Salpingo-oophorectomy specimens for endometrial c... [Hum Pathol. 2012] - PubMed - NCBI

Salpingo-oophorectomy specimens for endometrial cancer staging: a comparative analysis of representative sampling versus whole tissue processing

Abstract

Involvement of the ovary and/or fallopian tube by an endometrial cancer is uncommon but clinically significant because this is one of the indications for adjuvant chemotherapy. The authors evaluated whether the routine microscopic evaluation of the adnexal organs in this setting should be of the entire specimen or of representative sections. Slides and reports were reviewed for 105 consecutive patients who underwent a staging salpingo-oophorectomy (205 ovaries, 210 tubes) for endometrial carcinoma/carcinosarcoma. The study period encompassed the periods before and after an institutional policy change from discretionary (predominantly representative) adnexal sampling to obligatory total processing. Ninety-four and 111 ovaries (and 92 and 118 tubes) were entirely and representatively processed, respectively. Even when using the most expansive definition of ovarian gross abnormality (definition with the highest sensitivity and lowest specificity for microscopically confirmed cancer), we still identified 4 (of 148; 2.7%) grossly normal ovaries and 3 (of 187; 1.6%) grossly normal tubes that were found to harbor microscopic cancers. There was no significant increase in the number of grossly occult cancers detected after the policy change, and 5 (71%) of the 7 grossly occult cancers were in the representatively sampled group. Approximately 3.76 more blocks per patient were required for total overrepresentative processing, and the total cost of these additional sections was estimated to be $25.57 per patient. In conclusion, the 1.6% to 2.7% of grossly normal adnexa that proved to be cancerous represents, at least theoretically, the risk for misdiagnosis and understaging that is associated with representative sampling, at relatively modest savings. However, the findings in this study do not provide direct evidentiary support for routine complete processing.