Definition and Etiology

Sinusitis refers to inflammation of the paranasal sinuses, from primary microbial infection or secondary bacterial infection associated with dental disease or other sinus disease. Lined by respiratory mucosa, the paranasal sinuses are at risk for developing diseases affecting the respiratory tract.¹¹⁷³ Sinus empyema, accumulation of pus within a sinus cavity, may result from bacterial or viral infection.

Incidence of sinusitis is relatively low, likely less than 0.5% of disease in equine practice.¹¹⁷⁴ In 256 horses with sinonasal disease, primary sinusitis (24%) and dental disease—associated sinusitis (22%) occurred with similar frequency and were the most common cause of sinusitis, whereas sinus cysts (13%), sinonasal neoplasia (8%), ethmoid hematoma (8%), sinonasal trauma (6%), mycosis (5%), and rostral maxillary tooth infection (4%) were the other more common causes.¹¹⁷⁵ The frontal and maxillary sinuses are more commonly involved, with the corresponding conchal sinuses and the sphenopalatine sinus affected less often. Median age of horses affected with sinonasal disease is 7 to 11 years.¹¹⁷⁵

Acute or chronic upper respiratory tract infections of viral or bacterial origin can result in primary sinusitis; Streptococcus species infection is the most common. Maxillary sinusitis caused by disease of the third (caudal roots) through sixth cheek teeth (modified Triadian numbers 109 to 111, 209 to 211) often results from alveolar periostitis, patent infundibula, or fractured or split teeth.¹¹⁷³ Dental defects permit access of food material or bacteria to the tooth root and sinus cavity, with extension to the frontal sinus likely through the frontomaxillary opening. Reported neoplasms include osteoma, osteosarcoma, adenocarcinoma, lymphosarcoma, squamous cell carcinoma, and fibroma.^1176-1179

Clinical Signs and Differential Diagnosis

Signs vary depending on cause, location, and extent of sinus involvement, with unilateral nasal discharge (serous, mucoid, mucopurulent, purulent) being the most frequent sign.^1175,1180 Physical examination of a horse with sinusitis should include observation for epiphora, facial asymmetry, altered nasal airflow, abnormal breath odor, mandibular lymphadenopathy, and sinocutaneous fistula.

Nasal discharge is typically unilateral because the nasomaxillary opening is located rostral to the caudal edge of the nasal septum. Intermittent or continuous, nasal discharge need not be related to a previous upper respiratory infection. Malodorous discharge is commonly associated with dental sinusitis and sinonasal mycosis, whereas mucopurulent discharge is more commonly associated with sinus cysts.¹¹⁷⁵ Sanguineous discharge is common with ethmoid hematoma and sinonasal trauma; however, guttural pouch mycosis, pulmonary hemorrhage, nasal turbinate necrosis, and neoplastic or granulomatous lesions should also be considered. Other differential diagnoses of nasal discharge should include guttural pouch empyema or mycosis; acute pharyngitis (strangles, rhinopneumonitis, and influenza); neoplasia or necrosis of the turbinates; ethmoid hematoma; and pulmonary disease.

After nasal discharge, submandibular lymphadenopathy is the most common sign, particularly when microbial infection is a component of sinusitis. Facial swelling occurs with frequency similar to that of lymphadenopathy, typically when sinus drainage is obstructed or there is an expansile mass within the sinus, or as an acute sign after facial trauma. Occasionally exophthalmos can occur with marked sinus distortion.

Patency of the nasomaxillary opening generally precludes facial distortion. Loss of patency occurs when inspissated exudate accumulates or mucosal lining tissue reaction obstructs the opening. Expansion of the sinus may result in reduced nasal airflow, particularly ipsilaterally, caused by distortion of the architecture of the nasal passages, and in such instances abnormal respiratory noise rather than nasal discharge may be the primary presenting sign.¹¹⁷⁸

Epiphora occurs if there is nasolacrimal duct involvement from trauma, or compression or destruction of the duct from underlying sinus disease. Approximately one third to one half of horses with dental sinusitis, sinus cyst, and sinonasal trauma have epiphora as a presenting sign.¹¹⁷⁵

Percussion of the affected sinus may reveal dullness or pain, although normal resonance does not preclude the possibility of sinusitis. If there is bone thinning over gas above a fluid line (as can occur with some maxillary sinus cysts), percussion may elicit increased resonance.

Careful examination of the oral cavity for signs of dental or periodontal disease should be performed when any of these signs are present. Particular attention should be paid to examination of the occlusal surface of the teeth with a very fine dental pick (e.g., 22-gauge needle); however, it should be recognized that periapical abscess formation can occur without defects in the occlusal surface. Accuracy of diagnosis of dental disease is substantially enhanced by use of intraoral endoscopy and CT.

Clinical Pathology

The hemogram in animals with sinusitis generally remains within the normal range, although acute sinusitis of infectious origin may be associated with neutrophilia. With chronic sinusitis there may be concurrent hyperfibrinogenemia. Sinus fluid obtained by percutaneous centesis should be examined cytologically (including a Gram stain) and submitted for microbial culture and susceptibility testing to differentiate among bacterial, fungal, and neoplastic disease. Flecks of feed material indicate sinusitis secondary to dental abnormalities.

Laboratory Aids and Definitive Diagnostic Tests

A presumptive diagnosis of sinusitis can be made from the physical examination and associated clinical signs.^1175,1181 Procedures most helpful in establishing a diagnosis of sinus disease in 85 horses were radiography (92%), endoscopy (38%), percutaneous centesis (21%), and examination of the oral cavity (20%).¹¹⁷⁷

Endoscopic examination of the nasal cavity may permit identification of exudate draining from the nasomaxillary opening or in advanced cases may reveal distortion of the nasal cavity secondary to sinus enlargement. Middle meatus examination is an important component of endoscopic examination. Diagnosis beyond recognition of the potential source of the nasal discharge is limited unless there is an obvious mass or abnormal tissue is observed. Sinoscopy can be accomplished by using an arthroscope¹¹⁸² or flexible endoscope¹¹⁸³ inserted through small trephine holes in either the maxillary or the frontal sinus. Examination of the rostral compartment of the maxillary sinus requires a separate portal unless the bony septum between the rostral and caudal compartments has been destroyed. Observation may be limited by fluid or tissue especially in the rostral compartment of the maxillary sinus but can potentially be enhanced after sinus lavage and aspiration.

Standard radiographic projections include the standing lateral, dorsoventral, and right and left oblique views.^1184,1185 Radiographic findings include fluid lines within the sinus, space-occupying soft-tissue densities, areas of decreased bone density, fractures, or dental abnormalities. Dental root disease is identified radiographically by a loss of continuity of the lamina dura and lysis of the tooth root or surrounding bone, combined with new bone formation and cement deposition.¹¹⁸⁴ Anatomy of normal equine skulls as demonstrated by CT and MRI has been described.^1186-1188 Diagnostic accuracy, especially determination of the extent of involvement of structures within the skull, can be enhanced by CT.^1189,1190 CT imaging enhances diagnostic accuracy of tooth involvement.¹¹⁹⁰ Findings associated with dental caries include hypoattenuation of cementum, destruction of enamel, and filling of the infundibular cavity with gas, whereas with dental decay there is gas accumulation in the root area or fragmentation of the root, and sinus mucosal thickening. Additional changes with sinusitis typically involve the maxilla with endosteal sclerosis, thickening, periosteal reaction, and deformation, especially involving the facial crest.¹¹⁹⁰ Scintigraphic examination may improve specificity in identification of dental involvement in sinusitis.^1191,1192

Percutaneous sinus centesis may provide a definitive diagnosis and allow an avenue for subsequent therapy. Cytologic evaluation, with concurrent microbial culture and antibiotic susceptibility testing, may elucidate the cause of the sinusitis.¹¹⁷⁷ Isolation of a single organism such as Streptococcus species generally indicates a primary sinusitis, whereas polymicrobial infection is more compatible with sinusitis of dental origin. Visual examination of the oral cavity, especially intraoral endoscopy, and careful probing of the occlusal surfaces with a dental pick may identify dental abnormalities.

Necropsy Findings

Affected sinuses contain fluid or tissue of variable color and consistency. Fluid character ranges from clear and odorless with cystic sinus disease to white, yellow, or green purulent fluid with a variable, but often putrid, odor in sinusitis resulting from other causes. Sinusitis of dental origin has a characteristically pungent and unpleasant odor. Granulomatous lesions have been reported to appear as large lobular gelatinous masses filling the sinus cavity. The gross appearance of neoplastic lesions within the sinus cavity depends on the type of neoplasm. Neoplasia may cause surrounding soft-tissue and bony destruction, whereas large, benign space-occupying lesions may result in distortion of the nasal turbinates and nasal septum, as well as external facial bone distortion.

Treatment and Prognosis

Not infrequently, horses with a chronic mucopurulent nasal discharge from sinusitis have a history of response to antimicrobial therapy, followed by recurrence of the discharge after antibiotic therapy ceases. Definitive diagnosis of sinusitis can be accomplished using the techniques described earlier. Sinoscopy permits examination of the paranasal sinuses and in some instances facilitates treatment.^1184,1193

Suggested treatment for primary sinusitis or empyema involves daily lavage of the sinus through a percutaneous centesis site with 1 L of saline, to which a broad-spectrum antibiotic or antiseptic has been added. Once the results of culture and susceptibility testing are available, the appropriate antibiotic should be administered locally in the flush solution, as well as systemically, for 14 days. Resolution or reduction in the volume of nasal discharge is an indication of successful therapy. If little progress is made after 10 to 14 days or if drainage recurs, sinusotomy (trephination or bone flap technique, standing or recumbent anesthetized) may be required to resolve the condition.¹¹⁹⁴ The prognosis is generally favorable if primary sinusitis is not chronic and if the mucous membrane is not markedly thickened.¹¹⁹⁵ Chronic sinus disease (longer than 6 months) carries a poor prognosis, and, for resolution to occur, surgical removal of the thickened, infected mucous membrane is required; creation of sinonasal drainage by sinus fenestration is also recommended. Isolation of Pseudomonas species from a sinus aspirate generally indicates an unfavorable prognosis.¹¹⁸⁰

Sinusitis that results from secondary factors is generally not responsive to medical management. Such conditions include diseased teeth, granulomas, or neoplasia; surgical removal of the inciting cause is required, and adjunctive treatment may be required. The prognosis for sinusitis associated with dental abnormalities is usually favorable once the diseased tooth has been removed.^1195,1196 If the periodontal ligament is intact, endodontic therapy can be used to save the tooth. This is accomplished by surgical apicoectomy and retrograde occlusion of the root canal after debridement of the pulp. In geriatric horses with dental-associated sinusitis, when economic constraints limit surgical options, sinusotomy and periodic sinus lavage and antibiotic therapy have been used successfully to manage nasal discharge.

The prognosis for resolution of granulomatous lesions is generally guarded and depends on surgical access and extent of the lesion. Neoplastic lesions are often well established and have metastasized, either locally or regionally, by the time they become clinically apparent; the prognosis for resolution is generally guarded to poor.¹¹⁷⁶ In a series of 16 horses with sinus neoplasia, 11 were euthanized because of the extent of the lesion, four lesions recurred after surgical removal, and one horse with squamous cell carcinoma was successfully treated and had no recurrence at a 2-year follow-up evaluation.¹¹⁷⁷

Prevention and Control

Prevention of sinusitis in horses is difficult because of the variety of causes. Isolation of horses from those with upper respiratory bacterial or viral diseases may be of benefit in preventing primary sinusitis. Regular dental care and a proper diet may help circumvent sinusitis caused by dental abnormalities, although many cases most likely result from a variety of causes not yet defined or over which the owner or veterinarian has no control.

Definition and Etiology

Also termed progressive ethmoidal hematoma¹¹⁹⁷ and hemorrhagic nasal polyps,¹¹⁹⁸ ethmoid hematomas are slowly expanding angiomatous masses that appear to originate principally from the mucosal lining of the ethmoid conchae. Smaller hemangiomas arising from the mucosal lining of the frontal, maxillary, and sphenopalatine sinuses have been recognized, but the relationship between these benign endothelial tumors and ethmoid hematoma is uncertain. A relationship between paranasal sinus cysts and ethmoid hematoma has been suggested,¹¹⁹⁹ but distinctly different histologic features characterize each disorder,¹²⁰⁰ seemingly arguing against a common cause and the likelihood that these lesions are variants of each other. The cause of ethmoid hematoma is unknown, and it remains a relatively uncommon condition.^1201-1205 Although reported in a 4-week-old foal and in 3-year-old horses, most affected horses are older than 8 years—generally thoroughbred, Arabian, or warmblood horses.^1201-1207

Clinical Signs

A blood-tinged nasal discharge with intermittent epistaxis from one or both nostrils is the most common clinical sign.¹²⁰⁸ Unilateral or bilateral, epistaxis varies from blood-tinged mucoid or mucopurulent discharge to blood spots or a trickle of blood. Fulminant or fatal epistaxis as can occur with guttural pouch mycosis is uncommon. If the hematoma occupies the choana(e) or nasal cavity, a mucopurulent, occasionally malodorous nasal discharge with some blood discoloration is more commonly seen. Typically these horses have a history of abnormal respiratory noise, both inspiratory and expiratory, especially during exercise. With nasal cavity involvement, airflow is usually reduced or may be absent on the affected side. Facial distortion or asymmetry is uncommon and is more likely to occur when the hematoma occupies the frontal and maxillary sinuses. Less commonly there may be an associated history of coughing, choking, ptyalism, increased respiratory effort during resting breathing, and either head shyness or head shaking.¹¹⁹⁷ If the hematoma has expanded into the paranasal sinuses, percussion yields a dull sound.

Laboratory Aids and Definitive Diagnostic Tests

Confirmation requires endoscopy of the ethmoid conchae and skull radiography (Figs. 31-45 and 31-46); however, the origin and extent of the mass can be delineated more accurately by computed tomographic examination of the skull. Sinoscopy may be of diagnostic value in horses with ethmoid hematoma involving the paranasal sinuses without protrusion into the nasal cavity. Rarely, ethmoid hematoma infiltrates the nasal conchae; these lesions, identifiable by CT, may be missed by sinoscopy. Because ethmoid hematoma is bilateral in approximately 30% of affected horses, it is prudent to examine both the left and right ethmoidal conchae. The ethmoidal labyrinth is visible approximately 25 cm from the nares, with the endoscope positioned in the ventral nasal meatus and the viewing tip deflected dorsally. The rostral surface of the ethmoidal concha does not protrude beyond the caudal nasal cavity and has a bulbous shape and a moist pink to pale red mucosal covering.

