PCOS and Hyperandrogenism: Treating by the Problem, Not the Diagnosis

What PCOS actually is

A diagnosis of exclusion built from three features, of which you need only two. PCOS is defined by the Rotterdam criteria: a woman needs two of three features — irregular or absent ovulation, clinical or biochemical hyperandrogenism (hirsutism, acne, or a raised testosterone), and polycystic ovaries on ultrasound — once other causes of these findings have been excluded. Note the trap in the name: the "cysts" are not true cysts but arrested small follicles, and you do not need them to make the diagnosis. Underneath sits a self-reinforcing hormonal loop. Many women with PCOS have insulin resistance, so the pancreas pushes out more insulin to compensate (hyperinsulinaemia). That excess insulin does two unhelpful things: it drives the ovarian theca cells to make more androgens, and it lowers the liver's production of sex-hormone-binding globulin (SHBG), so more of those androgens circulate free and active.

On top of the insulin story, the pituitary drive is disordered: LH tends to run high relative to FSH. The high LH further stokes ovarian androgen production, while the relatively low FSH means no single follicle is selected to mature and ovulate — so follicles stall as the small "cysts" seen on the scan. The three headline consequences all fall out of this: anovulation (irregular or absent cycles), hyperandrogenism (hirsutism and acne), and the metabolic burden of insulin resistance. The molecular biology of these hormones — how androgens, oestrogen and the hypothalamic–pituitary–gonadal (HPG) axis actually work — belongs to the Endocrine section; here we take that physiology as given and act on it.

Problem 1 — irregular cycles and protecting the endometrium

The danger of a missed period is not the bleeding you don't see — it's the lining that keeps growing. In a normal cycle, ovulation produces progesterone, which matures the endometrium and then, when it falls, sheds it in a clean withdrawal bleed. A woman who does not ovulate makes oestrogen but little or no progesterone, so the endometrium is stimulated by oestrogen month after month with nothing to oppose it. Unopposed oestrogen drives the lining to thicken and, over years, raises the risk of endometrial hyperplasia and endometrial cancer. So the goal in a woman not currently seeking pregnancy is to guarantee a regular shedding of the lining — at least a bleed every few months. The tools: a combined oral contraceptive (which supplies a progestogen and imposes a regular withdrawal bleed), a cyclical oral progestogen for part of each month, or the levonorgestrel intrauterine system (IUS), which keeps the endometrium thin locally and is an elegant one-device solution. This is the same endometrial-protection logic taught in the Contraception chapter, applied here for a non-contraceptive reason.

Problem 2 — hirsutism and acne (hyperandrogenism)

The combined oral contraceptive is first-line here too, and for a satisfying mechanistic reason: the oestrogen component raises hepatic SHBG, which mops up free testosterone, and the progestogen suppresses LH-driven ovarian androgen production. Less free androgen means less signal reaching the hair follicle and sebaceous gland. When the pill alone is not enough after several months, add an anti-androgen. Spironolactone is the workhorse — beyond its aldosterone-blocking diuretic role it also blocks the androgen receptor. Cyproterone acetate is a potent anti-androgen (often combined with oestrogen). Finasteride, a 5α-reductase inhibitor, blocks conversion of testosterone to the more potent dihydrotestosterone (DHT) at the follicle. Because all of these can feminise a male fetus, an anti-androgen must never be used without reliable contraception. For facial hair specifically, topical eflornithine slows the growth of the hair itself. These agents overlap heavily with the Dermatology section, where hirsutism and acne are worked up in full.

Problem 3 — insulin resistance and metabolic health

PCOS is as much a metabolic diagnosis as a gynaecological one. Because insulin resistance sits at the root, weight management is the single most effective intervention across every domain — even modest weight loss can restore ovulation, lighten hirsutism, and improve the metabolic profile. Pharmacologically, metformin is the mainstay: by improving insulin sensitivity it lowers the hyperinsulinaemia that fuels ovarian androgen output, and can help restore regular cycles and ovulation. Its use in PCOS is largely off-label but long-established and widely accepted. It also nudges the metabolic profile in the right direction. Where weight is the dominant concern, the GLP-1 receptor agonists (as used in the Endocrine and obesity chapters) are increasingly used for their substantial weight loss. Crucially, women with PCOS carry a raised lifetime risk of type 2 diabetes and the metabolic syndrome, so screening for impaired glucose tolerance and cardiovascular risk factors is part of care, not an optional extra — this is where PCOS management overlaps directly with the Endocrine section's approach to insulin resistance and diabetes.

Problem 4 — infertility and ovulation induction

When the goal flips to pregnancy, the strategy inverts entirely: instead of suppressing the ovary with a contraceptive, you now want to coax it into releasing one egg. Weight optimisation comes first. The first-line drug is now letrozole, an aromatase inhibitor: by blocking the conversion of androgens to oestrogen, it lowers oestrogen feedback to the pituitary, which responds by raising FSH and driving a follicle to mature. Trials show letrozole achieves higher live-birth rates than the older clomifene in PCOS, which is why it has moved to first-line. Clomifene, a selective oestrogen receptor modulator that blocks oestrogen feedback at the hypothalamus to raise FSH/LH, remains an alternative, and metformin may be used as an adjunct to improve ovulation. If oral agents fail, injectable gonadotropins directly stimulate the ovary — powerful but carrying the real danger of ovarian hyperstimulation syndrome (OHSS), where too many follicles respond at once. Ovulation induction, monitoring, and the OHSS risk are covered in depth in the Fertility chapter; the load-bearing point here is simply that combined contraception and ovulation induction are opposite ends of the same menu, chosen by whether the woman wants to prevent or achieve a pregnancy.