Cryptococcal infections in general are rare. They are most commonly seen in the setting of immunosuppression with disseminated disease or occasionally as localized disease, namely meningitis, in the immunocompetent. Laryngeal cryptococcosis is a rare condition with only 20 previous cases reported. Here, we report two further cases of cryptococcal infection localized to the larynx. Both occurred in association with high-dose inhaled corticosteroid use. Laryngeal cryptococcosis, despite its rarity, is important as it may easily be confused with laryngeal malignancy, both macroscopically and histologically. These cases provide further evidence of inhaled corticosteroid local side effects and should strengthen the clinician’s resolve to use inhaled corticosteroids at their lowest effective dose.

Cryptococcus neoformans is an encapsulated yeast that lives as a saprobe in nature. While other types of Cryptococcus exist, it is only C. neoformans that is considered pathologically significant. It tends to be found in association with certain trees such as eucalypts and rotting wood. It is also consistently isolated from soil contaminated by guano from birds, especially pigeons, chickens, and turkeys. The route of infections remains somewhat unclear, but aerosolised particles that infect and disseminate after alveolar deposition in the lungs is thought to be the most likely portal of entry (1). At diagnosis, however, most cases do not have a defined epidemiological exposure or chest X-ray changes (1).

Given Cryptococcus’ relative ubiquity as an organism and rarity as a disease, it is thought that host factors, play an important role in the development of symptomatic disease. Between 15 and 30% of HIV/AIDS patients in sub-Saharan Africa will contract cryptococcal meningitis during the course of their illness (2). Other predisposing conditions for systemic disease include organ transplantation, lymphoproliferative disorders, hematological malignancy, systemic corticosteroid treatment, cirrhosis, and sarcoidosis (even without steroid treatment). An association with diabetes is not definitely known. A deficiency in cellular immunity seems to be the most important risk factor for cryptococcal disease (1). That said, one-third of patients lack any obvious immune deficit (1).

In immunosuppressed patients the pattern of disease tends to be disseminated with multiple sites being involved at the time of diagnosis, most commonly the lung and meninges, though the disease can involve practically any site. Similarly, mycotic infections of the larynx tend to arise, in immunocompromised hosts and also tend to occur as a result of dissemination of the organism, or at least contiguous spread, as is the case with laryngeal candidiasis, which usually coexists with oral candidiasis (3). That said, there are increasing reports of mycotic laryngitis with a normal immune status (Candida, aspergillosis, coccidiomycosis, histoplasmosis) (4).

The previously reported and current cases of laryngeal cryptococcosis are summarized in Table 1 with baseline characteristics listed in Table 2.

Of the previously reported cases, only three had possible cryptococcal exposure. One patient had exposure to chicken manure (5), another had exposure to pigeons (6) and bird droppings prior to presentation (7). Several patients had a history of immunosuppresion: four with significant systemic immunosuppression in the form of HIV/AIDS (8–11); three patients with diabetes mellitus (12–14); and one patient was steroid-dependent and on immunosuppressant medication for rheumatoid arthritis (15). Importantly, 8 of the 20 cases (i.e., 40%), were using inhaled corticosteroids (7, 9, 16–20). Two of these patients were also on systemic steroids as a confounder for obstructive airways disease (16, 17). Our cases describe a further two immunocompetent patients with isolated laryngeal cryptococcosis who were exposed to inhaled corticosteroids. These cases further suggest that inhaled corticosteroids, particularly in high doses, are a potential risk factor for developing isolated laryngeal cryptococcosis in immunocompetent patients.

In immunocompetent patients with isolated laryngeal disease, the role of localized immunosuppression and disruption of the laryngeal mucosal barrier is important. Factors described as causing such a disruption include previous radiotherapy, gastro-esophageal reflux, traumatic intubation, smoking, and inhaled corticosteroids (3). Five of the reported patients had a history of smoking (9, 13, 14, 18, 21). Previous case reports have emphasized local immunosuppression from inhaled high-dose corticosteroids as a risk factor for laryngeal cryptococcal disease (9, 18–20). Our cases have similarities to those reported elsewhere in the literature with the exception of a stronger link to the use of high-dose inhaled corticosteroids.

Inhaled corticosteroid likely creates a localized immunosuppression within the larynx which along with its observed local irritation effect (22) causes an impairment of the epithelial barrier and allows the ubiquitous Cryptococcus to colonize and infect via direct inoculation. Studies suggest that up to 90% of inhaled drug is deposited in the upper airway of patients using inhalant devices, and it is this deposition that accounts for the local side effects of inhaled corticosteroids (23). Interestingly, descriptions of similar mycotic infections of the lungs, in immunocompetent patients are rare.

