College of American Pathologists
Printable Version

  Anatomic Abstracts





cap today

December 2001

Morcellated kidney specimens: sampling guidelines
As laparoscopic surgery becomes more prevalent and more pathology specimens are derived from this procedure, criteria for evaluation will need to be established, especially in oncologic specimens, where staging is important. In this study, morcellated renal specimens resulting from laparoscopic nephrectomy/nephroureterectomy were evaluated retrospectively, and sampling strategy models were generated based on a statistical model that described the relationship of the fraction of specimen submitted, the tumor-to-kidney volume ratio (TKR), and the probability of identifying tumor within the sampled fraction of specimen. Preoperative radiologic imaging (for tumor size) and total kidney specimen weight were used to determine the TKR, and the sampling strategies were compared. Using a 75 percent probability sequential approach and a threshold TKR of 0.15 or greater suggested that five percent of the morcellated specimen needed to be submitted initially to identify the tumor when no suspicious lesions were grossly visible. The authors presented an algorithm for optimal sampling of specimens based on the TKR. They also concluded that pTNM cannot be assigned for renal cell and urothelial carcinomas and invasion cannot be definitively ruled out for urothelial tumors. The authors also compared radiologic and pathologic tumor staging for renal cell carcinoma removed by traditional nephrectomy and determined that although radiologic staging for RCC was good, overstaging and understaging can occur in five to 35 percent of cases.

Rabban JT, Meng MV, Yeh B, et al. Kidney morcellation in laparoscopic nephrectomy for tumor: recommendations for specimen sampling and pathologic tumor staging. Am J Surg Pathol. 2001;25:1158-1166.

Reprints: Dr. Joseph T. Rabban, University of California, San Francisco, Dept. of Pathology, Box 0102, 505 Parnassus Ave., San Francisco, CA 94143;

H. pylori and gastric cancer
In a Japanese study of 1,526 patients, gastric cancer developed in 2.9 percent of those subjects infected with Helicobacter pylori and none of those who were not infected. Patients underwent endoscopy with biopsy at enrollment in the study and one and three years later. H. pylori infection was assessed by histological examination, serologic testing, and rapid urease tests. Findings included significantly higher risk for the development of carcinoma in those patients with severe gastric atrophy, corpus predominant gastritis, and intestinal metaplasia. Gastric cancers developed in 4.7 percent of patients with non-ulcer dyspepsia, 3.4 percent of those with gastric ulcers, and 2.2 percent of those with gastric hyperplastic polyps. None of the patients with duodenal ulcers developed gastric cancer. Intestinal and diffuse types of carcinoma were identified in infected patients. An accompanying editorial by Fox and Wang discusses the role of H. pylori as a carcinogen. The authors point out that eradicating H. pylori may delay or prevent the development of gastric cancer in patients at risk. They conclude that although confirmatory studies are needed, H. pylori eventually may be viewed as an important carcinogen and a target for cancer prevention.

Uemura N, Okamoto S, Yamamoto S, et al. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med. 2001;345:784-789.

Reprints: Dr. N. Uemura, Department of Gastroenterology, Kure Kyosai Hospital, 2-3-28 Nishi-chuo, Kure City, Japan;

Fox JG, Wang TC. Helicobacter pylori—not a good bug after all. Editorial. N Engl J Med. 2001;345:829-831.

Objective criteria for distinguishing flat urothelial carcinoma in situ from normal nonreactive urothelium
Subjective assessment of flat urothelial lesions has often resulted in an under- or overdiagnosis of urothelial carcinoma in situ. To identify reproducible morphologic objective criteria, the authors used an image analysis system to measure several nuclear features (area, diameter, roundness, ellipticity, and optical density) in 20 cases each of CIS, urothelial dysplasia, and normal urothelium with adjacent lymphocyte controls. They did not analyze reactive urothelial lesions and umbrella cells. The most useful, statistically significant morphologic parameter was mean nuclear area of the largest 25 percent of nuclei. The mean upper quartile nuclear area relative to lymphocytes was 2.2 times (range, 1.4 to 2.8 times) in normal urothelium, 2.9 times (range, 1.8 to 3.6 times) in urothelial dysplasia, and 4.9 times (range, 4.0 to 7.6 times) in CIS. The nuclear size measurements for urothelial dysplasia and normal urothelium overlapped, but the separation between CIS and urothelial dysplasia was essentially complete. Pathologists can assess the size of urothelial nuclei without an image analysis system by comparing the size of urothelial nuclei in the specimen to the size of the nuclei of lymphocytes, which are almost always present in bladder biopsies. CIS nuclei are approximately five times the size of lymphocytes, whereas normal nonreactive urothelial nuclei are about two times the size of lymphocytes. Because they did not evaluate reactive epithelium, the authors do not recommend applying these morphometric measurements in assessing markedly inflamed urothelium.

