Covid 19 pandemic: Paranasal diagnostic imaging in patients with olfactory loss

Erkan Yıldız (  dr.erkanyildiz@hotmail.com ) Afyonkarahisar Şuhut State Hospital, Department of Otorhinolaryngology https://orcid.org/0000-0002-0265-7327 Aydın Balcı Afyonkarahisar Healty Science University,Department of Chest Disease https://orcid.org/0000-0002-6723-2418 Okan Selendili Manisa Alaşehir State Hospital, Department of Chest Disease https://orcid.org/0000-0002-4902-974X Selçuk Kuzu Afyonkarahisar Healty University,Department of Otorhinolaryngology https://orcid.org/0000-0002-0511-9874


Introduction
New Coronavirus disease  is an infectious disease caused by the SARS-CoV-2 virus , which started in Wuhan, China and spread all over the world [1]. Coronavirus disease has been declared as a pandemic by the World Health Organization (WHO) on March 11, 2020[2]. It starts as a mild respiratory infection in most people. It is quite severe in patients with chronic disease such as hypertension, diabetes, chronic renal failure, asthma, Chronic obstructive pulmonary disease (COPD) and patients with advanced age. The most common symptoms of the disease are fever, cough, headache, myalgia, dyspnea and diarrhea [3]. Smell and taste disorder started to draw attention as the symptom that started as the earliest and ended as the latest [4]. The disease starts as a simple respiratory infection and causes pneumonia in a short time. The virus, which causes the radiologically ground glass appearance in the patient's lung, progresses to ARDS (Acut Respiratory Distress Syndrome) and causing deaths [5]. In diagnosis, a rapid antigen test or a SARS-CoV-2 identi cation with Polymerase Chain Reaction (PCR) is used. There is no speci c treatment for COVID-19 disease. Although treatment guidelines vary from country to country, hydroxychloroquine therapy, azithromycin, favipravir, lopinavir + ritonavir, remdesevir, ivermectin and tocilizumab drugs such as plasma treatment is used. Although effective medication and treatment-seeking studies continue, the gold standard treatment has not been established yet [6][7][8].
COVID-19 infects the respiratory epithelium. Hence it starts with symptoms in this area. In the Coronavirus epidemic, the most noticeable symptom recently has been the loss of smell and taste [9]. Recently, reports of hyposmia or anosmia have been declared from many countries. The olfactory loss rate in COVID-19 is between 34-68%.
Otolaryngolists thought it might be a strong and early marker of COVID-19 infection, focusing on loss of smell and taste [10,11].
Post-viral olfactory loss has been described in many viral infections. Post-viral olfactory dysfunction (PVOD) is the most common cause of acquired olfactory dysfunction [12,13]. Post-viral olfactory dysfunction is mostly caused by Rhinoviruses, In uenza, Parain uenza and older types of Coronaviruses [14]. COVID-19 may also appear as the rst symptom. The cause of this loss can be considered as mucosal in ammation and edema in the nose and complete Page 3/11 closure of the airway in the olfactory region. It is not yet known whether smell loss is long-term in COVID-19. However, in some patients, it has been reported to continue even after the disease has passed [4,10]. Although there are many new studies on COVID-19 related to loss of smell, no study has yet been conducted on the radiological ndings in the paranasal sinuses [4,10,11]. In our study, patients who had olfactor loss in COVID-19 were tested both for smell and the olfactory regions were examined by taking Paranasal Sinus Tomography (CT). The most frequently used scent tests are "The University of Pennsylvania Smell Identi cation Test (UPSIT)", "Connecticut (CCCRC) Olfactory Test", "Sni n 'Sticks" Olfactory Test " [15][16][17]. In our study, we used the Connecticut (CCCRC) Olfactory Test, which is easy and fast to apply, and we tried to elucidate the pathophysiology of COVID-19 olfactory disorder by comparing the degrees of odor disorder and radiological ndings.

Materials And Methods
In this study, 40 patients diagnosed with COVID-19 in our center between March 2020 and May 2020, and who have a smell disorder, were evaluated. Any of these patients who previously had either an acute smell disorder that prevented smell due to any otolaryngology disease (chronic rhinosinusitis, allergic rhinitis, nasal polyposis, a deviated septum etc.) or Alzheimer's disease, or non-chronic smell disorders such as Parkinson's disease, or drug-induced smell disorders, were not been selected as patients. Patients in the study with fever, malaise, myalgia, dyspnea and cough as known Covidien-19 symptoms, as well as acute olfactory dysfunction since a month, and was diagnosed via the Polymerase Chain Reaction ( PCR ) with Covidien-19 as processor was selected as a patient. The diagnosis of PCR was made with Applied Biosystems GeneAmp® PCR System 9700 device (Thermo Fisher Scienti c, USA). Olfactory regions were interpreted radiologically by evaluating the Paranasal Sinus Tomographs (CT) of COVID-19 patients, who were referred to otolaryngology due to odor disorder. Those with complaints of olfactory disorders were diagnosed by making an odor test. A two-stage "Connecticut (CCCRC) Olfactory Test" was used for the odor test. The CCCRC test was performed in two stages as a butanol threshold test and an odor identi cation test as previously described. In the CCCRC test, scores below 5.75 were considered as olfactory disorders. It was accepted that the degree of odor loss in the disease increased as the CCCRC test score decreased. In addition, these patients were evaluated in terms of age, gender, smoking, additional disease, disease severity, treatments and mortality . The degree of the disease was divided into 1) mild (no pneumonia or mild pneumonia), 2) moderate (dyspnea, hypoxia or severe progress in lung ndings within 24 hours), and 3) severe (respiratory failure, shock, multi organ failure). Permission to study was given by the University of Health Sciences Afyonkarahisar Non-invasive Clinical Research Ethics Committee from (05.11.2020 / KAEK-2-2020-5) and the Republic of Turkey Ministry of Health, the Scienti c Research Commission (29/04/2020) . The study was carried out in accordance with the Declaration of Helsinki.

