Emerging respiratory infections (Proceedings)

Recent years have seen the emergence of previously undescribed respiratory infections in dogs and cats. Although these infections remain rare, the potential exists for substantial morbidity and mortality. While there are several respiratory infections that might be described as "emerging" (eg, Pneumocystis jirovecii, Streptococcus equi subspecies zooepidemicus, respiratory coronavirus), we will focus on only several specific infections including influenza and virulent systemic calicivirus (VS-FCV).

Influenza viral infection

Until less than a decade ago, dogs and cats were not thought to be susceptible to illness due influenza. Influenza viruses, single-stranded negative-sense RNA viruses in the family orthomyxovirdae, are named numerically according to the hemagglutinin (H) and neuraminidase (N) expressed on the virus. While many influenza viruses are adapted to a particular species, they may be able to infect multiple different species or add species affinity. All influenza viruses are highly susceptible to genomic change through either genomic drift (mutation) or genomic shift (interchange of entire segments of genome from one virus to another). Canine influenza is a high morbidity, low mortality infection that is reported from most of the USA. Feline infection with virulent avian influenza has not been reported in the USA, but is important not only as a very high mortality infection, but as a possible zoonotic risk. Both cats and dogs have been infected with the triple reassortment H1N1 virus often described as "swine flu".

Canine influenza

Canine influenza was first described in a kennel of racing greyhounds in 2004. Apparently, equine influenza was able to "jump species" from the horse to dog in a fashion that allowed subsequent spread within the canine species. The H3N8 virus seems to be adapting well to dogs, and it growing slightly more distant from the original equine virus. While the original description was of a severe hemorrhagic pneumonia with a relatively high mortality, it now seems that the typical course of infection is less severe. In fact, unless dogs are densely housed or concurrently infected with other pathogens, canine influenza may be largely another potential cause of Canine Infectious Respiratory Disease complex (i.e., CIRD, aka, infectious tracheobronchitis, aka kennel cough). The incidence and prevalence of CIV as a cause of CIRD is yet to be determined, although there are a few epidemiologic studies now beginning the early stages of investigation. It remains to be determined how often kennel cough is due to H3N8 infection in the USA.

The list of differential diagnoses for acute cough in dogs includes numerous cardiac and respiratory diseases. Infectious causes are also many, and include viruses (e.g. canine influenza virus, respiratory corona virus, distemper virus, parainfluenza virus, adenovirus 2), bacteria (e.g. Bordetella bronchiseptica, Mycoplasma spp., Streptococcus equi subsp. zooepidemicus), protozoa, mycoses, and parasites. A history of recent exposure to other dogs with cough or to group housing situations is supportive of a diagnosis of infectious tracheobronchitis. Influenza has a shorter incubation period (< 1 week) compared to Bordetella bronchiseptica (up to 2 weeks). Because most dogs are naïve to influenza antigens, spread of disease within an exposed population of dogs may be more extensive than the spread of other common causes of CIRD. Compared to CIRD associated with Bordetella, dogs with influenza are often described as more likely to act lethargic, have a soft, moist cough, have purulent nasal discharge, and are more likely to be febrile. Dogs with influenza may be more likely to develop pneumonia than dogs with Bordetella, but this tendency is not well documented.

Systemic screening tests (CBC, serum biochemical panel, urinalysis) and thoracic radiographs are indicated in all patients with signs suggestive of pneumonia. It should be noted that neither a normal CBC nor body temperature can be used to rule out a diagnosis of pneumonia because only half of patients with pneumonia have a fever or leukocytosis on presentation. If thoracic radiographs confirm the presence of bronchopneumonia and the patient is sufficiently stable, then a tracheal wash and culture of collected fluid should performed prior to initiating antibiotic therapy. While influenza is a viral infection and will not be confirmed on tracheal wash, in the sickest dogs secondary bacterial infection is common. Definitive diagnosis may not be necessary in dogs with milder clinical signs due to influenza. However, a definitive diagnosis may be useful for developing recommendations for other exposed dogs and for kennel or shelter situations. Serology, ELISA for antigen detection, virus isolation, and PCR can be used to confirm a diagnosis of influenza. Advantages of serology are that blood is a simple specimen to collect and infection can be detected even after viral shedding has ceased. Paired serology is the most sensitive means of confirming influenza infection. The main weakness of serology is that confirmation must be delayed pending collection of convalescent serum to confirm rising antibody titers or seroconversion. More timely results are possible with antigen detection methods and PCR. Preliminary data using nasal swabs for specimens indicates that PCR is more sensitive in detecting virus than antigen detection by ELISA or virus isolation. There are commercial ELISA kits made for confirmation of human influenza A which are readily used for dogs, but as with other antigen detection methods depend on viral shedding to be positive. Other specimens that can be submitted for virus isolation or PCR are pharyngeal swabs, tracheal wash fluid, or lung tissue. Results from any test for viral detection can be falsely negative because of the relatively short period of shedding after the development of signs in most patients. For best results, samples are collected from febrile dogs very early in the course of disease. When a kennel or shelter is involved, multiple dogs should be tested (minimum of 3-5, or 10-30% of the affected population). Some laboratories offer combination PCR tests which simultaneously test for a variety of pathogens involved in CIRD; these combination tests are especially useful in outbreak investigation and are less useful on a single ill pet.

