#19,221
Last January (see Netherlands: NVWA Announcement on Avian Flu Antibodies Detected In Dairy Cow) we learned of the detection of H5N1 antibodies in a European cow on a Dutch farm following an investigation into 2 sick barn cats, one of which died from H5N1.
A week later, we saw a cryptic report from NOS (Nederlandse Omroep Stichting) - the Netherland's Public Broadcasting Network - indicating that a total of 5 Dairy Cows Had Tested Positive for H5N1 antibodies.
None of the cows were reported sick, and none tested positive by PCR, suggesting these were past infections, not current (see ECDC Statement On H5N1 Antibodies in Dutch Dairy Cow: ECDC Risk Assessment Remains Unchanged).
Since then, we've heard very little about this apparent outbreak.
Yesterday, the ECDC Journal Eurosurveillance published the first detailed account of the investigation, where we learn that - on this one farm - there were literally scores of animals with detectable H5N1 antibodies, not five.
Reassuringly, `oropharyngeal swabs collected from members of the farmer's family tested negative for avian influenza virus', but there is only a narrow window (which had likely passed) where PCR tests would be useful in detecting the virus.
The bottom line is that more than half the livestock tested on this one farm carried antibodies for H5N1, and the source of their infection, and the mode of in-farm transmission, remain murky.If antibody testing was performed on the family, it wasn't reported in this study. Notably, there is also no mention of testing of livestock on any surrounding farms. Hopefully there are other studies in the pipeline.
Monika Z Ballmann1 , Luca Bordes1 , Kim M Bouwman1 , Marc Y Engelsma1 , Sandra Venema-Kemper1 , Sylvia BE Pritz-Verschuren1 , Rene Heutink1 , Marit Roose1 , Marcel AH Spierenburg2 , Arco van der Spek2 , Hendrik IJ Roest3 , Evelien A Germeraad1
Since March 2024, a large-scale outbreak of highly pathogenic avian influenza (HPAI) H5N1 virus has affected dairy cattle in the United States (US) [1]. In contrast, no evidence of HPAI H5N1 infection in dairy cattle associated with the European genotype DI.2.1 has been reported up to date. Here we describe a case investigation and a farm-level survey that resulted in the first reported detection of antibodies against HPAI H5N1 in dairy cattle in Europe.
Case description and virological investigation of the cat
On 24 December 2025, nasal, rectal and conjunctival swabs were collected and submitted for diagnostic testing from a 4-month-old domestic cat living on a dairy cattle farm and exhibiting clinical signs, lethargic and respiratory signs, consistent with HPAI virus infection. The cat died 2 days later. All three swab samples tested positive for influenza A virus by RT-qPCR [2], and the virus was identified as HPAI H5N1 clade 2.3.4.4b genotype DI.2.1 using Illumina whole genome sequencing [3]. Phylogenetic analysis demonstrated that the virus clustered with HPAI H5N1 viruses associated with outbreaks among wild and domestic birds across Europe since October 2025 (Figure 1).
The HPAI H5N1 virus detected in the cat did not contain the PB2 M631L substitution that has been frequently reported in the B3.13 genotype in the US, infecting dairy cattle. The PB2 E627K substitution, a well-characterised marker of mammalian adaptation, was present. Furthermore, the PB2 I292V substitution, which was rarely observed in earlier European HPAI H5N1 viruses, but now predominantly identified in the DI genotypes, was detected. No other previously identified genetic markers known to influence virulence, host specificity or binding of host proteins were identified in the cat sequence.
Investigation of dairy cattle and milk
Following the detection of HPAI H5N1 virus in the cat, dairy cattle from the same farm were tested for avian influenza viruses and antibodies as part of an epidemiological investigation. On 15 January 2026, 20 individual milk samples (midstream milk from all four udder quarters in a single tube) and a bulk tank milk sample were collected. No viral RNA was detected in any of the milk samples by RT-qPCR targeting the matrix gene (M-PCR) [2]. However, antibodies against influenza A virus were detected in nine individual milk samples and the bulk tank milk sample with nucleoprotein (NP)-ELISA (ID Screen Influenza A Antibody Competition Multispecies, Innovative Diagnostics (ID), Grabels, France).
The farm was revisited on 22 January 2026: individual milk (n = 70) and serum (n = 72) samples were obtained from all lactating dairy cows and serum samples from the 38 youngstock (aged 1–2 years)
(SNIP)
Discussion
Since March 2024, a large-scale outbreak of HPAI H5N1 clade 2.3.4.4b viruses belonging to the B3.13 and D1.1 genotypes has been reported in dairy cattle in the US [1]. While Eurasian-lineage HPAI H5N1 clade 2.3.4.4b viruses have infected multiple mammalian species [6], including cattle under experimental conditions [7,8], to our knowledge, we report the first detection of H5N1-specific antibodies in European dairy cattle.
