Olfactory and Gustatory Dysfunctions in Patients With COVID-19 in Wuhan, China

Background: The coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 isspreading all over the world.The main symptoms of COVID-19 include fever, cough,fatigue, and myalgia. However, there are few reports onolfactoryand gustatory dysfunctions in patients with COVID-19. Objective: To investigate the incidence of olfactory and gustatory dysfunctions in patients with conrmed COVID-19 infection, in Wuhan, China. Methods:In this retrospective study,we collected 81 conrmed cases of COVID-19 from the Renmin Hospital of Wuhan University, from February 2020 to March 2020, and analyzed the demographic characteristics, clinical manifestations (including olfactory and gustatory dysfunctions), laboratory ndings,and comorbidities. Results: A total of 81 conrmed COVID-19 patients were enrolledin this study (38 males). The most prevalent symptoms include cough, myalgia, and loss of appetite. On admission, 25 (30.9%) of all patients reported either olfactory dysfunction (OD) or gustatory dysfunction (GD), and 7 (8.6%) reported both OD and GD. 13.6% and 25.9% of allpatients reported OD and GD, respectively. OD and GD were not associated with disease severity. Pearson correlation analysisidentied some factors are positively correlated with OD and GD, including headache or dizziness (r = 0.342, P = 0.002), dark urine (r = 0.256, P = 0.021), IgM titer (r = 0.305, P = 0.01), and diabetes (r = 0.275, P = 0.013). In 81.8% of the cases with OD and 28.6% of the cases with GD, the symptomslasted for at least 1 month after discharge.3.6% of inpatients without OD developed OD after discharge. Conclusion:OD and GDare common in COVID-19.These symptoms appear early during thecourse of and may last for at month.The diabetics, and IgM titers.

The high spread and low fatality rate of COVID-19 makes its global pandemic. At present, COVID-19 became the largest epidemic since the H1N1 u outbreak in 1918. Early genetic tests emphasized that the genome sequence identity of SARS-CoV-2 and SARS-CoV is 79.5%. Both viruses infect cells through angiotensin-converting enzyme 2 (ACE2) [6]. The similarity of the gene sequence and infection route highly suggests that SARS-CoV-2 and SARS-CoV have similar clinical manifestations. Similar to SARS, the most common symptoms of COVID-19 are fever, cough, sputum, fatigue, and myalgia. Some cases reported gastrointestinal symptoms, such as diarrhea [2,7,8]. It is worth noting that SARS-CoV can be detected 60 to 66 h after infection and is most abundant in the olfactory bulb [9]. In addition, the virus may enter the brain through the olfactory bulb. Viral antigens can also be detected in brain areas connected to the olfactory bulb, such as the piriform nucleus and subliminal leather, basal ganglia, and midbrain [10,11]. Current research mainly focuses on the respiratory and gastrointestinal symptoms of COVID-19, while paying less attention to sensory symptoms such as olfactory and gustatory dysfunctions.
The purpose of this study is to investigate the incidence of olfactory and taste disorders in con rmed COVID-19 cases and to analyze the risk factors related to olfactory and gustatory dysfunctions in order to provide clues for the clinical diagnosis and treatment of COVID-19.

Patient enrollment and data collection
This descriptive study included 81 inpatients diagnosed as COVID-19 in the Renmin Hospital of Wuhan University, Wuhan, China. The diagnosis of COVID-19 was based on clinical symptoms, computed tomography (CT), real-time RT-PCR, and next-generation sequencing. The patient's admission date is February 1 to March 3, 2020. All patients participating in this study lived in Wuhan during the COVID-19 outbreak. Two well-trained investigators collected the data including demographic characteristics, clinical characteristics (including medical history, comorbidities, symptoms, olfactory and gustatory dysfunctions), preliminary laboratory ndings, treatment, and clinical outcomes.
According to the SARS-CoV-2 diagnosis and treatment guidelines (Versions 3-7) issued by the National Health Commission of China, the severity of COVID-19 patients was de ned. Severe cases were designated when the patients t one of the following criteria: 1) respiratory distress with respiratory rate ≥ 30/min; 2) oxygen saturation ≤ 93% at rest; 3) arterial partial oxygen pressure (PaO 2 )/oxygen absorption concentration (FiO2) ≤ 300 mmHg; 4) respiratory failure that needs mechanical ventilation; 5) shock or organ failure requiring ICU care.

