells. Disseminated intravascular coagulation (DIC) occurs and may contribute to the death.

Permeability of the intestine of the newborn foal to antibody disappears by 36 hours and in most cases much less. Hourly milking of the mare rapidly reduces the antibody content of the colostrum. The duration of the alimentary permeability in piglets has not been determined.

CLINICAL FINDINGS

CLINICAL PATHOLOGY

Hematological examination reveals acute anemia; erythrocyte counts, packed cell volumes, and hemoglobin concentrations are low and there is greatly increased erythrocyte fragility and sedimentation rate. Depending on the severity of the disease and its duration, there can be leukocytosis, attributable to neutrophilia and monocytosis, and the presence of nucleated red blood cells (in piglets and calves but rarely in foals). Affected mule foals, but not horse foals, are often thrombocytopenic.⁸ Isoimmune thrombocytopenia occurs rarely in foals and is not associated with neonatal isoerythrolysis, as it is in mule foals. In piglets, the erythrocyte count may be as low as 1 million/μL, the hemoglobin level below 2 g/dL, and thrombocytopenia is present. Serum biochemical analysis reveals an increased serum concentration of unconjugated bilirubin.

Diagnostic confirmation is achieved by demonstration of the presence of antibodies in the mare’s serum or colostrum that cause hemagglutination or lysis of foal red blood cells.¹¹ Tests to demonstrate hemagglutination or lysis of foal red blood cells exposed to mare serum or colostrum have been developed. Of these, the standard hemolysis test appears to have the greatest utility.¹² However, for practical purposes a positive direct antiglobulin test (direct Coombs’ test), confirming the presence of antibodies on the surface of red blood cells in a foal with anemia, provides a diagnosis of neonatal isoerythrolysis. False negative (foal has the disease but the Coombs’ test is negative) results occur occasionally because of the hemolytic nature of the antibodies. The same principle is applicable to all species. Detection of antibodies on the surface of the neonate’s red blood cells is possible using direct immunofluorescence flow cytometry.¹³ The test identifies the presence of antibodies on red cells in some instances when the Coombs’ test is negative.

The use of blood typing and other prepartum predictive tests in the prevention of the disease are discussed under control.

NECROPSY FINDINGS

In peracutely affected foals, there is marked pallor but only slight jaundice. The liver may be mildly swollen and friable but the spleen is greatly enlarged and is almost black due to the accumulation of lysed and lysing erythrocytes. In less severe cases jaundice is marked but pallor is only moderate in degree. The kidneys are usually pale and the urine is dark brown. The histopathological changes may include ischemic tubular nephrosis and periacinar hepatic necrosis and degeneration. Erythrophagocytosis is prominent and depending on the clinical course and therapeutic regime, there may be widespread hemosiderin deposition.

Hemoglobinuria is an important sign in piglets, and jaundice or port wine coloration of tissues occur constantly. The presence of blood-stained peritoneal fluid and an enlarged spleen is also typical of the disease in piglets.

TREATMENT

The aims of treatment are to:

The treatment of choice for neonatal isoerythrolysis depends on the severity of the disease. The choice of treatment should be based first and foremost on the severity of the clinical signs and secondarily on the hematocrit and red blood cell count. Foals or piglets with mild clinical signs (minimal lethargy, mild tachycardia, slight exercise intolerance) need only protection from environmental and nutritional stresses in order to recover. However, such animals should be carefully monitored to insure that their clinical condition does not worsen.

Severely affected animals need a transfusion of compatible blood to alleviate the anemia and intravenous fluids to insure adequate urine flow and minimize the risk of hemoglobin-nephrosis. In general, the younger the animal at the time the disease is evident, the more severe the disease and the more likely the need for intensive treatment. See Chapter 9 for a discussion of ‘transfusion triggers’.

CONTROL

The principles of control are:

Blood group typing permits the identification of mares that are at risk of developing antibodies against Aa or Qa antigens. If an Aa or Qa negative mare is mated to a stallion that has Aa or Qa factors then there is the potential for neonatal isoerythrolysis. If the stallion is Aa and Qa negative, then there is no risk of the disease caused by antibodies to these blood groups.

