Rhino-orbito-cerebral mucormycosis (ROCM) in eastern Iran: a six-year dual-center retrospective observational study of 329 patients

Abstract

Background

Invasive mucormycosis is a rapidly progressive angioinvasive mycosis with high mortality. We aimed to describe epidemiology, presentation, diagnostics, antifungal utilization, in-hospital outcomes, sex-based descriptive differences, and available sociodemographic variables among patients managed at referral hospitals in eastern Iran.

Methods

We conducted a retrospective observational study of all hospitalized rhino-orbito-cerebral mucormycosis (ROCM) cases at two tertiary hospitals in Mashhad, March 2018–March 2024. Inclusion required compatible clinical features plus pathological confirmation. Data were abstracted from charts, imaging archives, and pathology. Endoscopic, radiologic, and histopathologic findings; comorbidities; therapies; and in-hospital outcomes were summarized and compared using descriptive statistics.

Results

Among 329 patients (52.9% male), median age was 53 years (IQR 31–59); 71.4% lived in urban areas and 23.7% required ICU care. Diabetes was present in 45.6% and immunosuppressive therapy in 51.1%. Endoscopy (n = 326) most often showed necrosis (54.3%) and crusting (37.4%). Across 571 imaging procedures, mucosal thickening (68.1%) and sinus opacification (67.8%) predominated; orbital involvement occurred in 13.7%. Of 495 biopsies, the nasal septum was the commonest site (22.8%); histology reported necrosis in 49.1% and angioinvasion in 4.6%. L-AmB was administered to 56.8% and deoxycholate to 28.6%; combination regimens were less frequent. Overall, in-hospital mortality was 30.4%; survival to discharge occurred in 58.7%. Standardized 6-week or 12-week mortality could not be calculated from the available retrospective records. Unadjusted survival varied across antifungal regimens, but dual and triple antifungal groups were small; therefore, these descriptive differences should not be interpreted as evidence of comparative efficacy or superiority of combination antifungal therapy.

Conclusions

In this large retrospective cohort of referral-hospital ROCM, diagnosis commonly involved endoscopy, cross-sectional imaging, and tissue-based confirmation, and most patients received amphotericin B formulations with azole-containing combinations used in a minority. In-hospital mortality was 30.4%; however, lack of standardized severity staging, treatment-timing metrics, surgical debridement data, and antifungal dosing/duration precludes attributing outcomes to specific care processes. Observed differences between male and female patients in socioeconomic variables and regimen selection are descriptive and warrant evaluation in datasets with baseline severity and timing variables.

Introduction

Invasive fungal diseases cause substantial global morbidity and mortality, with recent estimates suggesting approximately 6.5 million invasive fungal infections and 3.8 million deaths annually, of which about 2.5 million are directly attributable to fungal disease [1]. Mucormycosis represents only a subset of this global burden but is clinically important because it is rapidly progressive, angioinvasive, and associated with tissue necrosis and high case-fatality rates [2].

Mucormycosis, caused by fungi of the order Mucorales, is an uncommon yet severe invasive mycosis that has emerged as a significant global health threat over the past two decades [3]. The incidence of this life-threatening infection has been steadily increasing, a trend attributed to the expanding population of immunocompromised individuals, including those with hematological malignancies, solid organ transplants, poorly controlled diabetes mellitus, and recipients of immunosuppressive therapies [4]. The ubiquitous nature of Mucorales in the environment, combined with their ability to rapidly invade blood vessels and cause devastating tissue destruction through angioinvasion and thrombosis, makes this infection a medical emergency requiring immediate recognition and aggressive multimodal treatment [5].

The epidemiological landscape of mucormycosis demonstrates striking geographical disparities, with India bearing the highest reported burden globally, estimated at approximately 140 cases per million population—a figure nearly 80 times higher than rates observed in developed nations [6]. This regional concentration is further exemplified by systematic reviews indicating that the majority of reported cases originate from Asia, with India contributing most significantly to the Asian case burden [7]. The COVID-19 pandemic dramatically exacerbated this global health challenge, leading to an unprecedented surge in COVID-19-associated mucormycosis (CAM) cases, particularly in regions with high baseline prevalence of diabetes mellitus and widespread corticosteroid usage [3].

The clinical presentation of mucormycosis varies significantly based on the anatomical site of infection, with rhino-orbital-cerebral mucormycosis (ROCM) representing the most common form, accounting for approximately 69.3% of cases [7]. Diagnostic challenges in mucormycosis are substantial, requiring a high index of clinical suspicion combined with characteristic endoscopic findings—most notably black necrotic eschar covering nasal mucosa—and detailed radiological assessment to determine disease extent [8]. Histopathological examination remains central to diagnosis, revealing characteristic broad, irregular, ribbon-like, pauci-septate or aseptate hyphae with right-angle or irregular branching, often accompanied by tissue invasion, necrosis, and angioinvasion when present [9, 10].

