Does sampling saliva increase detection of SARS-CoV-2 by RT-PCR? Comparing saliva with oro-nasopharyngeal swabs

Does sampling saliva increase detection of SARS-CoV-2 by RT-PCR?

Comparing saliva with oro-nasopharyngeal swabs

Ozlem Akgun Dogan

1,2

, Betsi Kose

, Nihat Bugra Agaoglu

1,3

, Jale Yildiz

1,5

, Gizem Alkurt

1,5

Yasemin Kendir Demirkol

1,2

, Arzu Irvem

, Gizem Dinler Doganay

1,5

, Levent Doğanay

Genomic Laboratory (GLAB), Umraniye Teaching and Research Hospital, University of Health

Sciences, Istanbul, Turkey.

Department of Paediatric Genetics, Umraniye Teaching and Research Hospital, University of Health

Sciences, Istanbul, Turkey.

Department of Medical Genetics, Umraniye Teaching and Research Hospital, University of Health

Sciences, Istanbul, Turkey.

Department of Microbiology, Umraniye Teaching and Research Hospital, University of Health

Sciences, Istanbul, Turkey.

Department of Molecular Biology and Genetics, Istanbul Technical University, Istanbul, Turkey.

Abstract Word Count: 250

Manuscript Word Count: 1988

*Correspondence address: Levent Doganay, MD Genomik Laboratuvar (GLAB), Ümraniye Eğitim

ve Araştırma Hastanesi, Elmalıkent Mah, Adem Yavuz Caddesi, No:1, 34760 Umraniye, Istanbul,

Turkey

Phone: +902166321818 – ext.1846

Fax: +90 216 6327111

E-mail: levent.dog[email protected].tr

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Abstract:

The gold standard method in the diagnosis of SARS-CoV-2 infection is the detection of viral RNA in

nasopharyngeal sample by RT-PCR. Recently, saliva samples has been suggested as an alternative due

to being fast, reliable and non-invasive, rather than nasopharyngeal samples. We compared RT-PCR

results in nasopharyngeal, oro-nasopharyngeal and saliva samples of COVID-19 patients. 98 of 200

patients were positive in RT-PCR analysis performed before the hospitalization. In day 0, at least one

sample was positive in 67% of 98 patients. Positivity rate was 83% for both oro-nasopharyngeal and

nasopharyngeal samples, while it was 63% for saliva samples (p<0.001). On day 5, RT-PCR was

performed in 59 patients, 34% had at least one positive result. The positivity rate was 55% for saliva

and nasopharyngeal samples, while it was 60% for oro-nasopharyngeal samples. Our study shows that

the sampling saliva does not increase the sensitivity of RT-PCR tests at early stages of infection.

However, on 5th day, viral RNA detection rates in saliva were similar to nasopharyngeal and oro-

nasopharyngeal samples. In conclusion, we suggest that, in patients receiving treatment, virus presence

in saliva, in addition to the standard samples, is important to determine the isolation period and to control

the transmission.

Keywords: SARS-CoV-2; coronavirus; RT-PCR; saliva

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Introduction:

The SARS CoV-2 infection, which has influenced the world since the end of 2019, causes serious

problems in both health and socio-economic fields.

1,2

Currently, it has affected approximately 6 million

people over more than 200 countries.

SARS-CoV-2 infection can cause serious lower respiratory tract

infections that can be fatal in some patients, however, many individuals remain asymptomatic during

the infection.

4,5

Especially asymptomatic individuals have been a major factor in increasing the

transmission rate of the disease and evolving it into a pandemic.

COVID-19 diagnosis is based on clinical findings, besides, the detection of the virus in patients'

specimens is of great importance in terms of monitoring the disease, guidance of treatment, and infection

control,

Quarantine, which is launched all over the world urging people to stay home, is considered as

the only way to reduce transmission, but it started to cause serious social and economic problems due

to loss of labor.

7,8

This situation reveals the need for a fast, reliable, easily applicable, and non-invasive

test that quickly identifies infected individuals to be isolated. Currently, gold standard method in the

diagnosis of SARS-CoV-2 infection is the detection of viral RNA in the nasopharyngeal swab sample

by Real Time Polymerase Chain Reaction (RT-PCR) analysis.