Procedures
Two investigators collected the information on epidemiology, clinical symptoms, laboratory test results, treatment, and clinical outcomes. Laboratory tests include blood routine examination (including white blood cells, lymphocytes, and platelets), coagulation function, lactate dehydrogenation, and immune indicators (including lymphocyte differential counts and in ammatory factors). Four weeks after the patients were discharged from the hospital, we conducted a telephone follow-up, asking the patients to record their clinical symptoms, especially the sense of smell and taste. Cases with unsuccessful followup were excluded.

Statistical analysis
Continuous variables were described as medians and interquartile range (IQR), and categorical variables were described as frequency rates and percentages. We used the chi-square (χ 2 ) test or the Fisher's exact test to compare categorical data. Mann-Whitney-Wilcoxon test was applied to compare nonnormally continuous variables. Pearson correlation analysis was used to examine the correlations between olfactory/gustatory dysfunctions and all indicators (including epidemiology, clinical symptoms, laboratory test results, treatment, and disease severity). The sample size varied due to missing data, and missing data were not imputed. The analyses regarding different factors were based on non-missing data. All statistical analyses were performed using SPSS software (V.23.0). Two-tailed P values were considered statistically signi cant at < 0.05. This study was approved by the Hospital Ethics Committee of the Renmin Hospital of Wuhan University (WDRY2020-K136).

Demographics and clinical characteristics
The demographic characteristics, clinical manifestations, and comorbidities are shown in Table 1.Of the 81 patients, 63 caseswere diagnosed as non-severe disease and 18 cases were severe. The median age was 58 years (IQR, 50.0 to 68.5 years). The median age of the severe group is 10.5 years higher than that of the non-severe group.Males accounted for 46.9% (38 of 81) of the total patients. 66.7% (12 of 18) of the patients were males in the severe group, comparedwith 41.3% (26 of 63)of males in the non-severe group.On admission, the most commonclinical manifestations were fever (72.8%) and respiratory symptoms (70.4%), while chest pain, fatigue, gastrointestinal symptoms, dark urine, and headache or dizziness accounted for 27.2%, 37%, 27.2%, 12.3%, 19.8%, and 14.8% of the patients, respectively.Compared with the non-severe group, the severe group had a higher proportion of dark urine (17.5% vs 27.8%).Among the 81 cases, 51.9% had at least one comorbidity, including hypertension, diabetes, coronary heart disease, stroke, cancers, and chronic pulmonary diseases. Compared with the non-severe group, more patients in the severe group had a history of stroke (0% vs 5.6%) and coronary heart disease (1.6% vs 11.1%). Data are presented as medians (interquartile ranges, IQR) and n/N (%).
*: Cancers referred to any malignancy.
∮ Chronic pulmonary disease includes tuberculosis, chronicobstructive pulmonary disease, and bronchiectasis. All cases were stable and no obvious bacterial infections.  The factors related to OD and GD are summarized in Table 3.We brought general conditions, symptoms, comorbidities, and laboratory tests into consideration and calculate the Pearson correlation coe cient.We foundsome factors are positively correlated with OD, including headache or dizziness (r = 0.342, P = 0.002), dark urine (r = 0.256, P = 0.021), IgMtiter (r = 0.305, P = 0.01), and diabetes (r = 0.275, P = 0.013).These factors are also positively correlated to GD (all r>0.2, P<0.05).The IgG titers were neither signi cantly related to patients with OD (r = 0.083, P = 0.496) nor with GD (r = 0.164, P = 0.176). But IgG titers were associated with the incidence of patients with either OD or GD (r = 0.269, P<0.05). OD: olfactory dysfunction; GD: Gustatory dysfunctions Table 4 describes the incidence of OD and GDone month after the patients were discharged from the hospital.6 patients had residual GD, and 11 patients reported OD.Of patients with OD or GD during hospitalization, 48.0% still complained OD or GD 1 monthafter discharge.In patients with no OD or GD during hospitalization, 3.6% deported OD or GD after discharge. We also analyzed the relationship between treatmentand the incidences of olfactory and urinary dysfunction after discharge (Supplementary Table 1), and found that treatmentshad no effect on the clinical outcomes of OD and GD after patients were discharged.