Fig. 31-45 Endoscopic view of left ethmoturbinate. A, Normal appearance. B, Ethmoturbinate obscured by an ethmoid hematoma.

Fig. 31-46 Lateral radiograph demonstrating an ethmoid hematoma.

Beyond the rostral surface the numerous pillars that form the ethmoidal conchae and separate the ethmoidal spaces (cellulae ethmoidales) are visible. Ethmoidal hematomas that project into the ventral meatus or through the choana into the nasopharynx often obscure the ethmoidal concha. Occasionally, unilateral ethmoid hematomas that have expanded into the nasopharynx may protrude into the contralateral ventral meatus, obscuring the view of the ethmoidal labyrinth on that side. The origin of hematomas that expand dorsally into the frontal sinus may not be visible on endoscopy, but hemorrhage that originates deep to the visible portion of the ethmoidal conchae may be evident or may be noticed from the region of the nasomaxillary opening in the middle meatus. Visible ethmoid hematomas can vary in color from deep red to red-purple or may have a yellow-brown or yellow-green-brown to bronze color. The surface is irregularly rounded, with small punctate hemorrhages or erosions, and may be partially covered in yellow-white mucopurulent material that may be admixed with blood. Often the floor of the ventral meatus and the regions of contact with the nasal cavity have pooled exudate of blood and mucopurulent matter. Manipulation of the visible surface of the hematoma with the tip of the endoscope may elicit bleeding or oozing.

Recognition of a discrete, often smooth-surfaced homogeneous radiodensity originating from the ethmoidal conchae and extending into the frontal, maxillary, or sphenopalatine sinuses or into the pharynx or nasal cavity is suggestive of an ethmoid hematoma. Radiography is beneficial in determining the extent of the hematoma and in identifying suspected ethmoid hematomas that are not visible by endoscopy; however, precise definition of the origin of any hematoma is difficult from radiographic projections. Small hematomas contained within the ethmoid labyrinth may not be visible on radiographs. Computed tomographic examination of the skull allows more accurate assessment of the origin of the ethmoid hematoma,^1207,1209 allows determination of the extent of involvement of the paranasal sinuses and conchae, and facilitates surgical planning.

Necropsy Findings

Most of the morphologic features have been described from surgical specimens; few skulls with intact ethmoid hematomas have been examined.¹¹⁹⁸ Except in regions with necrosis or secondary infection associated with contact with the sinus or nasal cavity walls, the hematoma is a smooth-surfaced saclike structure containing blood in various stages of organization. The sac lining is generally healthy respiratory mucosa originating from a pedunculated region of the mucosal covering of the ethmoturbinal or sinus wall. On section the contents are amorphous red-black to chocolate brown, and in larger masses some evidence of irregular compartmentalization by fibrous tissue exists, especially on the inner surface of the sac.

Morphologic features include an outer covering of respiratory epithelium (flattened columnar or cuboidal ciliated epithelium containing glands, and occasionally stratified squamous epithelium¹²⁰⁰) overlying an irregular zone of submucosal fibrous tissue, containing hemosiderophages, occasional plasma cells, and lymphocytes, less commonly neutrophils, that forms a pseudocapsule around hemorrhage in varying states of organization. There is typically variable organization of the fibrous tissue components. Endothelial cells do not show evidence of neoplasia. Thin endothelium-lined sinuses are often present within the myxomatous stroma. The respiratory epithelium is sometimes focally ulcerated and infiltrated with neutrophils, and occasionally there are squamous metaplastic changes. Ethmoid adenocarcinoma with a similar gross appearance to ethmoid hematoma has been reported in one horse.¹²¹⁰

Treatment and Prognosis

Because these masses slowly and progressively increase in size and can cause distortion of skull architecture if the paranasal sinuses are involved, removal is recommended. Treatment method depends on the location and size. Surgical ablation has been the preferred method; however, destruction of the hematoma by intralesional injection of formaldehyde solution is associated with less morbidity, although recurrence rates are similar to those with other methods.^1204-1212 Surgical access is usually achieved by sinusotomy and then hematoma ablation by curettage, cryosurgery, or use of an Nd:YAG laser^1213-1215; photoablation can also be accomplished through the biopsy channel of an endoscope. After sinusotomy, the pedunculated origin of the hematoma is identified by digital palpation and then dissected, frozen, or photoablated, and the hematoma removed. If the hematoma is friable, intact removal may not be possible, and hemorrhage may make observation of the origin of the mass difficult. After removal the paranasal sinuses and nasal cavity are packed with gauze to control postoperative hemorrhage.

Surgical curettage can have the disadvantage of being associated with marked intraoperative blood loss typically from the turbinates or sinus mucosa rather than the hematoma. Temporary occlusion of both carotid arteries¹²¹⁴ can substantially decrease blood loss until the sinus cavity is packed with gauze. Blood loss is minimized by cryosurgical extirpation and photoablation techniques, but these approaches are not always practical when initially dealing with large hematomas. There is also minimum blood loss with transnasal photoablation; however, this technique requires multiple procedures to destroy large masses but can be performed in the standing sedated horse. Photoablation (Nd:YAG laser, 100 W in noncontact technique) is effective in controlling remnants after surgical extirpation or subsequent regrowth.^1215,1216

Destruction of ethmoid hematomas by endoscopically guided intralesional injection of formalin in standing sedated horses may reduce the need for surgical ablation in many horses.^1211,1217 A catheter passed through the endoscope biopsy channel is advanced through the rostral surface of the mass toward its origin, then 10% formalin (4% formaldehyde solution) is injected intralesionally. Commercially available catheters with a beveled needle tip have been used, but relatively stiff plastic tubing that will slide through the biopsy channel works will penetrate the capsule of the ethmoid hematoma. Sufficient volume (from 10 to 100 mL) is injected until fluid leaks back alongside the catheter or leaks from the mass. For visibly pedunculated masses, it is best to inject the formalin at the origin or neck of the mass. Tissue necrosis and slough occur in 5 to 10 days and may be associated with nasal discharge. Repeat injections, typically at no longer than 10- to 14-day intervals may be necessary to destroy the mass. Removal of necrotic tissue can be facilitated by use of long grasping forceps and hydropulsion. Mycotic plaques may cover the treated site during healing but typically resolve without treatment; endoscopically delivered topical natamycin has been used if mycotic plaque is extensive and associated with malodorous, purulent discharge.¹²¹⁸

Occasionally progression of the ethmoid hematoma may result in weakening or loss of the cribriform plate or roof of the sphenopalatine sinus.¹²¹⁸ During mass removal, loss of this protective bony covering may result in intraoperative or postoperative neurologic complications. Recognition of loss of the integrity of these bony plates may not occur until sinus lavage. In a rare complication death occurred after intralesional formalin injection in a horse where the cribriform plate had been eroded by the ethmoid hematoma.¹²¹⁹ Erosion of calvarial bone is rare, but careful computed tomographic evaluation of these regions in horses with extensive hematoma formation involving the frontal or sphenopalatine sinuses is warranted.

Irrespective of treatment method, recurrence of the hematoma occurs in 30% to 50% of cases from several months to years after the initial surgery.^{1198,1203-1206,1211,1212}

Mycotic or Bacterial Nasal Granuloma

Infectious granulomas in the nasal cavity of ruminants are not common. Documented causes include the fungal organisms Rhinosporidium seeberi and other Rhinosporidium species (which cause rhinosporidiosis), Helminthosporium species (which cause maduromycosis), Drechslera rostrata, Aspergillus species, Phycomycetes species, Stachybotrys species, and Bipolaris species. Phycomycosis is discussed in the dermatology section in Chapter 40. Nasal granuloma caused by Nocardia species bacteria has also been reported.¹ There is no apparent age, breed, or seasonal predilection, and cases are sporadic. The major clinical signs are upper respiratory noise (stridor), dyspnea, and mucopurulent nasal discharge, sometimes with epistaxis. Affected animals may rub the nose, suggesting pruritus or irritation.^2-3 Nasal airflow may be reduced, and open-mouth breathing may occur in advanced cases. Hot or dusty weather may accentuate the signs, giving the appearance of seasonal exacerbation, but the lesions are progressive. The granulomas may be single or multiple, unilateral or bilateral, and located anywhere in the nasal cavity. They consist of 0.5- to 5-cm yellow to yellow-green or red nodules or polyps, which may be sessile or pedunculated. Rhinosporidiosis tends to be a single unilateral polyp in the posterior nasal cavity, and maduromycosis tends to occur in the anterior cavity, but these distinctions are not consistent. Red and black spots (spores) may occur on the masses, and some may become secondarily infected with bacteria and ulcerate. Differential diagnoses include allergic rhinitis, foreign bodies, tumors, nasal actinobacillosis, nasal actinomycosis, and Oestrus ovis infection in small ruminants.

Endoscopy, biopsy, and culture of the lesions aid in the diagnosis. Histopathologic analysis reveals granulation tissue containing eosinophils, mononuclear cells, round sporangia, and sometimes hyphae or filamentous bacteria.^1,2 The pathogenesis of the disease involves inoculation of eroded nasal mucosa with fungal spores or filamentous bacteria from the environment. The infectious agent causes a chronic delayed (type IV) hypersensitivity reaction, which eventually leads to the formation of a granuloma. Fungal granulomas can be more common in warm, wet climates. The granulomas can be difficult to treat, and although rarely fatal the disease is chronically debilitating, with salvage often being the most practical solution.

Recommended treatments include surgical removal of the granulomas when possible and long-term sodium iodide (NaI) therapy. NaI can be administered at a dose of 66 mg/kg IV as a 20% solution, repeated at 10- to 14-day intervals until remission or iodism occurs. Iodism is characterized by lacrimation, cough, and scaling of the skin. The use of antifungal drugs to treat this condition in ruminants has not been reported.

Allergic Rhinitis and Enzootic Nasal Granuloma

Allergic rhinitis occurs in cattle and in its chronic stages may lead to the formation of granulomas. A similar condition may occur in sheep.⁴ The inciting antigen is frequently a plant pollen or more likely a fungal spore.⁵ Once homocytotropic antibody (immunoglobulin E or possibly other classes in cattle) to the allergen has developed, subsequent exposure results in a localized, ongoing, immediate (type I) hypersensitivity reaction.^5-7 If recurrent exposure to the allergen occurs, repeated tissue damage by mast cell factors results in chronic epithelial, duct, and goblet cell hyperplasia and metaplasia, as well as mucous hypersecretion and granulomatous inflammation, suggesting that a type IV hypersensitivity reaction contributes to the chronic lesion.^5,6,8

Any breed may be affected, but Channel Island breeds and Friesians seem most susceptible.⁵ The disease occurs sporadically in the United States. A familial predisposition has been reported.⁹ Most affected animals are between 6 months and 2 years of age. The signs are initially seasonal, usually occurring in warm, moist conditions; they include rhinorrhea, sneezing, nasal pruritus, a sudden onset of dyspnea, and stertorous inspiration.^9,10 There is a profuse bilateral nasal discharge. Intense pruritus is characteristic and associated with sneezing, head shaking, and nose rubbing.¹⁰ In severe cases facial swelling, tachypnea, hyperpnea, and ulceration of the nasal mucosa may occur.^6,10 Nasal foreign objects may result from the animal’s attempts to scratch the nasal mucosa. Lacrimation, chemosis, and blepharitis may also be present. In the chronic stages (the “enzootic nasal granuloma”), the signs are more constant, with seasonal exacerbations.^6,8,10 The granulomas tend to be multiple, firm, white, raised nodules 1 to 2 mm in diameter with an intact mucosa, or pale pink flat plaques scattered throughout the nasal cavity. Differential diagnoses include fungal granulomas, foreign bodies, respiratory viruses, nasal actinomycosis or actinobacillosis, tumors, O. ovis infection (small ruminants), and irritation caused by inhalation of hot or irritant gases.

Endoscopy, biopsies, cultures, antigen detection tests for viruses, bacteria, or fungi, and serologic analysis can be used to rule out these differential diagnoses. Eosinophil counts in nasal secretions correlate with the susceptibility of the animal and activity of the disease, but no absolute level is diagnostic.⁵ Intradermal allergen testing has been suggested to aid in diagnosis, but interpretation of results needs to be done in conjunction with historical and clinical findings.⁷ This condition should be differentiated from fungal or bacterial granuloma because the therapy is different.

Treatment and control entail removal of the allergen, or removal of the animal from the allergen, and therapy to block the hypersensitivity reaction. Recommended drugs include various antihistamines, meclofenamic acid, and corticosteroids at standard antiinflammatory doses (0.05 to 0.2 mg of dexamethasone per kilogram IM or IV or 1 to 2.2 mg of prednisolone per kilogram IM or IV daily). Topical corticosteroid therapy can be considered in severe, acute occurrences of the disease. The adverse effects of corticosteroids on milk production and their potential to induce abortion or parturition should be considered before their use. Antihistamine therapy has had equivocal results.⁷

Nasal Foreign Bodies

Cattle are more prone than small ruminants to the acquisition of nasal foreign bodies. Foreign objects may be acquired as a result of attempts to scratch the nose in cases of allergic rhinitis, or because of the cow’s aggressive eating habits. Depending on the size and duration of residence of the object, signs may include head shaking, stridor, sneezing, snorting, frequent nose licking, unilateral decreases in airflow, foul odors, and serous, mucopurulent, or hemorrhagic discharges. Differential diagnoses include fungal granulomas, allergic rhinitis, tumors, nasal actinomycosis or actinobacillosis, and O. ovis (small ruminants). Many objects can be visualized on careful examination of the nasal cavity with an adequate light source, whereas some may require endoscopy for diagnosis and removal.

Nasal Trauma and Fractures

Trauma to the facial bones, sinuses, and turbinates may result from fighting, accidents caused by improper restraint, farm machinery accidents, human maliciousness, and passage of excessively large NGTs. Severe fractures can lead to facial swelling, SC emphysema, obstruction of airflow, stertor, and epistaxis. Secondary infection causes foul odors and mucopurulent nasal discharge. Differential diagnoses for the acute external swelling of the head with stertor include snakebite, actinobacillosis, actinomycosis, and phlegmon (Fusobacterium, Clostridium species). Unless the development of severe depression fractures, formation of sequestra, or severe obstruction of airflow occurs, surgery is usually not indicated. Radiographs confirm the diagnosis and help determine the need for surgical removal of potential sequestra or elevation and fixation of large displaced segments. Prophylactic antibiotics (typically penicillin, 22,000 U/kg IM or SC q12-24h) are recommended to prevent fracture infection and sinusitis, and NSAIDs (aspirin, 100 mg/kg PO twice; flunixin meglumine, 1.1 to 2.2 mg/kg IV daily or divided twice daily) may help relieve pain, swelling, and stridor. The prognosis is usually good.

