select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to study looking at clinically relevant levels of potentially active virus 10 days after a positive PCR test

A study published in International Journal of Infectious Diseases looks at persistence of clinically relevant levels of potentially infectious SARS-CoV-2 past the 10 day quarantine period.


Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, Respiratory Sciences, University of Leicester, said:

“This study just confirms what we knew already – using an alternative method – that there is often a long tail of virus shedding – which can extend beyond 10 days in a few people (at least 10%). 

“Also, its findings really apply to the original Wuhan virus, and the subsequently more transmissible alpha, delta and omicron variants likely result in more than 10% of infections remaining infectious after 10 days.

“Viral culture, as used by some of the studies cited, are not sensitive for virus detection, i.e. you need a certain minimum number of viable viruses to be detected in culture (using various means) and there may still be enough viable virus being shed be aerosol to cause infection in others, which are not sufficient to be detectable in standard virus culture systems – where the cell-line used will also impact on this sensitivity.

“Nevertheless, the main message is that any minimum duration of isolation (and it has just been reduced to 5 days in the UK) will pose a risk of allowing virus shedding to continue, albeit it at progressively decreasing viral loads. 

“But now with a well-vaccinated population, those lower levels of virus may be neutralised by existing vaccine-induced (or naturally-acquired) antibody responses in most people exposed to it – and pose less of a threat (though not zero) to others.”


Dr Nathalie MacDermott, NIHR Academic Clinical Lecturer, King’s College London, said:

This paper describes a test that identifies subgenomic RNA using a PCR technique. Subgenomic RNA is produced by virus that is actively replicating and therefore can be considered to be live and potentially infectious virus. Standard PCR techniques are very accurate in identifying SARS-COV2 RNA, however this RNA can be identified in both those with live, replicating virus and those with non-replicating viral remnants, making it difficult to distinguish who may continue to be infectious in individuals who continue to test PCR positive for some time after their first diagnostic PCR result. The ability to identify replicating and potentially infectious virus would be very useful in determining whether it is safe for individuals to end their isolation period, particularly when it comes to those working with vulnerable populations in the health and care sector and those individuals wishing to visit vulnerable family members. While it may be possible to conduct this test and determine infectiousness, the test still requires laboratory facilities to undertake a PCR test and so is limited by the capacity of labs to undertake additional PCR tests when they are already overwhelmed by testing samples purely for identifying infection. Should it be possible to develop this test into a form of rapid test similar to the lateral flow tests currently used by households around the country, then this could be a game changer in terms of determining safely when students and workers may end their isolation period following infection with COVID-19. This is particularly relevant in relation to the government’s decision to reduce the isolation period to 5 days, despite data from the UKHSA demonstrating that reducing the isolation period to 5 days followed by 5 days of daily lateral flow tests would still result in 15% of those released from self-isolation being infectious.”


Please note this comment is from an author of the work – not a third party comment: Prof Lorna Harries, Professor of Molecular Genetics at the University of Exeter Medical School, said:

“Our data suggest that after five days, about one in three people could potentially remain infectious for COVID-19; a figure that is backed up by the UK Health and Safety Authority’s own data.

“Our data arise from research published today (January 14) which uses a newly adapted test which can detect whether the virus is potentially still active. We applied that test to samples from 176 people in Exeter who had tested positive on standard PCR tests. Our data shows that at five days, 30 per cent of people still exhibited clinically-relevant levels of potentially active virus, meaning they may still be infectious.

“Deciding an isolation period is obviously a delicate balance between mitigating transmission and keeping society functional. However, this new move underlines the critical need to encourage people do those lateral flow tests, and abide by the outcome, to help mitigate the risk. Our government urgently need to ensure adequate supplies of lateral flow tests are available. Special care should be taken in vulnerable settings such as care homes, as older people with perhaps waning vaccine immunity may be much more at risk. A test that can detect specifically active virus such as ours may be useful in these specialised cases.”


Prof Paul Hunter, Professor in Medicine, UEA, said:

“This paper by Professor Harries’ group reports on the detection of SARS-CoV2 envelope gene subgenomic RNA for some time after onset and as suggested in that because these can be detected over time that people may remain infectious for longer than may be thought. However, as stated by the authors “the correlation between infectivity and E gene-derived sgRNA positivity is not always well conserved” or in other words even though people remain positive they may not still be infectious. Alexandersen et al.1 in particular suggest that subgenomic RNAs are not an indicator of viral replication.

