Sonography in Postmenopausal Bleeding

ostmenopausal bleeding is a serious and not uncommon clinical gynecologic occurrence that mandates evaluation. In fact, classic teaching has labeled postmenopausal bleeding as “endometrial cancer until proven otherwise.” Studies indicate that the incidence of malignancy in such patients ranges from 1% to 14%, and obviously this rate will vary depending on the years since menopause and the classic risk of factors such as obesity, hypertension, diabetes, and low parity.1 Transvaginal sonography can and should be considered as a first-line approach to this clinical problem because of the extremely high negative predictive value of a thin distinct endometrial echo when adequately visualized (Table 1).2–6 The overall risk of malignancy from these large prospective trials in postmenopausal women with bleeding is 1 in 917. Clearly the thinner the “cutoff” used, the fewer the cancers that will slip through the cracks, but there will be more patients needing alternative evaluation. The American College of Obstetricians and Gynecologists has opined7 that in postmenopausal women with bleeding, when present, “a thin distinct endometrial echo on transvaginal ultrasonography 4 mm or less has a risk of malignancy of 1 in 917, and therefore endometrial biopsy is not required” (Figures 1 and 2). The most likely diagnosis in such cases is an atrophic endometrium. There are some important clinical realities to this approach, however. Not all uteri lend themselves to a meaningful sonographic examination yielding an adequate depiction of the endometrial cavity (Figures 3 and 4). Previous surgery, coexisting leiomyomas, axial orientation, marked obesity, and adenomyosis can all result in an inability to find a reliable endometrial echo. In such cases, fluid enhancement by saline infusion sonohysterography will easily highlight the endometrial cavity. Sonohysterography should be thought of as a subset of transvaginal sonography to be used when an endometrial echo is not well visualized or is not thin and distinct. When spontaneous endometrial fluid is present, it may be thought of as a “naturally occurring” sonohysterogram. If the endometrium surrounding the fluid is thin (<2 mm) on each side and symmetric, it is compatible with atrophy.8 Sonohysterography, when needed, is a relatively simple, inexpensive, and well-tolerated office procedure. In those cases of inadequate visualization, it will yield an excellent depiction of the endometrial thickness. In those cases of a thick endometrial echo, it will distinguish focal from global lesions (Figures 5 and 6). Why is this capability important? Blind endometrial sampling has been and, unfortunately for many clinicians, is still the first “go-to” method for endometrial evaluation in patients with postmenopausal bleeding, and when positive for cancer or even atypical complex hyperplasia, Steven R. Goldstein, MD