Nasal Tumors and Polyps

Tumors and polyps of the nasal cavity and sinuses are rare in ruminants. Nasal tumors reported in cattle include osteomas and osteosarcomas of the sinuses, squamous cell carcinomas,¹¹ neuroblastomas, and adenocarcinomas of the ethmoid mucosa. Ethmoid adenocarcinomas are speculated to be caused by viruses on the basis of an endemic pattern in some cases.¹² They tend to occur in cattle 6 to 9 years of age and are frequently unilateral. Metastasis occurs to the lymph nodes and lungs. There is a report of a hemangiosarcoma involving the external naris of a cow.¹³ Signs common to all nasal tumors include mixed or inspiratory dyspnea, stridor, nasal discharge, epistaxis, foul breath odors, unilateral decreases in airflow, open-mouth breathing, and distortion of the facial bones. Differential diagnoses include fungal granulomas, atopic granulomas, foreign bodies, sinusitis, fractures, and nasal actinobacillosis and actinomycosis. Treatment has not been investigated.

The majority of nasal neoplasms in sheep and goats are adenopapillomas, adenomas, or adenocarcinomas.¹⁴ Squamous cell carcinoma¹⁵ and osteoma¹⁶ have also been reported. Nasal adenocarcinomas have also been described in goats.¹⁷

An enzootic form of nasal adenocarcinoma occurs in both sheep and goats and is associated with ovine nasal adenocarcinoma virus (ONAV)^18,19 or caprine nasal adenocarcinoma virus (CNAV),²⁰ respectively. These agents are β retroviruses that are closely related to, but distinct from, jaagsiekte sheep retrovirus (JSRV), the cause of ovine pulmonary adenocarcinoma (OPA).^20,21 Although nasal adenocarcinoma has been difficult to consistently reproduce experimentally, inoculation of kids with concentrated cell-free and bacteria-free filtrate containing virus from naturally infected goats has resulted in disease.²² Neoplastic transformation is limited to secretory epithelial cells of the nasal turbinates, but CNAV appears to have a wider tissue tropism than ONAV; in one study viral provirus incorporated into host DNA was found in many tissues of infected goats but was largely confined to tumor tissue of infected sheep.²⁰ There is no breed or sex predisposition for enzootic nasal adenocarcinoma; affected animals are most commonly young adults, but the tumor has been identified as early as 4 months of age.¹⁸ Signs include progressive inspiratory dyspnea; stridor; exercise intolerance; mouth breathing; serous, mucoid, or mucopurulent nasal discharge, which is typically profuse; tachypnea; decreased airflow; head shaking; sneezing; exophthalmos; and facial asymmetry.^14,17,18,21 The lesions may be unilateral or bilateral, originating in the olfactory region of the ethmoid turbinates. The neoplasia arises either from Bowman’s glands¹⁸ or serous glands of the nasal mucosa.¹⁷ The tumor begins as a small nodule that can grossly resemble the mucoid polyps that occur in animals with chronic rhinitis. Over time the adenocarcinoma grows into a soft, gray to grayish pink, mucoid, nodular, cystic mass. The tumor is benign but locally expansive, often entering the sinuses and eroding overlying bone.¹⁷ Histologically the tumor is typically classified as a low-grade adenocarcinoma, but it may also be identified as an adenopapilloma or adenoma.¹⁷ Initially affected animals eat and drink normally and maintain body condition, but as the tumor expands the animal begins to lose condition, and death eventually occurs as a result of inanition, asphyxia, or aspiration pneumonia. Necrosis or secondary bacterial infection of the tumor can occur and may lead to the production of foul-smelling, purulent discharge and systemic signs related to bacterial infection, such as fever, depression, and hyperemic mucous membranes.

Differential diagnoses for nasal neoplasia in small ruminants include nasal fungal or bacterial granuloma, actinobacillosis, actinomycosis, O. ovis infection, and sinusitis. Endoscopy and radiology are helpful in establishing an initial diagnosis of a nasal mass. Preoperative or antemortem pinch biopsies and exfoliative cytologic analysis frequently are nondiagnostic, and findings may be misleading.^14,15 Identification of tumors in multiple animals in a herd or flock supports a diagnosis of enzootic nasal adenocarcinoma. Definitive identification of infection with ONAV or CNAV can be difficult. Serologic tests are not reliable because animals do not consistently produce antibody to the viruses,²³ possibly because of the presence of endogenous retroviruses²⁴ that share epitopes with ONAV and CNAV and induce the development of immunologic tolerance. Recently PCR has been used to specifically identify and differentiate ONAV, CNAV, and JSRV in tissues of affected animals.²⁵ Surgical management of nasal adenocarcinoma in sheep has been described.²⁶

Definition and Etiology

O. ovis is parasite of the nasal passages and sinuses of sheep and, less commonly, goats. Goats are relatively resistant to infection and, even when housed with sheep, tend to have a lower prevalence of infection.²⁷ Occasionally humans and other animals are accidentally infected, with conjunctival infection the most common form of disease in humans.^28,29 The pathogenic stage of the parasite is the larval stage. The first instar larvae are deposited near the nostrils of sheep by the adult female fly and migrate into the nasal and ethmoid turbinates; in the ethmoids the larvae molt to second instars, which migrate to the sinuses before molting again and becoming third instar larvae. Third instar larvae, which are yellow-white in color and have a dark dorsal stripe with rows of spines on the ventrum of each segment, return to the nasal passages and are sneezed out onto the ground, where they pupate and eventually develop into adult flies.^30,31 The adult is active during warm months, and the parasite may overwinter either as a first instar larva in the host or as a pupa in the ground. Larvae can persist in the upper airways for weeks to months, and appear to be able to arrest development for a period of time if necessary to avoid climate extremes.³² Adult flies have a rudimentary mouthpiece and are not able to feed; thus the larvae must ingest adequate nutrients while in the host to support the life of the adult fly.³³

Clinical Signs

The larvae cause irritation of the nasal passages and sinuses, leading to mucoid to mucopurulent and sometimes blood-tinged nasal discharge, sneezing, nose rubbing, and inspiratory stridor. Adults flies cause annoyance by flying around the heads of animals; thus both stages can lead to decreased productivity by decreasing the time spent grazing by affected animals. Occasionally the larvae may cause sinusitis or pneumonia owing to secondary bacterial infection associated with irritation caused by the larval infestation. Differential diagnoses to consider include nasal foreign bodies, allergic rhinitis, nasal adenocarcinoma, fungal rhinitis, trauma, sinusitis, actinobacillosis, or actinomycosis.

Pathogenesis

The larvae cause direct irritation to the nasal passages and sinuses, and this irritation may predispose animals to the development of secondary bacterial rhinitis, sinusitis, and/or pneumonia. Interstitial pneumonia has also been seen on occasion in affected sheep, presumably induced by inhalation of parasite antigens and inflammatory mediators from the upper respiratory tract.³⁴

Epidemiology

Infection is more common in warm climates. Recent European surveys have identified O. ovis in 35% to 91% of sheep surveyed,^35,36 with seroprevalences of 46% to 69%.^27,36,37 A majority of flocks surveyed had at least one infected animal.^27,35 No North American reports describing the prevalence of O. ovis infestation have been published in recent decades.

Diagnosis

Diagnosis is usually presumptive, based on typical clinical signs. The larvae can also be identified via radiographs of the head or by endoscopy, but these tests may not be practical for field use. Serologic diagnosis has been used for research studies of the epidemiology of O. ovis infection, but these tests are unlikely to be available at most diagnostic laboratories.

Treatment and Prevention

The larvae are susceptible to ivermectin at 200 μg/kg PO.³⁸ Moxidectin at 0.2 mg/kg as a 0.1% oral drench was not effective against O. ovis³⁹; these investigators found that the same dose given as a 1% injectable solution was effective, but this route is not approved in the United States. Pour-on eprinomectin at 0.5 mg/kg (applied immediately after shearing)⁴⁰ and injectable doramectin at 200 μg/kg IM⁴¹ have also been shown to be effective in sheep, but these treatments are also not approved for use in sheep or goats in the United States. Because the adult fly should be killed by freezing weather, in areas with a season of freezing weather it is logical to treat after the first hard freeze, when sheep will no longer be susceptible to infection until warm weather returns. Prevention is aimed at regular strategic treatment with effective anthelmintics to prevent long-term infection with the larvae.

Definition and Etiology

Inflammation of the paranasal sinuses is most common in cattle and occurs infrequently in sheep and goats. Typically the frontal or maxillary sinuses are involved, and a variety of bacteria may be isolated. The proximate cause for infection is usually dehorning (frontal sinusitis) or infected teeth (maxillary sinusitis). Other causes include extension of actinomycosis or nasal neoplasia into the sinus, injuries to the horn, facial fractures, respiratory viruses (including malignant catarrhal fever [MCF], IBR, and parainfluenza viruses), sinus cysts,^44,45 lymphosarcoma,⁴⁶ and O. ovis (in sheep).^47-50

Clinical Signs and Differential Diagnosis

Sinusitis associated with dehorning may be acute or may occur weeks to months later; typically only one sinus is affected. Nonspecific clinical signs include anorexia, lethargy, reluctance to move, and fever. When sinusitis occurs acutely after dehorning, the portal of entry is frequently open and discharging pus, and the animal is often febrile (39.5° C to 40.5° C). In chronic sinusitis, signs may include unilateral or bilateral nasal discharge, mild stridor, changes in airflow, and foul breath odors that are frequently unilateral; fever is not common. The animal may hold its head at an odd angle (extended up or down, tilted) and may squint the eyelids as if in pain. ^49,51 With chronicity, frontal bone distortion, exophthalmos, and neurologic signs may occur.^51,52 In one report of 12 cattle with frontal sinusitis, four cattle had abnormal posture, with an extended head and neck, partially closed eyes, and a tendency to head-press or to rest the head on a stationary object; the other 8 cattle were apprehensive and intolerant of head manipulation.⁵¹ Extension of infection may involve the CNS.

Occasionally sinusitis irritates the animal sufficiently that it may rub its head on the ground, driving more debris into the sinus.⁵⁰ Maxillary sinus cysts have been observed in cattle.⁴⁴ Typical signs included unilateral facial swelling over the affected sinus; mucopurulent, nonfetid nasal discharge; and radiographic evidence of septal deviation. One cow had stertorous respiration with diminished airflow. Differential diagnoses for sinusitis include facial fractures, nasal tumors, actinomycosis, actinobacillosis, retrobulbar abscess, and lymphosarcoma.

Diagnosis

Diagnosis can usually be made based on clinical signs. Diagnostic aids may be useful in selected cases, particularly when there is no recent history of dehorning, or in cases of maxillary sinusitis. The hemogram is quite variable and is of little assistance in diagnosis. Percussion of the sinus may reveal a dull, full sound and may elicit pain. If the bone has been greatly thinned and has gas underlying it, percussion may produce a hyperresonant sound. Fractures, soft-tissue masses, dental disease, fluid in the frontal sinus, or lysis of bony septa may be evident on radiographs.⁵¹ Sinus centesis may yield purulent material, which should be cultured and examined cytologically. A small area over the affected sinus is clipped and surgically prepared, and local anesthetic is infiltrated subcutaneously. Then a small stab incision is made through the skin and periosteum, and a Steinmann pin is used to drill a small hole. Polyethylene tubing is inserted, and attempts are made to aspirate material. A small amount of sterile isotonic fluid can be injected and aspirated to obtain a washing. The stab incision is closed unless sinusitis is confirmed.

Treatment and Prognosis

Trephine sites for sinusotomy (Fig. 31-47) are as follows:

Fig. 31-47 Trephine sites for sinusotomy. A, Dorsal frontal sinus. B, Postorbital diverticulum. C, Rostral frontal sinus. D, Turbinate portion of frontal sinus. The maxillary sinus is trephined ventral to a line from the infraorbital foramen to the medial canthus (arrowhead). If draining tracts are present at the poll, an additional sinusotomy can be made in the cornual portion of the frontal sinus (arrowhead).

Cattle that have frontal sinusitis after dehorning should be treated by sinusotomy and drainage of the sinus.⁵³ Sinusotomy sites should be based on anatomic landmarks⁵¹ and modified as needed to accommodate any frontal bone distortion or wounds related to dehorning.⁵¹ Sinusotomy should be performed 3 to 4 cm from midline, intersecting a line drawn between the caudal aspect of the orbits. If draining tracts are present at the poll, an additional sinusotomy can be made in the cornual portion of the frontal sinus.⁵¹ Sinusotomy is performed after sedation and local anesthetic infiltration of the centesis site(s). A 2-cm—diameter circular piece of skin is excised, and a 19-mm (¾-inch) trephine used to create an opening into the sinus, through which purulent fluid should be evacuated and the sinus lavaged.

Additional trephine sites that permit access to other regions of the frontal sinus include the postorbital diverticulum, which is trephined approximately 4 cm caudal to the dorsal rim of the orbit, just above the temporal crest of the frontal bone; the rostral frontal sinus, which is trephined just caudal to a line between the centers of the orbits and to either side of the midline; and the turbinate portion of the frontal sinus, which is trephined just rostral to the line described and to either side of the midline.

Access to the maxillary sinus is achieved by trephining ventral to a line from the infraorbital foramen to the medial canthus. If an infected tooth is the cause of maxillary sinusitis, a sinusotomy is created with a trephine over the affected tooth to repel it; otherwise, the hole is usually made just dorsal and caudal to the facial tuberosity. The trephine site should be higher in the sinus of younger animals because the tooth roots are longer.

If the frontomaxillary and nasomaxillary openings are still patent, one trephine site may be sufficient, with the natural opening providing ventral drainage. In more chronic cases, two trephine sites (for ingress and egress) are needed. Another alternative in chronic sinusitis is the use of a curved steel sinus probe (1 cm diameter × 55 cm long), which is forcefully driven through the septal plates of the frontal sinus into the nasal meatus for ventral drainage.⁵⁴ The frontal sinus is very compartmentalized in mature sheep and goats, and effective drainage is difficult. Therapy should therefore be aggressive in these species, even in early cases, and double trephination or bone flaps for exposure and curettage should be considered.

If a tooth has been repelled, a roll of gauze or dental impression material should be used to occlude the alveolar socket to prevent feed material from entering the sinus. A strip of umbilical tape tied around the gauze roll or a wire in the dental material is passed through socket, sinus, and trephine hole and secured to the face by tying around another roll of gauze as a stent. These gauze packs are replaced each time the sinus is flushed. The sinus is lavaged daily with dilute antiseptic solutions such as 0.1% povidone iodine or chlorhexidine in saline, or 1:1000 potassium permanganate. Lavage is continued until infection is resolved. Enzymes (papain or 200,000 U of streptokinase and 50,000 U of streptodornase in at least 10 mL of normal saline solution) may help remove thick exudate.

