Abstract
Although cats had been considered resistant to disease from influenza virus infection, domestic cats and large felids are now known to be naturally und experimentally susceptible to infection with highly pathogenic avian influenza virus H5N1 (HPAIV H5N1). The virus causes systemic infection, lung and liver being the mainly affected organs. Infected cats show fever, depression, dyspnoea, and neurological signs, but subclinical infections have also occurred. Mostly, cats have been infected by direct contact with affected birds, especially by eating raw poultry; transmission from cat to cat may also occur. Little is known about the role of cats in the epidemiology of the virus. So far, no reassortment between avian and mammalian influenza viruses has occurred in cats, but experts fear that cats might give the virus an opportunity to adapt to mammals. This publication gives a review on avian influenza in cats with a focus on practical aspects for veterinarians.
Besides human infections, most known mammalian infections with highly pathogenic avian influenza virus H5N1 (HPAIV H5N1) have occurred in felids. Large felids and domestic cats can not only be infected by direct or indirect contact with infected birds (Keawcharoen et al 2004, Songserm et al 2006a, Leschnik et al 2007); the virus can also be transmitted horizontally from cat to cat (Kuiken et al 2004, Thanawongnuwech et al 2005). However, there are still a lot of open questions concerning the epidemiology and pathophysiology of this viral infection in cats. The close relationship between cats and humans is a cause for concern about the cat's role in the spread of H5N1 (Kuiken et al 2006). Even though most feline cases have occurred in Southeast Asia, infected cats were also found in Central Europe (Germany and Austria) (Leschnik et al 2007, Klopfleisch et al 2007a).
This article gives a review of the current literature to help veterinarians when confronted with a cat suspected of being infected with HPAIV H5N1, and to help answering cat owners' questions that arise in areas where infected birds have been found.
Aetiology
Influenza viruses are negative sense, single-stranded, segmented RNA viruses belonging to the family Orthomyxoviridae. Whereas influenza viruses types B and C are mainly human pathogens, influenza A viruses act as pathogens in many mammalian species including humans and in birds (Webster et al 1992). Influenza A viruses are classified into distinct subtypes according to different haemagglutinin and neuraminidase glycoprotein molecules expressed on the surface (Fouchier et al 2005). In avian influenza viruses, all different subtypes described until today are found. They can either lead to subclinical infection or cause serious systemic disease, depending on their pathogenicity. Among the subtypes H5 and H7, highly pathogenic variants may develop out of low pathogenic avian influenza viruses by mutation (Alexander 2000).
HPAIV subtype H5N1 was first detected in 1996 in domestic geese in China (Xu et al 1999, Li et al 2004). After several reassortment events, this avian virus not only caused serious disease in poultry, but also crossed the species barrier infecting people in Hong Kong in 1997 (Subbarao et al 1998, Webster et al 2002, Li et al 2004). During the subsequent years, different H5N1 genotypes emerged after a series of genetic changes, leading to fatal outbreaks in Asia in poultry in 2003/2004 (Li et al 2004, Chen et al 2006). Since then, HPAIV H5N1 has spread to many countries worldwide (World Health Organization 2007) resulting in high mortality in poultry and fatal infections in mammalian species, including humans (Vahlenkamp and Harder 2006).
Mammalian species known to be susceptible to HPAIV H5N1 are humans (Claas et al 1998, Tran et al 2004), ferrets (Zitzow et al 2002, Govorkova et al 2005), dogs (Songserm et al 2006b, Giese et al 2008), mice (Gao et al 1999), stone martens (Klopfleisch et al 2007b), pigs (Choi et al 2005), cynomolgus monkeys (Rimmelzwaan et al 2001, Kuiken et al 2003), civets (Roberton et al 2006), domestic cats (Kuiken et al 2004, Thiry et al 2007), tigers, and leopards (Keawcharoen et al 2004).
