ORLANDO – An endometrial echo complex thinner than 5 mm on ultrasound during initial evaluation does not necessarily rule out malignancy, according to a retrospective study of 250 postmenopausal women with biopsy-confirmed endometrial cancer.
The American College of Obstetricians and Gynecologists currently recommends no further diagnostic procedures in a woman with postmenopausal bleeding and an endometrial echo complex (EEC) less than or equal to 4 mm because the risk of malignancy is low. The committee opinion cited a number of reports comparing transvaginal ultrasonography with endometrial sampling that consistently found that an endometrial thickness of less than or equal to 4-5 mm in patients with postmenopausal bleeding reliably excluded endometrial cancer (ACOG Committee Opinion 440, August 2009).
"Although an EEC less than 5 mm may be reassuring, it does not rule out endometrial cancer and cannot supplant definitive histologic evaluation," Dr. Uma Chandavarkar said at the annual meeting of the Society of Gynecologic Oncologists.
A total of 40% of the 250 women had an EEC less than 5 mm in the study, regardless of histology type, said Dr. Chandavarkar, a fellow in gynecologic oncology at the University of Southern California in Los Angeles.
Interestingly, a greater percentage of patients with the more-aggressive type II endometrial cancers had an endometrial stripe less than 5 mm, 46% of 162 women, compared with 29% of 88 patients with type I disease. This was a statistically significant difference.
"Transvaginal ultrasound cannot replace definitive histologic evaluation," Dr. Chandavarkar said. "We recommend these patients are counseled that endometrial cancer may be missed without a biopsy."
"Some of the best studies you can do are often the simplest ones, and this study fits the criteria," said invited study discussant Dr. Richard Barakat, chair of gynecologic surgery at Memorial Sloan-Kettering Cancer Center, New York.
"It is important to do studies like this," Dr. Barakat continued, because "others have said women who are bleeding with an endometrial stripe less than 4 mm do not need an endometrial biopsy."
In the current study, radiologists performed the preoperative, pelvic ultrasound examination 3 months or less prior to hysterectomy with biopsy evaluation. They found no statistically significant differences by histology classification for ancillary ultrasound characteristics, including adnexal masses, myomatous masses, or free pelvic fluid.
Patients were treated from 1999 to 2009 at the University of Southern California/Los Angeles County Medical Center and the USC Norris Cancer Center. Patient demographics and clinical factors did not differ significantly by histology type. For example, mean patient age at time of endometrial cancer diagnosis was 59 years for type I cancer vs. 60 years for type II; mean age of menopause was 50 years and 51 years, respectively; and mean body mass index was 32 kg/m2 and 31 kg/m2, respectively. There likewise were no significant differences in gravidity, parity, race, or use of hormone therapy between groups. One exception was a significantly longer duration of hormone therapy among women diagnosed with type II vs. type I endometrial cancer, Dr. Chandavarkar said.
The cutoff for concern remains controversial.
A recent meta-analysis suggests the cutoff to rule out endometrial cancer should be 3 mm or less for women with postmenopausal bleeding (Obstet. Gynecol. 2010;116;160-7). However, their analytical methods were criticized in an editorial by Dr. Linda R. Duska, a gynecologic oncologist at the University of Virginia Health System in Charlottesville. Dr. Duska said the preponderance of data still supports an endometrial stripe thicker than 4 mm for endometrial biopsy sampling.
Automatic sampling of the endometrium should be discouraged, Dr. Steven Goldstein said in a review article (Obstet. Gynecol. 2010;116:168-76), citing an "extremely high" 10%-17% rate of incidental thick EEC findings in women without postmenopausal bleeding. Dr. Goldstein is professor of obstetrics and gynecology at New York University Medical Center.
In the current study, a relatively large patient cohort and the close timing of the ultrasound examination and biopsy were strengths of the study, Dr. Barakat said. In addition, correlation between EEC and type of endometrial cancer was another plus.
A retrospective design, EEC thickness data extracted from ultrasound reports over 10 years, and the criteria used to distinguish type 1 from type 2 cancers are potential limitations, he said.
Dr. Chandavarkar and Dr. Barakat said that they had no relevant financial disclosures.