IVF Success Rates by Age: What the Registries Report, Cycle by Cycle

Every autumn the same thing happens in fertility clinic waiting rooms: someone opens a national registry report on their phone, reads a single percentage, and decides in thirty seconds whether to hope. This year the spike in searches for IVF success rates by age has an extra push. As of September 2026, the annual national summary of assisted reproduction cycles has landed, the online success estimator has been refreshed with newer cycles, and a short video claiming that “90% of your eggs are gone by 30” has been viewed tens of millions of times.

Each of those things is partly true and easily misread. A registry number is only useful if you know what it counts, which denominator it uses, and how much of it really is about age rather than everything else that comes with age.

So this is the long version. Not the headline figure, but the cycle-by-cycle arithmetic behind it, graded by how strong the evidence actually is.

What changed recently in how IVF success rates by age are reported

Three developments explain the current wave of interest, and none of them is a breakthrough drug. The first is routine: under a US law passed in 1992, every fertility clinic performing assisted reproductive technology (ART, meaning any treatment where eggs and embryos are handled outside the body) must report each cycle to the CDC, which publishes a national summary and a searchable clinic-level report every year. That release cycle reliably sends people searching, because it is one of the few places where outcomes are audited rather than advertised.

The second is a quiet shift in what the registries choose to headline. For years the leading figure was the live birth rate per embryo transfer, meaning the percentage of times an embryo placed in the uterus resulted in a baby. The CDC national summary now also foregrounds the percentage of intended egg retrievals that eventually produce a live birth, counting every fresh and frozen transfer from that single stimulation. The NHS, drawing on the UK regulator’s registry, still quotes per-embryo-transferred figures by age band. Both are correct. They are simply different questions, and a reader comparing a 32% figure from one page to a 53% figure from another is usually comparing denominators, not clinics or countries.

The third development is the CDC’s IVF Success Estimator, an online tool built on those same registry cycles that lets a person enter age, height, weight, diagnosis and prior pregnancies to see a modeled probability. Each refresh draws on a more recent block of reporting years, and each refresh generates a new round of screenshots and arguments about whether the number is “right.” It is a population estimate, not a personal forecast, and the CDC says so plainly on the tool itself.

What has not changed is the shape of the curve. Across every registry, in every reporting year since the 1990s, the age of the egg provider remains the strongest single predictor of whether a cycle ends with a baby.

Why does age matter so much for IVF success?

Two biological facts, working together, produce the curve. The first is quantity. A person is born with a finite pool of immature eggs, and that pool, called the ovarian reserve, only shrinks. No new eggs are made after birth. The second, and more important for IVF, is quality. As eggs age, the machinery that separates chromosomes during the final stages of maturation becomes less reliable, so a growing share of eggs end up with an extra or missing chromosome. That condition is called aneuploidy, and an aneuploid embryo usually either fails to implant or ends in early miscarriage.

The quality effect is the one that quietly dominates the statistics. A 41-year-old and a 31-year-old can go through identical stimulation, produce the same number of eggs, watch the same number fertilize and reach the blastocyst stage (the five- or six-day-old embryo with roughly 100 cells that clinics prefer to transfer), and still face very different odds. The embryos look similar under a microscope. Chromosomally, they are not. Laboratory series that test embryos before transfer consistently find that roughly one in four blastocysts is aneuploid in the early thirties, rising to well over half by the early forties and higher still after 43. Those are observational lab data, but they are large and remarkably consistent across countries.

Quantity matters at a different point in the cycle. Fewer eggs retrieved means fewer chances for one of them to be the chromosomally normal embryo. That is why cumulative live birth rates, which count every embryo from one retrieval, fall more steeply with age than the chance of any single good-looking embryo implanting.

