Article Text

Invasive fungal disease of the sinus and orbit: a comparison between mucormycosis and Aspergillus
  1. Danielle Trief1,2,
  2. Stacey T Gray1,3,
  3. Frederick A Jakobiec1,2,4,
  4. Marlene L Durand1,5,
  5. Aaron Fay1,2,
  6. Suzanne K Freitag1,2,
  7. N Grace Lee1,2,
  8. Daniel R Lefebvre1,2,
  9. Eric Holbrook1,3,
  10. Benjamin Bleier1,3,
  11. Peter Sadow1,3,
  12. Alia Rashid1,2,4,
  13. Nipun Chhabra1,3,
  14. Michael K Yoon1,2
  1. 1Massachusetts Eye and Ear Infirmary, Boston, Massachusetts, USA
  2. 2Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA
  3. 3Department of Otolaryngology, Harvard Medical School, Boston, Massachusetts, USA
  4. 4David G. Cogan Laboratory of Ophthalmic Pathology, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts, USA
  5. 5Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts, USA
  1. Correspondence to Dr Michael K Yoon, Ophthalmic Plastic Surgery, 10th Floor, Massachusetts Eye and Ear Infirmary, 243 Charles Street, Boston, MA 02114, USA; Michael_Yoon{at}


Background/aims Invasive fungal infections of the head and neck are rare life-threatening infections where prompt diagnosis and intervention is critical for survival. The aim of this study is to determine the clinical characteristics and outcomes of invasive fungal disease of the sinus and orbit, and to compare mucormycosis and Aspergillus infection.

Methods A retrospective review was conducted from a single tertiary care eye and ear hospital over 20 years (1994–2014). Twenty-four patients with a confirmed pathological diagnosis of invasive fungal disease of the sinus and/or orbit were identified and their medical records were reviewed. The main outcome measures were type of fungus, location of disease, mortality and visual outcome.

Results Patients with orbital involvement had a higher mortality and higher likelihood of mucormycosis infection compared with those with sinus-only disease (78.6% vs 20%, p=0.01; 86% vs 30%, p=0.01, respectively). Patients with mucormycosis had a higher mortality (71%) than patients with Aspergillus (29%); however, this was not statistically significant (p=0.16). All patients with orbital involvement and/or mucormycosis infections were immunosuppressed or had inadequately controlled diabetes, and had a cranial neuropathy or ocular motility dysfunction. All five post-transplant patients with orbital infections died, while the two transplant patients with sinus infections survived.

Conclusions Patients with orbital fungal infections are more likely to be infected with mucormycosis compared with Aspergillus and have a higher mortality compared with infections sparing the orbit. History of transplant portends a dismal prognosis in orbital infections. Invasive fungal disease should be considered in any immunocompromised patient presenting with a new cranial neuropathy or ocular motility abnormality.

  • Infection
  • Orbit

Statistics from


Invasive fungal sinusitis (IFS) is a severe and often life-threatening infection. Affected patients almost universally have systemic immunosuppression or alteration, which is the primary risk factor for this infection. Even with rapid diagnosis and initiation of treatment, prognosis is poor with mortality ranging from 21% to 80%.1–5 Haematological malignancy, diabetes, immunosuppressive therapy after solid organ and bone marrow transplantation, AIDS, haemochromatosis and long-term corticosteroid use are all risk factors for IFS.5–7 Cases affecting patients without any risk factors for immunocompromise are decidedly rare.8–10

The rarity of these infections may prevent early recognition by clinicians. Presenting symptoms may initially be subtle, often resembling acute bacterial infections of the orbit or sinuses, non-infectious orbital inflammatory conditions (eg, idiopathic orbital inflammation) or facial nerve palsy.11 Misdiagnosis can cause delay in treatment or incorrect treatment with corticosteroids, which may worsen the infection.7 ,12

