Endometrial Hyperplasia

• The current WHO classification is binary: hyperplasia without atypia, and atypical hyperplasia / endometrioid intraepithelial neoplasia (EIN). This replaced the older four-tier simple/complex ± atypia scheme, and atypia — not architectural complexity — carries the risk.
• Hyperplasia without atypia progresses to carcinoma in only a small percentage and is generally managed medically with progestins.
• Atypical hyperplasia / EIN carries a substantially higher progression risk, and a considerable proportion of patients already harbour concurrent carcinoma at hysterectomy — which is why hysterectomy is the standard treatment when childbearing is complete.
• Sources of unopposed oestrogen include obesity (peripheral aromatisation), chronic anovulation / PCOS, nulliparity, late menopause, tamoxifen, oestrogen-only hormone therapy in a woman with a uterus, and oestrogen-secreting tumours.

Endogenous overproduction (mechanism: peripheral or gonadal oestrogen excess)

• Obesity: adipose aromatase converts adrenal androstenedione to oestrone; obesity also lowers sex hormone–binding globulin, raising free oestradiol. The single most commonly tested — and modifiable — risk factor.
• Chronic anovulation / PCOS: no corpus luteum means no luteal progesterone, so the endometrium sits in an unopposed proliferative state month after month. Also seen in perimenopause and in hypothalamic-axis irregularity.
• Oestrogen-secreting tumours: granulosa cell tumour of the ovary (inhibin-positive, Call–Exner bodies) and ovarian thecoma. A postmenopausal woman with bleeding and an adnexal mass is the classic vignette.

Exogenous exposure (modifiable)

• Unopposed oestrogen therapy: systemic oestrogen given to a woman with an intact uterus. ACOG advises adding a progestin (or using a combined regimen) whenever the uterus is present.
• Tamoxifen: antagonist in breast, partial agonist in endometrium. ACOG (Committee Opinion on tamoxifen and uterine cancer) recommends evaluating any bleeding rather than screening asymptomatic users.

Non-modifiable host factors

• Age and reproductive history: increasing age, perimenopause/postmenopause, nulliparity, early menarche, late menopause — all increase lifetime unopposed-oestrogen exposure.
• Lynch syndrome (MMR gene mutation): markedly elevated lifetime endometrial cancer risk; endometrial cancer is frequently the sentinel malignancy, preceding colorectal cancer. Suspect in a young, non-obese patient or with a suggestive family history.
• Diabetes/insulin resistance: hyperinsulinaemia lowers SHBG and raises IGF-1, amplifying mitogenic signalling.

Protective (examiners test these as the "least likely" option)

• Combined oral contraceptives, progestin-containing IUD, multiparity, and prolonged lactation — all deliver progestin exposure or suppress ovulation.

• Architectural change: glands proliferate faster than stroma, so the gland-to-stroma ratio rises and glands become crowded, branched and cystically dilated. This is hyperplasia without atypia — polyclonal, diffuse, and still hormonally responsive, which is why progestin alone usually reverses it.
• Why it bleeds: the thickened endometrium is never shed in a coordinated way. It outgrows its spiral-artery supply, undergoes focal ischaemic necrosis and asynchronous breakdown, and sloughs piecemeal — producing irregular, prolonged, unpredictable bleeding rather than a normal cyclic period.
• Genetic progression: sustained mitotic pressure accumulates replication errors. PTEN inactivation is the earliest and most characteristic hit (PTEN-null glands are detectable before morphologic change), followed by PAX2 loss, KRAS mutation, and MLH1 promoter hypermethylation with microsatellite instability.
• The clonal step: one gland population escapes and expands as a monoclonal lesion with cytological atypia — nuclear enlargement, rounding, loss of polarity, vesicular chromatin, prominent nucleoli. This is atypical hyperplasia / EIN, a true neoplasm rather than a hormonal response, and it need no longer depend on oestrogen.
• Invasion: further loss of tumour suppressors permits stromal invasion, giving type I (endometrioid) endometrial carcinoma — low-grade, oestrogen-driven, usually diagnosed early because it bleeds. Type II (serous, p53-mutant) arises in atrophic endometrium and is not part of this sequence.

