Hacker Read top | best | new | newcomments | leaders | about | bookmarklet login
Covid-19, duration of viral shedding, and infectiousness: meta-analysis (www.thelancet.com) similar stories update story
97.0 points by Manheim | karma 1163 | avg karma 6.65 2020-11-23 06:58:15+00:00 | hide | past | favorite | 76 comments



view as:

> Maximum shedding duration was 83 days in the upper respiratory tract

So covid-19 is transmissible for up to 83 days once infected?


Their interpretation does not make it sound like this is the case:

"Although SARS-CoV-2 RNA shedding in respiratory and stool samples can be prolonged, duration of viable virus is relatively short-lived. SARS-CoV-2 titres in the upper respiratory tract peak in the first week of illness. Early case finding and isolation, and public education on the spectrum of illness and period of infectiousness are key to the effective containment of SARS-CoV-2."


But:

> no live virus was isolated from culture beyond day 9 of symptoms despite persistently high viral RNA loads.

> This finding is supported by several studies demonstrating an association between viral load and viability of virus, with no successful culture from samples below a certain viral load threshold.

So the amount shed is too low to infect anyone beyond 9-10 days (I think, I am not a virologist, etc.)


>So the amount shed is too low to infect anyone beyond 9-10 days (I think, I am not a virologist, etc.)

Is it possible to acquire immunity against the virus (in a similar way that vaccines work) by being in contact with someone at say day 20+?


If it would be possible, pandemic would be over months ago and there would be no need to develop any vaccine.

It could be possible, just unlikely enough that immunity doesn't spread fast enough.

Sounds reasonable - but I'd still prefer to stay away from people who are shedding the virus.

There is also a similar theory that cloth masks might have the unintended positive side effect of building immunity. They are not perfect, so the virus might still reach its target. But the viral load could be reduced enough that instead of infection it leads to immunity.


PCR tests detect fragments of a virus, not necessarily an active/viable virus.

If I'm reading this right, the RNA of SARS-CoV-2 can be found in the upper respiratory tract for a very long time (even 83 days in the extreme case), but the live virus (capable of infecting others) was not discovered by any study beyond day 9.

Self-isolating/quarantining for 10 days after a positive test result, which many regions enforce, seems consistent with these findings (again, if I'm interpreting the results correctly).


I agree but the key is to get test results quickly. A five day delay in getting test results (which is common) seems to imply that an infectious patient may have already done their worst spreading by the time they start quarantining.

I agree accessibility and speed of testing is critical.

Unless they quarantine preemptively from the time they're tested.

In the UK, if you have a test, you are required to quarantine until you get the results. Is it not the case elsewhere??

In Australia that's the standard response too - if you're symptomatic (of anything vaguely resembling the wide range of COVID19 symptoms) then you a) isolate, and b) get a test, and c) isolate at least until the test results come back.

It's probably the rules everywhere, but I imagine that compliance is a lot lower for people waiting for a test result than it is for people that received a positive test (especially those with mild symptoms)

In the US it sometimes takes up to two weeks to get a result, in which case the test is pointless.

Lest readers assume this is typical, where I live at present (middle of the US) testing is efficient. Our kids' school runs their own rapid tests also. You can in addition buy saliva PCR tests from many vendors including Costco. These take 3 days max from purchase to result.

Right. I had to get a Covid-19 test in October as part of groundwork for another procedure. I could've gotten a rapid test that gave results same day, or the PCR test. I got the PCR test (because more accurate) and had results back within 2 days.

In my town, if a child receives a positive test result, the whole classroom is under quarantine.

So it's not pointless in that sense. You know who might have been exposed.

It's also not pointless for monitoring spread.


Not here in Poland.

Though, it would very logical as a precaution.


YES in Poland.

I know people in Poland who avoided getting a COVID19 test because they knew they would have to quarantine if they took the test, and would not have to quarantine if they didn't. That's perverse incentives for ya'.

It's very hard to enforce, of course, but if you get caught, there are (in abstract) severe fines.


I'd hesitate to make it a rule because some people take them proactively, or even routinely. But yeah if someone has symptoms, or think they were exposed, that would make sense.

In the US we don't really have "rules" so much as "suggestions"

Not in .il, because of mass-testing efforts and people needing to show a negative result before travel and having no reason to believe they are positive.

Quarantine is mandatory if you had exposure or returned from abroad.


I don't get this - if you're symptomatic beyond day 9, how can you not be contagious? A dry cough, for example, should be caused by the virus still being alive in the upper respiratory tract, i.e. you still should be able to infect others. Did they test aerosols, too?

> A dry cough, for example, should be caused by the virus still being alive in the upper respiratory tract

Why? I thought the body’s (potentially delayed) reaction caused the cough, not the virus


We have the instinct to cough to clear our air passages even without pathogens in certain cases. The virus is triggering these thru some irritation in order to spread to other people.

Right, but whatever process causes the coughing could well outlive the virus itself

Bother to provide any proof?


Naive guess: the damage was already done. Cough is a response to the damage caused by the virus, not to the virus itself.

