The management of endometrial hyperplasia: An evaluation of current practice

Abstract

Objective

To identify current management practices and evaluate subsequent outcomes of treatment for women diagnosed with endometrial hyperplasia.

Study design

All women with a histological diagnosis of endometrial hyperplasia at the Birmingham Women's Hospital were identified between October 1998 and September 2000. A retrospective case note review was performed for each woman using a standardised data abstraction sheet. Baseline characteristics including clinical presentation and treatment strategy were obtained. Results of subsequent endometrial tissue examinations were used to assess histological response to treatment and the need and indication for hysterectomy was used to assess clinical response.

Results

There were 351 women diagnosed with endometrial hyperplasia during the study period of which 84% presented with symptoms of abnormal uterine bleeding and 54% were postmenopausal. Complex endometrial hyperplasia was the most common diagnosis accounting for 60% of all cases. Eighty percent of women with atypical endometrial hyperplasia were treated by hysterectomy compared with 30% without evidence of cytological atypia (relative hysterectomy rate of 2.6, 95% CI 2.0–3.3). Hysterectomy was avoided in 138/172 (80%, 95% CI 74–86%) women managed conservatively during the study period. Overall 35/108 (36%, 95% CI 27–46%) of women managed conservatively had persistent or progressive disease identified (mean follow up 36 months). 20/143 (14%) women initially diagnosed with endometrial hyperplasia who subsequently underwent hysterectomy were found to have endometrial cancer, the majority of whom had been diagnosed with atypical disease (14/20, 70%).

Conclusion(s)

The majority of women with atypical endometrial hyperplasia were managed by hysterectomy and the substantial risk of diagnostic under-call supports this approach to treatment. In contrast, there is no consensus regarding the initial management of women with endometrial hyperplasia without cytological atypia.

Introduction

Endometrial hyperplasia is usually detected following investigation of women with abnormal uterine bleeding symptoms [1], [2], [3]. The diagnosis is made histologically, when disrupted, proliferative endometrium is identified. The female sex steroid estrogen stimulates endometrial proliferation in the normal menstrual cycle and a relative excess of this hormone compared to progesterone is thought to be the prime etiological factor [2], [4]. Endometrial hyperplasia is considered simple or complex according to the degree of architectural disruption seen. In addition, the condition is categorized according to the presence or absence of cytological atypia. Although endometrial hyperplasia has long been considered a precursor of endometrial cancer, oncogenic potential is though to be low in the absence of cytological atypia (<1–3%), but considerable when such changes are seen (30–50%) [5], [6], [7]. Atypical endometrial hyperplasia is also believed to arise de novo and in such situations is more likely to be a focal rather than global endometrial process [8].

The main aim of investigating women with abnormal uterine bleeding is to exclude endometrial hyperplasia and cancer. Endometrial hyperplasia is more common than endometrial cancer and affects a wider spectrum of women, i.e. both pre- and postmenopausal women [3], [9]. The traditional therapeutic approach for endometrial hyperplasia has involved short courses of high dose oral progestogens or hysterectomy [10], [11], [12]. However, the justification for such practices is questionable. For example, the tolerability and duration of high dose, unopposed, systemic progestogens are limited by acute and longer-term progestogenic side effects and there is scarce long-term effectiveness data to support their use. Newer, better-tolerated, medical approaches are now available and include the continuous use of systemic progestogens in combination with estrogen (e.g. combined oral contraceptive pill, hormone replacement therapy [13]) or local delivery of progestogens [14], [15], [16] (e.g. via the Mirena^® levonorgestrel releasing intrauterine system). Similarly, the widespread use of hysterectomy may represent overly aggressive management of endometrial hyperplasia in light of uncertainties over the natural history of the condition [2], [5], [6], [17] (i.e. spontaneous regression rates, malignant potential), false positive rates associated with endometrial sampling techniques [18] and the availability of potentially effective medical treatments to reverse the condition and abate symptoms.

In view of the controversies surrounding diagnosis and treatment of endometrial hyperplasia and the apparent lack of consensus regarding overall management, we undertook an observational study with the aim of identifying current management practices and subsequent outcomes of treatment for women diagnosed with endometrial hyperplasia.