Symptoms

• Abnormal uterine bleeding — the presenting complaint in the overwhelming majority. Mechanism: asynchronous ischaemic breakdown of an over-thick, unshed endometrium.
• Premenopausal: heavy or prolonged menses, intermenstrual bleeding, or oligomenorrhoea punctuated by episodes of flooding after months of amenorrhoea.
• Postmenopausal bleeding: any bleeding after menopause is abnormal and must be evaluated — the single most important trigger for endometrial sampling.
• Symptomatic anaemia: fatigue, dyspnoea on exertion, pallor and conjunctival rim pallor from chronic blood loss.
• Infertility or subfertility in the anovulatory/PCOS patient — the same absent ovulation that produces the hyperplasia prevents conception.
• Asymptomatic detection: an incidentally thickened endometrial stripe on pelvic imaging, or atypical glandular cells (AGC) on cervical cytology, which per ASCCP management guidance mandates endometrial sampling in women 35 and older or with risk factors for endometrial neoplasia.

Physical findings — deliberately unimpressive

• Normal-sized, non-tender uterus in most patients; hyperplasia is a mucosal process and does not enlarge the uterus. A bulky uterus should redirect you toward leiomyomata or adenomyosis.
• Signs of hyperandrogenism/insulin resistance: hirsutism, acne, acanthosis nigricans, central adiposity — pointing to PCOS as the source of chronic anovulation.
• An adnexal mass with signs of oestrogen excess (breast tenderness, postmenopausal bleeding) suggests a granulosa cell tumour.
• Pelvic pain, a fixed mass, or ascites is not hyperplasia — that picture implies invasive carcinoma and warrants urgent oncologic evaluation.

Step 1 — decide who needs sampling

• Any postmenopausal bleeding warrants evaluation (ACOG).
• Abnormal uterine bleeding at age 45 or older: proceed directly to endometrial sampling (ACOG Practice Bulletin on AUB in reproductive-aged women).
• Under 45: sample if there is unopposed-oestrogen exposure (obesity, PCOS/chronic anovulation, tamoxifen), failed medical management, persistent bleeding, or Lynch syndrome.
• Exclude pregnancy with hCG in reproductive-aged patients before anything else.

Step 2 — initial tests

• Transvaginal ultrasound: measures the endometrial stripe (double-layer thickness). In a postmenopausal woman with bleeding, a stripe of 4 mm or less has a very high negative predictive value for carcinoma and, per ACOG, makes biopsy unnecessary initially; >4 mm requires tissue sampling. This threshold does not apply to premenopausal women, whose stripe varies with cycle phase, nor to asymptomatic thickening found incidentally.
• CBC for anaemia; TSH/prolactin if anovulation is unexplained.

Step 3 — tissue is the diagnosis

• Office endometrial biopsy (Pipelle) is the initial diagnostic test of choice — sensitive, cheap, no anaesthesia. Its weakness is sampling error: it samples blindly and can miss focal lesions and polyps.
• Hysteroscopy with dilation and curettage is the gold standard and the mandatory next step when the office biopsy is non-diagnostic/insufficient, when bleeding persists despite a benign biopsy, when cervical stenosis prevents sampling, or when a focal lesion is seen on saline infusion sonohysterography.

Step 4 — classify the histology (WHO binary scheme)

• Hyperplasia without atypia: crowded glands, increased gland-to-stroma ratio, but nuclei resemble background endometrium.
• Atypical hyperplasia / EIN: the EIN diagnostic criteria require gland crowding with gland area exceeding stroma, cytology differing from the surrounding background endometrium, a lesion of at least ~1 mm, and exclusion of mimics and frank carcinoma. Loss of PTEN or PAX2 staining supports a clonal lesion.
• Any atypical diagnosis mandates gynaecologic oncology involvement, since occult carcinoma is common.

Hyperplasia WITHOUT atypia — medical, progestin-based (ACOG/SGO)

• Levonorgestrel-releasing intrauterine system (52 mg LNG-IUS): preferred first line. Delivers high local progestin, causing glandular atrophy and stromal decidualisation with minimal systemic exposure; regression rates exceed those of oral therapy.
• Systemic progestins: medroxyprogesterone acetate (cyclic or continuous) or megestrol acetate where an IUD is declined or contraindicated.
• Address the driver: weight loss, ovulation induction or cyclic progestin in PCOS, and adding a progestin to any oestrogen therapy in a woman with a uterus.
• Surveillance: repeat endometrial sampling at roughly 3–6-month intervals until two consecutive negative biopsies; persistence or progression on therapy is an indication for hysterectomy.

