Churg-Strauss syndrome (CSS) is a multisystem vasculitis of unknown etiology first described in 1951 using autopsy specimens from patients with a history of asthma, tissue eosinophilia, and systemic vasculitis (1). The majority of cases are idiopathic; however, recent reports have implicated leukotriene receptor antagonists (LTRA) and inhaled corticosteroids in the development of this rare condition (2-8). We present a patient with no prior history of allergic asthma who acutely developed a skin-limited ChurgStraus-like granulomatous vasculitis after initiating therapy with inhaled fluticasone and salmeterol (Advair Diskus).
A 59-year-old former coal miner with a history of smoking-related chronic obstructive pulmonary disease, silicosis, atherosclerotic vascular disease, and trauma-related peripheral neuropathy presented to his family physician in October 2001 with a five-day history of a progressive erythematous bullous-appearing eruption on his hands. This was associated with eye discomfort and intermittent chills. Two weeks prior to the development of these symptoms, he was started on inhaled fluticasone/salmeterol (Advair Diskus) for wheezing and occasional exertional dyspnea related to his silicosis and 35 pack-year smoking history. He had no past history of atopic disease, recurrent sinusitis, or chronic corticosteroid use. A presumptive diagnosis of herpes simplex infection was made, and the patient was placed on valacyclovir. Over the next two weeks, he continued to develop new lesions on his hands and was admitted to the hospital. Laboratory studies demonstrated a normal WBC count with no peripheral eosinophilia. Blood chemistries, ANA, and serum protein electrophoresis were negative. He remained afebrile during his admission, and blood and viral cultures of the blister fluid were negative. His lungs were clear on physical exam; therefore, a chest x-ray was not performed. A skin biopsy of the hand read by a local pathologist revealed "noncaseating microgranulomas." Since he continued to develop new lesions despite therapy with intravenous acyclovir and cefazolin, arrangements were made for an outpatient evaluation at our clinic following his discharge.
He was initially evaluated in our clinic within days of discharge from the hospital. Examination of both the palmar and dorsal aspects of his hands and fingers revealed multiple edematous-appearing violaceous nodules, some of which were ulcerated (Figs. 1, 2). There were also a few similar but smaller lesions on the forehead, back, and sclera of his right eye.
[FIGURES 1-2 OMITTED]
A biopsy was taken from one of the nodular lesions on his hand. On histopathologic exam, the overlying epidermis was completely necrotic but retained its overall configuration. A wedge of degenerated connective tissue extended to approximately the mid-dermal level. There was a perivascular and interstitial infiltrate extending from the superficial dermis into the subcutis, comprised of lymphocytes, macrophages, and numerous eosinophils (Fig. 3). Infiltration of vessel walls by these cells was identified. Several small extravascular granulomas were identified in the deep dermis and subcutis (Fig. 4). Direct immunofluorescence from perilesional skin showed positive deposits of IgG, IgM, C3, and fibrin in dermal and subcutaneous vessels. These findings were considered consistent with those encountered in cutaneous lesions of Churg-Strauss syndrome.
[FIGURES 3-4 OMITTED]
After reviewing the biopsy results, we recalled a recent case report by Termeer et al. (9,10) describing a patient with CSS associated with the use of a similar fluticasone/salmeterol inhaler. Suspecting a possible connection between these two cases, our patient's inhaler was discontinued and he was placed on a three-week prednisone taper (60mg-40mg-20mg). He stopped developing new lesions, and his skin was almost completely healed after one month. He is no longer using the fluticasone/salmeterol inhaler, and his skin has remained clear for over one year since his original eruption.
Also known as allergic granulomatous vasculitis, Churg-Strauss syndrome is diagnosed using criteria published in 1990 by the American College of Rheumatology (ACR) (11). Using this widely accepted standard, a patient must present with at least four of the following six features: moderate to severe asthma, > 10% peripheral blood eosinophilia, neuropathy, pulmonary infiltrates, paranasal sinus abnormality, and extravascular eosinophil infiltration on tissue biopsy (11). Two thirds of patients with CSS present with skin lesions which are classified into three categories: erythematous maculopapules resembling erythema multiforme, hemorrhagic lesions, and cutaneous and subcutaneous nodules (12).
The classical histopathologic findings include necrotizing vasculitis, dermal and subcutaneous infiltration by numerous eosinophils, and extravascular granulomas. These microscopic features are not unique, as similar findings can be observed in various disorders such as Wegener's granulomatosis, polyarteritis nodosa, lupus erythematosus, rheumatoid arthritis, and Takayasu's aortitis (13). Unfortunately, multiple names have been used to label this reaction pattern, including 'cutaneous extravascular necrotizing granuloma', 'Winkelmann granuloma', or more recently, 'palisaded neutrophilic and granulomatous dermatitis of collagen vascular disease' (14).
The incidence of CSS has been increasing in recent years following the introduction of LTRAs (5). Wechsler et al. in 1988 first described eight asthma patients treated with zafirlukast who developed CSS when their oral corticosteroids were tapered (6). Since then, similar cases have been reported in patients receiving both zafirlukast and the LTRAs montelukast and pranlukast (5,7,8,15). All of these cases were described in asthmatic patients who were treated with either oral or inhaled corticosteroids prior to taking the LTRA. It is controversial whether these cases represent an idiosyncratic drug reaction to the LTRA, or--as is more likely--an unmasking of pre-existing disease in patients on tapering doses of corticosteroids (4).