Diagnosis and Treatment of COVID-19
COVID-19 diagnosis and treatment for the Republic of Turkey Ministry of Health Science Board has bene ted from COVID-19 guidelines. Accordingly, patients with complaints such as fever, weakness, joint pain, headache, shortness of breath, cough, smell and taste disorder, and diarrhea were given a nasopharyngeal swab in the isolation rooms and diagnosed with a PCR device for the SARS-CoV-2 virus. Among these patients, an odor test was performed on patients with an odor disorder (hyposmia-anosmia) and a Paranasal Sinus CT was also taken. The "Connecticut (CCCRC) Olfactory Test" was used for the odor test. CT examinations with the device (MSCT, Philips Brilliance ICT 256; Philips Medical Systems, Netherlands) were performed on paranasal sinus with CT sections taken at 2 mm intervals (512x512matrix, voltage 100 kV, current 150 mAs). The images were made by bringing the head to hyperextension while the patient was in the prone position. Coronal CT sections were used for examination. Soft tissue thickening and fullness in the olfactory region of the paranasal Sinus CT coronal sections were accepted as positive tomography ndings. Hydroxychloroquine tablets 2 * 200 mg (5 days) + Azithromycin tablet (500 mg tablet was given on the rst day, and then for next 4 days, 250 mg / day was given as treatment, with Favipiravir 200 mg tablet (2 x 1600 mg loading, 2 x 600 mg ) for patients with severe pneumonia. All drugs were used for 5 days.

1) Butanol Threshold Test
For each trial, two identical glass bottles were presented to the patient. Solutions with a concentration of water and diluted butanol in one bottle were marked. The patients were asked to smell by plugging the single nostril and approaching the bottle with the tip of the nose. The strongest bottle of butanol contained 4% butanol. (Bottle 1) A total of 7 bottles were prepared by diluting 1/3 with deionized water. Results were scored between 1-7 for the CCCRC test.

