The PCOS and insulin resistance connection

PCOS is diagnosed by what appears on an ultrasound and a hormone panel: polycystic-appearing ovaries, irregular cycles, and androgen excess (acne, facial hair, hair thinning on the scalp). But that diagnostic picture is the result. It is rarely the cause.

In the overwhelming majority of PCOS cases we see - and the published literature is now closely in agreement - the driver underneath the hormonal picture is metabolic. Specifically: insulin resistance. The ovary is not the first organ to break. The ovary is responding to a whole-body metabolic signal that has been running hot for years.

If you have been told you have PCOS and the standard advice has been "take the pill for your cycle" or "lose some weight and try metformin," you have been offered a treatment plan for the symptoms, not the cause. This article walks through why insulin resistance is the engine of most PCOS, what the science actually supports, what the workup should include, and what a genuine reversal looks like - with three real cases.

The biology: why the ovary responds to insulin

Insulin is a growth and storage hormone. When insulin is high - as it is in insulin resistance, even when glucose looks normal - it drives two ovarian effects that produce the PCOS phenotype.

First, it stimulates the ovary to produce more androgens (testosterone, DHEA-sulphate). This is a direct effect at the level of the theca cells, which make androgens in response to luteinising hormone. Insulin amplifies that signal. The result is the classic PCOS picture: jawline and chin acne, hair along the upper lip and jaw, scalp hair thinning at the crown or temples, and in more severe cases, deeper changes.

Second, it lowers a protein called sex hormone binding globulin (SHBG). SHBG is the carrier protein that binds testosterone in the blood. When SHBG falls, free (biologically active) testosterone rises - even if total testosterone is only borderline. This is why many PCOS patients have "near-normal" total testosterone and still have full androgen-excess symptoms. The free fraction is what the skin, hair follicle, and ovary actually see.

At the same time, high insulin disrupts the ovulatory cycle itself. The delicate wave pattern of LH, FSH, oestrogen, and progesterone needed to ovulate is destabilised. Cycles lengthen. Ovulation becomes inconsistent or stops. Multiple follicles accumulate - the "polycystic" appearance on ultrasound is not actual cysts but un-ovulated follicles frozen at an early stage.

So the ovary, the skin, and the cycle are all downstream readers of the same single upstream message: insulin is too high.

The evidence

Two decades of research converge on this model:

• 70-80% of women with PCOS have measurable insulin resistance on proper testing
• Metformin improves ovulation in a substantial share of PCOS patients - not because PCOS is a diabetes, but because metformin reduces insulin
• Inositol (myo-inositol + D-chiro-inositol), which works directly on insulin signalling in the ovary, has strong clinical evidence for restoring ovulation and reducing androgens
• Lifestyle interventions that target insulin resistance - resistance training, lower-insulin-demand food patterns, sleep - produce cycle restoration and symptom reduction at rates comparable to or exceeding oral contraceptives, without side effects
• Women with PCOS have substantially higher lifetime risk of Type 2 diabetes, fatty liver, and cardiovascular disease - because they are, biologically, on an insulin-resistance trajectory that the ovary happens to reveal early

PCOS is not a disease of the ovary that happens to have metabolic complications. It is a metabolic disease that happens to express itself through the ovary in a pre-menopausal woman.

Why "lose weight" is the wrong advice

This is the advice every PCOS patient has heard, often delivered briskly. It is technically not wrong - weight reduction of 5-10% improves PCOS meaningfully in overweight patients. But it fails as advice for three reasons.

First, it is impossible for many PCOS patients to lose weight without first correcting insulin. High insulin actively drives fat storage and hunger. Telling an insulin-resistant woman to eat less and exercise more, without first reducing insulin, sets her up for a demoralising cycle of effort and non-response. Fix the insulin first, and weight follows - not the other way round.

Second, thin PCOS is common. Roughly a third of PCOS patients in our clinic are of normal weight. Their insulin resistance is there - fasting insulin elevated, HOMA-IR raised - but the scale and BMI do not flag it. The "lose weight" framing does not apply and is often met with "but I'm not overweight." They are, however, metabolically unwell.

