Anti-Müllerian hormone (AMH) is consistently elevated in women with polycystic ovary syndrome, typically running two to three times higher than in women without the condition. That elevation is not just a coincidence: AMH appears to be both a product of the hormonal disruption that defines PCOS and an active contributor to it, interfering with normal follicle development and ovulation. The relationship is tight enough that the most recent international PCOS guidelines now recognize AMH as a formal diagnostic tool, offering it as an alternative to pelvic ultrasound in adult women. But the connection goes deeper than diagnosis, touching on fertility treatment planning, metabolic health, reproductive aging, and even what may happen before birth.
Why AMH Runs High in PCOS
AMH is produced by granulosa cells, the cells that line developing ovarian follicles. In a typical ovary, AMH acts as a brake on early follicle development, slowing the transition from resting follicles to actively growing ones. That braking function makes it a reliable marker of how many growing follicles a woman has at any given time, which is why it is widely used as a gauge of ovarian reserve.
In PCOS, AMH levels climb for two reasons that compound each other. First, PCOS ovaries contain a larger-than-normal pool of small growing follicles, and more follicles means more AMH-producing cells. One study found that AMH rises at a consistent rate of roughly 0.2 ng/mL per additional follicle.1Human Reproduction. The relationships between AMH, androgens, insulin resistance and basal ovarian follicular status in non-obese subfertile women with and without polycystic ovary syndrome Second, and less intuitively, each individual granulosa cell in a PCOS ovary pumps out far more AMH than its counterpart in a normal ovary. Research comparing anovulatory PCOS ovaries to normal ovaries found that AMH production per granulosa cell was roughly 75 times higher in the PCOS group.2PubMed. Granulosa cell production of anti-Müllerian hormone is increased in polycystic ovaries So the elevation is not simply a matter of having more follicles; the follicles themselves are behaving differently.
Androgens appear to be a key driver of this overproduction. Excess androgens, a hallmark of PCOS, make granulosa cells more sensitive to follicle-stimulating hormone (FSH), which in turn pushes them to produce more AMH.3Human Reproduction Update. Interactions between androgens, FSH, anti-Müllerian hormone and estradiol during folliculogenesis in the human normal and polycystic ovary The excess AMH then feeds back into the problem: it inhibits FSH’s ability to push a dominant follicle to maturity, contributing to the arrested follicle development and anovulation that characterize the syndrome.2PubMed. Granulosa cell production of anti-Müllerian hormone is increased in polycystic ovaries The result is a self-reinforcing loop: androgens boost AMH, and AMH blocks follicle maturation, leaving a growing collection of small follicles that keep producing androgens and AMH.
AMH as a Diagnostic Tool
For years, diagnosing PCOS relied on the Rotterdam criteria: some combination of irregular periods, signs of excess androgens, and a polycystic-appearing ovary on ultrasound. The ultrasound requirement created barriers. Transvaginal ultrasound can be uncomfortable, is not always accessible in primary care, depends heavily on operator skill, and is less reliable in adolescents and people with a higher body mass. A simple blood draw measuring AMH sidesteps most of those problems.
The 2023 international evidence-based guideline for PCOS formally endorsed serum AMH as an alternative to ultrasound for identifying polycystic ovarian morphology in adults.4The Journal of Clinical Endocrinology & Metabolism. Recommendations From the 2023 International Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome The guideline was careful about when AMH testing actually adds value, though. If a woman already has both irregular cycles and clinical or biochemical signs of excess androgens, those two features are enough for a PCOS diagnosis on their own, and measuring AMH is unnecessary. AMH testing is most useful when only one of those features is present and the clinician needs a third criterion to confirm or rule out the diagnosis.5European Journal of Endocrinology. Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome The guideline also explicitly warns against performing both an ultrasound and an AMH test for the same patient, because using both raises the risk of overdiagnosis.
One practical challenge is that AMH thresholds differ between assay platforms. Automated assays and older manual assays can produce different numbers from the same blood sample, meaning a cutoff that works well on one platform may misclassify patients on another. Studies comparing multiple AMH assays have confirmed that diagnostic performance is comparable across platforms as long as platform-specific thresholds are used.6PubMed. Comparative assessment of five serum antimüllerian hormone assays for the diagnosis of polycystic ovary syndrome Efforts to standardize AMH measurement are ongoing, but in practice this means the specific number flagged as “elevated” can vary depending on the lab.
