key: cord-0983045-7mhpzztb
authors: Kandemirli, Sedat Giray; Altundag, Aytug; Yildirim, Duzgun; Tekcan Sanli, Deniz Esin; Saatci, Ozlem
title: Olfactory bulb MRI and paranasal sinus CT findings in persistent COVID-19 anosmia
date: 2020-10-19
journal: Acad Radiol
DOI: 10.1016/j.acra.2020.10.006
sha: e63a81850a49b443cd11bfde807f4256c331321e
doc_id: 983045
cord_uid: 7mhpzztb

BACKGROUND AND PURPOSE: There is limited literature consisting of case reports or series on olfactory bulb imaging in COVID-19 olfactory dysfunction. An imaging study with objective clinical correlation is needed in COVID-19 anosmia in order to better understand underlying pathogenesis. MATERIAL AND METHODS: We evaluated 23 patients with persistent COVID-19 olfactory dysfunction. Patients included in this study had a minimum 1-month duration between onset of olfactory dysfunction and evaluation. Olfactory functions were evaluated with Sniffin’ Sticks Test. Paranasal sinus CTs and MRI dedicated to olfactory nerves were acquired. On MRI, quantitative measurements of olfactory bulb volumes and olfactory sulcus depth and qualitative assessment of olfactory bulb morphology, signal intensity and olfactory nerve filia architecture were performed. RESULTS: All patients were anosmic at the time of imaging based on olfactory test results. On CT, Olfactory cleft opacification was seen in 73.9% of cases with a mid and posterior segment dominance. 43.5% of cases had below normal olfactory bulb volumes and 60.9% of cases had shallow olfactory sulci. 54.2% of cases had changes in normal inverted J shape of the bulb. 91.3% of cases had abnormality in olfactory bulb signal intensity in the forms of diffusely increased signal intensity, scattered hyperintense foci or microhemorrhages. Evident clumping of olfactory filia was seen in 34.8% of cases and thinning with scarcity of filia in 17.4%. Primary olfactory cortical signal abnormality was seen in 21.7% of cases. CONCLUSION: Our findings indicate olfactory cleft and olfactory bulb abnormalities are seen in COVID-19 anosmia. There was a relatively high percentage of olfactory bulb degeneration. Further longitudinal imaging studies could shed light on the mechanism of olfactory neuronal pathway injury in COVID-19 anosmia.

COVID-19 related olfactory dysfunction as an isolated symptom or in conjunction with other respiratory symptoms has been increasingly recognized. 1 A recent metaanalysis identified 45% of COVID-19 patients had olfactory dysfunction. 2 The olfactory dysfunction is sudden onset in majority of cases and is usually a transient entity with a median time to recovery ranging between 1-3 weeks. 3 No significant association with sinonasal symptoms had been identified, suggesting that pathogenesis of COVID-19 anosmia might differ from obstructive olfactory dysfunction seen in other viral upper respiratory tract infections. 1, 4 The pathogenesis of COVID-19 anosmia has not been fully defined, however plausible mechanisms are olfactory cleft inflammation/obstruction and/or olfactory bulb damage. 1, 4 Magnetic resonance imaging (MRI) dedicated to olfactory nerves is a useful anatomical imaging modality for evaluation of olfactory dysfunction related to post-viral infection, trauma, and neurodegenerative processes. 5, 6 A dedicated MRI study allows assessment of olfactory bulb volume, morphology and signal intensity, status of olfactory nerve filia and signal intensity of primary olfactory cortex, which is helpful to differentiate between different etiologies and predict prognosis of olfactory function recovery. 5 There is limited literature consisting of case reports on olfactory bulb imaging in COVID-19 olfactory dysfunction. Reported findings included enlarged olfactory bulbs with internal signal abnormality, olfactory bulb microhemorrhage/enhancement and essentially normal olfactory bulb signal intensity. [7] [8] [9] [10] However, in majority of the cases, images were not part of an MRI dedicated to olfactory nerves which limit reliability of the findings. Additionally, temporal course and objective assessment of the olfactory dysfunction was lacking in majority of the reports.

