Outcome of Total Surgical Debridement of Covid Associated Rhino-Orbito-Cerebral Mucormycosis Based on a New Surgical Staging System: A Cohort Study

Abstract

To propose a surgical staging system with management protocol for post-covid Rhino-orbito-cerebral mucormycosis (ROCM) with central skull base osteomyelitis. A prospective cohort study of a total of 193 post-covid ROCM patients was conducted between May 2021 and January 2022 at a tertiary care centre. Patients were assessed radiologically and staged from I to V. Follow up period was 16 months and the surgical outcome in terms of recurrent disease was assessed. A total of 193 patients (129 primary and 64 revision) were studied. Maxilla was found to be the epicenter of anterior disease (69.3%) and pterygoid wedge was noted to be the epicenter of posterior disease (85.6%). More than 65% of our patients, at the time of presentation, presented with involvement of the central skull base. Intracranial disease was noted in 13.9% of patients and the mortality rate was 6.2%. This staging system provides a systematic step-by-step protocol for the management of ROCM, with emphasis on meticulous disease clearance at the central skull base.

Introduction

Covid-19 Associated Mucormycosis (CAM) has been reported in many countries like Austria, Brazil, France, India, Iran, and the US [1, 2]. An expeditious increase of Mucormycosis in the Indian context appears to be due to the large uncontrolled diabetic population and excessive use of corticosteroids during treatment for Covid-19. This rare, intriguing entity became of immense public health importance due to its angio-invasive nature leading to high morbidity and mortality rate [3].

The recent surge of Rhino-orbito-cerebral mucormycosis (ROCM) may be attributed to hypoxia, hyperglycemia (diabetes, steroid-induced), acidic medium (metabolic acidosis, diabetic ketoacidosis), high ferritin levels and decreased phagocytic activity due to immunosuppression [4]. Covid 19 associated ROCM is the most frequent presentation of the disease [5]. Given its proximity to the central skull base, the infection travels through fissures and suture lines to the Haversian system of the compact bone. Further angioinvasion and arterial thrombosis lead to tissue necrosis. Therefore, a high index of suspicion, accurate radiological assessment with the aid of Magnetic Resonance Imaging (MRI) and pathological confirmation forms the key to diagnosis.

The steep rise of cases and lack of clear-cut surgical guidelines in the management of invasive skull base mucormycosis prompted us to come up with an evidence-based protocol for ROCM, after reviewing our own consecutive patients. Following a meticulous analysis of the disease extent, we proposed a classification system to map the extent of the disease. Thus, guiding the surgeon to further plan the surgical approach. This staging system can standardize reporting of imaging, treatment protocol and could be utilized for disease prognosis and prediction of complications. The above protocol further describes surgical strategies of aggressive debridement with respect to anatomical landmarks to curtail mortality. The present study is the first attempt in the literature to propose a clinical-surgical management algorithm and multidisciplinary treatment strategies including skull base debridement for ROCM with Central Skull Base Osteomyelitis (CSBO) and cranial bone involvement.

Methods

A prospective cohort study was conducted on 193 adult patients, who presented to our tertiary referral center with clinically and microbiologically proven ROCM and history of recent Covid-19 infection. The period of recruitment was between May 2021 and January 2022. Informed consent was obtained from all individual participants included in the study.

A total of 129 primary cases were included in this cohort study with 64 cases previously operated elsewhere. Cases were categorized as proven ROCM if clinical, radiological, microbiological, or pathological features were positive. In revision cases on antifungal therapy, PCR-based molecular techniques have been shown to have high sensitivity for confirming fungal strain. Non-covid ROCM with CSBO were excluded from the study. All patients were followed up for an average of 16 months following surgery (Till May 2023).

Magnetic resonance imaging (MRI) of Paranasal sinuses with Orbit and Brain (T1W, T1W STIR (short tau inversion recovery) with contrast, T2W, T2W FS and Diffusion Weighted Imaging (DWI)) with Computed Tomography of Paranasal Sinus (CT PNS) was performed to assess the extent of the disease. Patients with clinical features suggestive of palatal involvement were evaluated using Cone Beam Computed Tomography Scan (CB-CT) of maxilla. In cases with evidence of gross bone erosion and frontal table involvement, an MDCT with 3-D facial reconstruction was performed. A KOH smear and tissue biopsy was done for all patients pre-operatively. Glycemic control and other co-morbidities were strictly monitored and managed.