The uterus, by contrast, ages gracefully. Donor-egg cycles, where a younger person’s eggs are fertilized and transferred into an older recipient, produce live birth rates that barely move between recipients in their thirties and their late forties. The Mayo Clinic and the Cleveland Clinic both describe this pattern; it is the cleanest natural experiment we have that the problem lives mostly in the egg, not the womb.

IVF success rates by age: what the registries report, in one table

The table below sets two registry perspectives side by side. The first column uses the per-embryo-transferred live birth rate the NHS quotes from UK registry data for 2019, the most recent full year on that page. The second column summarizes the CDC’s national figures for cycles using a patient’s own eggs, expressed as the percentage of intended egg retrievals that led to at least one live birth after all fresh and frozen transfers from that retrieval. The CDC groups ages slightly differently and updates annually, so those figures are given as approximate ranges rather than false precision.

Age of egg provider	Live birth per embryo transferred (UK registry, via NHS)	Live birth per intended egg retrieval, cumulative (CDC national summary, own eggs, approx.)
Under 35	32%	About 50–55%
35–37	25%	Around 40%
38–40	19% (38–39)	Roughly one in four
40–42	11%	Roughly one in eight (41–42)
43–44	5%	Under 5% (over 42)
Over 44	4%	Under 5% (over 42)
Any age, donor eggs	Not age-banded	Roughly half of transfers, largely independent of recipient age

Read across a row and the two columns tell one story from two angles. Under 35, a single embryo has about a one-in-three chance, but because a retrieval typically yields several embryos, the whole retrieval carries better-than-even odds. After 42, both columns converge on a hard number: fewer than one cycle in twenty ends in a live birth with a person’s own eggs, and the reason is almost entirely the aneuploidy curve described above.

Two cautions before anyone circles a number. These are population averages that blend every diagnosis, every clinic and every embryo grade. And the UK figures are per embryo, so a transfer of two embryos, now uncommon, is counted twice in the denominator.

Per transfer, per retrieval, or cumulative: which IVF success rate should you read?

Most confusion about IVF statistics is a denominator problem, so it is worth slowing down here. A cycle has several places where you could stop counting, and each gives a legitimately different rate.

The live birth rate per cycle started counts everyone who began stimulation, including the roughly one in ten who never reach retrieval because the ovaries respond poorly or too strongly. It is the most conservative figure and the one that best answers “if I start, what are my odds this time?”

The rate per egg retrieval drops the cancelled cycles from the denominator and asks what happens once eggs are actually collected. The rate per embryo transfer goes one step further, counting only people who had an embryo to transfer. This is the number that looks best on a clinic website, and it is also the one that silently excludes the cycles that failed earliest, which are disproportionately the cycles of older patients.

The cumulative live birth rate per retrieval, the figure the CDC now emphasizes, counts one stimulation and every fresh or frozen transfer that follows from it, sometimes over a year or more. It is the fairest single measure of what one round of stimulation is worth, and it is the reason the CDC’s under-35 figure sits so far above the UK per-embryo figure.

Then there is the cumulative rate across multiple retrievals, which large observational cohorts from the UK and Australia have reported. In those datasets, a person under 35 who continued through three complete cycles reached a cumulative live birth rate in the range of 60–70%, while someone starting at 40–42 plateaued closer to 30% even after several rounds. These are observational data with real survivorship bias, since people who stop are often those with the poorest prognosis, but they are the closest thing to an answer for “if I keep going, where does this end?”

The rule of thumb: whenever you see a percentage, find the sentence that says “per” something. If it is not there, the number is decoration.

What the evidence actually says, graded by strength

Not every statement in this article carries the same weight, so here is an honest audit.

Strong, consistent registry evidence. The decline in live birth rates with the age of the egg provider is documented in mandatory national registries covering millions of cycles across the US, UK, Europe and Australia over three decades. Registries are observational, not randomized, but the effect is so large, so consistent and so biologically plausible that no serious body disputes it. The same registries show that donor-egg outcomes track the donor’s age, not the recipient’s.