Infections in felids
In felids, several outbreaks of infection with HPAIV H5N1 have been reported so far. The first outbreak was noted in 2003, when two tigers and two leopards suffering from high fever and respiratory distress died in a zoo in Suphanburi, Thailand (Keawcharoen et al 2004). Further evidence that felids are susceptible to influenza A H5N1 infection arose when in 2004 three domestic cats from a household in Thailand, where 14 cats had died, were tested positive for influenza A H5N1 and when a clouded leopard died in a zoo in Chonburi, Thailand, from infection with influenza A H5N1 (Enserink and Kaiser 2004, ProMED-mail 2004a,b). One month later, a tiger at the same zoo was found to be infected, but recovered from the disease (Enserink and Kaiser 2004, ProMED-mail 2004b). During an outbreak in a tiger zoo in Sriracha, Thailand, a total of 147 tigers died or were euthanased (Thanawongnuwech et al 2005). Furthermore, the virus was detected in a domestic cat in Thailand that had died showing high fever, dyspnoea, convulsions, and ataxia (Songserm et al 2006a). Experimental infections with H5N1 virus isolated from a fatal human case confirmed that cats can develop severe clinical signs after intratracheal inoculation or after feeding on infected chicken (Kuiken et al 2004). These findings are remarkable, as clinical disease resulting from infection with influenza viruses had not been noticed in cats before (Paniker and Nair 1970, 1972, Hinshaw et al 1981).
The first cases of HPAIV H5N1 infection in domestic cats in Europe were detected during the outbreak of avian influenza on the German Isle of Rügen in February 2006, where three free-roaming cats were found dead harbouring the virus (Klopfleisch et al 2007a). At approximately the same time, three cats that did not show clinical signs of influenza tested positive for influenza A H5N1 in an animal shelter in Graz, Austria, after an infected swan had been brought to the shelter (Leschnik et al 2007).
Epidemiology
The incidence of avian influenza in felids seems to be associated with the occurrence of infections in poultry or wild birds in the surrounding area (Keawcharoen et al 2004, Leschnik et al 2007, Klopfleisch et al 2007a). Phylogenetic analyses have shown that the examined virus isolates from cats and tigers were highly similar to the virus circulating in poultry at the same time and that the viruses found in felids were of avian origin, indicating that no genetic reassortment with mammalian influenza viruses had occurred. Several point mutations have been identified that are associated with higher virulence in mammals, but none of them seems to be essential for an infection in felids (Keawcharoen et al 2004, Amonsin et al 2006, 2007, Weber et al 2007).
Within the H5N1 subtype, at least two genetically and antigenically distinct lineages (clades 1 and 2) exist in non-overlapping geographic distributions in Asia. Clade 1 was mainly isolated in Vietnam and Thailand, whereas clade 2 was mainly found in China and Indonesia (World Health Organization Global Influenza Program Surveillance Network 2005). From there, the ‘Qinghai-like’ sublineage spread westwards to the Middle East, Europe, and Africa and split up into three different subclusters (Yingst et al 2006, Salzberg et al 2007, Weber et al 2007). All Asian cases reported in felids were caused by infections with clade 1 viruses, until in February 2006, an outbreak in domestic cats in Iraq and the cases in Germany were attributed to Qinghai-like clade 2 viruses (‘EMA’ clade 2), demonstrating that cats may be susceptible to different circulating H5N1 viruses (Yingst et al 2006, Weber et al 2007).
Virus transmission
Virus transmission has mostly been ascribed to direct contact of felids with infected birds, particularly through eating of raw poultry (Keawcharoen et al 2004, Kuiken et al 2004, Songserm et al 2006a). Both inhalation and ingestion seem to be possible routes of virus entry there (Rimmelzwaan et al 2006, Yingst et al 2006). In addition, indirect viral transmission to cats may occur after contact with contaminated birds' faeces, as was suspected in the cases in Graz (Leschnik et al 2007). Kuiken et al (2004) showed that horizontal transmission from experimentally inoculated cats to other cats is possible by direct contact. Most likely, horizontal transmission also occurred under natural circumstances in the outbreak in Sriracha tiger zoo (Thanawongnuwech et al 2005).
So far, no case of virus transmission from cats to other species including humans has been observed. Antibody development without development of clinical signs, however, was found in two of 58 people who had been in contact with infected tigers (Thanawongnuwech et al 2005).
Infected felids were found to excrete virus via respiratory, digestive, and urinary tract, as shown by virus detection from pharyngeal, nasal, and rectal swabs as well as urine and faecal samples (Rimmelzwaan et al 2006, Yingst et al 2006, Songserm et al 2006a, Klopfleisch et al 2007a). Virus shedding may occur before the onset of clinical signs (Kuiken et al 2006, Rimmelzwaan et al 2006). In an experimental study, virus excretion started at day 3 after infection and lasted until day 7 when the animals were euthanased (Kuiken et al 2004, Rimmelzwaan et al 2006). Subclinically infected cats are assumed to excrete virus less than 2 weeks (Leschnik et al 2007).