Treatment outcomes in mucormycosis remain disappointingly poor despite advances in antifungal therapy and surgical techniques, with overall mortality rates ranging from 40 to 50% across most series [7, 11]. Mortality varies significantly based on underlying host factors, with rates reaching 66% in malignancy patients and up to 76% in those with hematological diseases [12]. Current ECMM/MSG ERC global guidance emphasizes early imaging to define disease extent, prompt surgical intervention when feasible, and immediate initiation of effective antifungal therapy. High-dose L-AmB is strongly recommended as first-line therapy, while isavuconazole and posaconazole are Mucorales-active azoles used as salvage or continuation therapy; amphotericin B deoxycholate is discouraged because of toxicity but may remain relevant where lipid formulations are unavailable [10].

Given these substantial knowledge gaps and the urgent need for comprehensive epidemiological data to inform clinical practice and public health policy, we conducted a comprehensive analysis of mucormycosis patients from two largest general hospitals in eastern Iran and the primary referral centers for mucormycosis in the region. The primary objectives of this study were to: (1) characterize the demographic and clinical features of ROCM in our regional population; (2) describe observed differences between male and female patients in risk factors, clinical presentation, diagnostic findings, and treatment patterns; (3) analyze patterns of antifungal drug utilization and in-hospital outcomes descriptively; (4) report available sociodemographic variables, specifically residence and employment status; and (5) provide comprehensive data on endoscopic, radiological, and histopathological findings to contribute to the understanding of ROCM in the Middle East region. Because socioeconomic data were limited to residence and employment status, these variables were interpreted descriptively rather than as a comprehensive assessment of socioeconomic status.

Methods

Study design and setting

This retrospective observational study was conducted at two tertiary-care teaching hospitals affiliated with Mashhad University of Medical Sciences: Imam Reza (AS) Hospital and Ghaem Hospital, Mashhad, Iran. The study period spanned six years, from March 2018 to March 2024.

Participants and selection criteria

All patients diagnosed with mucormycosis and hospitalized at the participating centers during the study period were screened for eligibility using a census sampling approach due to the low incidence of this condition.

Inclusion criteria comprised patients with ROCM, defined by (1) compatible clinical manifestations involving rhino-sinus, orbital, and/or cerebral structures and (2) tissue-based mycological confirmation documented in contemporaneous pathology or microbiology reports. Case adjudication relied on the documented diagnostic report in the medical record rather than retrieval or re-review of the original diagnostic modality for every case. When original slides, culture results, species-level identification, or modality-specific source documents were unavailable, cases were retained only if the clinical record contained a tissue-based report explicitly supporting mucormycosis in a clinically compatible ROCM syndrome. This approach is consistent with research definitions of proven invasive mold disease, which recognize histopathologic, cytopathologic, or direct microscopic evidence of hyphae in tissue together with associated tissue damage [9].

Exclusion criteria included: (1) incomplete medical records, (2) patients who discontinued treatment against medical advice, (3) patients with mucormycosis involving anatomical sites other than rhinocerebral-orbital regions, and (4) cases where diagnostic or therapeutic procedures could not be completed.

Case definition

ROCM was defined as a compatible rhino-sinusitis syndrome with orbital and/or cerebral involvement plus tissue-based mycological confirmation documented in the clinical record. In routine care at the participating centers, pathologic confirmation was rendered when tissue histopathology and/or direct microscopy demonstrated tissue-invasive fungal hyphae morphologically compatible with Mucorales. The histopathological criteria used in report abstraction included broad, irregular, ribbon-like, pauci-septate or aseptate hyphae with right-angle or irregular branching, often associated with tissue necrosis and/or angioinvasion when documented. These features were used to distinguish Mucorales-compatible morphology from Aspergillus-like molds, which more typically show narrower, regularly septate hyphae with acute-angle dichotomous branching [9, 10]. Because culture, molecular confirmation, species-level identification, and exact diagnostic modality were not consistently retrievable across charts and archives, cases were not stratified by confirmation modality. Centralized slide re-review was not performed; therefore, residual diagnostic misclassification with other invasive fungal sinusitides, including aspergillosis or mixed mold infection, cannot be completely excluded.

Data collection and variables

Data were collected using a study-specific structured checklist. Sources included medical records, pathology archives, imaging databases, and hospital information systems. The following variables were systematically collected across several key domains.

Demographic and clinical characteristics included age and sex, department of admission, and ICU admission status. We use the term sex rather than gender because the available variable reflected recorded male/female biological classification, whereas sociocultural gender identity was not captured [13]. Underlying comorbidities were documented, encompassing diabetes mellitus, immunosuppressive therapy, non-hematological malignancies, and chronic kidney disease. Immunosuppressive therapy was defined as documented exposure to systemic corticosteroids, chemotherapy, transplant-related immunosuppression, biologic or other immunomodulatory therapy, or treatment-associated neutropenia when recorded. Because the retrospective records did not consistently distinguish all subcategories, this variable was analyzed as a composite exposure. Risk factors such as recent trauma and dental procedures, particularly tooth extractions, were also recorded. Socioeconomic factors included residential area classification as urban or rural and employment status.