A systematic imaging study with objective clinical correlation is needed in COVID-19 anosmia in order to better understand the underlying pathogenesis. In this study, we evaluated a group of patients presenting with persistent COVID-19 related anosmia/hyposmia despite resolution of other COVID-19 related symptoms. Patients were evaluated with objective olfactory tests, paranasal sinus CT, and MRI dedicated to olfactory nerves. We assessed the olfactory cleft opacification; olfactory bulb volume, morphology, and signal intensity; olfactory nerve filia architecture and primary olfactory cortex signal abnormalities.

This was a prospective study including consecutive patients presenting to a dedicated Smell and Taste Center Clinic with persistent COVID-19 related olfactory dysfunction between May and June 2020. Patients had persistent stable or mildly improved olfactory dysfunction after resolution/recovery of other COVID-19 related symptoms. COVID-19 infection at the time of initial symptoms was confirmed by polymerase chain reaction (PCR) with a swab test. Patients included in this study had a minimum 1-month duration between onset of olfactory dysfunction and evaluation at our center after symptom onset with a range of 1-4 months.

Pediatric and pregnant patients, patients with history of head trauma, preexisting smell and taste alterations, allergic rhinitis and chronic rhinosinusitis were excluded.

Institutional review board approval and informed consent were obtained.

Medical history with a specific focus on olfactory dysfunction for its onset and course, sinonasal symptoms (nasal congestion, rhinorrhea) were obtained. Patients underwent detailed examination of the nose and olfactory region. Olfactory functions were evaluated with Sniffin' Sticks Test battery (Burghart Messtechnik, Germany), which has 3 components to assess olfactory threshold (T), discrimination (D) and identification (I). Each component has a scale of 1-16, and TDI is a composite score representing the sum of these 3 scores. Normosmia is defined for scores ≥ 30.5, hyposmia for scores between 16.5 and 30.5, and anosmia for scores <16.5. 11

High-resolution paranasal sinus CTs (128x2-slice dual-source CT scanner, Siemens, Flash Definition, Erlangen, Germany) were acquired to assess cribriform plates for evidence of any prior trauma and nasal passage for obstructive causes. Additional reformatted images of olfactory cleft were created with small field of view with 0.4 mm section-thickness and 0.1 mm increment. Olfactory cleft aeration pattern was grouped as normal, partial or total opacification.

MRI dedicated to olfactory nerves was acquired with a 3 Tesla MRI unit (3 Tesla Magnetom MRI unit, Siemens, Erlangen, Germany) using a 32-channel head coil. Ultrahigh resolution T2-space sagittal images (TR: 1000 ms, TE: 136 ms, flip angle: 110°, slice-thickness: 0.6 mm, slice oversampling: 0%, FOV: 200×200 mm, matrix: 320×320, phase oversampling: 30%, band width: 150 Hz/pixel, voxel size: 0.3× 0.3×1 mm 3 , time of acquisition: 6.08 min, echo-train duration: 440 ms) and coronal T2 images covering anterior pole of the olfactory bulb to the primary olfactory region (TR: 6550 ms, TE: 99 ms, flip angle: 150°, slice-thickness: 1 mm, distance factor: 0, FOV: 100×100 mm 2 , matrix: 269×384, phase oversampling: 56%, bandwidth: 289 Hz/pixel, voxel size: .6× .6× .6 mm 3 , time of acquisition: 8.19 min, turbo factor: 17) were acquired. Additional conventional sequences for whole brain were obtained.

Olfactory bulb volume was calculated based on sum of sequential region of interest on consecutive slices using MPR with Syngo.Via Software (VB10B, Siemens). For olfactory volume, cut-off value of 58 mm 3 were used as minimal-normal OB volume as described by Buschhüter et al. 12 Olfactory sulcus depth was measured on coronal T2 images, by drawing a line tangent to the inferior borders of gyrus rectus and medial orbital gyrus and measuring the depth to the deepest point of the olfactory sulcus. 5 Olfactory bulb morphology was evaluated on high-resolution coronal T2 sections.