Third, the advice is metabolically incomplete. Losing weight on a calorie-restricted high-carbohydrate diet is different from losing weight on a protein-forward, lower-insulin-demand plan. The first often fails or rebounds. The second consistently produces the hormonal changes we want.

The instruction should not be "lose weight." It should be "reduce your insulin." Weight is often a consequence of success, not the path.

The workup that actually describes your PCOS

If you have been told you have PCOS and your workup was an ultrasound and a few hormones, you have been under-worked-up. A complete panel for a new PCOS presentation should include:

• Fasting insulin and HOMA-IR
• Fasting glucose and HbA1c
• Triglyceride/HDL ratio
• Total testosterone, free testosterone, DHEA-S
• SHBG
• LH and FSH (with the LH/FSH ratio)
• Prolactin and 17-hydroxyprogesterone (to rule out other mimics)
• Thyroid full panel, including anti-TPO (PCOS and Hashimoto's frequently co-exist)
• Vitamin D, B12, Ferritin
• hs-CRP
• Pelvic ultrasound
• ALT, AST, and ultrasound for fatty liver (silent in most PCOS)

The single most important test on this list, for most patients, is fasting insulin. It explains nearly everything the standard hormone panel does not.

The protocol: treating the insulin, not the ovary

A root-cause PCOS protocol has four pillars. Each of them is an insulin-reducing intervention wearing a different costume.

Pillar 1: Food designed around insulin demand

Not low-calorie. Low-insulin-demand. In practice:

• Protein: 1-1.2 g per kg of target body weight per day. Most PCOS patients arriving at our clinic are eating 40-50% of what they need.
• Carbohydrate: moderate, with a tight cap on refined forms - white rice, biscuits, sugary tea, juice, noodles, maida-based foods.
• Fat: not restricted. Healthy fats - nuts, seeds, olive oil, fatty fish - replace the calorie space.
• Meal timing: a 10-hour feeding window helps insulin rest. Early dinner (before 8 PM) is a high-value behavioural change.

Cycle regularisation in our cohort typically begins within three to four months of consistent protein-forward eating.

Pillar 2: Resistance training

The single most important behavioural change for PCOS, and almost no one is told. Muscle is the main tissue that absorbs glucose under insulin's direction. More muscle means lower insulin demand means lower ovarian androgens.

The target is twice or three times weekly, progressive. Bodyweight, dumbbells, or a gym - the modality matters less than the consistency and the progression. Cardio is good for the heart but is not the primary lever for PCOS; resistance training is.

Pillar 3: Inositol and targeted supplementation

Myo-inositol with d-chiro-inositol in a 40:1 ratio is the most evidence-based supplement for PCOS. Trials consistently show improvement in ovulation, androgens, and metabolic markers within three to four months. Typical dose is 2 g myo-inositol twice daily plus 50 mg d-chiro-inositol twice daily.

Supporting nutrients that often matter:

• Vitamin D to 50-70 ng/mL
• Magnesium (particularly in patients with sleep issues or anxiety)
• Omega-3
• N-acetylcysteine in some cases for ovulation support

Pillar 4: Sleep, stress, and circadian structure

PCOS is particularly sensitive to sleep debt and stress physiology, because both elevate cortisol, which further elevates insulin and androgens. A consistent sleep window - typically 10:30 or 11 to 6:30 or 7 - is structurally necessary. Shift-work PCOS is a distinct challenge; it requires specialised accommodation.

A genuine daily stress practice - breathwork, walking in daylight after lunch, journalling, whatever is sustainable - is not a soft recommendation. It is a clinical intervention.

Where medication fits

Medication has a legitimate but limited role in PCOS.

Metformin reduces insulin and is a reasonable adjunct in more severe or slower-responding cases. It is not first-line for most of our patients, because lifestyle interventions move the biology at least as effectively, but it has a place - especially in patients trying to conceive or with significant metabolic load.

The oral contraceptive pill masks the cycle. It does not treat the PCOS. It can be useful for symptom control (cycle irregularity, acne), but it should be understood as symptom management, not disease treatment. The underlying insulin resistance continues. Many PCOS patients spend a decade on the pill and discover their actual disease only when they come off to try to conceive.

