Michael R. Moyer
Rosenthal Director, Shelter Animal Medicine, University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA; and Bridgewater Veterinary Hospital, Inc., Bensalem, PA, USA
Influenza Virus Overview
Influenza viruses belong to the Orthomyxoviridae ("straight mucus"), and are further subdivided into Types A, B, and C. Canine Influenza Virus is a Type A virus, as are Avian, Swine, Equine, and seasonal influenza virus of people. There is a non-seasonal influenza of people in the Type B category, and an uncommon form of human and swine influenza in the Type C category. Migratory waterfowl are considered the natural hosts of the influenza virus, and in those birds this virus is found in the gastrointestinal epithelium. It is so well adapted to its host that low-path strains of influenza virus causes no overt symptoms of disease (high path avian influenza strains can cause a wide range of clinical pathology, including GI, cardiac, pulmonary, CNS, peripheral nervous tissue damage in water fowl). The migratory patterns of waterfowl assure that this virus has a worldwide distribution, and because of the ease with which viral particles in waterfowl feces can be distributed, it is no marvel that influenza viruses spill into other species.
In mammals, influenza is a respiratory disease, with inoculation of the respiratory epithelium (usually via nasal or tracheal epithelial contact) causing binding of the HA to the epithelia cells' surface glycoproteins. Ratios of particular glycoside residues are species dependent, and alter the binding affinity of the HA. Mutations in the HA regions responsible for binding to epithelial cell glycoproteins may alter the ability of a viral particle to bind to a new host species with a different ratio of expressed surface sialic acid residues. Binding is followed by a rapid influx of proton (H+) via the M2 ion channel in the viral capsule; an acid environment is essential for viral replication, as the RNA polymerases are effective only in an acid intracellular environment. In mammals, the entire respiratory tract epithelium can be infected, from the nasal passages all the way down to the terminal alveoli, and influenza virus can replicate within alveolar macrophages, too.
Canine Influenza Virus
Canine Influenza Virus (CIV) was first isolated from an outbreak of respiratory illness in a greyhound kennel, where 8 of 22 affected dogs died from severe acute pulmonary hemorrhage. One of these dogs yielded a sample from lung tissue, which was positive for influenza virus. Previous field and experimental work had established that dogs could be infected with avian influenza viruses, but in none of those instances were dogs proven capable of infecting other dogs - horizontal transmission. In the 2004 greyhound outbreak, it was not immediately clear whether this was a spillover infection of equine H3N8 into greyhounds, or whether this particular influenza had established itself in a new host and was now capable of horizontal transmission. Further examination of banked sera from other respiratory outbreaks in greyhounds demonstrated presence of CIV as early as 2000 in greyhounds. Since 2004, this disease has been demonstrated in 35 states and is consider enzootic in Florida, New York, New Jersey, Pennsylvania, and Colorado. Retrospectively, an outbreak of respiratory disease in English foxhounds in 2002 was demonstrated to be an H3N8 equine influenza virus, though horizontal transmission did not appear to be established in the dogs in that event. Additionally, equine and canine sialic acid linkages were compared in that retrospective work, demonstrating that dogs and horses share preferential expression of SA alpha2,3 linkages, which may explain why equine influenza may have spread to dogs on multiple occasions.
While CIV is the focus of much current interest, it also illustrates the importance of continued vigilance and diagnostics for other contagious respiratory pathogens of dogs. Canine Infectious Respiratory Disease Complex is a constellation of bacterial and viral pathogens, which can cause a range of overlapping or identical signs in dogs, from subclinical or inapparent signs to severe hemorrhagic pneumonia. It is beyond the scope of this lecture to describe these in any detail, other than to mention them by way of a listing:
Viral pathogens: adenovirus, herpes, respiratory corona, distemper, parainfluenza, and canine influenza
Bacterial pathogens: Bordetella, Streptococcus zooepidemicus, Mycoplasma spp.
