Okay, I wasn’t going to do this. I promised myself that in this issue I would talk about something other than a disease outbreak at an animal shelter. I’d hate to have to rename my column “Outbreaks ‘R’ Us.” But along with terrible hurricanes, last summer also brought news of a new canine respiratory infection spreading inexorably across the nation. First seen among greyhounds, canine influenza has now been documented at shelters, boarding kennels, and within the general pet population in many states. Inevitably, shelters are fielding questions from within their own ranks and from the public. Whether or not this disease has struck in your community, it’s one we all need to know about.
Although I didn’t realize what it was at the time, I first heard of canine influenza through a posting to the e-mail list of the Association of Shelter Veterinarians on April 15, 2005 (tax day and canine influenza—how lucky can a shelter vet get?). Kelly Ann Rada, the veterinarian at the time for the Jacksonville Humane Society in Florida, wrote, “We are having a rash of kennel cough cases … We always have our 3-6 usual cases, but the number is 24 now between our shelter and animals out in foster. We’ve only got 4 healthy dogs available for adoption! Plus, these kennel cough dogs are sicker than our usual cases … inappetant and lethargic, with 3 cases of pneumonia needing fluid support.”
A flurry of advice went out. Had she considered canine distemper? Perhaps an unusually virulent outbreak of Bordetella kennel cough? Among the replies was a suggestion that she contact Cynda Crawford, a researcher at the University of Florida. Dr. Crawford had recently identified a new strain of influenza as the culprit in an outbreak of severe respiratory disease at a nearby greyhound track. Samples were submitted, and the influenza virus was identified in the sick dogs from the Humane Society.
Six months later, a few cases were still being seen at the shelter. I interviewed Dr. Rada about her experience with managing this new disease in a shelter environment.
Can you tell me more about how you first suspected something unusual was going on?
Normally, the majority of our kennel cough dogs are happy, bouncy, bright, and alert. Suddenly, some of the dogs were developing high fevers, were lethargic, didn’t want to eat, with profuse nasal discharge dripping out of the cage. I found you can’t necessarily tell the difference with any one case between canine influenza and “normal” kennel cough, but overall we saw more severe and more frequent disease. It would have been helpful if we had been keeping baseline records of respiratory disease for comparison, and that would be one recommendation I’d make to others.
Do you know how it got started?
We think we might have a couple of index cases that came in at the end of March [2005]. Interestingly, these weren’t your typical unkempt stray animals with no history. Most of the dogs that got sick were adult, well-vaccinated, owner-relinquished animals. We only had a handful of puppies that ever were sick. That may be because puppies are housed in a separate area from the adult dogs.
How contagious was it?
Very contagious. We found about 80 percent of exposed dogs—those from the same ward as an infected animal—got sick, though we did have a few animals that were exposed and never came down ill. We’re cursed with a very poor ventilation system, and that may have been part of why it spread so easily.
How did you manage it in the shelter? That must have been difficult under the circumstances you describe.
It was challenging. At that time, we had a contract to provide animal sheltering for the county, so we couldn’t close admissions. We didn’t want to depopulate and didn’t have a good, quick test to see who was infected. [Diagnosis requires paired serum titers two weeks apart.] We also knew we were going to limited-admission and “no-kill” within a few months. We decided this would be a good test case: what do we do with a treatable, curable illness that will still tax our resources? We decided we would treat adoptable animals in foster homes with no other dogs. Our isolation only housed six dogs, [and] we reserved [the area] to treat those with severe pneumonia. We set up standard protocols so animals could be started on treatment even in my absence. We had wonderful success in terms of recovery. There were only two fatalities—both dogs who had preexisting conditions.
How did you go about isolating sick dogs?
Our isolation is not on a separate ventilation system, but at least it is a separate ward set well apart from the main dog area. We only allowed a limited number of employees access, and used gowns, gloves, and foot baths. Our isolation is so small, though; we weren’t able to isolate all sick dogs until we went to limited-admission recently. Even though it’s not perfect, now that we have it down to a few dogs, our isolation area seems to be doing the trick.
How did you treat the affected dogs?