Letrozole or clomiphene for ovulation induction is a fertility tool, not a PCOS treatment, and is best used within a broader root-cause frame.

Anti-androgens (spironolactone) can reduce symptoms of androgen excess (acne, hirsutism) in severe cases while the root-cause work takes effect.

None of these medications is wrong. All of them are incomplete if used alone.

Case study - Anjali, 27, Mumbai

Anjali came in because her cycle had drifted from a regular 28 days in her teens to 45-60 days over the last three years. She had persistent jawline acne, new facial hair along the upper lip, and scalp thinning at the temples she had begun trying to hide. She was not overweight - BMI 22, waist 78 cm.

Her gynaecologist had diagnosed PCOS on ultrasound and prescribed the oral contraceptive. She had been on it for a year. Her cycle was regular by virtue of the pill. Everything else - acne, hair, mood - was unchanged or worse.

She wanted to stop the pill. She wanted to understand her actual biology.

Extended panel:

• Fasting insulin 19, HOMA-IR 4.2
• HbA1c 5.5, fasting glucose 94
• Triglyceride/HDL 3.2
• Total testosterone upper normal, Free testosterone high, SHBG 22 (low)
• DHEA-S moderate elevation
• Vitamin D 15, Ferritin 14, B12 310
• hs-CRP 2.1
• ALT 38, Ultrasound: Grade 1 fatty liver, polycystic-appearing ovaries

Anjali's PCOS was entirely explained by the underlying biology. Her total testosterone was not particularly high, but her SHBG was low, making her free testosterone (the active molecule) high - which produced the skin and hair picture. Her insulin was running the show. Her fatty liver was telling the same story from a different angle.

Protocol over eight months: came off the oral contraceptive (with a clear understanding that the cycle might take three to four months to return), inositol 2 g twice daily with d-chiro 50 mg twice daily, protein-forward food plan, resistance training three times a week, iron repletion, vitamin D to therapeutic, sleep at 11 PM lights-out, omega-3, B12.

Timeline:

• Month 2: cycle returned at 38 days
• Month 4: cycle at 32 days, acne visibly improved
• Month 6: cycle at 29 days, acne cleared except around menses, new hair regrowth visible at temples
• Month 8: fasting insulin 7, HOMA-IR 1.5, SHBG 44, free testosterone normal, ALT normalised

Anjali's PCOS did not go "into remission" in the sense that her ovaries were now different. Her insulin biology was different, and the ovary stopped being pushed into the PCOS phenotype. She will maintain the protocol indefinitely; the biology rewards consistency.

Case study - Rupa, 32, Bengaluru

Rupa had been trying to conceive for eighteen months. Her gynaecologist had diagnosed PCOS three years earlier but framed it as "mild" because her cycles were "only slightly irregular." She had been put on a short course of metformin at diagnosis and discontinued it when she did not tolerate it well. The conversation since had been: "Try for a year, and if nothing happens, we'll look at letrozole."

BMI 28, waist 89 cm. Persistent mild acne, some scalp thinning.

Panel:

• Fasting insulin 23, HOMA-IR 5.4
• HbA1c 5.7, fasting glucose 96
• Total testosterone upper normal, Free testosterone high, SHBG 18
• LH/FSH ratio 2.4 (elevated)
• Anti-Mullerian hormone (AMH) 8.2 (high-normal, consistent with PCOS)
• Vitamin D 11, Ferritin 16
• hs-CRP 3.1
• ALT 48, Ultrasound: Grade 2 fatty liver, multiple peripheral follicles bilaterally

Rupa's picture was of a moderately severe metabolic PCOS. Her fertility plan had been thin. Letrozole in an insulin-resistant, fatty-liver patient might have produced ovulation but would not have addressed her chances of a healthy pregnancy, or her longer-term trajectory.