Specific treatment for influenza in people is with antivirals such as oseltamivir phosphate (i.e., Tamiflu). To be effective, the drug must be given within 12-48 hours of the onset of signs, and the major benefit is a decrease in the median duration until improvement of signs by just over 1 day. No controlled efficacy trials are available for dogs, and dose and toxicity studies have not been performed in dogs although the drug has been used by individuals to treat dogs with upper respiratory tract signs and parvo virus infection. It is unlikely that a sufficiently early diagnosis will be possible in most cases for this drug to have practical benefit. Additional considerations which limit the potential utility of the antiviral drug in dogs are the possibility of encouraging viral resistance and the appropriateness of use of a sometimes scarce human resource, particularly in the absence of data supporting any benefit or for the dog with mild clinical signs.

Effectiveness of non-specific treatments for mild, infectious cough has not been extensively explored. One limited field study of dogs with infectious tracheobronchitis found no benefit to the administration of steroids. Cough suppressants are often prescribed. Potential benefits are reducing secondary inflammation caused by persistent coughing and allowing rest for the patient and owners. A potential disadvantage is decreasing an important airway clearance mechanism in a patient who likely has disruption of normal ciliary function. Cough suppressants are always contraindicated in animals with pneumonia.

Although primary viral pneumonia due to influenza does occur on occasions, most dogs with pneumonia related to influenza have secondary bacterial infections. Therefore, dogs with severe pneumonia due to influenza are treated according to general recommendations for treating bacterial bronchopneumonia. Initially, antibiotics should be broad-spectrum. Modification to treatment is then based on results of cultures of airway specimens and response to treatment. A variety of bacteria have been isolated from dogs with influenza, including Streptococcus equi subsp. zooepidemicus and Gram-negative organisms that are resistant to commonly prescribed antibiotics. Antibiotics to consider for initial treatment of severe pneumonia include the combination of ampicillin with sulbactam and either a fluoroquinolone or an aminoglycoside, or meropenem. In a hospital, shelter, or other group housing environment, it is critical to immediately isolate any acutely coughing dog and to disinfect all areas of contact. Influenza may be shed for as long as 10 days in infected dogs, and can be shed by asymptomatic dogs.

There is now a vaccine against canine influenza. The killed vaccine is given in 2 doses, 2-4 weeks apart and is labeled to "Aid in the control of disease associated with canine influenza virus infection. The product has been demonstrated to reduce incidence and severity of lung lesions as well as the duration of coughing and virus shedding." It should be recognized that the vaccine does not provide sterilizing immunity. While there is no official position from groups such as the AAHA on the utility of H3N8 vaccination, it is likely it would be considered a conditional vaccine much like the vaccine for B. bronchiseptica. Vaccination for other respiratory pathogens may decrease the severity of illness in dogs in kennel or shelter settings when exposed to multiple respiratory pathogens.

In addition to H3N8 canine influenza, dogs are susceptible to infection with other types of influenza A. Infection with the H1N1 triple reassortment virus (North American influenza) and the H3N2 avian influenza has been reported in pet dogs, albeit very rarely.

