Indian Journal of Otology

: 2019  |  Volume : 25  |  Issue : 1  |  Page : 1--5

Epidemiology and predictive factors of fungal malignant otitis externa

Rim Bechraoui, Mohamed Dhaha, Houda Chahed, Amal Ksentini, Rim Zainine, Mohamed Ben Amor, Najeh Beltaief, Ghazi Besbes 
 Department of Ear, Nose, Throat and Maxillofacial Surgery, Rabta Hospital, Tunis, Tunisia

Correspondence Address:
Dr. Mohamed Dhaha
Brazil Road, Bad Saadoun, Tunis


Context and Aims: A recent upsurge of severe fungal forms of malignant otitis externa (MOE) has been recorded. Understanding the misleading clinical aspect of the disease allows earlier diagnosis, avoiding inadequate treatments that may complicate the adapted management. Our objective was to study the epidemiological and clinical aspects of the MOE in our department and to analyze some predictive factors of the fungal form of the disease. Materials and Methods: Medical data of 100 patients treated for MOE between 2004 and 2013 were retrospectively reviewed. Epidemiological findings were recorded. Clinical aspects and presentations were studied and noted. Analysis of some predictive factors of the fungal form of the disease was done. Results: Incidence of MOE was 10 new cases every year. The average age was 67.41 years, and the sex ratio was 0.92. All patients have diabetes. Otalgia was a constant symptom. Otorrhea was noted in 63% of cases. Hearing impairment and tinnitus were noted, respectively, in 23% and 18% of cases. Prior antibiotherapy was recorded in 33% of cases for a mean period of 21 days. Stenosis of the external auditory canal was observed in 92% of cases. Granulation tissue was present in 55% of patients. Twenty patients presented peripheral facial nerve palsy. Four others presented palsies of the VI, IX, and X nerves. Forty-one patients had mycological sampling. Twenty-six among them were diagnosed with fungal MOE. Identifying fungal agent took a mean delay of 40 days. Aspergillus and Candida were incriminated each one in half of the cases. Male gender and prior quinolone administration were predictive factors of fungal MOE with, respectively, P = 0.006 and P = 0.06. After multivariate analysis, only male gender was significantly correlated to fungal growth (P = 0.032). Conclusion: Management of fungal forms is challenging due to difficult and late diagnosis. An early mycological sampling is advisable with special attention to males and patients with prior quinolone administration.

How to cite this article:
Bechraoui R, Dhaha M, Chahed H, Ksentini A, Zainine R, Amor MB, Beltaief N, Besbes G. Epidemiology and predictive factors of fungal malignant otitis externa.Indian J Otol 2019;25:1-5

How to cite this URL:
Bechraoui R, Dhaha M, Chahed H, Ksentini A, Zainine R, Amor MB, Beltaief N, Besbes G. Epidemiology and predictive factors of fungal malignant otitis externa. Indian J Otol [serial online] 2019 [cited 2019 Jul 20 ];25:1-5
Available from:

Full Text


Malignant otitis externa (MOE), also known as necrotizing otitis externa, is a severe infection of the external auditory canal (EAC) that progressively spread along the soft tissue and bone of the skull base becoming a life-threatening condition.[1] Elderly diabetic population is classically affected by the disease. More recently, many cases of MOE have been reported in other immunocompromised populations such as AIDS and malignancy.[2] Pseudomonas aeruginosa is the main pathogen incriminated. Since the 1900s, many changes in the epidemiology of the disease have been reported. Apart an increase in the incidence of the disease, an upsurge of fungal MOE has been recorded.[3] Understanding the misleading clinical aspect of the disease allows earlier diagnosis, avoiding inadequate treatments that may complicate the adapted management.[4] In fact, prior inadequate antibiotherapy has been reported to be secondarily responsible for fungal MOE.[5]

In this case series, we tried to study the epidemiological and clinical aspects of the MOE in our department during a time period from the year 2004–2013. We focused on the relatively rare fungal MOE and tried to analyze the predictive factors leading to this clinical form.