Pathogenesis
After transmission, the virus spreads locally to the lower respiratory tract and can cause severe pneumonia (Keawcharoen et al 2004, Songserm et al 2006a, Klopfleisch et al 2007a). Predominant involvement of the lower respiratory tract and inability of the virus to attach to cells of the upper respiratory tract may be a reason why cats excrete virus at relatively low concentrations (Kuiken et al 2004, Van Riel et al 2006).
Unlike other influenza viruses, which are usually restricted to the respiratory tract in mammals, HPAIV H5N1 not only replicates in respiratory tissue, but can also lead to systemic infection causing severe necrosis and inflammation in many organs (Rimmelzwaan et al 2006). Two ways of virus spread to extra-respiratory tissue are presumed. The pattern of virus distribution in the body provides evidence of virus entry via viraemia. Another route may be virus entry from the intestinal lumen via nerve fibres into intestinal tissue, which is indicated by the finding of ganglioneuritis of the intestinal nervous plexi in cats that had been fed on virus-infected chicks (Rimmelzwaan et al 2006).
Clinical signs
The incubation period is known to be shorter after direct experimental infection by respiratory and oral routes (1–2 days) than after cat-to-cat transmission (5 days) (Kuiken et al 2006, Rimmelzwaan et al 2006). Clinical signs observed in affected felids include pyrexia, depression, laboured breathing, conjunctivitis, protrusion of the third eyelid, and neurological signs such as convulsions and ataxia (Keawcharoen et al 2004, Kuiken et al 2004, Thanawongnuwech et al 2005, Songserm et al 2006a). Respiratory signs are caused by severe pulmonary changes (consolidation, haemorrhage, oedema) and pleural effusion, visible at necropsy (Keawcharoen et al 2004, Thanawongnuwech et al 2005, Songserm et al 2006a). Histopathology reveals extensive inflammation and necrosis in lung tissue leading to interstitial pneumonia and diffuse alveolar damage (Rimmelzwaan et al 2006, Songserm et al 2006a). Neurological signs result from cerebral and cerebellar congestion and non-suppurative meningoencephalitis accompanied by vasculitis (Thanawongnuwech et al 2005, Songserm et al 2006a). Diarrhoea, described in affected poultry (Perkins and Swayne 2001) and also in humans infected with HPAIV H5N1 (Apisarnthanarak et al 2004, Tran et al 2004), has not been observed in cats.
Laboratory abnormalities in tigers included severe leukopenia and thrombocytopenia and increased activities of the liver enzymes alanine aminotransferase and aspartate aminotransferase (Thanawongnuwech et al 2005). Markedly increased liver enzyme activities have also been found in aqueous humour samples from cats taken during post-mortem examination (Klopfleisch et al 2007a). Histopathology of the liver shows multifocal necrotising hepatitis, explaining the increase in liver enzyme activities and the generalised icterus which can also be seen at necropsy (Thanawongnuwech et al 2005, Rimmelzwaan et al 2006, Klopfleisch et al 2007a). Serosanguinous nasal discharge observed in severely affected tigers may have been caused by severe thrombocytopenia (Thanawongnuwech et al 2005). Multifocal haemorrhage has been described in numerous organs such as lungs, heart, thymus, stomach, intestine, liver, tonsils, lymph nodes, kidneys, and diaphragm, as well as pancreas (Keawcharoen et al 2004, Rimmelzwaan et al 2006, Yingst et al 2006, Klopfleisch et al 2007a). Sudden death may occur as soon as 2 days after onset of clinical signs (Songserm et al 2006a).
Infection with HPAIV H5N1 can also result in subclinical infection. In Graz, three cats excreted virus after contact with an infected swan and two cats developed antibodies to influenza A H5N1 virus, but none of them showed signs of influenza (Leschnik et al 2007). Anecdotal reports also support the existence of subclinical infections. In an unpublished study carried out by the National Institute of Animal Health in Bangkok, eight of 111 cats (7%) were found to carry antibodies to influenza A H5N1 (Butler 2006). An unpublished study by Nidom suggested that 20% of 500 cats tested in Indonesia had antibodies to influenza A H5N1 (Mackenzie 2007).