Feline influenza

As with dogs, cats were thought immune to influenza until recently. Unlike the story with dogs, cats do not (yet) seem to have a type of influenza virus adapted specifically to the species and readily spread within the cat population. Instead, cats have been documented to become infected with the virulent avian influenza H5N1, with the triple reassortment H1N1, and experimentally with other types of influenza A. Cats in Asia (and in at least a few cases, Europe) have become infected naturally with H5N1, but so far no reports exist of infection in North America.

There are many types of avian influenza, some of which are more pathogenic than others. The highly pathogenic avian influenza H5N1 gained international attention after it was transmitted from infected birds to humans in Asia. This virus was responsible not only for tremendous morbidity and mortality in Asian poultry, but also for high-mortality (at least 50%) infections in humans.

Most feline infections are likely acquired when the cat eats an infected but uncooked birds, but cat-to-cat transmission is also possible via feco-oral or respiratory routes. Experimental infections have been transmitted by direct inoculation of the virus parenterally or through the respiratory tract. Some of the first reports of this virus infecting cats came from Thailand, where tigers and leopards housed in a zoo became infected. At first attributed entirely to consumption of uncooked and infected poultry, it soon became clear that the virus could be passed between cats (including domestic cats). At the time, the fear was that this lethal virus would gain the ability to be passed not only from poultry to humans, but then from human-to-human resulting in a human pandemic. Experimentally, the H5N1 virus isolated from a human was able to produce a lethal infection in domestic cats. The obvious question then becomes, can infected cats spread this highly pathogenic disease to humans? It seems that cat-to-cat transmission is dependant to the level of virus shedding; it is not known how much shedding would be required (if possible) to spread this infection from cats to humans.

It is believed that viral replication initially occurs in the respiratory and GI tract, with virus spreading through viremia to most organs. It is apparent that some cats can be subclinically infected and remain well; these cats most likely have initial exposure to lesser viral loads. In those with greater exposure, extensive pulmonary damage and multifocal organ hemorrhage and necrosis are responsible for mortality. Neurologic signs including ataxia and seizure in naturally infected cats likely results from non-suppurative encephalitis. More common signs are non-specific, and include fever, depression, elevation of the third eye lid, conjunctivitis, increased respiratory effort, nasal discharge, and icterus. Sometimes, sudden death is observed within days of infection.

There are many more likely causes for an acute onset of respiratory signs in cats (eg, herpes virus, calicivirus, bacterial pneumonia). Documented infections in birds in the area, high fever, and access to the outdoors all increase the index of suspicion for influenza. Viral isolation from oropharyngeal or rectal swabs or necropsy specimens or RT-PCR are the typical methods of confirmation. Immunohistochemistry can be used on infected organs. Serologic diagnosis using hemagglutination inhibition is also possible. Treatment is largely supportive, although the wisdom of treating a cat with confirmed infection is uncertain considering the possible zoonotic potential of this disease. If treatment is perused gloves, mask and goggles should be worn and the cat should be kept in isolation. Theoretically, antiviral drugs like Tamiflu might be useful but the drug did not protect tigers. The appropriate dose for cats is unknown, and effective dose of this antiviral varies between species. Should virulent avian influenza be identified regionally, it would be wise to keep cats indoors and to feed only cooked or processed cat foods as there is no preventative vaccine.

A few pet cats in the USA belonging to a families in which people were infected with the triple reassortment H1N1 virus have become naturally infected with the virus. For at least some of these cats, close contact with the infected owner was reported. The infected cats have developed respiratory signs which have ranged from a relatively mild, self limiting infection to a fatal infection. The role, if any, of cats in spread of the North American triple reassortment influenza remains to be determined.

Virulent systemic calicivirus?

Typical feline calicivirus (FCV) infection is a very common cause of upper respiratory infection in cats throughout the world. Like many other RNA viruses, FCV is prone to mutation and rapid change. There are numerous strains of FCV resulting in variable disease manifestations and varied antigenicity. The typical oral/respiratory disease is most severe in unvaccinated kittens and results in self-limiting oral ulceration, nasal ± ocular discharge, and sometimes sneezing. Virulent systemic feline calicivirus (VS-FCV) results from hyper-virulent viral mutants. Unlike typical FCV, these virulent strains seem to cause more severe disease in adult (and often in FCV vaccinated) cats than kittens. VS-FCV results in a disease presentation quite distinct from that of typical FCV infection. Infected cats develop high fevers; swelling (edema) of the face and limbs; alopecia, crusting, and ulceration of the skin (especially the face, ears, and feet); and death. Mortality rates approach 50% even with supportive care. The virus is shed through feces and sloughed skin/hair as well as nasal, oral, and ocular secretions. Mildly affected cats (often kittens) can pass a virulent and potentially fatal form of VS-FCV, so all exposed cats must be considered potentially contagious. The virus itself is readily spread by fomites, and may be carried by veterinary personnel to other cats in the facility or even to their own pets at home. Facilities with documented cases of VS-FCV may need to temporarily shut down feline admissions to stop the spread of infection. Fortunately, while "ordinary" URI due to FCV are common, this severe disease is very rare.