AMH Levels Differ Across PCOS Phenotypes
PCOS is not a single uniform condition. Depending on which features are present, it is classified into four phenotypes, labeled A through D. Phenotype A is the “classic” form with irregular cycles, hyperandrogenism, and polycystic ovaries all present. Phenotype B has irregular cycles and hyperandrogenism but no polycystic ovaries on imaging. Phenotype C combines hyperandrogenism with polycystic ovaries but regular cycles. Phenotype D has irregular cycles and polycystic ovaries but no androgen excess.
AMH levels follow a clear hierarchy across these groups. A meta-analysis found the highest levels in phenotype A, with a mean around 11.5 ng/mL, followed by phenotype D at about 9 ng/mL, phenotype C near 8 ng/mL, and the lowest in phenotype B at roughly 6.3 ng/mL.7PubMed Central. Anti-Müllerian Hormone Levels Across Phenotypes in Polycystic Ovary Syndrome: A Systematic Review and Meta-analysis The pattern makes biological sense: phenotypes that include polycystic ovarian morphology (A, C, and D) have more small follicles producing AMH. Phenotype B, which lacks that feature, produces the least.
This phenotype variation matters clinically. Phenotype D, the “non-hyperandrogenic” form, is often the hardest to diagnose because the most recognizable sign of PCOS, excess androgens, is absent. Recent research on phenotype D has shown that AMH, testosterone, and the LH-to-FSH ratio are all significantly higher in these women than in controls, and that age-specific AMH cutoffs can identify the phenotype with reasonable accuracy across reproductive age groups.8PubMed. AMH levels and diagnosis in PCOS phenotype D In other words, AMH testing may be especially valuable for catching the forms of PCOS that are easiest to miss.
The Adolescent Diagnosis Problem
Diagnosing PCOS in teenagers is tricky. Irregular periods are normal in the first few years after the first menstrual cycle, mild acne is common in puberty, and ovarian ultrasound can show a polycystic-appearing ovary in perfectly healthy adolescents. Because of this overlap, the 2023 guideline does not recommend using AMH or ultrasound to define polycystic ovarian morphology in adolescents within eight years of their first period.
Still, researchers have explored whether AMH could help in this group. A meta-analysis of studies in adolescents estimated that an AMH cutoff of roughly 6 to 7 ng/mL could distinguish those with PCOS from those without, with a specificity of about 81% and sensitivity around 66%.9PubMed. Anti-Müllerian hormone levels in the diagnosis of adolescent polycystic ovarian syndrome: a systematic review and meta-analysis The specificity is decent, meaning a positive result is fairly reliable, but the sensitivity leaves something to be desired: about a third of adolescents with PCOS would be missed. For now, AMH in teenagers is considered a supporting clue rather than a standalone diagnostic criterion, and a clinical reassessment later in young adulthood is the recommended path for borderline cases.
What AMH Means for Fertility Treatment
High AMH in PCOS is a paradox in the fertility clinic. On one hand, it signals a large pool of recruitable follicles, which sounds like an advantage when you are trying to conceive. On the other hand, that same abundance creates a heightened risk of ovarian hyperstimulation syndrome (OHSS) during IVF, a potentially serious complication where the ovaries overreact to stimulation medications. In one study, roughly 22% of women with PCOS undergoing controlled ovarian stimulation developed OHSS, compared with under 5% of non-PCOS women.10PubMed Central. Serum AMH Level to Predict the Hyper Response in Women with PCOS and Non-PCOS Undergoing Controlled Ovarian Stimulation in ART
Pre-treatment AMH values help fertility specialists calibrate stimulation protocols. In the PCOS group from that same study, an AMH cutoff of about 6.85 ng/mL predicted hyperresponse with moderate sensitivity and specificity, while in non-PCOS women the threshold was lower, around 4.85 ng/mL. These numbers are not precise enough to rely on in isolation, but combined with follicle counts and clinical judgment, they help guide how aggressively to stimulate.
Counterintuitively, among women with PCOS undergoing IVF, having very high AMH does not translate to better pregnancy outcomes. A systematic review and meta-analysis found that PCOS patients in the highest AMH quartile had lower odds of clinical pregnancy and live birth compared to those in the lowest quartile, despite retrieving more eggs.11PubMed Central. Association between serum AMH levels and IVF/ICSI outcomes in patients with polycystic ovary syndrome: a systematic review and meta-analysis Fertilization rates were also slightly lower at the highest AMH levels. The likely explanation is that extremely high AMH reflects a more severely disrupted hormonal environment, where egg quality may suffer even as egg quantity is abundant. For women with PCOS planning IVF, the practical message is that more is not always better when it comes to AMH.
