Recovery rates of persistent post-COVID-19 olfactory dysfunction using psychophysical assessment: A longitudinal cohort study

Abstract

Objectives: Persistent olfactory dysfunction (OD) following loss of smell associated with SARS-CoV-2 infection is a major feature of long COVID. Perspectives on the prevalence of persistent OD predominantly rely on self-reported olfactory function. Few studies have tracked longitudinal rates of recovery using psychophysical assessment among patients presenting for evaluation of persistent OD beyond a window of acute recovery. Data anchored in standardized testing methods are needed to counsel patients who fail to acutely regain their sense of smell. This study aims to quantify the degree of persistent OD in post-COVID-19 patients who experience subjective and psychophysical OD.

Methods: We grouped participants presenting for OD evaluation into cohorts based on both subjective and psychophysical olfactory status at a baseline assessment and assessed their olfactory abilities with a visual analogue scale and the Sniffin' Sticks extended test at baseline and 1-year time points. Participants had confirmed a history of COVID-19 by lab evaluation or clinical diagnosis if lab evaluation was not available.

Results: Baseline olfactory evaluation was completed by 122 participants, 53 of whom completed the 1-year follow-up assessment. Among participants presenting with perceived OD, 74.5% had confirmed psychophysical OD at baseline, with 55.1% at 1-year follow-up. Participants had reliable trends in self-rated versus psychophysically tested olfactory function at both time points. The total threshold, discrimination, and identification (TDI) score improved by +3.25 points in the cohort with psychophysical OD (p = 0.0005), with this improvement largely attributable to an increase in median threshold scores (+2.75 points; p = 0.0004).

Conclusions: OD persists in a significant number of patients who fail to acutely recovery their sense of smell after COVID-19, with many demonstrating lingering deficits at 1-year. Improvements in threshold, but not discrimination or identification, most significantly mediate improvement of total TDI score at follow-up.

Keywords: long COVID; olfaction; post‐COVID condition; smell dysfunction.

Figure 1

Pie charts with pooled subjective olfactory dysfunction (sOD) and psychophysical olfactory dysfunction (pOD) participants (no controls) showing overall distribution of olfactory status at (A) baseline and (B) 1‐year psychophysical assessment according to threshold, discrimination, and identification (TDI) score groupings (TDI < 16: anosmia; TDI ≥ 16.25 and ≤ 30.5: hyposmia; TDI ≥ 30.75: normosmia).

Figure 2

Spaghetti plot representing trends in participant threshold, discrimination, and identification (TDI) scores from baseline to 1‐year psychophysical assessment. Thin lines represent trends for specific participants. Bold lines represent trends for larger olfactory cohorts. Olfactory groupings represented by color blocks: green (normosmia), blue (hyposmia), and red (anosmia). OD, olfactory dysfunction.

Figure 3

Spaghetti plots representing trends in olfactory subdomains of T (A), D (B), and I (C) from baseline to 1‐year psychophysical assessment. Thin lines represent trends for specific participants. Bold lines represent trends for larger olfactory cohorts. There is overlap for some data points in the graphs given that threshold scores have a finite number of possibilities ranging from 1 to 16, and discrimination and identification test scores are integers ranging from 0 to 16. p‐Values represent post hoc Tukey's HSD comparison of psychophysical olfactory dysfunction (pOD) and subjective olfactory dysfunction (sOD) groups at 1‐year.

Figure 4

Correlation between visual analogue scale (VAS) score for olfactory status and psychophysically measured threshold, discrimination, and identification (TDI) score for study participants at Baseline and 1‐year assessment.