Confirmation of VS-FCV depends on isolation of identical viral strains (usually from oropharyngeal swabs or necropsy specimens) from more than one affected cat. No test can differentiate VS-FCV from typical FCV in a single cat. Additionally, cats may harbor more than one pathogen simultaneously. These multiple infections may act synergistically to worsen disease manifestation, or one or more pathogens may simply be present without contributing to clinical signs. Currently, only supportive care is available to treat infected cats. A recent publication describes the use of virus specific phosphorodiamidate morpholino oligomers (PMO) to block RNA viral replication. This treatment is still experimental, although intriguing.

Prevention of feline URI is dependent on both management practices to minimize exposure and vaccination. Minimization of crowding and stress, cleanliness, and routine disinfection are all crucial when cats are housed in groups, such as in catteries or shelters. A thorough discussion of shelter design and management is beyond the scope of this talk. A valuable resource with more detail on this topic can be found at the University of California – Davis Koret Shelter Medicine Program website. As an undeveloped virus, FCV is resistant to many routine disinfectants but is susceptible to a 5% bleach solution diluted 1:32 (1/2 cup per gallon) or potassium peroxymonosulfate. It should be noted that there are no disinfectants which work in the presence of organic debris, so basic cleaning must precede disinfection. Recently, a vaccination for virulent systemic FCV has become available (CaliciVax®, Fort Dodge Animal Health). This killed virus vaccine incorporates one of several strains of FCV known to cause severe systemic disease. Generally, hypervirulent strains of FCV have arisen from new genetic mutations in each group of cats infected. Although the new vaccine demonstrated protection from challenge with the same virulent strain used in its development, to our knowledge challenge has not be attempted (and therefore protection has not been demonstrated) with any other virulent strain.

Suggested readings

Barr MC. Update: Avian Influenza: An Emerging Feline Threat? Vet Med. 100(10):727. 2005.

Crawford PC, et al. Transmission of equine influenza virus to dogs. Science 2005; 310:482-485.

Deshpande MS, et al. Evaluation of the efficacy of a canine influenza virus (H3N8) vaccine in dogs following experimental challenge. Vet Ther. 10(3), 103-12, 2009.

Hurley KE, et al. An outbreak of virulent systemic feline calicivirus disease. J Am Vet Med Assoc 2004;224:241-249.

Karaca,K, et al. Evaluation of the ability of canarypox-vectored equine influenza virus vaccines to induce humoral immune responses against canine influenza viruses in dogs. Am J Vet Res. 68(2):208, 2007.

Payungporn S, et al. Influcnza A virus (H3N8) in dogs with respiratory disease, Florida. Emerg Infect Dis. 14(6):902-8, 2008.

Radford AD, et al. Feline calicivirus. Vet Res. 38(2):319. 2007.

Rimmelzwaan GF, et al. Influenza A virus (H5N1) infection in cats causes systemic disease with potential novel routes of virus spread within and between hosts, Am J Pathol. 168:176. 2006.

Smith AW, et al. Virus-specific antiviral treatment for controlling severe and fatal outbreaks of feline calicivirus infection. Am J Vet Res. 69(1):23, 2008.

Song D, et al. Experimental infection of dogs with avian-origin canine influenza A virus (H3N2). Emerg Infect Dis. 15(1):56-8, 2009.

Thiry E, et al. Highly pathogenic avian influenza H5N1 virus in cats and other carnivores. Vet Microbiol. 122(1-2):25. 2007

Yoon K-J, et al. Influenza virus in racing greyhounds. Emerg Infect Dis 2005;11:1974-1976.