express less proapoptotic genes (e.g., BLC2L13 (mitochondrion-­localized protein with conserved B-cell lymphoma-2 homology motifs), CASP2 (caspase-2), CARD4 (caspase recruitment domain family member-4)) involved in nuclear factor-kappa B (NF-κB) regulation [8]. Aggression of respiratory epithelial cells following exposure to the triggering factors (see below) could induce production of cytokines, such as interleukin (IL)-25. Indeed, IL-25 is elevated in EGPA patients’ blood, where it could drive CD4+ T lymphocytes toward a skewed T-helper cell type-2 (Th2) phenotype, thereby maintaining eosinophil proliferation via cytokines synthesized by Th2 cells [9]. High levels of eosinophilic cytotoxic proteins (e.g., eosinophil cationic protein, eosinophil peroxidase), directly involved in damaging tissues, are also found in EGPA patients’ sera, urine, and tissues [10].

As in asthma patients, EGPA patients’ CD4+ T lymphocytes, mostly Th2 phenotype, are activated. EGPA patients’ sera and bronchoalveolar lavage (BAL) fuid also contain elevated concentrations of Th2 ILs (e.g., IL-4, IL-5, and IL-13) [11]. IL-5 plays a central role in eosinophil survival and maturation, and strong associations have been found between IL-5 expression and clinical parameters of EGPA activity (Birmingham Vasculitis Activity Score, eosinophilia). Moreover, EGPA pathogenesis also seems to implicate Th1 and Th17 responses [12].

Although the precise role of B lymphocytes in EGPA has not yet been fully elucidated, their involvement in EGPA pathogenesis is being more thoroughly investigated, in light of the ef cacy of anti-CD20 monoclonal antibodies against GPA and MPA [13, 14]. Higher percentages of CD80+, CD27+, or CD95+ B cells and lower rates of CD19+ B cells have been found in patients with frequently relapsing EGPA [15]. Furthermore, immunoglobulin (Ig)G4 levels, an indirect surrogate of B-lymphocyte activation, were elevated during EGPA fares [16]. In addition, ANCA positivity, found in one-third of EGPA patients, is also indicative of B-lymphocyte activation, and direct EGPA pathogenesis via anti-MPO-ANCAs was demonstrated in vitro and in mouse models [17, 18].

Epidemiology

Incidence and Prevalence

EGPA is a rare disease, the prevalence of which ranges between 7.3 and 17.8/million inhabitants [19, 20], with an annual incidence of 0.9–2.4/million inhabitants [21, 22].

EGPA prevalence seems to be similar in Europe, Japan, and the United States [20]. The mean age at diagnosis is ~50 years [23], which is lower than that for MPA, GPA, or polyarteritis nodosa (PAN). Childhood-onset EGPA is extremely rare [24]. No male/female sexual predominance has been reported.

Triggering Factors

Because of the asthma prodrome, allergens have been suspected of being EGPA triggers. However, when frequent allergens were systematically tested, less than a third of EGPA patients had documented allergies [25]. Thus, although potentially responsible molecules may still not have been identi ed, allergens alone do not seem to be responsible for the onset of EGPA manifestations.

Although no common EGPA-triggering event or cause has yet been identi ed, many environmental factors implicated in its pathogenesis have been described, e.g., inhaled antigens (dark diesel fumes, grain dust, cereal dust, dust), desensitization, or vaccinations [26, 27]. However, these exposures may merely be disease accelerators in patients, with undiagnosed EGPA already present, rather than its true cause.

Some medications, mainly leukotriene receptor antagonists (e.g., montelukast) or, more recently, anti-IgE antibodies (e.g., omalizumab), are still considered potential triggers, but because they are often used to taper glucocorticoid doses, their role remains unclear [28–32]. In a study on 79 EGPA patients, montelukast use was signi cantly associated with the risk of developing EGPA within 3 months (OR 4.5 (95% CI 1.5–13.9)) [29]. However, that association was not speci c, as other drugs used to treat asthma, e.g., long-lasting β2-agonists and oral glucocorticoids, were also associated with EGPA with comparable risk ratios. These associations may be confounded by indication due to gradually worsening asthma.

Clinical Manifestations

EGPA generally evolves in three stages: rst with asthma or sometimes symptoms attributed to allergy, then eosinophilia and lung in ltrates, and, nally, systemic vasculitis manifestations. The time between the rst manifestations and overt EGPA is usually around 9 years, but all three stages can simultaneously appear. Its main clinical manifestations are described in Table 7.1.

General Symptoms

Constitutional symptoms, like fever, asthenia, or weight loss, are common before systemic manifestations, and, as for patients in our EGPA series at diagnosis, 38.9% were febrile with myalgias, 29.8% had arthralgias, and almost 50% had lost weight (mean ± standard deviation (SD) loss during the preceding 3 months, 7.8 ± 4.7 kg) [23].

