Format

Send to

Choose Destination
See comment in PubMed Commons below
Am J Surg Pathol. 2011 Feb;35(2):289-94. doi: 10.1097/PAS.0b013e3182073ac0.

Significant variation in the assessment of cervical involvement in endometrial carcinoma: an interobserver variation study.

Author information

1
Department of Pathology, Belfast Health and Social Care Trust, Northern Ireland. glenn.mccluggage@belfasttrust.hscni.net

Abstract

The histologic assessment of cervical involvement in endometrial carcinoma may be problematic for a number of reasons, but an accurate evaluation of this is important for correct staging, dictating the need for adjuvant therapy, and prognostication. In this study, we assessed interobserver variation in the evaluation of cervical involvement in hysterectomy specimens of endometrial carcinoma among 6 specialist gynecologic pathologists. Seventy-six cases of endometrial carcinoma enriched for cases exhibiting some perceived issue in the assessment of cervical involvement were used. In all the cases, a single slide of the primary tumor in the uterine corpus and a single slide of the cervix were circulated among the 6 participants who filled in a proforma. On the basis of the responses, the tumors were staged according to the 1988 International Federation of Gynecology and Obstetrics (FIGO) staging system (I, IIA, IIB) and the 2009 FIGO staging system (I, II). Using the 1988 FIGO staging system, the unweighted and weighted κ values between individual observers ranged from 0.3115 to 0.6139 (average 0.4675) and from 0.3492 to 0.6533 (average 0.5065), respectively. The κ values between observers for the 2009 FIGO staging system ranged from 0.3481 to 0.6862 (average 0.4908). Although enriched for problematic cases, our study shows that there is at most a fair-to-good agreement among specialist gynecologic pathologists in the assessment of cervical involvement in endometrial carcinoma. Problematic factors include determination of the junction between the lower uterine segment and upper endocervix, the distinction between "floaters" and true cervical glandular involvement, the distinction between cervical glandular involvement and stromal involvement, and the distinction between cervical glandular involvement and reactive non-neoplastic lesions of the endocervical glands. There is a need for specialist pathology groups dealing with gynecologic cancers to develop and disseminate recommendations regarding the assessment of cervical involvement in endometrial carcinoma.

PMID:
21263250
DOI:
10.1097/PAS.0b013e3182073ac0
[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Lippincott Williams & Wilkins
    Loading ...
    Support Center