Treatment is directed at limiting the progression of infection with a combination of surgical debridement, antifungal therapy and reversal of immunocompromising factors, when possible. However, given the rarity of this infection, the optimal treatment regimen has not been clearly defined.13 Current clinical recommendations are based on small case series and retrospective reviews.14 ,15

While invasive Aspergillosis and mucormycosis infections are often grouped together under the umbrella of IFS, studies have found very different clinical outcomes between these two fungal infections.1 Given the limited nature of published reports, the purpose of this study is to review our institution's experience treating patients with IFS. Differences between the clinical course for patients with invasive mucormycosis and aspergillosis and the differences between outcomes for patients with orbital and sinus involvement were specifically analysed.

Patients and methods

A retrospective review of patients with a diagnosis of invasive fungal disease of the orbit and sinonasal cavity treated by the Ophthalmology and Otolaryngology services of Massachusetts Eye and Ear Infirmary from January 1994 through February 2014 was conducted.

For inclusion in this study, patients required a histopathological diagnosis of invasive fungal disease, defined as histopathological evidence of angioinvasion or invasion of surrounding soft tissue (figure 1). Data collected included past medical history, with an emphasis on immunocompromising factors such as malignancy or diabetes, age, gender, presenting signs and symptoms (including cranial neuropathies and central retinal artery occlusion (CRAO)), medical and surgical treatments, time from symptom onset to diagnosis and time to death where applicable. The outcome of a patient who was discharged on hospice care was counted as a mortality.

Figure 1

Histopathological photomicrographs of fungal infections. (Top) Mucormycosis (zygomycosis). The hyphae are readily seen in the wall of an orbital vessel in a H&E stained section and are non-septate, wide (7–30 μm), and often branch at 90° angles. The inset displays the density of the matted hyphal elements (H&E, ×100, inset ×200). (Bottom) Aspergillus. Hyphae of Aspergillus are narrow (2–4 μm) and septate. They branch at acute, often 45° (insert) angles (Grocott's methenamine silver, ×200, inset ×400).

Infections were analysed based on their causative fungal organism and the presence or absence of orbital involvement, defined by clinical findings (ie, the presence of ophthalmoplegia and/or proptosis) and supported by radiographic evidence of orbital invasion. One patient with concomitant mucormycosis and aspergillosis, one patient with dematiaceous without further speciation and one patient infected with an indeterminate fungal species were excluded from the subanalysis of the fungal genera. For indeterminate fungal species, histopathological, immunohistochemical and genetic testing could not definitively determine the fungal species. Fisher's exact tests were used to calculate statistical significance between the groups.


Twenty-four patients with invasive fungal sinus with or without orbital disease were identified (see online supplementary table). There were 16 men and 8 women ranging in age from 30 to 80 years (mean 55). Fourteen patients had disease affecting the orbit while 10 spared the orbit. The fungal aetiology was an agent of mucormycosis in 14 patients (58%), Aspergillus in 7 (29%), both mucormycosis and Aspergillus in 1 (4%), dematiaceous fungus in 1 and indeterminate fungus in 1. In total, 13 patients (54%) had a fatal outcome, 11 of whom had orbital disease. Of the patients who died, 10 had a pathological diagnosis of mucormycosis, 2 had Aspergillus and 1 had a fungal infection of indeterminate species. The majority (11 of 14) of the orbital infections were secondary to mucormycosis, and the majority of orbit sparing sinus infections (6 of 10) were secondary to Aspergillus.

