Page 39 - The diagnostic work-up of women with postmenopausal bleeding

the use of outpatient hysteroscopy), as well as doctor and patient preferences.The
preferences of doctors in relation to diagnostic procedures for endometrial cancer
in women with PMB have not been investigated. Furthermore, guidelines need to
meet the expectations of the patients; most women want to rule out endometrial 2 cancer with a certainty of 100% and they are prepared to undergo rather invasive
and painful diagnostic tests in order to achieve this.31 However, a post-test probability of 0.0% seems virtually impossible and one should also keep in mind that the risk of endometrial cancer in a population of asymptomatic postmenopausal women is reported to be 0.2%.32
Clark et al determined the most cost-effective strategy for diagnosing endometrial cancer.29 They constructed a decision model and evaluated 12 different strategies for the initial investigation of PMB.With a cancer probability of 10%16,17 the strategy withTVS as the initial test with a cut-off of 4 mm followed by endometrial sampling was most cost-effective. Unfortunately, a cut-off value of 3 mm was not considered in their evaluation. More importantly, in this decision model, the assumptions made regarding test accuracy were based on the available systematic reviews. Systematic literature reviews in diagnostic research report the accuracy of tests, and thereby assist clinicians in their decision-making. However, there are limitations to this approach, as the analysis of such data often does not allow reviewers to explore the diagnostic information gained from combinations of tests. In clinical practice, tests are commonly combined in diagnostic sequences and disease probabilities are usually estimated in a hierarchical manner, first combining information from the history and examination, followed by additional information obtained from other diagnostic procedures (e.g.TVS, endometrial sampling). Studies of test accuracy often do not take this clinical paradigm into account, but tend to report test results in isolation and disregard the history and examination. In addition, they usually analyse a single test at a time, without taking into account of what is known from previous testing.
There is considerable variability in endometrial thickness and the likelihood of endometrial cancer across women. Individual patient characteristics, including age, time since menopause, obesity, hypertension, diabetes mellitus and reproductive factors, are associated with a higher prevalence of endometrial cancer.33-38 However, current policy is not based on these risk factors,but only on endometrial thickness.7,10-13 Breijer et al developed an algorithm for diagnostic pathways in women with PMB.39 This algorithm includes the calculation of the pre-test probability of endometrial cancer based on individual patient characteristics and the diagnostic approach to benign pathology, both of which require further research.
Diagnostic evaluation
37