The most important disadvantage of this

method is the presence of limited trained personnel available in sampling during the outbreak. High risk

of nosocomial infections that such personnel are exposed to is an additional obstacle. Moreover, in terms

of patients, discomfort experienced, especially, in repeated tests is the most frequently reported problem.

The abovementioned disadvantages have led the researchers to study on non-invasive, easy to self-

applicable sampling methods for massive screening.

10-14

As the main source of transmission of SARS-

CoV-2 infection is salivary droplets, viral RNA RT-PCR in saliva samples have been suggested as

possible alternative testing for diagnosis.

The main advantages of saliva sampling are self-availability,

no need for specialized staff and the comfort of the procedure. However, although turnaround time and

self-applicability of the tests are important, the reliability of the sampling method to be chosen should

also be tested very well. It is obvious that the use of tests with a high false negativity rate will adversely

affect the course of the pandemic.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Here, we compared the results of RT-PCR in nasopharyngeal, oro-nasopharyngeal, and saliva samples

in patients diagnosed with COVID-19 to investigate their possible relationships with clinical findings.

Methods:

Patients:

A cross-sectional study was conducted in repurposed Genomic Laboratory (GLAB), Umraniye

Teaching and Research Hospital, in Istanbul, with a total of 200 consecutive patients who met the

possible case definition for COVID-19 and hospitalized with moderate-severe disease.

According to

the diagnostic algorithm provided by The Turkish Ministry of Health, the possible cases were defined

as those who presented with;

*History of fever or acute respiratory symptoms,

and

*Travel history from an endemic area of COVID-19 within 14 days,

*History of contact with an individual who was confirmed or suspected having COVID-19,

*Presence of hospitalization requirement due to respiratory tract infection.

Within the scope of the present study, after hospitalization saliva, oro-nasopharyngeal and

nasopharyngeal samples were taken from all 200 patients within the first 24 hours, and it was defined

as day 0 sample. On day 5, patients were resampled.

Demographic characteristics, symptoms at presentation, comorbid diseases, and clinical findings during

hospitalization were collected for each patient. All subjects provided informed consent, and the study

was approved by the ethics committee of Umraniye Teaching and Research Hospital

(B.10.1.THK.4.34.H.GP.0.01/167)

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Patients admitted to ICU, not giving consent to study, incapable of providing saliva sample and patients

under age of 18 were excluded.

Sample Collection

In all patients, an oro-nasopharyngeal sample was taken with a cotton swab used for the viral RT-PCR

test before hospitalization as a standard diagnostic approach. In the scope of our study, oro-

nasopharyngeal samples were taken with cotton swab and nasopharyngeal samples with dacron swab.

Details of the sampling processes are given in Figure 1. Before the saliva collection, participants were

given brief explanations about the difference between saliva and sputum, then they were asked to give

saliva samples prior to other samples by a drooling technique. They spit approximately 1 mL into the

falcon tubes containing the viral transport medium (VTM, Innomed VTM001) used in standard

sampling. Single trained healthcare professional took samples to avoid possible variations in the

collection technique. All samples were transferred to our laboratory within 1 hour of sampling stored in

the refrigerator, and RT-PCR was performed on the day they were collected.

RT-PCR Workflow:

ORF1ab and N gene of SARS-CoV-2 were targeted for the diagnosis of the infected patients. For this

purpose, The Direct Detection of SARS-CoV-2 Detection Kit was used (Coyote Bioscience Co., Ltd).

The kit procedure was based on the detection of the conserved region of ORF1ab coated with a pair of

specific primers and a fluorescently labelled probe, and the N gene of SARS-CoV-2 by the RT-PCR

method. Since this kit did not need a separate RNA extraction, the samples in the VTM medium had

briefly vortexed and taken directly into the study. Biorad CFX 96 Real-Time PCR systems were used.

FAM channel for ORFlab gene, ROX channel for N gene, and HEX channel for internal RNase P gene

of human control were selected. These channels should have the logarithmic growth period with the Ct

value ≤29, for a positive result.

