Olfactory Dysfunction: A Highly Prevalent Symptom of COVID-19 With Public Health Significance
Olfactory Dysfunction: A Highly Prevalent Symptom of COVID-19 With Public Health Significance
Olfactory Dysfunction: A Highly Prevalent Symptom of COVID-19 With Public Health Significance
Otolaryngology–
Head and Neck Surgery
T
Abstract he 2019 coronavirus disease (COVID-19) is caused by
Objective. Coronavirus disease 2019 (COVID-19) is a global a novel coronavirus, referred to as severe acute respira-
pandemic affecting millions of individuals, killing hundreds of tory syndrome coronavirus 2 (SARS-CoV-2).1 COVID-
thousands. Although typically described with characteristic 19 was initially and is still most commonly described by char-
symptoms of fever, cough, and shortness of breath, greater acteristic symptoms of fever, cough, and shortness of breath.
understanding of COVID-19 has revealed myriad clinical However, COVID-19 is now recognized to be highly heteroge-
manifestations. Olfactory dysfunction (OD)—hyposmia and neous in severity—ranging from asymptomatic (or symptoms
anosmia—has recently been recognized as an important too mild to be noticed) to severe acute respiratory distress syn-
symptom of COVID-19 and increasingly gained traction as a drome and death—and in symptomatology, which extends to
public health tool for identifying COVID-19 patients, in par- include otolaryngologic symptoms. While patients with the
ticular otherwise asymptomatic carriers who, unawares, may well-known and characteristic symptoms of COVID-19 may
be major drivers of disease spread. The objective of this be readily identified (by themselves or others), patients experi-
study is to review the scientific evidence about anosmia in encing nonclassical symptoms of COVID-19 may be missed
COVID-19. and may serve as vectors for transmission, and therefore persis-
tence, of the disease. In the past few weeks, anecdotal reports
Data Sources. PubMed, Google Scholar, and Web of Science. from around the world have suggested that olfactory dysfunc-
Review Methods. Comprehensive literature search of primary tion (OD)—in particular, anosmia—may be an important
studies pertinent to the objectives of this review using the symptom of COVID-19 and also a significant predictor of
chosen data sources. asymptomatic COVID-19 carriers.
Conclusions. Current evidence shows that OD is highly pre- Methods and Results
valent in COVID-19, with up to 80% of patients reporting The PubMed, Google Scholar, and Web of Science data-
subjective OD and objective olfactory testing potentially bases were searched with combinations of search terms
showing even higher prevalence. OD is frequently accompa- including COVID-19, SARS-CoV-2, coronavirus, anosmia,
nied by taste dysfunction. Up to 25% of COVID-19 patients hyposmia, and olfactory dysfunction for any articles that
may experience sudden-onset OD as the first symptom. A were indexed up to April 18, 2020. While preprints were
large proportion of COVID-19 OD cases may resolve over considered for discussion, only peer-reviewed articles are
the period of a few weeks. presented in the evidence table. Due to the rapid prolifera-
Implications for Practice. Sudden anosmia should be consid- tion of peer-reviewed scientific evidence that may not yet
ered a symptom of COVID-19. Assessing for sudden-onset be database-indexed, online forums from the American
anosmia may increase sensitivity of COVID-19 screening Academy of Otolaryngology–Head and Neck Surgery and
strategies, in particular for identifying patients at the earliest American Rhinologic Society were also used to identify
stages of disease. Since many cases of OD due to COVID-
1
19 may resolve in the short term, conservative management, Department of Otolaryngology–Head and Neck Surgery, University of
including observation, is reasonable, while advanced imaging Cincinnati Medical Center, Cincinnati, Ohio, USA
2
is unnecessary. Klinik für Hals-, Nasen-, Ohren- Krankheiten, Hals-und Gesichtschirurgie,
Kantonsspital Aarau, Switzerland
studies. Only studies of COVID-19 patients were included, A high prevalence for OD in COVID-19 has also been
although case reports were not included in the final evi- described using objective testing of olfactory function. In a
dence table. Only articles in English were included. A total study by Moein et al,9 the mean University of Pennsylvania
of 6 studies representing the existing peer-reviewed evi- Smell Identification Test score in 60 Iranian COVID-19