 Materials and Methods

We retrospectively reviewed the medical data of all patients treated for MOE between 2004 and 2013 in the ear, nose, and throat and maxillofacial surgery department in the Rabta Hospital, Tunis. One hundred patients' records were exploitable. Diagnosis of MOE was based on clinical history, ear examination, and imaging or scintigraphy findings. Prior auricular treatments administrated before hospitalization were noted. All patients have beneficiated of a microscope examination of the ear. Neurovestibular clinical testing consisting of a checking of all cranial nerves with special focus on the 7th and 8th nerves was performed. An evaluation of the general state and the associated comorbidities was done systematically in all cases.

All patients have beneficiated of a blood test, bacteriological and mycological sampling before intravenous (IV) treatments. Biopsies of the EAC were not systematic and were performed in cases of suspicious granulation tissue. Computed tomography (CT) scans of temporal bones were systematically performed. Scintigraphy was added when positive diagnosis was doubtful. Magnetic resonance imaging (MRI) was performed mainly if intracranial complications were suspected.

A probabilistic antibiotherapy-targeting P. aeruginosa was initially administrated in most cases and then it was adapted to the antibiogram. An antifungal treatment was administrated in cases of fungal MOE. A hyperbaric oxygen therapy was indicated as an adjuvant treatment for some patients.

Fungal cases of MOE, confirmed on ear sampling cultured for fungi and bacteria, were isolated. These patients were investigated in search of predictive factors of fungal origin. Different factors were studied, including gender, prior administration of quinolone, presence of granulation tissue, vascular thrombosis, extension to the deep facial spaces, cranial nerve palsies, and backfill of the middle ear.

Data are presented as mean or rates. The relation between qualitative variables was analyzed using the Chi-square test when possible and using Fisher's exact test otherwise. Comparison between two mean values was done using the Student's t- test. A value of 0.05 was defined as significant for all statistical tests. For statistical analyses, we used SPSS 20.0 software for Windows (SPSS, Chicago, IL, USA).


The annual incidence of MOE in our department was 10 new cases every year. From 2004 to 2013, 100 cases of MOE were recorded: 52 women and 48 men with a sex ratio equals to 0.92. The mean age in the series was 67.41 years, ranging between 28 and 87 years. About 43% of patients were older than 70 years [Figure 1]. All patients have diabetes: 11% were Type I and 89% were Type II. Sixteen among them were poorly controlled. The main other comorbidities were renal failure in five patients, idiopathic thrombocytopenic purpura in one case, and hemorrhagic rectocolitis in one other case. The other associated comorbidities are summarized in [Table 1].{Figure 1}{Table 1}

The mean delay for seeking medical consultation was 1.76 months ranging between 15 days and 7 months.

Otalgia was reported in all patients. It started 15 days to 5 months before the first consultation (average delay = 2.5 months). It was persistent despite analgesic treatment in 95% of cases. Otorrhea was recorded in 63% of cases. Hearing impairment and tinnitus were noted, respectively, in 23% and 18% of cases. Two patients reported vertigo, twenty presented with facial nerve palsy, two reported swallowing disorder, and four lockjaws. Six patients reported intermittent fever. Loss of body condition, manifesting in anorexia and asthenia, was observed in two patients.

A trigging factor was identified in only six cases. It was essentially after bathing, extraction of ear wax or ear intervention.

Thirty-three percentage of patients had a prior antibiotherapy before hospitalization for a mean period of time of 21 days, ranging between 7 and 45 days. Ciprofloxacin was administrated in 80% of these patients. It was associated to local fluoroquinolone instillation treatment in six cases.

Clinical examination showed stenosis of the EAC in 92% of cases and otorrhea in 62% of cases. Granulation tissue was observed in 55% of patients. It was emerging from the posterior and inferior wall of the auditory canal in 58% and 52% of cases. The tympanic membrane was seen in only 31 cases. It was congestive in 15 cases, dull in 13 cases, and normal in 3 cases.

Facial nerve palsy was objectified in 20 cases, it was the most cranial nerve concerned. Apart from the facial nerve, four other patients had cranial nerve palsies. One patient had insolated trigeminal nerve palsy. A second had a simultaneous impairment of VI, IX, and X cranial nerve manifesting with swallowing disorder, hoarseness, diplopia, and convergent ipsilateral strabismus. Another patient showed X and IX nerve palsies. Hypoglossal nerve was impaired in one case.