Parenteral antibiotics and NSAIDs (aspirin, 100 mg/kg PO twice daily or flunixin meglumine, 1.1 to 2.2 mg/kg IV daily or divided twice daily) are indicated if systemic signs are present. In the absence of microbial culture and susceptibility results, penicillin (22,000 U/kg IM or SC q12-24h) is recommended as the antibiotic of choice because Arcanobacterium (Actinomyces) pyogenes is the most common organism isolated from cattle with chronic frontal sinusitis resulting from dehorning. Pasteurella multocida is the most common organism isolated from infections of the frontal sinus not associated with dehorning.⁵¹ Penicillin may be effective against some isolates of P. multocida; alternatively, oxytetracycline can be administered if P. multocida is suspected (11 mg/kg IV or SC q24h, or 20 mg of long-acting oxytetracycline per kilogram SC q72h). If bacterial culture and susceptibility results are available, antimicrobial therapy should be modified accordingly. Early cases often resolve in 10 to 14 days, with a good prognosis. Long-term therapy (weeks) is frequently needed in chronic cases; the prognosis is more guarded, and salvage is often the best option.

Prevention and Control

Dehorning ruminants as neonates, particularly if a “closed” method such as a dehorning iron is used, is the most effective way to prevent frontal sinusitis. In larger cattle surgical dehorning with primary skin closure achieved under aseptic conditions minimizes the likelihood of sinusitis.⁵⁵ When this is not practical, dehorning should be avoided in rainy, windy, or dusty conditions, and fly control must be used. The dehorning of mature sheep or goats leaves massive wounds that typically take 4 to 6 weeks to close by second intention and so are susceptible to infection; special care, such as bandaging for the initial 7 to 10 days, must be taken.⁵⁶ Sinusitis did not occur in goats aged 2 to 24 months dehorned with a technique in which primary skin closure was achieved.⁵⁷

Definition and Etiology

Bovine herpesvirus 1 (BHV-1) is an enveloped DNA virus that is classified as an alphaherpesvirus in the family Herpesviridae. It is associated with multiple, distinct disease syndromes of cattle that include IBR, conjunctivitis (see Chapter 39), infectious pustular vulvovaginitis (IPV), balanoposthitis, abortion (see Chapter 43, encephalomyelitis (see Chapter 35), and mastitis.^93,94 Although only a single serotype of BHV-1 is currently recognized, three subtypes have been identified on the basis of restriction endonuclease cleavage patterns. These subtypes are referred to as BHV-1.1 (respiratory infections), BHV-1.2 (respiratory and genital infections), and BHV-1.3 (neurologic infections). BHV-1.3 has been reclassified as a distinct herpesvirus and is designated as BHV type 5.⁹³ Only the respiratory manifestations of infection with BHV-1 are discussed here. The respiratory form is characterized by rhinitis, tracheitis, and pyrexia and is referred to as IBR. BHV-1 can also cause pneumonia as part of a severe generalized infection in newborn calves.⁹⁵ The virus is also of great importance in the bovine respiratory disease complex because of its role in enhancing secondary bacterial bronchopneumonia by causing respiratory injury and immunosuppression,⁹⁶ as discussed later.

Definition and Etiology

BRSV is an enveloped RNA virus that is classified as a nonhemagglutinating pneumovirus of the family Paramyxoviridae. This virus was named for the characteristic cytopathic effect it produces in vitro and in vivo, which is the formation of syncytial cells. BRSV shares many similarities in its biology and epidemiology with the human respiratory syncytial virus (HRSV). HRSV is considered to be the most important respiratory tract pathogen of infancy and early childhood. Although BRSV and HRSV are closely related, they are distinct viruses.¹³⁸ RSV has also been isolated from sheep and goats with respiratory disease. Studies using RNAse mismatch cleavage analysis indicate that ovine RSV may be distinct from HRSV and BRSV, whereas caprine RSV may be more closely related to BRSV.^139,140 The possibility of interspecies transmission has not been well defined. However, a European serologic survey undertaken to determine whether nonbovine species were likely to harbor BRSV infection found that only goats, in addition to cattle, were commonly seropositive.¹⁴¹ These results indicate either that goats are commonly infected with BRSV or that antigenic similarities between caprine RSV and BRSV led to production of cross-reactive antibodies in many goats. Although two major antigenic subtypes, A and B, of human RSV have been clearly defined, research has shown that BRSV is less variable than human RSV at the genetic level, as indicated by evaluation of nucleotide sequences in genes for viral proteins or by digestion of viral RNA with restriction enzymes.¹⁴² However, in spite of minimal to moderate differences among BRSV isolates at the genetic level, isolates can differ notably in their reactivity with monoclonal antibodies directed against individual BRSV proteins^143,144; this indicates that apparently minor genetic differences could lead to changes that allow the virus to evade host immunity. Differences in molecular weights of viral proteins also divide isolates into two major groups.¹⁴⁵ A study of genetic variability in BRSV isolates collected over several decades showed that genetic change occurred over time^146,147 and that BRSV isolates commonly included in commercial vaccines differed a relatively large amount from some isolates collected from recent natural outbreaks.¹⁴⁶ Studies have also shown that genetic differences in BRSV isolates can be grouped by geographical origin, with some European isolates differing to a notable degree from isolates from the United States or Japan.^146,147 Most important, European research identified genetic changes in a region of the BRSV G glycoprotein (the viral attachment protein) that was previously thought to be highly conserved and difficult or impossible for the virus to change.¹⁴⁶ These changes appear to be driven by vaccination, indicating that antigenic pressure induced in cattle by vaccination can lead BRSV to change in unexpected ways. These data indicate that commercial vaccines may need to be reformulated periodically to include isolates of the virus that are similar to isolates currently circulating in natural populations.

Clinical Signs

Clinical signs of BRSV infection can vary from inapparent to severe in an infected group of cattle.^148-151 Signs of disease are limited to the respiratory system. Infected animals may display elevated rectal temperature of 40° C to 42.2° C (104° F to 108° F), depression, decreased feed intake, elevated respiratory rates, ptyalism, cough, and nasal and lacrimal discharges. Signs of disease may progress rapidly, and early signs may be missed. Thoracic auscultation may reveal increased bronchial and bronchovesicular sounds; fine crackles or wheezes may be heard, particularly in the middle or dorsocaudal lung fields. Absence of bronchovesicular sounds may be noticed in the dorsal or dorsocaudal lung fields if an emphysematous bulla is present, or if a bulla has ruptured, leading to pneumothorax. In later stages of the disease, dyspnea can become pronounced and is characterized by increased expiratory effort and mouth breathing. SC emphysema and intermandibular edema are sometimes noted. Dramatic reduction in milk production has been reported in dairy cattle. Duration of disease is variable (1 to 2 weeks). A biphasic clinical course has been described but does not appear to be a consistent finding. Experimental infections of lambs with ovine RSV caused a mild primary pneumonia and was also capable of causing lower respiratory tract lesions in calves and deer.¹⁵²

Pathogenesis

The means of transmission appears to be contact with infected cattle and aerosols. The incubation period is 3 to 5 days. Infection with BRSV can cause bronchitis, bronchiolitis, alveolitis, and interstitial pneumonia.^148-151 After experimental infection, BRSV is found in cells of the nasal, tracheal, and bronchial epithelium by 2 days postinfection and in bronchiolar and alveolar epithelial cells by day 4 postinfection.¹⁵³ The virus causes epithelial cells to fuse, resulting in the characteristic multinucleated cells, or syncytia, which are seen in airways and alveoli. Epithelial cells, which may undergo apoptosis after BRSV infection,¹⁵³ slough into the lumen of airways and are phagocytized by neutrophils or alveolar macrophages.¹⁵³ Infection of alveolar macrophages and circulating peripheral blood mononuclear cells occasionally occurs,^154,155 and BRSV infection of macrophages decreases important functions of these cells including Fc receptor expression, phagocytosis, phagosome-lysosome fusion, and production of factors that induce neutrophil chemotaxis.^156,157 Infection leads to bronchitis, bronchiolitis, alveolitis, and sometimes AIP; these changes lead to clinical signs referable to small airway and alveolar disease (expiratory dyspnea, auscultable wheezing) and hypoxemia as evidenced by arterial blood gas analysis.¹⁵⁸

The severity of disease after BRSV infection is related to host immunity. Both humoral and cell-mediated immunity contribute to protection. Clinical signs of infection are less severe in animals with moderate to high levels of serum antibody against BRSV,^159,160 and more extensive pathology is seen in calves depleted of CD8⁺ T cells before infection.¹⁶¹ Protection after intranasal vaccination is related to the rapidity in onset of nasal BRSV-specific IgA production,¹⁶² indicating that mucosal immunity is likely also important in protection against naturally occurring disease. Host immunity after exposure to BRSV can minimize disease, and clinical signs are usually the most noticeable in calves under 6 months of age.^149,163 However, ruminants appear to be susceptible to reinfection with BRSV throughout their lives, and adult cattle can develop severe disease after BRSV infection.^149,150 Because genetic variability among BRSV isolates is not great, reinfection does not seem to be solely a result of viral strain variation. Although data investigating the duration of BRSV immunity after natural infection are limited, it appears that natural infection does not confer lifelong immunity, and some cattle may be reinfected annually,¹⁶⁴ although clinical signs are not always apparent.

While host immunity can protect ruminants from severe disease after BRSV infection, the host immune response can also contribute to disease. Evidence for BRSV immunopathogenesis comes from both natural outbreaks and experimental challenge studies. Cattle with severe disease after natural infection have a higher proportion of degranulated mast cells in their lungs than less severely affected cattle,^165,166 indicating that mast cell mediators contribute to physiologic and pathologic changes seen in severe cases. Serum tryptase, a preformed enzyme present in mast cell granules, was significantly increased in the serum of cattle experiencing severe naturally occurring BRSV infection,¹⁶⁶ also supporting a contribution of mast cells in severe BRSV-induced disease. In experimentally challenged cattle, BRSV-specific IgE has sometimes been associated with disease severity.^167,168 Exposure to the fungus Saccharopolyspora rectivirgula (formerly M. faeni), the spores of which induce pulmonary hypersensitivity in cattle and other species, enhanced BRSV-specific IgE production in calves, indicating that environmental allergens can also affect the nature of the immune response to BRSV infection.¹⁶⁸ Recent research showed that calves infected with BRSV before infection with Histophilus somni developed higher levels of H. somni–specific serum IgE than did calves infected with H. somni alone.¹⁶⁹ Thus BRSV infection can lead to production of virus-specific IgE, which may be enhanced by coexposure to allergens, and BRSV infection may also increase IgE production in response to infection with other agents. In these cases, cross-linking of IgE bound to mast cells by binding of BRSV or other specific antigen is expected to lead to bronchoconstriction, pulmonary edema, and other signs of mast cell mediator release and immediate hypersensitivity.

Perhaps the most convincing evidence that BRSV can trigger immunopathogenesis is the fact that certain BRSV vaccines have on rare occasion been shown to cause enhanced disease when vaccinated animals are subsequently infected with BRSV. Vaccine-enhanced disease has been seen after both natural^170,171 and experimental^172,173 infection. Vaccine-enhanced disease has also occurred in human infants vaccinated with a formalin-inactivated RSV vaccine,¹⁷⁴ and formalin-inactivated BRSV vaccines have caused enhanced disease in some^172,174 but not all¹⁵⁸ studies. Inactivated BRSV vaccines have also been shown to prevent disease after infection^175,176; therefore not all inactivated BRSV vaccines enhance disease. It appears that the dose of BRSV protein, the adjuvant included in the vaccine, and probably also other factors related to the formulation of the vaccine affect whether enhanced disease is likely after infection of cattle given a particular BRSV vaccine.¹⁷⁷ Apparent vaccine-enhanced disease has also been reported in cattle that received a modified live vaccine in the early stage of a natural BRSV outbreak.¹⁷⁰ It is important to note that BRSV vaccines are used very commonly, and vaccine-enhanced disease is very rare; thus vaccination is still recommended as part of any plan to control disease caused by BRSV. Many gaps exist in our understanding of what aspects of the immune response contribute to disease severity after BRSV infection or vaccination. Most research completed to date has focused on enhanced disease in calves that received formalin-inactivated BRSV. The data suggest that when enhanced disease occurs in calves vaccinated with formalin-inactivated BRSV, it is related to a relatively strong response by T helper type 2 (Th2) cells, as evidenced by decreased production of the Th1 prototype cytokine interferon gamma,¹⁷⁸ increased production of BRSV-specific IgE,¹⁷⁹ and pulmonary eosinophil infiltration.¹⁷³ The association of BRSV-specific IgE production with disease severity in some experimental studies indicates that at least some individual ruminants will produce IgE after BRSV vaccination or infection, which may contribute to severe disease in these individuals.

The presence of co-infection with other pathogens such as M. haemolytica¹⁸⁰ or bovine virus diarrhea (BVD) virus (BVDV)¹⁸¹ can also increase the severity of disease after BRSV infection. Although genetic variation among BRSV isolates is not extensive, experimental challenge with certain isolates of BRSV can lead to serious disease,^158,182 whereas challenge with other isolates leads to only minimal disease.^183,184 This indicates that, in spite of limited differences among viral isolates at the genetic level, variation among BRSV isolates may still contribute to variations in disease severity.

Epidemiology

Prevalence of antibodies to BRSV in the cattle population of the United States ranges from approximately 60% to 80%.¹⁵¹ The virus has been recognized in association with respiratory tract disease in nursing beef calves, in dairy calves, and in cattle entering into feedlots. BRSV was demonstrated to be involved in 14% to 71% of respiratory disease outbreaks in North American and European studies of calf pneumonia outbreaks involving several farms.¹⁸⁵ Seroconversion to BRSV has been significantly associated with treatment for respiratory disease in feedlot cattle,^186,187 and cattle with low antibody titers to BRSV at feedlot entry have increased risk of developing disease.¹⁸⁶ Although feedlot cattle are commonly infected with BRSV after arrival, a recent survey of necropsy findings at 72 Canadian feedlots found BRSV at postmortem of only 11% of the cattle that died or were euthanized because of pneumonia.¹⁸⁸ The low rate of isolation at necropsy suggests that if the virus contributes to the development of pneumonia in feedlot cattle, it is often no longer present by the time the animal dies or is euthanized.

