Nodular histiocytic hyperplasia of the endometrium.

Nodular Histiocytic Hyperplasia of the Endometrium Full access Masaharu Fukunaga From the Department of Pathology, Jikei University School of Medicine, Tokyo, Japan (Dr Fukunaga); and the Second Department of Obstetrics and Gynecology, Toho University School of Medicine, Tokyo, Japan (Dr Iwaki) From the Department of Pathology, Jikei University School of Medicine, Tokyo, Japan (Dr Fukunaga); and the Second Department of Obstetrics and Gynecology, Toho University School of Medicine, Tokyo, Japan (Dr Iwaki) A case of nodular histiocytic hyperplasia of the endometrium is described. A 45-year-old Japanese woman was found to have an enlarged uterus during her annual checkup. Hysterectomy and bilateral salpingo-oophorectomy specimens revealed uterine leiomyomas, adenomyosis, and acute salpingitis. A 5-mm, well-demarcated, elevated endometrial nodule (an incidental finding) was present and consisted of round or polygonal histiocytic cells with eccentric nuclei and pale or granular cytoplasm. The nuclei were ovoid, reniform, or crescent-shaped and had fine chromatin and inconspicuous nucleoli, and the cytoplasm contained single or multiple vacuoles. Immunohistochemically, the histiocytic cells were positive for vimentin, CD68, and lysozyme and were negative for cytokeratin, S100 protein, estrogen and progesterone receptors, and CD10. Nodular histiocytic hyperplasia in the endometrium is considered to be a reactive process. Differentiation from neoplasms, including signet-ring cell carcinoma, in curettage specimens is critical to avoid unnecessary surgical resection. Nodular histiocytic proliferation has been observed in various anatomic sites. This phenomenon is rare in the endometrium and can be confused with inflammatory and neoplastic conditions.1–14 These histiocytic cells often show signet-ring cell change, and the differentiation between nodular histiocytic proliferation and signet-ring cell carcinoma is very important for the choice of treatments. A case of nodular histiocytic hyperplasia is described, and the differential diagnoses and pathogenesis are discussed. REPORT OF A CASE A 45-year-old, gravida 2, para 2 Japanese woman was found to have an enlarged uterus on an annual checkup. On admission, abdominal computed tomographic scan and magnetic resonance imaging revealed adenomyosis. An abdominal total hysterectomy and a bilateral salpingo-oophorectomy were performed. The patient has been well without disease for 8 years after surgery. PATHOLOGIC FINDINGS Macroscopically, the enlarged uterus, measuring 9 × 6 × 4 cm and weighing 350 g, revealed multiple submucosal and intramuscular masses, measuring up to 2 cm in greatest dimension. The cut surfaces of the masses had a well-demarcated, myoma-like appearance. The cervix, endometrium, ovaries, and fallopian tubes were unremarkable. Microscopically, the uterine masses were conventional leiomyomas (Figure 1), and foci of adenomyosis were also noted. The endometrium was in the secretory phase with shedding features. A 5-mm, well-demarcated, elevated endometrial nodule was noted incidentally (Figure 1). The nodule consisted of clusters of round or polygonal cells with eccentric nuclei and abundant pale or granular cytoplasm. The nuclei were ovoid, reniform, and crescent- or spindle-shaped with fine chromatin and inconspicuous nucleoli, and the cytoplasm contained single or multiple vacuoles (Figure 2). No pigments were identified in the cytoplasm. Some cells had a signet-ring cell appearance. The cells were negative for periodic acid–Schiff, mucicarmine, and von Kossa stains. No atypia or pleomorphism was found, and no mitotic figures were seen. Endometrial glands were present within and at the periphery of the nodule. No vascular structures were found in the nodule. The shedding endometrium also contained clusters of the vacuolated cells. These cells were not seen in the background endometrium, and no features of pyometra or endometritis were evident. No decidual cells were present. The fallopian tubes revealed moderate eosinophilic and neutrophilic infiltrates and edema in the mucosa. The uterine cervix and ovaries were unremarkable. Immunohistochemically, the vacuolated cells in the nodule of the endometrium were positive for CD68 (monoclonal; Dakopatts, Glostrup, Denmark; 