Long-term effects of SARS-CoV-2 infection in patients with and without chemosensory disorders at disease onset: a psychophysical and magnetic resonance imaging exploratory study

Abstract

A preserved sense of smell and taste allows us to understand many environmental "messages" and results in meaningfully improvements to quality of life. With the COVID-19 pandemic, it became clear how important these senses are for social and nutritional status and catapulted this niche chemosensory research area towards widespread interest. In the current exploratory work, we assessed two groups of post-COVID-19 patients who reported having had (Group 1) or not (Group 2) a smell/taste impairment at the disease onset. The aim was to compare them using validated smell and taste tests as well as with brain magnetic resonance imaging volumetric analysis. Normative data were used for smell scores comparison and a pool of healthy subjects, recruited before the pandemic, served as controls for taste scores. The majority of patients in both groups showed an olfactory impairment, which was more severe in Group 1 (median UPSIT scores: 24.5 Group 1 vs 31.0 Group 2, p = 0.008), particularly among women (p = 0.014). No significant differences emerged comparing taste scores between Group 1 and Group 2, but dysgeusia was only present in Group 1 patients. However, for taste scores, a significant difference was found between Group 1 and controls (p = 0.005). No MRI anatomical abnormalities emerged in any patients while brain volumetric analysis suggested a significant difference among groups for the right caudate nucleus (p = 0.028), although this was not retained following Benjamini-Hochberg correction. This exploratory study could add new information in COVID-19 chemosensory long-lasting impairment and address future investigations on the post-COVID-19 patients' research.

Keywords: MRI brain volumetric analysis; Post-COVID-19 patients; Smell/taste impairment.

Fig. 1

Comparison of UPSIT scores between COVID-19 patients in Group 1 and 2 (A males and females together; B females only). p-values from Wilcoxon Mann–Whitney rank tests

Fig. 2

Comparison of UPSIT scores of COVID-19 patients from Group 1 and 2 with the 5th percentile of a normative population matched by sex. p-values from Wilcoxon matched-pairs signed-rank tests (p-values > 0.05 are not reported)

Fig. 3

Comparison of the Taste Strips Test (TST) global score among groups: Group 1, Group 2 and the healthy control one (Ctrl). p-values from Wilcoxon matched-pairs signed-rank tests (p-values > 0.05 are not reported). A significant difference was found between Group 1 and Ctrl (p = 0.005). The normal range score is 9–16. The red line indicates the minimum normal score (9)

Fig. 4

Comparison of the Taste Strips Test (TST) individual scores, for each taste quality assessed, among groups: Group 1, Group 2, and the healthy control one (Ctrl). Sweet (A), bitter (B), sour (C), and salty (D). p-values from Wilcoxon matched-pairs signed-rank tests (p-values > 0.05 are not reported). For sweet, sour, and salty, a significant difference emerged between Group 1 and Ctrl (p = 0.004, p = 0.028, and p = 0.024, respectively). In addition, for salty taste also Group 2 showed a significant difference with Ctrl (p = 0.002)