sharon sanders
Editor-in-Chief & President
Objective assessment of long-term impact of COVID-19 on multiple sensory functions
Background
This ongoing study investigates the impact of post-acute sequelae of SARS-CoV-2 infection (PASC) on broad sensory functions.
Methods
Sixty subjects aged 27–78 years were recruited who had contracted COVID-19 between 1/17/2020 and 12/21/2023 and had persistent symptoms (4.3–52.9, median = 27.48 months). Quantitative sensory assessments included (1) smell: 9-Item NIH toolbox odor identification, detection threshold to phenyl–ethyl alcohol, and retronasal candy test; (2) taste: modified NIH toolbox; (3) chemesthesis: nasal menthol lateralization thresholds and oral capsaicin identification; (4) hearing: pure-tone audiometry, otoacoustic emissions, words-in-noise recognition, and Dichotic Digits Test; (5) vestibular/balance: video head impulse testing, Subjective Visual Vertical, vestibular perceptual thresholds, and modified Romberg balance test; and (6) cognitive assessment: The Self-Administered Gero-Cognitive Exam, Digit Symbol Substitution Test, and Trail-Making Test.
Results
Overall, subjects self-reported high and overlapping dysfunctions: 67.3% smell, 63.6% taste, 56.5% balance and dizziness, 31.8% auditory, and 51.3% brain fog or cognitive dysfunctions, with varying discrepancy to the measured and confirmed deficits (smell 65.5%, taste 16%, vestibular 31.6%, hearing 53.4%, and cognition 19.1%). Significant associations were found between confirmed vestibular and auditory impairments (p = 0.04), and cognitive impairments with central components of vestibular (p = 0.03) and auditory (p = 0.01) impairments, but not with their peripheral components. Similarly, strong associations were found between identification components of smell and taste tests (p = 0.003), which may involve more central processing, but not with threshold tests (a more peripheral process). Hospitalization significantly associated with smell (p = 0.05), cognitive (p = 0.009), vestibular (p = 0. 001), and peripheral auditory (p = 0.01) dysfunctions. Age significantly associated with auditory (p = 0.02), vestibular (p = 0.01) and central olfactory (p = 0.03) dysfunctions.
Conclusions
COVID-19 impacts sensory systems broadly and differently, both peripherally and centrally, driven in part by aging, initial disease severity, and underlying post-COVID cognitive dysfunctions. Subjective symptoms may not always be corroborated by measured deficits.
Background
Since the outbreak in December 2019, the COVID-19 pandemic has ravaged the world, with over 775 million cases and more than 7 million deaths (as of March 2024, WHO). While most current patients with new COVID-19 infections report symptoms resembling those of the flu or common cold, neurological symptoms are still common among people with COVID-19, including sensory dysfunction. Smell dysfunction has become a hallmark symptom of COVID-19 and can present in isolation (the only symptom) or precede the occurrence of other symptoms [1,2,3,4]. The mechanism of smell dysfunction is unclear, potentially involving peripheral or central pathways. Expression of ACE2, the receptor targeted by SARS-CoV-2, is found not in olfactory sensory neurons (OSNs) but instead in supporting cells (sustentacular cells) and horizontal basal cells in the olfactory epithelium [5]. Sustentacular cells are functionally important for mucus composition (glucose, ion exchange, and gradients) and for structurally supporting OSN dendrites [6, 7]. Experimental manipulation of sustentacular cells has been shown to cause anosmia [6, 7]. Central mechanisms may also be involved, given reports of inflammation in the olfactory bulb [8].
Related to smell dysfunctions, early reported COVID-19-related “taste dysfunction” has been speculated to be due to impaired flavor perception consequent to retronasal smell impairment. However, recent quantitative measurements suggest that loss of taste function (beyond flavor perception) and chemesthesis (trigeminal function) may indeed exist in some COVID-19 patients [9,10,11,12]. There is no previous evidence that viral upper respiratory infections, such as flu or cold, would cause loss of taste or chemesthesis functions, and it remains unclear how these dysfunctions manifest in different taste sub-modalities.
In other sensory systems, up to 15–20% of patients report subjective changes in hearing, tinnitus, and/or dizziness [13, 14]. Evidence of potential auditory and vestibular impairments is additionally supported by several case studies of COVID-19-related sudden hearing loss or vertigo and cross-sectional studies identifying sensorineural hearing loss (SNHL) and/or changes in cochlear function [14,15,16,17,18]. While there is a strong precedent of viral infection affecting inner ear function (e.g., Zika, West Nile, herpes simplex) [19], the exact impact of COVID-19 on inner ear function is unknown. Several mechanisms have been suggested, including direct neural effects, neural inflammation, and hypoxic insults [13, 18, 20].