Measurement of alloantibodies in the serum or colostrum of at-risk mares is useful in identifying mares at increased risk of having affected foals. Serum from at-risk mares is collected during the last month (preferably 3–5 weeks before expected parturition) of pregnancy and examined for the presence of antibodies against the blood of the sire or, if a sample of the sire’s blood is not available, a range of blood group factors including Aa and Qa. Mares that have such alloantibodies causing hemolysis at >1:16 are not permitted to suckle at risk newborn foals. If the titer is between 1:2 and 1:16, it is measured again 1–2 weeks before anticipated parturition to determine if the titer is rising, in which case the mare is likely carrying a foal with an incompatible blood group. Equine blood typing and detection of isoantibodies is performed by specialized laboratories in a number of countries:

The jaundiced foal agglutination test (JFA) is useful in determining the compatibility of mare’s colostrum and foal’s red blood cells (Table 34.1).^16,17 In this test foal red blood cells are added to serial dilutions (1:2 through 1:32) of colostrum and the presence of agglutination examined. Agglutination at dilutions of 1:16 in horses and 1:64 in mules are considered significant and the foal should not be permitted to receive the mare’s colostrum.¹ The foal should be fed colostrum from another, compatible, mare or from a colostrum bank. The mare should be milked out every 2–4 hours until the JFA titer is less than 1:16 or for 36 hours, after which the concentration of antibodies in the milk is negligible,¹ and the foal can be permitted to suckle its dam. It is critical to the successful use of this test that it is performed before the foal is permitted to suckle the mare. If an incompatibility is detected the foal should be fed colostrum from a mare that does not have a positive jaundiced foal agglutination test to the current foal’s red cells.

Table 34.1 Method for performing the jaundiced foal agglutination test¹³

Avoidance of vaccines based on whole blood or cellular parts of blood is recommended, and if they have to be used it should be as far away as possible from parturition and should be restricted to one injection and one booster.

ETIOLOGY

The disease is acute and non-contagious. The cause of the vasculitis that characterizes purpura hemorrhagica is likely the deposition of complexes of antigen and immunoglobulin in the walls of capillaries and small blood vessels. The disease appears to be immune complex-mediated and due to a type III hypersensitivity reaction. The common association of the disease is with Strep. equi infection of the upper respiratory tract. The high concentrations of antibodies to Strep. equi M protein in affected horses, and the presence of complexes of IgA and streptococcal M protein in sera are evidence that the disease is associated with an immune reaction to streptococcal protein.^1-3 The immune complexes are not found in the serum of horses recovering from Strep. equi infection that do not have purpura hemorrhagica. However, in many instances there is no history of streptococcal infection.⁴ There is a suggestion that the disease might be associated with an adverse reaction to therapeutic drugs. Vaccination using modified live Strep. equi M protein or killed Strep. equi vaccines is strongly suspected of inducing the disease.

EPIDEMIOLOGY

Purpura hemorrhagica is an uncommon, non-contagious, sporadic disease of horses. It has been recorded in pigs and cattle. Estimates of the incidence of the disease are uncommon and imprecise. There was an incidence of 27 cases in 1438 horses housed over a 3-year period in a Swedish Army remount facility.⁵ All cases of purpura hemorrhagica followed upper respiratory infections, of which 11 were typical of strangles. Only a small proportion of horses are affected but the incidence is highest when extensive outbreaks of strangles occur, possibly because of reinfection with streptococci of horses already sensitized by previous infection.

Of 53 horses with purpura hemorrhagic treated at a referral center, 17 had been exposed to or infected with Streptococcus equi, 5 had been vaccinated with Streptococcus equi M protein, 9 had been infected with Corynebacterium pseudotuberculosis, and 5 had a history of apparently infectious respiratory disease of undiagnosed cause. Fifteen of 53 horses had no history of recent infectious disease.⁴

There is a strong suspicion among clinicians of an association between purpura hemorrhagica and vaccination.⁶ However, vaccination against Strep. equi infection has not been clearly demonstrated to be a risk factor for purpura hemorrhagica. Certainly, cases of purpura do occur in horses that have been vaccinated against strangles, but presumably there was a high probability that such horses were at increased risk of developing strangles. The consensus is that vaccination with vaccines containing M protein or avirulent Strep. equi is associated with increased risk of purpura. Edema of the lower limbs does occur after vaccination with streptococcal M protein and may represent a mild form of the disease.⁷ It is recommended by some authorities that horses with high serum antibody titers to streptococcal M protein not be vaccinated against strangles,⁶ although definitive data to support this recommendation are not available.

There does not appear to be breed, age, or sex predisposition to the disease. Horses as young as 6 months of age, and possibly younger, can be affected.

The case fatality rate with appropriate treatment is approximately 10%.⁴ Purpura accounted for 2 and 8% of 2028 and 1245 deaths among horses shipped from Great Britain and the United States, respectively, to South Africa during the Boer War.⁸ This mortality rate was before the advent of antimicrobials or corticosteroids which have presumably decreased the case fatality rates.