Due to varied presentation and lack of universal surgical guidelines, a staging system named after the senior author [Janakiram Staging for Post-covid Mucormycosis (JSPM)] was proposed and is depicted in Table 1. Cases were assigned to the respective stages and a suitable approach was decided. All the involved and surrounding tissues were sent for histopathology to assess the degree of involvement and confirm the diagnosis. Anti-fungal therapy was administered pre and post operatively with Liposomal Amphotericin-B depending upon the severity of disease and drug tolerance. Maintenance therapy was given with oral Posaconazole.

Table 1.

Janakiram staging for Post Covid Mucormycosis (JSPM) with surgical protocol

Stage	Areas involved		Surgical approach
Stage I	Rhino-palatal
IA	Osteomeatal complex, Middle turbinate, Inferior turbinate, Maxillary sinus, Ethmoid sinus, Frontal sinus		Endoscopic Medial Maxillectomy and complete Spheno-Ethmoidectomy
IB	Limited Unilateral or Unilateral with Central Palatal Involvement (With preserved palatal Mucosa)		Endoscopic Modified Denker’s Approach (EMDA) with dissection of Pterygopalatine fossa and Infratemporal fossa
Stage II	Rhino-orbital
IIA	Involvement of bony orbital walls-medial and inferior orbital wall (with or without orbital fat involvement)		Endoscopic Orbital Decompression + Trans-cutaneous Retro-orbital Amphotericin B injection (TRAMB)
IIB	Involvement of extra-ocular muscles (Extraconal compartment)		Trans-nasal trans-orbital endoscopic Globe & Optic nerve Sparing Orbital Debridement—with TRAMB
IIC	Involvement of posterior orbit including superior orbital fissure (SOF) and Orbital Apex (OA)		Endoscopic Trans-pterygoid clearance with TRAMB
IID	Involvement of anterior and posterior globe		Orbital Exenteration
Stage III	Central skull base osteomyelitis
IIIA	Median—Pterygoid wedge, Planum Sphenoidale, Sella, Clivus, Sphenoid sinus, Floor and Keel of sphenoid sinus and Rostrum		Endoscopic midline trans-sphenoidal approach to clear the Sella, Clivus, floor and lateral recess of Sphenoid
IIIB	Paramedian Paramedian Anterior Compartment (PAC)—Medial and Lateral pterygoid plates, V 2 branch of Trigeminal nerve and Quadrangular space Paramedian Middle Compartment (PMC)—Cavernous sinus Paramedian Posterior Compartment (PPC)—Meckel's cave, Medial Petrous		Trans-pterygoid Supra-petrous approach to Meckel's cave and cavernous sinus
IIIC	Lateral V3, Greater Wing of Sphenoid, Base of Temporal Lobe		Transpterygoid Supra-petrous approach to base of temporal lobe
	Lateral Orbital Wall, Temporal Fossa		Endoscopic + Hemi/Bicoronal Approach
III D	Skin, Palatal involvement with mucosa		Trans-facial approaches—Total maxillectomy with Endoscopic skull base debridement
STAGE IV	Involvement of bones of cranium (Ethmoid (cribriform and crista galli), Frontal, Parietal, Zygoma and Superior orbital wall) with/ without extradural abscess		Combined Approach (Endoscopic + External)
IVA	Involvement of bones of cranium without extradural abscess	Unilateral frontal Bilateral frontal Zygoma Temporal bone Parietal bone	Endoscopic + Lynch Howarth approach
IVB	Involvement of bones of cranium With extradural abscess		Endoscopic + Bicoronal incision with Pericranial flap
		Ethmoid bone	Anterior craniofacial resection (ACFR)
Stage V	Intracranial Involvement		Endoscopic/Transcranial Approaches
VA	Frontal and Temporal Intracranial abscess		Trans nasal Endoscopic Approach
VB	Intracranial: Dural or Pial Involvement		Trans nasal Endoscopic + transcranial approaches

Post-operative endoscopic assessment of vascularity and mucosalization was done weekly for the first two months and monthly for the next six months. The treatment outcome in terms of residue or recurrence was evaluated with serial MRI scans after one, three and sixth months after surgery.

