Implications for clinical outcomes

Type 1 hypersensitivity is believed to be one pathophysiologic component of the inflammation in AFRS. Multiple randomized, placebo-controlled trials have demonstrated the efficacy of allergen-specific immunotherapy to reduce symptoms and medication requirements in allergic rhinitis and asthma. However, there is a lack of randomized, controlled trials for the treatment of AFRS with immunotherapy. The peer reviewed literature contains low-level evidence that suggests immunotherapy may reduce inflammation and reduce steroid requirements in AFRS patients. This evidence derives from case reports, small cohort studies without comparator groups, and a single cohort study with an external control group. Immunotherapy should be considered for AFRS patients with significant allergic symptomatology, steroid dependence, or multiple polyp recurrences.

Allergic fungal rhinosinusitis

AFRS is a well recognized label for a form of polypoid chronic rhinosinusitis (CRS) that is characterized by type 1 hypersensitivity, the accumulation of dense inspissated “eosinophilic mucin,” and trapped fungi within the sinuses. AFRS was described as a distinct pathologic entity that resembled allergic bronchopulmonary aspergillosis (ABPA) with thick, dark, inspissated mucus filling the paranasal sinuses similar, grossly and microscopically, to that seen in the bronchial passages. The nature of this condition, its diagnosis and appropriate treatment continue to generate controversy despite 30 years of investigation.

The previously accepted hallmark for the diagnosis of AFRS was the characteristic sinus contents first responsible for the description of the disease. Grossly, this “allergic mucin” or eosinophilic mucin is thick, tenacious, and darkly colored. Viewed histologically, this mucus consists of onion-skin laminations of necrotic and degranulating eosinophils in a background of mucin, occasional small hexagonal crystals of lysophospholipase (Charcot-Leyden crystals), as well as fungal hyphae. Aspergillus and the dematiaceous fungi are most commonly found in AFRS mucus. Although once called “allergic” Aspergillus sinusitis, the preferred terminology for this condition is now “allergic fungal rhinosinusitis” or AFRS. However, investigators have noted that in some cases, the eosinophilic mucin evacuated from the sinuses does not have identifiable fungal elements. Additionally, Ferguson described an AFRS-like condition with slightly different clinical features and proposed the term “eosinophilic mucin rhinosinusitis” (EMRS) to describe cases in which fungus was not identified histologically. Some patients with clinical features of AFRS may have demonstrable fungus within their eosinophilic mucin, yet do not have allergy. Others have eosinophilic mucin yet no allergy and no evidence of fungi within the mucin. The report of Ponikau and colleagues suggesting that most, if not all, CRS was a hypersensitivity response to fungi and that fungi could be universally cultured from nasal secretions also further clouded the distinction between AFRS and AFRS-like CRS. AFRS is probably overdiagnosed because of clinical similarity to other forms of “eosinophilic mucin chronic rhinosinusitis” (EMCRS). The overlapping clinicopathologic features of chronic polypoid rhinosinusitis with eosinophilic mucin have led to questions about the underlying pathophysiology of AFRS. If patients with the clinical features of AFRS do not have allergy and/or do not have fungus in their eosinophilic mucin, how should these patients be classified? Is fungus the extrinsic trigger for inflammation? Is allergy an important component of the pathogenesis of AFRS? Currently it is safe to state that the clinicopathologic distinction of AFRS from other forms of EMRS requires further investigation. Questions regarding the proper diagnosis, classification, and pathogenesis of AFRS have yet to be resolved and have important implications for treatment.

Allergic fungal rhinosinusitis

AFRS is a well recognized label for a form of polypoid chronic rhinosinusitis (CRS) that is characterized by type 1 hypersensitivity, the accumulation of dense inspissated “eosinophilic mucin,” and trapped fungi within the sinuses. AFRS was described as a distinct pathologic entity that resembled allergic bronchopulmonary aspergillosis (ABPA) with thick, dark, inspissated mucus filling the paranasal sinuses similar, grossly and microscopically, to that seen in the bronchial passages. The nature of this condition, its diagnosis and appropriate treatment continue to generate controversy despite 30 years of investigation.