Statistical Analysis
All values were calculated as mean ± standard deviation. The measurement-evaluation and statistical analysis methods of the research was as follows: the data obtained was evaluated with descriptive statistics (Arithmetic mean, median, standard deviation, percentage distributions). When comparing the mean between groups, the normal distribution suitability was rst evaluated by the Kolmogorov Smirnov and Shapiro Wilk tests. When comparing the percentage distributions of categorical data between groups, a Chi Square test and one wow anova were used.  (Figure 1a, 1b). There were 12 patients (%30), 20 patients (%50) and 8 patients (%20) graded according to the severity of the disease. There were 11 patients with hypertension, 9 with diabetes mellitus, 3 with COPD (Chronic obstructive pulmonary disease) , 1 with asthma, 1 with multiple sclerosis, 1 with Alzheimer's, 1 with Epilepsy, and 1 with Heart Failure. 11 patients did not have additional diseases. Hydroxychloroquine was given to all patients in the treatment. 36 patients (100%) were given En uvir, 6 patients were given Azithromycin and 3 patients were given Favipravir. Three patients needed intensive care and these 3 patients died. (Table 1     COVID-19 disease affects the respiratory epithelium. The disease starts like a cold and can turn into pneumonia in a short time. If the disease progresses further, it progresses to Acute Respiratory Distress Syndrome (ARDS) and leads to mortality. It is believed that viral load is important in worsening the illness. The viral load was reported to be highest in the lower respiratory tract -in the bronchoalveolar uid (93%), secondarily in the upper respiratory tract, nasopharynx (60%), and thirdly in the pharynx (30%). The disease is diagnosed by bronchoalveolar lavage uid, nasopharyngeal or swabs from the oropharynx, via an RT-PCR test. In addition, it can be diagnosed as a rapid antigen test with serological methods. The disease has been reported to be more severe in males, smokers, elderly and those with additional diseases [21]. In our study, the relationship of the disease to gender was not detected. Elderly and additional diseases were in uential in mortality (p <0.05, p = 0.046, respectively). No relation was found between smoking and the degree of smell disorder. (p> 0.05) Imaging in COVID-19 pneumonia is much more important than many viral diseases. There are even authors who report that the disease is as important as the PCR test. The classic nding in thorax CTs that are routinely seen in patients is the "ground glass view" [22,23]. COVID-19 recently reported cases with pulmonary imaging, Abdomen CT and Brain CT. Encephalitis has been reported in the brain as in some viral infections. A possible way for this to occur is via nasopharynx, sphenoid sinus, frontal sinus and cerebrospinal uid [24].
Post-viral odor loss has been identi ed in many viral diseases, primarily In uenza and Rhinoviruses [12][13][14]. Patients with sudden olfactory loss were even reported in the MERS-CoV outbreak in 2012 [25]. In the COVID -19 pandemic, which started in China on December 2019, patients with sudden olfactory loss have been reported in countries such as China, Italy, Spain, Singapor and USA where pandemics are frequently seen. The rate of 5% was seen in a study in China and The American Academy of Otolaryngology-Head and Neck Surgery" in the study were reported as striking as the rst symptom in 73% Health 26.6% patients. [25][26][27][28] It is obvious that the disease causes much more olfactory disorders than other viral infections. In this, the direct damage to the olfactory epithelium of the virus or the neural retrograde pathway can have an effect on the olfactory region containing the odor receptors in the region. From a study, the nucleic acid of the virus has been detected in both the brain tissue and the cerebrospinal uid (CSF) [29,30]. In the spread of the virus in the body, into many tissues and nervous system of the body, the angiotensin converting enzyme 2 (ACE 2) and transmembrane serine protease 2 (TMPRSS2) proteins plays a role. However, in a new study, it was shown that ACE 2 and TMPRSS2 proteins were not found in the olfactory region in humans. Therefore, there is no role of these proteins in olfactory damage. On the contrary, damage to basal cells was detected. Therefore, the possible mechanism in olfactory disorder appears to be the damage of these cells. Therefore, the biochemical reaction in the formation of odor cannot take place [31][32][33][34]. In our study, when we evaluated Paranasal CTs taken from patients with odor loss, pathological ndings were observed in 2 patients' CT and were statistically signi cant (p <0.05).
Since odor disorder is a subjective concept, objective tests were needed in its diagnosis. There are two types of odor tests, psychophysical and electrophysiological tests. In the diagnosis of smell disorder, smell tests are used in practice.
These tests are carried out in two stages as odor threshold tests and odor odorant tests. In the threshold determination test, a scent bottle containing fragrances such as phenyl ethyl alcohol (PEA) or butyl alcohol (butanol) 4% and another bottle containing only water are presented to the patient. Odor detection tests are quantitative tests. Patients are asked to identify fragrances above the threshold. The most used of these are The University of Pennsylvania Smell Identi cation Test (UPSIT), Sni n' Sticks, Connecticut odor detection test-CCCRC, OSIT-J (Odor Stick Identi cation Test for Japanese, Daiichi Yakuhin, Co., Tokyo, Japan), B-SIT (the Brief Smell Identi cation Test) and the Crosscultural smell identi cation test (CC-SIT). The most widely used of these tests is the UPSIT test [10,15,35]. These tests can also be used in the diagnosis of olfactory disorders in COVID-19 . In a study, the UPSIT test was used in patients with COVID-19 smell disorder, and according to this test, most patients were found to have a loss of smell ranging from mild microsmia to anosmia. The test scores were not related to age, degree of disease, nor additional diseases [10]. In another study, a smell and taste change survey was conducted on social media, and a signi cant result for COVID-19 was found in those with this symptom [35]. In our study, the CCCRC test, which is more suitable for Turkish society, was used. A correlation was found between both clinical and PNS CT nding on the positivity of the CCCRC score. As patients' illness worsened, an increase in olfactory loss was observed (p <0.05). A statistically signi cant relationship was found between the CCCRC score and PNS CT ndings (p = 0.012).
In literature, single Paranasal imaging in patients with COVID-19 with olfactor impairment has been reported in a single patient in the USA. A thickening and drainage disorder in the olfactory region was reported in this patient in coronal sections. It is estimated that the olfactory bulb is retained through the cribriform plate depending on the involvement of the respiratory epithelium in patients with COVID-19 [11]. In a study related to post-infectious olfactor loss, the infectionrelated olfactory bulb volume was shown to decrease38. In our study, in the majority of patients with COVID-19 with olfactory disorders, obstructive thickening and loss of aeration were observed, especially in the Paranasal Sinus CT, which disrupted the olfactory region drainage. Thus, the cause of olfactory loss in these patients was con rmed by imaging. In addition, a signi cant relationship was found between the degree of odor disorder and the positivity of CT ndings.
Since there were 40 patients in the study, the study needs to be done in larger series and in multiple centers.

Conclusion
COVID-19 is a disease that causes pneumonia illness and causes serious mortality, such as respiratory failure, from common cold symptoms such as fever, dyspnea and cough. Sudden olfactory loss can be diagnostic in this disease.
Paranasal Sinus CT imaging, which is used to illuminate the pathophysiology of odor disorder in these patients, can be a new diagnostic tool for the clinician as a highly diagnostic method. For clear information on this subject, wider participation and multicenter research is needed.

Declarations Funding
The authors declared that this study has received no nancial support.

Con ict of interest statement
No con ict of interest was declared by the authors.