Statistical Analysis

The normality of the variables was assessed using Kolmogorov–Smirnov test. Quantitative variables have been expressed as mean and median (IQR). Qualitative variables have been summarized using proportions. The distribution of the diseases in various sites between primary and revision cases was compared and the difference in proportions was statistically assessed using Chi-square test. Similarly, orbital involvement between primary and revision cases were also assessed. A p-value of less than 0.05 was considered to be statistically significant. Analysis was done using SPSS software version 25.

Results

The mean age was 48.9 years (10.6). Out of the total 193 patients, 154 (79.8%) were males and 39 (20.2%) were females. A total of 129 cases (66.8%) accounted for primary and 64 cases (33.2%) for revision cases. Associated Co-morbidities were Diabetes Mellitus (n = 175), Steroid intake (n = 152), Oxygen support (n = 84), Hypertension (n = 74) and ventilatory support (n = 8).

Stage wise radiological and intra-operative distribution of primary and revision cases as per JSPM is depicted in Fig. 1a, b. After analysis of both the distributions, it was noted that in primary disease stage I was most common followed by stage IIIC. Similarly in revision cases stage IIIC numbered the most followed by stage IVA.

Fig. 1.

a, b Comparison between graphical representation of radiological and intraoperative distribution of primary and revision cases as per JSPM staging (graphs were made using excel spread sheet)

Most common site among the primary 129 cases was maxilla (66%) and least common site was orbital apex (3.1%). Among the 64 revision cases, most common site was pterygoid wedge (75%) and least common was cavernous carotid and skin (both 3.1%).

Anatomically the disease was classified as anterior (stage I to stage IID) and posterior (stage IIIA–VB). Out of the total 193 cases, 38.8% (n = 75) patients were staged as anterior disease and 61.2% (n = 118) patients were staged with posterior disease.

Involvement of maxilla was found to be 69.3% (n = 52) in anterior disease(n = 75) compared to 51.7% (n = 61) in posterior disease (n = 118) (p value = 0.015 statistically significant), thus suggesting that, maxilla forms the epicentre of anterior disease.

Similarly, out of 118 patients with posterior disease, n = 101 patients (85.6%) showed involvement of pterygoid wedge as compared to n = 3 patients (4%) among the anterior disease (n = 75, P value < 0.001), thereby concluding, pterygoid wedge forms the epicentre of posterior disease.

Comparing the site distribution of disease in primary and revision cases and their statistical significance is elaborated in Table 2. It was observed that maxilla was significantly more involved in primary than revision cases.

Table 2.

Comparison of site distribution of disease in primary and revision cases

Sites	Primary (n = 129)	Revision (n = 64)	‘p’ value
Maxilla	86 (66.7%)	27 (42.2%)	0.001
Pre-maxilla	23 (17.8%)	11 (17.2%)	Not significant
Retro-maxilla	42 (32.65)	34 (53.1%)	0.006
Palate	38 (29.5%)	27 (42.4%)	0.078
Sphenoid sinus	60 (46.5%)	47 (73.4%)	< 0.001
Pterygoid wedge	56 (43.4%)	48 (75.0%)	< 0.001
GWS	41 (31.8%)	44 (68.8%)	Not significant
Clivus	24 (18.6%)	24 (37.5%)	0.004
Superior-orbital fissure	14 (10.9%)	18 (28.1%)	0.002
Orbital apex	11 (8.5%)	16 (23.4%)	0.004
Paraclival carotid	20 (16.5%)	18 (28.1%)	0.038
Cavernous sinus	6 (4.7%)	3 (4.7%)	Not significant
Frontal bone	3 (2.3%)	13 (20.3%)	< 0.001
Crista galli	5 (3.9%)	7 (10.9%)	Not significant
Temporal fossa	1 (0.8%)	16 (25.0%)	< 0.001
Zygoma	2 (1.6%)	16 (25.0%)	< 0.001
Temporal abscess	10 (7.8%)	10 (16.6%)	Not significant
Frontal abscess	6 (4.7%)	6 (9.4%)	Not significant

The retro-maxilla was involved in 53.1% of revision and 32.6% of primary cases (p value < 0.006). Subsequently the pterygoid wedge was also involved in 75% of revision and 43.4% of primary cases. This substantiates that the spread of disease to posterior skull base is faster from the retro-maxilla.

Out of the n = 75 patients with anterior disease, n = 65 patients were primary and n = 10 were revision cases. The frequency of involvement of sites is given in Table 3. Maxilla is involved in 73.8% of primary cases whereas only 40% of revision cases showed involvement of maxilla. This is possibly because they have already been operated and maxillary clearance was given previously.