Recent clinical observations also reveal that a significant portion (73–76%) of COVID-19 patients may still have long-term and persistent symptoms 60–180 days after diagnosis [1]; the most common symptom to persist is fatigue or muscle weakness, as well as neurological disorders; one study reported that over 91% of patients, whether or not they had a neurological diagnosis when first hospitalized for COVID-19, had such problems 6 months after discharge [21]. It is unclear how multisensory dysfunctions manifest among these long COVID patients or how the prolonged vs acute phases differ. We aimed to use a comprehensive set of sensory tests to determine the long-term impact of COVID-19 on broad sensory functions (smell, taste, chemesthesis, hearing, and vestibular/balance function).
Methods
Recruitment
Patients with an ICD10 code U09.9 (“Post COVID-19 condition, unspecified”) were recruited through electronic medical records direct messaging. All interested subjects were then asked to complete a questionnaire probing the presence and severity of sensory dysfunctions (see Additional file 1). To be eligible, they must have longstanding complaints (minimum 90 days) in at least one sensory aspect: smell, taste, hearing, balance, or dizziness. Exclusion criteria (to rule out non-COVID-related pre-existing conditions) included:
This was a longitudinal prospective cohort study approved by the Ohio State Biomedical Sciences Institutional Review Board (approval number: 2022H0414). All subjects signed consent and authorization forms for participation and to enable their data to be included in this study. Olfactory, taste, chemesthesis, auditory, and vestibular functions were tested extensively: (1) smell: 9-Item NIH toolbox odor identification (ID) [22], Ortho nasal odor detection threshold (ODT) to phenyl–ethyl alcohol (PEA) [23], and retronasal candy test [24, 25]; (2) taste: modified NIH toolbox [26]; (3) chemesthesis: nasal menthol lateralization thresholds [27, 28], oral perception of capsaicin solution, which activates TRPV1 trigeminal nociceptors; (4) hearing: pure-tone audiometry (PTA) [29], otoacoustic emissions (OAE) [30], words-in-noise (WIN) recognition [31, 32], and Dichotic Digits Test (DDT) [33, 34]; (5) vestibular/balance: video head impulse testing (vHIT) [35], Subjective Visual Vertical (SVV) [36, 37], vestibular perceptual thresholds [38, 39], and modified Romberg balance test [40]; (6) cognitive assessment: Self-Administered Gero-Cognitive Exam (SAGE) [41], Digit Symbol Substitution Test (DSS) [42], and Trail-Making Test (TMT-A and TMT-B) [43]. The testing was split in two sequential sessions: (1) chemosensory testing, (2) hearing, vestibular and balance testing, with each session taking about 2–3 h. Please see Additional file 2: Supplementary Methods [44,45,46,47,48,49] for details of these tests.
Statistical analyses
Fisher’s exact test (univariate analysis) was used to examine the association between each pair of complaints including smell, taste, auditory, vestibular, and cognitive. Sensitivity and specificity of each measure, as well as the combination of measures, relative to patients self-reported complaints of sensory dysfunction were determined. Differences between subgroups (i.e., those reporting different symptoms) were examined using t-tests, ANOVAs, or non-parametric equivalents when normality of data was not identified. To quantify relationships between test measures, regression analyses or Pearson correlations were performed. Given that these analyses are preliminary and exploratory in nature, post-hoc corrections for multiple comparisons were not applied; results should be interpreted with caution. These analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).
Results
Sixty subjects (aged 27–78 years, median 51 years) with long COVID symptoms lasting more than 3 months were recruited. They contracted COVID-19 between 1/17/2020 and 12/21/2023, with symptoms persisting until enrollment for 4.3 to 52.9 months (median = 27.5 months). A majority of subjects reported only one confirmed COVID-19 infection; however, nineteen subjects had multiple infections, with three reporting three infections, one having four infections, and one having more than four infections. Since Delta, a variant of concern [50], dominated US circulation from June 26 to Dec 18, 2021, we estimated n = 22 were pre-Delta variants (original, Alpha, and other); 27 were post-Delta variants (mostly Omicron family); and only five were likely the Delta variant. The majority of patient cases were considered as mild per CDC standard, as only eight were hospitalized due to COVID-19 infection, among which two were intubated. The majority of subjects were fully vaccinated (n = 47) at the time of study enrollment, but only 21 subjects were fully vaccinated prior to their COVID-19 infection. As the initial step following recruitment, patients were asked to complete a REDCap questionnaire assessing a range of sensory complaints and cognitive function; however, only n = 55 completed.
