Avian H5N1 Influenza in Cats

See allHide authors and affiliations

Science  08 Oct 2004:
Vol. 306, Issue 5694, pp. 241
DOI: 10.1126/science.1102287


During the 2003 to 2004 outbreak of avian influenza A (H5N1) virus in Asia, there were anecdotal reports of fatal infection in domestic cats, although this species is considered resistant to influenza. We experimentally inoculated cats with H5N1 virus intratracheally and by feeding them virus-infected chickens. The cats excreted virus, developed severe diffuse alveolar damage, and transmitted virus to sentinel cats. These results show that domestic cats are at risk of disease or death from H5N1 virus, can be infected by horizontal transmission, and may play a role in the epidemiology of this virus.

The relationship between avian influenza A virus and its hosts has changed markedly in recent years, with important consequences for human health (1). The most recent example is the 2003 to 2004 avian influenza A (H5N1) virus outbreak in Asia, which not only caused vast mortality in poultry, but also resulted in 39 officially reported cases of direct bird-to-human transmission, of which 28 were fatal (2). During this outbreak, there were also anecdotal reports of fatal H5N1 virus infection in domestic cats and zoo felids after they had fed on virus-infected chickens (3). This is unusual, because domestic cats are generally considered to be resistant to disease from influenza A virus infection (4).

To determine the pathogenicity of this virus for domestic cats, we experimentally infected 4- to 6-month-old European Shorthair cats with H5N1 virus by different routes and examined them by virological and pathological techniques. Each group of cats was placed in a separate, negatively pressurized isolator.

First, we intratracheally inoculated three cats with 2.5 × 104 times the median tissue culture infectious dose (TCID50) of a H5N1 virus isolated from a fatal human case in Vietnam (A/Vietnam/1194/04). The cats showed clinical signs, including significantly raised body temperature from 1 day post-infection (dpi) onwards (P < 0.05, one-way analysis of variance) (fig. S1) and decreased activity, protrusion of the third eyelid, conjunctivitis, and labored breathing by 2 dpi. One cat died unexpectedly at 6 dpi. The cats excreted virus by 3 dpi at relatively low titers (Fig. 1A), likely because the infection predominantly involved the lower respiratory tract. On necropsy at 7 dpi, they had multiple or coalescing foci of pulmonary consolidation (Fig. 1B), which consisted histologically of diffuse alveolar damage, resembling that from H5N1 virus infection in humans and nonhuman primates (Fig. 1C) (5). H5N1 virus infection was confirmed as the cause of these lesions by virus isolation and immunohistochemistry (Fig. 1D). In contrast, three cats inoculated with an influenza A (H3N2) virus isolate from a human case in the Netherlands (A/Netherlands/18/94)—the most prevalent subtype of influenza A virus in humans—showed no evidence of virus infection or disease. These results show that this H5N1 virus can productively infect domestic cats, cause diffuse alveolar damage, and result in clinical disease or death.

Fig. 1.

Virus excretion and pulmonary lesions in cats infected with influenza A (H5N1) virus. (A) Pharyngeal virus excretion (mean and standard deviation) in cats inoculated with H5N1 virus either intratracheally (solid circles), by contact with H5N1 virus–inoculated cats (open circles), or by feeding on chicks inoculated with H5N1 virus (solid triangles). Control animals were either fed chicks inoculated with PBS solution (open triangles) or inoculated intratracheally with H3N2 virus (solid squares). (B to D) Left: The lung of a cat fed with an H5N1 virus–inoculated chick has (B) multiple foci of consolidation (arrowheads), consisting histologically of (C) diffuse alveolar damage, with (D) expression of influenza virus antigen in inflamed tissue, shown by immunohistochemistry. Right: The lung of a cat fed with a PBS-inoculated chick.

Second, we tested whether cats could be infected with H5N1 virus through horizontal transmission by placing two sentinel cats in contact with the intratracheally inoculated cats above at 2 dpi. Third, we determined whether cats could by infected with H5N1 virus by feeding on virus-infected birds. To test this, we inoculated 1-day-old chicks intratracheally with 2.5 × 104 TCID50 of H5N1 virus and, after euthanasia at 1 dpi, fed one chick to each of three cats. In both the sentinel cats and the cats fed on infected chicks, virus excretion (Fig. 1A), clinical signs (fig. S1), and pulmonary changes were similar to those of intratracheally inoculated cats. In contrast, two cats fed on chicks inoculated with phosphate-buffered saline (PBS) solution showed no evidence of virus infection or disease (Fig. 1, A to D, and fig. S1). These results show that cats can be infected with H5N1 virus both by horizontal transmission and by feeding on virus-infected birds.

The implications of these findings are, first, that during H5N1 virus outbreaks, domestic cats are at risk of disease or death from H5N1 virus infection, either because of feeding on infected poultry or wild birds (6) or because of contact with infected cats. Second, the role of cats in the spread of H5N1 virus between poultry farms, and from poultry to humans, needs to be reassessed. Third, cats may form an opportunity for this avian virus to adapt to mammals, thereby increasing the risk of a human influenza pandemic.

Supporting Online Material

Fig. S1

References and Notes

View Abstract

Navigate This Article