Clinical presentation and diagnostic findings encompassed endoscopic examination results, specifically documenting mucosal changes including paleness, necrosis, and crusting, as well as the presence of polyps, nasal discharge, and ulceration. Imaging studies were comprehensively documented, including the types of imaging modalities utilized such as CT, MRI, chest radiography, and ultrasound, along with the anatomical areas examined. Radiological findings were systematically recorded, including orbital involvement, mucosal thickening, sinus opacification, bone destruction patterns affecting sinus, orbital, and nasal/oral regions, and the presence of cavernous sinus thrombosis.

Pathological assessment involved documentation of biopsy sites and the anatomical distribution of tissue sampling. Histopathological findings were recorded, specifically noting the presence of angioinvasion, inflammatory changes, and tissue necrosis patterns.

Treatment and outcome variables included documentation of antifungal therapy regimens, encompassing amphotericin B deoxycholate, L-AmB, L-AmB plus posaconazole, L-AmB plus posaconazole plus caspofungin, and amphotericin B with other triazoles as recorded in the source charts. For interpretive purposes, step-down or continuation therapy was considered Mucorales-active only when posaconazole or isavuconazole was specified. Combination antifungal therapy was defined as the use of two or more systemic antifungal agents. This term was distinguished from combined medical-surgical management, which refers to antifungal therapy plus surgical source control. In-hospital outcomes were comprehensively tracked, including survival to discharge, in-hospital mortality, self-discharge against medical advice, and transfers to other healthcare facilities.

Statistical analysis

Data analysis was performed using R version 4.5.1 and SPSS version 23. Continuous variables were summarized using medians and interquartile ranges when distributions were skewed or non-normal.

Descriptive statistics are presented as frequencies and percentages for categorical variables. Continuous variables are expressed as mean ± standard deviation or median with interquartile range (IQR). No formal hypothesis testing was used to draw causal inferences; comparisons were interpreted descriptively.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and received approval from the Ethics Committee of Mashhad University of Medical Sciences (approval code: IR.MUMS.MEDICAL.REC.1401.479). Given the retrospective nature of the study, informed consent requirements were waived by the ethics committee, with patient confidentiality maintained throughout all study procedures.

Results

Demographic characteristics and clinical variables

Table 1 presents comprehensive demographic and clinical characteristics for 329 patients with a nearly balanced sex distribution of 174 males (52.9%) and 155 females (47.1%). Continuous variables are summarized using medians and interquartile ranges.

Table 1.

Demographic findings, predisposing factors, and treatment regimen of patients diagnosed with mucormycosis

Variable	Total, 329 (100%)	Male, 174 (52.9%)	Female, 155 (47.1%)
Age
Years, [IQR]	53.00 [31.00–59.00]	54.50 [40.00–68.00]	49.50 [31.00–64.00]
Department of Admission
Department of Infectious Diseases and Tropical Medicine	47 (14.2%)	27 (15.5%)	20 (12.9%)
Department of Internal Medicine	116 (35.2%)	61 (35.1%)	55 (35.5%)
Department of Otorhinolaryngology	90 (27.3%)	45 (25.9%)	45 (29.0%)
Other	76 (23.1%)	41 (23.6%)	35 (22.6%)
ICU admission
Yes	78 (23.7%)	47 (27.0%)	31 (20.0%)
No	251 (76.3%)	127 (73.0%)	124 (80.0%)
Risk factors
Diabetes mellitus	150 (45.6%)	80 (46.0%)	70 (45.2%)
Immunosuppressive therapy	168 (51.1%)	100 (57.5%)	68 (43.9%)
Non-hematological malignancies	80 (24.3%)	31 (17.8%)	49 (31.6%)
Chronic kidney disease	31 (9.4%)	15 (8.6%)	16 (10.3%)
Trauma	3 (0.9%)	2 (1.1%)	1 (0.6%)
Tooth extraction	5 (1.5%)	2 (1.1%)	3 (1.9%)
Sociodemographic characteristics
Region of residence
Urban	235 (71.4%)	125 (71.8%)	110 (71.0%)
Rural	94 (28.6%)	49 (28.2%)	45 (29.0%)
Job status
Employed	164 (49.8%)	109 (62.6%)	55 (35.5%)
Unemployed	159 (48.3%)	62 (35.6%)	97 (62.6%)
Student	6 (1.8%)	3 (1.7%)	3 (1.9%)
Outcome
Survived - Discharged	193 (58.7%)	98 (56.3%)	95 (61.3%)
Survived - Self-discharged/checkout	14 (4.3%)	7 (4.0%)	7 (4.5%)
Expired	100 (30.4%)	56 (32.2%)	44 (28.4%)
Referred for continuation of inpatient treatment/investigations	22 (6.7%)	13 (7.5%)	9 (5.8%)
Drug treatment
Administered drugs
Amphotericin B deoxycholate	94 (28.6%)	59 (33.9%)	35 (22.6%)
L-AmB	187 (56.8%)	101 (58.0%)	86 (55.5%)
L-AmB plus posaconazole	28 (8.5%)	11 (6.3%)	17 (11.0%)
L-AmB plus posaconazole plus caspofungin	11 (3.3%)	1 (0.6%)	10 (6.5%)
Amphotericin B plus other recorded triazole	9 (2.7%)	2 (1.1%)	7 (4.5%)

The overall median age was 53.00 years [IQR: 31.00–59.00]. The majority of patients were admitted to the Department of Internal Medicine (35.2%) and Department of Otorhinolaryngology (27.3%), while 76.3% did not require ICU admission. Notable comorbidities included diabetes mellitus in 45.6% of patients and composite immunosuppressive therapy exposure in 51.1% of cases. Non-hematological malignancies were present in 24.3% of patients.