Oval or inverted-J shape of olfactory bulbs was considered as normal ( Fig. 1a) . Multiple areas (≥2) of olfactory bulb contour lobulation, rectangular shape or atrophic appearance were considered as abnormal (Fig. 1b) . Olfactory bulb signal intensity was assessed with contralateral gyrus rectus taken as the reference point. Hyperintense signal abnormality was assessed for its morphology either as diffuse or punctate focus (Fig. 2, 3) . A single focus of abnormal signal was not considered as abnormal as bulb region is prone to artifacts. 13, 14 Images were also reviewed for presence of punctate hypointense regions in the olfactory bulb suggestive of microhemorrhages (Fig. 4) . 7 Primary olfactory cortex and visualized olfactory tracts were evaluated for presence of abnormal signal intensity on T2 and FLAIR images.

Olfactory nerve filiae were assessed on sagittal T2-space images through the medial and lateral aspects of the bulb. Fine architecture of the filia with uniform distribution at regular intervals was considered as normal ( Table 1 . Interval between onset of olfactory dysfunction and evaluation at our center ranged between 1-4 months.

At the time of evaluation at Smell and Taste center, all 23 patients were anosmic based on TDI scores. Median threshold score was 1 (interquartile range 1-2.25), median discrimination score was 2 (interquartile range 0-3), median identification score was 2 (interquartile range 0-4), and median TDI score was 4 (interquartile range 1-8.5).

Information on TDI scores is presented in Table 1 .

On paranasal sinus CT, none of the cases had opacification of ethmoid air cells or nasal cavities. Olfactory clefts showed normal aeration in 6 cases, partial opacification in 16 cases and were totally opacified in a single case. (Table 2 ). (Table 2) .

Based on the cut-off value of 58 mm 3 described by Buschhüter et al.; olfactory bulb volumes were below normal in 10 cases. 12 There was no significant difference between the TDI scores based on cut-off value of 58 mm 3 for bulb volume (p=0.180).

Median olfactory sulcus depths were 6.8 mm (interquartile range 5.4-8.1 mm) on the right and 6.3 mm (interquartile range 4.6-7.6 mm) on the left. Based on a cut-off value of 7.5 ml (48), 14 cases had shallow olfactory sulci. There was a significant negative correlation between total TDI scores and right olfactory sulcus depth (r=-0.567, p=0.006). No significant correlation between TDI scores and left olfactory sulcus depths were found. There was no significant correlation between sulcus depth and olfactory bulb volumes (p>0.05).

Morphological classification of the olfactory bulbs and olfactory nerves for groups are presented in Table 2 . Type J morphology (usual anatomical shape) was seen in 8 cases (34.8%). Mild contour irregularity with general preservation of J shape was seen in 2 cases. There was deformed J shape in 5 cases with rectangular morphology in 8 cases.

Olfactory bulb signal intensity was normal in 2 cases. Diffusely increased signal intensity of the olfactory bulbs was seen in 9 cases, and in 7 of these cases this increased signal extended to involve the olfactory tract (Fig. 7) .

In 16 cases, there were multiple ( ≥2) hyperintense foci in olfactory bulbs. In 5 of these cases, there was a hypointense halo around the hyperintense foci. Scattered hypointense foci suggestive of microhemorrhages were noted in an additional 4 cases (Table 2) . No significant differences in TDI scores, olfactory bulb volumes and olfactory sulcus depths were found when MRI findings were grouped based on bulb signal or shape pattern (p>0.05).

Olfactory nerve filia architecture was grossly normal in 13 cases. There was evident clumping in 8 cases. There was thinning and scarcity in 2 cases (Table 2) . No significant differences in TDI scores, olfactory bulb volumes and olfactory sulcus depths were found when MRI findings were grouped based on filia architecture pattern (p>0.05).

In 5 cases, there was increased cortical signal intensity in the primary olfactory cortex (Fig. 7b ).

