It was a busy week for H5 in U.S. Dairy One Health news this week, with the block buster information coming from the BioRx release of the USDA and collaborators’ initial analysis of diagnostic, epidemiologic and genomic analytics information of initially reported cases –
This dataset strongly implies a single introduction of an H5N1 2.3.4.4b clade virus into cattle late in 2023 or early 2024 in the Texas panhandle-eastern New Mexico area, with subsequent silent spread until initial clinical diagnosis in late March 2024. Spread to most other infected states was documented retrospectively through infected cattle movements.
Please read the entire report for a more accurate description of the findings and the many implications drawn from the data already collected. Existing shared data provides a wealth of information leading to some profound insights pieced together by a top-notch team of USDA and associated scientists.
In my opinion it is dangerous to over-generalize conclusions based on what is likely a too-limited dataset. Samples were collected over a relatively short period of time largely in herds subsequently tied together epidemiologically. Retail milk testing and limited wastewater sewage sampling suggest that many more herds may be infected across a wider geographic area. We have no public sequence or epidemiological data on additional herds likely contributing to more likely widespread genomic evidence across time and space. These additional samples may or may not fit in the “single crossover” phylogenetic tree.
A “post-game report” of the USDA preprint and a separate CDC study of the human conjunctivitis case from the NEJM (Highly Pathogenic Avian Influenza A(H5N1) Virus Infection in a Dairy Farm Worker | New England Journal of Medicine (nejm.org)) was released in Ars Technica yesterday and is worth reading for its expert analyses of the current state of knowledge related to this outbreak:
We still don’t understand how one human apparently got bird flu from a cow | Ars Technica
From this article:
Virologist Angela Rasmussen, an expert in emerging infectious diseases at the Vaccine and Infectious Disease Organization at the University of Saskatchewan, agreed with the assessment. "[I]t does seem like the most likely explanation is that it was from a cow," Rasmussen told Ars. "That fits with what has become very apparent about this outbreak: it's incredibly undersampled and much bigger than we currently have been able to quantify. To really understand this, we need to really increase sampling and surveillance as well as increase the number of sequences available for these analyses."
In a release yesterday from Colorado, State veterinarian Dr. Maggie Baldwin stated that the first confirmed case in Colorado (4/26) is not epidemiologically linked to the earlier cases and that a second herd is under investigation (NOT A CONFIRMED CASE):
What to know about bird flu found in Colorado dairy cattle (coloradosun.com)
I have not seen any information on phylogenetics of the first case to date, although it’s only been 8 days since confirmation. In NVSL’s defense, their staff is undoubtedly swamped with a huge total case load at this time with H5 dairy, poultry, and wild bird, as well as other “more routine” casework.
In the same Colorado Sun article, Dr. Baldwin provided the following dairy movement information:
“There is a tremendous amount of movement of dairy cattle across the United States and between states,” Baldwin said. Colorado has 106 dairies with about 200,000 cows. Baldwin said in the first quarter of this year, dairy farmers exported about 34,000 cattle from Colorado and imported about the same number. In other words, about 17% of Colorado’s population of dairy cattle turned over in three months.
While the first quarter and Colorado data may be somewhat of an aberration, it’s obvious that the dairy industry is extremely mobile. It’s not hard to imagine that asymptomatic infection(s) in one part of the country can easily disseminate nationwide in a short period of time to larger herds utilizing interstate split stage production models.
In summary we’ve grown our essential food animal production systems into intricate and efficient systems that can also rapidly spread infectious agents much faster than our current detection systems can handle! Utilizing paper-based (or even electronic based) interstate movement reporting of clinical signs-based disease assessments by accredited veterinarians is much too slow and ineffective. Even adding molecular testing requirements based solely on recognition of clinical signs to initiate sampling will not compete with modern dynamics of disease spread.
The future lies with ongoing agent and species-agnostic sensing in high-risk and accessible streams (think wastewater surveillance and milk and product flows). However, getting there will be a legal and political minefield, requiring a huge change in mindset for our production-processing, regulatory, and laboratory communities.
Until next week, watch for periodic article reposts here and/or on LinkedIn
John