Most patients resided in urban areas (71.4%) and showed differences between male and female patients in employment status, with 62.6% of males employed compared to 35.5% of females, while 62.6% of females were unemployed compared to 35.6% of males. Regarding treatment outcomes, 58.7% of patients survived to discharge, while 30.4% expired during their hospital stay. The predominant antifungal therapy was liposomal amphotericin B (L-AmB), administered to 56.8% of patients, followed by amphotericin B deoxycholate in 28.6% of cases, with combination therapies being less frequently utilized. The data suggests this represents a mixed population of patients with various underlying conditions and risk factors receiving antifungal treatment across multiple hospital departments.

Endoscopic findings

Endoscopic examination was performed in 326 patients (171 males, 155 females), with findings summarized in Table 2. The most common finding was necrosis, observed in 177 patients (54.3%), with a higher prevalence in males (58.5%) compared to females (49.7%). Crusting was documented in 122 patients (37.4%): 73 males (42.7%) and 49 females (31.6%). Nasal discharge was present in 103 patients (31.6%), slightly more common in females (34.8%) than males (28.7%). Paleness was documented in 72 patients (22.1%), with similar rates in males and females (21.6% and 22.6%, respectively). Polyps and ulceration were rare findings, present in 19 patients (5.8%) and 2 patients (0.6%) respectively.

Table 2.

Endoscopic findings of the patients

Endoscopy findings	Total, 326 (100%)	Male, 171 (52.5%)	Female, 155 (47.5%)
Paleness	72 (22.1%)	37 (21.6%)	35 (22.6%)
Polyp	19 (5.8%)	10 (5.8%)	9 (5.8%)
Crusting	122 (37.4%)	73 (42.7%)	49 (31.6%)
Discharge	103 (31.6%)	49 (28.7%)	54 (34.8%)
Ulceration	2 (0.6%)	2 (1.2%)	0 (0.0%)
Necrosis	177 (54.3%)	100 (58.5%)	77 (49.7%)

Imaging modalities

A total of 571 imaging procedures were analyzed across the cohort. Computed Tomography (CT) scan was the most frequently utilized imaging modality, accounting for 425 procedures (74.4% of all imaging). Within the CT category, imaging of the nose and sinuses represented the majority with 238 procedures (56.0% of CT scans), followed by imaging of other anatomical areas with 174 procedures (40.9%), and brain imaging with 13 procedures (3.1%). Magnetic Resonance Imaging (MRI) was performed in 70 cases (12.3% of total imaging), with the vast majority consisting of orbit or brain scans at 66 procedures (94.3% of MRIs), while only 4 scans (5.7%) targeted other regions. Chest X-rays comprised 75 procedures (13.1% of total imaging), and there was a single ultrasound examination of the lungs and heart (0.2% of total imaging).

Pathological imaging findings

Pathological findings on imaging studies were analyzed across all 329 patients, as detailed in Table 3. The most prevalent findings were mucosal thickening (68.1%, n = 224) and sinus opacification (67.8%, n = 223), which were observed at similar frequencies between male and female patients. Mucosal thickening was present in 66.7% of males and 69.7% of females, while sinus opacification occurred in 66.7% of males and 69.0% of females.

Table 3.

Pathological findings of the imaging

Pathological findings of imaging	Total, 329 (100%)	Male, 174 (52.9%)	Female, 155 (47.1%)
Orbital involvement
Yes	45 (13.7%)	20 (11.5%)	25 (16.1%)
No	284 (86.3%)	154 (88.5%)	130 (83.9%)
Mucosal thickening
Yes	224 (68.1%)	116 (66.7%)	108 (69.7%)
No	105 (31.9%)	58 (33.3%)	47 (30.3%)
Sinus opacification
Yes	223 (67.8%)	116 (66.7%)	107 (69.0%)
No	106 (32.2%)	58 (33.3%)	48 (31.0%)
Sinus bone destruction
Yes	5 (1.5%)	4 (2.3%)	1 (0.6%)
No	324 (98.5%)	170 (97.7%)	154 (99.4%)
Orbital bone destruction
Yes	24 (7.3%)	14 (8.0%)	10 (6.5%)
No	305 (92.7%)	160 (92.0%)	145 (93.5%)
Other Nasal / Oral bone destruction
Yes	7 (2.1%)	5 (2.9%)	2 (1.3%)
No	322 (97.9%)	169 (97.1%)	153 (98.7%)
Cavernous sinus thrombosis
Yes	3 (0.9%)	1 (0.6%)	2 (1.3%)
No	326 (99.1%)	173 (99.4%)	153 (98.7%)

Orbital involvement was documented in 45 patients (13.7%): 20 males (11.5%) and 25 females (16.1%). Orbital bone destruction was observed in 24 patients (7.3%): 14 males (8.0%) and 10 females (6.5%). Sinus bone destruction occurred in 5 patients (1.5%): 4 males (2.3%) and 1 female (0.6%). Other nasal/oral bone destruction was present in 7 patients (2.1%): 5 males (2.9%) and 2 females (1.3%). Cavernous sinus thrombosis was rare, occurring in only 3 patients (0.9%), slightly more frequent in females (1.3%) than males (0.6%).