Strong laboratory and observational evidence. The rise in embryo aneuploidy with maternal age comes from very large series of embryos tested before transfer. The direction and rough magnitude are settled; the exact percentage at any given age varies by laboratory technique and patient population.

Moderate evidence, randomized. Whether genetically screening embryos before transfer (PGT-A, preimplantation genetic testing for aneuploidy) improves the chance of a baby is a question that has been put to randomized trials. The largest found no improvement in cumulative live birth per retrieval overall, with a possible per-transfer benefit in the 35–40 age band and no clear benefit in younger patients. Expert bodies describe the evidence as mixed and advise individualized discussion.

Moderate evidence, randomized. Transferring one embryo at a time rather than two slightly lowers the chance of a birth from that single transfer but sharply cuts twin pregnancies and their complications, with equivalent cumulative live birth when the second embryo is transferred later. This is why single embryo transfer is now the default in most guidelines.

Weak or absent evidence. Most treatment “add-ons” marketed to improve implantation, from endometrial scratching to various supplements, have either failed in randomized trials or never been tested in them. The UK regulator maintains a public rating system in which the majority of add-ons are graded as lacking evidence of benefit.

Expert opinion. Advice about how many cycles to attempt at a given age, and when to consider donor eggs, rests on registry trends plus clinical judgment rather than trials. Reasonable clinicians differ, and the decision belongs with the treating team and the patient.

What stage of IVF is most likely to fail?

People imagine IVF failing at one dramatic moment, usually the pregnancy test. In reality it fails as a funnel, and knowing the shape of the funnel is more useful than any single percentage.

Consider a typical retrieval that yields ten eggs. Not all will be mature; on average around eight are. Of those, roughly six to seven will fertilize when sperm is introduced, whether by conventional insemination or by ICSI (intracytoplasmic sperm injection, where a single sperm is injected directly into the egg). Of the fertilized eggs, perhaps half, three or four, will keep dividing normally for five or six days to reach the blastocyst stage. And of those blastocysts, the share that is chromosomally normal depends almost entirely on age: perhaps three of four in the early thirties, one of two around 38, one of four or fewer in the mid-forties.

Then comes the step where the largest absolute number of cycles is lost: implantation. Even a chromosomally normal, high-grade blastocyst transferred into a receptive uterus implants and progresses to a live birth only about 60–65% of the time in large observational series. Nobody fully understands the remaining third. Some of it is subtle embryo abnormality that testing cannot detect, some is the endometrium, some is simply the ordinary inefficiency of human reproduction, which is inefficient in nature too.

So the honest answer has two parts. Measured by where individual embryos are lost, the sharpest attrition is between fertilization and blastocyst, and then between blastocyst and chromosomal normality in older patients. Measured by where a cycle that reached transfer most often ends without a baby, the answer is implantation.

The funnel also explains why a 28-year-old and a 42-year-old with the same egg count have such different odds. The 28-year-old has more embryos surviving each narrowing of the funnel, so a single retrieval more often yields a second or third frozen embryo to try if the first transfer fails. For the 42-year-old, the funnel frequently ends with zero.

At what age are 90% of your eggs gone? The viral claim, checked

The clip that is currently circulating is not invented. It traces back to a 2010 mathematical model, published in PLoS One and indexed in PubMed, that pooled histological counts of ovarian follicles from hundreds of ovaries across the lifespan. The model estimated that the ovaries hold a peak of several million immature eggs around the fifth month of fetal life, roughly one to two million at birth, and a few hundred thousand at puberty. By age 30, the model put the remaining pool at about 12% of the pre-birth peak, and by 40 at about 3%.

So yes: by the late twenties, around 90% of the eggs a person ever had are gone. But look at what that sentence measures. The 90% is calculated against a fetal maximum that was never available for reproduction. Most of the loss happens before birth and before puberty, through a normal process of programmed cell death. A 30-year-old with 12% of her peak pool still has, by that same model, something on the order of 100,000 immature eggs, and only one is needed each month.