Table 1 presents the number and percentage of individuals self-reporting sensory issues alongside confirmed deficits based on specific sensory tests. Overall, there was a high prevalence of self-reported sensory complaints. Subjects’ self-reporting of smell and taste complaints largely overlapped. Of those who reported ongoing chemosensory impairments (38/55 = 69.1%), 92.1% (35/38) reported taste impairment, while 97.3% (37/38) reported smell impairment. Among those reporting smell impairment, 17/37 (46%) reported a predominantly diminished sense of smell (anosmia or hyposmia), while 20/37 (54%) reported smell distortion (parosmia/phantosmia) in addition to a diminished sense of smell. Among those reporting taste impairment, 19/33 (57.6%) reported a predominant diminished sense of taste (hypogeusia), while 14/33 (42.4%) reported taste distortion (dysgeusia) in addition to diminished taste. Dizziness and balance issues were reported by 56.5%, hearing problems by 31.8%, and cognitive difficulties, particularly brain fog, by 51.3%. Finally, 7 subjects (16%) reported subjective unilateral or bilateral tinnitus....
snip
Conclusions
Current research suggests that long COVID affects everyone and every sensory system differently, shaped likely by initial disease severity, aging and cognitive impairments post-COVID-19. For chemosensory dysfunctions, the type of dysfunction (diminished vs distorted), as well as common central processing (i.e., identification), may significantly impact patients’ symptom profile and sensory testing outcome. Strong associations were found between confirmed vestibular and auditory dysfunction, and to cognitive impairment, implicating likely involvement of central mechanisms. Combinations of different sensory tests may better reflect self-reported dysfunctions, suggesting added value in multi-modal assessments. A future detailed study on the systematic impacted by long COVID—both objective and subjective—could further help to answer why the symptoms subjects experience vary drastically and to improve patient care.
https://link.springer.com/article/10.1186/s12916-026-04726-x
- Research
- Open access
- Published: 27 February 2026
- Volume 24, article number 136, (2026)
- Ahmad Odeh,
- Veronica L. Formanek,
- Christina Smith,
- Nidhi Jha,
- Junichi Tajino,
- Jessica H. Lewis,
- Lauren Gastineau,
- Shivani Patel,
- Songzhu Zhao,
- Lai Wei,
- Aaron C. Moberly,
- Daniel M. Merfeld,
- Christopher T. Simons,
- Megan J. Kobel &
- Kai Zhao
Background
This ongoing study investigates the impact of post-acute sequelae of SARS-CoV-2 infection (PASC) on broad sensory functions.
Methods
Sixty subjects aged 27–78 years were recruited who had contracted COVID-19 between 1/17/2020 and 12/21/2023 and had persistent symptoms (4.3–52.9, median = 27.48 months). Quantitative sensory assessments included (1) smell: 9-Item NIH toolbox odor identification, detection threshold to phenyl–ethyl alcohol, and retronasal candy test; (2) taste: modified NIH toolbox; (3) chemesthesis: nasal menthol lateralization thresholds and oral capsaicin identification; (4) hearing: pure-tone audiometry, otoacoustic emissions, words-in-noise recognition, and Dichotic Digits Test; (5) vestibular/balance: video head impulse testing, Subjective Visual Vertical, vestibular perceptual thresholds, and modified Romberg balance test; and (6) cognitive assessment: The Self-Administered Gero-Cognitive Exam, Digit Symbol Substitution Test, and Trail-Making Test.
Results
Overall, subjects self-reported high and overlapping dysfunctions: 67.3% smell, 63.6% taste, 56.5% balance and dizziness, 31.8% auditory, and 51.3% brain fog or cognitive dysfunctions, with varying discrepancy to the measured and confirmed deficits (smell 65.5%, taste 16%, vestibular 31.6%, hearing 53.4%, and cognition 19.1%). Significant associations were found between confirmed vestibular and auditory impairments (p = 0.04), and cognitive impairments with central components of vestibular (p = 0.03) and auditory (p = 0.01) impairments, but not with their peripheral components. Similarly, strong associations were found between identification components of smell and taste tests (p = 0.003), which may involve more central processing, but not with threshold tests (a more peripheral process). Hospitalization significantly associated with smell (p = 0.05), cognitive (p = 0.009), vestibular (p = 0. 001), and peripheral auditory (p = 0.01) dysfunctions. Age significantly associated with auditory (p = 0.02), vestibular (p = 0.01) and central olfactory (p = 0.03) dysfunctions.