Inhaled corticosteroids, especially fluticasone, have also been implicated in the development of CSS (2-4,10). Termeer et al. (9,10) recently reported a patient with recurrent bronchial infections and airway obstruction who was prescribed prednisolone and a fluticasone/salmeterol inhaled discus four weeks prior to the development of CSS. The patient initially responded to interferon alpha 2a therapy, but his disease relapsed when the interferon dose was tapered. After finding an association between inhaled fluticasone and CSS, his inhaler was switched to budesonide and his condition significantly improved. Termeer's subject, like our patient, had no prior history of asthma or chronic corticosteroid use, and had started the inhaler just prior to the onset of his disease. Also, both patients improved dramatically after discontinuing their inhalers.
Termeer's patient, like most others previously reported in the literature, also fulfilled the ACR criteria for CSS (9,10). In contrast, with the exception of his skin biopsy findings, our patient met none of the other criteria. He had no worsening of his pulmonary symptoms in association with his skin eruption, no evidence of sinusitis, and his blood eosinophil count was within normal limits. His peripheral neuropathy was the result of a severe injury sustained in an accident years before. Despite his unusual presentation, we believe the temporal association of his skin disease with the inhaled fluticasone/salmeterol combination raises strong circumstantial evidence associating this medication with his skin eruption. As discussed above, many cases of CSS associated with LTRAs and inhaled steroids are probably related to the unmasking of preexisting disease; however, Termeer's case and ours had no previous history of asthma or chronic systemic steroid use. We believe this case report adds to the growing evidence that medications such as fluticasone may be directly involved. Although it is possible that the salmeterol played a role in this process, it has not been previously associated with CSS, and we believe it is unrelated.
It is unclear why the skin was the only organ system affected in our patient, and we could find no other reports of skin-limited disease. All previously reported cases either met the criteria for CSS or had multiple manifestations of the disease. Our patient may simply represent one end of the spectrum in which only the skin is affected. As noted earlier, the histological skin changes seen in our patient have been associated with other systemic conditions; however, he had no evidence of any systemic process reported to trigger this tissue response (13,16).
In conclusion, reports of drug-related forms of CSS have become increasingly common. We present the first case of skin-limited Churg-Strauss-like granulomatous vasculitis related to inhaled corticosteroid use.
(1.) Churg J, Strauss L. Allergy granulomatosis, allergic angiitis, and polyarteritis nodosa. Am J Pathol 1951;27.
(2.) Le Gall C et al. Inhaled corticosteroids and Churg-Strauss syndrome: a report of five cases. Eur Respir J 2000; 15:978-81.
(3.) Bili A et al. Seven cases of complete and incomplete forms of Churg-Strauss syndrome not related to leukotriene receptor antagonists. J Allergy Clin Immunol 1999; 104:1060-5.
(4.) Lilly CM et al. Asthma therapies and Churg-Strauss syndrome. J Allergy Clin Immunol 2002; 109:S1-19.
(5.) Wechsler ME et al. Leukotriene modifiers and Churg-Strauss syndrome: adverse effect or response to corticosteroid withdrawal? Drug Saf 1999; 21:241-51.
(6.) Wechsler ME et al. Pulmonary infiltrates, eosinophilia, and cardiomyopathy following corticosteroid withdrawal in patients with asthma receiving zafirlukast. JAMA 1998; 279:455-7.
(7.) Kinoshita Met al. Churg-Strauss syndrome after corticosteroid withdrawal in an asthmatic patient treated with pranlukast. J Allergy Clin lmmunol 1999; 103:534-5.
(8.) Wechsler ME et al. Churg-Strauss syndrome in patients receiving montelukast as treatment for asthma. Chest 2000; 117:708-13.
(9.) Termeer C et al. Low-dose interferon alfa-2b for the treatment of Churg-Strauss syndrome with prominent skin involvement. Arch Dermatol 2001; 137:136-8.
(10.) Termeer C et al. Churg-Strauss syndrome associated with fluticasone therapy. Arch Dermatol 2001; 137:1527-8.
(11.) Masi AT et al. The American College of Rheumatology 1990 criteria for the classification of Churg-Strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum 1990; 33:1094-100.
(12.) Davis MD et al. Cutaneous manifestations of Churg-Strauss syndrome: a clinicopathologic correlation. JAAD 1997; 37:199-203.
(13.) Finan MC, Winkelmann RK. The cutaneous extravascular necrotizing granuloma (Churg-Strauss granuloma) and systemic disease: a review of 27 cases. Medicine (Baltimore) 1983; 62:142-58.
(14.) Chu P et al. The histopathologic spectrum of palisaded neutrophilic and granulomatous dermatitis in patients with collagen vascular disease. Arch Dermatol 1994; 130:1278-1283.
(15.) Green RL, Vayonis AG. Churg-Strauss syndrome alter zafirlukast in two patients nor receiving systemic steroid treatment. Lancet 1999; 353:725-6.
(16.) Wilmoth GJ, Pemiciaro C. Cutaneous extravascular necrotizing granuloma (Winkelmann granuloma): confirmation of the association with systemic disease. JAAD 1996; 34:753-9.
ADDRESS FOR CORRESPONDENCE:
Joseph English, III
COPYRIGHT 2003 Journal of Drugs in Dermatology
COPYRIGHT 2003 Gale Group