CIV Virology, Pathology
CIV attaches to the respiratory epithelial cell; the M2 ion channel allows protons to enter the endosome. The acidification of the cell is essential for RNA polymerase to copy the eight-strand negative-sense RNA material, hijacking the host cell's ribosomal machinery. Significant virulence enhancement for influenza viruses are noted with a variety of mutations in the HA, NA, and polymerase subunit PB1 genes. These mutations can occur by a variety of mechanisms, substitution mutations, recombination, and reassortment; it is this high degree of mutability that confers influenza's evasion of host immunity though antigenic drift and shift, and allows introduction to non-host species with the possibility of neo-host adaptation (as is the case for CIV host-adapting from equine influenza virtually intact). Assembled virons are released from the epithelial cell via a neuraminidase-facilitated process. In the case of CIV, the log viral shedding is relatively small as compared to other species, and may reflect incomplete host-adaptation on the part of the virus (E. Dubovi, personal communication). Nevertheless, because the virus is in all nasal secretions, clinically affected dogs are remarkably infective in group housing situations where much clinical work with this virus has been conducted. The incubation period can be as short as 24 hours, with a peak of viral shedding at approximately four days post inoculation. Shedding is essentially finished by 7 days post infection.
Replication of the virus in the respiratory epithelium extends into the alveoli and alveolar macrophages, causing extensive destruction and inflammation. Incapacitation of the respiratory ciliary clearance mechanism is compounded by lobar consolidation/hepatization within the lungs. The striking appearance of pulmonary lesions are identical to what is seen in porcine influenza, and explains how the cough of dogs with CIV is often more severe and of a longer duration than other agents of CIRDC. Accumulations of mucus, necrotic epithelial cells, fluids in concert with native or opportunistic flora will support secondary bacterial infections in the upper or lower airway or both.
Current Epidemiology of CIV
While CIV has been confirmed in 34 states as of June 2010, our appreciation for the epidemiology for this pathogen is still very incomplete, owing to a variety of factors (lack of awareness among many clinicians of the diagnostic approach to confirm CIV, no reliable surveillance data pooling multiple laboratory data, the need for either kennel owners, shelter management, or clients to pay for the costs of testing clinically affected dogs limits the number of samples submitted, and no ability to model canine social dynamic behavior to allow the use of existing computer models to approximate the effects of vaccination, social distancing, or changes in virulence on the dog population).
CIV Modes of Transmission
Because CIV replicates in the respiratory epithelium, viral particles are easily shed in all respiratory and oral secretions. Spread is possible by direct contact, fomites (including personnel), and aerosol. The distance limits for aerosol spread are not fully known, but may be as much as 50 feet. Coughing and barking dogs may generate sufficient aerosol for transmission as a part of their pathology, or cleaning techniques (hosing, spraying) may produce aerosol particles. Because of the typical methods of dog handling in mass kennel facilities, many opportunities exist for fomite transmission. Contaminated surfaces, common leashes, muzzles, other restraint equipment may be implicated in spread of this (and other) diseases.
Dogs at Risk for CIV Infection
While initially described in Greyhounds, it appears that CIV infects all breeds and mixes of dogs without discrimination. There is no gender or age bias. The acute hemorrhagic form has only been described in the Greyhound, although several outbreaks of Streptococcus zooepidemicus equi in shelters have occurred with super-infections of CIV; clinical signs of S. zooepidemicus are very similar to the hemorrhagic form of CIV in the Greyhound. Because CIV is still novel to much of the dog population, nearly all dogs are susceptible, and morbidity in several outbreaks approaches 100%.
The respiratory signs of CIV include conjunctivitis, rhinitis, sneezing, spontaneous cough, and dyspnea. Depression and fever have been observed. Anecdotally, kennel attendants, veterinary technicians, and veterinarians familiar with CIV describe the cough of as much more severe, anti-tussive resistant, and persistent than other forms of CIRDC. Signs of CIV lag begin at approximately 3 days post inoculation, peak at approximately 10 days post inoculation, and diminish by approximately 14 days post inoculation.
Infected dogs may develop a range of signs and severity of disease, with about 20% of the infected dogs showing no outwards signs, but still may be capable of shedding virus. The majority of dogs demonstrate moderate symptoms, but would be easily managed as outpatients in a home setting, or can be isolated and symptomatically treated in a kennel/hospital setting, expecting a full recovery. A smaller group - perhaps 6 to 10 % - develop severe disease that required hospitalization for pneumonia and respiratory distress. Some of these patients will require oxygen supplementation, and a minority will be candidates for ventilatory support. A very small percentage will die from CIV or complications thereof.