PATHOGENESIS

The basis of the disease process is an aseptic vasculitis of capillary walls that is accompanied by extravasation of plasma and blood into the tissues. Thrombocytopenia does not occur, nor is there a defect in coagulation in most cases. Prolonged clotting times (activated clotting time, partial thrombin time, and thromboplastin time) occur in severely affected horses with infarctive purpura hemorrhagica.³ Skin lesions predominate but other organs, including the kidney, muscles, and gastrointestinal tract, are affected.^3,9

CLINICAL FINDINGS

Affected horses are usually depressed and have reduced or absent appetite. The temperature is elevated in approximately 60% of cases,⁴ as is the heart rate. Extensive subcutaneous edematous swellings are the characteristic sign of the disease. They occur most commonly about the face and muzzle, but are often present on other parts of the body and are not necessarily symmetrical in distribution. The swellings may appear suddenly or develop gradually over several days. They are cold and painless, pit on pressure, and merge gradually into normal tissue without a definite line of demarcation. There is no discontinuity of the skin, although it may be tightly distended and even ooze red-tinged serum. Swellings about the head may cause pressure on the pharynx with subsequent dyspnea and dysphagia. Lesions in the lungs are usually not clinically apparent without radiographic or ultrasonographic examination of the chest. Extensive edema of the limbs occurs in almost all cases.⁴ Rare cases of the disease in horses do not have edema.¹⁰

Submucous hemorrhages occur in the nasal cavities and mouth, and petechiae may be present under the conjunctiva in over 80% of cases. Hemorrhage and edema of the gut wall may cause colic but in most cases there is no diarrhea or constipation. Severely affected skin, and especially that of the legs, may slough and leave granulating wounds.

Infarctive purpura hemorrhagica is an uncommon manifestation of the disease characterized by infarction of multiple tissues including the gastrointestinal tract and muscle.³ Affected horses have signs of colic and muscle swelling. The course is usually over 3–5 days and death, which is the most common outcome, is associated with severe colic and rapidly deteriorating metabolic status.

The course of the disease is usually 1–2 weeks and many animals die from blood loss, dyspnea due to laryngeal or pharyngeal swelling, and secondary bacterial infections. Relapses are uncommon among appropriately treated horses.

CLINICAL PATHOLOGY

There are no characteristic abnormalities detected on routine hematological or biochemical examinations of affected animals. Hematological changes are typically a mild anemia (usually <32% but >20%, <0.32 L/L but >0.20 L/L) with a neutrophilic leukocytosis and hyperfibrinogenemia. The platelet count is normal. Hypergammaglobulinemia can be present. There is an elevation in serum activity of creatine kinase (CK) and aspartate aminotransferase (AST) in affected horses, likely a result of muscle lesions, in approximately 25–30% of cases.⁴ Horses with infarctive purpura have marked elevations in serum activity of creatine kinase and aspartate aminotransferase, neutrophilia, and in severely affected horses, there is evidence of disseminated intravascular coagulation.³

Diagnostic confirmation is achieved by skin biopsy, especially of early lesions, and reveals leukocytoclastic vasculitis.⁷ Immunofluorescence staining of sections of skin may reveal the presence of antibodies, antigens, or complement in the walls of small blood vessels.

NECROPSY FINDINGS

Ecchymotic and petechial hemorrhages are present generally throughout the body. The subcutaneous swellings contain plasma which may be blood-stained, or sometimes whole blood. The lungs are edematous and congested. Histologically, the changes are also dominated by bland hemorrhage but a leucocytoclastic vasculitis is usually observed in scattered vessels. Sample of lung, muscle, and gastrointestinal tract, in addition to skin, should be examined via light microscopy to check for the presence of vasculitis.

Horses with infarctive purpura have dark red to black, multifocal coalescing hemorrhages in skeletal muscles.³ Hemorrhages also occur in the lungs and gastrointestinal tract. Histological examination reveals coagulative necrosis of muscle and other tissues. There is inflammation of the blood vessels.³

TREATMENT

The principles of treatment are to reduce inflammation of the blood vessels, remove the inciting cause, and provide supportive care. Because of the possibility that the disease is due to an adverse drug reaction, administration of any drugs that the horse is receiving at the time the disease develops should be discontinued.

Reduction of inflammation of the blood vessels involves mitigation of the immune response and removal of the source of the antigenic stimulus. The immune response, and its associated inflammatory reaction in blood vessels, should be treated with corticosteroids such as dexamethasone (0.05–0.2 mg/kg, IV or IM every 24 hours) or prednisolone (0.5–1 mg/kg, IM or IV every 24 hours). Prednisolone might not be as effective as dexamethasone. The dose of corticosteroid can be gradually reduced as the clinical signs improve, and the drug can be given orally. Non-steroidal anti-inflammatory drugs (phenylbutazone 2.2 mg/kg orally or IV every 12 hours, or flunixin meglumine 1.1 mg/kg orally or IV every 12 hours) may reduce inflammation and provide some analgesia.

