Ponikau et al. confirmed the presence of sinus eosinophilia in the majority (96%) of their patients with CRS by means of histological analysis of 101 consecutive patients.

In the same study, they also found fungal organisms, as examined on the basis of culture (96% of patients) and histology (81%), in the sinus mucus of patients with CRS, suggesting that these organisms might be involved in the disease process of CRS.

However, to their surprise, fungal organisms were also detected in the nasal mucosa of the majority of healthy control subjects. They concluded that the combination of eosinophilia and the presence of fungi explain the chronic inflammation in 96% of the patients with CRS.

As further proof of their theory, Ponikau et al. highlighted their observation that in 51 randomly selected patients given the diagnosis of CRS and treated with intranasal amphotericin B lavage, 75% experienced a significant improvement of nasal symptoms, especially nasal discharge and nasal obstruction and 36% had a polyp-free nasal endoscopy.

In those where a control CT scan was performed, they observed an improvement of the sinus opacification. The authors admit that the potential weakness of their pilot study is the fact that they did not include a placebo group.

The statement of the Ponikau group from the Mayo Clinic that the majority of the CRS cases are caused by an abnormal eosinophilic response of the patient to fungi initiated an intense controversy about the validity of the fungal hypothesis.

In 1976, Safirstein described a 24-year-old woman with allergic bronchopulmonary aspergillosis (ABPA) associated with nasal obstruction, nasal polyps, and nasal cast formation.

Millar et al. and Katzenstein et al. mentioned the histological similarity between sinus mucoid material and mucoid impaction of the bronchi in patients with ABPA, and they named it ‘‘allergic aspergillus of the maxillary sinus’’ and ‘‘allergic aspergillus sinusitis,’’ respectively.

The latter described the typical mucincontaining numerous eosinophils, sloughed respiratory cells, cellular debris, Charcot-Leyden crystals, and scattered fungal hyphae resembling Aspergillus species.

Waxman et al. described the clinical features of a young adult patient with allergic aspergillus sinusitis, showing a history of asthma and recurrent polyposis, radiographic evidence of pansinusitis, and the typical mucinous material as described by Katzenstein et al.

The majority of their patients had positive skin tests for Aspergillus (60%), 85% had IgE serum levels, and 85% had precipitins to Aspergillus. Robson et al. introduced the term ‘‘allergic fungal sinusitis’’ (AFS) after they described a case of an expansive tumor of the paranasal sinus caused by the rare fungal pathogen Bipolaris hawiiensis.

Corey et al. stressed the importance of the host’s immunological status, local tissue condition, and histopathological examination to differentiate among different forms of fungal disease. They differentiate between:

1. Allergic fungal sinusitis as the sinus counterpart of ABPA; patients showing chronic sinusitis can be atopic and show elevated IgE levels and eospinophilic counts in the peripheral blood.

2. Fungal ball or aspergilloma due to massive fungal exposure or local tissue anoxia. Patients are not immunocompromised.

3. Invasive or fulminent fungal sinusitis occurring in immunocompromised patients.

Other authors also define AFS (previously allergic aspergillus sinusitis) as a chronic sinusitis with nasal polyposis in young immunocompetent patients, showing diffuse expansive sinus disease on CT scan, with the typical allergic mucine. All their patients had positive IgE RAST to fungal antigens.

Taking into account the immune status of the patient, Bent et al. categorize fungal sinusitis into five subgroups: the role of the fungi, the presence of tissue invasion, the cause, and the affected sinus. A similar classification for fungal sinusitis was already published earlier by Ence et al.

1. Invasive fungal sinusitis is an acute fungal sinusitis affecting one sinus in an immunocompromised patient, showing tissue invasion.

2. Indolent fungal sinusitis is a subacute sinus infection with variable tissue invasion of one or more sinuses in a nonatopic immunocompetent patient.

3. Mycetoma or fungal ball is a chronic saprophytic sinusitis of one sinus without tissue invasion in a non-atopic immunocompetent patient.

4. AFS is a chronic fungal sinusitis in an immunocompetent atopic patient, where the fungus acts as an allergen involving multiple sinuses with a unilateral predominance without tissue invasion.

The patient must demonstrate the characteristic allergic mucine and have evidence of fungal etiology, either by direct observation in the surgical specimen, or by recovery of the organism in cultures of the sinus content.

5. AFS like syndrome: these patients have the same features as AFS patients, however, without the presence of fungi. Cody et al. found that 40% of these patients with allergic mucin have AFS-like syndrome.

Ferguson named this AFS-like syndrome ‘‘Eosinophilic Mucin Rhinosinusitis’’ (EMR) stating that the driving force is not a fungus but a systemic dysregulation associated with upper and lower eosinophilia.

In 1995, deShazo et al. described the criteria for the diagnosis of AFS in his study as follows:

1. Sinusitis of one or more paranasal sinuses on x-ray film.

2. Identification of allergic mucin by rhinoscopy or at the time of the sinus surgery or subsequently on histopathological evaluation of material from the sinus.

3. Documentation of fungal elements in nasal discharge or in material obtained at the time of surgery by stain or culture.