Anatomical distribution of biopsies

A total of 495 biopsies were obtained from various anatomical sites across the cohort, as shown in Table 4. The nasal septum was the most frequently biopsied site (22.8%, n = 113), followed by the right middle turbinate (11.5%, n = 57) and left middle turbinate (10.1%, n = 50). Other commonly sampled locations included the right inferior turbinate (8.4%, n = 42), nasal floor (8.0%, n = 40), and hard palate (7.6%, n = 38).

Table 4.

Anatomical sites which biopsies were obtained

Sites of biopsy	Total, 495 (100%)	Male, 271 (54.7%)	Female, 224 (45.3%)
Nasal Septum	113 (22.8%)	61 (22.5%)	52 (23.2%)
Right middle turbinate	57 (11.5%)	28 (10.3%)	29 (12.9%)
Left middle turbinate	50 (10.1%)	27 (10.0%)	23 (10.3%)
Right inferior turbinate	42 (8.4%)	23 (8.5%)	19 (8.5%)
Left inferior turbinate	36 (7.2%)	20 (7.4%)	16 (7.1%)
Nasal roof	28 (5.6%)	16 (5.9%)	12 (5.4%)
Nasal floor	40 (8.0%)	24 (8.9%)	16 (7.1%)
Right lateral nasal wall	30 (6.0%)	16 (5.9%)	14 (6.2%)
Left lateral nasal wall	33 (6.6%)	19 (7.0%)	14 (6.2%)
Hard palate	38 (7.6%)	20 (7.4%)	18 (8.0%)
Soft palate	28 (5.6%)	17 (6.3%)	11 (4.9%)

The distribution of biopsy sites showed a slight male predominance overall, with males accounting for 271 biopsies (54.7%) compared to 224 biopsies (45.3%) in females. The nasal septum remained the most frequently biopsied site in both males and females, representing 22.5% of male biopsies and 23.2% of female biopsies. The distribution of biopsy sites was broadly similar between male and female patients.

Histopathological findings

Histopathology report variables were available for 326 cases, and we abstracted prespecified fields from the report text (angioinvasion, necrosis, inflammation). Angioinvasion was reported in 15 cases (4.6%) and absent in 311 cases (95.4%). Inflammation was observed in 26 biopsies (8.0%), while the remaining 300 biopsies (92.0%) showed no signs of inflammation. Necrosis was reported in 160 cases (49.1%), while it was absent in the other 166 cases (50.9%).

Treatment outcomes by antifungal regimen

Table 5 presents treatment outcomes for patients receiving different antifungal therapy regimens. The majority of patients (56.8%) received L-AmB as their primary treatment, which demonstrated the highest absolute survival rate with 129 patients (69.0%) surviving to discharge, though 58 patients (31.0%) in this group expired. Amphotericin B - Deoxycholate was administered to 94 patients (28.6%), resulting in 64 survivors (68.1%) and 30 deaths (31.9%). Combination therapies were less frequently employed, with L-AmB plus posaconazole used in 28 patients (8.5%), yielding 21 survivors and 7 deaths. The least common treatments included L-AmB plus posaconazole plus caspofungin triple therapy in 11 patients (3.3%) and amphotericin B plus other recorded triazoles in 9 patients (2.7%). Overall, across all treatment modalities, 229 patients (69.6%) were alive at hospital outcome (discharged, self-discharged, or referred), while 100 patients (30.4%) expired. These regimen-specific outcomes are unadjusted descriptive findings and should not be interpreted as evidence of comparative efficacy, particularly because the dual and triple antifungal groups were small and treatment selection was likely influenced by clinical severity, comorbidities, drug availability, and survivor bias.

Table 5.

Outcomes according to antifungal treatments in patients diagnosed with mucormycosis

Outcome	Amphotericin B - Deoxycholate	L-AmB	L-AmB plus posaconazole	L-AmB plus posaconazole plus caspofungin	Amphotericin B plus other recorded triazole
Alive at hospital outcome (discharged/self-discharged/referred)	64 (68.1%)	129 (69.0%)	21 (75.0%)	9 (81.8%)	6 (66.7%)
Expired	30 (31.9%)	58 (31.0%)	7 (25.0%)	2 (18.2%)	3 (33.3%)
Total	94 (28.6%)	187 (56.8%)	28 (8.5%)	11 (3.3%)	9 (2.7%)

Figure 1. Summary of predisposing conditions, diagnostic findings, antifungal regimens, and in-hospital outcomes in 329 patients with rhino-orbito-cerebral mucormycosis. (A) Predisposing conditions by sex. (B) Key diagnostic findings across endoscopic (n = 326), radiological (n = 329), and histopathological (n = 326) modalities. (C) In-hospital outcome by antifungal regimen. (D) Overall in-hospital outcome distribution. AmB-D,Amphotericin B Deoxycholate; L-AmB, Liposomal Amphotericin B; Posa, Posaconazole. *Regimen-specific outcomes are unadjusted and subject to confounding by indication; they should not be interpreted as evidence of comparative efficacy.