If you get shot, you still have a bullet wound when the bullet comes out.


Damaged cells are cleaned up pretty fast. Of course, it's the virus causing damage, but we cough for the purpose to clear up the passages. In addition, the respiratory viruses probably trigger the cough as a vehicle to spread to other people.

What you feel as symptoms are your immune responses, you can not "feel" the actual virus. The immune reponse should go on for longer than the virus is functional.

Also, there has to be above a certain viral load to be contagious. I am however far from qualified to even guess what the thresholds are and I would think that it is mostly a matter of definition.


Coughing is a taxing activity - it wastes energy, and causes discomfort. We cough for a reason - I doubt it's from the solidarity to infect others. Of course we can sense the virus, but the irritation it causes. Do all viruses, which replicate in the nasopharyngeal tract, trigger coughs? Maybe some viruses evolve to cause irritation similar to air passage obstruction, which triggers the instrict to clear your throat up?

Our bodies are not perfect or omniscient. We cough because it is evolutionarilly favored as a way to dislodge irritants, and viruses are evolutionarily favored to cause that coughing response strongly in order to spread. Coughing too long seems like a very possible consequence of those.

The irritation still persists from the damaged tissue after the virus is eliminated. It takes time to rebuild.

So, senescent cells are just left for days to build up damage? We have mechanisms to clear up dead cells, many of them which actually killed themselves via apoptosis and were not directly killed by the virus.

Live virus doesn't really cause symptoms as such, although they evolve to take advantage of it. The virus is basically re-purposing then melting a bunch of cells. The body's immune response is killing the viral structures and turning whole areas into the cellular equivalent of a battleground, killing off anything that is or is likely to be controlled by the virus.

Many of the worst symptoms aren't related to the presence or absence of the virus itself. If someone is turning blue from lack of oxygen, it isn't that the virus is doing anything here and now. It is that their lungs are non-functional from damage that the virus already did (note: past tense). You could remove every trace of active coronavirus from them and while they'd probably not get worse they also wouldn't get better very quickly. New lungs don't grow overnight.


This is not true. Viruses can directly kill the cells, by interfering with biological processes, necessary for cellular survival.

The clarification is helpful, but everything said was in fact true. Omitting this clarification doesn't negate everything else they said.

What the poster said above is incorrect. Covid does produce damage directly to the lung and neural cells, which especially evident as loss of smell and taste. The severe, deadly reaction is indeed of the immune system.

That's what I saw in many explainer videos, too.

Are also the loss of smell and messed up taste effects of the immune system? I caught Covid at the end of October, and those symptoms lasted well beyond the final negative molecular swab test, so it seems the case.

PCR tests for RNA fragments. After 9 days it's likely you are surveying the battle field remains not actual live virus.

So help me understand. What about cases where somebody gets covid, goes to the hospital on day 15, respirator day 20, dies day 25.

Was somebody in that profile only contagious up until day 9?


They say yes.

Correct. Most of the “long Covid” effects appear to be more related to immune dysregulation as a response to the virus, causing the so-called cytokine storm and leading to damage in the lungs, heart, vasculature, etc. (Lots of support to this, I’ll look for a good review and update this post with an edit.)

Edit: https://www.medrxiv.org/content/10.1101/2020.05.27.20115238v...

Check out Figure 1 and the Discussion section for a good overview.


I think that everything you said regarding the "Long Covid" is correct, but I don't think that it necessarily follows that patients dying of covid ARDS have ended viral replication and are no longer contagious.

From what I understand the actual 'viral' phase of the disease isn't usually that long so potentially they were only contagious up to day 9. After that their immune system has usually wiped out the active virus.

Everyone is different but by the time people are in the hospital on respirators they're usually suffering from the follow on effects of either direct viral damage or their own immune system going nuts.


So why don't they put covid patients on a bucket of immunosuppressants after day 9?

> data on the shedding of infectious virus in asymptomatic individuals are too scarce to quantify their transmission potential in order to inform policy on quarantine duration in the absence of testing.

This has long been a key question: how much of community transmission is due to truly asymptotic individuals? Michael Mina style Lick-a-Stick daily tests could answer this question and remove the potential errors surrounding symptom self-assessment.


That question was answered in a study published last week.

"There were no positive tests amongst 1,174 close contacts of asymptomatic cases."

https://www.nature.com/articles/s41467-020-19802-w


Interesting paper but I'm not convinced that these are asymptomatic cases. I think these are primarily recovered cases that still have trace amounts of non-infectious viral RNA.

The originally linked Lancet paper makes a distinction between "viral shedding" and "viral RNA shedding". I think the unanswered question involves individuals in the first week of their newly acquired infection (EDIT: after the incubation period). I hope that doesn't come across as nit picking; the scale and scope of the Wuhan study is impressive but an equivalent study is needed in an area with ongoing community transmission.


Do tests test for the RNA or the live virus?

The PCR test amplifies RNA fragments. If after a certain amount of replication cycles the amount of detected RNA crosses a threshold, it is considered a positive.