Table 3.

Anterior disease—sites involved in primary and revision cases

Site	Primary (n = 65)	Revision (n = 10)
Pre-maxilla	13 (20.0%)	1 (10%)
Maxilla	48 (73.8%)	4 (40.0%)
Anterior ethmoid	35 (53.8%)	0 (0.0%)
Retro-maxilla	25 (38.5%)	1 (10.0%)
Frontal sinus	17 (26.2%)	3 (30.0%)
Palate	20 (30.8%)	4 (40.0%)

Similarly, out of the 118 patients with a posterior disease n = 64 were primary and n = 54 were revision cases. The different sites involved and their frequency is given in Table 4. In primary and revision cases, involvement of pterygoid wedge and sphenoid was significant.

Table 4.

Posterior disease -sites involved in primary and revision cases

Site	Primary (n = 64)	Revision (n = 54)
Sphenoid	57 (89.1%)	47 (87.0%)
Pterygoid wedge	53 (82.8%)	48 (88.9%)
Greater wing of sphenoid	40 (62.5%)	44 (81.5%)
Clivus	24 (37.5%)	24 (44.4%)
Paraclival carotid	20 (31.2%)	18 (33.3%)
Cavernous Sinus	6 (9.4%)	3 (5.6%)
Frontal Bone	3 (4.7%)	13 (24.1%)
Crista galli	4 (6.2%)	7 (13.0%)
Temporal fossa	0 (0.0%)	16 (29.6%)
Zygoma	1(1.6%)	16(29.6%)
Temporal abscess	7 (10.9%)	9 (16.7%)
Frontal abscess	5 (7.8%)	6 (11.1%)
Cavernous carotid	3 (4.7%)	2(3.7%)
Petrous carotid	2 (3.1%)	3 (5.6%)

With respect to orbital disease, individual wall involvement as well as number of exenterations were more in revision cases. In those with superior orbital fissure syndrome, ptosis and restricted eye mobility resolved partially during follow up, but visual loss persisted.

In the entire cohort of 129 primary cases, n = 7 patients had disease residue. Among the 64 revision cases, only n = 2 cases recurred, both patients were operated for temporal lobe abscess and subsequently underwent craniotomy and abscess drainage. Currently they are on follow up and disease free.

Amongst 193 patients, n = 12 (6.2%) patients died, out of which n = 5 patients were primary cases and n = 7 were revision cases. The cause of death is charted in Table 5.

Table 5.

Details of cause of death and time interval from surgery

	Stage of disease	Cause of death	Time interval from surgery
Primary cases
1	IIIC	Acute MI	10 days
2	VB	Systemic fungal disease	1 month
3	VB	Acute PE	20 days
4	IIIC	Carotid blowout	2 months
5	VB	Cardio-respiratory arrest	15 days
Revision cases
1	VB	Frontal abscess/meningitis	16 days
2	IVA	Acute PE	2 months
3	VB	Pre-existing liver failure	1 month
4	IIIB	RF	2 months
5	IIIC	RF	2 months
6	IIIC	Systemic fungal disease	1 month
7	IIIC	Systemic fungal disease	1 month

MI Myocardial infarction, PE Pulmonary Embolism, RF Renal Failure

Discussion

ROCM is an invasive, life-threatening opportunistic infection in immunocompromised patients. The most common clinical signs noted are facial oedema, loss of vision, proptosis and nasal discharge [6]. The exact pathogenesis of rhino-cerebral mucormycosis and its locoregional spread are not clearly understood.

Pattern of Spread

It is generally believed that Mucorales infection results from inhalation of fungal spores which are normal commensals of the airway [7–9]. Spores can inoculate the nasal mucosa, spreading to the paranasal sinuses, orbit and finally the intracranial fossa, when the host is immunocompromised [8, 9]. As stated by Sannathimmappa et al. [10], excess steroid administration for the treatment of covid infection in the presence of uncontrolled diabetes facilitated germination of fungal spores.

Mucorales are vaso-tropic and are notorious for their ability to cause extensive vessel thrombosis and tissue infarction [9, 11]. The damage and penetration through endothelial cells lining the blood vessels is a critical step for dissemination [11, 12]. In diabetics and those with impaired immune systems, the fungal inhibitory responses are suboptimal, and germination ensues [13, 14].