Interpretation of Results:

In the RT-PCR results of the samples, the internal RNase P gene was positive for all samples. If both

the ORFlab and the N gene were positive, the result was considered as presumptive positive. If both the

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

ORFlab and the N gene were negative, the result was considered as presumptive negative. If one of them

was positive and the other was negative, the test for this sample was repeated. If the same result was

achieved again, a new nasopharyngeal and oropharyngeal swab was requested.

Statistical Analysis:

SPSS.22 package program (IBM Company, Armonk, NY, USA) was used for statistical analysis. Mean,

median and standard deviation were used for descriptive statistical information. Categorical variables

were analysed with chi square test. The results of the tests performed in three different samples on the

0th and 5th days were compared among themselves using the Cohran Q test. All calculated P values are

double sided and for significant statistical results, p <0.05 was accepted.

Results:

98 of 200 (49%) patients in the study group were positive in RT-PCR analysis performed on the samples

taken as a standard diagnostic procedure before the hospitalization. The clinical and demographic

characteristics of the patients with positive results are presented in Table 1. Although 102 patients met

the possible case definition, the viral RT-PCR tests on admission were negative (Figure 2).

At day 0 RT-PCR analysis, at least one sampling method showed positivity in 66 of the above mentioned

98 patients. The sensitivity rate was observed as 55/66 (83%) for both oro-nasopharyngeal and

nasopharyngeal samples, while it was 35/66 (63%) for saliva samples (Table 2) and the difference of

the sensitivity rates among the sampling methods was statistically significant (p< 0.001) (Figure 3a)

The mean Ct values determined for FAM and ROX were shown in Figure 4.

On day 0, 6 (9%) patients had positive result only in nasopharyngeal swab sample, 5 (8%) was only

positive in oro-nasopharyngeal sample and additional 3 (5%) patients had a positive test results only in

saliva samples. Those three patients who had only saliva positivity had a history of COVID-19 positive

household contact and had mild involvement in thorax CT.

In the RT-PCR analysis on day 0, of 102 patients with a previous negative result, 95 remained as

negative, while in 7 (6·9%) patients positive RT-PCR results obtained in at least one sample type (Figure

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

2). While the oro-nasopharyngeal sample was positive in all 7 patients, nasopharyngeal samples were

positive in 3. There was only one patient whose saliva sample was positive. Oro-nasopharyngeal and

nasopharyngeal samples in this patient were also positive.

On day 5, we were able to collect samples in 59 of 200 (30%) patients. Of 59 patients, 20 (34%) had at

least one positive result in saliva or in one of the swab samples. The sensitivity rate was determined as

11/20 (55%) for both saliva and nasopharyngeal samples, while it was 12/20 (60%) for oro-

nasopharyngeal samples (Figure 3b). In the statistical analysis, there was no difference between the

sample types in terms of the sensitivity (p=0.368). The mean Ct values determined for FAM and ROX

were shown in Figure 4. On day 5, there were 5 patients who showed RT-PCR positivity only on the

saliva sample. These patients had longer prior history with positive RT-PCR results in oro-

nasopharyngeal samples, home-treated with hydroxychloroquine for five days, clinically deteriorated

with radiological findings, tested positive again in oro-nasopharyngeal samples and hospitalized.

Discussion:

Our study demonstrates that saliva sampling did not improve diagnostic sensitivity in patients who had

a negative result in initial testing before hospitalization. Our results also revealed that at the early stages

of the infection, saliva sampling had a lower sensitivity to detect viral RNA compared to other sampling

methods. In contrast to our findings, three recent studies suggest that the results of RT-PCR analysis in

saliva samples are compatible with the results of the nasopharyngeal and oro-nasopharyngeal

samples

11,13,14

. These studies had either limited number of patients (23 and 12 patients) or limited number

of RT-PCR positive patients (21 RT-PCR positive patients). In this study we consecutively recruited

200 inpatients who presented with clinical signs compatible with COVID-19 and at initial testing 98 of

them revealed SARS-COV-2 RT-PCR positivity. In this study the sensitivity rate for saliva samples was

63% and this was significantly lower than nasopharyngeal and oro-nasopharyngeal swabs. However,

our data on the 5th day showed that the viral RNA detection rates in saliva samples were similar to those

of nasopharyngeal and oro-nasopharyngeal samples. Two recent studies supported our results and

showed that SARS-CoV-2 detection from saliva was more consistent during extended hospitalization

and recovery

11,14

. Thus, we suggest that RT-PCR analysis in saliva samples may be beneficial when a

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

follow up test beyond day 5 is needed especially in hospitalized patients (e.g. before discharge).