What the viral framing gets wrong is the implied link between percentage remaining and fertility. Natural fertility does not drop 90% by 30; it declines gently through the early thirties, more steeply after about 35, and sharply after 40. That curve tracks egg quality, the aneuploidy problem, far more closely than egg count.

Count still matters for IVF in a practical sense, because stimulation can only recruit from the follicles available in a given month, and a smaller reserve usually means fewer eggs retrieved. Ovarian reserve is estimated clinically with an antral follicle count on ultrasound and a blood test for anti-Müllerian hormone (AMH), a protein released by small follicles that roughly tracks how many are waiting. A low AMH predicts fewer eggs per retrieval. It does not predict the quality of those eggs, and it is a poor predictor of natural conception in people who are not yet trying.

The accurate one-line version: most eggs are gone before you are born, and the ones that matter for IVF are judged by their chromosomes, not their number.

What are the chances of IVF working the first time?

This is the question people actually want answered, and it needs an age attached before it means anything. Using registry data for a first complete cycle with a person’s own eggs, counting every transfer from that one retrieval, the picture looks like this: for someone under 35, a little better than a coin flip; at 35–37, around two in five; at 38–40, about one in four; at 41–42, roughly one in eight; and after 42, fewer than one in twenty.

If instead you ask about the first embryo transfer alone, which is what many people mean by “the first time,” the numbers are lower because they no longer include the frozen embryos waiting for a second attempt. The UK per-embryo figures in the table above are the right reference: about one in three under 35, falling to about one in twenty after 43.

Several features make a first cycle more or less likely to succeed, independent of age. A previous live birth, whether natural or through treatment, improves the odds in every registry. A diagnosis of tubal blockage or male-factor infertility, where IVF directly bypasses the problem, tends to carry better outcomes than diminished ovarian reserve or unexplained infertility of long duration. Body mass index at the upper and lower extremes is associated with lower success in observational data; the mechanisms are debated and the effect is modest compared with age.

It also helps to remember what a first cycle is for beyond the pregnancy test. It is the only reliable way to learn how a particular person’s ovaries respond to stimulation, how many eggs mature and fertilize, and how many embryos reach blastocyst. Clinicians routinely adjust the approach for a second cycle based on the first, which is one reason cumulative rates across two or three cycles are considerably higher than a single attempt at any age.

None of this makes the first cycle a rehearsal. For a good share of people under 38 it is the only cycle they will need. It simply means a negative result carries information, not a verdict.

IVF success rates over 40: what the data say and what they don't

The over-40 band is where headline percentages do the most damage, because a single figure hides a cliff inside it. At 40, the cumulative chance of a live birth from one retrieval with a person’s own eggs sits somewhere around one in five to one in four. By 42 it has roughly halved. By 44 the registries report low single digits, and above 45 the number of own-egg live births in any national dataset is small enough that percentages become unstable.

The driver is the aneuploidy curve. Laboratory series report that by 42 or 43, three in four blastocysts carry a chromosomal error, and at 44 or 45 the figure approaches nine in ten. The eggs are not fewer in a way that stimulation can fix; they are less often capable of producing a viable embryo, and no medication changes that.

What the data also show is that the range around the average is wide. A 41-year-old with a good antral follicle count and a prior pregnancy does not carry the same odds as a 41-year-old with low reserve and a decade of unexplained infertility, even though the registry averages them together. Clinic-level reports and the CDC estimator can help separate those cases somewhat, but no model resolves the uncertainty completely.

Two options change the arithmetic for this group. The first is accumulating embryos across several retrievals before any transfer, sometimes called embryo banking, so that the low per-embryo probability is multiplied over more embryos. Observational cohorts suggest this raises cumulative live birth for people in their early forties, at the cost of more cycles of stimulation. The second is donor eggs, which reset the odds to those of the donor’s age, typically a live birth in roughly half of transfers regardless of the recipient’s age.