Doxycycline had been my standard drug of choice for kennel cough, but it was not effective in these cases. Dr. Crawford was able to provide us with antibiotic sensitivity data for the secondary bacterial infections she found in some cases, and we chose our antibiotics based on that information. We did not use cough suppressants because many of the dogs had a productive cough. In a few dogs that were doing very poorly and seemed headed for euthanasia, I tried an injection of steroids. I was nervous about doing this, but spoke to a vet at a boarding kennel who said she had success with it in some cases. I wouldn’t do that in all dogs, but if the alternative is euthanasia, it may be worth a try. I got a good response within 48 hours. We did not see any of the horrible, hemorrhagic pneumonia reported in greyhounds.
Did you stay open for adoptions? It seems like that would risk spreading the disease out into the community.
We knew if we closed down adoptions, we would have to euthanize some dogs—either dogs awaiting adoption or dogs coming in. We decided that we would be very clear with the media and the public and remain open for limited adoptions. We did not allow casual adopters—only if they had a specific interest, and only to people who did not have other dogs at home. Even if they weren’t sick, we had them sign a waiver. We knew the incubation period was two to five days and told people to isolate until then. At our shelter, all animals are spayed before they leave. Between that and the minimum stray hold period, most animals are already past their incubation period by the time they go home. We had no problems with this system. Not only was there no backlash, but I was amazed at the number of people still willing to adopt dogs given the illness.
Was there any local publicity about the outbreak?
Luckily, we were very proactive from the beginning. Even before we had a definite diagnosis, we were open with the media and local vets. When we realized we had influenza, we felt an obligation to share that with the veterinary community. Letting them know was challenging because there’s no real forum to easily reach all the area veterinarians. I sent letters out to all the addresses I could find. I did get some very positive feedback—for many, that was the first time they had been contacted by the shelter. We got the best results when we put together a fact sheet and had the state veterinarian release it.
How did you finally get the outbreak under control?
In the beginning we reexamined all our cleaning protocols and rehauled everything, but it became clear after a while it wasn’t for a lack of cleaning. Everyone wants to know the silver bullet, and unfortunately there wasn’t a solution like that. As long as we were still treating and bringing these animals in, we weren’t able to stop it. Since March we’d been dealing with 12 to 24 cases at any given time.
After October 1, we stopped taking in strays, and we were finally able to immediately isolate all sick dogs. … We [did] daily rounds and everyone [was] trained to look for signs of disease, and I think that helped. In a perfect world we would have closed intake at the very start, but it just wasn’t practical at the time. Since this disease is likely endemic in our area, I couldn’t guarantee that it wouldn’t come right back in.
Canine Influenza FAQ
What is canine influenza and where did it come from?
Canine influenza is a viral respiratory infection of dogs. It is closely related to equine influenza virus, which has been known for over 40 years as a cause of respiratory disease in horses. Canine influenza is believed to have “jumped species” from horses to dogs relatively recently; blood samples from before the year 2000 show no evidence of exposure, but since 2000, there has been increasing evidence of infection, first in greyhounds and more recently in pet dogs.
Is it common for a virus to suddenly infect a new species?
It’s not all that common, but it’s certainly not unheard of. Most of the time, when a virus moves into a new species, it creates a “dead end” infection; that is, it does not spread readily from animal to animal within the new species. However, once in a while a virus adapts not only to cause disease but also to spread from one animal to another, which is when large-scale epidemics can occur.
Those of you who have been around for a while may remember when this happened with canine parvovirus around 1978. Canine parvo is believed to have evolved from the closely related feline panleukopenia, and within just a few years it spread worldwide in the canine population. Other recent and well-publicized examples include the coronavirus that caused SARS in humans, believed to have originated from a strain that normally infects bats; it was perhaps transferred through civet cats. SARS proved capable of both infecting humans and spreading directly from one person to another.
Right now many people are worried about avian influenza H5N1. This virus has made the first jump—a human exposed to an infected chicken can contract the disease—but has not yet made the jump allowing it to be spread human to human. (Scarily enough, cats can also contract this infection when exposed to infected birds, and can spread disease from one cat to another. See “Who’s Afraid of the Big Bad Bird Flu?” for more information on avian flu.)