Risk factors

While 13 of the patients (54%) carried a diagnosis of diabetes, 8 of these 13 also had medication-related immunosuppression (chemotherapy or post-transplant antirejection medications). Of the five patients who had diabetes without other risk factors, three survived. All five patients with diabetes as their only risk factor had poorly controlled diabetes. Three of the five had a recent haemoglobin A1c >9.5. The other two patients, in whom a haemoglobin A1c was not recorded, had multiple episodes of diabetic ketoacidosis or finger stick readings of 400 mg/dL. There were seven patients who had a history of solid organ or stem cell transplant. Five had orbital infections (all mucormycosis), and they all died from their infection. Four of the transplant recipients had no other risk factors, and three of the four died. The one transplant patient without other risk factors who survived did not have orbital disease. Chronic intranasal steroid use for chronic sinusitis was the only apparent risk factor for two patients; both had invasive sinus disease without orbital involvement and both survived. Nine of the 24 patients had a haematological malignancy or dysplasia. Seven carried a diagnosis of leukaemia, one patient had multiple myeloma and one patient had myelodysplastic syndrome. No other active malignancies were present.

Signs and symptoms

All patients with orbital involvement (14) had an optic neuropathy, cranial neuropathy or ocular motility dysfunction. Twelve (85.7%) reported decreased vision, and six ultimately declined to no light perception. Twelve patients (85.7%) had motility deficits, and 10 (71.4%) had proptosis on presentation. Notably, all patients with visual loss also had ophthalmoplegia and/or proptosis. Six patients had a documented CRAO, all of whom had a diagnosis of mucormycosis (one had both mucormycosis and Aspergillus), and five of whom (83%) died.

Of the 10 patients with orbit sparing disease, 8 (80%) reported sinus pain or pressure. One patient with AIDS was admitted for persistent vomiting and developed periorbital swelling during her hospital course. Another patient was admitted for febrile neutropenia in the setting of leukaemia and developed facial swelling, a facial droop and nose bleeds during his admission.

The average time from symptom onset to diagnosis was 10.3 days for patients with orbital involvement and an indeterminate duration for patients with orbit sparing disease since many of these patients had symptoms from chronic sinus disease for years.


Of the 24 patients, 23 were initially treated with antibiotics for presumed bacterial sinusitis and/or orbital cellulitis. Five patients were treated with steroids, two of whom were treated for presumed giant cell arteritis (GCA).

Also, 20 of the 24 patients (83.3%) received liposomal amphotericin B (AmBisome, Astellas Pharma, Northbrook, Illinois), including all patients with a diagnosis of mucormycosis. In patients where a diagnosis of invasive Aspergillus was made, amphotericin was switched to voriconazole after this agent was demonstrated to be superior. The mean number of surgeries was 2.5 for patients with orbital involvement and 2 for patients with orbit sparing disease. All patients had endoscopic sinus debridement except one patient who elected for palliative care after a bedside intranasal biopsy was diagnostic for mucormycosis. Four patients underwent orbital exenteration, three of whom had a fatal outcome.

Four patients with orbital mucormycosis received hyperbaric oxygen treatments. Three of these four patients died.

Mucormycosis vs Aspergillus

Patients with mucormycosis had a higher mortality (71.4%) than patients with Aspergillus (28.5%); however, this was not statistically significant (p=0.16) (table 1). All patients with mucormycosis had systemic immunosuppression or uncontrolled diabetes, whereas one of the patients with angioinvasive aspergillosis had chronic sinus disease without other risk factors. A significantly higher number of patients with mucormycosis had orbital invasion compared with Aspergillus (78.6% vs 14.2%, respectively, p=0.02).

Table 1

Characteristics of patients with mucormycosis compared with Aspergillus

Orbital disease vs orbit sparing (sinus-only) disease

Patients with orbital infection had a higher mortality compared with those with orbit sparing disease (78.6% vs 20%, respectively, p=0.01) (table 2). Patients with orbital involvement were more likely to be infected by mucormycosis alone than orbit sparing disease (79 vs 30%, respectively, p=0.01). Conversely, patients with orbit sparing disease were more likely to be infected by Aspergillus alone (7% vs 60%, respectively, p<0.01).

Table 2

Characteristics of patients with orbital and orbit sparing disease


This 20-year retrospective case series represents one of the largest series of invasive fungal sinus and sino-orbital disease from a single institution. In total, 14 of our 24 patients had orbital disease, and these patients had a higher mortality compared with orbit sparing disease (78.6% vs 20%, respectively, p=0.01). The vast majority of patients with orbital disease were infected by mucormycosis alone (79%) while the majority of patients with orbit sparing disease were infected by Aspergillus (60%).

