Headline writers are probably going to screw up my day tomorrow by sensationalizing this, but a new paper in Animals (Fiorito et al) describes SARS-CoV-2 infection in cattle in Italy. (It’s an MDPI journal so I take anything published there with a big grain of salt given their low standards, but this one seems ok). It’s also similar to a pre-print from Germany that we’ve known about for a while.

Why do I say “take a deep breath and relax”?

I’ve talked about the need to consider spillback into animals since 2020. However, there are different concerns with different situations. Livestock raise some big concerns because there are a lot of them, they live in large groups in close proximity to people and we have contact with food product from them. However, we have to consider what the results of surveillance really mean. Data are great…we just need to think about how to incorporate those results into our understanding of this virus and what threats might be present (or not).

So, let’s look at what this study tells us.

It’s a nice but small study of cattle from a farm where 13 of 20 workers had diagnosed or suspected COVID. These are situations where it’s great to do animal surveillance since there’s a clear chance of exposure of the animals.

They collected nasal and rectal swabs, as well as milk, from 24 lactating cattle. Those were tested by PCR for evidence of active infection.

  • All were negative.

We know that PCR testing is a challenge when doing animal surveillance, since infected animals only shed the virus for a short period of time. As we saw with our dog/cat surveillance, it can be a challenge to sample animals during the appropriate time. Animals can be infected but eliminate the virus by the time we hear about the exposure, arrange to get samples and get there to collect them.

They also collected blood samples, which allow us to look at previous exposure through the detection of antibodies against SARS-CoV-2. The specificity of the test (how likely a positive result is really SARS-CoV-2 infection) isn’t clear, but they reported positive results in 11 (46%), 14 (58%) or 13 (54%) of cattle, depending on the test. The 54% number was from a serum neutralization test…not the standard virus neutralization test that’s been reported elsewhere but probably a result that’s fairly reliable.

So, what does this tell us?

It suggests that human-cattle transmission was relatively common in lactating cattle on this farm, where there was an outbreak in people. It’s too bad they only tested a small number of cattle and all were lactating. Lactating cattle are presumably high risk because there’s a lot of human contact a couple times a day for milking. It would have been nice to see if there was evidence of infection in other cattle and calves (and seeing low rates in low risk cattle would also have provided more weight to the positive results reported here). Regardless, this is pretty strong evidence that people infected cattle.

Did the cattle get sick?
There’s no evidence they had any problems. Lactating dairy cattle are pretty closely monitored so we can be pretty confident they didn’t have any overt signs of disease.

Could they have infected other cattle or people?

That’s hard to say. All were PCR negative. Whether that was because of timing or very low grade infection is impossible to say. However, given the low predicted susceptibility of cattle, lack of evidence of disease and zero prevalence by PCR despite high seroprevalence, it’s more likely that these cattle had very minor infections that resulted in them developing antibodies but not shedding enough virus to pose a risk for transmission. We need more study to evaluate this but nothing here suggests a major concern.

Can cattle be a reservoir?

For cattle to be a reservoir, they’d have to be able to infect other cattle and there would have to be ongoing transmission within the cattle population (vs a short term situation where cattle get infected and the virus burns through the group quickly). How susceptible they are, how well they can transmit the virus (if at all) and the number of cattle (to be able to sustain transmission) would impact this. Most likely, if cattle were able to transmit the virus, it would burn through the farm relatively quickly. Dairy cattle usually measure in the 10s or 100’s not thousands (like mink or wildlife), so sustained transmission seems unlikely .

What about cow-human transmission?

We don’t know. Most likely, cattle are ‘dead end hosts’ whereby they get infected and can’t pass on the virus. If they are able to spread it, the nature of dairy farms and the number of cattle, mean the risk would probably be short term and only to farm personnel (who are likely at greater risk of exposure from the other infected people, not the secondarily infected cattle). However, we need to look at this more. There’s been a lot of reluctance to test food animals, mainly because of fear of having to deal with positive results. So, our knowledge is still pretty superficial.

The take home?

Cattle are one of many species that this virus seems to be able to infect. Odds of it being relevant for cattle or public health are low but it’s worth further study.