All of the reported patients using high-dose inhaled corticosteroids, including our two cases, presented with sub-acute onset dysphonia and a dry cough. Dysphonia is reported in up to 50% of patients using inhaled corticosteroids. The pathogenesis of steroid inhaler-related dysphonia is variable, but commonly caused by local irritation in the larynx from deposited inhalant. This irritation is seen across drug and propellant types (22), though dose and frequency seem to have an impact (22, 24). Laryngeal candidiasis can cause dysphonia or present as altered taste and is seen in around 5–10% of inhaled corticosteroid users (22, 25, 26). Adductor palsies of the vocal cords are another cause of dysphonia and thought to develop due to localized myopathy due to laryngeal corticosteroid deposition (22, 25, 26). Fluticasone is implicated as the most common cause of local steroid side effects such as dysphonia and oral candidiasis. Although this may merely reflect its widespread use (22, 24, 27–29), it is the most potent of the inhaled corticosteroids with the greatest receptor affinity and longest half-life (24, 27, 30). One study reported more than double the rate of oral candidiasis in those taking fluticasone compared with beclomethasone (29).

The differential diagnosis for the macroscopic appearance of the laryngeal cryptococcosis includes other fungal infections and malignancy. This is especially true in patients presenting with painless progressive dysphonia. This differential is echoed histologically where the typical pathological appearance of pseudoepitheliomatous hyperplasia of the squamous mucosa, and granulomatous inflammation in the submucosal region is seen in a number of conditions, including other mycotic infections (due to Candida, Histoplasma, Blastomyces, Coccidioides, or Paracoccidiodes species) (3), granular cell tumor, and early squamous cell carcinoma. This is further complicated by the fact that there is often a low yield of yeast cells in the biopsy, and these may be missed if not specifically looked for. This problem is compounded by the rarity of laryngeal cryptococcosis as practitioners may fail to consider the condition.

Treatment was varied among the published cases. Most cases in the literature were medically managed with oral antifungal agents; however, three cases had surgical excisions (12) and three had a combination of both (7, 15, 17). Fifteen cases received a minimum of 6–8 weeks of oral fluconazole treatment. One patient refused treatment following initial surgical excisional biopsy (31). One patient received a month of intravenous amphotericin (2 g total) (5). One HIV/AIDS patient received 4 days of intravenous amphotericin followed by maintenance fluconazole (8). Another with HIV had amphotericin for 3 weeks then oral fluconazole for another 6 months (11). Interestingly, two immunocompetent patients achieved cure with excisional biopsy alone (12, 21). Of the non-HIV patients who were treated, all remained disease free on prolonged follow-up. Most had taken weeks to months for their voices to normalize.

Importantly, we believe risk factor modification (by means of dose reduction in inhaled corticosteroid) is a useful adjunct to medical management, especially in patients taking high doses. On diagnosis, patients should be commenced on antifungal treatment without delay.

Review of the literature has shown that while 8 of the 20 published cases described inhaled corticosteroid as a potential risk factor, only two cases included some description of the dose (16, 18). Of these eight cases, only two described reduction or cessation of inhaled steroid as part of their management. Certainly, the remainder of the cases that did not describe any dose reduction still had symptomatic or pathological resolution after appropriate treatment. However, the lack of dose description makes it difficult to infer that the inhaled steroids were causing local laryngeal immunosuppression in all cases.

At present, there is uncertainty surrounding the evidence for dose-reduction alone in the setting of high-dose inhaled steroids in immunocompetent hosts with cryptococcal laryngitis. The decision for dose reduction should be made on a case-by-case basis, based on the severity of the patient’s pulmonary status. A change to ciclesonide, a relatively new inhaled corticosteroid could also be of potential benefit in treating laryngeal cryptococcosis. It is an inactive pro-drug, with a smaller particle size, which is converted to an active metabolite in the airways, with potentially fewer upper airway adverse effects. A recent systematic review comparing ciclesonide to other inhaled corticosteroids in the treatment of asthma in children, however, found no difference in adverse effects (25). Both our cases were inhaling doses exceeding the maximum recommended dose. Dose reduction in addition to oral fluconazole led to successful resolution.

Laryngeal cryptococcosis, despite its rarity, is an important condition to diagnose, and as these two additional cases demonstrate, can occur in immunocompetent patients with possible local immunosuppression due to high dose inhaled corticosteroids. With correct diagnosis, risk factor modification (including inhaled steroid reduction), and oral antifungal treatment, it is a curable condition. It is especially important to be aware of the condition as a differential to early malignancy and other causes of inhaled steroid-related dysphonia.

All Authors were involved in the write-up of the manuscript, review, and editing. Dr Stanley contributed the clinical details of the two cases for discussion.

The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication.