College of American Pathologists
Printable Version

  Anatomic Abstracts





cap today

July 2003

Criteria for identifying serrated adenomas of the colon and rectum
The authors evaluated the histological characteristics ascribed in the literature to serrated adenomas and developed a practical working model for identifying the adenomas. They documented the frequency and location of serrated adenomas identified in an unselected series of individuals undergoing colonoscopic evaluation and the clinical characteristics of those individuals. They identified 140 consecutive individuals (prospective polyp data set; 97 male, 43 female; age mean, 63.3 years; age range, 29 to 98 years) with 255 polyps from 919 individuals undergoing colonoscopy. Additional polyps previously removed from these individuals were included for histological assessment (extended polyp data set, n=380). All polyps were assessed for eight selected architectural and cytological features. In the prospective polyp data set, 56 patients had 72 hyperplastic polyps, seven had nine serrated adenomas, three had four admixed polyps, and 98 had 170 conventional adenomas. No difference was found in the age, gender, or cancer association of the seven patients with serrated adenomas when compared with the other individuals with polyps. The prevalence of serrated adenomas was nine of 919 (one percent), with an average size of 5.8 mm. When assessing serrated adenomas histologically, the combination of nuclear dysplasia and serration of 20 percent or more of crypts provided the most accurate model for detecting these lesions (sensitivity, 100 percent; specificity, 97 percent). Other criteria provided supportive evidence but did not increase the diagnostic yield. The optimum model for the histological identification of serrated adenoma includes the presence of a serrated architecture in 20 percent or more of crypts in association with surface epithelial dysplasia.

Bariol C, Hawkins NJ, Turner JJ, et al. Histopathological and clinical evaluation of serrated adenomas of the colon and rectum. Mod Pathol. 2003;16(5):417-423.

Reprints: Dr. Robyn L. Ward, Dept. of Medical Oncology, St. Vincent’s Hospital, Victoria St., Darlinghurst, 2010, Australia;

Comparing voice-automated transcription to human transcription for generating pathology reports
Continuous speech systems were developed in 1994, with the latest commercially available editions claiming accuracy of up to 98 percent of speech recognition at natural speech rates. The authors evaluated the efficacy of one commercially available voice-recognition software system with pathology vocabulary in generating pathology reports, and they compared this with human transcription and drew cost-analysis conclusions regarding human versus computer-based transcription. To reach their goal, they simultaneously generated 206 routine pathology reports from the surgical pathology material handled at St. Joseph’s Healthcare, Hamilton, Ontario, using computer-based transcription and human transcription. They used a desktop 450-MHz Intel Pentium III processor with 192 MB of RAM, speech-quality sound card (Sound Blaster), noise-canceling headset microphone, and IBM Via-Voice Pro version 8 with pathology vocabulary support (Voice Automated, Huntington Beach, Calif.). The cost of the hardware and software was approximately $2,250 (Canadian). A total of 23,458 words were transcribed using both methods, with a mean of 114 words per report. The mean accuracy rate was 93.6 percent (range, 87.4 to 96 percent) using the computer software, compared with a mean accuracy rate of 99.6 percent (range, 99.4 to 99.8 percent) for human transcription (P<.001). Time needed to edit documents by the primary evaluator (M.A.) using the computer was an average of twice that needed for editing the documents produced by human transcriptionists (range, 1.4 to 3.5 times). The extra time needed to edit documents was 67 minutes per week (13 minutes per day). The authors concluded that computer-based continuous speech-recognition systems in pathology can be used successfully in pathology practice, even during the handling of gross pathology specimens. However, the relatively low accuracy rate of this voice-recognition software, with the resultant increased editing burden on pathologists, may not encourage its application on a wide scale in pathology departments with sufficient human transcription services, despite the potential for significant financial savings. Computer-based transcription represents an attractive and relatively inexpensive alternative to human transcription in departments where there is a shortage of transcription services and is likely to be more commonly used in pathology departments in the future.

Al-Aynati MM, Chorneyko KA. Comparison of voice-automated transcription and human transcription in generating pathology reports. Arch Pathol Lab Med. 2003;127: 721-725.

Reprints: Dr. Katherine A. Chorneyko, St. Joseph’s Healthcare, 50 Charleton Ave. East, Hamilton, Ontario, L8N 3Z5, Canada;

Intraparenchymal nevus cell aggregates in lymph nodes— a possible diagnostic pitfall
The authors conducted a study to draw attention to the finding of nevus cells in the parenchyma of lymph nodes and to alert pathologists to this as a potential diagnostic pitfall, especially in patients with concurrent melanoma or carcinoma. It is well-documented that nevus cells can be found within the fibrous capsule and trabeculae of lymph nodes, but it is less well-known that nevus cells can be found in the lymph node parenchyma. The authors reported the findings in 13 cases of nevus cell aggregates located in the cortical or medullary parenchyma of lymph nodes, or both. Seven of the 13 patients had a primary diagnosis of melanoma, three had no known malignancy, one had breast carcinoma, one had adnexal carcinoma of the skin, and one had squamous cell carcinoma of the tonsil. Of the seven patients with melanoma, four had axillary lymph node dissections and three had inguinal lymph node dissections. The patient with adnexal carcinoma had metastatic carcinoma in 14 of 20 lymph nodes that had been dissected; one of them also had intraparenchymal nevus cells. The patient with squamous cell carcinoma of the tonsil had an intraparenchymal nevus cell aggregate in one of the 21 dissected lymph nodes; all 21 were negative for carcinoma. Nests of intraparenchymal nevus cells ranged from clusters of only a few cells to 2.1-mm aggregates. Mitotic figures, prominent nucleoli, and lymphatic-vascular invasion were not detected in any of the melanocytic aggregates. The melanocytic cells of the nevus cells aggregates expressed S-100 protein or MART-1, or both, but not gp100 protein (HMB-45). Less than one percent of the nevus cells expressed Ki-67. Awareness that nevus cells can be present in nodal parenchyma, analysis of their morphologic features, including comparison with any previous or existing melanoma or carcinoma, and immunophenotyping will help pathologists establish the correct diagnosis in most instances.

Shah VI, Raju U, Chitale D, et al. False-negative core needle biopsies of the breast: an analysis of clinical, radiologic, and pathologic findings in 27 consecutive cases of missed breast cancer. Cancer. 2003;97: 1824-1831.

Biddle DA, Evans HL, Kemp BL, et al. Intraparenchymal nevus cell aggregates in lymph nodes: a possible diagnostic pitfall with malignant melanoma and carcinoma. Am J Surg Pathol. 2003;27(5):673-681.

Reprints: Dr. Victor G. Prieto, Dept. of Pathology, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030;

Anatomic pathology abstracts editors

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.