Discussion
This retrospective descriptive study included 81 con rmed cases of COVID-19 with a median age of 58 years. The median age of the severe group was 11.5 years older than that of the non-severe group. The ratio between males and females was about 0.88 (46.9% vs 53.1%). However, males accounted for 66.7% in the severe group, compared with 41.3% in the non-severe group. These results seem to suggest that males and older patients are more likely to develop severe disease, which is consistent with other reports [12]. On admission, fever and respiratory symptoms were the most common clinical manifestations. Compared with the non-severe group, there were more patients with dark urine in the severe group. This might suggest a potential relationship between urinary system dysfunction and disease severity. Among 81 patients, more than half had at least one comorbidity. Patients with cardiovascular and cerebrovascular diseases are more likely to develop severe illness. These above ndings are consistent with the previous studies [2,8,13]. This study focuses on the OD and GD in COVID-19 patients in Wuhan.
On admission or during hospitalization, patients self-reported OD, GD, and both accounted for 13.6%, 25.9%, and 8.6% of all subjects, respectively. The incidence of OD and GD in the severe cases was not signi cantly different from the non-severe cases, suggesting OD and GD are not related to the severity of the disease. In addition, 48.0% of patients had persistent OD or GD after the other clinical manifestations of COVID-19 disappeared.
Most of the previous trials focused on the patient's respiratory symptoms, sensory disturbances were largely ignored. There are currently few studies on the incidence of OD and GD in Asia. One study found that 5.6% of COVID-19 patients reported hypogeusia, while 5.1% reported hyposmia [14]. Compared with this study, we found a higher rate of hypogeusia (13.6%) and hyposmia (25.9%). A multicenter European study conducted a quantitative analysis of patients with OD and GD, showing that 66.2%, 13.5%, and 88.8% of patients suffer from anosmia, hyposmia, and gustatory disorders, respectively [15]. These results indicate that compared with Asian patients, olfactory disorders are more common in European patients.
The reasons for the different incidence of OD and OD may be as follows: 1) Quantitative measurement is more sensitive than self-report. Studies have shown that only 35% of patients are aware of their olfactory de cits [16]. 2) The different a nity of the virus to different populations may lead to clinical differences between patients in different regions. A study showed that ACE2 mutations can reduce the correlation between human ACE2 and SARS-CoV S-protein, thereby reducing the chance of infection [17]. ACE2 polymorphisms and the differences in expression levels between Asian and European populations may explain the difference in olfactory dysfunction between Asian and European populations [18]. 3) Different strains of virus may cause different clinical manifestations.
In addition, Pearson correlation coe cient showed that headache or dizziness, dark urine, IgM, and diabetes all showed a positive correlation with OD and GD (all P < 0.05). We noted that chronic rhinitis and certain neurodegenerative diseases may directly lead to taste or smell disorders. In this study, such patients had been excluded. Headache or dizziness are the most common symptoms of the nervous system. It has been reported that the incidence of headache and dizziness in COVID-19 patients were 13.1% and 16.8%, respectively [14]. OD and GD are positively related to dizziness and headache. It has been reported that ACE2 is highly expressed in the nasal goblet and ciliated cells [19]. The virus may infect the olfactory nerve early and cause OD, before other neurological manifestations. OD and GD were also signi cantly related to dark urine. This may suggest that dark urine, like OD/GD, were signs of early infection and damage to the urinary system. IgM is an early antibody produced by the immune system after infection. The interaction of infectivity, virulence, and immune response may explain the positive correlation between IgM and OD/GD. Consistent with other results, diabetes was signi cantly correlated with smell dysfunction [20]. The infection might exacerbate potential nerve damage in diabetic patients.
One month after discharge, some patients still suffer from OD and GD. A few patients who did not have OD or GD during hospitalization developed these symptoms after discharge. Considering the damage of the virus to the olfactory or gustatory nerve, this may be one of the possible sequelae of COVID-19 patients. Our analysis found that drug treatment is not related to the recovery of OD and GD in patients with COVID-19, suggesting that the OD and GD are most likely to be the primary symptoms caused by SARS-CoV-2, rather than the side effects of drugs. It is worth noting that SARS-CoV can be detected 60 to 66 h after infection and is most abundant in the olfactory bulb [21]. While the infection routes of SARS-COV-2 and SARS-COV are similar, OD may be an early symptom of COCID-19. Early screening of people with OG and GD, early detection of virus infections, and early isolation of COVID-19 patients can help prevent the spread of COVID-19.
In conclusion, we found that OD and GD are common symptoms of COVID-19. They appear early during the course of disease and may last for at least 1 month. Headache or dizziness, IgM titers, and diabetes are correlated with the occurrence of OD and GD. This retrospective study has some limitations. First, we