Does H5N1 spread from cats to dogs and vice versa?

The host range of H5N1 is impressive: birds, of course; but also many mammals, including dogs, cats, stone martens, ferrets, mice, rats, humans. There are undoubtedly others. Cats are probably infected when they eat infected birds. Dogs? Not clear. Humans? Birds, other people on rare occasions. What else? In fact we know incredibly little about how various hosts are infected. Do cats spread it from cat to cat and dogs, dog to dog? How about cats spreading it to dogs and vice versa? Of course dogs to humans or cats to humans is an important topic. So it's good to see some studies looking at the various combinations. A team at the Friedrich-Loeffler-Institut in Germany recently published a small study about dog to cat or cat to dog transmission using a highly pathogenic H5N1 isolated from a cat in 2006.

First, four dogs were inoculated and housed in the same containment room as three cats and an uninoculated dog:

The cats had the possibility of withdrawing and hiding, but they could also have direct contact with the dogs through 1 part of the cage fence. During the study the cats frequently had direct nose-to-nose contact with the dogs. For a realistic contact exposure setting, the cats were fed by using the dogs' food and water bowls without prior cleaning. (Giese et al., Emerging Infectious Diseases)

The inoculated dogs showed mild signs of infection (elevated body temps, conjunctivitis ["pink eye"]). These signs resolved in a few days. Viral RNA was present but the dogs did not appear to be shedding virus. No cats were infected nor was the uninoculated dog. Under the conditions of this study, then, there was no evidence of transmission of infection from three infected dogs to one uninfected dog or three cats.

Now the cats. The inoculated virus originally came from a cat and all three cats suffered more seriously from infection than the dogs. They became seriously ill and excreted virus from both respiratory and digestive tracts. Three contact dogs, housed similarly as in the other experiment, did not become infected, nor was there any antibody development over time that might indicate a covert infection.

These results are reassuring but not at all conclusive. The cats got quite sick and although no attempt was made to see if they could transmit to other cats, we know from previous studies this is possible. While no interspecies (dog to cat or cat to dog) occurred, this is one viral strain from a cat and all animals started out healthy. There were also not very many of them. Only one dog breed was used (beagles). As the authors point out, some breeds are unusually resistant or unusually sensitive to infections.

But this is a badly needed data point. Let's have some more. The paper, "Experimental Infection and Natural Contact Exposure of Dogs with Avian Influenza Virus (H5N1)," Matthias Giese et al., is at Emerging Infectious Diseases (.pdf here).