Pulmonary Manifestations

Asthma is the main EGPA characteristic, affecting 91–100% of the patients, most often before systemic vasculitis symptoms (mean ± SD interval 9.3 ± 10.8 years). It is a criterion in the American College of Rheumatology (ACR) classi cation criteria (Box 7.1). Asthma is often late-onset, appearing around 30–40 years of age, but can start during childhood. Asthma is generally severe and glucocorticoid-dependent, and its severity tends to increase 3–6 months before the systemic disease becomes apparent. In a retrospective study on 157 EGPA patients, asthma was mild, moderate, or severe in 17%, 26%, or 57%, respectively [38]. All patients required

high-dose, inhaled glucocorticoids and bronchodilators. The mean ± SD post-bronchodilator FEV1 (forced expiratory volume in 1 s) was 69.7 ± 24.9% of the predicted value, and post-bronchodilator FEV1/FVC (forced vital capacity) was <0.7 for 54% of tested patients.

Box 7.1 1990 American College of Rheumatology criteria for the classifcation of Churg–Strauss syndrome (Adapted from Masi et al. [2])

For classi cation purposes, a patient with vasculitis is said to have EGPA, when at least four of these six criteria are present. The presence of any four or more criteria yields 85% sensitivity and 99.7% speci city.

Compared to MPA, the alveolar hemorrhage frequency attributable to pulmonary vasculitis is lower (4%), but patients can sometimes experience massive hemoptysis (often associated with renal involvement). However, in most patients, alveolar hemorrhage is usually mild, with only bloody sputum or diagnosed by BAL. Eosinophilic, exudative, pleural effusions have been described, with rare pleural biopsies showing signs of vasculitis and in ltrated by eosinophils [39]. Lung in ltrates are common (38.6%) and can be unior bilateral, transitory and often disappear after a few days of glucocorticoids [40]. Phrenic paralysis has been described but is extremely rare [41].

Rarely, EGPA can be lung-limited and present without any systemic extra-respiratory organ involvement [42]. The diagnosis might be challenging, and a lung biopsy is needed to evidence eosinophilic, granulomatous, and necrotizing vasculitis.

Ear, Nose, and Throat (ENT) Manifestations

Asthma can be associated with other ENT symptoms. Maxillary sinusitis is common in EGPA and is one of the ACR classi cation criteria (Box 7.1). Allergic rhinitis and nasal polyposis are found in 17% and 41% of EGPA patients, respectively, but must be distinguished from the ENT manifestations of GPA, which are often destructive and severe. Rhinitis and polyposis are usually present in EGPA patients before the vasculitis symptoms become evident.

Although these ENT manifestations are considered EGPA symptoms, they could represent an underlying predisposition to the vasculitis, rendering patients more susceptible to developing vasculitis symptoms. Once a vasculitis fare has been successfully treated, most patients have polyps and symptoms at some time during the evolution of the disease, which supports the hypothesis advanced above of a predisposition.

Neurological Manifestations

Peripheral neuropathies are frequent (affecting 46–75% of EGPA patients), mainly mononeuritis multiplex (46% in our series) [23]. Therein, the most frequently involved nerves, in decreasing order of frequency, are the common peroneal, internal popliteal, then those of the arms (radial, cubital, and/ or median), and, nally, cranial nerves. Patients may complain of paresthesia or sometimes painful hyperesthesia before sensory and motor de ciencies become apparent, when mild, only super cial sensitivity is affected. According to the Five-Factor Score (FFS) [42], peripheral nerve involvement is not a poor prognosis factor and does not infuence mortality, but motor or sensory de ciency sequelae can worsen the functional prognosis because recovery is long and unpredictable, mostly for sensory symptoms.

Central nervous system involvement, indicating cerebral vasculitis, is uncommon (5.2% in our series [23]). Its clinical manifestations are nonspeci c: seizures, hemiplegia, brain, or subarachnoid hemorrhages. Although computed tomography (CT) scans can visualize hemorrhagic or some ischemic events, magnetic resonance imaging (MRI) can detect T2-weighted cortical and subcortical signals, more speci c to cerebral vasculitis. Since the FFS was revised in 2011, central nervous system involvement is no longer considered to be factor of poor prognosis [42].

Skin Manifestations

Small-vessel vasculitis or extravascular granulomas can lead to cutaneous involvement, which is common and affects 40–70% of patients in our EGPA series. Vascular purpura, preferentially affecting the lower limbs, is the most frequent skin sign, affecting 22–50% [23]. Subcutaneous nodules, often bilateral, symmetrical, red, and predominantly affecting the ngers, elbows, and upper limbs, are found in 10–30% of EGPA patients. When nodule biopsies are obtained, they can contain extravascular granuloma(s), which are nonspeci c for EGPA, as they also occur in other vasculitides or infammatory bowel diseases. Skin symptoms can also include Raynaud’s phenomenon, urticaria, livedo reticularis (3.9%), and gangrenous necrotic lesions.

Cardiac Manifestations

Heart involvement is one of the most severe EGPA manifestations and is the primary cause of death of EGPA patients. It is also an FFS prognostic item [42]. The pathophysiology of these cardiac manifestations includes coronary vasculitis, extravascular granuloma(s), and eosinophilic interstitial