A positive PCR test does not imply infectiousness. Many people who have a positive test are not just asymptomatic, they are not really infected to begin with.

A negative PCR test does not rule out disease. Some people who have a clinical diagnosis of COVID may produce a negative PCR result.


Can this be interpreted as most people who test positive for COVID-19 are not infectious anymore?

I don't think they drew that conclusion as it is probably difficult to say in which phase of the illness people get tested. Some have mild (or no) symptoms and probably go through their active 9-day period without getting tested at all. Others can have severe symptoms from an early onset, and probably get tested quite fast (hopefully). Some may get severe symptoms after an incubation period, only to get ill when their active 9-day period was over.

It should mean that a 10-day isolation for positive cases, combined with high frequency testing should have a massive impact on the spread of the virus. That the virus can be found in your body much much later, in an "inactive" state should not have much impact on the spread.


"No study detected live virus beyond day 9 of illness, despite persistently high viral loads," - is it possible that IgA antibodies in the mucosa neutralizes the virus even before it gets spread?

That could also explain the strange heterogenity of the virus spreading - maybe some people have more IgA or have IgA earlier and don't really spread a live virus?


I know enough about viruses to know that they're not alive by any meaningful definition, but not enough to understand how this use of 'live virus' is used in these kinds of studies.

I'm assuming it means a full, viable string of RNA, rather than (presumably harmless) fragments of same -- and the fragments we're finding being described here are definitely non-viable. Is that roughly right?


“Live” in this context seems to mean “active”.

Live means that it is readable by the ribosome.

Whether viruses are alive or not is purely a matter of definition.

Alive in this context means that it can infect host cells and replicate.

Contrary to what the sibling posts say, that doesn't just require intact viral RNA, which by itself is not infectious, but an intact viral assembly, i.e in this case a lipid hull with various accessory proteins, such as the nucleocapsid that packs the RNA and the spike protein that binds to host cells.


There many studies contradicting this. I need to find the links gain (check BNO reinfection tracker, it is one the most recent), but a recent Korean study showed a case of reinfection (or as I am often told to be a "relapse") when a healthy women had shown symptoms again 10 days after recovery, although genetically different. After that in 10 days (or similar time period, do not remember) she showed the symptoms again, third time, with the virus of the first serotype. Also, immunocompromised people keep shedding viruses for months, not only that, the viruses also mutate very rapidly in them.

This study says live virus has only been confirmed to be seen for at most 9 days. However, (presumably inactive) virus particles can be shed for months afterwards and can show up as PCR positive.

This meta analysis only includes studies of multi patients so there's pretty much a guarantee that there are individual case studies that differ from this.


> This study says live virus has only been confirmed to be seen for at most 9 days. However, (presumably inactive) virus particles can be shed for months afterwards and can show up as PCR positive.

The paper cites 34 replication cycles as the lower bound to obtain viable viral cultures. Whether a PCR test is reported as "positive" depends on the chosen replication cycle threshold (Ct value). As far as I'm aware, this "magic number" is usually chosen to be 35.

Beyond that, it is possible to do more cycles to demonstrate the existence of trace amounts of PCR fragments, but such a test would not be considered "PCR positive".


The issue is that a PCR test does not actually tell us if a virus particle is live or inactive. A shed virus particle can still have intact RNA for the region tested by PCR.

There are published case reports where someone has tested PCR positive during an infection, then PCR negative, and then PCR positive again later and the explanation given has been that they did not get re-infected and their infection did not flare back up but just that virus had been continually shed and the later test was able to capture that.

edit: s/DNA/RNA. The PCR process actually operates on DNA so the virus RNA is converted to DNA as the first step of RT-PCR.


Covid is not a DNA virus.

The PCR thresholds are chosen to minimize false negatives, not false positives. A sample that required 30+ replication cycles is unlikely to come from a person that is infected/infectious, yet it would be counted as "positive". The thresholds are derived from experiments under lab conditions, providing a safe lower bound.

That does put into question the validity of counting PCR test positives as a benchmark of viral load among the population, but as a safety measure for potentially infected individuals, the trade-off is worthwhile.


Yes I understand this, and I also believe that meta-analysis needs to be taken with a grain of salt, because it is never as reliable as an original research. In the case of pandemics of a dangerous and highly contagious disease, even individual cases, such as above are important. This is the reason why Asia is so successful in the pandemic management, and the west is a pathetic failure.

It’s good to know that we can now file assault charges on those maskless mouth breathers.

https://www.cbsnews.com/news/trump-backer-charged-with-assau...


How did they test if virus is 'live' (able to infect)?

In cell cultures.

To clarify, if I'm reading this correctly, when they say "day 9" they're referring to day 9 of illness, not day 9 after exposure.

Does anybody know what the data says about how long to wait for symptoms after being in contact with somebody who has it?


^^^THIS. I have a friend that believes that if everyone invited to an indoor house party goes and gets a test, the test comes back negative, and if they have quarantined since the test, that they can then party in the same room together with zero risk.

Legal | privacy