key: cord-0847697-qngjpoea authors: Keshavarz, Pedram; Haseli, Sara; Yazdanpanah, Fereshteh; Bagheri, Fateme; Raygani, Negar; Karimi-Galougahi, Mahboobeh title: A Systematic Review of Imaging Studies in Olfactory Dysfunction Secondary to COVID-19 date: 2021-08-30 journal: Acad Radiol DOI: 10.1016/j.acra.2021.08.010 sha: 2d36d5666285ec5064633fdebde04f9f7075dcd7 doc_id: 847697 cord_uid: qngjpoea RATIONALE AND OBJECTIVES: Hyposmia/anosmia is common among patients with COVID-19. Various imaging modalities have been used to assess olfactory dysfunction in COVID-19. In this systematic review, we sought to categorize and summarize the imaging data in COVID-19-induced anosmia. MATERIAL AND METHODS: Eligible articles were included after a comprehensive review using online databases including Google scholar, Scopus, PubMed, Web of science and Elsevier. Duplicate results, conference abstracts, reviews, and studies in languages other than English were excluded. RESULTS: In total, 305 patients undergoing MRI/functional MRI (177), CT of paranasal sinuses (129), and PET/CT or PET/MRI scans (14) were included. Out of a total of 218 findings reported on MRI, 80 were reported on early (≤ 1 month) and 85 on late (> 1 month) imaging in relation to the onset of anosmia. Overall, OB morphology and T2-weighted or FLAIR signal intensity were normal in 68/218 (31.2%), while partial or complete opacification of OC was observed in 60/218 (27.5%). T2 hyperintensity in OB was detected in 11/80 (13.75%) and 18/85 (21.17%) on early and late imaging, respectively. Moreover, OB atrophy was reported in 1/80 (1.25%) on early and in 9/85 (10.58%) on late imaging. Last, among a total of 129 CT scans included, paranasal sinuses were evalualted in 88 (68.21%), which were reported as normal in most cases (77/88, [87.5%]). CONCLUSION: In this systematic review, normal morphology and T2/FLAIR signal intensity in OB and OC obstruction were the most common findings in COVID-19-induced anosmia, while paranasal sinuses were normal in most cases. OC obstruction is the likely mechanism for olfactory dysfunction in COVID-19. Abnormalities in OB signal intensity and OB atrophy suggest that central mechanisms may also play a role in late stage in COVID-19-induced anosmia. Hyposmia/anosmia is relatively common among patients with coronavirus disease-2019 , both as an isolated symptom or concomitant with other systemic or respiratory symptoms (1, 2) . Olfactory dysfunction may present as an alteration in the intensity of perceived odor or in odor quality. While hyposmia/anosmia is a frequent symptom associated with COVID-19, it is usually transient, and spontaneously resolves within a few weeks (1, 2) . The first step in the olfactory perception is activation by odorants of sensory neurons that are located in the olfactory epithelium. This sensory epithelium is located in the olfactory clefts (OC), which are two narrow passages at the upper part of the nasal cavities (3) . The olfactory receptor neurons originating from the OC cross the cribriform plate to reach the olfactory bulb (OB) (3) . The exact pathophysiology of COVID-19-induced anosmia is not fully understood (1) (2) (3) . Various imaging modalities have been used to assess the olfactory dysfunction in COVID-19, mainly by imaging OB and OC, such as with magnetic resonance imaging (MRI) , computed tomography (CT) of paranasal sinuses (29) (30) (31) (32) , or with functional modalities such as positron emission tomography (PET) of brain (15, 22, 33) . These studies have reported a variety of findings, which are at times divergent. In the present report, we sought to systematically review the imaging studies that have investigated COVID-19-induced olfactory dysfunction and to categorize the data from these studies, including based on the timing of the imaging in relation to the onset of olfactory dysfunction in the course of COVID-19. We identified studies that have used imaging modalities such as brain CT, MRI, or PET in patients with