Why AMH is a really terrible egg-timer

It is time we all acknowledged the significant limitations of AMH testing as a reliable predictor of fertility and communicated this to our patients

---

At a conference I recently attended in Dublin one of the local clinicians presented a paper recommending that anti-Mullerian Hormone (AMH) testing be made available at no cost to all Irish women over 25 years of age.

Greater access to AMH screening, she contended, was a feminist issue since it enabled women to estimate their background fertility and make informed decisions about pregnancy planning.

The cost of such testing in a public hospital in Ireland is similar to Australia,  about $A80. The clinician mentioned, however,  that entrepreneurs in the private sector were charging the equivalent of $A400 for the test.

To illustrate the value of the test, she presented cases of women in their late 20s who were prompted by an adverse AMH level to reschedule their plans to start a family.

I did not make myself popular when I later challenged the speaker as to usefulness of this test in the general population. However, this is exactly what the evidence has been telling us now for several years, and it is time we all acknowledged the significant limitations of this test as a reliable predictor of fertility and communicated this to our patients.

BACK TO BASICS

AMH is produced within the ovarian follicle by cells known as granulosa cells. All women start out with a limited number of follicles from which an ovum can eventually develop. Five months before she is born, the female fetus has the highest number of follicles she will ever have- about seven million.

Over time, more and more follicles self-destruct, so that by her first menstrual period only 200,000 remain. AMH levels rise rapidly at puberty, peak in a woman’s 20s and then decline until menopause.

A simple blood test allows a woman’s AMH levels to be ranked as a percentile against those of the same age.

One obvious problem is that the database from which “normal” AMH graphs are derived comes largely from a population of women attending fertility clinics and cannot reliably be applied to the general population.^1,2

Another real problem is that the number of follicles in the ovaries form only part of the story. Every menstrual cycle sees the best quality ova selected for the maturation process which could eventually result in a pregnancy. Regardless of the size of the reservoir, if the ova are of poor quality then successful pregnancy is unlikely. 

WHAT IS AMH USEFUL FOR?

AMH estimation was developed for, and is most useful in, the setting of assisted conception. In a sub-fertile population, AMH is invaluable in predicting the success of invitro-fertilisation (IVF) and in guiding doctors as to the recommended dosages of the hormones used in such techniques.

Additionally, studies do suggest that for women in their 40s, their level of AMH can broadly predict the age at which they are likely to reach menopause. However, these predictions have an eight to 12-year leeway, making them useless as a predictor for any individual woman.³ The age at which a woman’s mother went through menopause is probably a more accurate guide. 

AS A FERTILITY TEST?

For several years now there has been an increasing promotion of the AMH test to women generally as an indicator of their likely fertility – the so called “egg-timer” test.⁴

The premise is simple. Aa low AMH level could indicate that the remaining number of follicles is limited, and that the woman should be encouraged to either get on with her plans for pregnancy or lay down some frozen ova as a reproductive insurance policy.

The converse of this is that women with a normal AMH can perhaps be reassured that time is on their side. Unfortunately, neither of these assumptions holds up when we examine the evidence.⁵

The problem is that while AMH can estimate the numbers of follicles it tells us nothing of the quality of the ova within them. You only need one good ovum to conceive and every IVF doctor can tell stories of natural pregnancies occurring in women with undetectable AMH levels as they wait to start treatment.

Interestingly, a high level of AMH can suggest polycystic ovarian syndrome (PCOS) which can also be associated with difficult conceiving. And the biggest problem of all with the test is that the quality of the ova declines well before the quantity, so a “normal” AMH can be falsely reassuring for a woman in her 40s.

In one 2017 study of nearly 750 healthy women, wanting to conceive and with no known fertility problems the subjects had initial AMH levels estimated and were then followed up for 12 months.⁶

There was no significant difference in the length of time to conception between those with low AMH levels at the beginning of the study and those with normal.

To the surprise of the researchers, this finding also held for the older women in the group who were aged 38 to 44 years. Several other studies have reached the same conclusion.^7,8

THE FINAL MESSAGE

Though invaluable in an assisted conception setting, AMH is a poor indicator of fertility potential in the general population and is really no better that chronological age in this group. Unfortunately, though most Australians are vaguely aware that fertility in women decreases with age, they tend to overestimate the age at which that decline occurs by about 10 years.⁹

The bottom line is that for most women fertility has begun to decrease by age 35, regardless of the AMH level.

Unlike my Irish colleague, I believe that doctors should be actively discouraging their otherwise healthy female patients from taking this test. The results of AMH testing in this group can be both falsely concerning and falsely reassuring. Those are very dangerous qualities in a test on which so much depends.     

Dr Terri Foran is a sexual health physician with a special interest in women’s sexual and reproductive health and particular expertise in contraception, menopause and the management of sexually transmitted infections

References:

1.https://www.sciencedirect.com/science/article/pii/S0015028211003700?via%3Dihub

2.https://www.sciencedirect.com/science/article/pii/S0015028210024106?via%3Dihub

3. https://academic.oup.com/jcem/article/96/8/2532/2834584

4. https://www.mamamia.com.au/the-egg-timer-test-would-you-take-it-and-then-what/

5. https://www.smh.com.au/national/how-the-egg-timer-test-falsely-feeds-women-s-fertility-anxieties-20190104-p50ppb.html

6. https://jamanetwork.com/journals/jama/fullarticle/2656811

7. https://www.ncbi.nlm.nih.gov/pubmed/26406293

8. https://www.ncbi.nlm.nih.gov/pubmed/28393651

9. https://www.sciencedirect.com/science/article/pii/S0015028212023436?via%3Dihub