“Whatever the merits for or against  subgenomic RNAs being markers of viral replication that does not necessarily make this a useful indicator of infectiousness. All of the samples come from 2020 so in the early days of the pandemic and most will predate the alpha variant. We know from early studies that virus can be shed from as early as 4 days before to several weeks after symptom onset2. But that doesn’t necessarily equate to infectiousness. Very few studies have attempted to measure when cases infect others. One such study3 suggested that most infections were transmitted between 2 days before to 3 days after symptom onset. This would suggest that relying on subgenomic RNA would over estimate risk of transmission after this period. Also this work relates to infections pre-omicron. There is some evidence that omicron reaches peak viral shedding two or three days later than previous variants, but as yet no really good epidemiological evidence on when patients with omicron are most likely to infect others.

“Given the above I personally do not think this work, though interesting, will add much to our knowledge about when cases of covid are particularly infectious.

“In any event viral load, even viral load of potentially infectious virus is not a perfect estimate of infectiousness. Even people who are shedding potentially infectious virus are not necessarily going to infect others. How much a risk someone poses to others depends on the viral load being shed, the symptoms of the patient (ie. Whether they are coughing and sneezing) and their context and levels of immunity in potential contacts. The value of isolating potentially infectious individuals also depends to a large extent on how much infection is circulating in the community. At present we are probably only detecting about 1/3 of all infections so strict isolation policies, though reducing the risk of transmission somewhat, are unlikely to make a big difference to the trajectory of the UK epidemic going forward especially with the increased transmissibility of omicron.”  


1 – ‘SARS-CoV-2 genomic and subgenomic RNAs in diagnostic samples are not an indicator of active replication’ by Soren Alexandersen et al. Was published in Nature Communications on 27 November 2020

2 –‘SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis’ by Muge Cevik et al. was published in The Lancet Microbe on 1 January 2021

3 – ‘COVID-19 Transmission Dynamics Among Close Contacts of Index Patients With COVID-19, A Population-Based Cohort Study in Zhejiang Province, China’ by Yang Ge et al. was published in JAMA Internal Medicine on 23 August 2021


Prof Lawrence Young, Virologist and Professor of Molecular Oncology, Warwick Medical School, University of Warwick, said:

“This study uses a modified version of the PCR test (the detection of subgenomic viral RNAs so-called E sgRNA) as a proxy for the detection of active virus replication. Using this technique in swab samples from symptomatic patients taken over 8 months in 2020, the study shows that in a small selected group of individuals (32 from 72 who tested positive for E sgRNA), 4 of the 32 had high levels of this RNA consistent with on-going virus replication for more than 10 days. This means that these individuals are very likely to have remained contagious beyond 10 days and could have transmitted the virus to others.

“While this is a small study and the modified PCR used is not a definitive test for infectious virus, it does confirm previous observations that the infectious period varies from person-to-person and that some individuals can continue to be infectious for long periods. A recent study from Japan of 21 hospitalised cases with omicron infection has shown that peak infectiousness occurs at 3 to 6 days after diagnosis or symptom onset but found that no infectious virus was detected after 10 days in these vaccinated individuals. The current study was based on infections analysed in samples taken in 2020 when the original Wuhan variant was dominant and it’s likely that the infectious period with omicron will be different. Recent analysis from UKHSA suggests that the window between infection and infectiousness (the incubation period) may be shorter for omicron as compared to delta and this could mean that the infectious period is also shorter. Nevertheless, this current study reinforces concerns that reducing the self-isolation period to 5 days will increase the risk of highly infectious people spreading infection as they return to work or school.

“I don’t think that this test will be useful for everyday use. It’s a proxy for detection of virus replication and is not always an accurate reflection of infectivity. In this regard, lateral flow tests offer a much easier way of detecting infectious virus and, although not 100% accurate, are extremely good at identifying the majority of contagious individuals particularly when they are at their most infectious.”



‘Persistence of clinically-relevant levels of SARS-CoV2 envelope gene subgenomic RNAs in non-immunocompromised individuals.’ by Merlin Davies et al. was published in International Journal of Infectious Diseases at 4:00 UK time on Friday 14 January.



Declared interests

None received.

in this section

filter RoundUps by year

search by tag