COVID-19, confirmed with reverse transcription-polymerase chain reaction, and a clinical presentation with anosmia or olfactory dysfunction. Relevant studies were searched using MEDLINE (PubMed), Web of Science, Embase (Elsevier), Scopus, and Google Scholar databases (search was undertaken on May 26, 2021). The following medical subject headings and keywords were used: 'Coronavirus', 'COVID-19', 'SARS-CoV-19', '2019-CoV-19', 'radiolog*', 'radiograph*', 'Anosmia', 'Hyposmia', 'Olfactory Bulb', 'Tomography, X-ray Computed', 'CT', 'MRI', and 'PET'. The full list of the keywords that were used for search in the PubMed is provided in the Supplementary Appendix. Duplicates, conference abstracts, reviews, and studies in languages other than English were excluded. The following data from the eligible studies were collected by two independent investigators and were subsequently cross-checked. Disagreements between the investigators were resolved by a third investigator after discussions and reaching consensus. The following information were collected: first author's name, country where the study was conducted, study type, clinical characteristics of participants, clinical presentation of the olfactory dysfunction, findings on brain imaging, and clinical outcomes. After excluding duplicate search results, titles and abstracts of 1490 studies were assessed, and 50 eligible studies were assessed by full-text review. Thirty-one studies, including 7 cohort studies, 6 case series/reports, 2 case-control studies, 1 cross-sectional study, and 15 commentaries, letters to the editor, and short reports were included in the final analysis. The PRISMA flow diagram for study selection is shown in Figure 1 . Types of the studies and clinical characteristics of the patients are summarized in Table 1 Findings using various imaging modalities are summarized in Table 2 . In total, 305 patients undergoing a total of 320 imaging studies were included. Imaging modalities used to assess changes in the olfactory tract in COVID-19-induced anosmia included MRI/functional MRI (n=177), CT of the paranasal sinuses and brain (n=129), and PET/CT or PET/MRI scans (n=14). A summary of the findings on MRI in anosmia of COVID-19 is provided in Tables 2 and 3 . In total, 218 MRI findings were reported. Abnormalities on MRI can be divided into the following categories: changes in the morphology and signal intensity in OB, OC, or cortex, and abnormalities on functional MRI. With regard to the timing of imaging in relation to the onset of olfactory symptoms, the studies can be categorized to early imaging (≤1 month since the onset of olfactory dysfunction) and late imaging (> 1 month). Of the total of 218 MRI findings, 80 were detected in the early and 85 in the late phase, while the timing was not specified in the remainder (n=53) ( Table 4 ). Of the total of 218 findings on MRI, both morphology and signal intensity of OB were normal in 68/218 (31.2%) (Figure 2 ), which were the most common overall MRI finding (9) (10) (11) (12) (13) (14) (15) (22) (23) (24) (25) (26) (27) . OB morphology and signal was normal in 28/80 (35%) in the early and in 26/85 (30.6%) on late imaging, similarly constituting the most common findings on both early and late imaging. Other findings on MRI included OB enhancement following gadolinium injection in 18/218 (8.25%) (5, 7, 8, 28) and T2 hyperintensity in the OB/tract in 29/218 (13.30%) (4,10,12,17-21,26) ( Figure 3A ), while hypointense focus, consistent with microhemorrhage/methemoglobin deposition ( Figure 3B ), was identified in 17/218 (7.8%) (5, 10, 28) . T2 hyperintensity in OB was detected in 11/80 (13.75%) and 18/85 (21.17%) during the early and late imaging, respectively. Moreover, a reduction in the OB volume was detected in 10/218 (4.6%) (4, 6, 16, 19) . OB atrophy was reported in one case (1/80 [1.25%]) on early imaging and in 9 on late imaging (9/85 [10.58%]) ( Table 4) . Partial or complete obliteration of OC was the second most common finding