Protocol: full metabolic programme - protein-forward food plan, resistance training, inositol at full PCOS dose, vitamin D to therapeutic, iron, omega-3, liver support for fatty liver, sleep restructuring; brief reintroduction of slow-release metformin at a low dose (500 mg daily) that she tolerated well this time; gluten elimination trial (mild response); preconception folate and prenatal nutrients.

Four months:

• Cycle from 40-45 days to 30 days
• Fasting insulin 10, HOMA-IR 2.1
• SHBG 35, free testosterone normal
• ALT normalised

She conceived at month six without ovulation induction, and carried a healthy pregnancy to term. She remained on the protocol throughout pregnancy with modifications for her obstetrician's plan. Her post-partum metabolic profile was substantially better than her pre-pregnancy one - the programme held.

Case study - Kavya, 23, Hyderabad

Kavya was a young patient referred by her mother, who was our patient. She had had irregular cycles since menarche, stubborn acne since fifteen, recent onset of hirsutism along the jaw, and a recent weight gain of 6 kg that she could not explain. She was a student; her lifestyle had not changed.

Her gynaecologist had diagnosed PCOS on ultrasound and suggested she start the oral contraceptive. Her mother had seen enough of the cost of treating symptoms without treating causes and had brought her for a proper workup first.

Panel:

• Fasting insulin 15, HOMA-IR 3.1
• Total testosterone high, Free testosterone high, SHBG 20
• DHEA-S elevated (adrenal contribution)
• Vitamin D 13, Ferritin 9 (very low), B12 250
• hs-CRP 1.6
• TSH 3.8, Anti-TPO 65 (positive - early autoimmune thyroiditis)
• Cycle history: never regular, last menses ten weeks earlier

Kavya is the PCOS patient we see most often - young, never had a thorough workup, diagnosed on ultrasound alone, offered the pill. Her picture also showed the second-common pattern we see: PCOS co-existing with early Hashimoto's. Treating one without the other is a common way to stall progress.

Protocol: food plan built around her college life (simple, implementable at a hostel - eggs at breakfast, proper dal-and-protein at lunch, early dinner at home, no white rice at dinner), resistance training twice weekly at the college gym with a simple programme, inositol at PCOS dose, iron (her main deficiency), vitamin D, selenium (for the thyroid), omega-3. Thyroid was monitored; no thyroxine started because TSH was borderline and antibodies were at the lower end.

Six months:

• Cycle at 32-35 days for four consecutive months (first time ever)
• Acne cleared substantially
• Facial hair unchanged in first three months, then slow improvement
• Fasting insulin 8, HOMA-IR 1.6, SHBG 42
• Ferritin 52, Vitamin D 50
• TSH 2.1, Anti-TPO 28

Kavya may need medication at different points in her life. For now, she does not. The most important outcome in a young PCOS patient is not any single number, but the trajectory - she has gone from a system on a worsening path to a system on an improving one. That trajectory is what determines her twenties, thirties, and fertility to come.

What to do if you have PCOS

If you have been diagnosed or suspect you have PCOS:

1. Get the full panel. Fasting insulin, HOMA-IR, free testosterone, SHBG, vitamin D, ferritin, thyroid with anti-TPO, and a fatty liver screen. The standard PCOS workup is not enough.
2. Do not accept the oral contraceptive as a treatment. It is a symptom manager. If you choose it for your reasons, that is fine, but know that it is not addressing your disease.
3. Focus on insulin. Protein-forward eating, resistance training, inositol, vitamin D, sleep. This is not glamorous and it is not a quick fix, but it is the real treatment.
4. If you are trying to conceive, do not skip the metabolic work. A PCOS pregnancy on uncorrected insulin resistance carries higher rates of gestational diabetes, hypertension, and later child metabolic risk. The preconception year is the most leveraged year of a PCOS patient's life.
5. Plan for the long term. PCOS is a marker of lifetime metabolic risk. The good news is that women who address it well in their twenties and thirties are women who avoid diabetes, fatty liver, and cardiovascular disease in their forties and fifties. The work you do now is structural.

Your PCOS is not an ovarian problem you happen to have. It is a metabolic disease the ovary is revealing early. That frame will change everything about how you treat it, and how it responds.