Abstract
Acute invasive fungal sinusitis (AIFS) is a rapidly progressive infection predominantly seen in immunocompromised patients. There is an increasing incidence of AIFS post COVID-19 infection. Yet, there is sparse literature regarding spectrum of cross-sectional imaging findings (CT and MRI) in these patients, which is prudent for appropriate timely surgical intervention. This study aims to highlight spectrum of imaging findings in patients with post-COVID-19 AIFS integrating the clinical details from presentation to follow-up. We retrospectively reviewed the demographics, clinical details and radiological imaging of 31 histopathological proven cases of post COVID-19 AIFS. MR and CT images of these patients were retrieved from the PACS and analysed. 90-day follow-up of these patients was obtained. Statistical analysis was performed using descriptive statistics. Cross-sectional imaging showed nasal cavity involvement in all patients, bilateral in (15, 48%) cases; predominant involvement of maxillary (31, 100%) followed by ethmoid sinuses (29, 93.5%) was seen. MR showed patchy/complete loss of normal mucosal enhancement in the turbinates and sinuses as predominant imaging finding. Maxillary sinus walls erosions were seen in 28 cases (90%). Rhino-orbital and rhino-orbito-cerebral mycosis was seen in 24 (77%) and 14 (45%) respectively. Optic nerve-sheath complex was involved in 15 (48%) cases. Cerebral involvement was seen in form of meningeal enhancement, cerebritis, ischemic changes, cavernous sinus and intracranial arterial thrombosis and aneurysms. Comprehensive knowledge of imaging features of AIFS and recognition of extent of their spread allows radiologists to play pivotal role in alerting the clinician for appropriate therapy to avoid protracted and fatal outcome.
Keywords: Acute invasive fungal mycoses, Mucormycosis, Imaging features, Post COVID-19
Introduction
Acute invasive fungal sinusitis (AIFS) is rapidly progressive and most invasive form of fungal sinusitis. It is traditionally described in two distinct subsets of immunocompromised patients, i.e., in diabetes and in severe neutropenia. In the former, especially those with diabetic ketoacidosis, the fungi implicated in upto 80% cases belong to the order of Zygomycetes which includes Rhizopus, Rhizomucor, Absidia and Mucor [1]. Latter group consists of patients with hematologic malignancies, AIDS, patients undergoing systemic chemotherapy, systemic steroid therapy, BMT, or immunosuppressive therapy for organ transplantation. Infection in upto 80% of patients in this group is caused by Aspergillus species. The morality rate reported is 54% [2].
AIFS begins in nose and PNS after inhalation of fungal spores which are ubiquitous. Infection spreads rapidly to involve the extra-sinus structures including orbit, brain, cavernous sinus, carotid artery via direct spread, bony erosions, vascular or perineural invasion.
Cases of rhino-orbito-cerebral mycosis are on a rising trend in post-COVID-19 patients. The attributed causes can be associated comorbidities (DM, CKD) or immunocompromised states (For e.g., immunosuppressive therapy post renal transplant, use of systemic corticosteroids in the COVID-19 treatment, ICU stay).
Radiology plays a pivotal role in the early diagnosis and determination of the extent of spread of AIFS which requires immediate and aggressive treatment strategy. Therefore, our study aims to describe the spectrum of cross-sectional imaging findings in histopathological proven cases of AIFS.
Materials and Methods
Study Population and Design
This was a prospective observational descriptive study from 1 March 2021 to 31 August 2021 (8 months) in dept of Radiodiagnosis, Apollo hospitals, Hyderabad.
We reviewed the demography, clinical details and imaging findings of 31 consecutive histopathological proven cases of post-COVID-19 AIFS. MR and CT images of these patients were retrieved from PACS and analyzed. We obtained the follow-up of these patients for a 90-day time period from the time of initial imaging.
Inclusion Criteria
Patients with post COVID-19 AIFS infection proven histopathologically.
Patients who underwent MRI of PNS + Orbit + Brain with gadolinium ± CT scan of PNS and/or orbit.
Exclusion Criteria
Patients with previous unrelated sinus/orbit/head surgery.
Demographics and Clinical Details
Age and gender of patients were recorded.
Presenting symptoms, duration between RT PCR positivity for COVID-19 infection and onset of symptoms related to AIFS, CTSS, co-morbidities, history of systemic steroid use for treatment of COVID-19 infection and post discharge 90-day follow-up were documented.
Image Acquisition
MR imaging of CNS, orbit and brain including axial, coronal and sagittal T1Wt, T2Wt, STIR, DWI images and post-contrast T1Wt images (0.1 mmol per kg of intravenous gadolinium-based MR contrast medium) were acquired which was performed using 1.5 T Philips Archieva machine.
