Influenza A, like Omicron SARS-CoV-2, Is Similarly Detected in Saliva or Nasopharyngeal Samples via RT-qPCR

Abstract

After the Coronavirus pandemic, the importance of virus surveillance was highlighted, reinforcing the constant necessity of discussing and updating the methods for collection and diagnoses, including for other respiratory viruses. Although the nasopharyngeal swab is the gold-standard sample for detecting and genotyping SARS-CoV-2 and Influenza viruses, its collection is uncomfortable and requires specialized teams, which can be costly. During the pandemic, non-invasive saliva samples proved to be a suitable alternative for SARS-CoV-2 diagnosis, but for Influenza virus the use of this sample source is not recognized yet. In addition, most SARS-CoV-2 comparisons were conducted before the Omicron variant emerged. Here, we aimed to compare Influenza A and Omicron RT-qPCR analysis of nasopharyngeal swabs and saliva self-collection in paired samples from 663 individuals. We found that both nasopharyngeal swab and saliva collection are efficient for the diagnosis of Omicron (including sub-lineages) and for Influenza A, with high sensitivity and accuracy (>90%). The kappa index is 0.938 for Influenza A and 0.905 for SARS-CoV-2. These results showed excellent agreement between the two samples reinforcing saliva samples as a reliable source for detecting Omicron and highlighting saliva as a valid sample source for Influenza detection, considering this cheaper and more comfortable alternative.

Keywords: COVID-19; diagnostic techniques and procedures; flu; surveillance; virus detection.

Figure 1

Multiplex detection of SARS-CoV-2 and Influenza in nasopharyngeal swabs and saliva samples: (A) Positive cases of Influenza A and COVID-19 detected by epidemiological week from 2022. Values combine patients’ positive in NSP, saliva, or both. (B) Frequency of patients positive for Influenza A, SARS-CoV-2, Fluorona (coinfection), or negative diagnosis in each type of sample source. (C) Frequency of Omicron subvariants by epidemiological week.

Figure 2

Comparison of cycle threshold values between saliva and NS samples: (A) Comparison of CT from RT-qPCR of NPS and saliva to Influenza A (left) and SARS-CoV-2 (right). p-values derived from the Welch t-test, each dot is a patient, and lines connect samples from the same patients. Cases with no amplification were defined in Ct 40. (B) Correlation between Ct value from NPS and saliva for Influenza A (above) and SARS-CoV-2 (below). Correlations values derived from Pearson correlation; gray areas denote no amplification in one of the samples; and the line represents the theoretical curve for perfect correlation. (C) Comparison of CT from RT-qPCR of NPS and saliva for Influenza A, SARS-CoV-2, and human RNase P targets. p-values derived from the Welch t-test, the dark line is the median, the box extends from the first to the third quartile, and each dot is a single diagnosis.