Viral diagnostic techniques include PCR, viral isolation, and serologic assays, and point-of-care influenza A/B test kits. PCR and viral isolation samples can be obtained by swabs of the conjunctival sac and nares, deep pharyngeal swabs, sterile endotracheal intubation samples, trans-tracheal wash, and bronchio-alveolar lavage. Sterile sampling of lung tissue at necropsy is also a good source of viral material in the case of euthanized dogs or dogs that succumb to their disease. Yield for PCR and viral isolation for CIV has been highest with nasal swabs. Infected dogs with a competent immune system will produce rising IgG levels in 7 to 15 days post inoculation, with paired titers providing the clearest illustration of an acute infection. Point of care influenza test kits calibrated for human patients have been used in shelter outbreak investigations; a positive result would have a high predictive value, but because the kits are designed for a much higher viral threshold (influenza of people sheds at a much higher log dose then CIV in dogs), a negative result is not conclusive. Because the kits can deliver a result "patient side", there are nevertheless advantages to using them in an outbreak investigation, where a rapid detection would guide the management of a population of exposed dogs.
Physical examination, bacterial culture, and thoracic radiography remain important in all cases of CIRDC, and will capture the other causes of respiratory disease in dogs, as well as document pulmonary or bronchiolar changes. Several veterinary reference laboratories provide canine respiratory pathogen panels that cover a wide spectrum of CIRDC agents.
The short interval of viral shedding, combined with a lag in symptomatic signs, make viral detection extremely time-sensitive. Virus is virtually cleared by day 7 post inoculation, such that many symptomatic dogs (day 3 or 4) in a clinical setting may not be positive on swab samples with either PCR or viral isolation by the time they are examined. Dogs in this phase may not have mounted a sufficient IgG level to be detectable via serology, and may thus lie in a diagnostic window of uncertainty. A convalescent serum sample 7 to 10 days later will resolve the uncertainty; if positive, the dog had CIV, if negative, it did not. Because of this diagnostic window of uncertainty, symptomatic dogs with negative PCR/viral isolation cannot be considered CIV negative until confirmatory paired or convalescent serology is performed. This has significance for outbreak management in shelters, commercial kennels, boarding and dog daycare/training facilities.
Clinically affected dogs with CIV are treated with general supportive care, and those suspected or confirmed to have bacterial infections can be given antibiotics. A variety of cough suppressants have been used in cases of CIV, though the use of anti-tussives in cases of alveolar damage is thought to be counterproductive (clearance of mucoid debris is already compromised, it is thought that further suppression of cough may worsen lung damage). The cough of CIV is quite resistant to suppression, and may persist for up to 3 or 4 weeks. Various antivirals for influenza have been discussed, but none have been tested in the dog for efficacy with CIV, and their use cannot be recommended. The majority of patients remain systemically well, will eat and continue to function without special care, and can be managed on an outpatient basis. A small percentage will require more intensive care
CIV Prevention: Biosecurity
Effective control of CIV, or any other bacterial or viral disease for that matter, begins with fundamental biosecurity protocols. For veterinary hospitals, shelters, and commercial kennels, this begins with staff training for fomite minimization, hand sanitization, and proper cleaning and disinfection procedures. There have been cases where CIV in a facility was transmitted via fomite to pet dogs belonging to kennel staff or veterinary staff. CIV is an enveloped virus, and succumbs readily to disinfectants used properly, but the virus can persist in mucus accretions. "Resident" dogs or pet dogs allowed to co-mingle with boarding, hospitalized, or sheltered dogs are at risk of contracting and spreading infection. In the hospital setting, receptionists ought to schedule respiratory patients for time slots that minimize the chance for spread in the waiting room, or respiratory patients may need to be examined in a different space or even be examined in the parking lot (if not severely ill) to decrease the chance of spread through a hospital kennel.
Admitted respiratory patients require the highest level of isolation hygiene, use of personal protective equipment, and appropriate kennel space to contain the spread to other dogs. A separate respiratory isolation is often preferred to a combined isolation ward. If treating CIV patients, surveillance of other dogs within the facility is important, so that suspected spread can be identified.
Suspected CIRDC cases should be immediately isolated from the general population and screened for CIV (and other CIRDC agents). The isolation period should be approximately 10–14 days.