BRSV therefore represents an important virus in the bovine respiratory disease complex on the basis of its frequency of occurrence and predilection for causing infection of the lower respiratory tract. In general, morbidity rate tends to be high in outbreaks of BRSV, whereas case fatality rate is variable, ranging from none to as high as 20%.¹⁸⁵

Cattle are most likely the principal reservoirs of infection, although a European serologic survey of different species found that goats were also often seropositive,¹⁴¹ indicating that goats may be a source of BRSV infection for cattle, and vice versa. The mechanism by which BRSV persists in the cattle population is not known. Possibly a similar epidemiologic pattern to the one that has been described for HRSV also exists for BRSV. HRSV is capable of reinfecting the host throughout his or her life; however, severe lower respiratory tract disease occurs only in association with the initial exposure. Subsequent exposure results in mild upper respiratory tract disease. Similarly, adult cattle may periodically undergo subclinical to mild infections and serve as a source of infection for susceptible young stock. However, a recent study concluded that transmission among seropositive cattle was not a plausible mechanism of BRSV persistence in a dairy herd.¹⁸⁹ These authors suggested that persistent BRSV infection in individuals is a more plausible explanation of population persistence of BRSV. BRSV has been identified in B lymphocytes in tracheobronchial and mediastinal lymph nodes of calves 71 days after experimental infection,¹⁹⁰ so it may be that individual cattle harbor the virus long term and periodically begin shedding it, allowing it to periodically reappear in herds even if they are closed to new introductions. However, more research is needed before persistent infection is clearly proven to be a means by which BRSV remains established in herds.

Necropsy Findings

Grossly, BRSV infection can cause signs of AIP; therefore differential diagnoses for the typical gross lesions include other causes of AIP (see later discussion). Signs of AIP are most evident in the dorsocaudal lung; affected lung is heavy, with a rubbery texture, and fails to collapse when the thorax is opened. Individual lobules may appear dark, and interstitial or bullous emphysema is often present in the dorsocaudal lung in severe cases (Fig. 31-54). Pathologic emphysema must be differentiated from that sometimes caused by agonal breathing of cattle. Cranioventral lobes or lobules are often dark and collapsed because of atelectasis or consolidation (Fig. 31-55). Histologic lesions depend on the stage of infection. Neutrophilic and later mononuclear bronchitis, bronchiolitis, and alveolitis are present in infected animals. Syncytial cells may be seen in the airways or alveoli; intracytoplasmic inclusion bodies are rarely present.^149,160,182 Signs of AIP including alveolar epithelial hyperplasia, hyaline membrane formation, and interstitial inflammatory cell infiltrate, hemorrhage, and edema are seen in severe cases. Later, evidence of chronic bronchitis and bronchiolitis obliterans can be found.¹⁴⁹

Fig. 31-54 Postmortem photograph of bovine lung with interstitial emphysema often seen in severe BRSV infection, as well as other causes of acute interstitial pneumonia.

Photograph contributed by Dr. Amelia Woolums, University of Georgia, Athens, Ga.

Fig. 31-55 Postmortem photograph of lungs from calf with severe BRSV infection.

Photograph contributed by Dr. Laurel Gershwin, University of California, Davis, Calif.

Diagnosis

Infection is diagnosed by identification of the virus in nasal secretions, tracheal aspirates, or lung lavage fluid from live calves or in lung tissue collected postmortem. BRSV is difficult to isolate as it does not survive transport well; thus methods of identification that do not require the virus to be alive (such as immunofluorescence, IHC, and RT-PCR) are preferable to virus isolation for the identification of BRSV. IHC of formalin-fixed tissue is convenient because the virus can be identified in tissue processed for histopathologic evaluation. If virus isolation is to be attempted from samples collected in the field, it is recommended that that veterinarian contact his or her diagnostic laboratory before collecting samples so that samples are collected and transported in ways that optimize likelihood of success. Although virus is readily identifiable in lung tissue of cattle early in the course of disease (within approximately 8 days of infection), virus is less likely to be found in lung tissue by 10 to 15 days postinfection, even when IHC is used.^153,172,182

Seroconversion as evidenced by paired serology also supports a diagnosis; virus neutralizing or ELISA assays are most commonly used to identify BRSV-specific antibody. Identification of BRSV-specific IgM in a single acute sample is also diagnostic,¹⁹¹ but testing for BRSV-specific IgM may not be available at most diagnostic laboratories.

Treatment and Prevention

Treatment of BRSV is supportive and aimed at preventing secondary bacterial infection and limiting the inflammatory response in bronchioles and alveoli. Antimicrobial therapy appropriate for common secondary bacterial pathogens (see Table 31-10) is appropriate. Antiinflammatory therapy with NSAIDs (flunixin meglumine at 1.1 to 2.2 mg/kg IV daily or divided twice daily) is considered appropriate, but controlled studies of these drugs in animals with disease caused by BRSV are limited. Administration of one or two doses of steroid therapy (dexamethasone 0.05 to 0.2 mg/kg IV or IM once or twice) is appropriate in animals with severe respiratory distress or evidence of AIP. Intranasal oxygen insufflation is appropriate if available. Diuretic therapy with furosemide (0.5 to 1 mg/kg IM or IV once or twice daily) is warranted in animals suspected of having AIP. Although the prognosis for animals with uncomplicated BRSV infection is good, the prognosis for animals with AIP is guarded. Animals with severe respiratory distress should be handled with care, as even careful manipulation to administer treatment can lead to rapid respiratory decompensation and death.

Both modified live and inactivated BRSV vaccines for IM or SC administration are commercially available. Certain commercially available vaccines have been shown to protect calves against virulent experimental challenge.^175,176,192 Results of some well-designed trials of BRSV vaccine efficacy have also been published.^193-196 In some studies the effect of BRSV vaccination on all clinical respiratory disease (i.e., not only disease caused specifically by BRSV) was evaluated,^193,195,196 whereas in other studies protection against disease caused by BRSV was specifically evaluated by identification of clinical respiratory disease and simultaneous seroconversion to BRSV in vaccinated cattle.¹⁹⁴ Vaccination decreased all respiratory morbidity in some¹⁹⁵ but not other^193,195,196 groups of cattle evaluated. Vaccination decreased disease specifically caused by BRSV in large groups of calves if all calves in the group were vaccinated, but not if half of the calves in the group were vaccinated.¹⁹⁴ A recent large clinical trial evaluating the effect of vaccination of feedlot cattle with a modified live vaccine containing BHV-1, PI3, and BVDV, with or without BRSV, found that pens of cattle that received the vaccine including BRSV had lower overall morbidity and mortality and lower numbers of respiratory deaths than pens of cattle that received the vaccine that did not include BRSV.¹⁹⁶ There is no RSV vaccine licensed for use in small ruminants, and no controlled studies of the effects of extralabel administration of BRSV vaccines licensed for cattle on respiratory disease in sheep or goats have been published.

Definition and Etiology

Parainfluenza virus type 3 (PI3) is an enveloped RNA virus classified in the family Paramyxoviridae. The virus has been associated with respiratory tract disease in cattle, sheep, and goats. It hemagglutinates and hemadsorbs red blood cells of certain species, which means that serum antibodies can be identified by hemagglutination inhibition assays. Variation in virulence among strains of PI3 has been reported.²¹⁴ Significant similarities between bovine PI3 and human PI3 have led to efforts to use bovine PI3 as a modified live intranasal vaccine for humans; bovine PI3 was found to be safe and immunogenic in a clinical trial in human infants.²¹⁵

Clinical Signs

Uncomplicated PI3 infections result in subclinical to mild signs. Clinical signs, if present, may include fever, cough, nasal and ocular discharge, increased respiratory rate, increased bronchovesicular sounds, and wheezes.^216-218 The most important role of PI3 is in predisposing the respiratory tract to subsequent infection by other viruses and bacteria such as Mannheimia (Pasteurella) haemolytica.²¹⁹ Severity of signs increases with the development of secondary bacterial pneumonia, and if death occurs it is usually the result of secondary bacterial infection. Infection with PI3 is widespread in sheep and goats.^220-221 Only one serotype of ovine PI3 has been identified, and it is related to but distinct from the bovine strain. Most infections are inapparent to mild.

Pathogenesis

Infection with PI3 can lead to signs referable to both upper and lower respiratory tract infection. After experimental infection of calves, clinical signs are evident by 2 days postinfection, and signs peak 4 days postinfection. Virus is found in the nasal passages, trachea, and bronchiolar and alveolar epithelial cells. The virus damages the pulmonary mucociliary apparatus²²² and depresses several important functions of alveolar macrophages such as Fc receptor expression, phagocytosis, and microbicidal activity,²²³ which are important in predisposing infected animals to secondary bacterial pneumonia.

Epidemiology

The widespread prevalence of antibodies to this virus indicates that it commonly circulates in ruminant populations.^191,221,224 This finding suggests the possibility of repeat infections or at least the persistence of antibodies after infection. Inapparent or subclinical infections with PI3 are common; in one report, 28 groups of calves seroconverted to PI3 over 8 months, but respiratory disease was seen in association with PI3 infection in only four of these groups.²²⁵ Although infection can often be inapparent, if environmental and managerial practices are suboptimal, PI3 may become an important initiator of respiratory tract disease. Infection appears to spread rapidly in susceptible cattle housed at high population densities and in close contact. In seroepidemiologic surveys evaluating groups of calves in herds experiencing outbreaks of respiratory disease, seroconversion to PI3 has been associated with 14% to 38% of outbreaks evaluated.^111,112,226 In some surveys PI3 is the virus most commonly isolated from the lungs of calves that die or are euthanized because of pneumonia.²²⁷ Feedlot cattle commonly seroconvert to PI3 soon after feedlot arrival, and seroconversion has been associated with treatment for respiratory disease in some¹⁸⁶ but not all¹⁸⁷ cases. In spite of the fact that feedlot cattle frequently become infected with PI3 after feedlot entry, a recent survey of necropsy findings at 72 Canadian feedlots found PI3 at postmortem examination of only 4% of the cases with pneumonia.¹⁸⁸ The low rate of isolation at necropsy suggests that if the virus contributes to the development of pneumonia, it may no longer be present by the time the animal dies or is euthanized.

Necropsy Findings

Lesions of PI3 infection alone are rarely seen during postmortem examination. Experimental PI3 infection results in congestion of the respiratory mucosa, swelling of lymph nodes associated with the respiratory tract, and lobular consolidation concentrated in the cranioventral lung.^218,219 Bronchiolitis and alveolitis are seen histologically with both proliferative and degenerative changes in the epithelial cells of the bronchioles and alveoli. Syncytia and intranuclear and intracytoplasmic inclusion bodies may be seen.^218,219 In many respects, pathologic features of PI3 infection are similar to those caused by BRSV, although the lesions produced by the latter are generally more extensive.

Diagnosis

PI3 can be isolated from nasal swabs of infected animals. Unlike BRSV, which is also in the paramyxovirus family, PI3 is not particularly difficult to isolate. Diagnosis can also be confirmed with paired serology, with hemagglutination inhibition or virus neutralizing assays most commonly used.

Treatment and Prevention

There is no specific treatment for infection with PI3; as for other respiratory viruses of ruminants, administration of appropriate antibiotics to prevent secondary infection with likely bacteria is recommended. As for any viral respiratory tract infection, supportive care is indicated, such as providing readily available good-quality feed and water and avoiding or postponing additional stressors such as movement or mixing of animals. Both inactivated and modified live PI3 vaccines are available for parenteral administration, and modified live PI3 vaccines are available in combination with BHV-1 for intranasal administration; these vaccines are labeled for use in cattle but not in sheep or goats. A univalent modified live intranasal PI3 vaccine is available in Europe.²¹⁸ Experimental challenge studies have shown that both parenteral and intranasal vaccines can decrease viral shedding and clinical signs after challenge.^216-218228 No field trials have specifically evaluated the effect of PI3 vaccines to decrease respiratory disease. One study reported that a modified live BHV-1/PI3 vaccine licensed for use in cattle decreased clinical signs and viral shedding when administered before experimental challenge of sheep, but the vaccinated sheep also appeared to become latently infected with BHV-1.²²⁸ The investigators noted that induction of latent BHV-1 infection could be an important negative side effect of vaccination, as sheep could theoretically spread BHV-1 to other in-contact ruminants, such as cattle. Although the practice appears to occur commonly, administration of BHV-1/PI3 vaccines to sheep or goats constitutes an extralabel use of these products.

Definition and Etiology

M. haemolytica (formerly Pasteurella haemolytica) is a gram-negative aerobic bacteria of the family Pasteurellaceae. There are at least 12 serotypes of M. haemolytica, and some ruminant isolates are untypable with currently available laboratory tools. Five serovars previously classified as M. haemolytica have been determined to be separate species and have been reclassified as Pasteurella trehalosi or Mannheimia glucosida.^255,256P. trehalosi causes pneumonia or systemic disease with multiorgan infection in sheep,²⁵⁷ whereas M. glucosida is a low virulence opportunistic pathogen of sheep.²⁵⁸ Some serotypes of M. haemolytica are pathogenic, whereas others are nonpathogenic commensals of the ruminant nasopharynx, and the serotypes that are pathogenic for cattle are not the same as those that are pathogenic for sheep or goats.^259-261 Serotype A1 is the most common isolate from pneumonic lungs of cattle, and serotype A6 is the next most common^262,263; serotype A2 is the most common isolate from the nasopharynx of normal cattle.²⁶⁴ Serotype is also relevant to pathogenicity of isolates from sheep and goats; both M. haemolytica and P. trehalosi can be isolated from the nasopharynx of normal small ruminants,²⁶⁵ but M. haemolytica serotype A2 is the most common isolate from sheep and goats with pneumonia, and serotypes A7, A9, and several others have also been associated with disease.^259,261,266

Clinical Signs

Cattle, sheep, or goats infected with M. haemolytica display a dull or depressed attitude and lose interest in eating. Fever, tachypnea, and depression are often the only abnormal signs early in the course of infection^267-269; coughing is not a prominent sign in the acute stage of disease^268,269 unless co-infection with another agent such as BHV-1 or BRSV is present. Animals may display evidence of thoracic pain, such as standing with elbows abducted or catching the breath before expiration; these signs are a result of the painful fibrinous pleuritis caused by M. haemolytica. Because M. haemolytica is a gram-negative organism, endotoxin (lipopolysaccharide) is produced by the agent, and thus cattle with mannheimiosis may show signs of endotoxemia, including fever, tachycardia, tachypnea, salivation, respiratory distress, pale or dark mucous membranes with prolonged capillary refill time, and cool extremities. Thoracic auscultation may reveal harsh or loud bronchovesicular sounds consistent with pulmonary consolidation,²⁶⁸ particularly over the cranioventral lung. Disease after M. haemolytica infection can be fatal, particularly if it is not treated; it can also lead to chronic pneumonia associated with secondary invasion with agents such as P. multocida, Mycoplasma bovis, or Arcanobacterium (Actinomyces) pyogenes.

It is important to remember that naturally occurring disease caused by M. haemolytica is commonly preceded by infection with a viral agent such as BHV-1, BRSV, or BVD; therefore the clinical signs described previously for these and other respiratory viruses may also be present in animals infected with M. haemolytica.

Pathogenesis

M. haemolytica is a normal inhabitant of the nasal pharyngeal mucosa,^96,265 but not the lung, and is considered an opportunistic pathogens. Calves and lambs become infected at an early age and carry Pasteurella as a minor part of the upper respiratory tract flora. Only a small percentage of nasal swabs yields positive results for M. haemolytica serotype A1 in healthy, unstressed calves,²⁷⁰ but if several areas of the nasal mucosa are cultured at necropsy, it is often possible to isolate M. haemolytica from animals that previously had negative findings on nasal swabs.²⁷¹ The stress of transportation or viral infection causes a breakdown of the defense mechanisms that hold the nasal mucosa infections in check, resulting in a rapid proliferation of virulent M. haemolytica serotype A1. A greater number of calves will yield positive nasal mucosa swab M. haemolytica results during and after transport, and there is a large increase in the numbers of M. haemolytica in positive samples.^270,272 BHV-1 and PI3 viruses have been shown to have the same effect as transportation on Pasteurella populations of the nasal mucosa. M. haemolytica has been demonstrated in the tracheal air of stressed, healthy calves harboring the organism on their nasal mucosa. Some of these inhaled organisms are deposited deep within the lung and normally are cleared within hours. But under conditions of impaired pulmonary defenses caused by stress, nutritional deficiencies, or preexisting viral infection, M. haemolytica is able to proliferate rapidly within the lung and with the aid of its virulence factors and toxins to produce a severe lobar necrotizing fibrinous pleuropneumonia. Calves infected with respiratory viruses, including BHV-1, PI3, BVDV, and BRSV, or Mycoplasma species have increased susceptibility to severe bronchopneumonia when exposed to M. haemolytica.⁹⁶ Similarly, infection of lambs with PI3 or adenovirus followed in several days by M. haemolytica causes severe pneumonia.^220,273

Once M. haemolytica becomes established in the lungs, interactions between the bacteria and the host defenses result in tissue damage and elimination of the invader. A major virulence factor of M. haemolytica is an exotoxin that is lethal to ruminant leukocytes, the leukotoxin. Leukotoxin, which is produced by M. haemolytica during the logarithmic phase of growth, causes cytolysis of ruminant platelets, lymphocytes, macrophages, and neutrophils.²⁷⁴ There is diversity in the gene encoding the leukotoxin molecule among M. haemolytica serotypes, and genetic diversity is related to differences in toxicity. For example, the leukotoxin encoded by some pathogenic bovine serotype A1 strains differs from the leukotoxin encoded by ovine A1 strains by only one amino acid but is substantially more toxic for bovine leukocytes than for ovine leukocytes.²⁶⁰ Leukotoxin binds to cells via CD18, the beta subunit of the β₂ integrins CD11a/CD18 (LFA-1), CD11b/CD18 (Mac-1), and CD11c/CD18 (CR-4)^275,276; although leukotoxin can bind to cells from nonruminant species, only ruminant cells are killed by the toxin.²⁷⁴ Contact with leukotoxin increases expression of LFA-1 on ruminant leukocytes, making the cells even more sensitive to injury by leukotoxin.²⁷⁷ At low concentrations, leukotoxin induces leukocyte death by apoptosis, whereas at higher concentrations the toxin causes cell lysis.²⁷⁸M. haemolytica expressing a mutant nontoxic leukotoxin induced less lung pathology in calves as compared with M. haemolytica producing functional toxin, but clinical signs were not different between the two groups of calves, indicating that leukotoxin is not the only virulence factor of importance in causing disease.²⁷⁹

Another virulence factor that certainly contributes to disease resulting from M. haemolytica is lipopolysaccharide, or endotoxin. As in all species, exposure of ruminants to endotoxin from gram-negative bacteria induces a multitude of responses leading to inflammation, including initiation of the complement and coagulation cascades, activation of endothelial cells and recruitment of neutrophils, and activation of neutrophils and alveolar macrophages, leading to their production of proinflammatory cytokines, which further amplify the inflammatory response. The proinflammatory cytokines TNF-α, IL-1β, and IL-8 are expressed in the lungs of cattle within 48 hours of M. haemolytica infection.²⁸⁰ Calves exposed to a preparation containing M. haemolytica endotoxin develop leukopenia, fever, tachypnea, diarrhea, and dyspnea as early as 2 hours after exposure.²⁸¹ Endotoxin potentiates the effects of leukotoxin by inducing increased expression of β₂-integrins on leukocytes, which contain CD18, the receptor for leukotoxin.^277,282 Bovine alveolar macrophages first exposed to endotoxin were susceptible to death induced by concentrations of leukotoxin too low to cause cell death alone; and macrophages exposed to both endotoxin and leukotoxin produced more of the proinflammatory cytokines IL-8 and TNF-α than did macrophages exposed only to leukotoxin.²⁸² These data indicate that leukotoxin and endotoxin work in synergy to cause disease in ruminants. The production of IL-8 is of particular importance, because IL-8 is a major inducer of neutrophil chemotaxis.²⁸³ The massive influx of neutrophils induced by IL-8 and other inflammatory mediators is a key factor in lung tissue destruction caused by M. haemolytica. Calves experimentally depleted of neutrophils are completely protected from gross and microscopic lesions of the severe fibrinonecrotic bronchopneumonia that is induced by intratracheal inoculation of M. haemolytica in calves with normal neutrophil levels.²⁸⁴ Lysis of neutrophils results in the release of lysosomal products, including elastase, collagenase, and reactive oxygen intermediates. These chemicals are bactericidal but also capable of destroying the neutrophils themselves and surrounding tissues. Neutrophil-mediated damage to the endothelial cells results in exudation and thrombosis, which produce the classic lesions of necrosis and fibrinous exudation.²⁸⁵

In addition to leukotoxin and endotoxin, M. haemolytica possesses other factors that contribute to its virulence. The bacteria have a polysaccharide capsule that aids in attachment and prevents phagocytosis by neutrophils,²⁸⁶ and iron-regulated outer membrane proteins (IROMPs) that bind transferrin and alter the function of neutrophils.^287,288 The bacteria produce adhesions that mediate attachment to host cells,²⁸⁹ and neuraminidase produced by the bacteria may aid in host colonization by decreasing the viscosity of respiratory mucus²⁹⁰ and decreasing the repellant negative charge on host cells by cleaving sialic acid residues.²⁹¹

Epidemiology

In many studies over several decades, M. haemolytica has been found to be the most common bacterial isolate from feedlot cattle with fatal fibrinous bronchopneumonia.^188,292-296M. haemolytica was isolated postmortem from 25% to 30% of cattle that died or were euthanized because of pneumonia in two studies evaluating causes of mortality in feedlot cattle.^188,296 Seroconversion to M. haemolytica or M. haemolytica leukotoxin has been significantly associated with treatment for respiratory disease or “undifferentiated fever” in feedlot cattle in some^186,297 although not all²⁹⁸ studies. Booker and colleagues found that the odds of developing undifferentiated fever (case definition similar to that usually used for undifferentiated respiratory disease) was 2.8 times greater for cattle that developed a fourfold rise in antibody titer to M. haemolytica leukotoxin, as compared with cattle that did not seroconvert to leukotoxin.²⁹⁷ And whereas O’Connor and co-workers found no association between seroconversion to M. haemolytica and treatment for undifferentiated respiratory disease, they did find that vaccination against M. haemolytica was significantly associated with protection against the development of undifferentiated respiratory disease in feedlot cattle in their study, which indirectly indicated a role for M. haemolytica in the development of respiratory disease.²⁹⁸ Although M. haemolytica is recognized as the bacteria most commonly contributing to fibrinous pneumonia in feedlot cattle, one study that used BAL to identify bacteria present in the lungs of feedlot cattle before antimicrobial treatment for acute pneumonia found that M. haemolytica was not identified more frequently in the lungs of cattle with pneumonia than normal controls, whereas P. multocida was significantly associated with bronchopneumonia.²⁹⁹ In the study by Allen and colleagues, none of the calves died or were euthanized because of the naturally occurring disease, indicating that disease was not severe. Because M. haemolytica is relatively more likely to cause acute fatal pneumonia than other bacterial pathogens of the ruminant lung, it may be overrepresented in necropsy surveys in which only fatal cases of disease are sampled.

In contrast to the situation in feedlot cattle, M. haemolytica is not commonly associated with outbreaks of pneumonia in calves. At postmortem examination of 43 calves from 34 outbreaks of respiratory disease, M. haemolytica was isolated from only two calves¹¹¹; in another study, M. haemolytica was isolated from calves in 4 of 14 outbreaks of respiratory disease.¹¹² Although M. haemolytica is not as commonly isolated from calves as are other bacterial pathogens, it can be a cause of fatal disease; in one study, only M. haemolytica was significantly more often isolated from calves with fatal pneumonia as compared with calves with subclinical pneumonia.³⁰⁰

Necropsy Findings

Infection with M. haemolytica causes fibrinopurulent bronchopneumonia (Fig. 31-56). The infection is aerogenous, so disease occurs primarily in the cranioventral lung, but in severe cases, a majority of the lung may be affected. Affected areas of lung are dark red to purple or gray-brown (Fig. 31-57), firm, and heavy; discolored areas may be wedge-shaped owing to thrombosis of a vessel supplying the affected region during the severe inflammatory response (Fig. 31-58). The interlobular septa are expanded by clear to yellow gelatinous material that represents proteinaceous fluid that has leaked from blood vessels in the lung. The inflamed areas of lung are covered with yellow fibrin that may adhere to the pleura of the thoracic wall, and the pleural cavity usually contains straw-colored fluid, which may be voluminous.^267-269 In cases that have been ongoing for a few days, firm, gritty lumps that are dry and crumbly on sectioning may be identified; these represent areas of necrotic tissue.²⁶⁹ Bronchial lymph nodes may be swollen, wet, and dark red. Histologically, alveoli are filled with edema and fibrin, and there is massive infiltration of neutrophils and macrophages. “Oat cells” can be seen, which are necrotic leukocytes with streaming of the chromatin.¹⁸⁸ Foci of coagulative necrosis are often found rimmed by a basophilic border of leukocytes; coagulative necrosis may expand to fill entire lobules. Hemorrhage is often present both within and between alveoli, and an occasional vessel may be found to contain a thrombus. Interlobular septa and lymphatics can be found dilated with edema and fibrin. Bronchioles are filled with leukocytes and may have foci of epithelial necrosis.^188,268,269M. haemolytica may also produce pneumonia characterized by firm, dark red pulmonary consolidation without fibrinous pleuritis; this is especially common in dairy calves or sheep and goats. Animals that survive the acute stage of pneumonia caused by M. haemolytica may have multiple abscesses and pleural adhesions; in these cases other bacteria such as P. multocida, Mycoplasma bovis, or Arcanobacterium (Actinomyces) pyogenes may be contributing to pathology.

Fig. 31-56 Postmortem photograph of fibrinopurulent bronchopneumonia caused by Mannheimia haemolytica. Note extensive dark red consolidated cranioventral lung and fibrin on pleural surface.

Photograph contributed by Dr. Amelia Woolums, University of Georgia, Athens, Ga.

Fig. 31-57 Postmortem photograph of necrotizing bronchopneumonia caused by Mannheimia haemolytica. Abnormal lung is dark red to gray-brown in color.

Photograph contributed by Dr. Amelia Woolums, University of Georgia, Athens, Ga.

Fig. 31-58 Postmortem photograph of infarct (dark red triangular lesion at uppermost edge of tissue) in lung of calf with necrotizing bronchopneumonia caused by Mannheimia haemolytica.

Photograph contributed by Dr. Amelia Woolums, University of Georgia, Athens, Ga.

Diagnosis

Diagnosis of pneumonia due to M. haemolytica is most reliably made by culture of lungs with typical gross pathology at postmortem evaluation of animals that die or are euthanized as representative cases in outbreaks. Antemortem diagnosis in individual animals can be made by collecting fluid by transtracheal aspiration (TTA) or BAL and submitting the fluid for aerobic bacterial culture. These methods are not practical for widespread use in the field, but they may be warranted for valuable individual animals or to identify the cause of an outbreak of respiratory disease in several animals. Cytologic evaluation of TTA or BAL samples would be expected to reveal septic purulent inflammation characterized by large numbers of neutrophils that may be degenerate, exhibiting toxic change, and possibly containing intracellular bacteria. Thoracocentesis can be a useful diagnostic procedure to identify fluid accumulation in the thorax; fluid collected shows a high percentage of neutrophils with a high (>3 g/dL) total protein content if bacterial pleuropneumonia is the cause of fluid accumulation. Pleural fluid collected by thoracocentesis can also be submitted for aerobic bacterial culture to confirm infection with M. haemolytica. In a hospital with the necessary equipment, transthoracic ultrasound evaluation can be used to confirm the presence of consolidated lung tissue and pleural effusion with fibrin.³⁰¹ Thoracic radiography, if possible, is expected to show evidence of consolidation of the cranial lung and possibly pleural fluid accumulation.³⁰² Evidence of pneumonia identified by either radiography or transthoracic ultrasound evaluation has been shown to correlate strongly with findings on postmortem examination of calves with bacterial bronchopneumonia.^301,302

Collecting bacteria by nasal swabs could be a reliable means of diagnosis if M. haemolytica isolated from swabs can be serotyped,³⁰³ but because nonpathogenic M. haemolytica can be present in nasal passages of normal animals and can be distinguished only by serotyping, simple identification of the bacteria in nasal swabs as identified by culture is not diagnostic.

Serologic tests can be used to identify serum antibodies to M. haemolytica, and seroconversion identified by paired serologic testing could be used to confirm infection with M. haemolytica in one or more ruminants. However, these tests are most commonly used for research applications, and they may not be available at all diagnostic laboratories.

Treatment and Prevention

Treatment of M. haemolytica requires administration of an antimicrobial effective against the organism. Many antimicrobials are currently labeled for the treatment of M. haemolytica (see Table 31-10). The decision to choose any one of the many approved products is based on multiple factors including regional susceptibility of M. haemolytica isolates, the number of times it is possible to treat animals, withdrawal times, and cost. See the discussion on pp. 630–633 for further information on treatment of bacterial bronchopneumonia.

