Page 35 - The diagnostic work-up of women with postmenopausal bleeding

they found a sensitivity of 98% and a specificity of 35%, giving a LR for a negative test
result of 0.06.Thus, a cut-off level of 3 mm reduced a pre-test probability of 10%
to a 0.7% post-test probability.The conclusion of this article was that the use ofTVS measurement of endometrial thickness remains justified, but with a recommended 2 cut-off level of 3 mm.
Outpatient endometrial sampling
Clark et al conducted a systematic review and meta-analysis to determine the diagnostic accuracy of outpatient endometrial sampling in detecting endometrial hyperplasia.22 Postmenopausal women were included in two studies, in which they represented 25% of the (combined) patient sample. In these two articles three different diagnostic devices had been used for endometrial sampling: Accurette®, Pipelle® and Vabra® aspirator.The pre-test probability of 14.3% was increased to a post-test probability for a positive result of 66.7% (95% CI 42.3–83.9%). In 34 of 149 women the endometrial sampling was unsuccessful, with a failure rate (sampling not possible for technical reasons) of 17% (25/149) and an inadequate sampling rate (not enough tissue obtained for a pathologic diagnosis) of 7% (9/124).This review demonstrated that endometrial sampling is moderately accurate in diagnosing (pre) malignant endometrial pathology. A positive test result was more accurate than a negative test result (sensitivity 91.9%, with a specificity of 99.7%). Clark et al concluded that the more clinically significant the endometrial pathology is, the better the diagnostic accuracy of outpatient endometrial sampling will be and, hence, the more clinically useful the test. Additional endometrial assessment should be undertaken with technical failure or inadequate sampling, especially if symptoms persist.22
Dijkhuizen et al performed a meta-analysis to assess the accuracy of endometrial sampling devices in the detection of endometrial cancer and atypical hyperplasia.23 With respect to the diagnosis of endometrial cancer, they identified seven studies that were limited to postmenopausal women.The pooled data from these studies showed a sensitivity of 95% and a specificity of 99.5%, giving a post-test probability for a positive test of 95.5%. Outpatient endometrial sampling therefore appeared to be a highly sensitive technique for diagnosing endometrial cancer.With regard to inadequate sampling (0 to 54% of cases in the studies they reviewed), they concluded that an inadequate sample is an indication for fur ther investigation, based on an ar ticle of Farrell et al which demonstrated that of those women for whom the result of the Pipelle was ‘insufficient’ 20% had uterine pathology after fur ther investigation, 3% with endometrial cancers.27
Diagnostic evaluation
33