Saliva as a Reliable and Non-invasive Sample for Detecting Influenza A in Severe Acute Respiratory Infection Cases

Background
Nasopharyngeal swab sampling remains the gold standard for influenza diagnosis; however, it has several limitations, including dependence on medical staff, invasiveness, potential for nosocomial transmission, and occupational exposure risk. Non-invasive alternatives, such as saliva and nasal vestibular swabs, may improve patient comfort and participation in clinical studies. In addition, diagnosis with reverse transcription real-time quantitative polymerase chain reaction (RT-qPCR) is often delayed because it requires trained laboratory technicians and facilities with appropriate laboratory settings. Although rapid diagnostic devices such as the GenPad® offer potential alternatives to RT-qPCR, their performance with non-invasive samples remains insufficiently explored. This study addresses the two key questions for influenza detection in severe acute respiratory infection (SARI) cases: (i) whether saliva or nasal vestibular swab samples serve as suitable alternatives to nasopharyngeal swab samples, and (ii) whether the GenPad® provides a reliable option for detecting influenza using saliva samples.
Methodology
A prospective observational study was conducted with 16 inpatients classified as having SARIs and diagnosed with influenza between December 2024 and March 2025 in Japan. Paired saliva and nasal vestibular swab samples were collected 1-9 (median = 3.5) days after symptom onset. RT-qPCR testing was performed according to the National Institute of Infectious Diseases protocol. Saliva samples were also tested using the GenPad® system. Comparisons between sample types and diagnostic methods were analyzed using the exact McNemar's test.
Results
Among the 16 influenza-positive patients, saliva samples demonstrated higher sensitivity (87.5%) than nasal vestibular swabs (31.3%) in RT-qPCR when compared with the diagnostic results obtained from nasopharyngeal swabs. A comparison of RT-qPCR results between saliva and nasal vestibular swabs revealed a total agreement of 43.8%, with exact McNemar's test showing a significant difference (p = 0.0039). While nasal vestibular swabs showed inconsistent results, saliva samples consistently tested positive, particularly within seven days of symptom onset (100% positive agreement). The GenPad®, a rapid diagnostic device, showed promising performance (92.9%) using saliva samples compared to RT-qPCR.
Conclusions
Saliva is a reliable non-invasive alternative to nasopharyngeal swabs for influenza detection in SARI cases, particularly within seven days of symptom onset, whereas nasal vestibular swabs show lower sensitivity. Additionally, the GenPad® provides comparable performance to RT-qPCR using saliva samples, offering a rapid, portable diagnostic option. These approaches may mitigate discomfort, minimize infection risk for healthcare workers, and improve testing capacity. However, the absence of influenza-negative controls and the small sample size (n = 16) substantially limit the assessment of diagnostic accuracy and specificity. As a result, the broader applicability of our findings should be interpreted with caution, and further studies are required to validate these observations.