CT scan was performed on 128 Slice Phillips Ingenuity MDCT scanner using a routine CT PNS protocol with 130 kVp and 150–220 mA tube current. CT angiography brain was performed in 1 patient after administration of 1 ml/kg low osmolar iodinated IV contrast medium using a pressure injector.
Image Analysis and Interpretation
Review was performed to look for sites, extent of involvement including extra sinus extension, orbital and cerebral involvement and signal characteristics. Each case images were analyzed by 3 senior consultant radiologists having an experience of 15 years plus in cross sectional imaging.
Presence of soft tissue opacification and their T2 signal characteristics in nasal cavity and PNS were documented. On post-contrast MR images, the pattern of soft tissue and mucosal enhancement and any extra sinus extension including pterygopalatine fossa, pterygoid muscles, infratemporal fossa, face, orbit and brain were evaluated. Bone involvement in the form of rarefaction/erosion/frank bone destruction were documented.
Proptosis, preseptal cellulitis, extraconal and retro-orbital fat stranding, EOM(s) involvement, soft tissue extension into orbit, involvement of optic nerve-sheath complex and SOV and walls of orbit including apex were considered as evidence of orbital involvement.
Meningeal enhancement, cerebritis (area of T2/FLAIR hyperintensity with or without diffusion restriction and no/minimal heterogenous enhancement, adjacent to the thickened and enhancing meninges), intracranial aneurysm, loss of flow void in major intracranial vessels, ischemic changes (evaluated on DWI images and ADC map), granuloma and cavernous sinus involvement{increased diameter (> / = 10 mm), loss of flow void, abnormal signal characteristics, convexity of lateral wall, thickening and non-enhancement on post contrast images} were considered as evidence of brain involvement.
Institution Ethics Committee
IEC clearance was obtained. Requirement of inform patient consent was waived off.
Statistical Analysis
The study data from the study proforma sheath was entered into the Master Chart on MS excel sheet. Data was analyzed using SPSS 27.0 version (Statistical Package for Social Sciences) computer software. All the qualitative and quantitative parameters were represented with frequencies and percentages.
Results
Demographics and Clinical Details
Among total of 31 patients, 25 (80.6%) were males and 6 (19.4%) were females with an average age of 49 years ranging from 22 to 75 years. 8 (25.8%) patients were below 40 years age.
Mean duration from date of testing positive for COVID-19 infection to onset of symptoms related to AIFS was 15 days (ranging from 4 to 22 days). Most common presenting complaint was facial swelling in 10 (32.2%), followed by facial pain in 8 (25.8%). Other presenting complaints included fever, nasal block, hoarseness of voice, diplopia, headache, drowsiness, ptosis, epistaxis. Four patients presented with decreased vision.
The average CTSS at diagnosis of COVID-19 infection was 14/25 (ranging from 4/25 to 23/25). Twenty-nine (93.5%) patients had history of systemic steroid use for treatment of COVID-19 infection. Comorbidities (Table 1) were present in 28 (90.3%) patients out of which 25 (80.6%) were diabetic.
Table 1.
Predisposing factors for post COVID 19 acute invasive fungal sinusitis
| Predisposing condition | Number of cases | Percentage |
|---|---|---|
| Systemic steroid use | 29 | 93.5 |
| Diabetes mellitus | 28 | 90.3 |
| Oxygen support | 24 | 77.4 |
| Other comorbidities (includes hypertension, one case of post renal transplant, CKD and hypothyroid with hypertension each) | 16 | 51.6 |
Imaging
All 31 patients had undergone MRI of PNS, orbits and brain with gadolinium-based contrast and 16 patients also underwent NECT scan of PNS and 1 patient had CT angiography of brain additionally.
Nasal Cavity Involvement
Soft tissue lesion in nasal cavity was present in all except 4 cases (87.1%). Lesions were bilateral in 15 (48.3%) cases. They showed heterogeneous signal intensity on T2W MR images except in 2 cases which were T2 hyperintense. All these lesions showed a heterogeneous post-contrast enhancement.
Loss of enhancement of turbinates were seen in all cases, common pattern being patchy loss of enhancement of the mucosa, showing ‘black turbinate sign’ as described by Safder et al. [3] (Fig. 1a). Nasal cavity involvement is detailed in Table 2.
Fig. 1.
a Case of fungal sinusitis. Axial T1 post-contrast image at the level of nasal cavity, maxillary sinus showing normally enhancing mucosa (green arrow), non enhancing soft tissue mass (blue arrow) filling left maxillary sinus with absent normal enhancement of left inferior turbinate (orange arrow); b Case of acute invasive fungal sinusitis with possible nasopharyngeal wall involvement. Axial T2 STIR image at the level of nasopharynx shows thickening of posterior nasopharyngeal wall with STIR hyperintensities suggesting possible involvement
Table 2.