Milord RA, Lecksell K, Epstein JI. An objective morphologic parameter to aid in the diagnosis of flat urothelial carcinoma in situ. Hum Pathol. 2001;32:997-1002.

Reprints: Dr. Jonathan I. Epstein, Dept. of Pathology, The Johns Hopkins Hospital, Weinberg Bldg., Rm. 2242, 401 N. Broadway St., Baltimore, MD 21231; jepstein

Differential diagnosis of ovarian tumors: role of patterns and
cell types

Awareness of the broad range of lesions that may exhibit particular patterns or contain one or more cell types is crucial in formulating a differential diagnosis of ovarian tumors. The following patterns are considered: moderate-to-large glandular and hollow tubular; solid tubular and pseudotubular; cords and ribbons; insular; trabecular; slit-like and reticular spaces; microglandular and microfollicular; macrofollicular and pseudomacrofollicular; papillary; diffuse; fibromatous-thecomatous; and biphasic and pseudobiphasic. The following cell types are considered: small round cells; spindle cells; mucinous cells comprising columnar, goblet cell, and signet ring cell subtypes; clear cells; hobnail cells; oxyphil cells; and transitional cells. The morphologic diversity of ovarian tumors poses many challenges; thus it is important to be knowledgeable about the occurrence and frequency of these patterns and cell types in various tumors and tumor-like lesions. A specific diagnosis usually can be made by evaluating routinely stained slides, but, much less often, the diagnosis requires special staining, immunohistochemical staining, or, very rarely, ultrastructural examination. Clinical data, operative findings, and gross features of the lesions may provide decisive diagnostic clues.

Young RH, Scully RE. Differential diagnosis of ovarian tumors based primarily on their patterns and cell types. Semin Diagn Pathol. 2001;18:161-235.

Reprint information unavailable.

Diameter of perineural invasion as an independent predictor of recurrence in radical prostatectomy specimens
The prognostic significance of perineural invasion is controversial, perhaps due to the paucity of analyses of specific quantitative aspects of PNI. The authors analyzed clinical and pathologic parameters from 640 patients with prostate carcinoma who were treated with radical prostatectomy. The clinical parameters studied were patient age and preoperative serum PSA level, and the pathologic parameters included surgical margin, pelvic lymph node status, presence of seminal vesical invasion, Gleason score, extracapsular extension, total tumor volume, presence of PNI, and maximum diameter of PNI foci if present. None of the patients received preoperative hormonal therapy or radiotherapy. The diameter of the largest PNI tumor focus was measured perpendicular to the long axis of the neural Schwann cell nuclei using an ocular micrometer. Although PNI was detected in 477 cases, it was not an independent predictor of disease progression. PNI diameter, however, was an independent predictor of prognosis. Cases with PNI diameters of more than 0.25 mm were associated with substantially higher rates of progression than those with PNI diameters of less than 0.25 mm. PNI diameter was also significant in the subset of cases with cancer confined to the prostate and those with extracapsular extension without seminal vesical invasion or lymph node metastasis, regardless of grade.

Maru N, Ohori M, Kattan MW, et al. Prognostic significance of the diameter of perineural invasion in radical prostatectomy specimens. Hum Pathol.

Reprints: Dr. Thomas M. Wheeler, Dept. of Pathology, MS 205, 6565 Fannin Rd., Houston, TX 77030-2707