It’s been known for a while that it’s been possible to recover viral genetic material from the poop of people with COVID-19. Could this be used to track and quell outbreaks?

Fast and accurate tests for COVID-19 remain challenging; I’ve written about the main methods before and each has its strengths and weaknesses.

But it’s been known for a while that many people who have COVID egest viral material in their stool. Or, put another way, you can find viral RNA their poo if you’ve a mind to look. The presence of viral RNA is not reliable enough to have much use as a diagnostic test at an individual level because not everyone who gets infected poops out the virus (or, more accurately, its debris) but there has been talk of testing sewage at a community level to track the spread of the disease. Trials have been conducted in a number of countries but a recent one shows you can do more than just track outbreaks.

Can you predict outbreaks with poop?

Yes. And the point is testing sewage for evidence of the virus is not just accurate, it’s predictive. In a fascinating study just out of Yale the researchers found:

  • During an outbreak SARS-CoV-2 RNA was detected in all sewage sludge samples;

  • When adjusted for time lag, the viral RNA concentrations correlated precisely with the COVID-19 outbreak curve;

  • The sewage testing tracked to hospital admissions as well as positive tests;

  • RNA concentrations were an accurate indicator seven days before COVID-19 lab test data on patients, and predicted local hospital admissions data three days in advance.

Going through the motions...

The process is very simple. A small sample of raw sewage is analysed using RT-qPCR in a laboratory. It’s the same way throat swabs are analysed currently. The test doesn’t detect the virus itself – the chances of intact virions surviving in sewage are pretty much zero – the test is looking for the molecular fingerprint of COVID – genetic material specific to the virus.

We already knew wastewater in Milan contained viral RNA at the same time community transmission of COVID-19 was confirmed and similar findings were made in Valencia. But this study brings a bit more to the party.

Is this a gamechanger?

It could be. The cool thing is the Yale study shows predictive value. I’ve already looked at the biology of relaxing ‘lockdowns’ and the likelihood further will be needed on a regional / local / community level as cases rise and fall.

This sort of testing could proactively inform decisions to implement or relax public health interventions by tracking outbreak dynamics at a community level – and far more accurately than the ‘apps’ currently being trialled.

[And I can’t help but observe that many people made political comments on that post about the biology of lockdowns without realising viruses really don’t give a rat’s arse about your politics, you’re just another host, and also that those furiously Tweeting about privacy issues arising from being tracked by an app are mostly doing it from the tracking devices they’ve carried in their pockets for bleedin’ years…]

Can COVID spread via poop?

The primary route of transmission is respiratory droplets and close person-to-person contact.

Yes, there have been reports of GI symptoms. Plus SARS-CoV-2 uses ACE2 as a receptor, and ACE2 mRNA is highly expressed in the GI tract. But the precise mechanisms by which SARS-CoV-2 interacts with the GI tract remain occult.

There have been a handful of reports of viable virus being recovered from patient stool samples and more showing recovery of viral RNA (BIG difference) – but these reports are based on small numbers, prevalence in stool specimens has varied widely and there are methodological inconsistencies. Which is unsurprising – these are case reports of patients treated on the frontline during a pandemic and not formally designed research studies.

Despite looking quite hard there’s no real evidence this virus spreads via the faeco-oral (‘turd to tongue’) route. And the notion it could remain viable in sewage and get into potable water really is a non-starter.