Atypical hyperplasia / EIN — surgical (ACOG Committee Opinion on EIN; NCCN Uterine Neoplasms)

• Total hysterectomy is definitive, because a substantial share of these patients already harbour carcinoma. Bilateral salpingo-oophorectomy is individualised by menopausal status.
• Do not perform supracervical hysterectomy, morcellation, or endometrial ablation — each risks leaving or disseminating occult cancer.
• Fertility-sparing alternative (strongly desired childbearing, cancer reasonably excluded, oncology co-management): LNG-IUS, sometimes with a systemic progestin, plus serial endometrial biopsy every ~3 months, then hysterectomy after childbearing.

Of the disease

• Progression to endometrioid (type I) endometrial carcinoma — the complication that defines the entity. Low risk with hyperplasia without atypia; substantially higher with atypical hyperplasia/EIN. Signalled by persistent bleeding despite progestin therapy or a biopsy showing atypia on follow-up.
• Concurrent occult carcinoma at hysterectomy in a large minority of patients whose biopsy showed only atypical hyperplasia — the reason biopsy grading cannot be trusted to exclude cancer and why oncology involvement precedes surgery.
• Iron-deficiency anaemia from chronic blood loss: microcytic indices, low ferritin.
• Acute haemorrhagic AUB with haemodynamic instability — an emergency. Tachycardia, hypotension and orthostasis in a patient passing clots; resuscitate, transfuse, and proceed to hormonal control, tamponade, or D&C.
• Infertility, from the underlying chronic anovulation, and reduced implantation on a hyperplastic endometrium.

Of medical therapy

• Progestin side effects: breakthrough bleeding, bloating, breast tenderness, weight gain and mood change; these drive nonadherence, which is itself a treatment failure risk.
• LNG-IUS complications: expulsion (recurrent bleeding, missing strings), and uterine perforation at insertion — new pelvic pain with a non-palpable device requires imaging.
• Treatment failure or relapse after progestin discontinuation, especially if obesity and anovulation are unaddressed — mandating the scheduled surveillance biopsies rather than symptom-based follow-up.
• Missed cancer during fertility-sparing management, the reason serial sampling every few months is non-negotiable.

Of surgery

• Standard hysterectomy morbidity: haemorrhage, infection, ureteric or bladder injury, venous thromboembolism, and adhesive bowel obstruction. VTE presenting as sudden dyspnoea and hypoxaemia postoperatively is an emergency.
• Surgical menopause if oophorectomy is performed in a premenopausal woman: vasomotor symptoms, genitourinary atrophy, and accelerated bone loss.
• Loss of fertility — permanent, and the reason the decision is framed around completed childbearing.

• Unopposed oestrogen is the unifying mechanism — obesity (peripheral aromatisation), chronic anovulation/PCOS, nulliparity, late menopause, tamoxifen, oestrogen-only therapy with a uterus, and granulosa cell tumour. Progestin, in any form, is protective.
• Atypia — not architectural complexity — decides management. The WHO scheme is binary: hyperplasia without atypia (progestin) versus atypical hyperplasia/EIN (hysterectomy). Do not pick management based on "simple versus complex."
• Best next step for postmenopausal bleeding: evaluate every episode. Transvaginal ultrasound with a stripe ≤4 mm is reassuring per ACOG; >4 mm, or any persistent bleeding, requires endometrial biopsy. A thin stripe never excuses ignoring recurrent bleeding.
• Best next step when the office biopsy is insufficient, or bleeding persists despite benign histology: hysteroscopy with D&C, not repeat empiric hormones — Pipelle sampling misses focal lesions.
• Sample any woman 45 or older with abnormal uterine bleeding, and younger women with unopposed-oestrogen risk factors or failed medical therapy (ACOG).
• Atypical hyperplasia frequently coexists with invasive carcinoma already present in the uterus — hence total hysterectomy, and never supracervical hysterectomy, morcellation, or endometrial ablation (ablation destroys the tissue you need to survey).
• The association examiners love: Lynch syndrome, where endometrial cancer is often the sentinel malignancy — think of it in a young, thin patient with a strong family cancer history.
• Common distractors: treating an obese anovulatory woman's irregular bleeding with unopposed oestrogen (contraindicated); attributing postmenopausal bleeding to atrophy without sampling; and confusing this oestrogen-driven type I pathway with type II serous carcinoma, which is p53-mutant, arises in atrophic endometrium, and is unrelated to hyperplasia.