Involvement of nasal cavity by post COVID 19 acute invasive fungal sinusitis
| Imaging finding | Number of cases | Percentage |
|---|---|---|
| Soft tissue lesion | 27 | 87 |
| Turbinates | 31 | 100 |
| Erosions | 26 | 83.8 |
| Loss of enhancement | 31 | 100 |
| Lateral wall emphysematous changes | 4 | 12.9 |
Nasal cavity was involved in all cases; was bilateral in 15(48.3%) cases
PNS Involvement
Maxillary sinus was the most commonly involved sinus (31, 100%) followed by ethmoid (29, 93.5%), sphenoid (24, 77.4%) and frontal (14, 45.1%) sinuses. Bilateral involvement was most common in ethmoid sinuses (24, 77.4%). Pan-sinusitis was seen in 5 (16.1%) cases. Involvement of sinuses is detailed in Table 3.
Table 3.
Involvement of paranasal sinuses in post COVID 19 acute invasive fungal sinusitis MRI and CT
| Sinus involved | Involvement of sinus (Unilateral) | Bilateral involvement | ||
|---|---|---|---|---|
| Maxillary sinus | 31 | 100% | 22 | 70.9% |
| Ethmoid sinus | 29 | 93.5% | 24 | 77.4% |
| Sphenoid sinus | 24 | 77.4% | 13 | 41.9% |
| Frontal sinus | 14 | 45.1% | 6 | 19.3% |
| Multiple sinuses involvement | ||||
| Maxillary + ethmoid | 23 | 74% | ||
| Ethmoid + sphenoid | 18 | 58% | ||
| Maxillary + ethmoid + sphenoid | 24 | 77.4% | ||
| Pansinusitis | 5 | 16.1% | ||
Soft tissue lesions in sinuses were T2 heterogeneously hyperintense with few hypointense foci within in all cases. On post contrast, lesions were non-enhancing in 16 (51.6%) (Fig. 1a) and showed subtle heterogenous enhancement in 15 (48.4%) cases. Mucosa of PNS showed patchy loss of enhancement. Involvement of maxillary sinus walls was seen in 28 (90.3%) cases.
Extra Sinus Extension (excluding orbit and brain)
Extra sinus extension was seen in the form of involvement of retro antral fat (27, 87.1%), pterygopalatine fossa (26, 83.9%), pterygoid muscles (26, 83.9%), pterygoid bones (23, 74.2%), preantral fat (23, 74.2%), infratemporal fossa (19, 61.3%) and nasopharyngeal wall (9, 29%). (Figs. 1, 2) (Table 4).
Fig. 2.
Case of acute invasive fungal sinusitis with involvement of pterygopalatine fossa and pterygoid muscles on the left. a, b CT axial and reformatted oblique sagittal sections at the level pterygopalatine Fossa showing replacement of left pterygopalatine fossa fat with soft tissue density lesion (red arrow) and destruction of the posterior wall of the left maxillary sinus. Soft tissue densities noted in maxillary sinuses. c, d T1 post-contrast axial sections without and with fat suppression at the left maxillary sinus and pterygoid fossae showing recent thickening in left maxillary sinus with mucosal enhancement and heterogeneously involving left pterygoid muscles (orange arrow)- suggesting involvement of pterygoid fossa
Table 4.
Extrasinus extension (other than orbit and brain) in post COVID 19 acute invasive fungal sinusitis MRI and CT
| Number of cases | Percentage (%) | |
|---|---|---|
| Preantral fat | 23 | 74.2 |
| Retroantral fat | 27 | 87.1 |
| Pterygoid bones | 23 | 74.2 |
| Pterygopalatine fossa | 26 | 83.9 |
| Pterygoid muscles | 26 | 83.9 |
| Infratemporal fossa | 19 | 61.3 |
| Nasopharyngeal wall | 9 | 29 |
Orbital Involvement
Orbital involvement was seen in 24 (77.4%) cases. Presence of soft tissue lesion in orbit was noted in 6 (19.3%) cases, proptosis was present in 11 (35.4%), preseptal cellulitis in 18 (58%), orbital fat stranding in 19 (61.3%) cases. EOMs were involved in 15 (48.4%) cases. Optic nerve-sheath complex and SOV involvement was present in 15 (48.4%) cases and 8 (25.8%) cases respectively (Fig. 3). Orbital involvement is detailed in Table 5.
Fig. 3.
