The trusted source for
healthcare information and
ABSTRACT & COMMENTARY
Langer and colleagues compared the sensitivity and specificity of transvaginal ultrasonography and endometrial biopsy for the detection of endometrial disease in 448 postmenopausal women who received estrogen alone, cyclic or continuous estrogen-progesterone, or placebo for three years. Concurrent ultrasonographic and biopsy results were available for 577 examinations in the 448 women99% of whom were undergoing routine annual follow-up. Endometrial thickness was less than 5 mm in 45% of the examinations, 5-10 mm in 41%, more than 10 mm in 12%, and not measured in 2%. It was higher in the women receiving estrogen alone than in the other groups. Biopsy detected 11 cases of serious disease: one case of adenocarcinoma, two cases of atypical simple hyperplasia, and eight cases of complex hyperplasia. Biopsy also detected simple hyperplasia in 20 cases. At a threshold value of 5 mm for endometrial thickness, transvaginal ultrasonography had a positive predictive value of 9% for detecting any abnormality, with 90% sensitivity, 48% specificity, and a 99% negative predictive value. With this threshold, a biopsy would be indicated in more than half of the women, only 4% of whom had serious disease. Langer et al conclude that transvaginal ultrasonography has a poor positive predictive value but a high negative predictive value for detecting serious endometrial disease in asymptomatic postmenopausal women.
Endometrial cancer is the most common gynecologic malignancy, accounting for almost 40,000 cases annually in the United States. Each year, thousands of American women develop endometrial hyperplasia, a proportion of which is premalignant. Studies in the 1970s indicated that, compared with dilatation and curettage, office endometrial biopsy was approximately 95% accurate in predicting serious endometrial pathology.
But, endometrial biopsy is an invasive procedure associated with anxiety and discomfort. Proof of equivalence in detection of serious endometrial histology for a less invasive procedure with similar cost, such as transvaginal ultrasound, would be an advance. Unfortunately, in the present study, transvaginal ultrasound proved to have an unacceptably low positive predictive value. It was disappointing as a test for monitoring the safety of unopposed estrogen replacement therapy. Therefore, one cannot recommend its use for screening of asymptomatic women.
In an accompanying editorial, Fleischer, a radiologist whose area of interest is gynecologic ultrasonography, is considerably more upbeat about the use of transvaginal sonography in detecting endometrial pathology (Fleischer AC. N Engl J Med 1997;337:1839). He states that "transvaginal ultrasonography of the endometrium is a useful test in women presenting with unexplained vaginal bleeding and is also useful in monitoring the response of the endometrium to hormone replacement therapy."
He mainly emphasizes the problems with performance and interpretation of transvaginal ultrasonography, including the variation in threshold for normal endometrial thickness among heavier women and those receiving hormone replacement therapy, improper measurement of the endometrium, interobserver variability, and additional information that may be obtained from the test.
In my view, based on the information presented in this study and the editorial, further study is warranted in both asymptomatic women and women with vaginal bleeding. Until more convincing information becomes available, I will not personally use transvaginal ultrasonography for monitoring asymptomatic women on hormone replacement therapy.
In a recent study of transvaginal ultrasonography compared with endomtrial biopsy for detection of endometrial disease, the former had:
a. high negative and low positive predictive value.
b. high positive and low negative predictive value.
c. high positive and high negative predictive value.
d. low positive and low negative predictive value.
e. none of the above.