Avian influenza in cows – How much of a risk is it?

Most of us will be familiar with the concerns raised by avian influenza A virus (IAV) and the risks of wild bird populations spreading bird flu into poultry flocks. IAVs are, however, increasingly being detected in mammals, so posing a new threat to both livestock and human populations. These viruses are primarily categorized by two groups of surface proteins: hemagglutinin or “H” proteins (H1–H16), and neuraminidase or “N” proteins (N1–N9). The different potential combinations of “H” and “N” give rise to many different subtypes, with H5N1 emerging in 1997.  Multiple lineages of H5N1 have evolved since. The highly pathogenic Clade 2.3.4.4b H5N1 was first detected in 2020 and soon became the predominant clade circulating in wild birds worldwide and spreading from them to cause disease outbreaks in domestic poultry. 

This particular clade also demonstrates an enhanced capacity for mammalian adaptation and cases have now been confirmed in a wide range of species. In the USA these currently include bears, seals, skunk, various rodents, alpacas and a number of carnivores ranging from tigers to domestic cats. In March 2024 American authorities including USDA announced that H5N1 had been identified for the first time in dairy cattle. It has subsequently spread widely within the USA and H5N1 has currently (December 2025) been confirmed in over 1000 dairy herds located in 19 states.

In these infected herds about 10-20% of the lactating cows generally show clinical signs, with little evidence of disease in non-lactating animals including dry cows, heifers and bulls. Birds with H5N1 suffer high morbidity and mortality and this also applies to domestic cats. The latter suffer highly pathogenic lesions in their central nervous system and lungs causing severe neurological and respiratory symptoms. Illness and deaths amongst farm cats is often, therefore, the first sign that a farm has been exposed to the virus. In contrast, common symptoms in cows are similar to those associated with mastitis including low appetite, reduced milk production and abnormal milk which becomes thickened and discoloured. Most cows recover with supportive treatment, with a mortality/culling rate ≤2%. A concern is, however, that the virus shows tropism for lactating mammary epithelium with high concentrations of virions shed into milk, so posing a potential food safety risk to anyone consuming unpasteurized milk or cheese.

Epidemiological evidence from the USA suggested that the outbreak in cattle probably began following contamination of feed or equipment by droppings or carcasses from wild birds, followed by efficient local transmission. Within herds the spread of H5N1 is directly cow-to-cow, although there may also be infection through contaminated milking equipment. The virus can also spread from dairies to nearby poultry facilities, with the main risk factors thought to be shared personnel, vehicles and/or equipment between premises. Animal movement is therefore a recognized risk for disease transmission. Since 2024 there have also been 71 diagnosed human cases in the USA. These were primarily in agricultural workers having close contact with livestock, of whom 41 were associated with dairy herds and 24 with poultry operations. Reported symptoms have ranged from conjunctivitis to more typical flu symptoms, such as fever, chills, coughing, and sore throat/runny nose. However, some serious illnesses have occurred and two people have died. There has been no indication so far of person-to-person transmission. 

Until recently the particular form of the virus infecting American cows (AI A(H5N1) clade 2.3.4.4b genotype B3.13) had never been detected in the UK or Europe. In Great Britain this negative result was based on sequencing of outbreaks on poultry premises, testing a proportion of wild bird cases and, in 2024, a representative selection of mastitis cases in cows. In March 2025 a case of H5N1 was confirmed in milk from a single sheep in the UK. This animal was identified following routine surveillance of co-located livestock on a Yorkshire farm where avian influenza had been confirmed in captive birds. An individual dairy cow in the Netherlands has also now tested positive, with antibodies to H5N1 present in its blood. These are thought to have developed following an avian influenza infection in December 2025, when the cow had symptoms of mastitis and a reduced milk yield. The reason for this investigation was the detection of H5N1 in a farm cat. Twenty cows were sampled initially, with just one testing positive for antibodies but not virus.  All the remaining lactating animals on the farm have since been tested at Wageningen Bioveterinary Research.  Based on PCR none of them carried the virus but results of the antibody tests are not yet available. The most likely source of infection is again thought to be wild birds. The cat could either have eaten a sick bird or been given infected milk from the cow. 

Previously the risk of this particular clade of H5N1 virus reaching the UK was considered to be low. There is no trade in live cattle from the USA and, although some migratory wild birds from the USA reach West Scotland and NW England via the East Atlantic flyway, not many species are involved and the frequency is low. Some raw dairy products including cheese and colostrum are imported for human consumption or to feed to calves. Colostrum for calves requires brief pasteurization at 72^oC for 15 secs. Although standard pasteurization inactivates the virus, the effectiveness of this treatment is uncertain. 

The Dutch result now raises concern that transmission from wild birds to domestic animals including both cows and cats is indeed possible in both the UK and elsewhere in Europe, requiring a greater level of future vigilance. Although clinical signs in cows can be mistaken for mastitis, unexplained deaths, neurological and/or respiratory symptoms noticed in birds or other mammals present on farms including cats should be reported so that appropriate testing can be undertaken promptly. This is vitally important due to the zoonotic potential of this virus and also its ability to spread to other livestock premises. Staff working on poultry units should already be aware of the biosecurity precautions necessary, but this knowledge must be extended to other farm workers. The USA have two federal orders in place requiring testing of all lactating cows prior to interstate movement and that dairy processing facilities nationwide test raw (unpasteurized) milk samples from their milk silos. No H5N1 vaccines have currently been approved for dairy cattle, but vaccines are under development in the United States and field trials are underway.

In conclusion, the discovery of H5N1 in dairy herds and its transmission to humans has highlighted the need for regular testing of both animals and milk and for rigorous biosecurity protocols on farms. Successfully controlling viral diseases in animals requires a holistic One Health approach involving government, academia and industry to organise the necessary surveillance and control procedures and to undertake vital underpinning research.

Written by: Claire Wathes, Emeritus Professor at the Royal Veterinary College.

Claire is a member of the BISAS Scientific Committee, a cross discipline group of animal scientists who meet regularly to discuss topical matters which may impact on our members.