Cases of acute invasive fungal sinusitis with orbital involvement. a CT coronal sections in bone and soft tissue window at the level of orbit and maxillary sinuses showing soft tissue density mass lesion filling the right maxillary sinus and ethmoid sinuses. Evidence of soft tissue density lesion in right orbit along the medial wall with few pockets of air in right orbit—Suggesting orbital involvement. b Axial T2 STIR image the level of orbits showing effacement of left retroorbital fat which is replaced by a heterogeneously hyperintense soft tissue density lesion (green and orange arrows), along with pre-septal soft tissue thickening and edema (grey arrow) c Postcontrast axial images at the level of orbits show proptosis of the right eyeball with bulky, non-enhancing medial rectus (yellow arrow) and non-enhancing right optic nerve sheath complex (orange arrow). Enhancing pre-septal and zygomatic area soft tissue thickening noted. Abnormal nodular enhancement of the left optic nerve sheath complex (blue arrow) noted. d Axial T2 weighted image at the level of orbits show soft tissue intensity lesion in Bilateral ethmoid sinuses, proptosis of right eyeball, effaced retroorbital fat, thickened and mildly laterally displaced medial rectus, thinned out optic nerve sheath complex and thickening of pre-septal and prenasal soft tissues. e Coronal T1Wt image at the level of maxillary sinus and orbits show soft tissue mass lesion in bilateral maxillary sinuses and right side of nasal cavity. Right inferior rectus appears bulky with mildly ill-defined outlines and subtle effacement of retroorbital fat suggesting early orbital involvement. f Coronal T2 STIR images at the level of maxillary sinuses and orbits in another case show heterogeneously hyperintense soft tissue lesion in bilateral maxillary sinuses and right side of nasal cavity. Right orbit shows heterogeneous soft tissue lesion filling the retro orbital fat Predominantly along the medial and inferior aspects
Table 5.
Orbital involvement in post COVID 19 acute invasive fungal sinusitis on MRI and CT
| Number of cases | Percentage | Imaging features | |
|---|---|---|---|
| Wall including orbital apex | 23 | 74.2 | Frank bone destruction/erosion/rarefaction |
| Proptosis | 11 | 35.4 | Distance from anterior margin of the globe to inter-zygomatic line exceeding 21 mm on axial sections |
| Preseptal cellulitis | 18 | 58 | Diffuse soft tissue stranding/STIR hyperintensity anterior to the orbital septum/Soft tissue lesion extension |
| Orbital fat involvement | 19 | 61.3 | Fat stranding/STIR hyperintensity/Soft tissue lesion extension |
| Extraconal | 19 | 61.3 | |
| Retro-orbital | 15 | 48.4 | |
| Extraocular muscles involvement | 15 | 48.4 | Increased muscle bulk/altered signal intensity/ abnormal enhancement |
| Soft tissue in orbit | 6 | 19.3 | Soft tissue lesion with signal characteristics similar to lesion in nasal cavity/paranasal sinuses |
| Optic nerve sheath complex | 15 | 48.4 | Increased diameter of optic nerve/abnormal enhancement of optic nerve sheath complex |
| Superior ophthalmic vein | 8 | 25.8 | Prominence/ loss of flow void/increased diameter |
Orbital involvement was seen in 25 (80.64%) cases
Intracranial Involvement
Intracranial involvement was seen in 14 cases (45%). Most common finding was meningeal enhancement (11, 35.4%) (Fig. 4a, b). Cerebritis was seen in 4 (12.9%) cases. One case had an intracranial aneurysm. Loss of flow void in major intracranial vessels was seen in 5 cases (16.1%) (Fig. 5d) (all these cases had a loss of flow void in Internal carotid artery, an additional loss of flow void in middle cerebral artery in 1 patient, one patient had left frontal cortical vein thrombosis). Acute ischemic changes in brain were seen in 6 (19.3%) patients, out of which 4 were arterial infarcts (Fig. 5c), 1 patient had venous infarct and 1 case was combined arterial and venous infarcts. Fungal granuloma (Fig. 4) was seen in 1 patient (3.1%); cavernous sinus thrombosis (Fig. 6c) was seen in 4(12.9%) cases (Table 6).
Fig. 4.
Cases of rhino- orbito-cerebral AIFS. Postcontrast axial section at the level of frontal horns a and post-contrast coronal sections at the level of left optic canal b show thickened dura with enhancement along the right frontal dura (arrows) c Coronal post-contrast image the level of cavernous sinus showing bulging lateral walls of left cavernous sinus with non-enhancing cavernous segment of right internal carotid artery and heterogeneously enhancing right cavernous sinus—suggestive of cavernous sinus thrombosis
Fig. 5.
Cases of acute invasive fungal sinusitis with cerebral infarction and ICA thrombosis a Maximum intensity projection of intracranial arteries superior projection and MR angiography of intracranial arteries source image at the level of circle of Willis b show absence of flow related enhancement in right internal carotid artery with faintly enhancing right MCA which is Filling through anterior communicating artery c Diffusion weighted imaging shows large infarct in Right MCA territory d axial T2Wt image at the level of sphenoid sinus in another case showing hypointense lesions within the thickened mucosa in sphenoid and left side of Nasal cavity. The same section show absent flow void the petrous segment of left internal carotid artery—suggestive of thrombosis
Fig. 6.
Case of rhino-orbito-cerebral mucormycosis with fungal granuloma. a Axial section of brain at the level of basal gangliaT2Wt axial and b FLAIR sequence show a well-defined heterogeneously hyperintense focal lesion in the region of left Caudate nucleus, the corresponding Diffusion weighted image c and ADC map d show rim of restricted diffusion at the periphery. e Pre-contrast T1Wt image shows peripheral curvilinear hyperintensity. f which is enhancing on postcontrast scan suggesting possibility of fungal granuloma
Table 6.
Intracranial involvement in post COVID 19 acute invasive fungal sinusitis on MRI and CT
| Number of cases | Percentage | |
|---|---|---|
| Meningeal enhancement | 11 | 35.4 |
| Cerebritis | 4 | 12.9 |
| Aneurysm | 1 | 3.2 |
| Loss of flow void in major vessels | 5 | 16.1 |
| Ischemic changes | 6 | 19.3 |
| Granuloma | 1 | 3.2 |
| Cavernous sinus involvement | 4 | 12.9 |
Cerebral involvement was seen in 14 (45%) cases
Clinical Management
Patients underwent nasal endoscopy followed by KOH mount/histopathology of biopsy specimen and treatment as per the discretion of MDT involving radiologists, otorhinolaryngologists, ophthalmologists, neurosurgeons and infectious diseases specialists. Treatments included course of intravenous antifungals (for e.g. liposomal amphotericin B, posaconazole, isavuconazole), Functional endoscopic sinus surgery with debridement with or without maxillectomy, CaldwelLuc procedure, orbital exenteration/evisceration.
Histopathology
Surgical specimens were sent for the histopathological examination and reports documented. All cases were confirmed to be invasive zygomycosis, out of which 13 cases were further confirmed to be caused by mucor. Two cases (6.4%) had combined mucormycosis and aspergillus infection. In our paper we had focused on Post covid AIFS and not on the further subspecies within. Hence, our imaging finding are applicable all post covid Invasive zygomycosis infection.
Follow-Up
An average 90-day follow up of 28 patients (3 lost to follow up) were obtained. It showed no further progression of disease (no new symptoms/no new radiologic findings) in 15 patients (35%). 4 patients (12.9%) with rhino-sino-orbital disease had progression to involve brain (with contralateral orbit involvement in 1 case). Two patients (6.4%) reported back with osteomyelitis of maxilla. The mortality was recorded to be 22.5% (7 patients).
Discussion
AIFS is defined as presence of tissue invasion by fungal elements over an acute clinical course of less than 4 weeks. In our study we had considered Acute invasive fungal sinusitis (AIFS) to include all the causative fungal species (mucor, Rhizopus and aspergillus etc.). However, on reviewing the literature, many studies in both pre and post covid era had focused mainly on the Mucor species as the cause of AIFS. In this pictorial review, we aimed at presenting the spectrum of imaging features of AIFS to highlight the role of radiology in aiding the decision-making process on early and appropriate treatment of this potential lethal condition.
Previous study carried out by El-Kholy et al. [4] (post COVID times) showed that the most common associated disease/ co-morbidity in mucormycosis patients was DM (27.8%). Pre COVID-19 pandemic study of mucormycosis carried by Therakathu et al. [5] also shows 91% of patients had co-morbidity of DM. Our study results are concordant with this, with DM being the most common co-morbidity (90.3%). In addition, we documented the history of systemic steroid use for the treatment of COVID-19 infection (93.5%), which was not given due consideration in any of the previous studies.
Most common presenting symptoms was facial swelling (32.2%), followed by facial pain (25.8%). Previous publications El-Kholy et al. [4] showed facial pain as the most common presenting symptom. Therakathu et al. [5] showed headache (88%) and El Naaj et al. [6] showed pain resembling sinusitis followed by facial swelling and fever as common presenting symptoms.
Our study revealed Rhino-sinus, rhino-sino-orbital involvement in 100% and 77.4% of cases respectively, similar to 100% and 80.6% in study by El-Kholy et al. [6]. Cerebral involvement was seen in 45% of our cases, which is higher than the percentage involvement in study by El-Kholy et al. [4] (27.8%).