In the US, HPAI virus infection in dairy cattle appeared to be mostly asymptomatic or associated with non-specific clinical signs, mastitis, a rapid decline in milk production, reduced feed intake and rumination, lethargy and mild elevation in body temperature [9]. In our report, one lactating cow had clinical mastitis in all four udder quarters. Although the observed clinical signs were consistent with HPAI virus infection, no samples from the symptomatic animal at the time of clinical manifestation were available for testing for influenza virus; therefore, the involvement of other pathogens cannot be excluded.
During the investigation, antibodies were detected in more than 50% of the animals suggesting a predominantly subclinical presentation and highlights that infection in dairy cattle may remain undetected without targeted surveillance. While milk is a sensitive matrix for detecting HPAI virus during acute infection in dairy cattle [10], our results indicate that serum could be more suitable for serological surveillance, as antibody titres were higher in serum (VNT titres > 2,048 and up to 40 by HI) than in milk (VNT titres up to 128 and HI titres 0) from the same individual. This finding is consistent with findings from a previous study indicating VNT titres up to 813 in serum and up to 512 in milk [7]. Antibody waning may occur at different rates in serum and milk over time, which may explain the relatively large difference in VNT and HI titres observed in this investigation, but longitudinal studies assessing antibody waning in both serum and milk are currently lacking.
The exact source and route of virus introduction in the cat and cattle remain unclear. The farm is in an area frequented by overwintering birds that grazed on the same grasslands as the cattle. Increased wild bird mortality was observed in the area during October and November 2025. Bird carcasses were removed from the grassed pastures by the farmer but not submitted for virological testing. Around the last weekend of November 2025, dairy cattle and youngstock were moved indoors for the winter, while the cat continued to have access to the fields. The cat may have acquired the virus through consumption of an infected bird or contaminated milk from infected cattle. Cattle exposure may have occurred via contaminated environment, feed or direct contact with wild birds.
The high proportion of dairy cows and youngstock with antibodies suggests either a high level of primary exposure or the possibility of cow-to-cow transmission. Further investigation is required to clarify transmission dynamics.
In this study, oropharyngeal swabs collected from members of the farmer's family tested negative for avian influenza virus. Although, since the onset of the outbreak of HPAI H5N1 virus in cattle in the US, occasional human infections have been reported. To date, no sustained human-to-human transmission has been documented. Current public health assessments consider the risk of clade 2.3.4.4b H5N1 viruses to the general population to be low, with human cases limited to sporadic infections following close contact with infected poultry or dairy cattle [11,12]. Consumption of raw milk or unpasteurised dairy products may also pose a risk of virus exposure [13].
Conclusion
We report detection of antibodies in Europe against the Eurasian lineage of H5N1 virus in dairy cattle, identified after a cat living on the same farm tested positive. Infection of a mammalian livestock species with frequent human contact warrants continuous attention, as it increases opportunities for virus transmission and subsequent adaptation and underscores the importance of integrated surveillance on the animal-human interface.
Despite the aggressive spread of H5N1 in dairy cattle in the United States since 2024, livestock surveillance in Europe remains limited, based (in part) on the belief that the B3.13 genotype in North America was somehow unique in its ability to infect cattle (see DEFRA Risk Assessment Of HPAI H5N1 Occurring in Cattle In the UK).
This, despite the fact that over the summer of 2024 the FLI successfully infected cattle with a contemporary European H5N1 virus, finding it replicated efficiently in bovine mammary tissue and could produce adaptive mutations (PB2 E627K) during replication, and the 2025 discovery that genotype D1.1 could also infect cattle.
Last October - more than 18 months after the first detection of HPAI H5 in U.S. cattle - OFFLU released an 11-page statement (see OFFLU Guidelines for High Pathogenicity Avian Influenza Virus Risk Mitigation in Cattle) which called for sweeping changes, including:
- Implementing risk-based bulk milk surveillance
- Enforcing pasteurization and safe disposal of waste milk
- Applying flexible movement controls with testing and quarantines
- Improved biosecurity & milking practices
- Protecting workers with PPE
How diligently these guidelines are now being followed in Europe - and the rest of the world - is unknown.
This report should be (another) wake up call that livestock outside of the United States can no longer be assumed to be somehow immune to H5N1, and that bulk milk testing may not be sensitive enough to reliably detect the virus in dairy cows.
Assuming anyone is listening, that is.