In addition to appropriate antimicrobial therapy, treatment to prevent the adverse effects of endotoxin should be considered for ruminants suspected or confirmed to have pneumonia caused by M. haemolytica. The NSAID flunixin meglumine can ameliorate the inflammatory response to endotoxin, and treatment with flunixin meglumine has been shown to improve outcome in individual animals infected with M. haemolytica. However, administration of NSAID drugs such as flunixin meglumine has not been shown to be reliably cost-effective in the treatment of large numbers of feedlot cattle with respiratory disease. Thus a consideration of the value of the animals treated may be necessary before NSAID therapy can be justified in an entire group of calves.

Prevention of infection and disease caused by M. haemolytica is approached by three avenues: administering antimicrobials prophylactically to animals at high risk of disease; increasing host immunity by ensuring adequate passive transfer and administering vaccines against M. haemolytica; and minimizing factors such as viral respiratory tract infection, mixing of animals from various sources, and long-distance shipment, which increase the susceptibility of animals to infection and disease caused by pathogenic serotypes of M. haemolytica.

Prophylactic or metaphylactic administration of antimicrobials effective against M. haemolytica is a reliable means of decreasing morbidity and mortality associated with respiratory disease in calves at high risk for disease because of recent weaning, uncertain immune status, mixing with cattle from a variety of sources, and long-distance shipment. Administration of tilmicosin to groups of such high-risk calves either before shipment or on arrival at feedlots decreased the proliferation of M. haemolytica serotype A1 in the nasopharynx of calves and was associated with decreased treatment for respiratory disease as compared with calves not treated with tilmicoson.³⁰⁴ In a later study, administration of florfenicol at arrival decreased colonization of the nasopharynx with M. haemolytica A1 and delayed treatment for respiratory disease.²⁷⁰ Further discussion of the role of metaphylactic administration of antimicrobials to control bovine respiratory disease can be found on p. 639.

Mixing recently weaned calves from multiple sources and shipping them long distances is a well-known precursor to outbreaks of fibrinous pneumonia³⁰⁵; and because M. haemolytica is the species of bacteria most commonly associated with fatal fibrinous pneumonia, it follows that efforts to minimize stresses and improve immunity for calves moving through the marketing chain should lessen disease caused by M. haemolytica. Recent studies have indicated that preconditioning, wherein vaccination and stressful procedures such as weaning and castration are carried out well in advance of mixing and shipment, can decrease costs associated with fibrinous pneumonia in feedlot cattle.³⁰⁶ However, preconditioning should not be considered a guarantee against all respiratory disease; disease that is sometimes severe can occur in preconditioned calves. Further discussion of the role of preconditioning to prevent bovine respiratory disease is continued on p. 638.

Vaccination can be used as part of a plan to decrease disease caused by M. haemolytica. Many experimental and commercially marketed vaccines have been tested for efficacy in preventing either disease resulting from experimental infection with M. haemolytica or naturally occurring respiratory disease, and the published literature is too extensive to describe in detail here. In summary, some research indicates that vaccination can lessen disease, other research shows that vaccination has no effect, and a few older reports show that M. haemolytica vaccination can enhance disease after infection.³⁰⁷ Although commercially marketed vaccines are tested for safety, and no published reports exist that describe enhanced disease caused by currently available products, the fact that M. haemolytica bacterins historically caused enhanced disease is a reason to avoid the use of autogenous products that are not thoroughly tested for safety before administration. To induce protective immunity, a vaccine against M. haemolytica must contain leukotoxin; additional benefit may be gained by including other antigens associated with the bacterial cell wall.

Experimental challenge studies have shown that currently available commercial vaccines can lessen disease caused by infection with M. haemolytica.^308-310 However, data from well-designed field trials, which are a more relevant measure of the value of vaccination in the field, have reported mixed results. Most of the published field trials that have evaluated currently available (as of 2007) vaccines have evaluated a leukotoxin-rich bacterial culture supernatant vaccine (Presponse, Fort Dodge Animal Health, Fort Dodge, Ia). Of four trials carried out by different investigators in different regions, vaccinated calves had decreased respiratory morbidity (treatment for respiratory disease) in two trials,^311,312 but not in the other two trials.^313,314 Mortality resulting from respiratory disease was lower in vaccinated calves in two trials^312,313 but not in the other two trials.^311,314 Fewer vaccinated calves required a second treatment for respiratory disease (repulls) in three trials^312-314; this variable was not examined in the fourth trial. Two trials found an economic advantage to vaccinating calves with this vaccine^312,313; the other two trials did not measure costs related to vaccination. In all of these trials calves were vaccinated at feedlot arrival; in some cases subsets of calves also received a vaccination before arrival.^312,314

Another recent trial evaluated a commercially available vaccine containing leukotoxin and cell-associated antigens (Pulmo-Guard, Boehringer Ingelheim, St. Joseph, Mo.), and found that vaccination of calves at feedlot arrival had no effect on respiratory morbidity or mortality or on the number of calves that required a second treatment for respiratory disease.³¹⁵ A trial testing an M. haemolytica vaccine currently available in Canada (Pneumo-Star, Biostar Inc., Saskatoon, Saskatchewan) found that vaccination at arrival was associated with decreased respiratory morbidity.²⁹⁸

This short summary of clinical trials testing the efficacy of M. haemolytica vaccines yields a clue to the reason for the ongoing debate regarding whether these vaccines are useful: about half the time field trials indicate that the vaccines decrease illness or death related to respiratory disease, and about half the time they do not. Therefore there is support for using M. haemolytica vaccines, and there is support for not using them. Note also that no large-scale clinical trial has compared two different M. haemolytica vaccines, so it is not possible to give an evidence-based recommendation in favor of any vaccine over another. It must be remembered that in none of these clinical trials was the actual cause of respiratory disease determined, which is typical of large-scale field trials. Therefore vaccination was expected to decrease all respiratory disease, not just respiratory disease caused by M. haemolytica. This may be an unreasonable expectation for a vaccine from an immunologic perspective, but from a practical perspective, this is what producers expect. In general, if vaccination against M. haemolytica is used, evidence best supports its use in “lightweight” (300 to 600 lb) calves at high risk for disease. In one trial, vaccination decreased disease in calves with relatively high morbidity, but no effect was seen in a group of calves with low respiratory morbidity.³¹¹M. haemolytica vaccination is also recommended as part of many preconditioning programs, particularly if calves are to be sent directly to the feedlot without any backgrounding period.³¹⁶

Vaccination of sheep or goats using commercially available vaccines labeled for use in cattle is not likely to provide reliable protection against disease, as serotypes of M. haemolytica that most commonly cause disease in sheep and goats are not included in vaccines for cattle. Moreover, although vaccination with sheep-associated serotypes can provide protection against infection, cross protection among different serotypes is not reliable,²⁶¹ indicating that an effective vaccine will likely need to contain antigens from multiple serotypes. There is a vaccine labeled for use in sheep and goats in the United States (M. haemolytica–P. multocida Bacterin, Colorado Serum Co., Denver, Colo.), but published clinical trials testing the efficacy of this product are lacking.

Definition and Etiology

P. multocida is a gram-negative aerobic bacteria of the family Pasteurellaceae. Like M. haemolytica, P. multocida can be found in the nasopharynx of healthy ruminants. Although P. multocida is regularly isolated from the lungs of ruminants with bronchopneumonia, there has been debate over the years as to whether this species is a primary pathogen—that is, capable of causing disease alone—or whether some other primary stressor or insult is required to occur before this agent can participate in disease. As with M. haemolytica, experimental challenge of calves with P. multocida alone does cause clinical signs and pathologic change in the lung similar to that seen in natural outbreaks of disease (described later),^268,317,318 but large numbers of the bacteria must be administered in a way that bypasses the upper respiratory tract (most commonly by intratracheal instillation). Alternatively, prior administration of a viral respiratory pathogen makes animals more susceptible to disease caused by experimental P. multocida infection. These findings indicate that some insult that weakens the ability of the respiratory tract to resist advancement of these bacteria into the lower airways is necessary for most if not all cases of naturally occurring disease to occur. However, because P. multocida can cause lung lesions that can be severe,³¹⁷ and because the bacteria can exacerbate disease caused by primary viral infection, it is logical to consider this agent when planning strategies to treat and control ruminant respiratory disease.

P. multocida is a diverse species of bacteria that is classified into five serogroups (A, B, D, E, and F) based on antigenic differences in the bacterial capsule. Serogroups B and E cause hemorrhagic septicemia, a disease predominantly seen in Asia and Africa³¹⁹; these serogroups are rarely isolated in the United States. Serogroup A is by far the predominant serogroup associated with ruminant respiratory disease,^263,320 although serogroups D and F may be common in some regions, particularly in sheep.³²¹ In addition to the alphabetic serogroup designation, isolates may also receive a numeric designation based on cell wall antigen types (for example, P. multocida A3 has been commonly isolated from cases of bovine pneumonia).³¹⁸ A recent study characterized 153 P. multocida isolates from cases of pneumonia and mastitis in cattle in England and Wales based on the outer membrane proteins expressed by the bacteria. This research indicated that relatively few strains of P. multocida cause the majority of disease in cattle. Moreover, although a few strains isolated from cattle had also been associated with disease in swine and poultry, the majority of strains were uniquely associated with their host species of origin.³²⁰ Other research indicated that certain strains of P. multocida had a predilection for the respiratory tract of sheep, whereas other strains were associated with the reproductive tract.³²¹ Taken together, these data suggest that differences among strains of P. multocida are related to the type of disease caused and the host likely to be affected. Therefore merely isolating P. multocida without further characterizing the isolate could make it difficult to know with confidence whether the type of bacteria isolated was actually contributing to disease. It has been suggested that strain variation is related to differences in severity of respiratory disease occurring in ruminants infected with P. multocida.³¹⁷

Clinical Signs

Calves infected with P. multocida alone display clinical signs of fever, increased respiratory rate, and sometimes depression, coughing, and mucoid to mucopurulent nasal discharge.^268,317,318 Loud or harsh bronchovesicular sounds may be heard over the cranioventral lung fields owing to pulmonary consolidation, and coarse crackles may be heard as a result of air moving through exudate in large airways. Because P. multocida is a gram-negative organism that produces endotoxin, signs of endotoxemia (tachypnea, tachycardia, fever) may also contribute to signs that accompany infection with the organism. Compared with calves infected with M. haemolytica, calves infected with P. multocida tend to have less severe signs, and signs of disease last for a shorter time.^268,317P. multocida is more commonly isolated from young calves with pneumonia^111,112 than with other bacterial respiratory pathogens such as M. haemolytica, but it is less commonly isolated from feedlot cattle with acute bronchopneumonia.¹⁸⁸ Thus P. multocida is considered to be of relatively greatest importance in contributing to calf pneumonia. P. multocida has been found to overgrow M. haemolytica in the lungs of calves experimentally challenged only with M. haemolytica²⁶⁸; therefore P. multocida may also be an important contributor to cases of subacute to chronic pneumonia in feedlot cattle initiated by other organisms.

Pathogenesis

Little is known regarding the pathogenesis of P. multocida in ruminant respiratory disease. In addition to lipopolysaccharide (LPS), the organism also has a capsule that allows it to resist phagocytosis. Outer membrane proteins, particularly iron-regulated outer membrane proteins, are likely to contribute to the ability of the bacteria to establish and proliferate within the host.^322,323 The prominent role of P. multocida in enzootic calf pneumonia (ECP) and rather minor role in acute bronchopneumonia of feedlot cattle suggests that prolonged impairment of the respiratory defense mechanisms is necessary for this organism to establish in the lungs in sufficient numbers to create bronchopneumonia. It is likely that the organism is chronically inhaled in small numbers into the lungs of calves with persistent damage to the respiratory defenses from infectious agents such as viruses or mycoplasmas and environmental damage from inadequate housing and ventilation, allowing it to colonize and produce an expanding lesion. This proposed pathogenesis agrees with the insidious onset that is commonly observed with ECP, and would also fit the association of P. multocida with chronic or ongoing pneumonia in feedlot cattle. It is important to remember that culture of P. multocida from pneumonic lung does not exclude the possibility of other bacteria such as M. haemolytica being the primary pathogen, because P. multocida has been shown to overgrow M. haemolytica in challenge studies using large doses of pure cultures of M. haemolytica.²⁶⁸

Epidemiology

P. multocida is commonly isolated from the lungs of calves that die or are euthanized because of bronchopneumonia, with mycoplasmas being the only bacteria isolated more often in surveys involving multiple farms experiencing outbreaks of calf pneumonia.^111,112 In contrast, M. haemolytica is more frequently isolated postmortem from feedlot cattle with fibrinous pneumonia than is P. multocida.^188,296 However, one study of bacterial isolates from BAL of feedlot cattle found that P. multocida was the only species isolated more frequently from cattle with clinical signs attributed to acute pneumonia as compared with normal cattle.²⁹⁹ A notable aspect of this study was that animals were sampled before treatment with antimicrobials; in surveys of bacteria identified postmortem, animals have usually received antimicrobial treatment before death, which may bias the microbiologic findings.

Necropsy Findings

P. multocida produces a purulent bronchopneumonia with plum-colored cranioventral consolidation and purulent exudate on cut section within the airways; when calves are experimentally infected with P. multocida alone, the lesion is usually not extensive and is typically confined to the cranioventral lung lobes.^268,317,318 Histologically, bronchopneumonia characterized by infiltration of neutrophils and macrophages into alveoli and airways is seen. Microscopic evidence of abscesses may be present. Fibrin deposition with expansion of lymphatics and interlobular septa with edema, and focal areas of coagulative necrosis, has been reported in calves infected with P. multocida,^317,318 but this lesion is more typical of infection with M. haemolytica.²⁶⁸

Diagnosis

Bronchopneumonia caused by P. multocida is diagnosed as described for M. haemolytica. The species is most commonly identified by culture of lung lesions identified at postmortem examination of affected animals. Because of the association of P. multocida with chronic or ongoing pneumonia, identification of this agent may indicate that the animal has been affected with bronchopneumonia for several days to weeks.

Treatment and Prevention

Treatment of P. multocida is as described for M. haemolytica. Several antimicrobials are labeled specifically for the treatment of P. multocida, and products labeled for the treatment of M. haemolytica are also likely to be effective against this organism (see Table 31-10). Because P. multocida seems to be associated with chronic or ongoing pneumonia, effective treatment of infection with this species may require longer therapy than the 3 to 5 days historically recommended for ruminants with bronchopneumonia, but this recommendation has not been tested in controlled studies and would be an extralabel use of some antimicrobials.

Because prior insult to respiratory defense mechanisms appears necessary for infection with P. multocida to cause disease, prevention of disease is likely to be aided by undertaking efforts to prevent other primary respiratory injury. This would include efforts to prevent infection with viral respiratory pathogens and to establish management practices that help minimize respiratory tract disease.

