College of American Pathologists
Printable Version

  Anatomic Abstracts





cap today

September 2004

Michael Cibull, MD, professor of pathology and laboratory medicine and director of surgical pathology, University of Kentucky Medical Center, Lexington
Subodh Lele, MD assistant professor of pathology and laboratory medicine, University of Kentucky Medical Center
Melissa Kesler, MD, hematopathology fellow, University of Texas Southwestern Medical Center at Dallas

Human cytomegalovirus infection of the GI tract in select patients
Use of immunoperoxidase stains for MART1 and MelanA in lymph nodes
Fungal sinusitis: histologic spectrum and correlation with culture
Use of routine cytogenetic studies in the evaluation of suspected lymphomas

Human cytomegalovirus infection of the GI tract in select patients

Human cytomegalovirus infection is usually reported in immunocompromised patients. This study reports 11 cases of human cytomegalovirus (HCMV) infection of the gastrointestinal tract diagnosed in apparently immunocompetent hosts. The median age of the patients studied was 76 years, and the major presenting symptoms were diarrhea, epigastric pain, and abdominal discomfort. The large intestine was involved in six cases, the stomach in four, and the lower esophagus in one. Endoscopy revealed ulcers or hypertrophic folds in the GI tract and single ulcers or erosions in the colon and rectum. Light microscopy showed chronic inflammatory infiltrate in the lamina propria in all cases. The diagnosis of HCMV infection was based on the histological and immunohistochemical identification of HCMV inclusion bodies in a variety of cell types, including epithelial, endothelial, stromal, and smooth muscle cells. Classical inclusions, characterized by an "owl's eye" appearance, and atypical inclusions were found. No apparent causes of immunodeficiency were detected for all patients at the time HCMV infection was diagnosed. At followup, however, four patients were found to harbor a malignant tumor—in the pancreas, lung, Vater's papilla, or extrahepatic bile duct—at an interval of two to five months after the diagnosis of HCMV infection. HCMV infection of the GI tract might be an early clue to the presence of immunologic defects induced by an underlying neoplasia, particularly in the elderly.

Maiorana A, Baccarini P, Foroni M, et al. Human cytomegalovirus infection of the gastrointestinal tract in apparently immunocompetent patients. Hum Pathol. 2003;34:1331-1336.

Reprints: Dr. Antonio Maiorana, Anatomia Patologica, Policlinico, via del Pozzo, 41100 Modena, Italy

Use of immunoperoxidase stains for MART-1 and MelanA in lymph nodes

MART-1 and MelanA are sensitive markers of melanocytic differentiation and are used to help detect melanoma micrometastases in sentinel lymph nodes. However, the false-positive rates of these antibodies have not been adequately evaluated. The authors examined 217 lymph nodes (LNs) from patients with no history of melanoma: 117 sentinel lymph nodes (SLNs) from breast cancer patients, 79 LNs from other nonmelanoma malignancy patients, and 21 reactive LNs. Capsular melanocytic nevi were identified in five SLNs from five breast cancer patients using both antibodies. Two of these five SLNs with capsular nevus also contained MART-1 and MelanA-positive cells within lymph node parenchyma. Individual immunoperoxidase-positive cells were also identified within the parenchyma of lymph nodes without capsular nevus (nine LNs with MART-1 and three LNs with MelanA). The false-positive rate was 5.1 percent for MART-1 and 2.4 percent for MelanA. The authors concluded that MART-1 or MelanA-positive cells may be present in lymph nodes from patients without melanoma. Therefore, MART-1 and MelanA-positive cells in SLNs from melanoma patients, without corresponding atypia or hematoxylin-and-eosin findings, should be interpreted with caution.

Yan S, Brennick JB. False-positive rate of the immunoperoxidase stains for MART-1/MelanA in lymph nodes. Am J Surg Pathol. 2004;28:596-600.

Reprints: Dr. Jeoffry B. Brennick, Dept. of Pathology, Dartmouth-Hitchcock Medical Center, 1 Medical Center Drive, Lebanon, NH 03756;

Fungal sinusitis: histologic spectrum and correlation with culture

Fungi are important etiologic agents of sinusitis. However, features of fungal sinusitis, including histologic spectrum, diagnostic mishaps, incidence, and fungal types, have not been systematically studied. The authors retrieved, from 1996 through 2001, 788 surgical pathology sinus specimens from 384 cases. Fungal sinusitis was diagnosed in 58 specimens (seven percent) from 47 cases (12 percent). Four histologic categories of fungal sinusitis were identified: allergic fungal sinusitis in 34 cases (copious mucin, abundant eosinophils, Charcot-Leyden crystals [so-called allergic mucin], with rare noninvasive fungal hyphae); mycetoma/fungus ball in 11 cases (tightly packed fungal hyphae without allergic mucin or tissue invasion); chronic invasive fungal sinusitis in one case (tissue granulomas with fungal hyphae); and acute fulminant fungal sinusitis in one case (fungal vascular invasion). The diagnosis was initially missed in 16 of 34 (47 percent) cases of allergic fungal sinusitis despite typical features; incorrect classification was noted in 47 percent of cases. Sixty-seven percent of cases had positive fungal cultures, dematiaceous fungi being the most common. Allergic fungal sinusitis accounted for the majority of cases of fungal sinusitis. Although misdiagnosis or incorrect classification of fungal sinusitis is rather frequent, awareness of the distinctive morphologic features of this entity may prevent these errors.

Granville L, Chirala M, Cernoch P, et al. Fungal sinusitis: histologic spectrum and correlation with culture. Hum Pathol. 2004;35:474-481.

Reprints: Dr. Laura Granville, Dept. of Pathology, Methodist Hospital and Baylor College of Medicine, 1 Baylor Plaza, Houston, TX 77030

Use of routine cytogenetic studies in the evaluation of suspected lymphomas

While cytogenetic analysis is critically important for diagnosing and classifying acute leukemia, it is not commonly used in lymphoma diagnosis, even though many lymphomas have characteristic cytogenetic abnormalities. To evaluate the utility of routine cytogenetic studies in this setting, the authors reviewed the results of G-banded karyotyping studies in 279 lymph node or tissue biopsies submitted for suspected lymphoma. Complete cytogenetic studies were obtained in 177 cases (63.4 percent), including 115 (69.3 percent) of 166 hematolymphoid neoplasms. Abnormal karyotypes were obtained in 97 (84.3 percent) of the hematolymphoid neoplasms. In at least three cases (2.6 percent of successfully karyotyped malignancies), the findings directly and substantially contributed to the final diagnoses (two atypical Burkitt lymphomas and one splenic marginal zone lymphoma). In many more cases, characteristic but nonspecific findings were present. Abnormalities of suggested prognostic importance were identified in 14 of 32 (44 percent) cases of follicular lymphoma and eight of 33 (24 percent) cases of diffuse large B-cell lymphoma. The authors concluded that most karyotyped lymphomas will yield abnormal findings, including many that are not completely specific but that support the diagnosis, some that add prognostic information, and a few that help establish a diagnosis that might otherwise have been missed.

Cook JR, Shekhter-Levin S, Swerdlow SH. Utility of routine classical cytogenetic studies in the evaluation of suspected lymphomas: results of 279 consecutive lymph node/extranodal tissue biopsies. Am J Clin Pathol. 2004;121:826-835.

Reprints: Dr. S. Swerdlow, Dept. of Pathology, Division of Hematopathology, Room 606-PUH, University of Pittsburgh School of Medicine, 200 Lothrop St., Pittsburgh, PA 15213