dence on OD in COVID-19 were identified (Table 1). patients was 20.98 (indicative of severe microsmia) com-
pared to 34.10 (indicative of normosmia) in 60 age-matched
Discussion controls. Moein et al9 found that 59 COVID-19 patients
(98%) were experiencing at least some hyposmia: 8 (13%)
Incidence of Olfactory Dysfunction in Relation
with mild microsmia, 16 (27%) with moderate microsmia,
to the COVID-19 Pandemic 20 (33%) with severe microsmia, and 15 (20%) with anos-
In March 2020, anecdotal reports from around the world mia. In comparison, only 18% of 60 age-matched controls
described a dramatic increase in the recent incidence of OD, showed evidence of hypmosia, all of whom were quantified
in particular anosmia, which seemed to parallel the spread as having mild microsmia. While the etiology of OD in
of COVID-19. A study by Hopkins et al2 showed that over COVID-19 remains unknown, 2 case reports have shown
February and March 2020 in the United Kingdom, the inci- negative magnetic resonance imaging findings, with normal-
dence of patient-reported OD followed an exponential appearing olfactory bulbs,10,11 although 1 case report found
increase similar to the incidence of COVID-19. A study by mucosal obstruction of the olfactory clefts on computed
Bagheri et al,3 available as a preprint, showed that the inci- tomography.11
dence of patient-reported OD across the 31 provinces of
Iran since the onset of the COVID-19 epidemic there was Olfactory Dysfunction as a Predictor of COVID-19
highly correlated with the incidence of COVID-19 during In addition to being a highly prevalent symptom of COVID-
that time. These studies provided strong circumstantial evi- 19, anecdotal reports have suggested that anosmia may also
dence for the relationship between OD and COVID-19. be used as a predictor of COVID-19. In their study, Yan et
al8 reported that in patients presenting with flu-like symp-
Olfactory Dysfunction as a Symptom of COVID-19 toms, OD was associated with COVID-19 with an adjusted
Although the prevalence of nasal symptoms such as conges- odds ratio of 10.9. In contrast to the context of concomitant
tion or rhinorrhea in COVID-19 is reported on the order of flu-like symptoms, a case report by Gane et al12 described a
5%,4 OD has been reported as an increasingly prevalent COVID-19 patient who presented with sudden-onset anos-
symptom of COVID-19. A short communication by Vaira et mia but no other symptoms. Gane et al12 also described a
al5 first reported chemosensory dysfunction—anosmia or series of 10 new patients presenting with OD during the
ageusia—in at least 19.4% of a cohort of 320 Italian COVID-19 pandemic, half of whom had other viral prodro-
COVID-19 patients. Another study of 59 hospitalized Italian mic symptoms and the other half reporting only anosmia.
COVID-19 patients found that 23.7% of patients com- Consistent with the observation of Gane et al,12 a recent
plained of OD, the majority with concomitant taste altera- study led by Dominique Salmon, MD, PhD (Hôtel Dieu,
tion as well.6 Reported at the height of the COVID-19 Paris), and Alain Corré, MD (Hôpital Fondation Adolph de
epidemic in Italy and with the stated intention of raising Rothschild, Paris)—the results of which were directly com-
awareness of chemosensory dysfunction as a symptom of municated to the authors of this commentary while the
COVID-19, the authors hypothesized that the prevalence of manuscript was in review—found that 94% of patients pre-
chemosensory dysfunction could be even higher given their senting to them with anosmia but no other nasal symptoms
limitations for assessment. and none of the classical respiratory symptoms of COVID-
In fact, subsequent studies have suggested that preva- 19 tested positive for COVID-19.4 These findings suggest
lence of OD in COVID-19 is much higher. A multicenter that during the COVID-19 pandemic, isolated sudden-onset
European study of 417 COVID-19 patients by Lechien et OD could be used to screen for and identify asymptomatic
al,7 representing the first systematically collected data pub- carriers.
lished regarding OD in COVID-19, found that 85.6% of The utility of anosmia to identify asymptomatic carriers
patients reported subjectively decreased sense of smell in must also be interpreted in the context of how often
association with COVID-19, which was closely correlated COVID-19 patients will only exhibit anosmia as a symptom.
