Egg Quality and Age — What You Can Change, and What You Cannot (2026)

Egg quality is the chance that an egg, once fertilised, becomes an embryo with the right number of chromosomes. Age is the main thing that changes it, it cannot be measured before the egg is collected, and nothing has been shown to reverse the age effect.

Researchers attribute the fall in live births with age to two separate things: a higher proportion of chromosomally abnormal eggs, and fewer eggs left (the background to Bentov 2014). The first is quality; the second is quantity, or ovarian reserve. They are often blurred together, and the confusion runs both ways.

Because no test grades an egg before it is used, every figure about egg quality is really a proportion: of the embryos tested in IVF labs at a given age, what share had the wrong number of chromosomes (aneuploidy). That is why the numbers below all come from IVF patients, not the general population.

Steadily, then steeply. The largest dataset, 15,169 embryo biopsies in the United States (Franasiak 2014), found aneuploidy was lowest between ages 26 and 30, rose predictably after 26, and was above 40% even in women aged 23 and under. The figure that matters most to a person planning treatment is the chance of having no normal embryo at all among those tested: 2 to 6% for women aged 26 to 37, 33% at 42, and 53% at 44.

Indian data point the same way. A Bangalore series of 1,501 embryos from 488 patients tested between 2016 and 2020 found 46.8% aneuploid overall - about 28% for women under 30, rising to about 67% above 40 (Nair 2022). A Mexican clinic's 1,341 embryos gave almost the same curve, 26.9% under 30 to 62% over 40 (Ponce-Nájera 2025).

One correction worth making, because it circulates widely and appeared on an earlier version of this page: that aneuploidy rises "about ten percent a year". The Mexican paper reported an odds ratio of 1.10 per year of age - the odds rise by about a tenth each year, which is not the same as ten percentage points. Its own figures, 27% to 62% across more than a decade, show the difference.

Read the other way round, even above 40 about one tested embryo in three was normal in the Bangalore series. The usual problem is not that a normal embryo cannot exist; it is that more eggs are needed to find one, and the chance of a cycle producing none rises sharply in the early 40s.

No. AMH and the antral follicle count estimate how many eggs are likely to respond to stimulation - quantity - not whether those eggs are chromosomally normal. NICE's 2026 fertility guideline (NG257) uses them to predict ovarian response (1.18.4), advises against using AMH to predict natural conception (1.18.3), and names age as the initial predictor of the chance of pregnancy (1.18.2).

The research behind that is direct. Among 750 women aged 30 to 44 trying naturally, 84% of those with a low AMH conceived within 12 cycles against 75% with a normal AMH - not a significant difference (Steiner 2017). Pooling data from 5,705 IVF patients, AMH and follicle count predicted a poor response to stimulation well, but neither added anything to age in predicting an ongoing pregnancy (Broer 2013).

So a woman of 42 with a reassuring AMH has the aneuploidy figures of a 42-year-old, and a woman of 31 with a low AMH has fewer eggs, not worse ones. More on what the number does measure is in our guide to AMH and what diminished ovarian reserve means.

After collection, the lab records how many eggs are mature, but it cannot see their chromosomes. Testing happens later, on the embryo, through PGT-A. That tests embryos rather than improving them, and NICE NG257 1.48.1 (2026) advises against offering PGT-A to improve live birth rates; our PGT-A evidence review sets out why. In India, the ART Act 2021 s.25(1) says pre-implantation genetic testing "shall be used to screen the human embryo for known, pre-existing, heritable or genetic diseases only"; how that applies to screening for age-related chromosome errors is a question for the clinic and the regulator, not one this page can settle.

Nothing has been shown to reverse the age-related rise in chromosome errors. Most supplement and hormone trials measured pregnancy or response to stimulation rather than egg chromosomes, and part of this literature has had to be withdrawn.