Removal of the source of the antigenic stimulus of the disease is difficult, especially in cases when an antecedent infection or disease is not readily identified. On the assumption that purpura hemorrhagica is often a sequela to Strep. equi infection, and the suspicion that occult Strep. equi infection is present and the source of antigen associated with the disease; affected horses are usually treated with penicillin (procaine penicillin, 20 000 IU/kg, IM every 12 hours, or potassium penicillin, 20000 IU/kg, IV every 6 hours) until the clinical signs resolve. Treatment with antibiotics might need to be continued for as long as 20 days.⁴

Supportive care includes bandaging of swollen limbs, care of wounds, hydrotherapy, and intravenous fluid administration. Swelling of the head and pharynx may necessitate placement of a nasogastric feeding tube to permit enteral feeding of dysphagic horses. Respiratory distress can develop very rapidly and emergency tracheotomy may be required to relieve respiratory distress and prevent asphyxiation.

CONTROL

There are no specific preventive measures. However, control and prevention of upper respiratory tract infections in horses should lead to a reduction in the incidence of purpura hemorrhagica. Careful consideration should be given to the use of vaccines containing streptococcal M protein or avirulent Strep. equi in horses at low risk of developing strangles. Although the relationship between M protein containing vaccines and purpura hemorrhagica is not definitive, circumstantial evidence and the opinion of authorities in the field support such an association. Measurement of serum antibodies to M protein might be useful in determining the need for vaccination of horses in endemic areas or at high risk.⁶ Horses with antibody titers >1:3200 should not be vaccinated.⁶

ETIOLOGY

The disease is caused by type I (immediate) hypersensitivity to salivary antigens introduced into the skin by the bites of sandflies and other insects. There may be a lesser role for type IV (cell-mediated) hypersensitivity in the disease. Culicoides brevitarsus is the cause in Australia,¹ C. pulicaris in the United Kingdom and Europe,² and C. obsoletus in Canada.³ Stomoxys calcitrans, the stable fly, and Simulium spp. cause the disease. The distribution of the skin lesions is related to the feeding habits of the inciting insect. For instance, C. pulicaris has a predilection for landing at the mane and tail, and this is where the lesion is most commonly seen.

EPIDEMIOLOGY

The prevalence of the disease varies depending on environmental factors, and possibly characteristics of the local horse population. Up to 60% of horses are reported affected in areas of Queensland, Australia, 22% in Israel, and 18% of Icelandic horses in Norway.⁴ The prevalence in Switzerland is very low in regions above 1000 m and 1.6% in lower areas.⁵

The disease is quite common worldwide in areas where hot and humid summer weather favors the causative insects: Sweden, the United Kingdom, Japan, Israel, Hong Kong, North America, Australia, the Philippines, India, and France. Most cases occur during summer and lesions disappear during cooler weather. Lesions disappear when the horses have been stabled in insect-proof barns for several weeks or are moved outside the geographical range of the inciting insect.

The disease is characteristically sporadic and affects only a few of a group of horses. However, because the predilection to the disease is likely inherited, there may be multiple cases among related animals on a farm.^5,6 The prevalence of the disease increases with age; 3.4% of Icelandic horses 1–7 years of age compared to 32% of horses older than 14 years were affected.⁴

PATHOGENESIS

Reaginic antibodies (IgE) produced in response to exposure to proteins in insect saliva bind to mast cells in the skin and, when exposed to the antigen, are associated with degranulation of the mast cell. Horses with sweet itch have IgE antibodies that react with constituents of the salivary gland of Culicoides sp., whereas horses that do not have the disease have IgG, but not IgE, antibodies against Culicoides salivary gland antigens.⁶ Horses that have not been exposed to Culicoides sp. do not have either antibody to the insect salivary gland antigen.⁶ Degranulating mast cells and intradermal or subcutaneous lymphocytes release various vasoactive substances and cytokines that cause inflammation and accumulation of eosinophils in the skin of affected areas and eosinophilia.⁷ The distribution of the lesions on patients reflects the insects’ preferred feeding sites. Ponies with seasonal allergic dermatitis have greater numbers of circulating CD5+ and CD4+ T-lymphocytes than do normal animals.⁸ Increased numbers of CD3+ T-lymphocytes, most of which are CD4+, and eosinophils are present in the skin of affected ponies after injection of Culicoides antigen.⁸ Furthermore, eotaxin and monocyte chemoattractant protein (MCP) 1, but not MCP-2 or MCP-4, mRNA expression is upregulated in skin biopsies of sweet itch lesions, demonstrating a mechanism for accumulation of eosinophils and T-2 lymphocytes in the lesions.⁹

CLINICAL FINDINGS

Lesions are usually confined to the base of the tail, rump, along the back, withers, crest, poll, ears and, less commonly, ventral midline. In severe cases the lesions may extend down the sides of the body and neck and onto the face and legs.

Pruritis is intense, especially at night, and the horse scratches against any fixed object for hours at a time. In the early stages slight, discrete papules, with the hair standing erect, are observed. Constant scratching may cause self mutilation, severe inflammatory lesions, and loss of hair. Scaliness and loss of hair on the ears and tail-base may be the only lesions in mildly affected horses.

CLINICAL PATHOLOGY

Affected animals have eosinophilia and thrombocytosis.

Diagnosis is facilitated by skin biopsy, fungal culture, and parasitological examination of skin scrapings, and intradermal sensitivity testing. Skin biopsy of early lesions, before trauma masks the true picture, reveals edema, capillary engorgement, and eosinophilic and mononuclear perivascular infiltration. Fungal culture and parasitological examination of skin scrapings are useful only in that they rule out dermatophycosis, onchocerciasis, and strongyloidosis. Intradermal skin testing demonstrates immediate and delayed sensitivity reactions to extracts of Culicoides and Stomoxys spp.¹

TREATMENT

The principles of treatment are removal of the inciting cause and suppression of the hypersensitivity reaction.

Removal of the inciting cause is achieved by preventing horses from being exposed to the inciting insects. This can be achieved by relocating the horse to a geographical region where the insects do not occur, stabling of the horse in an insect-proof stable during the periods of the day (early evening) when the insects are most active, or applying agents that kill the insect or otherwise prevent them from alighting on and biting the horse.

Suppression of the immediate hypersensitivity reaction or its sequelae can be achieved by administration of corticosteroids (prednisolone, 1 mg/kg every 24 hours initially then reducing to as low a maintenance dose as possible). Theoretically, hyposensitization may be effective, but the only controlled clinical trial to date did not demonstrate a beneficial effect, although the placebo effect on the owners was impressive.¹⁰

CONTROL

Prevention of the disease necessitates protection against sandfly bites by stabling in insect-proof quarters. Continuous spraying of the horses with insecticides or repellents may be of some value. A 4% permethrin pour-on gives effective protection. Most horses need only one application a week; others need an application every second day.¹¹

ETIOLOGY

Heaves, previously referred to as chronic obstructive pulmonary disease, is caused by inhalation by susceptible horses of dust particles found in barns, bedding, and feed materials such as dusty hay. The inhaled particles include endotoxin, mites, plant debris, inorganic materials, and conidia and fragments of molds. Faenia rectivirgula (formerly known as Micropolyspora faeni), Aspergillus fumigatus, and Thermoactinomyces vulgaris are molds commonly associated with respiratory disease in susceptible horses, as evidenced by experimental studies involving inhalation of mold or mold fragments by horses.¹ Molds contain a number of inflammatory substances including various allergens, glucans, mycotoxins, and proteases and it is not clear which of these agents are the inciting cause of heaves.² Furthermore, dust containing mold also contains endotoxin. Endotoxin contamination of molds contributes to the airway response to inhalation of preparations of molds used in experimental studies³ and inhalation of endotoxin alone produces airway inflammation and impaired respiratory function in horses in a dose-dependent manner, with heaves-susceptible horses having an exaggerated response at lower doses.⁴ However, the response to endotoxin is less than that of susceptible horses exposed to hay dust containing endotoxin indicating that endotoxin alone is not sufficient to cause the clinical signs of heaves. Other compounds in hay dust are integral to the development of heaves.⁵

It is emphasized that there is not one causative agent acting alone but rather a range of agents that, when inhaled in sufficient concentration by susceptible horses, induce airway disease. It is likely that heaves is associated with the potentiating interactions among several agents present in barn or hay dust and is not simply a response to one agent.^3,6 The mechanisms underlying development of airway inflammation and respiratory dysfunction are provided under ‘Pathogenesis’. Viral infections and 3-methylindole intoxication are not considered important causes of heaves.⁷

EPIDEMIOLOGY

PATHOGENESIS

Susceptible horses, when exposed to adequate concentrations of respirable dust in the breathing zone, develop airway inflammation, excess mucus accumulation, and respiratory dysfunction within hours to days of exposure. The putative inciting agents include fungal spores and debris, endotoxin, and particulate matter; however, the exact role for each of these agents is unclear, partly because of the difficulty in obtaining uncontaminated material. For instance, fungal material is commonly contaminated with endotoxin which has a synergistic interaction with hay dust in inducing airway neutrophilia in susceptible horses.³

The mechanisms underlying these responses to inhalation of dust are not well defined but can be considered in the contexts of immune and inflammatory responses, mucus secretion, and pulmonary dysfunction.

