Toquet, et al: Eosinophilic fasciitis 1811 From the Departments of Pathology, Internal Medicine, INSERM U539, and Dermatology, University Hospital of Nantes, Nantes, France. C. Toquet, MD, Department of Pathology; M.A. Hamidou, MD, Senior Physician, Department of Internal Medicine; K. Renaudin, MD, Department of Pathology; A. Jarry, PhD, INSERM U539; P. Foulc, MD; S. Barbarot, MD, Department of Dermatology; C. Laboisse, MD, Professor of Pathology; J-M.G. Mussini, MD, Department of Pathology. Address reprint requests to Dr. C. Toquet, Department of Pathology A, CHU Hôtel Dieu, 44093 Nantes Cédex, France. E-mail: claire.toquet@chu-nantes.fr Submitted June 20, 2002; revision accepted January 17, 2003. Eosinophilic fasciitis (EF) was first described in 1974 by Shulman as an autonomous syndrome characterized by diffuse fasciitis with hyperglobulinemia and eosinophilia 1 . As tissue and blood eosinophilia did not appear as constant criteria, the term Shulman syndrome was preferred to EF 2 . More than 200 cases have been published in the literature. Some striking clinical similarities with scleroderma and inflammatory myopathies have been observed. However, unlike scleroderma, Raynaud’s phenomenon and visceral involvement are classically absent in EF, and patients respond generally well to corticosteroid therapy 3,4 . Morphologically, this disease belongs to the “fasciitis- panniculitis syndromes,” defined as a fibrous and inflamma- tory thickening of subcutaneous septal-fascial-perimysial collagenous scaffold 5 . The pathophysiology of the disease is still unknown. However, immune-mediated mechanisms appear to play a pivotal role 6-8 . The immune origin of this disease is supported by the successive detection of elevated immunoglobulins and circulating immune complexes in patients with active EF and finally the occurrence of EF in chronic graft-versus-host disease 6,8 . In addition, autoim- mune mechanisms have been proposed on the basis of the association of EF with other autoimmune disorders 7 . Interestingly, in their attempt to define the cytokine network potentially playing a role in the cascade of events leading to tissue fibrosis, Viallard, et al showed the overex- pression of type 1 and type 2 cytokines from peripheral blood cells 9 . In addition, they concluded that a thorough immunophenotypic characterization of the local inflamma- tory infiltrate was mandatory to understand the pathophysi- ology of EF. We characterized the inflammatory infiltrate and demon- strated the predominance of macrophages and CD8+ T lymphocytes. Cytotoxic properties were found in 14% of CD8+ T lymphocytes, as shown by granzyme B expression. Our results suggest a cytotoxic cellular immune response in EF that could be triggered by infectious or environmental agents. MATERIALS AND METHODS Patients. Eleven patients were selected from a series of 25 patients seen In Situ Immunophenotype of the Inflammatory Infiltrate in Eosinophilic Fasciitis CLAIRE TOQUET, MOHAMED AMINE HAMIDOU, KARINE RENAUDIN, ANNE JARRY, PHRYNÉ FOULC, SÉBASTIEN BARBAROT, CHRISTIAN LABOISSE, and JEAN-MARIE GILBERT MUSSINI ABSTRACT. Objective. Eosinophilic fasciitis (EF) is histologically characterized by a fibrous and inflammatory thickening of subcutaneous septal-fascial-perimysial collagenous scaffold. This study aims to define the immunophenotype of inflammatory cells of fascia and muscle underlying the in situ immune response in EF. Methods. In 11 cases of EF, we determined the phenotype of inflammatory cells, expression of MHC class I and class II antigens, and C5b9 membranolytic attack complex (MAC) deposits by immuno- histochemistry analysis of fascia tissue. Muscle biopsies from 9 patients with active dermatomyositis and 5 with active polymyositis were used as controls. Results. In all patients but one, the inflammatory infiltrate was mainly composed of macrophages associated with CD8+ T lymphocytes (CD4/CD8 ratio < 1) and few eosinophils. Cytotoxic proper- ties were found in 14% of CD8+ T lymphocytes, as shown by granzyme B expression. MHC Class I antigens were overexpressed (5/7) by muscle fibers, with a paratrabecular reinforcement in 4 cases. MHC class II antigens were not expressed by muscle fibers except in one case. C5b9 MAC deposits were not detected. Conclusion. Our in situ characterization of inflammatory infiltrate demonstrates the predominancy of macrophages and CD8+ T lymphocytes. Some of these CD8+ lymphocytes contain granzyme B, thus suggesting a cytotoxic cellular immune response in EF, which could be triggered by infectious or environmental agents. (J Rheumatol 2003;30:1811–5) Key Indexing Terms: SHULMAN EOSINOPHILIC FASCIITIS FASCIITIS-PANNICULITIS IMMUNOPHENOTYPE INFLAMMATORY INFILTRATE Personal, non-commercial use only. The Journal of Rheumatology Copyright © 2003. All rights reserved. www.jrheum.org Downloaded on January 20, 2023 from