on MRI, and was observed in 60/218 (27.5%) (8, 10, 14, 16, 18, 22, 23) . OC was normal in 5/218 (2.3%) (8, 22) . OC obliteration was detected in 28/80 (35%) on early imaging and in 22/85 (25.9%) on late imaging. No enhancement in the mucosal lining of OC was detected on early imaging, while this change was only detected in one case (1/85 [1.17%]) in the late phase (16) . Studies assessing central olfactory centers were scarce, with signal alteration in the olfactory cortices reported in 10/218 [4.6%]) (4, 10, 11, 12, 26) . Hyperintensity in the orbitofrontal and entorhinal cortices (12) , in the right rectus gyrus (4) and in the left caudate and parahippocampus (11) have been reported. T2 hyperintensity in cortex both on early and late imaging were reported in 6.3% and 5.9% cases, respectively. Bilateral inferior frontal lobe hypodensity involving the straight gyrus was reported in one patient (20) . Only one report on the use of functional MRI in anosmia secondary to COVID-19 has been published, where strong blood oxygen level-dependent signal within the piriform and right uncal cortices was detected (24) . Findings on CT of paranasal sinuses can be categorized into two main groups: changes in the paranasal sinuses (opacification of the sinuses) or in OC (partial or complete opacification). Among a total of 129 CT scans included in the present review, paranasal sinuses were evaluated and reported in 88 of 129 (68.21%) (10, 29, 30) . In the majority of these scans (77/88 [87.5%]), paranasal sinuses were normal (10, 30) , with opacification of the paranasal sinuses only observed in 11/88 [12.5%]) (10, 30) . Moreover, out of the total 129 CT scans, OC was normal in 100/129 (77.5%), while in 29/129 (22.5%) complete or partial opacification of OC was reported (10, (29) (30) (31) (32) . Few studies using PET/CT for assessment of olfactory dysfunction in COVID-19 have been performed. Reduction in fluorodeoxyglucose uptake in the left insula, left inferior frontal gyrus, left hippocampus and left amigdala compared with the contralateral side were identified in one patient (15) . Additionally, hypometabolic activity of the left orbitofrontal cortex was detected in a patient who underwent PET/CT for evaluation of COVID-19-induced olfactory dysfunction (33) . Alterantions in glucose metabolism in the olfactory tract and higher brain networks have also been reported in 12 pateints with anosmia secondary to COVID-19 (22) . There was heterogeneity in what has been defined as "control" in the three case-control studies included in the current review. Some studies included normosmic individuals without COVID-19 as the control group (8), whereas others included normosmic patients with COVID-19 as controls (14) . In one study, patients with olfactory dysfunction but without COVID-19 were included as controls for imaging evaluation of neurological manifestations of COVID-19 (12) . Elezier et al (8) controls with olfactory dysfunction without COVID-19 (p=0.028). OB volume was similar between the two groups (p value not reported). In this systematic review, opacification of OC together with normal OB morphology and signal intensity were the most common imaging findings in patients with olfactory dysfunction secondary to COVID-19, while paranasal sinuses were normal in most cases. OC opacification was detected in most anosmic patients with COVID-19 compared to the normal OC in normosmic controls, with resolution of OC opacification correlating with improved olfaction. Taken together, these data suggest obstruction of OC, which likely results in mixed conductive and peripheral sensory-neural anosmia, as the most likely mechanism for olfactory dysfunction in COVID-19. An increase in the frequency of abnormalities in signal intensity and morphology of OB on late versus early imaging suggests that central mechanisms for anosmia may also play a role, especially in the late phase -a postulate that is supported by functional abnormalities detected on PET and functional