CLINICAL FINDINGS

The degree to which horses are affected varies considerably. Minimally affected horses have airway inflammation evident on endoscopic or cytological examination of the airways, but few other signs on physical examination, whereas severely affected horses have very obvious clinical signs.

The usual history is that of chronic cough in a stabled horse.¹² Typically, the disease is precipitated by exposure to hay and stabling, and disease remission occurs in most horses when pastured and removed from hay. There may be a history of reduced exercise tolerance.

Affected horses are usually bright and alert and have a normal appetite and rectal temperature. Severely affected horses appear anxious and have a greatly increased respiratory effort.

Coughing is common in horses with heaves, although it is neither particularly specific nor sensitive as an indicator of the disease. Coughing may consist of a single cough every few seconds to minutes or there may be a paroxysm of coughing. The cough can also be elicited by digital massage of the larynx and proximal part of the trachea because horses with airway inflammation have increased sensitivity of the cough reflex. Stimulation of the larynx or proximal trachea by digital massage does not elicit coughing in normal horses. The cough becomes more pronounced and wheezing with exercise. It also occurs more frequently when the horse is exposed to cold air, physical activity, excitement, and when placed in a dusty environment, or if dusty feed is offered. The amount of coughing, which must be counted over at least 15 minutes and preferably 1 hour for accurate determination of its severity, correlates closely with the amount of mucus in airways, maximal change in pleural pressure (a measure of bronchoconstriction), and neutrophil count in bronchoalveolar lavage fluid.⁴¹ Coughing is more frequent in horses with heaves, and affected horses often have paroxysmal coughing especially after barn cleaning and feeding.

An intermittent, bilateral mucopurulent to serous nasal discharge is a common sign in affected horses.

The resting respiratory rate is increased from a normal of 12/min up to 24–36/min. There is a pronounced effort during expiration and markedly affected horses have an obvious abdominal component to respiration. Normal horses have a biphasic pattern of airflow during inspiration and expiration while affected horses lack the second phase of respiration.⁵³ Longstanding cases develop a ‘heave-line’ in the flank due to hypertrophy of the abdominal oblique musculature. It is evident as a trough or furrow along the costal arch. In advanced cases the nostrils may be visibly dilated during inspiration and the force of the expiratory effort causes the anus to protrude.

Heart rate is commonly within the normal range or only slightly increased. In horses with heaves, the heart rate is significantly higher during exercise than in healthy horses.

Abnormal lung sounds are one of the most frequent abnormalities detected on clinical examination and the sensitivity of this finding can be increased from 70% to almost 90% by auscultating the thorax while the horse breathes for 60 to 120 seconds with an airtight plastic bag over its nostrils.¹² The bag should be large enough to enable the horse to breathe unhindered (10–15 L) and should not leak. Accumulation of carbon dioxide in the bag increases the horse’s respiratory rate and tidal volume and accentuates lung sounds. Auscultation of the lungs in the early stages of the disease may reveal only a slight increase in the amplitude of normal breath sounds. Abnormal lung sounds become audible as the disease progresses. Wheezing and crackling sounds occur at the end of inspiration and the end of expiration. These abnormal sounds are audible over most of the lung but are usually easiest to detect over the upper one-half of both lung fields. Auscultation of the trachea usually reveals moist sounds characteristic of fluid in the trachea. Some affected horses have quieter than expected lung sounds.

Percussion of the thorax may reveal an increase in the area of resonance by as much as one to two intercostal spaces caudally. However, the area of resonance delineated by percussion is too labile and ill-defined to be of diagnostic value.

Endoscopic examination of the upper airways, trachea, and bronchi reveals an abundance of mucopurulent material in the trachea which, in severe cases, is also present in the nasopharynx. The amount of mucus can be graded on a 0–5 scale⁴¹:

Observation of tracheal mucus of grade 4 or 5 has a high specificity (92%) but low sensitivity (52%) for detection of heaves.⁴⁷

Radiographic examination of the thorax usually reveals evidence of bronchial disease with some evidence of interstitial disease. Radiography is more useful in ruling out other diseases, such as granulomatous or interstitial pneumonia, than in confirming heaves.