The previously accepted hallmark for the diagnosis of AFRS was the characteristic sinus contents first responsible for the description of the disease. Grossly, this “allergic mucin” or eosinophilic mucin is thick, tenacious, and darkly colored. Viewed histologically, this mucus consists of onion-skin laminations of necrotic and degranulating eosinophils in a background of mucin, occasional small hexagonal crystals of lysophospholipase (Charcot-Leyden crystals), as well as fungal hyphae. Aspergillus and the dematiaceous fungi are most commonly found in AFRS mucus. Although once called “allergic” Aspergillus sinusitis, the preferred terminology for this condition is now “allergic fungal rhinosinusitis” or AFRS. However, investigators have noted that in some cases, the eosinophilic mucin evacuated from the sinuses does not have identifiable fungal elements. Additionally, Ferguson described an AFRS-like condition with slightly different clinical features and proposed the term “eosinophilic mucin rhinosinusitis” (EMRS) to describe cases in which fungus was not identified histologically. Some patients with clinical features of AFRS may have demonstrable fungus within their eosinophilic mucin, yet do not have allergy. Others have eosinophilic mucin yet no allergy and no evidence of fungi within the mucin. The report of Ponikau and colleagues suggesting that most, if not all, CRS was a hypersensitivity response to fungi and that fungi could be universally cultured from nasal secretions also further clouded the distinction between AFRS and AFRS-like CRS. AFRS is probably overdiagnosed because of clinical similarity to other forms of “eosinophilic mucin chronic rhinosinusitis” (EMCRS). The overlapping clinicopathologic features of chronic polypoid rhinosinusitis with eosinophilic mucin have led to questions about the underlying pathophysiology of AFRS. If patients with the clinical features of AFRS do not have allergy and/or do not have fungus in their eosinophilic mucin, how should these patients be classified? Is fungus the extrinsic trigger for inflammation? Is allergy an important component of the pathogenesis of AFRS? Currently it is safe to state that the clinicopathologic distinction of AFRS from other forms of EMRS requires further investigation. Questions regarding the proper diagnosis, classification, and pathogenesis of AFRS have yet to be resolved and have important implications for treatment.

Pathophysiology

A hypersensitivity to fungus is believed to underlie the pathogenesis of AFRS, but the nature of this hypersensitivity is still debated. The dominant theory to explain the pathogenesis of AFRS was adopted from the model of ABPA pathogenesis : in short, a combination of Gell and Coombs type 1 and type 3 hypersensitivity to fungal allergens causes sinonasal inflammation. This inflammation leads to polyp formation and mucus accumulation in the sinuses, thus explaining the clinical findings in the disease. This paradigm was reinforced by the clinical association of AFRS with allergy and the detection of elevated serum levels of total and fungal antigen-specific IgE and IgG in AFRS patients. Work by Manning and Holman and Stewart and Hunsaker lent early support to the theory that connected AFRS pathogenesis to ABPA. Manning and Holman analyzed serum from patients with culture-confirmed AFRS ( Bipolaris ) and nonsinusitis controls and found that 82% of AFRS cases had Bipolaris -specific IgE by radioallergosorbent test inhibition and 94% had Bipolaris -specific IgG by ELISA inhibition. Stewart and Hunsaker analyzed fungal-specific serum IgE and IgG levels in nonatopic controls, allergic rhinitis patients, non-AFRS polyp patients, AFRS-like patients, and AFRS patients. Patients in the AFRS and AFRS-like group had elevated serum levels of IgE and IgG to multiple fungi. Compared with those patients with only nasal polyps, the AFRS and AFRS-like groups had significantly elevated levels of fungus-specific IgG and IgE. More recent studies have also examined humoral immune responses in patients with EMCRS including AFRS cases. Pant and colleagues found that serum fungal-specific IgG was increased in these patients regardless of the presence of allergy or fungal elements in their eosinophilic mucin. Patients with AFRS had increases in fungal-specific IgE and total IgE but these were no different from a control group with allergic rhinitis. Pant and colleagues found a poor correlation between fungal species present in the eosinophilic mucin of AFRS patients and the specific fungal allergy (42%). However, elevated fungal-specific IgG3 appeared to be a distinguishing serologic feature that separated EMCRS and AFRS patients from those with fungal allergic rhinitis and other forms of CRS. Serum IgE levels could be used to distinguish EMCRS from AFRS. Finally, Luong and colleagues found that peripheral blood mononuclear cells from AFRS patients are stimulated by fungal antigens to secrete TH2-type cytokines. These more recent studies continue to support a role for humoral immune factors in AFRS and EMCRS, but do not resolve the controversy about AFRS pathogenesis.