1:50) (Figure 3), and some were positive for lysozyme (polyclonal; Dakopatts; 1:2000). These cells were uniformly negative for cytokeratin (CAM 5.2, monoclonal; Becton Dickinson, Mountain View, Calif; 1:1), epithelial membrane antigen (monoclonal; Dakopatts; 1:100), CD45RA (monoclonal; Dakopatts; 1:50), HLA-DR (monoclonal; Nichirei, Tokyo, Japan; 1:1), α1-antitrypsin (polyclonal; Dakopatts; 1:1000), S100 protein (polyclonal; Dakopatts; 1:100), CD10 (monoclonal; Dakopatts; 1:50), and estrogen (monoclonal; Novocastra, Newcastle upon Tyne, United Kingdom; 1:20) and progesterone receptors (monoclonal; Novocastra; 1:50). COMMENT There seems to be at least 2 types of histiocytic cells, including cells with signet-ring cell change, in the endometrium.179111314 The first type is of endometrial stromal origin and often shows vacuolated decidual change.791114 The second type is of histiocytic origin.1257 The first type is a foamy histiocyte, which is often associated with hyperestrogenic lesions, including endometrial hyperplasia and adenocarcinoma,171114 and xanthogranulomatous endometritis.34 The foamy histiocytes contain single or multiple cytoplasmic vacuoles and dark small nuclei, and can be positive for estrogen and progesterone receptors. They are usually aggregated or loosely scattered in the endometrial stroma. Histiocytes of the second type, which were found in the present case, cytologically resemble Langerhans cells; the nuclei are displaced, grooved, lobulated, reniform, or ovoid, and the cytoplasm is granular and less abundant than that of foamy histiocytes and contains single or multiple vacuoles. These cells are not associated with endometrial hyperplasia or adenocarcinoma. They are immunohistochemically positive for histiocytic markers, including CD68 and lysozyme. The differential diagnosis of nodular histiocytic hyperplasia of the endometrium includes xanthogranulomatous endometritis, Langerhans cell histiocytosis, malakoplakia, and signet-ring cell carcinoma. Xanthogranulomatous endometritis is unlikely because it shows granulomas composed of histiocytes, other inflammatory cells, multinucleated giant cells, and cholesterol clefts.34 Langerhans cell histiocytosis can be differentiated from nodular histiocytic hyperplasia in that histiocytes in the former are associated with eosinophilic infiltrates and are positive for S100 protein.812 Signet-ring cell carcinoma can be ruled out because histiocytes in nodular hyperplasia are negative for periodic acid–Schiff, mucicarmine, and cytokeratin. The absence of Michaelis-Gutmann bodies, other inflammatory cells, and giant cells exclude the possibility of malakoplakia.6 An endometrial stromal cell origin could be excluded because histiocytes in nodular hyperplasia lacked estrogen and progesterone receptors, CD10, and characteristic vascular structures. The mechanism of the development of histiocytic nodular hyperplasia may involve a complex interaction of elements, such as obstruction, pyometra, inflammation, and a lipid source.125911 In the present case, it was an incidental finding, and it may represent a reactive response to the adenomyosis or uterine leiomyoma, but this possibility is not likely. If so, histiocytic nodules would be noticed more frequently. The nodules might not have originated in the endometrium, and the absence of vascular structures within the nodules and their detached nature suggest that the nodules might be introduced or transported from extrauterine sites. The acute salpingitis might be associated with nodular histolytic hyperplasia. Nodular histiocytic hyperplasia in the endometrium is considered to be a reactive process. Differentiation from neoplasms, including signet-ring cell carcinoma, in curettage specimens is critical to avoid unnecessary surgical resection. Copyright: College of American Pathologists 2004 Figure 1. Endometrial nodule (top) and submucosal leiomyoma (bottom) (hematoxylin-eosin, original magnification ×20). Figure 2. Note the cluster of vacuolated cells in the endometrium (hematoxylin-eosin, original magnification ×400). Figure 3. Vacuolated cells were positive for CD68 (original magnification ×400) Contributor Notes Reprints: Masaharu Fukunaga, MD, Department of Pathology, Jikei Daisan Hospital, 4-11-1, Izumihoncho, Komaeshi, Tokyo, 201-8601, Japan ([email protected])