4. Absence of diabetes, previous or subsequent immunodeficiency disease, and treatment with immunosuppressive drugs.

5. Absence of invasive fungal disease at the time of diagnosis or subsequently.

From the criteria for the diagnosis of AFS listed by deShazo and Swain, for these authors absence of atopy, asthma, nasal polyps, elevated IgE levels, and serum fungal precipitins do not exclude the diagnosis of AFS.

Furthermore, bilateral involvement of the sinus on x-ray examination does not exclude the diagnosis either. On the basis of immunopathological findings in ABPA and AFS, Corey et al. concluded that both represent Gell and Coombs type I and type III response.

In AFS, IgG antibodies, in addition to elevated IgE antibodies, to the specific fungus in the serum can be demonstrated.

Therefore, they suggest the following immunological workup: total eosinophil count, total serum IgE, fungal antigen-specific IgE in vitro testing and/or skin test, fungal antigen-specific IgG (if available), and precipitating antibodies (if available).

In 1998, Manning et al. showed that AFS is an antigen, IgE-and IgG-mediated, hypersensitivity response with a late-phase eosinophilic inflammatory reaction.

On the basis of immunohistocytochemistry studying major basic protein (MBP) eosinophil-derived neurotoxin (EDN) and a neutrophils mediator (neutrophils elastase) in tissue samples of CRS, they also showed that in all cases there was evidence that MBP and EDN mediatorrelease predominated over neutrophils elastase, proving that AFS is a predominantly eosinophilic-driven disease.

In a controversial publication, Ponikau et al. reevaluated the recurrent criteria for diagnosing AFS in CRS. By using a novel method of mucous collection and fungal-culturing technique, the authors demonstrated allergic mucin in 96% of 101 consecutive surgical cases of CRS.

In the majority of their patients, they were not able to find an IgE-mediated hypersensitivity to fungal antigens. Since the presence of eosinophils in allergic mucin, and not a type I hypersensitivity, was likely the common denominator in the pathophysiology of AFS, they proposed a change of terminology from AFS to ‘‘eosinophilic fungal rhinosinusitis (EFR).’’

Similar results were found by Braun et al. Other authors had their doubts about the validity of the Mayo Clinic hypothesis. Marple questioned whether fungi are indeed ubiquitous and are present within 100% of normal noses, and wondered what separates those patients who develop AFS from the normal population.

He also questioned if the fungal screening methods used in the study were so sensitive that normal fungal colonization was mistaken for AFS, or if CRS merely represents an early form of clinically recognized AFS.

Although it is generally accepted that eosinophils play an important role in the development of both AFS and some forms of CRS, the factors that ultimately trigger eosinophilic inflammation remain in question.

Riechelmann et al. disagree with the EFR theory. They were able to show the presence of fungi only in 50% of the patients with nasal polyposis when using the most sensitive detection techniques.

Ragab et al., using the same culture technique used by Ponikau et al., were able to show positive fungal cultures in 44% of the middle meatal lavage and in 36% of the nasal cavity lavage of patients with CRS. It seems, therefore, that the rate of positive lavages is dependent of the site of collection of the sample.

The question whether fungi are present in the upper airways inducing an eventual eosinophilic response may not be relevant because the presence of these fungi can be a mere epiphenomenon of an unknown cause that initially induced the CRS.

The fungi may have not been adequately removed by the mucociliary clearance and ultimately resulted in an eosinophilic response. Novey et al. showed that a normal person inhales about 50 million spores a day.

With normal mucociliary clearance, these fungal spores are removed adequately and do not have the time to germinate and release their toxins. Once the fungi are not cleared because of an unknown cause, fungi start to colonize the sinuses and may contribute to the maintenance or amplification of the disease.

The therapeutical results with antifungal agents such as amphotericin B lavage or nasal spray do not strongly support the role of fungi in CRS, as only in 35% to 43%, respectively, of the nasal cavities become disease-free.

Bernstein et al., who are recently studying the molecular biology and immunology of nasal polyps, were unable to demonstrate that fungi play a principal role in CRS.

Their data support the hypothesis that Stapylococcus aureus releases a variety of enterotoxins (superantigens) in the nasal mucus that induce an interaction of antigen-presenting cells and lymphocytes, resulting in an up-regulation of inflammatory cells (lymphocytes and eosinophils) following an up-regulation of cytokines (TFN, IL- 1b, IL-4, and IL-5).

Bachert et al. described IgE antibodies to S. aureus enterotoxins in polyp tissue, linked to a polyclonal IgE production and aggravation of eosinophilic inflammation. A similar mechanism was described by Perez-Novo et al. in aspirin-sensitive nasal polyposis patients.

If this hypothesis proves to be true, then the classification of fungal sinusitis needs to be reconsidered and the definitions redefined. It also illustrates that the constancy of the classifications based on the hypothetical causes is not very reliable.

Finally, Ferguson described a visible growth of fungus (in AFS or EFR the fungus is not visible to the naked eye) within the nasal cavity of an asymptomatic individual and uses the term ‘‘saprophytic fungal infestation’’ for this condition.