Pathways for Spread
Nasal cavity is the primary site of infection [7]. There is rapid progression over a few days with fungi invading the mucosa, submucosa, blood vessels and bony walls of nose and PNS.
Imaging, with its ability to detect subtle evidence of invasive disease has crucial role in early diagnosis and management [3]. CT and MRI are necessary radiological modalities for investigation, with MRI having a higher soft tissue resolution. It also depicts advanced disease involving orbit or brain, extra axial space, meninges, and cavernous sinuses. CT scan is better for the detection of bony erosion/destruction.
On imaging, AIFS in initial stages may show minimal mucosal thickening or relatively small soft tissue thickening in nasal cavity and sinuses. On CEMRI, loss of enhancement of mucosa of turbinates (Black turbinate Sign) [3] is seen due to vascular/direct invasion of tissue resulting in ischemic necrosis. 3% of our cases show imaging features restricted to these early findings, in concordance with Therakathu et al. [5].
In AIFS, there is predilection for unilateral involvement of ethmoid and sphenoid sinuses as described by Therakathu et al. [7], Aribandi et al. [8] and Fatterpekar et al. [9]. Also, bone destruction has been described as a late finding in AIFS in the previous literature [11, 12]. However, our study showed bilateral involvement of sinuses in 87% of cases, with maxillary sinus being the most common (100%). This is likely due to destruction of wall of maxillary sinus from adjacent disease process in nasal cavity, which explains the greater percentage of involvement of maxillary sinus compared to cases of mucormycosis unrelated to COVID-19.
Finding of large number of cases with bone destruction in our study maybe due to bone destruction early in the course of disease or due to quick progression to advanced disease, indicating aggressive nature of the disease per-se which may be further catalysed by immunomodulation by COVID-19 infection.
16% of our cases showed an additional extra sinus extension (excluding orbit and brain) (Table 4). Peri-antral fat obliteration is subtle sign of extension [13] and must be carefully viewed in all patients at risk for AIFS. 50% of patients with isolated rhino-sinus disease who also had signal intensity changes in peri-antral fat presented to us with residual/recurrent disease post conservative surgery during follow up. There was no recurrent/residual lesion in patients without pre-operative peri-antral involvement. This highlights the need for due consideration of this subtle finding before planning surgery. 6.4% of our cases had involvement of peri-antral fat without adjacent bone destruction. Various theories have been put forward to explain early peri-antral soft tissue involvement without adjacent bone erosions [13]. Fungal inclination to cause thrombosis resulting in vascular congestion, thereby manifesting on CT/MR as peri-antral oedema; alternatively, peri-antral soft tissue involvement could be due to fungal elements spreading along blood vessels or spread through perivascular spaces into peri-antral fat.
Hosseini et al. [14] proposed pterygopalatine fossa as the main conduit for extension of rhino nasal mucormycosis to other sites. Pterygopalatine fossa communicates with nasal cavity via sphenopalatine foramen, infratemporal fossa via pterygomaxillary fissure and inferior orbital fissure opens into orbital apex forming pathway for spread of disease process. Hosseini et al. [14] further suggested that in a patient with signs of orbital involvement or progressive ocular involvement, debridement of pterygopalatine fossa and inferior orbital fissure should form an integral part of surgery to clean up the major reservoir of the fungi and prevent invasion of cranial cavity. Our study 83.9% had involvement of pterygopalatine fossa, out of which 73% showed extension of disease process to infratemporal fossa and 92% to orbit. All patients whose infratemporal fossa were involved had involvement of pterygopalatine fossa.
Spread of fungal infection to orbit causes extensive tissue damage leading to potential permanent vision loss and death if not treated [15]. 92% of our cases with orbital involvement had medial wall erosion and 96% had pterygopalatine fossa involvement, suggesting the possibility of an aggressive disease spread directly through bony wall erosion or through infraorbital fissure from the pterygopalatine fossa. 1 patient with involvement of extraocular fat by disease process had intact orbital walls. These findings suggest the possibility of pterygopalatine fossa being the conduit for spread of infection.
Various potential ways of spread of AIFS to brain have been proposed which include osseous erosion (of skull base), perivascular extension (across the cribriform plate of ethmoid bone to frontal lobes or through the orbital apex to the cavernous sinuses) or hematogenous spread from arterial wall invasion leading to occlusive ischemic damage and/or abscesses formation. Perineural extension being an unusual mode of spread [16, 17].
Meningeal enhancement with intracranial invasion can be subtle and must be carefully sought for in all cases with suspicion of intracranial involvement [8]. Leptomeningeal enhancement adjacent to the cribriform plate was seen in 7 (22.5%) cases out of which 3 showed erosion of cribriform plate and 4 cases had an intact cribriform plate, suggesting a possibility of perineural/perivascular spread as well.