In studies (Spellberg et al. and Ibrahim et al.) it was observed that the presence of nonviable R.oryzae can lead to endothelial damage, in part explaining the inefficiency of antifungal drugs and surgical debridement as the mainstay of treatment [11, 12]. Eradication is difficult due to the aimless cycle of arterial invasion and occlusion, tissue hypoxia and necrosis, thereby making the delivery of intravenous drugs to the site of infection difficult [12, 15].

The perivascular and peri-sinus spread play a crucial role in dissemination of disease beyond normal sinus walls. It is thought to be along the neurovascular bundles of pterygopalatine and infratemporal fossae, via the multiple cranial foramina [9]. Inferior orbital nerve involvement can explain the initial pain and paresthesia in this region [9]. The mucor then spreads along the nerve to the facial soft tissues with an intact anterolateral wall.

Potential extra-arterial routes of orbital spread in mucormycosis include direct spread from the orbit, infraorbital nerves, nasolacrimal duct and the lamina papyracea. Sclera represents a natural barrier that inhibits direct fungal invasion. Vascular routes of ocular invasion include retrograde propagation from the cavernous sinus or ophthalmic veins or the embolization via the ophthalmic, central retinal, and ciliary arteries. Thus, ischemia may result from occlusion of the internal carotid artery, ophthalmic artery, central retinal artery, or posterior ciliary artery. The internal carotid artery is one of the most frequently thrombosed vessels and is often found to be invaded [16].

Among our patients, the primary ROCM cases demonstrated significant signs of early osteomyelitis in the pterygoid wedge and floor of the sphenoid sinus. The authors firmly believe that the posterior spread of the disease starts with the nidus in the pterygoid wedge and spreads towards the central skull base. The spread from the pterygoid wedge can also occur by venous sinusoids and Haversian canals. Fungus becomes established in the marrow and provokes thrombophlebitis leading to bone sequestration, which assists in the proliferation of hyphae, thus resulting in tissue hypoxia. In our series, the clival bone demonstrated maximum involvement due to high density of venous sinusoids.

Mucosal involvement of frontal sinus was noted from the orbit, septum and ethmoid sinus. In addition, direct erosion of the frontal bone leads to subperiosteal abscess, subdural collections, meningitis, and encephalitis.

Intracranial involvement can occur through the superior orbital fissure, ophthalmic vessels, perineural route via olfactory nerves through the cribriform plate or the carotid artery [17, 18].

Due to an unexpected spike in the number of cases, lack of human resources, inadequate essential supplies, unavailability of universal guidelines and paucity of infrastructure, we experienced an initial struggle in the management of ROCM at our centre. Therefore, the need for an effective and systematic surgical staging system was realised and implemented. Our study proposes a detailed protocol for the surgical management of invasive skull base fungal osteomyelitis, based on the radiological assessment of disease and its progression. After thorough analysis, a novel staging system was designed to standardize the reporting and management. This could facilitate the surgeon in counselling the patients scientifically, regarding the extent of involvement, probable surgical outcome and post-operative rehabilitation.

The surgical treatment for mucormycosis over time was ‘debridement till healthy bleeding (15). However, we stress on ‘defining the debridement with anatomical boundaries’ for better surgical outcome. Unlike malignancies, the disease pattern in ROCM may not always be contiguous. Therefore, a higher stage does not necessarily mean involvement of structures in the lower stages [19]. But in overlapping stages, patients are automatically staged into the highest possible stage.

In the proposed staging system entire region of sinuses to skull base is divided into anterior and posterior compartments.

The anterior compartment is subdivided into rhino-palatal and rhino-orbital, which forms stage I and II of our classification respectively. The rhino-palatal stage is again subdivided into stage Ia and Ib.

Stage Ia includes the premaxillary region and paranasal sinuses except sphenoid sinus and Ib involves limited palatal disease with preserved mucosa (Fig. 2a).

