Imaging in Rhino-Orbital-Cerebral Mucormycosis in COVID-19 Patients - A Retrospective Study

Introduction

Mucormycosis is a fungal infection and the fungi are ubiquitous in their distribution. It affects those with uncontrolled diabetes, especially ketoacidosis, prolonged administration of injectable steroids, chemotherapy, organ transplantation, haematological malignancies, iron overload, injury to the skin, and chronic ailments.¹ The condition gained prominence as ‘Black funguses’ during the second wave of COVID-19 exhibiting high mortality. The spores of mucor gain entry through the nasal cavity. There is an extensive, rapid destruction of the tissues. The disease begins as an infection of the nasal mucosa and sinuses, progresses to involve eye, superior orbital fissure and orbital apex. Ultimately there is cerebral involvement with spread of infection through superior orbital fissure and cribriform plate.

The present study aimed to evaluate the role of contrast CT and contrast MRI in diagnosing the disease and also evaluating the prognosis of the ROCM. The study was also utilized to guide planning of surgery by knowing the extent of the disease and to identify the therapeutic response to different modalities of treatment.

To evaluate the role of CT and MRI in diagnosing and monitoring the treatment of Rhino-orbitocerebral mucormycosis, and to assess the outcome of the disease. 

Materials and Methods

Retrospectively, assessment of 60 patients was done over the period of one year at our institute, KBU Faculty of Medical Sciences, Kalaburagi. The diagnosis of ROCM was based on contrast-enhanced (CE) CT and Contrastenhanced MRI images, and was subsequently confirmed by biopsy and potassium hydroxide (KOH) staining. Multi-slice CT images of the involved areas were acquired through Seimens Somatom 16 slice CT and through Seimens Essenza 1.5 T equipment. The pre- and post-contrast images were used for accurate visualization of the disease. All 60 patients were hospitalized for COVID-19 at the time of imaging. Thirty-five patients had a history of diabetes and 20 patients were being treated with mechanical ventilation for COVID-19 pneumonia. All patients were treated with injectable corticosteroids (treatment duration of 10–14 days). All patients were treated in the ICU for ROCM with the use of IV amphotericin B. Thirty patients underwent surgical debridement with orbital exenteration. Fourteen patients succumbed to the disease.

Limitation of the present study was that only a few patients were correlated with CT findings. It would have been beneficial if CT was correlated for all the patients.

Important radiological progressive or resolving findings would be done if patients follow up would be done adequately.

Demographic data, clinical presentation, investigations (USG and CT scan reports), intraoperative findings, complications, and details of hospital stay were recorded. All the data was analysed using SPSS, Version 24.0

Inclusion criteria

All hospitalized patients for COVID-19 at our institute at the time of imaging.

Exclusion criteria

1. Patients who were COVID-19 positive but not being treated in ICU for COVID-19/ROCM.

2. Patients who were not being treated with injectable corticosteroids.

3. Patients who were not being treated with IV amphotericin in the ICU for ROCM

4. Patient whose diagnosis changed afterwards.

5. Patients lost to follow up. 

Results

The epicentre of the lesion in 65% of the cases was middle turbinate. There were areas of absent enhancement within the mucosal thickening. It correlated with pathological involvement of the disease by fungal hyphae. In 70% of cases, the ethmoid air cells were commonly involved, followed by the involvement of the maxillary sinuses (55%). In 20% of the patients, there was a permeative / moth-eaten pattern of destruction and it was predominantly seen in the medial wall and floor of the orbit. In majority of the cases, there was involvement of the middle turbinate. Some cases exhibited mild widening of the pterygopalatine fissure also. In 15% of patients, with post-operative orbital clearance, tethering of the medial rectus muscle was noted. In post-contrast study, the involved extra-ocular muscles appeared bulky with relatively increased enhancement. In our study, the frequency of optic nerve involvement was 5%. The route for the intracranial extension was through the optic canal and cribriform plate. There was mild ballooning of the sinus in 10% of the patients who presented with ptosis. There was absent flow void in the cavernous segment of the internal carotid artery (ICA) due to thrombosis in 8% of cases. Fourteen patients succumbed to the disease.

Discussion

Mucormycosis affects the sinuses, turbinates, bone and soft tissues, orbits, optic nerve and spreads intracranially. The present study showed mucosal thickening of the sinus on post-contrast MR images. Characteristically, there was no enhancement within the mucosal thickening. These regions were involved by fungal hyphae.

There was involvement of middle turbinate in 65% of cases and they appeared mildly enlarged and bulky. They also exhibited hypointense areas. There was absence of enhancement in post-contrast images. There was focal mucosal non-enhancement. It indicated presence of areas of devitalized necrotic mucosal soft tissue in ROCM.² Infiltration of the peri-antral fat planes may represent the earliest imaging evidence of invasive fungal disease, especially when it is seen as the sole radiological finding in an appropriate clinical setting.³ In our study, we found involvement of the periantral fat plane as the earliest predictor of the disease even in the absence of involvement of the sinuses, and progression of the disease into the orbito-cerebral region.

Destruction of bony sinus wall with tiny air foci within it and infiltration of the peri-antral soft tissue was frequently noted. The presence of air within the bony structures was uncommon in other infections.⁴

The predominant bony regions involved in our study in decreasing order of frequency were, the anterior wall of the maxillary sinus, medial wall of the orbit, medial and lateral pterygoid plates, greater and lesser wings of sphenoid bone, maxillary bone, hard palate, zygoma, and anterior skull base. Marrow edema was noted on STIR/FAT SATSEQ within the affected bones.

Most cases in our study showed involvement of the middle turbinate. Some exhibited mild widening of the pterygopalatine fissure also. There was significant hyperintensity of the medial and lateral pterygoid muscles. There was significant fatty stradding and hyperintense signals in the retro-orbital fat due to inflammation and oedema.

Involvement of optic nerve has been noted.⁵ It was noted in cases involving the optic canal or orbital apex. There can be spread of infection from the orbital apex, posteriorly through the superior orbital fissure into the cavernous sinus, and through the inferior orbital fissure across the pterygopalatine fossa into the infratemporal fossa.⁶ CT and MRI showed a spectrum of findings in rhinocerebralmucormycosis.⁷ Complete debridement of necrotic tissue improves survival.⁸ The major specific finding in ROCM was the periantral fat involvement suggestive of angioinvasive disease. Focal non enhancement of the mucosa or turbinate corresponded to areas of active disease / fungal hyphae. In CT images, tiny focal air lucencies seen within the infiltrated bone favoured the diagnosis of fungal osteomyelitis. Altered signal intensity was seen around the pterygoid plates involving the pterygoid muscles on MR images. Orbital extension was predominantly through the medial wall of orbit. Intracranial extension of infection from the orbital apex posteriorly, through the superior orbital fissure into the cavernous sinus and through the inferior orbital fissure across the pterygopalatine fossa into the infratemporal fossa was noted. There was also extension along the cribriform plate into the basifrontal region. Involvement of cavernous sinus and cavernous segment of RT ICA was seen.

Conclusion

Contrast CT and MRI plays a vital role in accurately establishing the extent of the disease and its correlation during surgery. Sequential MRI provided additional information to manage the cases either medically or surgically. Assessment of disease extent by imaging is crucial for planning surgical debridement.

Conflicts of Interest

None