Our study had involvement of cavernous sinus in 12.9% of cases, which is in concordance with the pre COVID-19 statistics of Therakathu et al. [5], while Joshi et al. [10] in COVID-19–related mucormycosis study found a higher percentage of cases with cavernous sinus involvement (36%), likely due to differences in demographics. In our study, cavernous sinus involvement without orbital involvement was seen in one case and it was due to direct spread of disease process through destruction of sphenoid sinus wall. One case had perineural spread via cranial nerves III to V (Fig. 4d). Rest of these cases showed involvement of sphenoid sinus and orbit, either of which being the possible source of infection spread to cavernous sinus.
With progression of disease, adjacent cerebritis, granulomas and cerebral abscess might be seen [4]. Our study showed cerebritis in 28.5% and granulomas in 7% of cases with intracranial involvement. None of our cases showed progression to cerebral abscess on imaging.
Ischemic changes in brain parenchyma were seen in 19.3% of cases, most common cause being internal carotid artery thrombosis. Cortical venous thrombosis and direct spread of lesion were found to be the cause in one case each.
Pre COVID-19 case reports of mycotic aneurysm of internal carotid artery in invasive fungal sinusitis have been published [18–20]. One of our cases which had bilateral maxillary, left ethmoid and sphenoid sinus involvement had mycotic aneurysm of ICA at presentation.
Mortality rate reported in COVID-19–related AIFS by El-Kholy et al. [4] is 36.1% while in our study in the follow up, the mortality is 22.5%. Pre COVID-19 studies have reported a higher mortality of 40–60% [21, 22].
Limitation of our study is its retrospective nature in a single institute and non-availability of CT scan of all cases and lack of controls (non-Covid 19 AIFS cases). Dose, duration and type of steroid used couldn't be documented as we had focused mainly on imaging spectrum in this paper.
In conclusion, comprehensive knowledge of the imaging features of AIFS and the recognition of extent of their spread allows the radiologists to play a pivotal role in alerting the clinician for appropriate therapy to avoid protracted and fatal outcome.
Acknowledgements
We extend our sincere gratitude to Dr Samiya Razvi, DCH, MD, MS for promoting scientific temperament in our institute.
Abbreviations
- AIFS
Acute invasive fungal sinusitis
- PNS
Paranasal sinuses
- DM
Diabetes mellitus
- CKD
Chronic kidney disease
- ICU
Intensive care unit
- PACS
Picture archiving and communication system
- CTSS
CT severity score
- STIR
Short tau inversion recovery
- DWI
Diffusion weighted imaging
- FLAIR
Fluid attenuation inversion recovery
- NCCT
Non-enhanced contrast computed tomography
- EOM
Extra ocular muscle
- SOV
Superior ophthalmic vein
Funding
No funds, grants, or other support was received.
Declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical Approval
Ethical Clearance has been obtained from the institutional ethics committee.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Equal first authorship by Noamaan Muhammed and Sneha Hiriyanna.
References
- 1.Chakrabarti A, Denning DW, Ferguson BJ, et al. Fungal rhinosinusitis: a categorization and definitional schema addressing current controversies. Laryngoscope. 2009;119:1809–1818. doi: 10.1002/lary.20520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Roden MM, Zaoutis TE, Buchanan WL, Knudsen TA, Sarkisova TA, Schaufele RL, et al. Epidemiology and outcome of zygomycosis: a review of 929 reported cases external icon. Clin Infect Dis. 2005;41(5):634–653. doi: 10.1086/432579. [DOI] [PubMed] [Google Scholar]
- 3.Safder S, Carpenter JS, Roberts TD, et al. The “Black Turbinate” sign: an early MR imaging finding of nasal mucormycosis. AJNR Am J Neuroradiol. 2010;31(4):771–774. doi: 10.3174/ajnr.A1808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.El-Kholy NA, El-Fattah AMA, Khafagy YW. Invasive Fungal sinusitis in post COVID-19 patients: a new clinical entity. Laryngoscope. 2021 doi: 10.1002/lary.29632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Therakathu J, Prabhu S, Irodi A, Sudhakar SV, Yadav VK, Roopa V. Imging features of rhinocerebral mucormycosis: study of 43 patients. Egypt J Radiol Nucl. 2018;49:447–452. doi: 10.1016/j.ejrnm.2018.01.001. [DOI] [Google Scholar]
