Assessment of saliva as a non-invasive and reliable complement to nasopharyngeal swabs for COVID-19 diagnosis

Objectives: Nasopharyngeal/oropharyngeal (NP/OP) swabs, the gold standard for COVID-19 diagnosis, pose challenges including patient discomfort, need for trained personnel, and biosafety risks. This study evaluated the diagnostic performance of saliva as a less invasive alternative compared to the gold standard. Materials and Methods: Paired NP/OP and saliva samples were collected from 515 suspected COVID-19 patients. NP/OP swabs underwent standard RNA extraction and reverse transcription-polymerase chain reaction (RTPCR). Saliva samples were analyzed by RT-PCR, either directly (saliva-direct [SD]) or after RNA extraction (saliva-RNA [SR]). A subset of discordant samples (saliva-positive, NP/OP-negative), with cycle threshold values ≤32, was investigated using Nanopore sequencing. Statistical analysis: Sensitivity, specificity, and Cohen’s Kappa were used to assess agreement between NP/OP and each saliva protocol. McNemar’s test compared detection rates between methods, and paired t-tests/Wilcoxon signed-rank tests compared Ct values between NP/OP and saliva-RNA assays. Results: Of 515 participants, NP/OP swabs yielded 114 positives (reference standard), SR yielded 117, and SD yielded 147. SR sensitivity/specificity was 61.4%/88.28% compared to NP/OP. SD sensitivity/specificity was 70.17%/83.29%. Cohen’s Kappa was 0.50 (0.399–0.581) for SR and 0.48 (0.395–0.569) for SD. 16.69% of saliva-positive samples were NP/OP negative. Nanopore sequencing confirmed severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) RNA in 78.26% of discordant saliva samples. Conclusions: Saliva shows promise as a complementary, adjunct specimen for COVID-19 diagnosis, potentially enhancing case detection. While sensitivity was lower than NP/OP, the direct saliva method demonstrated improved detection. The simplified collection and reduced biosafety risks associated with saliva offer significant advantages.