key: cord-275439-cdlcv1c9 authors: Iwasaki, S.; Fujisawa, S.; Nakakubo, S.; Kamada, K.; Yamashita, Y.; Fukumoto, T.; Sato, K.; Oguri, S.; Taki, K.; Senjo, H.; Hayasaka, K.; Konno, S.; Nishida, M.; Teshima, T. title: Comparison of SARS-CoV-2 detection in nasopharyngeal swab and saliva date: 2020-05-19 journal: nan DOI: 10.1101/2020.05.13.20100206 sha: doc_id: 275439 cord_uid: cdlcv1c9 We prospectively compared the efficacy of PCR detection of SARS-CoV-2 between paired nasopharyngeal and saliva samples in nine COVID-19 patients. SARS-CoV-2 was detected in saliva in 8 of 9 (89%) patients and in all 11 samples taken within 2 weeks after disease onset. Viral load was equivalent at earlier time points but declined in saliva than nasopharyngeal samples. PCR negativity was also concordant in all 27 saliva samples from 24 patients between nasopharyngeal and saliva samples. These results suggest that saliva is a reliable noninvasive alternative to nasopharyngeal swabs and facilitate widespread PCR testing in the face of shortages of swabs and protective equipment without posing a risk to healthcare workers. Rapid detection of the novel coronavirus SARS-CoV-2 is critical for the prevention of outbreaks coronavirus disease 2019 in communities and hospitals. The diagnosis of COVID-19 is made by PCR testing of samples collected by nasopharyngeal or oropharyngeal swabs, with the nasopharyngeal route being the standard with a sensitivity for the virus in the range of 52-71% [1] [2] [3] [4] [5] . However, swab sample collection requires specialized medical personnel with protective equipment and poses a risk of viral transmission to healthcare workers. Although sputum specimen is a noninvasive alternative, sputum production is seen in only 28% of COVID-19 patients 6 . The angiotensin converting enzyme 2 (ACE2) is the main host cell receptor for SARS-CoV-2 entry to the human cell 7, 8 . ACE2 is highly expressed on the mucous of oral cavity, particularly in epithelial cells of the tongue 9 . These findings explain mechanisms that the oral cavity is high risk for SARS-CoV-2 infection, transmission occurs through saliva before the onset of symptoms, and the impairment of the sense of taste 10 . Thus, it is reasonable to use saliva as a diagnostic sample, and recent studies have shown that SARS-CoV-2 is detected in saliva [11] [12] [13] . Moreover, Wyllie et al. demonstrated the saliva to be more sensitive for SARS-CoV-2 detection patients than nasopharyngeal swabs 14 . However, few studies compared viral load between nasopharyngeal and saliva samples. We herein compared the diagnostic value of saliva and nasopharyngeal samples using prospectively collected paired samples. Nasopharyngeal swab samples and saliva samples were simultaneously collected from patients suspicious of COVID-19 and from patients who were referred to our hospital with the diagnosis of COVID-19. This study was approved by the Institutional Ethics Board and informed consent was obtained from all patients. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 19, 2020. . https://doi.org/10.1101/2020.05.13.20100206 doi: medRxiv preprint Nasopharyngeal samples were obtained by using FLOQSwabs (COPAN, Murrieta, CA, USA). The swab was passed through the nostril until reaching the posterior nasopharynx and slowly removed while rotating. The swabs were placed in the saline. Saliva samples were self-collected by the patients except one patient, in whom saliva was collected by swab due to inability of self-collection, and spit into a sterile PP Screw cup 50 (ASIAKIZAI Co., Tokyo, Japan). 200 L Saliva was added to 600 L PBS, mixed vigorously, then centrifuged at 20,000 x g for 5 minutes at 4 o C, and 140 µl of the supernatant was used as a sample. We used the paired t-test to compare data. All P-values were 2-sided. We used Pearson's correlation to assess the relation between time from symptom onset and viral load and CT value. We performed analyses using Prism software version 6 (GraphPad, La Jolla, CA). P-value of 0.05 was used as the cutoff for statistical significance. Thirty-three patients were enrolled in this study, including 9 patients with COVID-19 and 24 COVID-19 suspicious patients. Most of COVID-19 patients had mild to moderate disease. Median age of COVID-19 patients was 70.5 years-old, ranging from 30 to 97 years-old. COVID-19 patients were admitted to our hospital after a diagnosis was made by nasopharyngeal samples. In COVID-19 patients, median day of sampling was 10 days (range, 7-19 days) after symptom onset. SARS-CoV-2 was detected in all 9 patients in nasopharyngeal samples and in 8/9 (89%) patients in saliva samples (Table 1 ). In one patient who showed saliva negativity, samples were taken 19 days after symptom onset. The mean ± SD of the viral load were 6.1 ± 1.3 and 4.2 ± 1.5 log10 gene copies/ml in nasopharyngeal and saliva samples, respectively, and significantly lower in saliva samples All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 19, 2020. . https://doi.org/10.1101/2020.05.13.20100206 doi: medRxiv preprint (P=0.018). When looking at relation of viral load and duration from symptom onset to sampling, viral load was equivalent between the two samples at earlier time points but declined in saliva at later time points ( Figure 1A) . In a real-time PCR assay, cycle threshold (CT) value is defined as the number of cycles required