In this study, we evaluated 23 patients with persistent COVID-19 olfactory dysfunction. All patients were anosmic at the time of imaging based on TDI scores. We noted a high percentage of olfactory cleft opacification (73.9% of cases). There was reduction of olfactory bulb volumes, change in bulb shape and signal abnormalities.

Some of the cases showed olfactory nerve filia clumping or scarcity. Additionally, 21.7% of cases had primary olfactory cortical signal abnormality.

Pathogenesis of olfactory dysfunction in COVID-19 disease is still incompletely understood, but there are some distinct features that separate this entity from other postviral olfactory dysfunction. Post-viral anosmia in the setting of upper respiratory tract infection can account up to 40% of the cases, and is usually related to mucosal congestion and nasal obstruction. 15 This can affect the airflow and impair the travel of odorants, despite an intact olfactory epithelium, and result in a conductive olfactory loss. 4 On the other hand, there is no significant association with sinonasal symptoms in COVID-19 anosmia, as also shown in our cohort. 16 This suggests that mechanisms other than sinonasal obstruction may play a role. Plausible mechanisms include direct damage to olfactory epithelium and olfactory bulb by the virus, and damage to olfactory epithelium secondary to inflammatory changes. 4 Angiotensin converting enzyme 2 (ACE2) receptors, which are the target molecules for SARS-CoV-2, are expressed by non-neuronal supporting cells of the olfactory epithelium, but not directly by the olfactory neurons, which might be the putative target site for the virus. 4 Another potential mechanism is direct damage to olfactory nerves and retrograde invasion of olfactory tracts. 4 Neurotropic potential of coronaviruses has been demonstrated, however absence of ACE2 receptors in olfactory neurons does not support this hypothesis. 17 The best example of retrograde olfactory neuroinvasion with anosmia is described in herpes simplex virus. Histological analysis of olfactory epithelium in herpes simplex encephalitis patients has shown diffuse inflammation and ragged appearance of the cells with hemorrhage along perineural sheaths. 20 However, pathological evidence for olfactory nerve invasion in COVID-19 is not available currently.

Olfactory clefts are lined by olfactory epithelium which also contain the olfactory receptor neurons. 21 Axons of these neurons cross the cribriform plate as olfactory filia and synapse in the olfactory bulbs. Various sinonasal pathologies can cause mucosal inflammation and obstruction in lower parts of the nasal cavities, resulting in blockage of odors to reach the olfactory cleft. 16 In our series, none of the cases had mucosal synechia in the nasal cavities.

As described above, there can be secondary inflammatory changes in olfactory clefts due to viral invasion which could result in mucosa edema with subsequent narrowing of the olfactory cleft. In our study, we identified partial and total olfactory cleft opacification in 69.6% and 4.3% of cases, respectively, supporting this hypothesis..

Review of the available paranasal sinus CT literature on COVID-19 anosmia shows that majority of patients did not have significant pathological changes with partial opacification of the olfactory cleft noted in some cases. 22

MRI of olfactory nerves provides useful anatomical details for evaluation of olfactory bulb, olfactory nerve filia and primary olfactory cortex in patients with anosmia. Majority of the literature on post-viral anosmia evaluated the olfactory bulb volume, which showed correlation with olfactory function. 23, 24 Olfactory bulb can also be assessed for its overall morphology, where loss of oval or inverted J shape is suggestive of abnormal structure. 23 Additionally, olfactory bulb can be assessed for presence of signal abnormalities in the form of hyperintensities indicating degeneration and hypointensities representing microhemorrhages. 7 Olfactory nerve filia can be assessed for clumping or scarcity, which suggests inflammatory changes and degenerative/traumatic loss, respectively.

Literature on olfactory bulb imaging in COVID-19 anosmia is limited, derived from conventional brain images instead of dedicated sequences. Additionally, there is variability in timing of imaging, and objective assessment of olfactory dysfunction. 7-10, 25, 26 Some of the initial case reports reported normal olfactory bulb volume with normal signal intensity. 8 whereas the available literature mainly reports acute findings in temporary anosmia. In this aspect, our findings may reflect changes in cases that had higher degree of olfactory epithelial damage and subsequent olfactory bulb involvement. There was no significant correlation between TDI scores and imaging findings except for olfactory sulcus depths, however considering the fact that majority of cases were anosmic with low TDI score, we may not have detected small differences between groups.

