Adequacy criteria for endometrial biopsy samples do not exist.
To assess the sensitivity of endometrial sampling for detecting neoplasia in the setting of extensive glandular and stromal breakdown.
Retrospective case-control study. Surgical pathology records between 1996 and 2005 at Beth Israel Deaconess Medical Center (Boston, Mass) were searched for endometrial samples with diagnoses containing the key words "menstrual" or "extensive breakdown." Hospital records for these women were parsed for demographics, clinical indications, and follow-up with rebiopsy within 6 months. Age cutoffs enriched the population for women at higher risk for carcinoma. A control group, consisting of 2 age-matched control patients for each test patient, was also studied; each control patient had an endometrial sample taken within a 6-month period and was not diagnosed with extensive breakdown, menstrual endometrium, or neoplasia on initial sampling.
Fifty-four cases were identified. The primary biopsy reports had benign descriptive diagnoses (ie, proliferative, secretory, polyp). Follow-up biopsies showed benign pathology in all cases and specific causes of bleeding--including polyp, leiomyoma, or endometritis--in 28 (52%) of 54. In the control group, neoplasia was found in 2 of the 108 follow-up biopsies. Only 5 other controls had specific diagnoses; all were polyps.
Extensive breakdown or menstrual-pattern endometrium may mask other specific benign pathologies but does not commonly mask cancer.