Sophisticated techniques for measuring pulmonary function, such as determination of tidal flow–volume loops, nitrogen washout or forced expiratory flow–volume loops, may identify mildly or subclinically affected animals but have limited day-to-day clinical utility.^53-55

Measurement of pleural pressure changes by insertion of an esophageal balloon is relatively simple and may be useful in monitoring response to treatment. Affected horses have pleural pressure changes during respiration greater than 6 cm H₂O.⁵⁴ Administration of atropine (0.02 mg/kg, IM or IV), isoproterenol (isoprenaline), or a β₂-adrenergic agonist such as terbutaline (0.04 mg/kg, PO) reduces the maximal change in pleural pressure of horses with heaves.

The course of the disease is dependent on the removal or continual presence of the precipitating cause. If the cause is removed in the early stages, complete recovery can occur. In the continual presence of the precipitating cause relapses occur commonly or the disease becomes progressive and affected horses become severely incapacitated. Bronchiectasis, evident on radiographic examination of the thorax, develops in horses with heaves of prolonged duration.⁵² With conscientious management and adequate housing, breeding animals and hunters or showjumpers with heaves can remain useful for many years.⁵⁶

CLINICAL PATHOLOGY AND SPECIAL EXAMINATIONS

There are no significant changes in the hemogram or serum biochemistry of affected horses. The PaO₂ is below normal in moderately to severely affected horses and the PaCO₂ is usually normal although it may be increased in severely affected horses. Blood oxygen tension measurements should be corrected for the temperature of the animal and the altitude. At approximately sea level, PaO₂ values of normal horses are usually greater than 90 mmHg (12 kPa), whereas affected horses have PaO₂ less than 82 mmHg (10.9 kPa). With increases in altitude the values in both normal and affected horses decrease.⁵⁴ The normal hypoxemia that occurs in horses during intense exercise is exacerbated by heaves.

Bronchoalveolar lavage fluid from affected horses during symptomatic episodes has a relative neutrophil count greater than 5 to 10%, and usually over 50%, of the absolute nucleated cell count.^12,57,58 It is recommended that horses not be considered to have airway inflammation unless >15% of cells in bronchoalveolar lavage fluid are neutrophils. During periods of remission, the bronchoalveolar lavage fluid of previously affected horses is not different from that of normal horses. Absolute nucleated cell counts in bronchoalveolar lavage fluid of affected horses are reported^12,57,58 but the values depend on the collection technique used.⁵⁹ The relative proportions of macrophages and lymphocytes in bronchoalveolar lavage fluid of affected horses are lower than those of normal horses. Eosinophil numbers in bronchoalveolar or tracheal aspirate fluid of affected horses may be mildly elevated (up to 10%), but are usually low (<3–5%). Higher values should raise the index of suspicion for Dictyocaulus arnfieldi or Parascaris equorum infestation. Aspirates of tracheal fluid reveal a profound neutrophilia (>90%).⁵⁷

Measurement or identification of precipitins in serum of horses is not useful in identifying horses with heaves.⁶⁰

Intradermal testing using putative allergens has been investigated as a means of identifying horses with heaves or of identifying antigens with which to hyposensitize affected horses with variable or undetermined efficacy.^1,61,62 Retrospective examination of records of horses with a history of heaves suggests that they are more likely to react, and react to a larger number, of intra-dermally injected allergens than horses without a history of heaves.⁶³ However, reactions to individual allergens cannot be used to determine hypersensitivity to particular allergens, although it is suggested that overall patterns of reactivity, with a history of exposure of the horse to these allergens, might be useful in guiding management of affected horses.⁶³ Contrary to these results, a prospective study demonstrated that horses with heaves did not have a greater rate of reaction to intradermal skin tests than did horses not affected by heaves.⁶⁴ Intradermal testing did not distinguish clinically relevant reactions from those that were not clinically relevant.⁶⁴ Horses with heaves have greater sensitivity to intradermal injection of histamine, which is commonly used as a positive control, than horses without heaves.⁶⁵ Overall, intradermal skin testing is neither useful in detecting horses with heaves nor in determining hypersensitivity to particular allergens in individual horses. Results of such testing might be useful in management of horses, but this has not been demonstrated. The usefulness of intradermal skin testing and subsequent administration of preparations of antigens selected on the basis of intradermal testing, in an effort to hyposensitize horses with heaves, has not been determined. The apparent lack of efficacy of intradermal testing might be because the extent of reactivity to intradermal injection of mold preparations does not correlate with the severity of pulmonary dysfunction after inhalation of the same preparation in horses with heaves.¹