There are some limitations to our study. Main limitation is the relatively low number of patients given the extensive work-up required for inclusion in the study like requirement for completing ENT examination, smell test, and olfactory MRI. There might be a selection bias, as patients with more severe respiratory/systemic COVID-19 infection may not have been included in the study, as these patients more likely have a protracted disease course and did not present to our Smell disorder outpatient clinic.

Additionally, we imaged patients after a median interval of 1-4 months from onset of olfactory dysfunction. So these imaging findings may reflect the subacute and chronic state of changes, rather than the acute changes. Assessment of olfactory bulb and olfactory filia for signal intensity and morphology requires extensive experience with high level of attention to details. Another point to consider is the correlation between neuromalacic hyperintense changes in the olfactory bulbs and olfactory loss is still vague, as neuromalacic signal changes could also be identified in patients with normosmia. 23

This is the first systemic olfactory nerve imaging study with objective olfactory tests available in persistent COVID-19 anosmia. Our findings support the observations of olfactory cleft inflammation in COVID-19 anosmia. Additionally, relatively high percentage of olfactory bulb degeneration suggests that direct/indirect injury to olfactory neuronal pathways also take place, specifically in cases with persistent post COVID-19 anosmia. Interval between anosmia onset and MRI*  Data is presented as median and interquartile range. 

COVID-19 and the Chemical Senses: Supporting Players Take Center Stage

Olfactory and gustatory dysfunctions in COVID-19 patients: A systematic review and meta-analysis

Prevalence and Duration of Acute Loss of Smell or Taste in COVID-19 Patients

Anosmia in COVID-19: Mechanisms and Significance

Imaging the olfactory tract (cranial nerve #1)

Olfactory bulb volume in the clinical assessment of olfactory dysfunction

Anosmia in COVID-19 Associated with Injury to the Olfactory Bulbs Evident on MRI

Olfactory Bulb Magnetic Resonance Imaging in SARS-CoV-2-Induced Anosmia: The First Report

Bilateral transient olfactory bulbs edema during COVID-19-related anosmia

Magnetic Resonance Imaging Alteration of the Brain in a Patient With Coronavirus Disease 2019 (COVID-19) and Anosmia

Normative data for the "Sniffin' Sticks" including tests of odor identification, odor discrimination, and olfactory thresholds: an upgrade based on a group of more than 3,000 subjects

Correlation between olfactory bulb volume and olfactory function

Visualization of the olfactory nerve using constructive interference in steady state magnetic resonance imaging

The olfactory system

Olfactory disorders following upper respiratory tract infections

Frequency and outcome of olfactory impairment and sinonasal involvement in hospitalized patients with COVID-19

Non-neural expression of SARS-CoV-2

entry genes in the olfactory epithelium suggests mechanisms underlying anosmia in COVID-19 patients

Postviral Olfactory Dysfunction

Olfactory mucosal findings and clinical course in patients with olfactory disorders following upper respiratory viral infection

Olfactory mucosa in herpes simplex encephalitis

Inflammatory obstruction of the olfactory clefts and olfactory loss in humans: a new syndrome?

Paranasal sinuses computed tomography findings in anosmia of COVID-19

MR Imaging-Based Evaluations of

Olfactory Bulb Atrophy in Patients with Olfactory Dysfunction

Prognostic Value of Olfactory Bulb Volume Measurement for Recovery in Postinfectious and Posttraumatic Olfactory Loss

MRI Evaluation of the Olfactory Clefts in Patients with SARS-CoV-2 Infection Revealed an Unexpected Mechanism for Olfactory Function Loss

Anosmia and olfactory tract neuropathy in a case of COVID-19

Sudden and Complete Olfactory Loss of Function as a Possible Symptom of COVID-19

Flowchart 23 cases with COVID-19 anosmia Sudden