2 Nov Episode 80 Saliva and rapid tests

Mon, 11/02/2020 - 21:54


  1. Saliva was found a reliable substrate for RT-PCR,  as compared to more invasive sampling such as naso-, or oro-pharyngeal (NP OP) sampling
  • First on various respiratory viruses, already in 2017  (Ref Saliva 1-2)
  • This was also confirmed for SARS-CoV-2 in a review of March 2020 (Ref Saliva 3-4).  In most cases symptomatic COVID patients were tested. 
  • A study that compared symptomatic and asymptomatic subjects concluded: Our results show that saliva samples provide relevant, reliable data, especially for asymptomatic individuals and symptomatic patients detected early, although the virus loads in saliva are lower than in NP swabs. The rare discrepancies between the saliva and NP samples occurred in low virus load patients, who were, perhaps, not very contagious (Ref Saliva 5).
  • A study on of saliva viral dynamics (Ref Saliva 6) concluded:
    • Comparable (slightly lower) performance, relative to NP/OP
    • Salivary viral load peaked during the first week of symptoms and gradually declined over time.
    • Surprisingly, there was no significant difference regarding the temporal viral load profile between mild and severe cases
  • In addition, a simplified RT-PCR protocol for saliva handling, requiring no additives or transport media, has been proposed by Moreno-Contreras, with excellent sensitivity as compared to NP and OP sampling  in a population of ambulatory patients has recently been proposed (Ref Saliva 7)


  1. Rapid tests show increasing sensitivity, but in most cases nasal, NP or OP sampling is recommended and sensitivity remains lower than RT-PCR.  It is unfortunately usually not always clear how representative the investigated patients are for the spectrum of COVID disease.
  • An early test (BIOCREDIT) was found to be very insensitive (Ref RDT 1).
  • The early systems used visual readings.  The later ones fluorometric readings.  
  • On 7 Aug two rapid tests were approved by the NH Division of Public health Services: Quidel Sofia and BD Veritor.  The latter is said to be less sensitive (Ref RDT2).
  • On 4 Sept a comparative study is published in medRxiv on over 250 NP or OP samples (Ref RDT3) :
    • The BD Veritor showed a positive percentage agreement (PPA) between 81.8%-87.5%  with RT-PCR
    • The BD Veritor was very close to Sofia as well  (agreement over 97%)
  • End of August two naked eye systems from Abbot were also approved (Ref RT4 and 5). According to the leaflet, NP swab should be used and the reading can be done with naked eye with a sensitivity of  91.4% (94.1% for samples with Ct values ≤33).    
  • The Japanese company Fujirebio has two types of antigen test: Espline for direct reading on NP swabs https://www.fujirebio.com/en/products-solutions/espliner-sarscov2 and Lumipulse for instrument mediated reading on NP swab or saliva https://www.fujirebio.com/en/products-solutions/lumipulse-g-sars-cov2-ag
  • The former (Espline) is clearly not suitable for testing on saliva, as the sensitivity was only 11.7 % as compared to molecular testing (Ref RDT 6).
  • The latter Lumipulse on NP samples exhibited 55.2% sensitivity and 99.6% specificity, as compared to RT-PCR (Ref RDT 7).
  • Despite this questionable performance,  Japanese authorities have chosen Lumipulse® for nasopharyngeal or saliva-based SARS-CoV-2 antigen testing of passengers at their main international airports (Ref RDT 8).


Conclusion: The fields of rapid antigen testing and using saliva as a convenient sample are evolving quickly, but until now, largely separately. From the literature that I could find, it is not yet clear whether both can be combined without too much loss of sensitivity.  But it is evident that there is an urgent need for rapid tests by preference on saliva, even if some patients with low viral loads would be missed.  Presumably, this will require instrumental reading.  At present, the technology is not yet enough advanced for home-based testing, without supervision by a trained health care worker.