Conclusions
COVID-19 impacts sensory systems broadly and differently, both peripherally and centrally, driven in part by aging, initial disease severity, and underlying post-COVID cognitive dysfunctions. Subjective symptoms may not always be corroborated by measured deficits.
Background
Since the outbreak in December 2019, the COVID-19 pandemic has ravaged the world, with over 775 million cases and more than 7 million deaths (as of March 2024, WHO). While most current patients with new COVID-19 infections report symptoms resembling those of the flu or common cold, neurological symptoms are still common among people with COVID-19, including sensory dysfunction. Smell dysfunction has become a hallmark symptom of COVID-19 and can present in isolation (the only symptom) or precede the occurrence of other symptoms [1,2,3,4]. The mechanism of smell dysfunction is unclear, potentially involving peripheral or central pathways. Expression of ACE2, the receptor targeted by SARS-CoV-2, is found not in olfactory sensory neurons (OSNs) but instead in supporting cells (sustentacular cells) and horizontal basal cells in the olfactory epithelium [5]. Sustentacular cells are functionally important for mucus composition (glucose, ion exchange, and gradients) and for structurally supporting OSN dendrites [6, 7]. Experimental manipulation of sustentacular cells has been shown to cause anosmia [6, 7]. Central mechanisms may also be involved, given reports of inflammation in the olfactory bulb [8].
Related to smell dysfunctions, early reported COVID-19-related “taste dysfunction” has been speculated to be due to impaired flavor perception consequent to retronasal smell impairment. However, recent quantitative measurements suggest that loss of taste function (beyond flavor perception) and chemesthesis (trigeminal function) may indeed exist in some COVID-19 patients [9,10,11,12]. There is no previous evidence that viral upper respiratory infections, such as flu or cold, would cause loss of taste or chemesthesis functions, and it remains unclear how these dysfunctions manifest in different taste sub-modalities.
In other sensory systems, up to 15–20% of patients report subjective changes in hearing, tinnitus, and/or dizziness [13, 14]. Evidence of potential auditory and vestibular impairments is additionally supported by several case studies of COVID-19-related sudden hearing loss or vertigo and cross-sectional studies identifying sensorineural hearing loss (SNHL) and/or changes in cochlear function [14,15,16,17,18]. While there is a strong precedent of viral infection affecting inner ear function (e.g., Zika, West Nile, herpes simplex) [19], the exact impact of COVID-19 on inner ear function is unknown. Several mechanisms have been suggested, including direct neural effects, neural inflammation, and hypoxic insults [13, 18, 20].
Recent clinical observations also reveal that a significant portion (73–76%) of COVID-19 patients may still have long-term and persistent symptoms 60–180 days after diagnosis [1]; the most common symptom to persist is fatigue or muscle weakness, as well as neurological disorders; one study reported that over 91% of patients, whether or not they had a neurological diagnosis when first hospitalized for COVID-19, had such problems 6 months after discharge [21]. It is unclear how multisensory dysfunctions manifest among these long COVID patients or how the prolonged vs acute phases differ. We aimed to use a comprehensive set of sensory tests to determine the long-term impact of COVID-19 on broad sensory functions (smell, taste, chemesthesis, hearing, and vestibular/balance function).
Methods
Recruitment
Patients with an ICD10 code U09.9 (“Post COVID-19 condition, unspecified”) were recruited through electronic medical records direct messaging. All interested subjects were then asked to complete a questionnaire probing the presence and severity of sensory dysfunctions (see Additional file 1). To be eligible, they must have longstanding complaints (minimum 90 days) in at least one sensory aspect: smell, taste, hearing, balance, or dizziness. Exclusion criteria (to rule out non-COVID-related pre-existing conditions) included:
- Olfactory, taste, or hearing dysfunctions due to any reason other than COVID-19.
- Any prior balance issue unrelated to COVID-19.
- History of severe head trauma or traumatic brain injury.
- Any previously existing neurologic conditions and severe health problems unrelated to COVID-19.