Diagnosis
Virus detection is possible from pharyngeal, nasal, and rectal swabs, from faecal and urine samples, from organ tissue, and pleural fluid (Rimmelzwaan et al 2006, Yingst et al 2006, Songserm et al 2006a). In subclinically infected cats, H5N1 has only yet been detected in pharyngeal swabs (Leschnik et al 2007). H5N1 virus RNA can be identified by real-time reverse transcriptase polymerase chain reaction (RRT-PCR) using primers specific for the haemagglutinin and neuraminidase genes (Keawcharoen et al 2004, Klopfleisch et al 2007a). Virus can also be isolated by inoculation into embryonated chicken eggs (Songserm et al 2006a) or cell cultures (Rimmelzwaan et al 2006) and subsequently be identified by RRT-PCR or haemagglutination and haemagglutination inhibition assays (Songserm et al 2006a). Immunohistochemistry can be used for H5N1 virus antigen detection in affected organs (Rimmelzwaan et al 2006, Songserm et al 2006a). Antibodies to influenza A H5N1 in serum samples can be detected by an haemagglutination inhibition test (Karaca et al 2005, Leschnik et al 2007).
Treatment and prevention
Not much is known about the efficacy of antiviral treatment in cats infected with HPAIV H5N1. Although the neuraminidase inhibitor oseltamivir has shown potent antiviral activity against HPAIV H5N1 in vitro (Hurt et al 2007) as well as in experimentally infected mice and ferrets (Leneva et al 2000, Govorkova et al 2007) and is recommended for treatment and prophylaxis of HPAIV H5N1 infection in people (Schunemann et al 2007), this treatment was unsuccessful in tigers during the outbreak in Sriracha tiger zoo in 2004. Oseltamivir (Tamiflu®; Roche) was administered to the tigers at a dose of 75 mg/60 kg twice daily (human dosage) for treatment and prophylaxis, but failed in symptomatic as well as asymptomatic animals. The treatment failure may have resulted from improper dosage or timing of drug administration; differences in pharmacokinetics and host metabolisms between humans and felids and even between large felids and domestic cats can be expected (Thanawongnuwech et al 2005, Amonsin et al 2006).
So far, there is no licensed influenza vaccine for cats on the market. However, there is some research in this regard. After experimental vaccination with fowlpox virus expressing avian influenza virus H5 haemagglutinin gene derived from an H5N8 influenza virus, cats developed high levels of antibodies to the homologous H5N8 antigen. After administration of a second dose, antibodies were shown to cross-react with a recent HPAIV H5N1 isolate (Karaca et al 2005). In another study, an anti-H5N1 antibody response was induced by injection of canine adenovirus expressing H5 haemagglutinin gene of a tiger isolate in one cat (Gao et al 2006).
As all known infections in cats were connected with the occurrence of HPAIV H5N1 in birds, contact with birds carrying the virus must be avoided to prevent infections in cats. It is, therefore, advised to keep cats indoors in areas with occurrence of highly pathogenic avian influenza in birds (Anonymous 2006, Duke 2006). Furthermore, cats should not be fed uncooked poultry meat (Advisory Board on Cat Diseases 2006).
Little is known about the actual risk for cats in the field to become infected with the virus and their role in the spread of H5N1. The risk of transmission from potentially infected cats to humans is unknown. One study showed, though, that there is no major risk for pet cats in areas with sporadic incidence of avian influenza in birds, as neither virus excretion nor antibodies were detected in 171 cats with outdoor access in areas in which infected birds had been found (Marschall et al 2008).
Preventive measures must still be taken to minimise the risk of transmission when confronted with a cat suspected of being infected with H5N1. Recommendations on the handling of cats suspected to be infected with HPAIV H5N1 are given in Table 1. The cat should be isolated, and contact with the cat should be restricted to a minimum. People in contact with the cat must wear protective clothing; surfaces should be decontaminated with standard medical disinfectant (Thiry et al 2007). Uncooperative cats should be sedated before handling (Advisory Board on Cat Diseases 2006). Further studies are necessary to learn more about the epidemiology and pathophysiology of HPAIV H5N1 infection in cats.
Recommendations on the handling of cats suspected to be infected with HPAIV H5N1
Restriction zone: area within 10 km radius around location of outbreak of avian influenza in birds for the duration of 30 days after discovery of infected birds.