There are some things in common among the chicken factories where avian influenza viruses tend to spring up, the wild animal markets where SARS made the transfer to humans, and the greyhound tracks where canine influenza first emerged. When stressed animals are crowded together, there are many opportunities for viruses to mutate and spread, sometimes into new species. For this reason, in shelters where many of the same crowded and stressful conditions exist, it’s important to keep species separate as much as possible. This is especially true for sick animals. Some shelters have only one isolation area where sick dogs and cats must be housed together; others have no isolation at all and house sick dogs and cats among the general population. If you can find creative solutions to avoid this—a donated trailer, a few cages in the laundry room, foster care, refurbished outdoor spaces in temperate climates—it will help protect all of us from species-jumping infections. This will also protect shelter animals from the infections we already know can cross species, such as Bordetella bronchiseptica (the main bacterial cause of kennel cough) and some strains of canine parvovirus that can infect cats.
Can canine influenza infect humans or any other species?
So far, there is no evidence that canine flu can infect humans. The closely related equine influenza has been around for many years and has never been considered a human health risk. Researchers are still investigating whether or not cats can be infected and spread the disease; no outbreaks of disease associated with canine influenza have been found in cats so far. There is no evidence at this point that the disease can spread to ferrets or back to horses.
What are the signs of infection?
Canine influenza can look just like kennel cough arising from other causes. This is especially true in shelters, where we are more likely to see relatively severe forms of the common kennel cough infections. The most common signs include nasal discharge and a dry or moist cough that can last up to about 30 days. Some dogs may experience fever, loss of appetite, and depression; a small percentage develop pneumonia. The incubation period is short—usually around three to five days. About 20 percent of exposed dogs will get infected without showing any signs of disease. Although fatalities are more common than with run-of-the-mill kennel cough, the rate is still low, estimated at about one to five percent. Aside from greyhounds being more susceptible to fatal infection, there are no known ways to predict which dogs will be at risk for severe disease. All ages seem to be at risk, and severe disease has been reported even in healthy, well-vaccinated pets.
In outbreaks among racing greyhounds, “acute hemorrhagic pneumonia” has been seen; this may present as sudden death with blood coming out of the nose and mouth. This has not been reported as a common feature among infections in pet or shelter dogs, and may reflect a different susceptibility in greyhounds. So if you are seeing sudden death in shelter dogs, some other infection is likely involved, although canine flu may be present as well. Contrary to rumors, canine influenza does not mimic parvo. If it looks like parvo, it probably is!
So how can we tell if we have canine influenza in our shelter?
When canine flu first shows up in a given population, none of the dogs will have immunity. Therefore, you would expect to see a noticeable increase in the number of cases, as almost all exposed dogs will get infected and most will show some signs. For shelters that euthanize for kennel cough, an increase in the sheer number of cases may be the only clue. Reports or complaints from local vets, adopters, or rescue groups may also alert the shelter to more severe disease. Shelters that routinely treat kennel cough in-house are likely to see a higher percentage of severely affected dogs than they would during an outbreak of the usual kennel cough. Because the incubation period is relatively short compared to that of some other common causes of kennel cough, you may see disease sooner after admission than usual. Keep in mind that there are other causes of unusually frequent or severe kennel cough; canine influenza is only one of several infections to consider.
Smaller shelters can easily recognize a sudden jump in the number of cases of kennel cough without any special record-keeping system, but large shelters might consider starting to track their numbers of kennel cough cases now. This could be a simple system such as keeping a tally of the number of dogs started on treatment—and adding a special category in your shelter software system to track “kennel cough” as a reason for euthanasia.
How can canine influenza be diagnosed?
There is no quick test to diagnose infection in a live dog. Infection is determined by finding antibodies in serum. Acute infection can be diagnosed by collecting two serum samples, one at the onset of signs and one two weeks later to detect rising antibody levels. A history of infection can be established by analyzing a single serum sample any time more than one to two weeks after exposure. Although a single sample cannot tell you whether or not a recent infection was caused by canine influenza, a high percentage of positive antibody titers in a shelter population is strongly suggestive that disease is being spread at the shelter, as it is unlikely in most areas that many dogs are coming in already infected.
Virus can also be detected from tissue samples taken at necropsy. If dogs die or are euthanized for severe kennel cough, this would be the quickest way to establish whether influenza played a role. Unfortunately, attempts to isolate virus from respiratory secretions of live dogs have not been successful so far. Researchers are still working on this, and a PCR (polymerase chain reaction) test may be available in the future.
Right now, Cornell’s Animal Health Diagnostic Center is the only place set up to look for antibody titers. More information about sample submission is available on their website at www.diaglab.vet.cornell.edu/news.asp. (You can also find this page by typing “Cornell canine influenza test” into Google.) UC Davis is working on developing diagnostic testing services, and testing will likely be more widely available in the future. In addition to testing for canine influenza, diagnostic testing for other canine respiratory infections—including Bordetella bronchiseptica, canine distemper, parainfluenza, and adenovirus—should be considered.