- Current pregnancy
This was a longitudinal prospective cohort study approved by the Ohio State Biomedical Sciences Institutional Review Board (approval number: 2022H0414). All subjects signed consent and authorization forms for participation and to enable their data to be included in this study. Olfactory, taste, chemesthesis, auditory, and vestibular functions were tested extensively: (1) smell: 9-Item NIH toolbox odor identification (ID) [22], Ortho nasal odor detection threshold (ODT) to phenyl–ethyl alcohol (PEA) [23], and retronasal candy test [24, 25]; (2) taste: modified NIH toolbox [26]; (3) chemesthesis: nasal menthol lateralization thresholds [27, 28], oral perception of capsaicin solution, which activates TRPV1 trigeminal nociceptors; (4) hearing: pure-tone audiometry (PTA) [29], otoacoustic emissions (OAE) [30], words-in-noise (WIN) recognition [31, 32], and Dichotic Digits Test (DDT) [33, 34]; (5) vestibular/balance: video head impulse testing (vHIT) [35], Subjective Visual Vertical (SVV) [36, 37], vestibular perceptual thresholds [38, 39], and modified Romberg balance test [40]; (6) cognitive assessment: Self-Administered Gero-Cognitive Exam (SAGE) [41], Digit Symbol Substitution Test (DSS) [42], and Trail-Making Test (TMT-A and TMT-B) [43]. The testing was split in two sequential sessions: (1) chemosensory testing, (2) hearing, vestibular and balance testing, with each session taking about 2–3 h. Please see Additional file 2: Supplementary Methods [44,45,46,47,48,49] for details of these tests.
Statistical analyses
Fisher’s exact test (univariate analysis) was used to examine the association between each pair of complaints including smell, taste, auditory, vestibular, and cognitive. Sensitivity and specificity of each measure, as well as the combination of measures, relative to patients self-reported complaints of sensory dysfunction were determined. Differences between subgroups (i.e., those reporting different symptoms) were examined using t-tests, ANOVAs, or non-parametric equivalents when normality of data was not identified. To quantify relationships between test measures, regression analyses or Pearson correlations were performed. Given that these analyses are preliminary and exploratory in nature, post-hoc corrections for multiple comparisons were not applied; results should be interpreted with caution. These analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).
Results
Sixty subjects (aged 27–78 years, median 51 years) with long COVID symptoms lasting more than 3 months were recruited. They contracted COVID-19 between 1/17/2020 and 12/21/2023, with symptoms persisting until enrollment for 4.3 to 52.9 months (median = 27.5 months). A majority of subjects reported only one confirmed COVID-19 infection; however, nineteen subjects had multiple infections, with three reporting three infections, one having four infections, and one having more than four infections. Since Delta, a variant of concern [50], dominated US circulation from June 26 to Dec 18, 2021, we estimated n = 22 were pre-Delta variants (original, Alpha, and other); 27 were post-Delta variants (mostly Omicron family); and only five were likely the Delta variant. The majority of patient cases were considered as mild per CDC standard, as only eight were hospitalized due to COVID-19 infection, among which two were intubated. The majority of subjects were fully vaccinated (n = 47) at the time of study enrollment, but only 21 subjects were fully vaccinated prior to their COVID-19 infection. As the initial step following recruitment, patients were asked to complete a REDCap questionnaire assessing a range of sensory complaints and cognitive function; however, only n = 55 completed.
Table 1 presents the number and percentage of individuals self-reporting sensory issues alongside confirmed deficits based on specific sensory tests. Overall, there was a high prevalence of self-reported sensory complaints. Subjects’ self-reporting of smell and taste complaints largely overlapped. Of those who reported ongoing chemosensory impairments (38/55 = 69.1%), 92.1% (35/38) reported taste impairment, while 97.3% (37/38) reported smell impairment. Among those reporting smell impairment, 17/37 (46%) reported a predominantly diminished sense of smell (anosmia or hyposmia), while 20/37 (54%) reported smell distortion (parosmia/phantosmia) in addition to a diminished sense of smell. Among those reporting taste impairment, 19/33 (57.6%) reported a predominant diminished sense of taste (hypogeusia), while 14/33 (42.4%) reported taste distortion (dysgeusia) in addition to diminished taste. Dizziness and balance issues were reported by 56.5%, hearing problems by 31.8%, and cognitive difficulties, particularly brain fog, by 51.3%. Finally, 7 subjects (16%) reported subjective unilateral or bilateral tinnitus....
snip
Conclusions
Current research suggests that long COVID affects everyone and every sensory system differently, shaped likely by initial disease severity, aging and cognitive impairments post-COVID-19. For chemosensory dysfunctions, the type of dysfunction (diminished vs distorted), as well as common central processing (i.e., identification), may significantly impact patients’ symptom profile and sensory testing outcome. Strong associations were found between confirmed vestibular and auditory dysfunction, and to cognitive impairment, implicating likely involvement of central mechanisms. Combinations of different sensory tests may better reflect self-reported dysfunctions, suggesting added value in multi-modal assessments. A future detailed study on the systematic impacted by long COVID—both objective and subjective—could further help to answer why the symptoms subjects experience vary drastically and to improve patient care.
https://link.springer.com/article/10.1186/s12916-026-04726-x