Low AMH: What It Means for Ovarian Reserve and Fertility

Anti-Müllerian hormone (AMH) is a protein made by the granulosa cells of small ovarian follicles, from pre-antral follicles to small antral follicles a few millimetres across. This pool of follicles, visible on ultrasound as the so-called antral follicle count (AFC), is the 'reservoir' from which cells are recruited for further growth in every cycle. The more such follicles there are, the more AMH reaches the blood. That is why AMH is considered an indirect marker of ovarian reserve: it does not count eggs directly, it measures the activity of the pool linked to them.

Unlike FSH, oestradiol or progesterone, AMH changes relatively little over the menstrual cycle, so in clinical practice it can be measured on any day of the cycle - a convenience the other sex hormones do not offer. 'Relatively little' does not mean 'not at all': some studies have observed swings of the order of ten to a few tens of percent between cycle phases, and assay variability adds to that. If the result is borderline and the decision matters, repeating the test, preferably in the same laboratory, is reasonable.

AMH rises through youth, usually peaks around the early twenties to thirties, and then declines gradually until it becomes undetectable around menopause. The decline reflects the natural depletion of the follicle pool, which is why it is the comparison with age, not the number alone, that decides interpretation. An AMH of 1.0 ng/ml in a 25-year-old and in a 41-year-old are two entirely different clinical situations: the first is clearly below what is expected for age, the second falls within the range typical for that stage of life.

There is no single universal threshold for 'low AMH'. Laboratories give their own reference ranges (often wide and age-dependent), fertility clinics use functional thresholds linked to the expected response to stimulation, and research papers use yet other cut-offs. In addition, different assays (immunoenzymatic, automated) are not fully interchangeable, so the same sample can give a slightly different result depending on the reagent kit. Comparing your result with an online table is therefore unreliable.

Mind the units: in Poland AMH is most often reported in ng/ml, in European literature in pmol/l. Roughly, 1 ng/ml corresponds to about 7.1 pmol/l. For example, in the JAMA study cited below, low AMH was defined as below 0.7 ng/ml, which is roughly 5 pmol/l; in the Danish comparison study, values below 5 pmol/l were considered very low.

This is the heart of most searches, and the answer from the best available data is reassuring, though with caveats. The key is to separate two situations: women who are trying to conceive naturally and have no history of infertility, and women undergoing infertility treatment (IVF). In the first group, low AMH did not significantly predict a worse chance of conceiving in the studies.

A Danish study comparing women referred for infertility treatment with women of the same age (20-39) with no history of infertility is consistent with this. The infertile women did not have lower AMH or fewer antral follicles than the control group, and the proportion of very low values (below 5 pmol/l) was similar in both groups. The authors interpret this as meaning that difficulty conceiving in younger women more often reflects the expected age-related decline in fertility than an over-representation of women with prematurely exhausted reserve.

So where does low AMH really make a difference? Above all in reproductive medicine. AMH is one of the best-documented predictors of the ovarian response to hormonal stimulation: the lower the result, the fewer eggs are usually obtained in a single cycle. A meta-analysis of 42 studies showed that both AMH and antral follicle count predict poor and excessive response to stimulation well (sensitivity of about 0.80 for predicting poor response with AMH; the ROC curves of the two methods did not differ significantly), although the cut-off value was a significant source of heterogeneity.

The most important nuance, however, concerns not the number of eggs but the ultimate goal, a live birth. A 2023 systematic review and meta-analysis (32 studies) showed that AMH is related to the live birth rate after IVF/ICSI in a non-linear way, but no threshold was found below which having a baby would be impossible. The authors explicitly stress that low AMH should not be the sole criterion for rejecting IVF treatment, especially in younger patients. At the same time, there is not enough data to assess AMH's ability to predict live birth after insemination or in women trying to conceive without treatment.

For people considering egg freezing (social freezing), the practical conclusion is similar: AMH helps estimate how many eggs can be obtained in one stimulation cycle and whether two cycles may be needed, but the chance of pregnancy from frozen eggs depends mainly on age at the time of retrieval. Low AMH is therefore an argument to decide earlier, not proof that it is too late.

The AMH result is sensitive to several factors that do not necessarily reflect permanent exhaustion of the ovaries. The best-documented example is hormonal contraception. In a Danish comparison of 228 users and 504 non-users of hormonal contraception (after adjusting for age), AMH was on average about 30% lower in women using contraception, as were antral follicle count (about 30% fewer) and ovarian volume (about 42% smaller); the relationship with duration of use was linearly negative, and none was seen with ethinylestradiol dose. The authors conclude that AMH and AFC may lose accuracy as reserve markers in women taking hormones.

It is also worth remembering that high AMH is not 'better' - values well above the range for age often accompany polycystic ovary syndrome, in which a large number of follicles goes hand in hand with ovulation disorders. We write more about this condition in the PCOS entry of our knowledge base.

First, AMH does not assess egg quality. Quality - understood as the risk of aneuploidy and the developmental potential of the embryo - depends mainly on the woman's age and is not measured by any routine blood test. A young woman with low AMH therefore usually has fewer eggs but statistically still a high proportion of good ones; a woman in her forties with 'normal' AMH may have many eggs, of which the proportion capable of fertilisation is low.

Second, AMH does not precisely predict the age of menopause. Low values statistically herald earlier menopause, but individual scatter is so large that it is not suitable as a stand-alone tool for planning one's life. We discuss the differences between perimenopause and menopause in a separate article.

Third, AMH does not tell you whether you ovulate. You can have high AMH and not ovulate (PCOS), or low AMH and ovulate regularly. Ovulation is assessed with other tools: cycle tracking, ovulation tests, luteal-phase progesterone (we discuss its interpretation in the article on progesterone) and ultrasound. Finally, AMH does not assess the male factor or tubal patency, and therefore does not replace the basic work-up of the couple.

A special case is suspected premature ovarian insufficiency (POI), meaning loss of normal ovarian function before the age of 40. The 2024 guideline from ESHRE, ASRM, CREWHIRL and IMS indicates that a single elevated FSH above 25 IU/l is enough to diagnose POI (with a suitable clinical picture), and that AMH and a repeat FSH may help where the diagnosis is uncertain. Low AMH alone does not diagnose POI - symptoms (cycle disturbance, hot flushes) and the hormonal profile decide.

What this text does not cover: it does not describe infertility treatment or stimulation protocols, does not replace a gynaecological consultation, and does not give individual chances of pregnancy. The data from the JAMA study concern women aged 30-44 who had no fertility problems and were just starting to try - the results need not carry over to patients with years of infertility, after cancer treatment or with a history of POI. Many of the cited studies are observational, and cut-offs, assays and populations differ between papers.

AMH is a very useful test in the hands of a reproductive medicine specialist planning stimulation or assessing risk after cancer treatment. As a stand-alone 'fertility test' ordered out of curiosity and read online, however, it is a generator of needless anxiety. If the result is low, the best move is not to hunt for a miracle supplement but to put the facts in order: age, cycles, any contraception, family plans and timelines. Then the number in the lab report becomes information for a decision, not a reason for panic.