There are several hypotheses why patients with orbital disease has a poorer outcome than those with orbit sparing disease. Mucormycosis is known to cause a more aggressive fungal infection,1 and in the current study, there was a higher proportion of patients with orbital disease infected with it. However, mucormycosis infection compared with other fungi did not lead to a statistically higher mortality rate in our study. A second explanation is that once the fungus enters the orbit it can easily access the intracranial space through the ophthalmic artery, optic canal or superior orbital fissure.6 A third theory is that the fungal infections that have demonstrated a propensity to spread beyond the confines of the sinuses and into surrounding spaces may be more inherently virulent.7 ,16

Thirteen of the patients in the current study (54.1%) had a diagnosis of diabetes mellitus. Five had uncontrolled diabetes without other concomitant immunocompromising conditions. Uncontrolled diabetics are known to have poor humoral immunity and neutrophilic dysfunction.17 An environment of low oxygen tension, hyperglycaemia and ketosis provides an excellent medium for the fungus to thrive.6 Diabetics also have poor iron metabolism.16 Iron overload is a risk factor for mucormycosis infection and patients with iron overload (haemochromatosis) or who are treated with iron chelating agents such as deferoxamine are at higher risk of developing invasive fungal infections.18 ,19 Patients with diabetes as their only risk factor fared better than patients with other immunocompromising conditions (40% mortality). This finding may reflect the potential reversibility of diabetic immune dysfunction.4 ,14

In contrast, 75% of patients who were organ transplant recipients without other risk factors died from the disease. Transplant patients had the worst prognosis. This likely reflects the severe and sometimes irreversible immunosuppression in these patients.20

All patients with mucormycosis were immunosuppressed or had diabetes. There are few case reports of invasive mucormycosis in immunocompetent hosts: these infections were secondary to trauma21 or high exposure load.9 In contrast, there were two patients in our study with Aspergillus whose only apparent risk factor for IFS was chronic intranasal steroid use. While invasive aspergillosis is most commonly seen in immunosuppressed individuals, there have been reports of infection in immunocompetent individuals22 and these individuals tend to have much higher survival rates.8 ,10 Additionally, compared with immunosuppressed patients, immunocompetent patients with IFS tend to have more chronic or indolent presentations. In one study of 35 immunocompetent patients with invasive orbital Aspergillus, symptoms persisted for an average of 1 year prior to diagnosis.10 Our two immunocompetent patients with invasive Aspergillus both had a diagnosis of chronic sinusitis and reported sinus pain for months prior to diagnosis of fungal infection. The immunocompromised or uncontrolled diabetics had a much more rapid disease course.

Haematological malignancy or dysplasia was seen in 9 of the 24 patients (5 of whom had mucormycosis infections). Although haematological malignancies are classically associated with Aspergillus infections, they may be associated with mucormycosis as well.2 Consistent with previous studies, the current study found a male predominance.16 The clinical significance of this is unknown.

In total, 23 of the 24 patients were initially treated for bacterial sinusitis. As sinus symptoms progressed despite broad-spectrum antibiotics, patients were reimaged and sinus biopsies were taken, leading to the diagnosis of invasive fungal infection. In contrast, all patients with orbital disease had an optic neuropathy, trigeminal hypesthesia or ocular motility dysfunction. Also, 5 of the 14 orbital cases were initially given high-dose steroids for presumed GCA, idiopathic orbital inflammation or presumed bacterial sinusitis with concurrent antibiotics. Given the rare nature of IFS and the fact that orbital fungal infections can have similar presentations to non-infectious inflammatory conditions, misdiagnosis is quite commonly reported in the literature.23 However, delay in diagnosis or treatment of these potentially deadly infections with steroids can lead to reduced survival.12 Clinicians should maintain a high suspicion for fungal infections in immunocompromised or diabetic patients who present with a new cranial neuropathy and orbital findings.

