Taste dysfunction in Long COVID

Abstract Persistent taste dysfunction is frequently reported in individuals with post-acute sequelae of infection by SARS-CoV-2 (Long COVID). The mechanisms and pathological correlates underlying this taste dysfunction are unknown. This study investigates the underlying pathology in 28 non-hospitalized subjects diagnosed with COVID-19 and who experienced taste disturbances more than 12 months after testing positive for SARS-CoV-2. To objectively establish the nature of taste deficit, we used the WETT taste test, which quantifies the subject’s ability to taste each of the five taste qualities: sweet, umami, bitter, sour, and salty. We then biopsied five to eight fungiform taste papillae (FP) in 20 of the 28 subjects. The FPs were analyzed histologically for overall taste bud structure and innervation, and by quantitative PCR (qPCR) for mRNA expression of markers for different taste receptor cells. Although all subjects had reported taste dysfunction, only three showed overall taste scores below the 10th percentile for a normal population adjusted for age and sex. However, 11 of the 28 subjects exhibited total loss of one or more taste qualities. Loss of PLCβ2-dependent taste qualities (sweet, umami, bitter) was significantly more common and was correlated with reduced expression of PLCβ2 and Tas1R3 mRNAs. Histological analysis revealed generally preserved taste bud structure and innervation, but with occasional disorganized taste buds and abnormal, isolated PLCβ2-positive cells in the epithelium. Our findings suggest long-term taste dysfunction after COVID-19 occurs rarely -- more frequently involving PLCβ2-dependent taste qualities -- but is not due to wholesale disruption of the taste periphery. Competing Interest Statement The authors have declared no competing interest. Footnotes ↵# Shared Senior authors The data were re-analyzed using the covid+ subject without taste dysfunction as a standard for normalization of the PCR. Also for psychophysical data we used the 10% limit as a cut-off for hypoguesia instead of 95% confidence interval. None of these changes in analysis resulted in changes of significance or conclusions. Minor changes also were made throughout the text to correct minor errors or improve clarity.