West Nile encephalitis: A new differential for neurological illness in dogs and cats

The discovery of West Nile virus (WNV) in the Northeast United States in 1999 and its subsequent successful establishment on the North American continent over the last three years has had an impact on human and animal populations nationwide.

West Nile virus continues to spread across the country affecting both people and animals. Although the virus has been predominately found in humans, horses and birds, there are now cases confirmed in companion animals.

While primarily an infection of birds and mosquitoes, the continental United States has seen an ever-widening epizootic involving avians, mammals and even reptiles. Each successive year of active transmission has seen an explosive increase in morbidity and mortality in birds, humans and horses as well as a widening list of other mammalian species identified with WNV-associated illness as the virus gained new territory.

In the initial year of its discovery, WNV encephalitis was identified in a cat by the Centers for Disease Control and Prevention (CDC). During the years 2000 and 2001, WNV infections in squirrels, rabbits and bats were identified as a consequence of public health surveillance efforts. No additional confirmed cases of WNV encephalitis were reported in small domestic companion animals until the summer of 2002.

Host range

WNV is one of the oldest known flaviviruses with an extensive historical geographic distribution in Africa and Asia. Previous experimental and serosurvey data suggested the WNV had an extremely broad host range that included amphibians, mosquitoes, ticks, birds and mammals but caused clinical illness in only a small percentage of avian species and of mammals, mainly humans and horses. This is generally consistent with the extensive surveillance data collected in the Northeast and Southeast United States during the first two years of the expanding epizootic and with more recent experimental infection data in dogs and cats.

Table 1: Reported Clinical Signs in Dogs with Positive IgM and Virus Neutralizing WNV Antibody

A serosurvey was done by the CDC in stray and owned dogs and cats in New York City after the 1999 epizootic to determine the level of exposure of these animals to WNV. Samples of 189 dogs and 12 feline samples were tested for neutralizing antibody by plaque reduction neutralization against WNV and St. Louis encephalitis virus. Five percent of the dogs had neutralizing antibody titers to WNV ranging from 1:80 to >/= 1:320 while none of the cats had neutralizing antibody.

More recent data presented by the CDC at the 2003 WNV Planning Meeting sponsored by the CDC and the American Society of Microbiology in February provided additional support for the conclusion that cats and dogs are susceptible to WNV infection, but that infection does not cause overt clinical signs in infected animals. The experiments also demonstrated that ingestion of infected mice by cats led to productive infection of the cats and suggested that exposure by consumption was an efficient route of infection. Infected animals developed only low virus titers and are considered to be incapable of transmitting virus to mosquitoes or other animals.

Table 2: Testing options for west nile virus

What happened in 2002?

The summer of 2002 was an explosive year for WNV in the United States and Canada. Virus expansion across the country continued at a rapid rate. Morbidity and mortality in birds, humans and horses increased exponentially over 2001 figures, and several new species were identified as having WNV-associated morbidity and mortality. These included llama, sheep, reindeer and a mountain goat. The toll WNV took on wildlife in Central and Southern regions of the country has largely been unquantified, but was substantial.

Additional firsts included a confirmed diagnosis of WNV encephalitis/myocarditis was made in an 8-year-old Golden Retriever and a wolf cub was confirmed to have died of WNV infection in Illinois. The Animal Health Diagnostic Laboratory at Cornell began identifying WNV neutralizing antibody positive serum samples from dogs and cats in the Midwest, Central west and Southeast in mid to late summer with clinical histories of an acute onset neurological illness. Of the 461 dogs tested during 2002, 60 had high serum neutralizing antibody titers to WNV. One dog showed a >/=four-fold rise in titer on paired serum samples and one 4-month-old puppy had a high SN antibody titer in serum and CSF.

A good correlation was also established between IgM positive dogs tested at the University of Nebraska and positive serum neutralizing titers obtained from the same samples at the Cornell lab (Galeota and Glaser, unpublished data). We have follow-up information for about 30 percent of the WNV positive dogs tested at Cornell for which histories were provided, all of which recovered. Samples were received from 45 cats, including a cougar. Four of the cats, including the cougar were SN antibody positive for WNV with high titers demonstrated. The cougar and one domestic cat died with no further diagnostic testing performed.

Clinical signs

Clinical signs in WNV antibody positive dogs reported on forms or during telephone interviews with veterinarians included weakness, ataxia, flaccid paralysis, decreased or absent patellar reflexes, inability to stand, muscle tremors, depression, hyperesthesia, seizures and fever. Most dogs were reported to be bright, alert and able to eat and drink. Clinical signs in the cats included seizures, ataxia, nystagmus and hyperesthesia. These signs did not differ significantly from signs reported for animals with no detectable WNV SN titers.

The level of the WNV transmission between all species in the Midwest states was exceedingly high as evidenced by the enormous numbers of equine and human cases occurring in those regions. Our experience last year is supportive of a possible role for WNV in neurological illness in dogs, and less definitively supportive of a role for WNV in neurological illness in cats in regions of high epizootic transmission. However, the data is incomplete as IgM capture ELISA for dogs and cats were not been widely available.