The decision between those paths is personal, medical and often emotional. Guidelines in several countries ask clinics to counsel patients over 42 frankly about own-egg odds. Whether to proceed, and for how many cycles, is a decision for the patient and the treating specialist together.

Why donor eggs change the picture, and what they reveal about age

Registry data on donor-egg cycles are the closest science has come to isolating the role of the egg from the role of everything else. In the CDC’s national summary, transfers of embryos created from donor eggs result in a live birth in roughly half of cases, and that proportion is nearly flat across recipient ages from the mid-thirties into the late forties. The Mayo Clinic and the Cleveland Clinic both describe donor-egg outcomes as depending primarily on the donor’s age.

That flatness is informative. It tells us that the endometrium, the lining of the uterus where an embryo implants, retains its receptivity well past the age at which a person’s own eggs have largely stopped producing viable embryos. It also tells us that the health conditions that accumulate with age, while they matter for pregnancy safety, are not the main reason own-egg IVF outcomes fall.

The pregnancies themselves are a different matter. Observational studies consistently find higher rates of hypertensive disorders of pregnancy, including preeclampsia, in donor-egg pregnancies compared with own-egg IVF pregnancies, even after adjusting for age. The leading hypothesis involves the immune interaction between the recipient and a genetically unrelated fetus, though the mechanism is not settled. Recipient age adds its own risks of gestational diabetes, hypertension and caesarean delivery. These are reasons for careful obstetric monitoring, not reasons to dismiss the option, and they are part of the counseling any reputable program provides.

Donor eggs are also the reason registries publish own-egg and donor-egg figures separately. A clinic that treats many older patients with donor eggs can show a high overall live birth rate that says little about what a 41-year-old using her own eggs should expect there. When comparing clinic reports, or entering data into the CDC estimator, it is essential to look at the own-egg, age-banded figures rather than the blended headline.

For some people, the decision to move to donor eggs is a relief; for others, a grief. Both responses are reasonable, and most programs offer counseling specifically for that transition.

Frozen embryos, single transfers and genetic testing: do they shift the age curve?

Three changes in laboratory practice over the past fifteen years have altered how IVF is done, and each is sometimes credited with improving success rates. The registry evidence is more nuanced.

Frozen embryo transfer has become the majority of transfers in the US, largely because embryo freezing (vitrification, an ultra-rapid method that avoids ice crystal damage) now has survival rates above 95%. Randomized trials comparing a fresh transfer with freezing everything and transferring later found similar live birth rates overall, with a benefit for freezing in people at high risk of ovarian hyperstimulation syndrome and possibly in those with polycystic ovary syndrome. Freezing does not change the chromosomal status of an embryo, so it does not move the age curve. What it does is make single embryo transfer practical, since remaining embryos can wait.

Elective single embryo transfer, transferring one embryo even when several are available, has cut twin rates dramatically. Twins carry substantially higher risks of prematurity, low birth weight and maternal complications. Randomized trials show that transferring one embryo and then a second one later, if needed, yields the same cumulative live birth as transferring two at once, with far fewer twins. Guidelines in the US, UK and Europe now recommend single transfer for most patients under 38 with good-quality embryos, and increasingly for older patients when embryos have been tested.

PGT-A, testing a few cells from the blastocyst for chromosome number before transfer, is the most debated. The logic is appealing, especially for older patients where most embryos are aneuploid: transfer only the normal ones and avoid failed transfers and miscarriages. The randomized evidence, however, has not shown a higher cumulative live birth per retrieval, because testing cannot create a normal embryo that was not there; it can only reorder which embryo is transferred first. There is a plausible per-transfer benefit and a reduction in miscarriage for the 35–40 group, and some risk of discarding embryos that might have produced a baby, given imperfect test accuracy. Professional societies describe it as an option to discuss, not a routine recommendation. Whether it suits a particular person is a decision for the treating team.