As compared with M. haemolytica, relatively little is known about protective immunity to P. multocida. Modified live vaccines can protect calves from disease caused by experimental challenge.^324,325 Antibody responses to several outer membrane proteins were associated with protection in one study, and these were induced by live but not killed vaccine.³²⁵ Several commercial vaccines are marketed that contain P. multocida in combination with M. haemolytica; one such vaccine is approved for use in sheep and goats as well as cattle (M. haemolytica–P. multocida Bacterin, Colorado Serum, Denver, Colo.). There are currently no vaccines marketed that contain only P. multocida. No published large-scale trials have specifically evaluated the effect of vaccination against P. multocida on respiratory disease in the field.

Definition and Etiology

H. somni is a gram-negative aerobic bacteria of the family Pasteurellaceae. The name of this organism was recently changed form the previous name, Haemophilus somnus.³²⁶ H. somni can be found on the genital and upper respiratory mucosa of normal ruminants,^327-330 but it can also cause a variety of diseases, including septicemia, thrombotic meningoencephalitis (TEME), endometritis, abortion and infertility, pneumonia, pleuritis, laryngitis, otitis, conjunctivitis, myocarditis, mastitis, and polyarthritis.³³¹ Only pneumonia and pleuritis resulting from this organism are considered here. Two bacteria very similar to H. somni, Haemophilus agni and Haemophilus ovis, have been isolated from sheep with septicemia, meningitis, mastitis, and reproductive abnormalities; DNA hybridization studies have indicated that these three organisms should all be classified as H. somni.³³² Although specific serotypes of H. somni have not been associated with disease as is the case for M. haemolytica and P. multocida, differences in virulence among different strains have been demonstrated in experimental challenge studies.³³³ Moreover, differences in expression of molecules considered to be virulence factors have been demonstrated when strains from healthy animals are compared with strains from diseased animals.³³⁴ Thus it appears that some isolates of H. somni are more likely to cause disease than others, but more research is needed to confirm the characteristics that define a pathogenic isolate.

Clinical Signs

As mentioned earlier, H. somni can cause disease in a variety of organ systems, and the clinical signs of infection will depend on the organ system affected. The possibility of concurrent infection of multiple systems should be considered in patients suspected to have disease caused by H. somnus. In a Canadian retrospective study from the early 1990s, the majority of animals presented to a regional diagnostic laboratory for necropsy as a result of haemophilosis had disease in more than one organ system.³³⁵ Signs of respiratory tract infection can range from mild to severe and can include fever, tachypnea, cough, nasal discharge, and depression. Severe cases of disease can be fatal.^333,336 Harsh or loud bronchovesicular sounds may be heard on thoracic auscultation because of pulmonary consolidation, particularly over the cranioventral lung fields. Affected animals may have signs of pain owing to fibrinous pleuritis,³³⁶ including reluctance to move, standing with elbows abducted, and catching the breath before expiration. The cell wall of H. somni contains lipooligosaccharide (LOS), which induces inflammatory responses identical to those induced by LPS from E. coli³³⁷; therefore animals with pleuritis or pneumonia caused by H. somni could also have clinical signs referable to endotoxemia, including tachypnea, tachycardia, dark or pale mucous membranes with prolonged capillary refill time, salivation, or dyspnea.

Pathogenesis

H. somni can live on respiratory or genital mucous membranes without causing disease, and it is not entirely clear what factors related to the pathogen or host must change for disease to occur. The physical and immunologic barriers presented by the upper respiratory tract are apparently of major significance; calves exposed to H. somni by aerosol did not develop disease,³³⁸ whereas instillation of the bacteria directly into the trachea or bronchi led to disease that could be severe.^333,336 As has been shown for M. haemolytica and P. multocida, primary infection by a viral respiratory pathogen is likely a predisposing factor that allows H. somni to advance and establish in the lower respiratory tract in many cases. Calves infected with BRSV before infection with H. somni had disease of greater severity than that seen in calves infected with either BRSV or H. somni alone.^169,336

H. somni has many features that help the bacteria escape the immune response. The bacteria have outer membrane proteins that bind to the Fc region of antibody, allowing them to escape opsonization.³³⁴ When the bacteria are ingested by neutrophils or macrophages, they are able to resist being killed.³³⁹ The bacterial LOS induces inflammatory responses similar to the LPS of other gram-negative bacteria,³³⁷ and H. somni is able to periodically change the structure and antigenicity of its LOS, which is likely to be important in evading the host immune response.³⁴⁰

An important aspect of the pathology caused by H. somni is the formation of vasculitis and vascular thrombi; this was first recognized in the context of the neurologic lesion caused by the bacteria, which is known as thrombotic meningoencephalomyelitis (TME), but vascular thrombi are also identified histologically in the lungs of animals with pneumonia caused by H. somni.³³⁶ For some time the exact cause of the vascular thrombosis typical of H. somni infection was not known, but recent research indicates that the bacteria cause death of vascular endothelial cells by inducing apoptosis (programmed cell death).³⁴¹ Apoptotic death of endothelial cells is mediated in part through the bacterial LOS, which interacts with the P2X7 purinergic receptor on endothelial cells, inducing activation of certain caspases, enzymes that activate the apoptotic pathway and initiate a cell suicide program.³²⁸ Although the entire pathway leading to vasculitis and thrombosis has not been completely elucidated, thrombosis is no doubt mediated in part by exposure of the vascular basement membrane on endothelial cell death, which leads to exposure of the basement membrane, activation of platelets and the coagulation cascade, and thrombosis.

Another important aspect of H. somni is the propensity of infection to induce IgE production by the host.^169,342 Infection with BRSV before infection with H. somni led to production of high levels of H. somni–specific IgE, which was associated with disease of increased severity as compared with control calves.¹⁶⁹ Cattle vaccinated with four different H. somni bacterins were all found to develop serum levels of H. somni—specific IgE that were significantly higher than levels seen in control unvaccinated cattle.³⁴² Because IgE mediates type I (immediate) hypersensitivity, animals that produce IgE against H. somni after vaccination or primary infection could have signs of an allergic or anaphylactic response on subsequent revaccination or reinfection. Such IgE-mediated hypersensitivity reactions have been proposed to contribute to the adverse reactions sometimes seen in cattle after vaccination for H. somni.³⁴² In addition to inducing IgE production, H. somni directly produces histamine,³⁴³ which could further contribute to the development of hypersensitivity-like signs during H. somni infection. Among other actions, histamine increases permeability of bronchial epithelium; this may aid the bacteria in moving out of the airway and into the lung parenchyma.

Epidemiology

Exposure to H. somni is common, with 25% to 100% of cattle in various populations having serum antibodies.³⁴⁴ A retrospective study of bovine carcasses submitted to a regional diagnostic laboratory in western Canada from 1970 to 1990 indicated that cases affected with the neurologic form of haemophilosis were decreasing, whereas cases with respiratory and/or myocardial disease caused by the organism were increasing.³³⁵ The reason for the apparent shift in type of disease caused in cattle could not be determined from the study, and a similar study has not been repeated to see if the trend is continuing.

It is common for feedlot cattle to have measurable antibody titers to H. somni at feedlot entry,^187,297 indicating that cattle are commonly exposed on the farm of origin or in transit to the feedlot. Seroconversion is also common in the first few weeks after feedlot entry,^297,298 indicating that cattle also continue to be exposed to H. somni in the feedlot. Several Canadian reports have indicated that H. somni can be a significant contributor to the development of respiratory disease or undifferentiated fever in feedlot cattle,^297,345,346 although studies of feedlot respiratory disease do not always find the bacteria involved. A causative association is typically inferred either by the association of seroconversion with respiratory morbidity and mortality during the period of study or by the association of H. somni vaccination with decreased respiratory morbidity and mortality. By these measures it appears that most disease resulting from H. somni occurs within the first 2 months of the feeding period,³⁴⁶ and perhaps even within the first 2 weeks.²⁹⁸ In some cases the association of seroconversion or vaccination with all morbidity and mortality is also evaluated, but one study found that when causes of mortality were separated into those in which it was biologically plausible for H. somni to be involved and those in which it was not (e.g., cattle with musculoskeletal injuries), vaccination was associated with decreased mortality attributable to H. somni but not with a decrease in all causes of mortality.³⁴⁶ A causative association for H. somni (or any respiratory pathogen) is sometimes assumed when animals with high antibody titers to the organism at feedlot entry ultimately have lower respiratory morbidity and mortality,²⁹⁷ although the case has been made that this is not a reliable marker of causation.²⁹⁸ A decreasing antibody titer to H. somni was associated with decreased likelihood of disease in one study²⁹⁷ and increased likelihood of disease in another¹⁸⁷; this demonstrates that it is not always easy to develop a unified theory regarding the role of infectious agents and the immune response from the available data. It is also noteworthy that high-quality field research on the role of H. somni in feedlot respiratory disease in recent years has come almost entirely from Canada; it is not clear whether this indicates that H. somni is a less significant contributor to disease in the United States or whether the research has just not been done.

H. somni is uncommonly isolated in surveys of the causes of pneumonia in young calves,^112,300 but it can cause bronchopneumonia that is significant; one necropsy survey found pneumonia in 12 of 15 calves under 8 weeks of age submitted because of disease resulting from H. somni, and in 8 of the 15 calves, pneumonia was the only lesion found.³³⁵

Necropsy Findings

H. somni produces lesions similar to that of P. multocida, and it may rarely produce lesions that resemble M. haemolytica, creating a fibrinous pleuropneumonia.^331,333,336 Grossly, plum or red to brown consolidated lobules are seen in the cranioventral lung (Fig. 31-59), sometimes with abscesses containing brown-red fluid material. Interlobular septa can be distended with edema and fibrin, and hemorrhage may be grossly visible.³³³ Purulent material is seen within airways on the cut surface of lung. The surface of the pleura may be flecked with fibrin, or in some cases fibrin deposition may be extensive, with varying amounts of straw-colored fluid in the pleural cavity. Bronchial lymph nodes may be enlarged, with ecchymotic hemorrhages on the cut surface.

Fig. 31-59 Postmortem photograph of lungs from feedlot steer with bronchopneumonia caused by Haemophilus somni. Note extensive dark red, consolidated cranioventral lung. Laceration of caudoventral lung is iatrogenic and not relevant to the lesion.

Photograph contributed by Feedlot Health Management Services, Okotoks, AB, Canada.

Histologically, inflammatory cells predominately made up of neutrophils infiltrate the alveoli and airways. Edema, hemorrhage, and fibrin can be seen in alveoli and interstitial spaces, and areas of coagulation necrosis surrounded by inflammatory cells may be found. Interlobular septa are expanded with fibrin, and thrombi are seen in blood vessels.^333,336

A massive fibrinous pleuritis with pleural effusion sometimes results from septicemic spread of H. somni. This condition can be differentiated from the fibrinous necrotizing lobar pleuropneumonia of shipping fever caused by M. haemolytica because the H. somni lesion involves only the pleural surface, not the lung.

Diagnosis

Bronchopneumonia or pleuropneumonia caused by H. somni is diagnosed as described for M. haemolytica. The bacteria can be difficult to isolate, so samples collected for culture should be transported to the diagnostic laboratory without delay, and the diagnostic bacteriology laboratory should be specifically requested to attempt to isolate H. somni if involvement of the agent is suspected. Samples are ideally taken from animals before antimicrobial treatment, as H. somni is susceptible to a wide variety of antimicrobials, and treatment makes it difficult to isolate the bacteria.³³¹ Serologic tests have been used to identify antibodies to H. somni for studies of the epidemiology of H. somni infection,^297,298 and seroconversion as measured via paired serology could be used to confirm infection in a group of cattle, but these assays may not be available at all diagnostic laboratories. IHC has been used to identify H. somni in association with lesions in formalin-fixed tissues,³⁴⁷ and this may be an additional test available at some diagnostic laboratories.

Treatment and Prevention

H. somni is susceptible to a variety of antimicrobials, including tetracycline, penicillin, sulfonamides, and erythromycin, as well as more recently developed antimicrobials, many of which are labeled for use in the treatment of H. somni (see Table 31-10). In a feedlot with unusually high morbidity and mortality because of haemophilosis, prophylactic treatment with oxytetracycline was administered at different time points within the first 2 weeks of the feeding period in an effort to decrease haemophilosis morbidity or mortality, to no avail (although morbidity from all causes of respiratory disease was decreased).³⁴⁸

Several H. somni vaccines are commercially available; all are killed whole bacteria preparations (bacterins). Vaccination can prevent disease caused by experimental challenge; in one study, vaccination of 10-week-old calves with a commercial bacterin significantly decreased clinical signs and pulmonary pathology resulting from H. somni challenge.³³³ Clinical trials of H. somni bacterins in the 1980s gave mixed results, with evidence that vaccination decreased respiratory morbidity and/or mortality in some trials but not in others^346,349,350; and very little research has provided more information since then. In a cow-calf herd with a high incidence of calf pneumonia, vaccination of calves at 3 and 5 weeks of age with an M. haemolytica and H. somni bacterin in combination with a modified live BRSV vaccine led to decreased treatment rates as compared with calves not vaccinated or calves vaccinated with either the bacterin or the BRSV vaccine alone.³⁵¹ The decrease in treatment rates only tended toward statistical significance; this may have been because small numbers of calves were enrolled in the study. The specific influence of H. somni vaccination in this trial could not be determined. The same authors found that vaccination of cows with M. haemolytica and H. somni bacterin 4 weeks and/or 7 weeks prepartum increased antibody titers to both pathogens in the serum of their calves at 3 days and 1 month of age. Furthermore, calves vaccinated at 1 and 2 months in the face of maternal antibody had significantly higher titers at 4 and 6 months of age than unvaccinated calves.³⁵² The effect of antibody titers on disease in the calves was not examined. A recent trial of the effect of vaccination on respiratory morbidity in feedlot cattle showed no effect of H. somni vaccination.²⁹⁸

The fact that H. somni vaccines induce cattle to produce IgE against the bacteria³⁴² indicates that vaccinated cattle may be at increased risk for a hypersensitivity reaction after a booster vaccine or infection. Although some older clinical trials showed a benefit associated with vaccination in terms of decreased respiratory morbidity and/or mortality, H. somni vaccination may also put cattle at risk for adverse reactions. Therefore it may be most prudent to recommend vaccination only for groups of cattle on operations in which significant morbidity or mortality caused by infection with H. somni has been confirmed.

As described for M. haemolytica and P. multocida, preventing primary injury to the respiratory tract by viral infection and limiting other causes of stress that suppress the host immune response are expected to help animals resist disease caused by H. somni.