Vaginal ultrasound as an initial test in postmenopausal vaginal bleeding
Vaginal ultrasound as an initial test in postmenopausal vaginal bleeding
Abstract & Commentary
Synopsis: In postmenopausal women with vaginal bleeding, vaginal ultrasound is highly sensitive, but less specific, for identifying those women with potentially serious disease.
Source: Smith-Bindman R, et al. JAMA 1998;280: 1510-1517.
Postmenopausal women commonly present with vaginal bleeding, prompting many clinicians to ask whether noninvasive testing may be recommended as the first step in workup. Smith-Bindman and colleagues from the University of California, San Francisco, conducted an exhaustive meta-analysis of prospective studies that collected endovaginal measurements of endometrial thickness prior to sampling endometrial tissue in postmenopausal women with vaginal bleeding. Thirty-five studies including a total of 5892 women met the review’s inclusion criteria.
When Smith-Bindman et al used 5 mm as the threshold for abnormal endometrial thickness (encompassing the width of both the anterior and posterior endometrial walls), vaginal ultrasound was found to be 96% sensitive for endometrial carcinoma (95%; CI, 94%-98%) and 92% sensitive for endometrial disease (cancer, polyp, or atypical hyperplasia; 95%; CI, 90%-93%). These sensitivities did not vary with hormone replacement status. If a postmenopausal woman with vaginal bleeding had a negative endovaginal ultrasound, her pretest probability of endometrial cancer was reduced by 90% regardless of hormone use. Ultrasound was 77% specific for women who were using hormone replacement (95%; CI, 75%-79%), and 92% specific for those who were not (95%; CI, 90%-94%).
Comment by Elizabeth Morrison, MD, MSEd
Some postmenopausal women with vaginal bleeding cannot or do not wish to undergo endometrial sampling. In others, sampling is attempted but is nondiagnostic. This helpful meta-analysis by a multidisciplinary research team clarifies the role of endovaginal ultrasound for these women. Clinicians may choose to use ultrasound to exclude many postmenopausal women who do not require endometrial biopsy, as ultrasound is a highly sensitive screen for significant endometrial disease.
The meta-analysis used sound methods to explore this important question. Smith-Bindman et al explicitly stated their comprehensive strategy for selecting relevant prospective trials so that results could be combined appropriately. A test for homogeneity was done, and validity was assessed by two independent reviewers.
The results are compelling. As Smith-Bindman et al state, "For a postmenopausal woman with vaginal bleeding with a 10% pretest probability of endometrial cancer, her probability of cancer is 1% following a normal endovaginal ultrasound result." This is not to say that all clinicians will wish to use ultrasound for screening all such patients. Some physicians will reasonably choose endometrial biopsy, hysteroscopy, or other invasive techniques for patients felt to be at moderate to high risk of endometrial cancer or atypia. Endometrial biopsy and ultrasound had similar costs in a recent cost analysis,1 so cost need not be a factor in the decision.
It is interesting that in this meta-analysis, sensitivity and specificity were not improved when 4 mm rather than 5 mm was used as an endometrial stripe cut-off. Using a 3 mm cut-off yielded excellent sensitivity and specificity but would be impractical because relatively few women have an endometrium less than 3 mm thick. Using a 6 mm cut-off yielded sensitivity and specificity below 90% for endometrial disease. Future studies may further illuminate the optimal threshold value for endometrial thickness on ultrasound. Meanwhile, this meta-analysis will help gynecologists and family physicians use vaginal ultrasound more confidently in the workup of postmenopausal women with vaginal bleeding.
Reference
1. Weber AM, et al. Am J Obstet Gynecol 1997;177: 924-929.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.