Influence of Saliva Collection Method on the Detection of SARS-CoV-2 and IgG Antibodies

Abstract

Background: Although, the nasopharyngeal swabs (NPS) are considered as gold standard for the clinical diagnosis of SARS-CoV-2 virus in the COVID-19 disease, they pose several limitations such as high risk of exposure, discomfort to the patients and requirement of trained healthcare professionals. Several studies have explored ‘Saliva’ as an alternate clinical specimen for the SARS-CoV-2 and IgG detection, reporting varied degree of sensitivity of detection. This study was aimed to investigate the influence of method of saliva collection on the sensitivity of SARS-CoV-2 detection.Methods: We performed a cross sectional study where patients were screened for the COVID-19 infection. Saliva was collected by four different methods (Expectoration, drooling, gargling and by using salivary swabs). SARS-CoV-2 viral load in the paired saliva and NPS were determined by using the reverse transcriptase quantitative PCR (RT-qPCR). IgG levels were assayed in the serum and saliva of the COVID-19 patients by nephelometry and ELISA resp.Findings: Out of 350 patients screened, 43 patients were included in the study which was found to be positive for COVID-19 as evidenced by RT-PCR in the NPS (positivity rate-12.2%). Expectorated saliva exhibited 78.5% sensitivity (11 out of 14 patients, median CT (saliva e gene 27.0±6.03 vs NPS e gene 29.5.0±6.18, p=NS). Drooling method showed 22.2% sensitivity (2 out 9 patients, median CT (saliva e gene 30.5±0.7 vs NPS e gene 30.0±1.5, p=NS)) whereas salivary swab method yielded 21.42% sensitivity (3 out 14 patients, median CT (saliva e gene 29.0±5.1vs NPS e gene 29.0±5.1, p=NS)). We found lowest sensitivity (16.66%) in the gargled saliva (1 out of 6 patients, median CT (saliva e gene, 24±7.5 vs NPS e gene 14.0, p=NS)). Furthermore, the sensitivity of SARS-CoV-2 detection was reduced to 18.1% (2 out of 9 patients) and 0.0% (0 out 3 patients) in the saliva collected by salivary swab and gargling method above the CT value 25.0 (NPS). Interestingly, salivary IgG showed better concordance with viral load as compared to the serum IgG (R 2 0.23 vs 0.04, p=0.044).Interpretation: Expectorated saliva is a better specimen as compared to the drooling, gargling and salivary swabs for SARS-CoV-2 viral detection for the diagnosis of COVID-19. Exposure of saliva to the upper and lower respiratory track may be the primary explanation to our findings. Also, salivary IgG can be a better predictor of the viral load as compared to the blood IgG. However, our finding needs to be confirmed in the larger sample size of the COVID-19 patients.Funding Statement: In house funding was given by Dr. D.Y. Patil Vidyappeth, Pimpri, Pune (India) to carry out the study.Declaration of Interests: All the authors declare no conflict of interest.Ethics Approval Statement: This study was approved by the Institutional Ethics Committee of Dr. D. Y. Patil Medical College and Hospital, Pune (India) (DYPV/EC/599/2020).