Fig. 2.

a MRI Brain with paranasal sinus and Orbit T2W STIR coronal image depicting the involvement of Stage I—involvement of all paranasal sinuses except sphenoid sinus. b MRI Brain with paranasal sinus and Orbit T1W coronal image with contrast showing the involvement of anterior and posterior globe—Stage Iid. c MRI Brain with paranasal sinus and Orbit T1W coronal image with contrast depicting the involvement of central skullbase—clivus, right pterygoid wedge and greater wing of sphenoid (Stage III), post medial maxillectomy status noted in right maxilla. d MRI Brain with paranasal sinus and Orbit T1W coronal image with contrast depicting gross erosion of frontal and parietal bones on the right side (Stage IV). e MRI Brain with paranasal sinus and Orbit T1W with contrast coronal image showing right sided temporal lobe abscess (Stage V)

Stage II of the classification is divided into 4 sub-groups. Stage IIa includes involvement of the medial and inferior orbital walls, IIb incorporates involvement of extraocular muscles. Disease extending to superior orbital fissure and orbital apex is grouped under Stage IIc. Stage IId marks the involvement of anterior and posterior globe (Fig. 2b). Oculoplastic surgeon administers Transcutaneous Amphotericin B (TRAMB) injection to all patients in stages IIa to IIc, at a dose of one ml of 3.5 mg/ml of Amphotericin for 3 days.

From stage III, posterior compartment with central skull base osteomyelitis (CSBO) commences (Fig. 2c). It is subdivided into A (Median), B (Paramedian), C (Lateral) and D (palate and/or facial skin).

In stage III disease, we noticed that those cases with involvement of clivus, floor of sphenoid sinus and bilateral greater wing of sphenoid, following complete surgical clearance gave the appearance of a ‘cross’ sign in the post-operative MRI. This is termed as ‘Holy Cross Sign’ (Fig. 3a, b).

Fig. 3.

a MRI Brain with paranasal sinuses and Orbit T1W with contrast coronal image depicting the involvement of clivus, sphenoid sinus with floor of sphenoid, bilateral greater wing of sphenoid. b Post-operative MRI Brain with paranasal sinuses and Orbit T2W STIR coronal image depicting clearance from clivus to floor of nasal cavity and bilateral greater wing of sphenoid—Holy Cross sign

Involvement of bones of the cranium is classified under Stage IV (Fig. 2d).

Stage V involves intracranial disease (Fig. 2e). The surgical approaches to each stage are explained in Table 1.

An extensive review of literature revealed multiple staging systems proposed by various authors; Honavar et al. (n = 2669) described a 4 tier staging based on anatomical progression of disease, Soni et al. (n = 145) proposed another 4 tier staging based on clinical and radiological evaluation, Vaid et al. (n = 65) described a severity grading system. A guideline for management for mucormycosis by Cornely et al. was found, but this study was on the overall management rather than surgical protocol [20–23]. The JSPM staging (n = 193) in the current study is based on radiological imaging and has defined a debridement protocol based on anatomical landmarks. A comparison with the recent staging systems and our staging is depicted in Table 6.

Table 6.

A comparison between current published staging systems and JSPM

Classifications	Stages	Based on	No. of patients	Central skull base and cranial bone debridement
Soni et al. [20]	I–IV	Clinical & radiological	145	Not specific
Honavar et al. [22]	1–4	Anatomical progression of ROCM	2669	Not included
Vaid et al. [23]	Mild to Very severe	Radiological	65	Not included
Naik et al. [24]	ROC (1–3)	CECT/MRI	10	Not included
JSPM Staging	I–V	Radiological	193	Included

A male preponderance (79.8%) was noted in our study which was consistent with other studies in literature [23, 24]. This could be due to factors like increased smoking and compromised airway mucosa with poor inherent immune response in males, but requires further studies. The mean age was 48.9 years. Young diabetic patients under 40 years consisted of 42% of the total cohort, indicating the rising trend of diabetes in the younger Indian population.

A total of 129 cases (66.8%) accounted for primary and 64 (33.2%) for revision cases. Due to the lack of amphotericin and delay at presentation, two-third of revision patients presented with advanced stages (60.9% in stage IIIc and IVa).

The spectrum of head and neck Mucormycosis consisted of 50% of the total presentations, as per a study conducted by Michael Dan [25]. In consensus with that, the patterns described in our study are; Rhino-palatal, Rhino-Orbital, central skull base osteomyelitis, cranial bone involvement and intracranial disease.

Out of the total cohort 9.4% of patients did not have any of the traditional risk factors. In a study by Patel et al. and another study by Soni et al. they reported 12% and 11% of total patients of ROCM without any co-morbidities respectively [20, 26]. Therefore, on afterthought, whether the disease is due to the viral infection or due the treatment for it, is not clear.

