The usefulness of a quantitative olfactory test for the detection of COVID-19

  1. Marcos A Lessa
  2. Stella M Cotta-Pereira
  3. Frederico A Ferreira
  4. Therezinha Marta P P Castiñeiras
  5. Rafael M Galliez
  6. Débora S Faffe
  7. Isabela de C Leitão
  8. Diana Mariani
  9. Erica R Nascimento
  10. Flávia S Lessa
  11. Isabela Brasil Succi
  12. Carlos A Pedreira

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Background

During the COVID-19 pandemic, olfactory dysfunction (anosmia or hyposmia) has been reported by many patients and recognized as a prevalent and early symptom of infection. This finding has been associated with viral-induced olfactory neuron dysfunction rather than the nasal congestion typically found in cold- or flu-like states. In literature, the prevalence of anosmia varies from 15% to 85%, and the studies, in general, were based on the subjective evaluation of patients’ self-reports of loss of smell (yes or no question). In the present study, we quantitatively evaluated olfactory dysfunction and the prevalence of fever in symptomatic patients suspected of having COVID-19 using a scratch-and-sniff olfactory test and infrared temperature testing with RT-PCR as the gold-standard comparator method to diagnose COVID-19 infection.

Methods

Outpatients had their forehead temperature checked with an infrared non-contact thermometer (temperature guns). After that, they received two olfactory smell identification test (SIT) cards (u-Smell-it™; CT, USA) that each had 5 scent windows and were asked to scratch with a pencil and sniff each of the 10 small circles containing the microencapsulated fragrances and mark the best option on a response card. Nasopharyngeal swabs were then collected for Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) to determine if the patients were positive or negative for COVID-19 infection. We considered the number of ‘hits’ (correct answers) ≤ 5 as positive for loss of smell (LOS) in the olfactory test; ≥ 6 hits was considered negative for LOS (i.e. normal olfactory function). All data were analyzed using Excel and Matlab software.

Results

In the present study, 165 patients were eligible for the olfactory test and nasopharyngeal swab collection RT-PCR. Five patients were excluded because of inconclusive PCR results (n=2) and missing data (n=3). A total of 160 patients completed all the protocols. The RT-PCR positivity rate for COVID-19 was 27.5% (n=44), and PCR+ patients scored significantly worse in the olfactory test (5.5±3.5) compared to RT-PCR-patients (8.2±1.8, p<0.001). 0/44 PCR+ patients presented with a fever (≥37.8°C). In contrast an olfactory SIT had a specificity of 94.8% (95% CI, 89.1 – 98.1), sensitivity of 47.7% (95% CI, 32.7 – 63.3), accuracy of 0.82 (95% CI, 0.75 – 0.87), positive predictive value of 77.8% (95% CI, 59.6 – 88.8), negative predictive value of 82.7% (85% CI, 78.7 – 86.7), and odds ratio of 16.7.

Conclusion

Our results suggest that temperature checking failed to detect COVID-19 infection, while an olfactory test may be useful to help identify COVID-19 infection in symptomatic patients.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.01.20.21250173: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Institutional Review Board StatementIRB: The study was approved by the local ethics committee from Clementino Fraga Filho University Hospital (CAAE: 30161620.0.0000.5257).
    Consent: Written informed consent was obtained from all participants.
    RandomizationHere, we randomly sorted 5 out of 10 odorants for each patient and calculated the parameters after passing through the complete dataset.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    All data was statistically analyzed Matlab software (R2019b) and confirmed with Medcalc® software.
    Matlab
    suggested: (MATLAB, RRID:SCR_001622)
    Medcalc®
    suggested: (MedCalc, RRID:SCR_015044)

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:

    The study’s limitations include that we recruited only outpatient that came into the clinic for a diagnostic test at variable points of the disease, that as mentioned above we not ascertained, and we did not investigate asymptomatic patients. More, our cohort did not include children, pregnant, or the elderly, so our results may not be directly extrapolated to these groups of patients. It is anticipated that a short 5- or 10-odorant smell identification test, patients with minor hyposmia cases may be missed compared to a 40-odorant UP-SIT test11,26. In conclusion, our study demonstrates that patients with positive SARS CoV-2 RT-PCR were highly associated with olfactory dysfunction, and 3.6-fold higher when tested with a short olfactory test than seen in self-reports. Taking together, these results suggest that quick olfactory tests may be useful to detect and COVID-19 infection in symptomatic patients.

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from scite Reference Check: We found no unreliable references.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2021.01.20.21250173 on medRxiv