Temporal bones CT showed evidence of MOE in 90% of cases. In the remaining cases, a technetium-99-based scintigraphy confirmed the diagnosis. The disease was limited to the tympanic bone in 46% of cases. Twelve patients needed MRI. This examination was required in 10 cases to evaluate complicated disease extended to the deep facial spaces and the brain. In the two other cases, MRI was requested for suspicion of malignant process. A backfill of the middle ear was observed in 90% of cases. The deep facial spaces were infiltrated in 41%. Concerned spaces were as follows: lateral pharyngeal space in 35% of cases, infratemporal fossa in 14%, the parotid gland in 13%, and retropharyngeal space in 3%. The rhinopharynx was involved in two cases.

Vascular thrombosis was noted in 12 patients. Internal jugular vein was the most concerned. It was observed in 37.5% of cases. Lateral and sigmoid sinuses were concerned, respectively, in 29.1% and 25% of cases. Cavernous sinus and internal carotid artery were concerned each one in 4.2% of cases.

All patients have beneficiated of bacteriological sampling since the 1st day of hospitalization before any antibiotherapy. Forty-one patients had mycological sampling. Two among them had mycological studies on biopsies from granulation tissue. A third one had, in addition, a mycological study on bone sequester.

These samplings were either systematically done since the 1st day of hospitalization (38 patients, 93% of cases), either motivated by the absence of improvement under IV antibiotherapy (3 patients, 7% of cases). The mean delay for practicing mycological sampling was 5.3 days ranging between 1 and 51 days.

Twenty-six patients were diagnosed with fungal MOE. Positivity rate was 63.4% on otorrhea sampling and 100% on granulation tissue biopsy studies and bone sequester. Identifying fungal agent took a mean delay of 40 days ranging between 5 and 90 days.

Thirteen (50%) of fungal MOE were Aspergillus infections and 13 (50%) were candidal [Figure 2].{Figure 2}

Aspergillus serology was performed in 16 cases. It was positive in 4 patients (25%), 3 among them had positive fungi culture growth (Aspergillus flavus in two cases and Aspergillus fumigatus in one case).

Aspergillus serology was negative in 12 cases. About 33.3% of them had positive culture growth of A. flavus. Aspergillus antigenemia was performed in 11 patients. It was positive in only 2 patients (18.1%) having, respectively, A. fumigatus and A. flavus infections.

Incidence of fungal MOE showed tendency toward increasing [Figure 3].{Figure 3}

Nearly 38.3% of males and 13.5% of females had fungal MOE. Prior quinolone antibiotherapy was noted in 40% among these patients. Nearly 22.2% of granulation tissue, 23.3% of middle ear backfill, 27.3% of vascular thrombosis cases, 38.5% of cases presenting infiltrations of deep facial spaces, and 20% of cranial nerves palsies were in relations with fungal MOE.

Statistical correlations between the previous variables (gender, prior quinolone antibiotherapy, presence of granulation tissue, backfill of the middle ear, vascular thrombosis, facial deep spaces infiltrations, and cranial nerves palsies) showed, in a univariate analysis, that only male gender and prior quinolone-based antibiotherapy were associated to fungal MOE with a P = 0.006 and 0.06, respectively [Table 2]a and [Table 2]b.{Table 2}

After multivariate analysis, only male gender was statistically significant (P = 0.032).