Lung biopsy demonstrates peribronchiolar lymphoplasmacytic inflammation, goblet cell metaplasia, alveolar fibrosis, and bronchial lumen exudate and neutrophils.¹² The severity of bronchiolar neutrophil and mast cell infiltration correlates well with the severity of the clinical signs.¹²

NECROPSY FINDINGS

The major findings are restricted to the lungs, which are pale, voluminous, and do not collapse when the chest cavity is opened. The tissue damage is primarily centered on airways which are less than 2 mm in diameter. Microscopically, a variable degree of alveolar emphysema is accompanied by a chronic bronchiolitis featuring diffuse epithelial hyperplasia, goblet cell metaplasia, peribronchiolar fibrosis, and cellular infiltration by lymphocytes, plasma cells, mast cells, and sometimes eosinophils. Plugs of mucus with entrapped neutrophils often occlude bronchiolar lumina.

DIAGNOSTIC CONFIRMATION

confirmation of the disease is based on the presence of a history and clinical signs consistent with the disease, in particular the response to stabling and pasturing, and demonstration of reversible airway obstruction. Objective confirmation can be achieved by measuring the response of maximal changes in pleural pressure in response to bronchodilator drug (atropine or glycopyrolate) administration.

TREATMENT

The principles of treatment are:

Heaves is an inflammatory disease caused by inhalation exposure to inciting agents. Bronchoconstriction is secondary to inflammation. Control of the disease is based upon preventing inhalation of inciting agents and suppression of inflammation by administration of corticosteroids. Relief of bronchoconstriction should be necessary only during acute exacerbations of the disease, and administration of bronchodilatory drugs for more than several days is not optimal treatment in most horses. Drugs used in the treatment of heaves are summarized in Table 34.2.

It is essential that the horse is not exposed to the inciting agents and irritant substances that could provoke or worsen the disease. Even relatively brief exposure of susceptible horses to the inciting agents, such as can occur if a horse is brought into a poorly ventilated barn to be fed, can result in airway hypersensitivity and the development or maintenance of clinical signs. Affected horses should be moved to a clean environment, ideally pasture, in which the concentration of airborne allergens is reduced to an absolute minimum. If the horse cannot be kept at pasture, then it should be housed in a well ventilated barn (see ‘Control’ for details), bedded with clean wood shavings or shredded paper, and fed a complete pelleted ration. If affected horses are fed hay, it should be thoroughly wetted to minimize the release of spores. Remission of clinical signs can be expected in 4–21 days if the environmental changes are adequate.⁶⁷ This may be all that is necessary to control the disease in many horses.

CONTROL

Housing horses in stalls with good air quality is essential in reducing the occurrence of the disease. Adequate ventilation is critical in maintaining good air quality in stalls. Few horse housing units have adequate ventilation^19,22 although a well-designed individual box stall can meet the needs of the horse both for air hygiene and thermal comfort.¹⁹ Many horse barns have inadequate open space for ventilation in still air conditions when the doors are closed at both ends of the building. When the release rate of spores is low, ventilation rates of 4 air changes per hour are satisfactory. However, suggested minima are 8–10 air changes per hour, airspace of 44 m³/head, and floor space of 9.2 m²/head.²³ In practical terms, if the upper half of the stable door is open, and faces open air and not into a barn, the natural ventilation should exceed the minimum specifications.²³ Hay and dusty feed materials should not be stored above stalls or in the same airspace as horses. Bedding should be changed frequently, preferably daily. Use of cardboard as bedding material is effective as part of an overall regimen to improve air quality.⁹²

A portable slit sampler is an accurate, quick, and simple semiquantitative method of assessing the mold contamination of source materials such as hay, straw, and other feeds and bedding collected from stables.¹⁸ The health hazard posed by any moldy source material depends on the types of organisms present and their abundance. The size of the respirable challenge from heated hays and straws arises from the prolificacy of the species involved and their small spore size. The highest respirable challenges are from the presence of thermotolerant and thermophilic mold species. The most critical factors in determining the microbial development in plant-based materials are water content and thermal environment. Hay baled at 15–20% moisture heats little, it is virtually dust-free and contains few spores. Baling hay with 20–30% moisture leads to temperatures of up to 35–45°C. At these temperatures, hazardous contamination may develop with the appearance of thermotolerant fungi and actinomycetes.¹⁸ The heaviest contamination of hay and straw occurs with baling at 35–50% moisture when spontaneous heating up to 50–60°C may occur. Microscopically, these hays show large numbers of fungal spores in the 2–5 μm size range.