Type 1 hypersensitivity and/or dysregulation of IgE metabolism is evident locally within AFRS tissue. Most patients with AFRS have detectable fungal specific IgE in their eosinophilic mucin, whereas most patients with other forms of EMCRS do not. Compared with non-AFRS sinus tissue (from nonallergic subjects) there is a local mucosal elevation in total and specific IgE to fungal and nonfungal antigens. These findings again suggest a role for IgE and type 1 hypersensitivity in the pathophysiology of the disease.

However, some argue that, because EMCRS and AFRS are so similar, allergy to fungus cannot be the primary pathophysiologic force driving the inflammation in AFRS. Studies comparing AFRS to other forms of EMCRS have found similar polyp histopathology, inflammatory cell infiltrate, tissue eosinophilia, and fungal-specific PBMC proliferation. However, circumstantial evidence suggests that both allergic and nonallergic fungal hypersensitivity may be important components of the underlying pathophysiology of both AFRS and EMCRS.

Immunologic mechanisms alone may not account for the development of the clinical features of AFRS. It has been suggested that AFRS develops in susceptible patients with a convergence of local anatomic as well as environmental factors. This may explain the propensity of AFRS to cause unilateral disease. The working theory has been that fungi entering the nose and sinuses and triggers an inflammatory response. This inflammation induces polyp formation and the accumulation of allergic mucin. Trapped fungi continue to stimulate the immune system in a vicious cycle. Over time, massive polyposis develops and mucoceles distort the sinonasal anatomy. This model is likely an oversimplification that does not take into account many possible pathophysiologic mechanisms that are poorly understood and likely to be complex.

In summary, although AFRS and EMCRS are phenotypically quite similar, a clinically important distinguishing feature of AFRS is type 1 hypersensitivity. Therefore, type 1 hypersensitivity to fungal antigens, as assessed by specific allergy tests, helps to distinguish AFRS from other forms of EMCRS and has implications for treatment.

Epidemiology and microbiology

Allergic fungal sinusitis may be the most common form of fungal sinusitis. Allergic fungal sinusitis accounts for about 7% to 12% of cases of CRS taken to surgery in the United States, though this may be an overestimation. Climactic factors that determine fungal growth and exposures may be responsible for the varied incidence in certain regions. In the United States, the highest incidence is in the south and along the Mississippi basin. In a large pathology series of CRS cases from India, AFRS accounted for 24% of cases. So clearly, the disease has a worldwide distribution, though there may be differences in the microbiology of the disease across continents.

AFRS develops primarily in young adults and adolescents. Older patients with the clinical features of AFRS may be more likely have some other EMCRS syndrome. Besides being younger than other CRS patients, AFRS patients may be more likely to come from disadvantaged groups with lower socioeconomic status. Affected patients are immunocompetent and have a history of atopy. By definition, AFRS patients have allergy that should be evident by skin or in vitro testing, but only about two-thirds of patients give a history of allergic rhinitis.

Aspergillus was initially believed to be the causative organism in AFRS, but further experience with cases in the United States showed that the dematiaceous fungi were most commonly found in AFRS mucus, The terminology for this condition subsequently changed from “allergic Aspergillus sinusitis” to “allergic fungal sinusitis.” In series of AFRS and nonallergic eosinophilic fungal sinusitis from other parts of the world, Aspergillus is still found to be a common isolate. The specific fungal organism has not been shown to be an important or predictive clinical characteristic, but the identification of fungus in eosinophilic mucin either via histopathology or culture is still considered to be important to make the diagnosis of AFRS.