How should we treat dogs with canine influenza?
As with all viral infections, treatment mainly relies on supportive care. Secondary bacterial infections probably play a role in causing severe disease, so antibiotic treatment may be needed, especially in dogs who develop green or yellow nasal discharge or pneumonia. A broad spectrum antibiotic should be used, as a wide variety of secondary bacterial infections have been found in dogs with influenza. Doxycycline, which is a common first-choice antibiotic for kennel cough, may not be effective against secondary bacterial infections associated with influenza. Cough suppressants should be avoided in dogs that have a productive cough.
Trials are underway to assess whether the antiviral drug Tamiflu is effective against canine influenza, and if so, at what dose. In the meantime, using Tamiflu at ineffective or partially effective doses raises the risk of developing drug-resistant strains of influenza, a phenomenon that has already been documented in humans. In the interest of maintaining the usefulness of this valuable drug, it would be best to wait for more information before deciding to use Tamiflu for treatment of canine influenza. The Jacksonville Humane Society and others have been able to successfully manage outbreaks with antibiotics and supportive care.
How does canine influenza spread, and how can we keep it under control in our shelter?
Like the other canine respiratory agents, canine influenza can be transmitted by direct contact or fomites or via airborne spread. Spread may be similar to that of canine distemper, transmission of which has been observed over distances of up to 20 feet, even from non-coughing dogs. The virus is easily killed by any commonly used disinfectant such as a quaternary ammonium compound or bleach solution diluted at a ratio of one part bleach to 32 parts water. As with distemper, continued spread is more likely a result of inadequate isolation of sick and carrier animals, rather than inadequate cleaning. To control an outbreak in a shelter, focus on quickly identifying sick animals and removing them from the general population. Isolation in an area with separate air supply is ideal, but a separate room even with shared ventilation will be significantly helpful. Sick dogs should be handled with the usual isolation precautions: separate cleaning and care supplies, gowns or other protective clothing, and gloves, as well as shoe covers, dedicated boots, or foot baths. Dogs may continue shedding virus up to a couple of weeks after recovery. Ideally, dogs should be kept isolated during this time, though they may be adopted out with appropriate medical waivers and adopter education. Although spread to the pets of shelter staff has not been reported, it is always a good idea to change clothing and wash hands between work and home. You don’t want your own pet to be the first case study!
Is there a vaccine available?
Not yet, although trials are underway. A vaccine against equine influenza is available for horses, though like many respiratory vaccines, it is not 100-percent effective. The fact that there is an equine vaccine suggests the same may be achieved for dogs, but the equine vaccine is not safe for use in dogs, nor has it been shown to be protective. Don’t use it! While we wait for a flu vaccine, vaccinating all dogs at intake against the respiratory agents we can ward off remains a good investment; the practice will also help reduce the occurrence of false alarms regarding canine influenza. This includes distemper, Bordetella, parainfluenza, and adenovirus-2. Canine distemper vaccine is always given as a subcutaneous injection, while the other three may be given as intranasal or injectible vaccines.
What should we tell the public?
Canine influenza has spread widely recently and is likely to continue doing so. Wherever it appears, few or no dogs will have immunity. Still, most cases so far have been associated with exposure in places where dogs are closely congregated, such as boarding kennels, shelters, and veterinary clinics. “Kennel cough” is called that for a reason—it mostly happens in kennels. Casual exposure at a location such as a dog park is less likely to result in significant disease, though of course it’s not impossible. Reasonable precautions include keeping dogs up-to-date on vaccinations against other respiratory infections, keeping dogs away from dog parks and other common dog-gathering places during and at least two weeks after any illness, and making sure that boarding kennels are clean, well-run, and armed with a plan for isolating dogs who develop kennel cough.
What is the bottom line?
It’s never good news to find we have yet another infectious disease to contend with in our shelters. On the other hand, this is not the new dog SARS or parvo. The death rate is low, and efforts to find effective vaccines and treatments are already underway. You can use this as an opportunity to review current procedures for managing canine respiratory disease, and to become better prepared for canine influenza should it strike your shelter.
Kate Hurley is the director of the UC Davis Shelter Medicine Program.
