Endometrial hyperplasia (EH) is a term which refers to the overgrowth or thickening of the endometrium, the innermost lining of the uterus. It is usually thought to be due to increased growth of cells as a result of exposure to unopposed estrogen, which promotes the growth of endometrial gland cells.

It is associated with conditions such as polycystic ovarian syndrome (PCOS), or obesity, in both of which estrogen levels are high. It is a potentially serious diagnosis because when the cells show atypical features, the EH may be a precursor or even a marker of the presence of uterine cancer.

Types of EH

or curettage (scraping the endometrium with a spoon-like instrument) is done to provide a tissue sample for examination under the microscope. However, it is still the case that many cases of actual cancer may be missed because of incomplete sampling, and only the final examination of the uterus following hysterectomy (removal of the uterus) shows the presence of an endometrial cancer.

The old WHO (1994) classification

• Simple hyperplasia:
  • without atypia 66% - Increased number of glands but regular glandular architecture
  • with atypia 2% - Crowded irregular glands
• Complex hyperplasia:
  • without atypia 11% - Simple hyperplasia with presence of cytologic atypia
  • with atypia 14% - Complex hyperplasia with cytologic atypia.

Defining the terms

Simple EH is characterized by uniform thickening of both gland cells and stroma, unlike that seen in normal uterine cycles, but without undue prominence of the gland cells and with rounded gland shapes.

Complex EH is characterized, on the other hand, by increased gland cell proliferation with lesser stromal involvement, leading to the crowding of glands. The glands themselves are of various shapes and sizes, and show marked and irregular branchings and buddings.

In both these cases, atypia may be present or absent. This term refers to abnormally high nuclear size with dense chromatin. It is often difficult to tell when simple hyperplasia ceases to be simple, so that this classification has proven to be far from reproducible.

In the original study (Kurman et al), simple hyperplasia had a 1% rate of progression to endometrial cancer, with the rates being 3%, 8% and 29% for complex, simple atypical and complex atypical EH respectively.

Another classification followed in Germany divided EH into cystic-glandular hyperplasia and adenomatous hyperplasia, with the latter being subdivided into grades I to III based on the presence of crowding and atypical changes in the nucleus.

Mutter classification

Later work by pathologists and clinicians led to the emergence of a third system (Mutter et al) which classifies EH as follows:

• Benign hyperplasia which corresponds to simple hyperplasia in the earlier classifications
• Endometrial intraepithelial neoplasia (EIN), in which glands are closely crowded with little stroma, the cells are abnormal, the lesion is more than 1 mm in diameter, and other invasive characteristics are not present. Other similar conditions must be excluded as well.

The advantages of this system are its better reproducibility and the close relationship between EIN and complex atypical EH in earlier systems, with respect to accurately predicting the risk of cancer development. This system is recommended by several expert bodies because of its prognostic value.

Revised WHO classification

The latest classification by the WHO (2014) attempts to simplify the diagnosis on microscopy, with greater accuracy and better management as a result. It also takes into account the presence of EIN in a high percentage of atypical EH. This is still the most widely-used classification in the world. It has only two categories, namely,

• Hyperplasia without atypia
• Atypical hyperplasia or EIN

The clinical advances reflected in this classification include the absence of genetic changes in EH without atypia, which is a wholly benign condition and capable of being completely reversed once the source of estrogen is removed and the patient is treated with adequate doses of progesterone. The rate of development of cancers in this group is low, about 1-3%, and usually accompanies long-term exposure to estrogen.

On the other hand, atypical EH is associated with cancerous mutations such as PAX2inactivation and mutations in PTEN, KRAS and CTNNB1 (β-catenin) genes, with micro satellite instability. It is associated, as already mentioned, with an extremely high risk of cancer of the endometrium in up to 60%. The treatment of both types is obviously quite different.

Atypical EH is diagnosed when the cells contain nuclear abnormalities which are characteristic of or suspicious of malignancy. These include:

• loss of polarity in the cells
• the loss of normal progression between the endometrial cell layers
• highly variable shapes and sizes of the cells
• round rather than columnar nuclei
• large nuclear diameter in comparison to cell size
• very dense nuclei with large nucleoli, and clumped chromosomal material rather than evenly dispersed
• highly eosinophilic cytoplasm which often draws the attention to the abnormal clump of cells

The incidence of EH without atypia is only about 5% in asymptomatic premenopausal women, while atypical hyperplasia occurs in about 1% in this group. EH occurs in about 10% of those who have abnormal uterine bleeding.

The most common form of endometrial cancer is endometrioid carcinoma, which shares the same risk factors as EH. It is usually the end-result of atypical EH and comprises about 85% of endometrial cancers. In fact, up to 60 of every 100 women with atypical EH already have actual endometrial cancers or will develop them within a few years, and so this diagnosis is best treated with a total hysterectomy, removing the whole of the uterus and the cervix. For this reason, it is extremely important to make an accurate diagnosis in this condition to provide the best form of management.