Ultrasound Procedure Detects Uterine Cancer And Disease

The study is published in the Nov. 4 issue of the Journal of American Medical Association.

The test, called an endovaginal ultrasound (EVUS), uses an ultrasound probe that is placed directly into the vagina to obtain a measurement of the uterine lining and detailed images of the uterus.

Researchers found that EVUS was able to identify 96 percent of uterine cancer and 92 percent of uterine disease in postmenopausal women who experience abnormal vaginal bleeding, an indicator of uterine problems.

"This is an excellent test in postmenopausal women who have vaginal bleeding, and it can accurately exclude significant underlying abnormalities, such as uterine cancer," said Rebecca Smith-Bindman, MD, UCSF assistant professor of radiology and lead author of the study. "If a woman's ultrasound results are normal, meaning that she has a thin uterine lining, then there is no need for subsequent invasive testing with a uterine biopsy."

She added that, although many gynecologists are knowledgeable about and may use EVUS in postmenopausal women with vaginal bleeding, few primary care physicians and generalists rely on the test because, until now, there has been a lack of sufficient data proving its efficacy. Smith-Bindman emphasized that physicians need to be thoroughly trained to perform an EVUS examination.

Most postmenopausal women with abnormal vaginal bleeding must undergo a uterine biopsy to determine if they have uterine disease or cancer, the fourth most common malignancy in women in the United States affecting more than 36,000 women and causing 6,300 deaths annually.

However, a standard biopsy, in which part of a woman's uterine tissue is removed to obtain a diagnosis, is uncomfortable, non-diagnostic, or cannot be performed in 2 to 28 percent of attempts, Smith-Bindman said.

Conversely, she added, an ultrasound test is painless, well tolerated, cost-effective, and has a higher accuracy rate for detecting uterine cancer than a biopsy because ultrasound imaging allows visualization of the entire uterine cavity, whereas the latter relies on blind sampling of the uterine tissue. In addition, EVUS eliminates the need for multiple unnecessary biopsies in postmenopausal women who continue to bleed although they have no serious uterine problems.

The ultrasound test is able to accurately measure the lining of a woman's uterus, which is an indicator of uterine cancer or other problems. According to Smith-Bindman, a woman whose uterine lining is greater than 5 mm thick is considered to be at risk of developing uterine cancer or disease, and therefore requires subsequent invasive tests such as a biopsy. However, if her uterine lining is less than 5 mm thick as measured by an ultrasound (0-5 mm is the normal thickness for postmenopausal women), then an invasive test is unnecessary.

During the study, UCSF researchers conducted a search using MEDLINE database of all the studies evaluating EVUS published from January 1966 through November 1996. Studies were included in the analysis if postmenopausal women who experienced vaginal bleeding underwent an EVUS prior to a uterine biopsy. Out of the 85 studies collected, a total of 35 English and non-English studies, including 5892 women, were analyzed.

The mean age of the women in the studies was 61 years and 94 percent experienced vaginal bleeding. Using a 5 mm threshold of uterine thickness to define an abnormal uterine measurement, 96 percent of women with cancer had an abnormal EVUS result, whereas 92 percent of women with uterine disease had an abnormal result. Hormonal replacement therapy, which often increases a woman's uterine lining, did not impact EVUS's ability to detect cancer, Smith-Bindman said.

However, the number of women without uterine problems who had an abnormal EVUS result (false positive test result), did vary by hormone replacement use. For instance, 23 percent of normal women who used hormone replacement therapy had an abnormal EVUS result; whereas 8 percent of normal women who did not use such drugs had an abnormal EVUS result.

Other researchers on the study include Richard Brand, PhD, UCSF professor of epidemiology and biostatistics; Vickie A. Feldstein, MD, UCSF assistant professor of radiology; Deborah Grady, MD, UCSF associate professor of medicine, epidemiology, and biostatistics; Karla Kerlikowske, MD, UCSF assistant professor of medicine, epidemiology, and biostatistics; Juergen Scheidler, MD, assistant professor of radiology in Heidelberg, Germany; Mark Segal, PhD, UCSF assistant professor of epidemiology and biostatistics; and Leslee Subak, MD, UCSF assistant professor of obstetrics and gynecology.