• CoQ10. The only randomised trial that measured egg aneuploidy gave 600 mg or placebo to women aged 35 to 43. It was stopped early over safety concerns about the egg biopsy it used, with 24 women completing (10 vs 14): aneuploidy was 46.5% against 62.8%, not a statistically significant difference, and the authors call the trial underpowered (Bentov 2014). A later review concluded the data "do not clearly prove" a benefit to human egg quality. Our CoQ10 article covers the pregnancy trials, including a widely cited one that has since been retracted.
• DHEA. The 2024 Cochrane update (28 trials, mostly in poor responders) found DHEA "likely results in little to no difference" in live birth or ongoing pregnancy. ESHRE's 2025 guideline does not recommend it for low or for normal responders. An earlier version of this page quoted a more favourable, older estimate; it has been superseded.
• Testosterone. Here the two bodies disagree. The same Cochrane update rated testosterone before IVF as likely to improve live birth in poor responders (moderate certainty, 8 trials, 716 women); ESHRE 2025 still says it is probably not recommended. Neither looked at egg chromosomes, and it is a prescription hormone - a conversation for the clinician managing the cycle, not a supplement decision.
• Growth hormone and myo-inositol. ESHRE 2025 says growth hormone is probably not recommended for low responders and not recommended for normal responders, and that myo-inositol is not recommended for low responders or for women without PCOS.
• Antioxidants in general. The Cochrane review (63 trials, 7,760 women) rated the evidence low to very low certainty. In March 2026 its editors noted that seven of the included studies had been retracted and two carried expressions of concern, while retaining confidence in the conclusions.

Not every reviewer reads this literature the same way. A 2024 narrative review in Reproductive BioMedicine Online (Hart) was more favourable to starting DHEA and CoQ10 before a cycle in poor responders. The Cochrane review and the ESHRE guideline, which grade the trials formally, are the stronger evidence, and they do not support it.

Mostly the number of eggs it takes. With a higher share of abnormal embryos, more eggs are needed to reach one normal embryo, and sometimes more than one cycle. That changes how a cycle is planned more than whether it is worth attempting.

It also has a cost, and in India most of it is paid out of pocket. In an ICMR-NIRRCH study of 148 couples at five tertiary hospitals, the mean out-of-pocket spend per IVF cycle was ₹2,30,668 in private hospitals and ₹1,10,104 in public ones, drugs made up 55% of it, and about 30% of couples faced catastrophic health expenditure (Patil 2026). Those are IVF cycle costs, not egg-freezing prices.

Time is part of the same decision. Spending months on a supplement plan before starting treatment is not a neutral wait: on Franasiak's figures, the share of women with no normal embryo moved from 33% at 42 to 53% at 44. Under the ART Act 2021 s.21(g), clinics provide ART to women above 21 and below 50.

None of this means IVF is the default answer. Whether it is indicated depends on how long you have been trying, the tests on both partners and the cause; sometimes the answer is testing, timing or a simpler treatment. The natural-conception numbers by age are set out separately in female fertility in your 20s, 30s and 40s.

Freezing keeps eggs at the age they were frozen, so the age at freezing matters more than anything else. It does not make a baby certain, and in the largest follow-up most women had not come back to use their eggs.

The clearest outcome data come from a Spanish programme in which 1,468 women froze eggs for reasons other than cancer and 137 (9.3%) came back to use them (Cobo 2016). Live birth per returning woman was 50% for those who froze at 35 or under and 22.9% at 36 or over. For the younger group, the cumulative chance rose sharply from 15.4% with 5 eggs used to 40.8% with 8, levelling off at 85.2% with 10 to 15; for women 36 and over it levelled at 35.6% with 11 eggs. The authors' conclusion was that at least 8 to 10 mature eggs are needed and that women freezing for age reasons should come younger. These are small groups with wide margins of error.

The UK regulator, the HFEA, says most patients under 38 have about 7 to 14 eggs collected in a cycle, that frozen-egg success numbers "tend to be quite low". In India, the ART Act 2021 covers freezing gametes for "infertility, disease or social or medical concerns". The process is on our egg freezing page, and costs and the law in more detail in egg freezing in India.

A useful test for any claim about egg quality is to ask what it was measured against.

• Better results in animals are not changes in human live births. The CoQ10 work in aged mice (Ben-Meir 2019) is real and is routinely quoted as if it were a human result.
• More eggs collected is not better eggs. It is more chances.
• A good embryo grade is not a chromosome test.
• A reassuring AMH at 41 does not change the aneuploidy figures for that age.
• A trial showing a sizeable difference without statistical significance has not shown a benefit. It means chance could not be ruled out, usually because the trial was too small.

Wanting to do something when the main factor is out of your hands is an ordinary human response, and nothing here is a verdict on anyone's chances. If a supplement is affordable and your clinician sees no reason against it, taking it alongside treatment is a personal choice; the thing worth guarding against is delaying treatment for it.