Fig. 1.

provides a graphical summary of predisposing conditions, key diagnostic findings, antifungal regimens, and in-hospital outcomes across the cohort

Discussion

This dual-center retrospective cohort of 329 patients provides regional epidemiological data on ROCM in eastern Iran, including host factors, diagnostic findings, antifungal utilization, and in-hospital outcomes. The principal findings were the substantial burden of diabetes mellitus and composite immunosuppressive therapy exposure, frequent use of tissue-based diagnosis and cross-sectional imaging, predominance of amphotericin B-based treatment, and an in-hospital mortality rate of 30.4%. Because the study lacked standardized disease-severity staging, treatment-timing data, surgical debridement variables, and time-defined mortality endpoints, the findings are best interpreted as descriptive epidemiological data rather than evidence of causal determinants of outcome.

Demographic characteristics and sex distribution

Our retrospective observational study demonstrated a male predominance of 52.9%, with a median of 53.00 years [IQR: 31.00–59.00], which is consistent with, though less pronounced than, many previously published series. This finding aligns with systematic reviews reporting male predominance ranging from 63% to 78% in various mucormycosis cohorts, including a large systematic review of 10,335 patients that found 66% were male [7], and a meta-analysis of 851 cases reporting 63% male patients [14]. The observed sex distribution in our study is particularly similar to findings from Iran by Arastehfar et al., who reported 55.5% male prevalence in their 164-patient cohort [15], suggesting potential regional consistency in demographic patterns within Iran.

The mean age closely corresponds to the median age of 47 years reported in the Iranian multicenter study [15] and falls within the typical age range described in other major series [16, 17].

The slight male predominance in our cohort is consistent with several published mucormycosis series, but our retrospective data do not establish biological or occupational mechanisms for this difference. Observed differences between male and female patients should therefore be interpreted descriptively.

Predisposing conditions and comorbidity profile

Diabetes mellitus was present in 45.6% of our patients, reaffirming its status as the foremost predisposing factor for mucormycosis. This prevalence, while substantial, is notably lower than rates reported in many other series, particularly those from India where prevalence often exceeds 80–90% [7, 17]. This difference may reflect regional variations in the underlying population at risk or differences in patient selection criteria. Our finding is more consistent with Iranian studies, where Salah et al. reported diabetes in 70.3% of 64 patients [18], and Arastehfar et al. found diabetes in 46.0% of adult patients [15]. Similarly, Jafarzadeh et al. reported that all six patients in their COVID-19-associated rhino-orbital mucormycosis series from Babol, northern Iran, had diabetes mellitus as the primary predisposing factor, with corticosteroid use further exacerbating glycemic control and fungal susceptibility [19]. The risk-factor profile of our cohort should also be interpreted within the known geographic heterogeneity of mucormycosis. Diabetes mellitus is frequently the dominant predisposing condition in India, Iran, Mexico, and parts of the Middle East/North Africa, whereas hematologic malignancy, hematopoietic stem-cell transplantation, neutropenia, and other forms of profound immunosuppression contribute a larger share of cases in many European, North American, and Australian cohorts. This regional contrast is also associated with differences in anatomic presentation, with diabetes more closely linked to sinusitis and rhinocerebral disease, and hematologic malignancy or neutropenia more often associated with pulmonary mucormycosis [2, 20].

Clinical outcomes and mortality

The overall mortality rate of 30.4% in our retrospective observational study falls within the reported range for mucormycosis but is lower than many published series. This mortality rate aligns with the wide range reported in literature, which varies from 25% to 80% depending on disease extent, underlying conditions, and treatment timeliness [8]. Systematic reviews have reported mortality rates ranging from 34% to 46% [7, 11], while studies from India during the COVID-19 pandemic reported rates between 16.4% and 33.9% [16, 17].

The in-hospital mortality rate in our cohort (30.4%) lies within the lower range of published estimates, but this study cannot identify causal drivers of outcome. We did not have standardized baseline severity staging, time from symptom onset to diagnosis, time to antifungal initiation, or surgery-related variables (performance, extent, timing), and therefore cannot evaluate whether outcomes reflect earlier-stage case mix, referral patterns, differences in host factors, or specific management processes. The expanding global pool of immunocompromised hosts—owing to increased immunosuppressive therapies, higher survival of hematologic malignancy patients, and COVID-19-related immune dysfunction and corticosteroid exposure—has produced a broader and potentially more heterogeneous population of mucormycosis patients [21], which may influence comparisons across series. L-AmB was used in over half of patients, consistent with guideline-preferred therapy [10]; however, regimen-specific survival comparisons in this retrospective dataset are descriptive and subject to confounding by indication. The lower diabetes prevalence (45.6%) in our cohort compared with many Indian series may also reflect regional differences in the underlying at-risk population rather than a protective effect.

