PCOS Macros: What the Trials Actually Compared, and What Changed Nothing

Does the Macro Split Actually Change How Much Weight You Lose With PCOS?

Weight loss improved the presentation of PCOS in the majority of studies regardless of which macronutrient composition was used, according to a systematic review of six trials and 137 women with PCOS that compared monounsaturated-fat-enriched, low-glycemic-index, high-carbohydrate, low-carbohydrate and high-protein diets against each other. The review did find one composition-specific weight difference — a monounsaturated-fat-enriched diet produced somewhat greater weight loss than the others in the single study that tested it — but its overall conclusion was that weight loss should be targeted “irrespective of diet composition,” through a reduced calorie intake built on adequate nutrition and food choices a person can sustain, rather than through hitting one specific protein, carbohydrate or fat percentage. Total energy balance is what mostly decides whether weight changes at all, and the right energy target is individual — shaped by starting size, activity, medication and metabolic rate — which is why no single calorie number belongs in a general article, and why the 2023 international guideline routes energy prescription through an individualised assessment with a dietitian rather than a fixed figure for every person with PCOS. This page sits inside the weight-loss section, which treats weight as one metabolic marker among several — insulin, inflammation, blood pressure — rather than the point of the article.

Which Macro Ratio Won on Which Specific Outcome?

Every diet composition tested in the 2013 review moved at least one PCOS-relevant marker, and no single composition won on all of them. Reading the six-trial review outcome by outcome, rather than as one pooled verdict, is what actually tells you which ratio matches which goal.

None of this means a low-glycemic-index diet is “the” PCOS diet, or that a high-carbohydrate diet is unsafe — free androgen index rising on higher-carbohydrate intake was one finding inside a small evidence base (six trials, 137 women total), not a settled dose-response relationship. It does mean that “which macro split is best” has a different answer depending on whether the marker you care about most is a cycle, a mood measure, or an insulin number.

What Is the “Ideal” PCOS Macro Split, According to the Trials That Tested One?

Three trials, each testing a different comparison, found three different things moved — and none of them was the scale by a different amount. In a 16-day crossover trial comparing a standard diet (56% carbohydrate, 31% fat, 16% protein) against a low-carbohydrate diet (43% carbohydrate) and a monounsaturated-fat-enriched diet (with fat raised to 17% above baseline) at matched calorie levels, fasting insulin was lower on the low-carbohydrate diet than on the standard diet, and the insulin response to a glucose challenge was lower on the low-carbohydrate diet than on the fat-enriched diet — while fasting glucose, insulin sensitivity itself, and circulating reproductive hormones did not change significantly across any of the three. Because calories were held equal across all three diets in this trial, the insulin differences came from composition alone, not from weight change.

Read the Goss trial carefully, because it is the one genuine “which macro split changes body composition” answer in this table: at equal calories, the lower-carbohydrate diet led to more fat loss specifically from visceral and intermuscular depots — the fat linked to insulin resistance — while the higher-carbohydrate standard diet lost lean mass instead of fat during the same eight weeks. That is a composition effect independent of total weight change, which is a different and more specific claim than “low-carb helps you lose weight faster.”

Does a High-Protein, Low-Carb Ratio Actually Outperform a Standard Ratio for PCOS?

Both diets in a 12-week trial produced statistically similar weight loss, but only the high-protein, low-glycemic-load diet produced a significant drop in insulin, HOMA-IR and high-sensitivity C-reactive protein, in a randomized trial of 60 overweight and obese women with PCOS assigned to either a conventional hypocaloric diet (15% protein) or a modified hypocaloric diet (30% protein, low-glycemic-load foods). LDL cholesterol fell by 24.5% in both groups. Testosterone dropped in both groups too — from 1.78 to 1.31 ng/mL in the higher-protein group and from 1.51 to 1.15 ng/mL in the conventional group — which the trial’s authors attributed to weight loss itself rather than to protein or glycemic load specifically, since both groups lost a similar amount of weight and both saw a similar-sized androgen drop.

Why Would Protein or Carbohydrate Ratio Change Insulin Without Changing Weight?

Two mechanisms explain the Douglas and Goss findings without needing weight change as the intermediate step. A meal lower in refined carbohydrate produces a smaller glucose spike, which means the pancreas releases less insulin to handle it — repeated at every meal, that adds up to a lower average insulin exposure across a day even before any weight changes. Separately, protein and fibre both slow gastric emptying and blunt the glucose response to a mixed meal, which is the mechanism behind the low-glycemic-load half of the Mehrabani diet doing more for insulin than the higher-protein half alone would predict. Neither mechanism requires fat loss to operate, which is why Douglas found an insulin difference in 16 days at matched calories and matched weight — too short a window for meaningful fat change, but long enough for a meal-by-meal insulin difference to show up in fasting blood work.

What Actually Changed Nothing in These Trials?

Fasting glucose, insulin sensitivity measured directly, and circulating reproductive hormone concentrations did not differ significantly by macro composition in the eucaloric Douglas trial, even though fasting insulin did. Follicle-stimulating hormone, luteinizing hormone and most lipid markers besides LDL were unchanged by protein ratio in the Mehrabani trial. And across the systematic review of all six trials, weight loss improved PCOS presentation under nearly every composition tested, with only one trial showing a composition-specific edge — the headline pattern underneath all three trials in the table above is convergence, not a single ratio pulling away from the rest. If a claim promises a macro ratio that melts fat other ratios cannot touch, none of the trials reviewed here support it; what they support is that composition changes specific metabolic markers, sometimes independent of weight, while the scale itself mostly tracks calories.

Who This Does Not Work For

These trials ran 8 to 16 weeks — none of them answer what happens to insulin or fat distribution after a year on a given ratio, and none tested very-low-carbohydrate or ketogenic ratios against the moderate reductions used here, so nothing in this evidence base tells you what a stricter split would do that a 40% carbohydrate diet does not already do. Someone managing a history of disordered eating is a poor fit for rigid macro tracking regardless of what a trial found, since the psychological cost of counting grams can outweigh a modest metabolic benefit — a conversation worth having with a clinician or dietitian before adopting any structured ratio, and one where “the research says 30% protein” is not a good enough reason to override that judgment. Pregnancy and breastfeeding are also outside every trial cited here, all of which recruited women who were neither.

Adherence, not the ratio itself, decided outcomes in every trial in this review — participants in each arm were coached, monitored and supported for the length of a study, conditions that do not automatically carry over to someone building a ratio alone from an app. A diet nobody can sustain past the study period will not reproduce these results no matter how well-evidenced the macro split looks on paper, and none of the four trials above measured what happened once the intervention ended. That is a real gap: short-term composition studies are the only evidence that exists, and “lasts 8 to 16 weeks under supervision” is not the same claim as “lasts.”

Treating insulin resistance in PCOS ranks interventions by measured effect size if the goal is specifically the insulin marker rather than the diet pattern that gets you there, and how much protein actually supports PCOS covers the practical target and food sources in more detail than fits here.

Sleep matters here too: untreated sleep apnea independently worsens insulin resistance in PCOS, by a margin large enough in one trial to work against any macro change made on the diet side. And because so much PCOS diet advice gets discussed inside a much larger conversation about a person’s weight, it is also worth reading how weight stigma shapes what actually gets said in these appointments before assuming a clinician’s dietary advice is evidence-based rather than assumption-based.

You may see this condition written as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 medical organisations renamed it. The trial findings above are unchanged by the name — this page uses PCOS because that is still what most people search.