with gustatory dysfunction. Of those reporting decreased In the 417-patient cohort of Lechien et al,7 11.8% reported
sense of smell, 79.6% reported a complete loss. A subse- OD as the first symptom of COVID-19. In comparison, OD
quent study by Yan et al8 described that of 59 patients pre- occurred at the same time as other COVID-19 symptoms in
senting to their institution with flu-like symptoms and 22.8% and after other COVID-19 symptoms in 65.4%. In
testing positive for COVID-19, 68% reported decreased another cohort of 237 COVID-19 patients with OD sub-
sense of smell, which was also closely correlated with mitted to the American Academy of Otolaryngology–Head
decreased sense of taste. In comparison, only 16% of a con- and Neck Surgery COVID-19 Anosmia Reporting Tool,
trol group of 203 COVID-19 negative patients presenting anosmia was reported to be the first symptom of COVID-19
with flu-like symptoms during the same time period in 27% of patients.13 Thus, up to a quarter of COVID-19
reported a decreased sense of smell.8 patients could potentially be identified at the earliest stages
14
Table 1. Table of Evidence.
Author Patient Population Key Outcome Measures Measurement Tool Main Results
Vaira et al5 320 patients with COVID-19 1. Anosmia Not described 1. 19.4% prevalence of chemosensory dysfunction, without distinction
2. Ageusia of anosmia vs ageusia
Giacomelli et al6 59 patients with COVID-19 1. OD Patient reports 1. 23.7% of patients reported OD with 78.6% of those patients also
2. Gustatory dysfunction reporting gustatory dysfunction
Lechien et al7 417 patients with mild to 1. OD prevalence Patient reports 1. 85.6% prevalence of OD
moderate COVID-19 2. Gustatory dysfunction 2. Of patients reporting OD, 79.6% reported anosmia
recruited from 11 hospitals 3. Associated nasal, systemic, 3. 88.0% of gustatory dysfunction
across Belgium, France, Spain, and pulmonary symptoms 4. OD occurred as the first symptom in 11.8% of cases
and Italy 4. Timing of OD relative to 5. OD occurred at the same time as other symptoms in 22.8%
other COVID-19 symptoms 6. OD occurred after other COVID-19 symptoms in 65.4%
5. Time to resolution of OD 7. In the subset of clinically cured patients, 44% reported resolution of
OD with almost three-quarters experiencing resolution within first 8
days after resolution of COVID-19
Yan et al8 59 patients with COVID-19 1. OD Patient reports 1. In COVID-19 patients, 68% reported OD and 71% reported taste
presenting with flu-like 2. Taste loss loss
symptoms and 203 patients 3. Resolution of OD 2. In COVID-19 negative patients, 16% reported OD and 17% reported
negative presenting with flu- taste loss, which was significantly less compared to COVID-19
like symptoms patients
3. In patients presenting with flu-like symptoms, OD was associated
with COVID-19 with adjusted odds ratio of 10.9
4. 74% of COVID-19 patients with OD experienced improvement of
OD with improvement of COVID-19
Kaye et al13 237 patients with COVID-19 1. Timing of anosmia relative to Physician reports 1. Anosmia occurred as the first symptom of COVID-19 in 27%
and anosmia from around the other symptoms and 2. Anosmia was present in 73% prior to diagnosis
world submitted to the AAO- diagnosis 3. Anosmia contributed to COVID-19 testing in 40%
HNS Anosmia Reporting Tool 2. Resolution of anosmia 4. At least some improvement of anosmia experienced by 27% with
mean time of 7.2 days
Moein et al9 60 Iranian patients with COVID- 1. OD Patient reports 1. In the COVID-19 cohort, 35% of patients had smell/taste complaints
19 and 60 approximately age- and UPSIT while 0% of controls had these complaints
matched controls 2. COVID-19 patients had significantly lower UPSIT score (mean =
20.98 reflective of severe microsmia) compared to controls (mean =
34.10 reflective of normosmia)
3. 98% of COVID-19 patients had objective evidence of OD while only
18% of patients in the control cohort had evidence of OD
Abbreviations: AAO-HNS, American Academy of Otolaryngology–Head and Neck Surgery; COVID-19, coronavirus disease 2019; OD, olfactory dysfunction; UPSIT, University of Pennsylvania Smell Identification
Test.
Sedaghat et al 15
Disclosures
Competing interests: None.
Sponsorships: None.
Funding source: None.