Antifungal treatment patterns

L-AmB was the most commonly used treatment modality (56.8%), followed by deoxycholate amphotericin B (28.6%). Males received deoxycholate amphotericin B more frequently (33.9% vs. 22.6%), while combination therapy with L-AmB and posaconazole was more common in females (11.0%) than in males (6.3%).

The 2019 ECMM/MSG ERC global guideline strongly recommends first-line high-dose L-AmB and strongly recommends surgical intervention when feasible. The same guideline discourages amphotericin B deoxycholate because of substantial toxicity, although it recognizes that deoxycholate may remain the only available option in resource-limited settings. Isavuconazole and posaconazole are recommended Mucorales-active azoles, particularly for salvage or continuation therapy [10].

In this study, combination antifungal therapy refers specifically to the use of two or more systemic antifungal agents, whereas combined treatment refers to antifungal therapy plus surgical source control. Azole-containing antifungal regimens were used in a minority of patients and are reported descriptively. These unadjusted regimen-specific outcomes should not be interpreted as evidence that dual or triple antifungal therapy improves survival. Current ECMM/MSG ERC guidance states that there are no definitive data to guide antifungal combination therapy in mucormycosis, and that available evidence is too limited to support more than a marginal recommendation [10]. More recent observational data similarly found no clear 6-week mortality advantage for upfront or salvage combination antifungal therapy compared with monotherapy [22]. In our cohort, apparent survival differences by regimen may reflect confounding by indication, survivor bias, drug availability, disease severity, or selection of clinically stable patients for Mucorales-active azole continuation. In this manuscript, azole step-down or continuation therapy refers specifically to Mucorales-active azoles, particularly posaconazole or isavuconazole, and not to azoles as a class [10, 23].

Surgical source control

Surgical debridement is a cornerstone of ROCM management and is strongly recommended, when feasible, alongside prompt initiation of effective antifungal therapy in the 2019 ECMM/MSG ERC global guideline [10]. In multiple case series, patients who underwent surgical debridement had substantially lower mortality than those managed with antifungals alone, and surgery has been identified as an independent predictor of favorable outcome in logistic regression models. Because surgical debridement variables (frequency, approach [endoscopic vs. open], extent [limited vs. extensive], and timing relative to antifungal initiation) were not consistently documented in the retrospective records available for abstraction, we could not report these parameters or evaluate their relationship with mortality. This omission is a major limitation that constrains the interpretability of our outcome data.

Diagnostic findings and clinical presentation

Endoscopic findings

Our endoscopic findings revealed necrosis as the most common feature (54.3%), followed by crusting (37.4%) and nasal discharge (31.6%). These findings are consistent with established literature describing black necrotic eschar as the hallmark endoscopic feature of rhino cerebral mucormycosis [8]. The presence of necrosis reflects the angioinvasive nature of Mucorales that leads to vascular thrombosis and tissue infarction [24]. Clinical guidance emphasizes that atypical dry necrotic tissue or crust formation in the nasal cavity should prompt suspicion for mucormycosis [25]. Early recognition of these endoscopic features is crucial for prompt diagnosis, as delays in treatment initiation significantly worsen outcomes.

Imaging characteristics

The imaging findings in our cohort, with mucosal thickening in 68.1% and sinus opacification in 67.8% of patients, are reported descriptively. Differences between these frequencies and those reported in other series should be interpreted cautiously because imaging indications, modalities, protocols, criteria, and reporting practices were not standardized in this retrospective study. These findings should not be interpreted as evidence of earlier-stage disease or as an explanation for mortality patterns [2, 10].

CT scanning was the most frequently utilized imaging modality (74.4%). MRI was performed in 70 cases (12.3%), predominantly for orbit and brain evaluation (94.3%), consistent with its superiority for assessing soft tissue, orbital, and intracranial extension [26].

Orbital involvement was documented in 13.7% of patients, which is lower than rates reported in some series [18]. This difference is reported descriptively and may reflect variation in case mix, referral patterns, imaging practices, documentation, or retrospective abstraction. The low rates of bone destruction (orbital 7.3%, sinus 1.5%) are consistent with literature emphasizing that Mucorales frequently spreads via angioinvasion through perivascular channels, foramina, and fissures, often traversing intact bony walls [8, 26]. Consequently, bone destruction is often a late finding on CT scans and its absence should not provide false reassurance against invasive disease.

Histopathological analysis

Angioinvasion was reported in only 15 cases (4.6%), substantially lower than rates reported in some other cohorts [27]. This low documented rate should be interpreted cautiously because angioinvasion is a focal histopathological finding whose detection depends on biopsy depth, sampling of viable tissue margins, number of sections examined, use of special fungal stains, and pathology reporting practices. The predominantly superficial endoscopic mucosal biopsies obtained during initial diagnostic workup may not have consistently captured deeper submucosal vasculature, and non-systematic application of PAS or GMS stains may have further contributed to under-ascertainment. Therefore, the 4.6% rate likely reflects documented angioinvasion in retrospective diagnostic biopsy reports rather than the true prevalence of vascular invasion, and it should not be used to explain the observed mortality rate [10].