Since it was first described by Chandler in 1968, the incidence of MOE seems to be on the rise. In the English literature, a recent retrospective study done by Chawdhary et al. showed a trend toward increasing of hospitalization rates for MOE between the years 1999 and 2013.[6] An increase in aging and diabetic population has been suggested by these authors as an explanation to this growing problem.[6] Besides, the improvement in the index of suspicion of the disease among generalist physicians has been also proposed to justify these changes in the epidemiology of the disease since 1990s.[2] However, some recent studies found a stable hospitalization rate for MOE, despite some year-to-year variability.[7]

MOE usually affects elderly population above 55 years old.[3],[8],[9] However, cases of MOE in younger adults and pediatric population have been reported. Sylvester et al., in a large cohort, counting 8300 cases of MOE, stratified patients according to age in three groups: the pediatric group (<18 years) representing 4.9%, adults (between 18 and 64) representing 62.4%, and the elderly (>64 years) representing 32.7%.[7] Speaking of sex ratio, most of the studies reported a predominance of the male gender.[4],[9],[10] It was not the case in our series, considering that the number of females exceeds slightly the number of males which is in accordance with the results found by Sylvester et al.[7] Diabetes mellitus is usually reported as the main predisposing factors to MOE estimated between 55% and 95% of cases.[2],[10],[11],[12]

About 26% of our patients had fungal MOE. Diagnosis was confirmed either on otorrhea sampling culture or on histological findings after biopsies of granulation tissue or bone sequester. This rate is considered moderate. Hamzany et al. reported a rate of 15% of fungal MOE.[12] For Glikson, this rate was relatively high reaching 35%.[10] Otorrhea sampling is contributing to diagnosis of fungal origin.[13] However, most authors recommend performing biopsies on granulation tissues and bone sequester.[3],[11],[14] Concerning patients with negative culture, fungal origin should be suspected when there is no improvement under antibiotherapy.[3],[11],[14] The delayed diagnosis represents another important issue as an extended time is usually needed before any proof of fungal origin.[12] The mean delay was 77 days in the series of Hamzany et al., in our study, it was 40 days. Therefore, fungal MOE is often managed as bacterial MOE for long time before adapted treatment.

Aspergillus was involved in 50% of infection. Candida was involved in all remaining cases. In the literature, Aspergillus is the main fungal agent reported notably A. fumigatus, Aspergillus niger, and A. flavus.[3],[14],[15],[16] Aspergillus serology, although reported as contributive,[17] was positive in only 4 cases (25%). Three among them had positive culture of fungi growth.

About 38.3% of males and 13.5% of females had fungal MOE. Male gender was associated to fungal origin in univariate and multivariate analysis. In fact, there was a predominance of male gender in the series of Hamzany et al. with 39 males and 21 females.[12] As well, in a literature review speaking of outcome predictors of treatment effectiveness for fungal MOE by Mion et al., the sex ratio was 19/6.[18] However, no statistical correlation analysis between male gender and fungal origin was performed in the literature to the best of our knowledge.

About 33% of patients had prior antibiotherapy before hospitalization. Ciprofloxacin was administrated in 80% of all cases and 40% of fungal MOE group of patients. Six among them received in addition ofloxacin or ciprofloxacin ear drops. In this study, these prior treatments were correlated to secondary fungal overgrowth in a univariate analysis (P = 0.06). In fact, this is a well-recognized complication of broad-spectrum antibacterial agents. Schrader reported in, a pediatric population, an association between prior ciprofloxacin ear drops administration and development of fungal overgrowth.[5] In fact, the incidence of fungal MOE has risen when the use of quinolone ear drops became a common clinical practice.[19] Besides, according to Bovo et al., instillation of antipseudomonal topical agents only increases the difficulty of isolating the pathogenic organism from the ear canal.[11]

A higher number of facial nerve palsy in patients with fungal MOE was reported: 75% of patients with MOE due to Aspergillus had facial nerve palsy versus only 24% of patients with MOE due to Pseudomonas.[12],[20] Besides, it seems that intracranial complications are more observed with fungal MOE as reported by Ress et al.[21] However, no studies concerning vascular thrombosis and fungal MOE were performed to the best of our knowledge. This attests of the invasive behavior of the fungal agents. In this study, none of these factors was identified as predictive of fungal origin.

Middle ear backfill was observed in 23.3% in the fungal MOE group. According to Mani et al., fungal MOE is more characterized by the participation of middle ear and mastoid cells, but no statistical correlation was found.