Compared to nasopharyngeal swabs, sampling saliva causes less discomfort in patients and reduces risk

for nosocomial infection among healthcare workers as patients can give saliva samples by themselves.

On day five, out of 59 patients, 5 (8%) had detectable viral RNA only in saliva sample. The fact that all

of these patients were using hydroxychloroquine prior to hospitalization, this may have hindered the

detection of the virus in nasopharyngeal and oro-nasopharyngeal samples. With this result, we suggest

that taking saliva samples along with the standard method before ending the isolation in individuals

treated with hydroxychloroquine can be effective in minimizing contamination in the community by

reducing the false negativity rate. In addition, we recommend saliva sampling in patients who show

progression despite at-home treatment. Such an approach would increase the detection of positive cases,

thereby enabling more accurate planning of treatment, follow-up, and subsequent discharge.

One of the most important issues affecting false negative rates in standard nasopharyngeal sampling is

the use of inappropriate techniques. In our study, taking all samples by single trained healthcare staff is

the strength of our study and enabled us to achieve a safer result. The most important limitation of our

study is that no study has been performed in asymptomatic individuals and mild cases that do not require

hospitalization. Therefore, it was not possible to make a comment about whether saliva samples can be

used in screening.

In conclusion, our study shows that the saliva sample is not as sensitive as standard nasopharyngeal

swabs in determining viral RNA and it does not improve detection rate in PCR negative patient group.

However, in the later stage of the disease, RT-PCR test from saliva samples might help detecting

deteriorating patients or determining the isolation period more effectively after treatment.

Acknowledgements: We thank to Murat Kaya for technical assistance

Conflict of Interest: None

Funding: None

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

References

1. Bedford J, Enria D, Giesecke J, et al. COVID-19: towards controlling of a pandemic. Lancet

2020; 395(10229): 1015-8.

2. Rothan HA, Byrareddy SN. The epidemiology and pathogenesis of coronavirus disease

(COVID-19) outbreak. J Autoimmun 2020; 109: 102433.

3. Zollner J, Rompe JD, Eysel P. [Biomechanical properties of synthetic lumbar intervertebral disk

implants]. Zeitschrift fur Orthopadie und ihre Grenzgebiete 2000; 138(5): 459-63.

4. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus

in Wuhan, China. Lancet 2020; 395(10223): 497-506.

5. Guan WJ, Ni ZY, Hu Y, et al. Clinical Characteristics of Coronavirus Disease 2019 in China.

The New England journal of medicine 2020; 382(18): 1708-20.

6. Wang Y, Kang H, Liu X, Tong Z. Combination of RT-qPCR testing and clinical features for

diagnosis of COVID-19 facilitates management of SARS-CoV-2 outbreak. J Med Virol 2020.

7. Remuzzi A, Remuzzi G. COVID-19 and Italy: what next? Lancet 2020; 395(10231): 1225-8.

8. Parmet WE, Sinha MS. Covid-19 - The Law and Limits of Quarantine. The New England

journal of medicine 2020; 382(15): e28.

9. Lippi G, Simundic AM, Plebani M. Potential preanalytical and analytical vulnerabilities in the

laboratory diagnosis of coronavirus disease 2019 (COVID-19). Clin Chem Lab Med 2020.

10. Wang W, Xu Y, Gao R, et al. Detection of SARS-CoV-2 in Different Types of Clinical

Specimens. Jama 2020.

11. To KK, Tsang OT, Chik-Yan Yip C, et al. Consistent detection of 2019 novel coronavirus in

saliva. Clin Infect Dis 2020.

12. To KK, Tsang OT, Leung WS, et al. Temporal profiles of viral load in posterior oropharyngeal

saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort

study. Lancet Infect Dis 2020; 20(5): 565-74.