Limitations

Several limitations should be acknowledged. The retrospective design limits our ability to establish causal relationships and may introduce selection bias, and we lacked long-term follow-up data for assessment of late complications and recurrence rates. Mortality was available as in-hospital mortality rather than standardized 6-week or 12-week all-cause mortality. This limits comparability with contemporary invasive fungal infection and mucormycosis studies, for which time-defined mortality endpoints such as 6-week and 12-week mortality are commonly used or recommended to improve cross-study comparison. Although we report in-hospital mortality and regimen-specific outcomes descriptively, we did not perform risk-adjusted modeling such as multivariable regression or propensity-based comparisons. Key determinants required for valid adjustment, including standardized baseline severity at treatment initiation, uniform documentation of disease extent and progression, and consistent time-stamped clinical milestones, were not available for all patients. The number of mortality events relative to clinically relevant predictors also raises concerns about model instability and overfitting. Antifungal regimens were selected based on clinical status, evolving practice patterns, and medication availability, introducing substantial confounding by indication. Therefore, the regimen-specific survival percentages should be interpreted as descriptive observations rather than causal estimates of treatment effectiveness.

Furthermore, although all included cases were documented as tissue-confirmed in the clinical record, we could not reliably extract the confirmation modality breakdown (histopathology vs. culture vs. direct smear) or species-level identification across the full study period, and centralized slide re-review was not performed; therefore, residual diagnostic misclassification with other invasive fungal sinusitides, including aspergillosis, cannot be completely excluded. Surgical debridement data, including performance, approach, extent, number of sessions, and timing relative to antifungal initiation, were not consistently documented for abstraction, precluding assessment of this guideline-recommended cornerstone of ROCM management. Similarly, antifungal dose (mg/kg/day), cumulative dose, total treatment duration, and step-down protocols were not uniformly available, so we report regimens by drug class and formulation only. Immunosuppressive therapy was analyzed as a composite exposure because the records did not consistently separate systemic corticosteroids, chemotherapy, transplant-related immunosuppression, biologic or other immunomodulatory therapy, and treatment-associated neutropenia. Finally, we lacked standardized metrics of disease severity at treatment initiation and time-to-diagnosis or time-to-treatment, precluding causal statements that outcomes reflect early diagnosis, standardized protocols, or other center-level processes. Observed differences in mortality across treatment regimens and between sexes should therefore be considered hypothesis-generating. Future prospective cohorts with standardized severity metrics, complete surgical documentation, dosing parameters, and timing variables are needed to enable robust risk-adjusted analyses.

Conclusion

In this large dual-center retrospective cohort of ROCM in eastern Iran, patients commonly had diabetes mellitus or composite immunosuppressive therapy exposure, and diagnosis was supported by endoscopic, imaging, and tissue-based findings. Amphotericin B-based regimens, particularly L-AmB, were the most frequently used antifungal treatments. In-hospital mortality was 30.4%, but the absence of standardized severity staging, time-to-diagnosis and time-to-treatment metrics, surgical debridement variables, antifungal dosing and duration data, and standardized 6-week or 12-week mortality endpoints limits causal interpretation and cross-study comparability. Observed differences between male and female patients and across antifungal regimens should be considered descriptive and hypothesis-generating. Prospective studies with standardized severity assessment, treatment-timing data, surgical documentation, antifungal dosing parameters, and time-defined outcomes are needed to better define determinants of ROCM outcome in the region.

Author contributions

MB, SB, and MA were responsible for data acquisition, supervision, and preparation of the final draft [Data curation], [Supervision], [Writing – original draft]. SB, KK, and SA contributed to the conceptualization, supervision of the study, and preparation of the final draft [Conceptualization], [Supervision], [Writing – original draft]. AZ, AT, HA, and AB performed data analysis, and revised the drafted manuscript [Software], [Formal Analysis], [Writing – review & editing]. KK, IR, AB, and SA prepared the original draft and assisted in data acquisition [Writing – original draft], [Data curation].All authors read and approved the final manuscript.

Funding

This research has not received any funding.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.However, data may be available upon reasonable request and with permission of the original data providers, subject to compliance with applicable confidentiality agreements and data protection laws, via contact through the Infectious Diseases Department, Mashhad University of Medical Sciences, Mashhad, Iran, at the following email address: [ketabi.kiana@yahoo.com](mailto: ketabi.kiana@yahoo.com).

Declarations

Ethics approval and consent to participate

The study was conducted in compliance with medical ethics principles and approved by the Ethics Committee of Mashhad University of Medical Sciences. Patient information was handled confidentially, analyzed anonymously. The study was approved with the ethics code IR.MUMS.MEDICAL.REC.1401.479.

Consent for publication

Not applicable. This study does not contain any individual person’s identifiable data.

Competing interests

The authors declare no competing interests.

Consent to participate

The requirement for informed consent to participate was waived by the Ethics Committee of Mashhad University of Medical Sciences due to the retrospective nature of the study and the use of anonymized patient data.

Declaration of Generative AI

During the preparation of this work, the authors used ChatGPT for fundamental grammar, spelling, and reference verification. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.