Necropsies and additional testing have rarely been pursued due to prohibitive costs. While dogs and cats are not considered at risk for development of clinical illness following WNV infection, it is probable that the infection rate was so high in these regions that a low rate of clinical illness in a percentage of infected animals became visible.

It is likely still true that cats and dogs will not commonly develop WNV-associated clinical illness following infection, but in areas with confirmed active WNV transmission, WNV infection should be included in the differential diagnosis in animals presenting with a sudden onset of neurologic signs which include weakness or ataxia. Collection of more information from laboratory confirmed clinical WNV-associated illness in dogs and cats is required to define this infection in these populations.

Diagnosis

The key to diagnosing WNV-associated illness in companion animals is first to include it on the differential and then to test for it. Because evidence suggests that dogs and cats can be infected with WNV with no clinical signs of illness, multiple testing strategies are required to achieve the highest probability of diagnostic accuracy.

Testing strategies will depend on the samples available. Testing in animals with mild signs of illness will rely heavily on serology. Animals in WNV-endemic areas may be antibody positive so this alone should not be considered proof of causality in an animal with a neurologic illness. A minimum standard of association of WNV with a clinical illness in a companion animal when serology is the only available test should be the presence of IgM antibody with a positive virus neutralizing antibody titer in serum and cerebral spinal fluid (CSF) or demonstration of a four-fold or higher rise in antibody titer in paired serum samples.

Paired serum samples should be submitted routinely and serial samples taken days apart may also be helpful. All antibody capture ELISAs, whether for IgM or IgG should initially be confirmed by serum neutralization, at least until the sensitivity and specificity of the ELISA tests can be determined. EDTA blood can also be submitted for reverse transcription – polymerase chain reaction (RT-PCR) detection of viral nucleic acid and for virus isolation. It is currently unknown how useful this will be for the detection of WNV as titers of virus in blood have generally been described as low and transiently present.

If an animal should die or be euthanized as a result of disease progression, a post mortem CSF sample should be collected and held pending rabies testing.

Ideally, the body should be submitted to a veterinary diagnostic laboratory for a full necropsy, histopathology and ancillary testing. Once rabies is ruled out, fresh tissues samples should be tested for WNV by RT-PCR and virus isolation should be attempted from kidney, brain and heart. Histopathology and immunohistochemical staining should also be pursued whenever possible.

Safety

There are several challenges facing individuals at all levels within the diagnostic process. While it is believed that cats and dogs with WNV infections represent no risk of horizontal transmission to humans or other animals, there is theoretically an increased risk of exposure to virus as a result of the necropsy procedure. Where rabies is a realistic possibility, minimal handling of the carcass is encouraged until a negative rabies result is obtained though routine public health laboratory testing.

Necropsy of an animal with a neurologic illness should always be approached with personal safety issues foremost in mind. Individuals electing to do necropsies in-house are encouraged to wear appropriate personal protective clothing. This would include at a minimum two pairs of impermeable gloves, a disposable gown, face-mask and a face-shield or goggles. Power saws should not be used at any time during the procedure. All surfaces should be thoroughly chemically disinfected. Most commonly used disinfectants are capable of inactivating WNV.

Challenges of responding

Challenges within the veterinary diagnostic and academic communities make rapid response to a virus infection of this nature difficult. Resources sufficient to provide the testing services required are limited. Very few laboratories have the ability to provide all of the testing services needed to obtain a definitive diagnosis. This is due in large part to the requirement for high security biocontainment laboratories (biosafety level 3) to perform virus isolation and serum neutralization testing and to a shortage of reagents required for ELISA-based test development.

In every year since 1999, testing capacity has been maximally challenged in both the human and animal diagnostic communities. More veterinary diagnostic laboratories are acquiring the ability to do some WNV serology, generally IgM and IgG antibody capture ELISAs.

These tests are extremely useful, but should not be used as definitive tests as reactivity detected in these assays can only be considered flavivirus specific.

One of the most challenging tasks of the veterinary community, both academic and clinical, is to thoroughly characterize emerging disease problems within the populations we serve. This includes getting careful histories, recording clinical signs and including this information on submission forms sent to the laboratory so that clinical data can be compiled and shared with the rest of the veterinary community.

This task is made more difficult in the face of economic constraints at both ends of the spectrum as laboratories are facing ever restrictive funding that precludes doing large amounts of surveillance testing for infectious disease for free and as pet owners are feeling the effects of turbulent economic times.

To date, we have a very incomplete picture of WNV-associated illness in companion animals. We currently have very little understanding of the number of animals exposed in a region and how that relates to the number of animals with potential WNV-associated illness. We have very little information about the details of disease progression and clinical outcome. Because clinical illness appears to be rare, it is only with the help of all members of the veterinary community that we will be able to define the clinical syndrome of WNV encephalitis in dogs and cats.

Suggested Reading

Dr. Glaser received her DVM degree from Cornell University in 1987. She completed a Ph.D. in virology in 1995 at the same institution. She has been a senior research associate at the Animal Health Diagnostic Laboratory since 1998 and has developed the WNV testing program at the laboratory. She routinely assists in investigations of infectious disease outbreaks in mammalian and avian populations and is engaged in research to develop rapid testing formats for the identification of viruses.