AMH and Insulin Resistance
Because PCOS and insulin resistance so often travel together, researchers have spent years trying to pin down whether AMH is directly linked to insulin resistance or whether the connection runs through other pathways. The answer, so far, is frustratingly murky.
A recent analysis found a negative association between AMH and markers of insulin resistance that persisted even after accounting for body weight, suggesting insulin resistance itself may independently suppress AMH to some degree.12PubMed Central. Beyond BMI: insulin resistance emerges as the key metabolic correlate of AMH in PCOS But when a broader meta-analysis pooled results from multiple studies, the overall correlation between AMH and insulin resistance was weak and did not reach statistical significance, with enormous variability between studies.13PubMed Central. Correlation between anti-mullerian hormone with insulin resistance in polycystic ovarian syndrome: a systematic review and meta-analysis Adding to the confusion, at least one study found no meaningful difference in AMH levels between PCOS women with and without insulin resistance, and no correlation between the two at all.14PubMed. Serum AMH levels and insulin resistance in women with PCOS
The inconsistency probably reflects how many variables are in play at once. AMH is primarily driven by follicle count and granulosa cell behavior, which are influenced most directly by androgens and gonadotropins. Insulin resistance may nudge AMH levels indirectly by worsening androgen excess or altering follicle dynamics, but it does not appear to be a major independent driver. In practice, this means AMH is not a useful marker of metabolic health in PCOS, and the two should be assessed separately.
How Treatment and Lifestyle Affect AMH
Because AMH reflects the underlying ovarian disturbance in PCOS, it can also shift when that disturbance is treated. Metformin, an insulin-sensitizing drug commonly prescribed for PCOS, has been shown to reduce AMH levels after several weeks to months of use. One study reported a significant drop from a mean of about 10 ng/mL to roughly 7.8 ng/mL after eight weeks of treatment.15PubMed Central. Effect of metformin on the anti-Müllerian hormone level in infertile women with polycystic ovarian syndrome A separate study found the reduction became more pronounced over six months and was associated with improved ovulation and pregnancy rates.16Acta Medica Academica. The Impact of Metformin Treatment on Serum Level of AMH and on Ovulation and Pregnancy Outcome in Women With PCOS Laparoscopic ovarian drilling, a surgical approach sometimes used when medication fails, has also been shown to bring AMH down substantially, from a mean of about 15.2 to 10.2 ng/mL in one report.17Human Reproduction. P-746 effect of laparoscopic ovarian drilling on reproductive, hormonal and metabolic parameters in anovulatory infertile PCOS women resistant to Letrozole/CC + gonadotropin based ovulation induction cycles
Exercise also appears to lower AMH in women with PCOS. A meta-analysis found that both aerobic and resistance exercise reduced AMH levels, with the effect being strongest in women who had a higher BMI and higher baseline AMH, and with longer exercise durations showing more benefit.18Middle East Fertility Society Journal. The effect of exercise on anti-Mullerian hormone levels in patients with polycystic ovary syndrome: a systematic review and meta-analysis A separate trial found that the decline in AMH after exercise tracked with reductions in a marker of oxidative stress, hinting at a possible mechanism involving oxidative damage to granulosa cells.19PubMed. Improvement of anti-Müllerian hormone and oxidative stress through regular exercise in Chinese women with polycystic ovary syndrome
Interestingly, one randomized trial comparing diet, exercise, and a combination of the two found that diet alone produced the most significant drop in AMH, and that the strongest predictor of the decrease was a fall in free testosterone rather than weight loss itself.20PubMed. Serum antimüllerian hormone in response to dietary management and/or physical exercise in overweight/obese women with polycystic ovary syndrome: secondary analysis of a randomized controlled trial Women whose AMH normalized also showed improvements in menstrual regularity and hyperandrogenism. This suggests that falling AMH levels in response to treatment are not just a lab curiosity but a signal that the ovarian environment is genuinely improving.
Does PCOS Delay Reproductive Aging?
One of the more intriguing aspects of the AMH-PCOS connection is what it implies about long-term reproductive lifespan. Because AMH is a marker of remaining ovarian follicle reserve, and because PCOS women start with much higher levels, a natural question is whether PCOS women enter menopause later than average.
