Evaluation of ambulatory diagnosis of abnormal uterine bleeding
In the ambulatory assessment of women with abnormal uterine bleeding, the main aim is to reach a diagnosis and thereby allow appropriate treatment. Excluding endometrial pathology, particularly carcinoma and hyperplasia, is of paramount importance. However there is no consensus as to which set of investigations should be used (hysteroscopy, ultrasonography and endometrial sampling). There is a lack of good quality research evidence on the accuracy of the various diagnostic tests in predicting endometrial lesions. This dearth of relevant evidence prompted the research presented in this thesis to address and answer the following research questions: • What is the accuracy of outpatient miniature hysteroscopy / ultrasonography in the identification of premalignant and malignant endometrial lesions? • What is the relative significance of hysteroscopic and ultrasonographic evidence of endometrial atrophy in relation to an insufficient sample on outpatient endometrial biopsy? • What is the risk of premalignant and malignant pathology among endometrial polyps? What is the significance of various risk factors associated with endometrial polyps? • What is the feasibility of multivariable analysis to evaluate combinations of diagnostic tests in the diagnosis of endometrial disease? Findings and Conclusions Positive hhysteroscopy is accurate in ruling in endometrial cancer and hyperplasia (the LR was 51.1 (95% CI 7.9-326.9). Using endometrial thickness >4mm at ultrasound scan, ultrasound is accurate in ruling out endometrial cancer and hyperplasia (the LR was 0.14 (95%CI 0.02-0.64). Insufficient sample on endometrial biopsy was more common among cases with hysteroscopic finding of endometrial atrophy (adjusted OR= 4.79, 95% CI 1.05-21.91, p=0.04) and less common among cases with sonographic endometrial thickness above 5mm (adjusted OR=0.19, 95% CI 0.07-0.53, p=0.001). Therefore insufficient sample may be considered a substitute to absence of pathology provided the hysteroscopic and sonographic endometrial assessment is consistent with endometrial atrophy. Hyperplasia was more frequent in endometrial specimens with polyps than in those without (9.7% vs 4.8%, OR=2.1, 95% CI 0.85-5.2), but the rate of carcinoma in the two groups was not statistically different (4.8% vs 3.2%, OR=1.5, 95% CI 0.46-5.0). Tamoxifen treatment was associated with endometrial polyps (adjusted OR= 8.16, 95% CI 2.01-33.2, p=0.003) but hormone replacement therapy was not (adjusted OR=1.35, 95% CI 0.65 – 2.81, p=0.42). The true clinical value of a test lies in the added information over and above what was already known from the history and examination. It is feasible to develop a stepwise multivariable analytic approach to explore the added value of tests (hysteroscopy or ultrasonography) over and above history when predicting endometrial hyperplasia or cancer. This analytic strategy needs to be applied in larger datasets to draw clinical conclusions.