Olfactory dysfunction in SARS-CoV-2 infection: Focus on odorant specificity and chronic persistence

Abstract

Background: Smell dysfunction has been recognized as an early symptom of SARS-CoV-2 infection, often occurring before the onset of core symptoms of the respiratory tract, fever or muscle pain. In most cases, olfactory dysfunction is accompanied by reduced sense of taste, is partial (microsmia) and seems to normalize after several weeks, however, especially in cases of virus-induced complete smell loss (anosmia), there are indications of persisting deficits even 2 months after recovery from the acute disease, pointing towards the possibility of chronic or even permanent smell reduction for a significant part of the patient population. To date, we have no knowledge on the specificity of anosmia towards specific odorants or chemicals and about the longer-term timeline of its persistence or reversal.

Methods: In this longitudinal study, 70 participants from a community in Lower Austria that had been tested positive for either IgG or IgM SARS-CoV-2 titers in June 2020 and a healthy control cohort (N = 348) underwent smell testing with a 12-item Cross-Cultural Smell Identification Test (CC-SIT), based upon items from the University of Pennsylvania Smell Identification Test (UPSIT). The test was performed in October 2020, i.e. 4 months after initial diagnosis via antibody testing. Results were analyzed using statistical tests for contingency for each smell individually in order to detect whether reacquisition of smell is dependent on specific odorant types.

Results: For all odorants tested, except the odor "smoke", even 4 months or more after acute SARS-CoV-2 infection, participants with a positive antibody titer had a reduced sense of smell when compared to the control group. On average, while the control cohort detected a set of 12 different smells with 88.0% accuracy, the antibody-positive group detected 80.0% of tested odorants. A reduction of accuracy of detection by 9.1% in the antibody-positive cohort was detected. Recovery of the ability to smell was particularly delayed for three odorants: strawberry (encoded by the aldehyde ethylmethylphenylglycidate), lemon (encoded by citronellal, a monoterpenoid aldehyde), and soap (alkali metal salts of the fatty acids plus odorous additives) exhibit a sensitivity of detection of an infection with SARS-CoV-2 of 31.0%, 41.0% and 40.0%, respectively.

Conclusion: Four months or more after acute infection, smell performance of SARS-CoV-2 positive patients with mild or no symptoms is not fully recovered, whereby the ability to detect certain odors (strawberry, lemon and soap) is particularly affected, suggesting the possibility that these sensitivity to these smells may not only be lagging behind but may be more permanently affected.

Keywords: Anosmia; COVID-19; Dysgeusia; Microsmia; SARS-CoV-2; Smell and taste dysfunction; Smell test.

Fig. 1

Graphical representation of the percentage of correctly identified different odors from the standardized smell test (CC-SIT-Burghart Messtechnik; Wedel, Germany).

Fig. 2

Odors ranked by the sensitivity (A) and specificity (B), as also shown in Table 3. Sensitivity signifies the probability that an antibody positive-person cannot detect a specific odorant and specificity values describe the probability that a healthy, antibody-negative person identifies a smell correct.