13. Pasomsub E, Watcharananan SP, Boonyawat K, et al. Saliva sample as a non-invasive specimen

for the diagnosis of coronavirus disease 2019: a cross-sectional study. Clin Microbiol Infect 2020.

14. Azzi L, Carcano G, Gianfagna F, et al. Saliva is a reliable tool to detect SARS-CoV-2. J Infect

2020.

15. Khurshid Z, Asiri FYI, Al Wadaani H. Human Saliva: Non-Invasive Fluid for Detecting Novel

Coronavirus (2019-nCoV). Int J Environ Res Public Health 2020; 17(7).

16. Doganay L, Agaoglu NB, Irvem A, Alkurt G, Yildiz J, Kose B et al. Responding to COVID-

19 in Istanbul: Perspective from genomic laboratory. North Clin Istanb. 2020;7(3):311-2.

doi:10.14744/nci.2020.30075

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

FIGURE LEGENDS

Figure 1: Schematic representation of the sampling methods used in the study (1a:

Oronasopharyngeal sampling, 1b: Nasopharyngeal sampling, 1c: Saliva sampling, 1d: Tube and

swab types) (This figure was designed by using BioRender program that is licensed)

Figure 2: Study workflow diagram

Figure 3: Sensitivity rate of each sampling methods in day 0 and 5 RT-PCR analysis.

Figure 4: Graphical representation of FAM and ROX Ct values detected in saliva,

nasopharyngeal and oronasopharyngeal samples in day 0 (a) and 5 (b) RT-PCR studies.

TABLE LEGENDS

Table 1: The clinical and demographic characteristics of the patients.

Table 2: RT-PCR study results of saliva, oronasopharyngeal and nasopharyngeal samples on

day 0 in 98 patients who were found positive in RT-PCR analysis before hospitalization.

Table 3: RT-PCR study results of saliva, oro-nasopharyngeal and nasopharyngeal samples on

day 5 in 59 patients. We would like to remind that of 59 patients, 20 (34%) had at least one

positive result in saliva or swab sample.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Figure1: Schematic representation of the sampling methods used in the study (1a:

Oronasopharyngeal sampling, 1b: Nasopharyngeal sampling, 1c: Saliva sampling, 1d:Tube and

swab types) (This figure was designed by using BioRender program that is licensed)

Figure 2: Study workflow diagram

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Figure 3: Sensitivity rate of each sampling methods in day 0 and 5 RT-PCR analysis.

Figure 4: Graphical representation of FAM and ROX Ct values detected in saliva,

nasopharyngeal and oronasopharyngeal samples in day 0 (a) and 5 (b) RT-PCR studies.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Table 1: The clinical and demographic characteristics of the patients.

Patients with

positive

RT-PCR results

n=98

Patients with

negative

RT-PCR results

n=102

Total

n=200

p value

Age

(yrs, mean ±SD)

55.8(±15.8)

54.1(±16.4)

54.9(±16.1)

0.434

Sex M/F

48/50

49%

58/44

57%

106/94

53%

0.264

Symptoms

of presentation

-Cough

69%

51%

120

60%

0.008

-Myalgia

73%

37%

110

55%

0.000

-Fever

45%

35%

40%

0.166

-Dyspnea

43%

46%

45%

0.647

-Loss of

smell&taste

0.131

Thorax CT (%)

0.001

-Normal

22%

15%

-Mild Infiltration

54%

58%

112

56%

-Moderate

Infiltration

33%

17%

25%

-Severe

Infiltration

Thorax CT

Moderate &

Severe

Normal & Mild

38%

18%

28%

0.001

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint

Table 2: RT-PCR study results of saliva, oronasopharyngeal and nasopharyngeal samples on day 0 in 98 patients who were found positive in RT-

PCR analysis before hospitalization.

Table 3: RT-PCR study results of saliva, oro-nasopharyngeal and nasopharyngeal samples on day 5 in 59 patients. We would like to remind that

of 59 patients, 20 (34%) had at least one positive result in saliva or swab sample.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 28, 2020. ; https://doi.org/10.1101/2020.07.26.20158618doi: medRxiv preprint