for the fluorescent signal to cross the threshold. The mean ± SD of the CT values were 24.2 ± 4.4 and 30.4 ± 4.9 in nasopharyngeal and saliva samples, respectively, and significantly higher in saliva samples (P=0.018). The CT values were equivalent between the two samples at earlier time points but higher in saliva at later time points ( Figure 1B ). All patients were treated with favipiravir 15 and underwent PCR examination when symptoms were relieved to determine the timing of discharge. Figure 2 We prospectively compared SARS-CoV-2 detection between nasopharyngeal samples and saliva samples in 9 patients with COVID-19. The virus was detected in saliva in 8/9 (89%) patients. Detection rate was consistent to other studies: 11/12 (92%) patients and 20/23 (87%) patients in Hong Kong 11, 13 , 25/25 (100%) in Italy 12 , and 36/38 (95%) in New Haven 14 . In a screening clinic in Australia, 39/622 (6.3%) patients had PCR positive nasopharyngeal swabs and 33/39 (85%) had virus in saliva 16 . Taken together, these results consistently support the use of saliva as an effective alternative to nasopharyngeal swabs for diagnosis and screening of COVID-19. It has been shown that salivary viral load peaks at onset of symptoms and is highest during the first week and subsequently declines with time 11, 13, 14 . Our results were consistent to these data; the virus was detected in all the saliva samples taken within 2 weeks after symptom onset. PCR cannot distinguish whether the virus is alive or dead. Recently, Korean researchers suggested that particle of the dead virus could persist in the nasopharynx and resulted in "false positivity" (http://www.koreaherald.com/view.php?ud=20200429000724). Interestingly, in our results, PCR results tended to become negative much quicker in saliva All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 19, 2020. . https://doi.org/10.1101/2020.05.13.20100206 doi: medRxiv preprint than in the nasopharynx, suggesting that dead virus particle in mouth is more efficiently cleaned by saliva. Saliva could be a better tool to determine virus clearance in COVID-19 patients. To our knowledge, two studies compared viral load between nasopharyngeal and saliva samples. The viral load was higher in saliva than in nasopharyngeal samples in one study 14 , whereas sensitivity determined by CT values was lower in saliva in one study 16 . Our results showed that the viral load was equivalent at earlier time points but lower in saliva than in nasopharyngeal samples at later points. Although it is difficult to compare viral concentration in saliva itself and in transport media of the swab samples, there are several possibilities to explain the inconsistency. First, salivary viral load appears to be highest during the first week after symptom onset and subsequently declines with time 11, 13, 14 . Our study did not include samples taken during the first week after onset. Second, differences in sampling methods may have contributed to the discrepant results. Specifically, the 1ml volume of our saliva samples in our study were much smaller in our study with compared to a third full cup of saliva collected in the Yale study. There was inconsistency of nasopharyngeal swab samples in the Yale study 14 , while our data showed less inconsistency, probably due to swab sampling performed exclusively by only three pulmonologists. Last, all patients in our study were treated with favipiravir 15 , which may be associated with rapid virus clearance from the oral cavity. Although our study has several limitations due to the small number of samples and the lack of samples within the first week of symptom onset, there have been few prospective studies to date comparing the two samples. Given the large benefits of saliva collection that does not require health worker specialists and protective equipment, our results together with recent studies support the use of saliva as a noninvasive alternative to nasopharyngeal swabs to greatly facilitate widespread PCR testing. Data are shown as mean ± SD (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 19, 2020. . https://doi.org/10.1101/2020.05.13.20100206 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 19, 2020. . https://doi.org/10.1101/2020.05.13.20100206 doi: medRxiv preprint Detection of SARS-CoV-2 in Different Types of Clinical Specimens Sensitivity of Chest CT for COVID-19: Comparison to Profiling Early Humoral Response to Diagnose Novel Coronavirus Disease (COVID-19) SARS-CoV-2 Viral Load in Upper Respiratory Specimens of Infected Patients Clinical features of patients infected with 2019 novel coronavirus in Wuhan Genome Composition and Divergence of the Novel -nCoV) Originating in China Functional assessment of cell entry and receptor usage for SARS-CoV-2 and other lineage B betacoronaviruses Sabino-Silva R, Jardim ACG, Siqueira WL. Coronavirus COVID-19 impacts to dentistry and potential salivary diagnosis Consistent detection of 2019 novel coronavirus in saliva Saliva is a reliable tool to detect SARS-CoV Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study Saliva is more sensitive for SARS CoV-2 detectionin COVID-19 patients than nasopharyngeal swabs. medRxiv Favipiravir versus Arbidol for COVID-19: a randomized clinical trial. medRxiv All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder The authors declare that they have no competing interests.