- 6.El-Naaj IA, Leiser Y, Wolff A, Peled M. The surgical management of rhinocerebral mucormycosis. J Craniomaxillofac Surg. 2013;41:291–295. doi: 10.1016/j.jcms.2012.03.019. [DOI] [PubMed] [Google Scholar]
- 7.Gillespie MB, O’Malley BW, Jr, Francis HW. An approach to fulminant invasive fungal rhinosinusitisin immunocompromised host. Arch Otolaryngol Head Neck Surg. 1998;124(5):520–526. doi: 10.1001/archotol.124.5.520. [DOI] [PubMed] [Google Scholar]
- 8.Aribandi M, McCoy VA, Bazan C. Imaging features of invasive and non-invasive fungal sinusitis: a review. Radiographics. 2007;27:1283–1296. doi: 10.1148/rg.275065189. [DOI] [PubMed] [Google Scholar]
- 9.Fatterpekar G, Mukherji S, Arbealez A, Maheshwari S, Castillo M. Fungal diseases of the paranasal sinuses. Semin Ultrasound CT MR. 1999;20(6):391–401. doi: 10.1016/S0887-2171(99)90023-9. [DOI] [PubMed] [Google Scholar]
- 10.Joshi AR, Muthe MM, Patankar SH, Athawale A, Achhapalia Y. CT and MRI findings of invasive mucormycosis in the setting of COVID-19: experience from a single center in India. AJR Am J Roentgenol. 2021 doi: 10.2214/AJR.21.26205. [DOI] [PubMed] [Google Scholar]
- 11.Del Gaudio JM, Clemson LA. An early detection protocol for invasive fungal sinusitis in neutropenic patients successfully reduces extent of disease at presentation and long term morbidity. Laryngoscope. 2009;119:180–183. doi: 10.1002/lary.20014. [DOI] [PubMed] [Google Scholar]
- 12.Gillespie MB, O’Malley BW. Role of middle turbinate biopsy in the diagnosis and management of fulminant invasive fungal rhinosinusitis. Laryngoscope. 2000;110:1832–1836. doi: 10.1097/00005537-200011000-00013. [DOI] [PubMed] [Google Scholar]
- 13.Silverman CS, Mancuso AA. Periantral soft tissue infiltration and its relevance to the early detection of invasive fungal sinusitis: CT and MR findings. AJNR Am J Neuroradiol. 1998;19:321–325. [PMC free article] [PubMed] [Google Scholar]
- 14.Hosseini SM, Borghei P. Rhinocerebral mucormycosis: pathways of spread. Eur Arch Otorhinolaryngol. 2005;262(11):932–938. doi: 10.1007/s00405-005-0919-0. [DOI] [PubMed] [Google Scholar]
- 15.Klotz SA, Penn CC, Negvesky GJ, Butrus SI. Fungal and parasitic infections of the eye. Clin Microbiol Rev. 2000;13:662–685. doi: 10.1128/CMR.13.4.662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sandron J, Hantson P, Duprez T. Intracranial brain parenchymal spread of mucormycosis through olfactory tract: a diffusion-weighted imaging-based concept. Acta Radiol Open. 2020;9(12):2058460120980999. doi: 10.1177/2058460120980999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kandpal H, Aneesh MK, Seith A, Sharma S. Symptomatic perineural extension of fungal sinusitis in an immunocompetent person: imaging features. Singapore Med J. 2008;49(7):e171–e174. [PubMed] [Google Scholar]
- 18.Hurst RW, Judkins A, Bolger W, Chu A, Loevner LA. Mycotic aneurysm and cerebral infarction resulting from fungal sinusitis: imaging and pathologic correlation. AJNR Am J Neuroradiol. 2001;22(5):858–863. [PMC free article] [PubMed] [Google Scholar]
- 19.Azar MM, Assi R, Patel N, Malinis MF. Fungal mycotic aneurysm of the internal carotid artery associated with sphenoid sinusitis in an immunocompromised patient: a case report and review of the literature. Mycopathologia. 2016;181(5–6):425–433. doi: 10.1007/s11046-015-9975-1. [DOI] [PubMed] [Google Scholar]
- 20.Dusart A, Duprez T, Van Snick S, Godfraind C, Sindic C. Fatal rhinocerebral mucormycosis with intracavernous carotid aneurysm and thrombosis: a late complication oftranssphenoidal surgery? Acta Neurol Belg. 2013;113(2):179–184. doi: 10.1007/s13760-012-0151-9. [DOI] [PubMed] [Google Scholar]
- 21.Herrera DA, Dublin AB, Ormsby EL, et al. Imaging findings of rhinocerebral mucormycosis. Skull Base. 2009;19(02):117–125. doi: 10.1055/s-0028-1096209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Turner JH, Soudry E, Nayak JV, Hwang PH. Survival outcomes in acute invasive fungal sinusitis: a systematic review and quantitative synthesis of published evidence. Laryngoscope. 2013;123(5):1112–1118. doi: 10.1002/lary.23912. [DOI] [PubMed] [Google Scholar]