Beyond age: the other factors that move an individual's IVF odds

Age is the biggest variable, but the registry models that power the CDC estimator include several others, and understanding them helps explain why two people of the same age can be quoted very different numbers.

A previous pregnancy, and especially a previous live birth, is one of the strongest positive predictors after age in every registry model. It signals that the biological pathway has worked before. Duration of infertility works in the opposite direction; longer unsuccessful trying is associated with lower success at any age, likely because it enriches for harder-to-treat causes.

The underlying diagnosis matters. Blocked or damaged fallopian tubes, and male-factor infertility addressed with ICSI, tend to carry outcomes at or above the age average, because IVF directly solves the problem. Diminished ovarian reserve carries outcomes below the age average, because it usually means fewer eggs and often reflects accelerated ovarian aging. Endometriosis and uterine factors sit in between and depend heavily on severity.

Body weight appears in the models as body mass index. Large observational datasets show lower live birth rates at both very high and very low BMI, with a modest effect that is dwarfed by age. Randomized trials of intensive weight-loss programs before IVF have generally not shown improved live birth rates, which suggests the association may partly reflect other factors. This is an area where the evidence is weaker than the confidence with which it is sometimes discussed.

Smoking is associated with lower ovarian reserve, fewer eggs retrieved and lower live birth rates in observational data, and with higher miscarriage risk; stopping is one of the few lifestyle changes with consistent support. Sperm quality contributes to fertilization and blastocyst rates, and paternal age above about 45 is associated in observational studies with slightly lower success and higher miscarriage, though the effect is far smaller than the maternal age effect.

Finally, the laboratory and clinic matter, which is why clinic-level registry reports exist. Differences in embryology skill and protocols are real, but they explain much less of the variation in outcomes than the age and diagnosis of the patients a clinic happens to treat.

What gender are most IVF babies? The sex ratio, explained

Among naturally conceived births, about 105 boys are born for every 100 girls, a ratio that has been stable across populations for as long as records exist. IVF births tilt this slightly, and in a direction that depends on how the embryos were created and when they were transferred.

Large registry analyses from Australia, the US and Europe, all observational, find that transfers at the blastocyst stage (day five or six) produce a modestly higher proportion of boys, in the range of 52–55%, compared with about 51% for transfers at the cleavage stage (day two or three). The most cited explanation is that male embryos tend to develop marginally faster in culture, so when embryologists choose the most advanced-looking blastocyst for transfer, they slightly favor males without intending to.

ICSI pulls in the other direction. Cycles in which a single sperm is injected into each egg show a slightly lower proportion of boys, sometimes below 50%, in the same registries. Why is unclear; one hypothesis involves the selection of sperm by the embryologist, another the fact that ICSI is used for male-factor infertility where sperm characteristics may already differ.

So the short answer is that most IVF babies are boys, but only just, and only for certain protocols. The shifts are a few percentage points on a population level and irrelevant to any individual pregnancy, where the chance remains close to even. No mainstream technique in routine IVF selects for sex, and in many countries, including the UK, choosing an embryo’s sex for non-medical reasons is prohibited by law. Where genetic testing identifies sex as a byproduct of screening for chromosomal disorders, laws and clinic policies govern whether that information can be used.

The sex-ratio data are a useful reminder of how observational registries work: they are excellent at detecting small population-level patterns, and poor at explaining them. The mechanisms behind the blastocyst and ICSI shifts remain hypotheses, and the honest evidence grade is observational with plausible but unproven explanations.

Common myths about IVF success rates by age, corrected

“Ninety percent of your eggs are gone by 30, so fertility collapses at 30.” The 90% figure compares the pool at 30 with a fetal peak that was never usable. Most loss happens before birth. Natural fertility declines gently through the early thirties and steeply only after about 37. Egg quality, not count, drives that curve.