Population-based cohort data suggests the answer is modestly yes. A study tracking AMH levels over ten years in PCOS women and controls estimated that the average age at menopause for PCOS women was around 51, compared with about 49 for controls, a roughly two-year extension.21Human Reproduction. Is polycystic ovary syndrome an exception for reproductive aging? The same approximate two-year difference appeared in a separate study that modeled predicted reproductive lifespan based on AMH trajectories.22PubMed. Prediction of age at menopause in women with polycystic ovary syndrome The PCOS group also showed an interesting pattern where the rate of AMH decline slowed over time, as though the large initial follicle pool was being depleted more gradually.
Two extra years is real but modest, and the confidence intervals around these estimates are wide. It does not mean PCOS protects against age-related fertility decline in any clinically meaningful way; the irregular ovulation that characterizes the syndrome is itself a significant fertility barrier. But it does complicate the interpretation of AMH in older women with PCOS: a 38-year-old with PCOS and a “normal” AMH level may have less reserve relative to her own baseline than the number suggests, while the same number in a woman without PCOS might look reassuring.
Genetics and the Prenatal AMH Theory
The question of why certain women develop PCOS has led researchers to look at both genetic variation in the AMH gene and the possibility that AMH exposure before birth plays a role. On the genetics side, several studies have examined variants in the genes coding for AMH and its receptor (AMHR2). A pilot study found that a specific AMH gene variant (rs10407022) was associated with high AMH levels in PCOS women, suggesting that genetic differences in AMH regulation could contribute to the syndrome’s development.23PubMed Central. Anti-Mullerian Hormone Gene Polymorphism in Polycystic ovary syndrome: A pilot study Another study in Chinese women with PCOS and insulin resistance found that diversity in AMH-related genotypes was connected to differences in androgen levels and disease susceptibility.24PubMed Central. Anti-Müllerian hormone gene polymorphism is associated with androgen levels in Chinese polycystic ovary syndrome patients with insulin resistance
That said, a comprehensive analysis of AMH and AMHR2 variants found no significant overall association between these gene variants and PCOS susceptibility when looking at allele, genotype, and haplotype levels together.25PubMed. The Role of AMH and AMHR2 Variants in Polycystil Ovary Syndrome: ‘A Comprehensive Analysis’ Some genotype-specific correlations with lipid profiles emerged, but the gene variants themselves did not reliably predict who would develop PCOS. The genetics of AMH in PCOS are clearly complex and likely involve many genes of small effect rather than a single smoking gun.
A more provocative theory comes from animal research. In a landmark mouse study, researchers found that pregnant women with PCOS had elevated AMH levels during pregnancy. When pregnant mice were treated with excess AMH to mimic this elevation, their female offspring developed a PCOS-like condition in adulthood, complete with irregular cycling, high testosterone, and altered brain signaling.26PubMed Central. Elevated prenatal anti-Müllerian hormone reprograms the fetus and induces polycystic ovary syndrome in adulthood The mechanism involved the excess prenatal AMH driving the mother’s body to produce more testosterone, which then crossed the placenta and masculinized the developing fetus. If this applies in humans, it would mean PCOS can, to some extent, program itself into the next generation through AMH. The idea is compelling but unproven in people; it remains an active area of investigation.
When AMH Testing Can Mislead
As useful as AMH testing is, there are situations where it gives a misleading picture. AMH levels naturally decline with age, so a cutoff that works well in a 25-year-old may not apply to a 35-year-old. Some PCOS research has attempted to create age-stratified thresholds to account for this, but no universally agreed-upon set of age-adjusted cutoffs exists yet. Hormonal contraceptives also suppress AMH to varying degrees, so testing while on the pill or other hormonal methods can produce a falsely low reading. Most guidelines recommend testing AMH in an untreated state for diagnostic purposes.
Body weight adds another layer of complexity. While AMH is not tightly correlated with BMI in the broader population, women with severe obesity may have somewhat lower AMH than similarly-aged women of normal weight, regardless of PCOS status. In an obese woman with PCOS, the AMH level might be elevated relative to non-PCOS women of the same weight but still fall below a threshold designed for the general PCOS population, potentially leading to a missed diagnosis.
Finally, a high AMH level alone does not automatically mean PCOS. Some women with high ovarian reserve but no hormonal disturbance simply have more follicles than average. The 2023 guideline was deliberate about situating AMH within a broader diagnostic algorithm rather than treating it as a standalone test. AMH is most informative when read alongside menstrual history, androgen levels, and clinical context, not in isolation.