Inflammatory pseudotumor of the posterior mediastinum

We report the case of a 33-year-old white female patient. Her personal history was unremarkable. She presented with a three-month history of progressive dysphagia, which was accompanied by vomiting after eating, weight loss (8 kg), and asthenia. Physical examination was normal. Laboratory test results showed microcytic hypochromic anemia (hemoglobin, 10.2 g/dL; mean corpuscular volume, 69 fL; and mean corpuscular hemoglobin, 22 pg). No other changes were found. Upper gastrointestinal endoscopy revealed punctal stenosis 23 cm from the maxillary dental arch, at the transition between the upper third and the middle third of the esophagus, the esophageal mucosa showing mild nodosity and stiffness upstream of the stenosis but no typical signs of neoplastic infiltration into the area of stenosis. The stenosis hindered the progression of the endoscope. Esophageal mucosal biopsies showed edematous esophagitis. An esophagogram showed stenosis of the esophagus at the cervicothoracic junction and esophageal dilation upstream of the stenosis, food residue being probably present (Figure 1). A CT scan of the chest showed solid tissue in the posterosuperior mediastinum, as well as circumferential involvement of the trachea and esophagus, together with significant narrowing of the esophageal lumen and esophageal dilation upstream (Figure 2). Diagnostic hypotheses included lymphoproliferative disease and gastrointestinal stromal tumor. Fiberoptic bronchoscopy showed no endotracheal lesion. Thoracoscopy showed a tracheoesophageal lesion. The lesion had ill-defined margins and was approximately 5 cm in diameter, as well as being white, hard, and fibrous. Biopsies of the lesion showed no malignancy. Given the diagnostic uncertainty and the persistence of the symptoms, a thoracotomy was performed, the entire lesion being resected and the thoracic esophagus being freed. Frozen section analysis was suggestive of gastrointestinal stromal tumor. The histopathological findings were consistent with a nonspecific chronic inflammatory process. Immunohistochemistry was positive for CD20, CD3 (SP7), and smooth muscle actin, and negative for β-catenin, CD30, CD15, and CD245, findings that were consistent with inflammatory pseudotumor. After surgery, the patient progressed favorably, the symptoms having disappeared. At this writing, 1 year had passed since oncologic control had been achieved. Inflammatory pseudotumor was first described by Brunn in 1932 and was so designated by Umiker et al. in 1954 because it can mimic malignancy clinically and radiologically. Since then, inflammatory pseudotumor has been reported in the literature under different names, such as histiocytoma, plasma cell granuloma, xanthoma, and inflammatory myofibroblastic tumor. Inflammatory pseudotumor is a rare tumor of unknown origin that predominantly affects the lung …