“IVF fixes age.” It does not. IVF bypasses blocked tubes, low sperm counts and ovulation problems. It cannot repair chromosomal errors in eggs, which is why own-egg live birth rates fall with age in every registry, including the best clinics.

“A clinic advertising 60% success is twice as good as one reporting 30%.” Almost always a denominator difference. Sixty percent per transfer in patients under 35 with tested embryos and 30% per cycle started across all ages can describe the same laboratory. Compare age-banded, own-egg, same-denominator figures from the national registry, not marketing pages.

“Freezing embryos lowers the chance of a baby.” Modern vitrification has survival rates above 95%, and randomized trials show frozen transfers produce live birth rates equal to fresh transfers overall, with advantages for certain patients.

“Genetic testing of embryos raises IVF success.” Randomized trials have not shown higher cumulative live birth per retrieval. Testing may reduce failed transfers and miscarriages in the 35–40 group by reordering which embryo goes first; it cannot create a normal embryo. Expert bodies call the evidence mixed.

“A low AMH means you cannot get pregnant.” AMH predicts how many eggs a stimulation will yield. It is a poor predictor of egg quality and of natural conception in people not yet trying. Low AMH at a young age still comes with a young age’s chromosomal odds.

“Transferring two embryos doubles your chances.” It raises the chance from that single transfer but increases twin pregnancies, which carry serious risks. Transferring one and then the other later gives the same cumulative live birth rate with far fewer twins, which is why guidelines favor single transfer.

“Success rates are falling because IVF is getting worse.” Per-transfer rates in some registries have plateaued partly because the average patient is older and single embryo transfer is now standard. Cumulative rates per retrieval have generally improved.

When to see a doctor: fertility evaluation and IVF warning signs

Two very different situations bring people to a clinician around this topic, and both have clear guidance.

The first is deciding when to seek a fertility evaluation. The Mayo Clinic, the NHS and the CDC align on the same thresholds: after 12 months of trying without success for people under 35; after 6 months for those 35 and older; and promptly, without waiting, for anyone over 40, anyone with known irregular or absent periods, a history of pelvic infection or endometriosis, prior cancer treatment, or a partner with a known sperm problem. Waiting costs more at 38 than at 28, because the registry curve is steepest in the late thirties and early forties. An evaluation does not commit anyone to IVF; it produces information.

The second is during treatment itself. IVF involves hormone medicines prescribed and dosed by the treating clinic, and a small minority of people develop ovarian hyperstimulation syndrome (OHSS), in which the ovaries over-respond and fluid shifts into the abdomen. Contact the clinic the same day, or go to an emergency department, for any of the following:

• Severe or rapidly worsening abdominal pain or bloating
• Rapid weight gain, for example several pounds in a day or two
• Shortness of breath, chest pain or difficulty lying flat
• Vomiting that prevents keeping fluids down
• Markedly reduced urine output or very dark urine
• Fever, chills or heavy vaginal bleeding after egg retrieval
• Pain, redness or swelling in one leg, which can signal a blood clot

After a positive pregnancy test, one-sided pelvic pain, shoulder-tip pain, dizziness or fainting are warning signs of an ectopic pregnancy, which occurs somewhat more often after IVF than after natural conception, and require urgent assessment.

Emotional distress deserves the same seriousness. Depression and anxiety are common during fertility treatment, and persistent low mood, hopelessness or thoughts of self-harm are reasons to speak to a doctor or the clinic’s counseling service promptly.

Every decision about whether to start IVF, which protocol to use, how many cycles to attempt, whether to test embryos and when to consider donor eggs belongs with the prescribing clinician and the patient together. Never adjust or stop a prescribed fertility medicine without speaking to the clinic that prescribed it.

Frequently asked questions

What are the chances of IVF working the first time?