Dogs in the apparently general population should be sorted into "exposed" and "not likely exposed". Those that were exposed should be quarantined for 7–10 days - any dogs breaking with symptoms should be moved to isolation and the "quarantine clock" reset to day 0. When the last symptomatic dog has been out of the quarantine group for 10 days, it is reasonable to consider the quarantine period complete. Isolated dogs can be considered recovered 14 days after the addition of the last symptomatic dog in the outbreak. Serology of the isolation group will confirm or rule out CIV; dogs with positive titers are very unlikely to be shedding CIV. Negative titers in asymptomatic dogs, if there have been no new cases introduced into isolation for 10 days, were probably not infected.
Ideally, dogs should not leave the facility except to go to homes without dogs or with dogs already vaccinated against CIV. Ideally, intake of dogs would be diverted to another facility, ward, or kennel. At the minimum, intake should go in to the "not exposed" cohort.
Since July 2009, a killed CIV vaccine for dogs has been available in the USA. This product requires a priming dose and a booster at 3 weeks, producing a non-sterile immunity that decreases the period of viral shedding and the number of viral particles, and in the experimental work has shown significant reduction in the clinical signs of disease. The vaccine may be useful for certain shelter facilities, commercial kennels, boarding facilities and dog daycare/training centers. It is highly recommended for the dogs belonging to kennel staff/vets/foster/rescue operatives because of the risk of fomite transmission.
1. Nikitin T, Cohen D, Todd JD, et al. Epidemiological studies of A/Hong Kong/68 virus infection in dogs. B World Health Organ 1972;47:471–479.
2. Kilbourne ED, Kehoe JM. Demonstration of antibodies to both hemagglutinin and neuraminidase antigens of H3H2 influenza A virus in domestic dogs. Intervirology 1975/76;6:315–318.
3. Songserm T, Amonsin A, Jam-on R, et al. Fatal avian influenza A H5N1 in a dog. Emerg Infect Dis 2006;12:1744–1746.
4. Amonsin A, Songserm T, Chutinimitkul S, et al. Genetic analysis of influenza Avirus (H5N1) derived from domestic cat and dog in Thailand. Arch Virol 2007;152:1925–33.
5. Maas R, Tacken M, Ruuls L, et al. Avian influenza (H5N1) susceptibility and receptors in dogs. Emerg Infect Dis 2007;13:1219–1221.
6. Zini E, Glaus TM, Bussadori C, et al. Evaluation of the presence of selected viral and bacterial nucleic acids in pericardial samples from dogs with or without idiopathic pericardial effusion. Vet J 2009;179(2):225–229.
7. Giese M, Harder TC, Teifke JP, et al. Experimental infection and natural contact exposure of dogs with avian influenza virus (H5N1). Emerg Infect Dis 2008;14:308–310.
8. Crawford PC, Dubovi EJ, Castleman WL, et al. Transmission of equine influenza virus to dogs. Science 2005;310:482–485.
9. Yoon K-J, Cooper VL, Schwartz KJ, et al. Influenza virus infection in racing greyhounds. Emerg Infect Dis 2005;11:1974–1975.
10. Payungporn S, Crawford PC, Kouo TS, et al. Influenza A virus (H3N8) in dogs with respiratory disease, Florida. Emerg Infect Dis 2008;14(6):902-908.
11. Available at: http://diagcenter.vet.cornell.edu/. Accessed 2005.
12. Daly JM, Blunden AS, MacRae S, et al. Transmission of equine influenza virus to English foxhounds. Emerg Infect Dis 2008;14:461–464.
13. Newton R, Cooke A, Elton D, et al. Canine influenza virus: cross-species transmission from horses. Vet Rec 2007;161:142–143.
14. Rosas C, Van de Walle GR, Metzger SM, et al. Evaluation of a vectored equine herpesvirus type 1 (EHV-1) vaccine expressing the H3 haemagglutinin in the protection of dogs against canine influenza. Vaccine 2008;26:2335–2343.
15. Karaca K, Dubovi EJ, Siger L, 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 2007;68:208–212.
16. www.sciencedirect.com/science/article/pii/S0898656800000668 (VIN Editor: link updated Aug 2011)