Clinical presentation

Symptoms of AFRS are similar to other forms of polypoid CRS. Patients with AFRS usually present with rhinosinusitis symptoms lasting months or years and they may not seek medical attention until complete nasal obstruction, headaches, visual disturbances, or facial distortion develop. Symptoms are frequently unilateral. Patients may report dark, thick nasal mucus drainage. A history of previous allergic rhinitis symptoms is common, but is not always present.

The physical examination findings in AFRS often reflect the advanced nature of disease at presentation. Proptosis or telecanthus are frequently seen at presentation, especially in younger patients. Intranasal examination will reveal polyps that are unilateral or bilateral, and it is common for the bulk of polyp disease to be asymmetric. On nasal endoscopy, inspissated yellowish or brown mucus may be visualized among the polyps.

Testing is important to establish evidence of atopy, as demonstration of type 1 hypersensitivity is required for diagnosis. This may be accomplished with skin testing or in vitro testing for antigen-specific IgE. In addition to fungal antigens, patients should be tested against a region-specific panel of seasonal and perennial allergens. Possible laboratory abnormalities in AFRS patients include peripheral eosinophilia and dramatically elevated total IgE levels. Skin testing or in vitro IgE testing will usually demonstrate IgE-mediated hypersensitivity to multiple fungal and nonfungal antigens.

Diagnostic criteria

The diagnosis of AFRS requires a combination of clinical, radiographic, microbiologic, and histopathologic information. Therefore, the diagnosis of AFRS cannot be made reliably until after surgical intervention. There is still no universally recognized set of diagnostic criteria for AFRS.

A variety of diagnostic criteria for AFRS have been proposed by various investigators and these criteria have been further refined by a recent consensus conference on definitions of rhinosinusitis. The classic and still widely accepted diagnostic criteria for AFRS were described by Bent and Kuhn, who proposed the following: type 1 hypersensitivity, nasal polyposis, characteristic CT scan findings, eosinophilic mucus without fungal invasion into sinus tissue, and a positive fungal stain of sinus contents removed at surgery. In the absence of better defined immunologic parameters to distinguish AFRS from other forms of EMCRS, the Bent and Kuhn criteria are still important.

An important criterion is the presence of eosinophilic mucin. Grossly, eosinophilic mucin is thick, tenacious, and darkly colored; it may appear similar to a fungus ball but microscopically the two are quite different. Microscopically, eosinophilic mucin consists of onion-skin laminations of necrotic and degranulating eosinophils in a background of mucin with occasional Charcot-Leyden crystals. Fungal hyphae are present but scarce, and special fungal stains may be needed for identification. Fungal hyphae do not invade tissue: the presence of fungal tissue invasion has been considered incompatible with a diagnosis if AFRS. However, there are reports of cases with clinicopathologic features of both AFRS and chronic invasive fungal sinusitis. Such cases highlight the difficulty of creating rigid diagnostic categories for fungal sinusitis, and suggest that there is a spectrum of overlapping disease features in fungal sinusitis. In AFRS, adjacent mucosa and polyps demonstrate a prominent eosinophilic inflammatory infiltrate. Many patients with polypoid CRS and eosinophilic mucin lack other important clinical characteristics of AFRS: demonstrable fungi and fungal allergy. These patients should not be classified as having AFRS.

As discussed above, there is significant phenotypic similarity between AFRS and EMCRS. Saravanan and colleagues found considerable clinical overlap in AFRS and EMCRS groups, but AFRS subjects were more likely to have bony erosion, heterogeneous opacity, and sinus expansion on CT scan. These findings are similar to those of Dhiwakar and colleagues who point out that the combination of nasal polyps, CT scan hyperattenuation, and elevated titers of anti- Aspergillus IgE have high predictive value for AFRS, though considered in isolation they are not specific. Clearly, considerable overlap exists between AFRS, EMCRS, and CRS from other causes, and the Bent and Kuhn criteria are still helpful to distinguish between these.