For a first complete cycle with a person’s own eggs, counting every transfer from that retrieval, registry data suggest a little over 50% under age 35, around 40% at 35–37, about one in four at 38–40, roughly one in eight at 41–42 and under 5% after 42. For the first embryo transfer alone, the odds are lower, about one in three under 35, because frozen embryos awaiting later transfer are not yet counted.

What stage of IVF is most likely to fail?

Implantation, the point where a transferred embryo must attach to the uterine lining, is where most cycles that reach transfer end without a baby; even a chromosomally normal blastocyst produces a live birth only about 60–65% of the time. Measured by embryo attrition, the sharpest losses occur between fertilization and the blastocyst stage, and in older patients between blastocyst and chromosomal normality.

At what age are 90% of your eggs gone?

By roughly age 30, according to a widely cited 2010 model of ovarian reserve, but the 90% is measured against a fetal peak of several million eggs that was never available for reproduction. Most of that loss happens before birth. A 30-year-old still has on the order of 100,000 immature eggs, and fertility at that age tracks egg quality far more than egg count.

What gender are most IVF babies?

Slightly more IVF babies are boys, but only by a few percentage points and only with certain techniques. Large observational registries find that blastocyst-stage transfers produce about 52–55% boys, possibly because male embryos develop marginally faster in culture, while ICSI cycles skew slightly toward girls. For any individual pregnancy the chance remains close to even, and routine IVF does not select for sex.

What are IVF success rates over 40 with your own eggs?

At 40, the cumulative chance of a live birth from one egg retrieval is roughly one in five to one in four; by 42 it has about halved, and by 44 registries report low single digits. The steep drop reflects the rising share of embryos with chromosomal errors, which reaches three in four or more by 43. Donor eggs reset the odds to those of the donor’s age.

How accurate is the CDC IVF success calculator?

The CDC IVF Success Estimator is a population model built from mandatory registry cycles, using age, height, weight, diagnosis and prior pregnancies to give an average probability for people with a similar profile. It is reasonably calibrated at the group level but cannot account for ovarian reserve tests, embryo quality or clinic differences, so it is a starting point for a conversation with a specialist, not a personal prediction.

What is the IVF success rate per embryo transfer by age?

The NHS, citing UK registry data for 2019, lists live births per embryo transferred as 32% under 35, 25% at 35–37, 19% at 38–39, 11% at 40–42, 5% at 43–44 and 4% over 44. These per-embryo figures are lower than cumulative per-retrieval figures because they count each embryo separately and do not include later transfers of frozen embryos from the same cycle.

Does freezing embryos lower IVF success?

No. With modern vitrification, embryo survival after thawing exceeds 95%, and randomized trials comparing fresh transfer with freezing all embryos for later transfer show similar live birth rates overall, with advantages for freezing in people at risk of ovarian hyperstimulation syndrome. Freezing does not alter an embryo’s chromosomes, so it neither improves nor worsens the underlying age-related odds.

Does genetic testing of embryos improve IVF success with age?

Randomized trials have not shown that PGT-A, testing embryos for chromosome number before transfer, increases the cumulative chance of a live birth from one retrieval, because it cannot create a normal embryo that was not already there. It may reduce failed transfers and miscarriages for people aged 35–40 by transferring a normal embryo first. Professional societies describe the evidence as mixed; the decision rests with the treating clinician.

How many IVF cycles does it usually take?

Large observational cohorts show cumulative live birth rates continue to rise across several cycles, reaching roughly 60–70% after three complete cycles for people starting under 35, and plateauing nearer 30% for those starting at 40–42. Many under 38 succeed in the first cycle. How many cycles to attempt depends on response to stimulation, embryo results and personal circumstances, and is a decision for the patient and treating team.

References

• CDC – IVF Success Estimator
• NHS – IVF: What happens and chances of success
• Cleveland Clinic – IVF (In Vitro Fertilization)
• MedlinePlus – In vitro fertilization (IVF)

This article is for general information only and is not a substitute for professional medical advice. Please consult a qualified doctor about your individual situation.