MOE is a growing health problem. Management of fungal forms of MOE is a real challenge because of the difficult and late diagnosis. Patients are usually seen in complication's stage. This study showed many clinical and epidemiological aspects of the disease in Tunisia. Male gender and prior quinolone administration were identified as predictive factors of fungal origin. Thus, an exhaustive and early investigation for fungi overgrowth should be undertaken in this category of patient.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Chandler JR. Malignant external otitis. Laryngoscope 1968;78:1257-94.
2Rubin Grandis J, Branstetter BF 4th, Yu VL. The changing face of malignant (necrotising) external otitis: Clinical, radiological, and anatomic correlations. Lancet Infect Dis 2004;4:34-9.
3Mani R, Belcadhi M, Krifa N, Abdelkefi M, Ben Said M, Bouzouita K, et al. Fungal necrotizing external otitis. Ann Otolaryngol Chir Cervicofac 2008;125:40-5.
4Guevara N, Mahdyoun P, Pulcini C, Raffaelli C, Gahide I, Castillo L, et al. Initial management of necrotizing external otitis: Errors to avoid. Eur Ann Otorhinolaryngol Head Neck Dis 2013;130:115-21.
5Schrader N, Isaacson G. Fungal otitis externa – Its association with fluoroquinolone eardrops. Pediatrics 2003;111:1123.
6Chawdhary G, Liow N, Democratis J, Whiteside O. Necrotising (malignant) otitis externa in the UK: A growing problem. Review of five cases and analysis of national hospital episode statistics trends. J Laryngol Otol 2015;129:600-3.
7Sylvester MJ, Sanghvi S, Patel VM, Eloy JA, Ying YM. Malignant otitis externa hospitalizations: Analysis of patient characteristics. Laryngoscope 2017;127:2328-36.
8Facon F, Braccini F, Ayache S, Laurent P, Layoun W, Pancrazi J, et al. Otite externe maligne: Approche diagnostique et therapeutique. J Fr Oto Laryngol (1977) 2000;49:270-5.
9Martel J, Guyot M, Darrouzet V. Otites Externes “Malignes” ou n…Crosantes Progressives; 2018.
10Glikson E, Sagiv D, Wolf M, Shapira Y. Necrotizing otitis externa: Diagnosis, treatment, and outcome in a case series. Diagn Microbiol Infect Dis 2017;87:74-8.
11Bovo R, Benatti A, Ciorba A, Libanore M, Borrelli M, Martini A, et al. Pseudomonas and Aspergillus interaction in malignant external otitis: Risk of treatment failure. Acta Otorhinolaryngol Ital 2012;32:416-9.
12Hamzany Y, Soudry E, Preis M, Hadar T, Hilly O, Bishara J, et al. Fungal malignant external otitis. J Infect 2011;62:226-31.
13Chabasse D, Contet-Audonneau N. Examen direct et place de l'histologie en mycologie. Rev Fr Laboratoires 2003;2003:49-54.
14Carfrae MJ, Kesser BW. Malignant otitis externa. Otolaryngol Clin North Am 2008;41:537-49, viii-ix.
15Bellini C, Antonini P, Ermanni S, Dolina M, Passega E, Bernasconi E, et al. Malignant otitis externa due to Aspergillus niger. Scand J Infect Dis 2003;35:284-8.
16Hollis S, Evans K. Management of malignant (necrotising) otitis externa. J Laryngol Otol 2011;125:1212-7.
17Persat F. Aspergillus serology, from yesterday to today for tomorrow. J Mycol Méd 2012;22:72-82.
18Mion M, Bovo R, Marchese-Ragona R, Martini A. Outcome predictors of treatment effectiveness for fungal malignant external otitis: A systematic review. Acta Otorhinolaryngol Ital 2015;35:307-13.
19Martin TJ, Kerschner JE, Flanary VA. Fungal causes of otitis externa and tympanostomy tube otorrhea. Int J Pediatr Otorhinolaryngol 2005;69:1503-8.
20Halsey C, Lumley H, Luckit J. Necrotising external otitis caused by Aspergillus wentii: A case report. Mycoses 2011;54:e211-3.
21Ress BD, Luntz M, Telischi FF, Balkany TJ, Whiteman ML. Necrotizing external otitis in patients with AIDS. Laryngoscope 1997;107:456-60.