It is a well-established fact that diabetes is the most common predisposing factor for Mucormycosis. Excessive use of corticosteroids especially in them, further impedes the host immunity. Among our total patients 90.6% of patients were diabetics and 78.7% were treated with corticosteroids.

‘Global guidelines for the management of mucormycosis (2019)’, a study by Cornely et al. [21] preferred MRI only in orbital and intracranial extension, whereas in our study MRI was the primary radiological investigation for all patients. In ROCM, the time interval between diagnosis and treatment is highly critical, because any delay in the treatment increases mortality by double fold [27]. CT scans tend to miss out on early diseases without bone erosion and thus increase this interval and accelerate the mortality rate.

Statistically significant number of cases with anterior disease had maxillary involvement and cases with posterior disease had involvement of pterygoid wedge, thus supporting our interpretation of maxilla and pterygoid wedge being the epicentre in anterior and posterior disease respectively. Eighty percent of the cohort presented with ROCM involvement of the central skull base, at the initial presentation (36.5% primary and 43.7% revision cases). Sphenoid sinus, clivus, pterygoid wedge, greater wing of Sphenoid and posterior orbit were significantly more involved in revision cases. Paraclival internal carotid artery was involved in 31.2% and 33.3% of primary and revision posterior disease respectively.

Figure 1a (Radiological staging) was compared to Fig. 1b (Intra-operative staging) and it was noted that, intra-operative staging was more advanced than radiological staging. This was possibly because the interval between imaging and surgery was more than two weeks, due to non-compliance for surgery and hence the disease progressed. This reiterates the importance of correlating radiological findings with intraoperative findings.

Perineural spread via involvement of vidian nerve (37.8%), V2 branch of trigeminal nerve (39.5%), facial nerve (2.0%) and olfactory nerves (1.5%) were observed in our study. In a study by Parsi et al. [28], perineural involvement of trigeminal root and middle cranial fossa via retrograde spread from infraorbital nerve and neurovascular bundle of pterygopalatine fossa has been described.

Inadequate clearance of frontal sinus in primary disease results in spread to frontal tables and crista galli, thus supporting the significant increased percentage (37.1%) of cranial bone involved cases in revision. Though temporal and frontal abscesses were also found more in revision patients (16.7% and 11.7% respectively), this finding was not statistically significant, probably due to the smaller number of cases. Almost all cases with involvement of zygoma and temporal fossa were seen among revision.

There are no definitive criteria to differentiate between residue and recurrence in Mucormycosis. In our study residual disease was seen in 4.6% of cases which is comparatively much less compared to other contemporary studies [19].

Out of the total cohort, 94.8% patients had no evidence of disease during follow-up.

A total of 12 patients (6.2%) died. The mortality rate was also comparatively lesser when compared to the recent studies [19, 23].

Surgery

Considering the abrupt dissemination of infection from the skull base to neurological structures and the poor bioavailability of drugs in necrotic tissue, aggressive rhino-orbital and skull base debridement is advocated to reduce morbidity and mortality. The primary trans-nasal access was through Endoscopic Modified Denker’s approach (EMDA), which aids in ‘four handed bi-nostril technique’. This technique allows better visualization and the advantage of a panoramic view. EMDA also facilitates trans-pterygoid approaches and aids in post-operative inspection of operated areas to pick up early disease changes.

Conclusion

The second wave of covid-19 infection was accompanied by a sudden surge in the number of patients with ROCM. Even though multiple staging systems have been described in the literature, clear-cut guidelines of management of central skull base osteomyelitis and cranial bones have been unclear. This staging system can help surgeons on the step-by-step approach to central skull base debridement respecting the anatomical landmarks with minimal complications. It helps in achieving better surgical outcomes in terms of minimal mortality rate.

The steep rise in cases of Human Immunodeficiency virus in the 1980s helped the medical fraternity in formulating diagnostic and treatment protocols, because of which the disease is on a decline now. Similarly, through this study conducted during a very wide spread of Mucormycosis, we have formulated principles in diagnostic and treatment strategies. Even though the rapid number of cases have come down, since Covid and Mucormycosis still persists in the society, this staging system may be used anywhere in the world to treat this fatal disease adequately.

Funding

This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Declarations

Conflict of interest

All authors report no conflicts of interest relevant to this article.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.