Interestingly, 6 of the 14 orbital cases had evidence of retinal ischaemia on initial examination (CRAO), and all of these patients had mucormycosis infections. This points to the angioinvasive nature of mucormycosis. If a CRAO is seen in an immunosuppressed patient, IFS, specifically mucormycosis, should be considered.

Of the 24 patients, 23 underwent endoscopic sinus debridement. Surgical debridement (endoscopic or open sinus surgery and/or orbitofacial debridement) of IFS has been shown to lead to increased survival.4 Antifungal agents take time to be effective. The advantages of surgical debridement are multifaceted: (1) removal of devitalised tissue; (2) debulking of infected tissue, theoretically allowing for faster action of antifungals against a smaller burden of fungi; and (3) provision of specimens for culture and pathology to confirm aetiology.5

Only four patients in this study underwent exenteration, three of whom succumbed to the infection. All four of these patients were infected with mucormycosis. Exenteration of the orbit is aimed at local control prior to spread of fungus intracranially, where intravascular invasion can be lethal. The role of exenteration in orbital fungal disease has been contested. It is a disfiguring procedure, although radical debridement of affected tissue may be life saving. While some studies have failed to show a statistical benefit,4 another found a benefit in patients who presented with fever.14 We advocate that the decision of exenteration should be made by the multidisciplinary team caring for the individual patient.

All patients with mucormycosis were treated with liposomal amphotericin B. Doses ranged from 5 to 10 mg/kg. Liposomal amphotericin appears to be superior to amphotericin B deoxycholate in the treatment of mucormycosis.18 We recommend that if there is suspicion for IFS, systemic amphotericin be administered empirically (prior to tissue diagnosis) in consultation with infectious disease specialists.

The majority of patients with Aspergillus without mucor were treated with voriconazole. In a randomised clinical trial of patients infected with Aspergillus, initial therapy with voriconazole led to better response and improved survival compared with amphotericin B.24 Patients were placed on liposomal amphotericin and then switched to voriconazole if the pathological diagnosis or culture revealed Aspergillus.

Several case reports and series have reported survival benefits of treatment with hyperbaric oxygen.7 ,25 Hyperbaric oxygen is thought to improve flow to ischaemic tissue and alleviate acidosis. The complications of hyperbaric oxygen include pneumothorax, seizures, nausea, tinnitus and visual abnormalities.13 In the present study, four patients with orbital mucormycosis received hyperbaric oxygen treatments, and three of the four died. It is difficult to determine whether this was a failure in treatment or selection bias for patients with worse disease. Future studies are needed to clearly define the role of hyperbaric oxygen in invasive fungal disease.

The mortality rate for both mucormycosis and Aspergillus continues to be high. The treatment regimen has remained similar, that is, debridement (endoscopic and/or open) with antifungal medication, although pharmacological developments (liposomal amphotericin and voriconazole) have been made. Studies have routinely found survival benefit when fungal infections are identified early. Early diagnosis is often challenging as patients may present with subtle and non-specific findings. However, as demonstrated by our study, careful consideration of fungal infection should be given to any immunosuppressed patient presenting with a new optic neuropathy, cranial neuropathy or ocular motility dysfunction. High-dose steroid treatment must be reserved until fungal disease is ruled out. Transplant and systemically immunosuppressed patients are especially vulnerable to IFS, and this infection should be considered in this population, along with diabetics, when this clinical scenario arises.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors All authors made substantial contributions to the conception of the work or to the acquisition/analysis/interpretation of the data. All authors gave final approval of the version published and agree to be accountable for all aspects of the work.

  • Competing interests None declared.

  • Ethics approval Massachusetts Eye and Ear Infirmary Institutional Review Board. The study was HIPAA compliant and adhered to the tenets of the Declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost Jonas