MRI studies. The exact pathogenesis of the olfactory dysfunction in COVID-19 is not fully elucidated. Possible mechanisms include neurotropism and invasion of the OB by the virus, inflammatory changes, and impairment of the olfactory epithelium via angiotensin converting enzyme 2 receptors that are expressed by the non-neural olfactory supporting cells (35) . Persistent olfactory dysfunction after sinonasal symptoms resolve suggests possible injury to the olfactory stem cells and impairment of the supporting cells (36) . Post-viral ansomia due to upper respiratory tract infections can occur in up to 40% of cases (37) , which is often secondary to diffuse sinonasal mucosal thickening. Nevertheles, sudden, transient olfactory dysfunction is common in COVID-19 (1, 4, 8, 12, 23, 24, 26, 29, 40) . Lack of concomitant sinonasal symptoms (i.e., nasal obstruction and mucosal congestion) in majority of the cases (38) indicates that diffuse sinonasal mucosal thickening may not be a contributing mechanism in most cases of COVID-19-induced anosmia. Consistent with this, CT of paranasal sinuses was normal in most patients with COVID-19-induced anosmia, with no mucosal thickening or obstruction detected (10, (29) (30) (31) (32) . Similarly, On MRI, OB and OC were normal in most cases with COVID-19-induced anosmia. Nevertheless, infection of the olfactory epithelial support cells (Bowman and sustentacular cells) plays a key role in ansomia of COVID-19 (39) . SARS-CoV-2 utilizes angiotensin converting enzyme 2 and transmembrane serine protease 2 to enter the respiratory epithelial cells, which act as a viral reservoir for COVID-19 (35) . The localized respiratory epithelial involvement and inflammation in the OC may appear as hyperintense mucosal thickening and obstruction, which indeed was the second most common abnormality on MRI in the present review. It is postulated that patients with obstruction in the OC may develop more severe olfactory dysfunction compared with patients with normal OC (29) . Inflammation of the OB with blood brain barrier breakdown can result in edema in OB, which appears as hypersignal intensity on T2-weighted images. Prolonged inflammation of the olfactory tract can be detected as thickening and clumped appearance of the olfactory nerve filia on MRI (5, 7, 8) . Additionally, microbleeding secondary to microvascular injury and microthrombosis by SARS-Cov-2 may appear as foci of hyposignal intensity within the OB (5,10,28). Overall, timing of the imaging studies during the course of COVID-19-induced olfactory dysfunction appears to be important in detection of abnormalities on imaging. While imaging of OB in most cases is normal within the early days after the onset of olfactory dysucntion, a decrease in the OB volume, detected as thinning or loss of normal oval shape and asymmetry compared with the contralateral OB (19) , is more frequently detected at later stages. Additionally, it is plausible that persistently low activity in the afferent olfactory neurons may result in further reduction in the OB volume. Indeed, it takes several weeks for structural changes in the OB to appear in the course of COVID-19-induced anosmia. This observation might suggest impairment in the olfactory stem cells in addition to damage to the epithelial support cells (36) , which in combination may lead to olfactory dysfunction and atrophy of OB. Finally, decreased metabolic activity of the orbitofrontal cortex on PET-CT (33) may be a result of direct viral neurotropism through the olfactory pathways (38) as suggested by cortical hyperintensity on FLAIR sequence on MRI (11, 20) . Additionally, deafferentation process and functional reorganization secondary to absent sensory stimulation of the olfactory pathways (22, 40) may also contribute to the observed hypometabolism on PET-CT. Ethics approval: not applicable. The authors of this manuscript declare no conflicts relevant to the content of this manuscript. OM enhancement Hypersignal cortex --1 -1 ---1 -4 -1 ---------1 -----1 --------1 ---1 19 --3 --------23 -17 4 --17 -5 5 -------1 12 - 4 - - - - - 1 1 - - - - - - - - 14 - - - - - 10 - - 1 - - - - - - - - - - - - 7 - 4 1 - - - 1 - - - - - - - - 1 - - - 1 - - - - 1 - 1 - - - - - - 1 - - - - - - - - 1 - - - - - - 3 - - - - 4 8 - - 1 - - - - - 1 - - 1 - - - - - - - - 1 - - - - - - - - 2 - 2 - - - - - 2 1 - - - - - - - - - 12 - 12 - - - - - Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The lancet Olfactory bulb signal abnormality in patients with COVID-19 who present with neurologic symptoms Inflammatory obstruction of the olfactory clefts and olfactory loss in humans: a new syndrome?. Chemical senses Magnetic resonance imaging alteration of the brain in a patient with coronavirus disease 2019 (COVID-19) and anosmia Anosmia in COVID-19 associated with injury to the olfactory bulbs evident on MRI COVID-19-induced anosmia associated with olfactory bulb atrophy Magnetic resonance imaging features of COVID-19-related cranial nerve lesions Loss of smell in patients with COVID-19: MRI data reveal a transient edema of the olfactory clefts Olfactory bulb magnetic resonance imaging in SARS-CoV-2-induced anosmia in pediatric cases Olfactory bulb MRI and paranasal sinus CT findings in persistent COVID-19 anosmia Smell and taste disorders in COVID-19 patients: objective testing and magnetic resonance imaging in five cases Olfactory bulb signal abnormality in patients with COVID-19 who present with neurologic symptoms Sudden anosmia and ageusia in a child: A COVID-19 case report Magnetic resonance imaging of COVID-19 anosmic patients reveals abnormalities of the olfactory bulb: preliminary prospective study Multimodality Imaging With PET/CT and MRI Reveals Hypometabolism in Tertiary Olfactory Cortex in Parosmia of COVID-19. Academic Radiology Olfactory bulb and mucosa abnormalities in persistent COVID-19-induced anosmia: a magnetic resonance imaging study Anosmia and olfactory tract neuropathy in a case of COVID-19 Bilateral transient olfactory bulb edema during COVID-19-related anosmia Olfactory bulb atrophy in a case of COVID-19 with hyposmia. Academic Radiology Anosmia caused by ischaemic olfactory infarction: false alert for COVID-19 infection Mild to Severe Neurological Manifestations of COVID-19: Cases Reports Structural and metabolic brain abnormalities in COVID-19 patients with sudden loss of smell. European journal of nuclear medicine and molecular imaging Sudden and complete olfactory loss of function as a possible symptom of COVID-19. JAMA otolaryngology-head & neck surgery Absent Blood Oxygen Level-Dependent Functional Magnetic Resonance Imaging Activation of the Orbitofrontal Cortex in a Patient With Persistent Cacosmia and Cacogeusia After COVID-19 Infection Olfactory bulb magnetic resonance imaging in SARS-CoV-2-induced anosmia: the first report An MRI of the Olfactory Tract in a Case of Post-COVID-19 Persistent Anosmia Clinical and Radiological Profiles of COVID-19 Patients with Neurological Symptomatology: A Comparative Study Clinical and Radiological Evaluations of COVID-19 Patients with Anosmia: Preliminary Report. The Laryngoscope Paranasal sinuses computed tomography findings in anosmia of COVID-19 Covid 19 pandemic: Paranasal diagnostic imaging in patients with olfactory loss Frequency and outcome of olfactory impairment and sinonasal involvement in hospitalized patients with COVID-19 PET/CT scan reveals hypoactive orbitofrontal cortex in anosmia of COVID-19 Anosmia and ageusia as the only indicators of Coronavirus Disease 2019 (COVID-19) Anosmia in COVID-19: a bumpy road to establishing a cellular mechanism COVID-19 and the chemical senses: supporting players take center stage Covid-19: Anosmia and Ageusia Might be Initial or Unique Symptoms Anosmia in COVID-19: Underlying Mechanisms and Assessment of an Olfactory Route to Brain Infection COVID-19 treatments and pathogenesis including anosmia in K18-hACE2 mice Anosmia in COVID-19: mechanisms and significance. Chemical senses