PCOS and trying to conceive: what can bring back ovulation

You open the cycle app and see a number that cannot be right. Day 62. The month before it was 48, the month before that 39.

The app shows you a fertile window. It guessed it. It measured nothing.

On the desk there are two packs of ovulation tests. Yesterday's strip was positive. The one from the day before too. The one from last week as well.

Many women know this pattern. It feels as if the body keeps getting ready and never arrives.

And with PCOS that is often exactly what happens. Follicles grow, they stall halfway, and the one moment everything is meant to lead up to does not take place.

So I am not starting with advice. I am starting with a piece of reassurance that is documented, and with a sentence that matters just as much: time is a real factor in this topic. Nothing in this article is a reason to postpone a reproductive medicine assessment.

The reassurance first: you are in the group with the ordered path

If you search online for PCOS and trying to conceive, you find two extremes. On one side pages telling you it is all no problem. On the other side forums full of despair.

Neither gets it right. The sober picture is a third one.

Depending on the diagnostic criteria, PCOS affects 8 to 13 percent of women of reproductive age, in some surveys more. The current Cochrane review on letrozole gives a range of 5 to 20 percent, depending on which definition is used. This is not a fringe group. It is one of the most common hormonal diagnoses there is.

What makes PCOS special in fertility medicine: the core of the problem is usually a single, clearly nameable step that does not take place. Ovulation. And for absent ovulation something exists that does not exist for many other causes, namely a step by step therapy that is internationally agreed and has been tested in large randomised trials.

If you want to know how PCOS is diagnosed in the first place and what forms exist, that is covered in detail elsewhere. This article assumes the diagnosis and deals only with the question of how ovulation can come back. For the basics: PCO syndrome, causes and symptoms.

One note belongs here anyway, because it concerns the diagnosis and not the treatment. There are conditions that look very much like PCOS, for example a mild inherited disorder of the adrenal hormones, non classic congenital adrenal hyperplasia. That changes the counselling around fertility. If this exclusion was never explicitly a topic for you, ask about it.

I often read these numbers out differently in the consulting room than they are quoted online. Not as a success rate. Rather as a description of what happened in a trial with a clearly defined population.

And that population was selected. Only women between 18 and 40 with at least one patent fallopian tube, a normal uterine cavity and a partner with a sperm concentration of at least 14 million per millilitre were included. Keep that last point in mind. It comes back in the final section.

What happens in the ovary when ovulation does not occur

Imagine a school class in which one child is to be chosen each month to give a presentation. Normally several put up their hand, one gets the slot, the others sit down again.

In PCOS many put up their hand. But there is nobody who reliably makes the choice. And so they all stay standing, hands raised.

That is exactly what you see on ultrasound. Many small follicles at the edge of the ovary, all at an early stage, none of them dominant. Not too many eggs. Rather many attempts without a decision.

Why the selection does not come about has been measured well. The control signal from the brain arrives at an altered rhythm.

A second amplifier is blood sugar metabolism. Insulin can stimulate androgen production in the ovary and at the same time lower the transport protein SHBG in the liver, which leaves more free testosterone in circulation. How this loop works in detail and what influences it is covered at length here: insulin resistance and female hormones.

How this fast rhythm arises is hard to measure in humans, because that would mean looking into the hypothalamus. There are animal data on this, and they belong clearly labelled.

Step one, lifestyle: what the numbers support and what they do not

There is a sentence almost every woman with PCOS who is trying to conceive has heard. Lose five percent first.

That sentence is well meant. It is not wrong either. But it is less precise than its confident tone suggests, and for a portion of readers it simply does not apply.

For this article I searched deliberately for the primary source of this five percent threshold. I did not find it. What I did find are three studies that together give a considerably more differentiated picture.

The more robust figure comes from a randomised trial, and it is more honest, because it shows not only what improved but also where the difference no longer held.

And if you are slim

A considerable proportion of women with PCOS are of normal weight. For this group the weight recommendation explicitly does not apply. They should not lose weight.

This is not a footnote but a safety question. In the study from Boston, LH fell as body fat percentage rose. That is an observation in a group and not an instruction for action. The authors themselves read their finding explicitly as a continuous spectrum and precisely not as two separate forms of disease. So what counts for you is not this study but the guideline: for women with PCOS who are of normal weight, weight loss is not recommended. Here the treatment path begins directly at step two.

The 2023 guideline also did something I consider one of the most important advances of this edition. It explicitly calls for care regarding weight stigma. Trying to conceive and pressure about weight are a mixture that can seriously harm people.

If you want to read on concretely at this point, without it tipping into calorie arithmetic: avoiding blood sugar spikes, losing weight with insulin resistance and why a calorie deficit alone is often not enough.

And now you know why I go into more detail on step one than most. It is the step with the weakest evidence and the highest emotional price.

Step two, inducing ovulation: why letrozole is in front today

If your mother had fertility treatment in the nineties, the tablet was called clomiphene. If you go to a fertility centre today, in many cases it is called letrozole.

That is not a change of fashion. It is the consequence of a very large trial and a very large review.

The two substances act at different points. Clomiphene blocks estrogen receptors in the brain and thereby simulates an estrogen deficiency for the control centre. The body answers with more FSH. The drawback: the same blockade can also affect the lining of the uterus and cervical mucus.

Letrozole takes a different route. It temporarily inhibits the enzyme aromatase and can thereby lower estrogen production itself. Here too the control centre answers with more FSH, but without blocking the receptors persistently.

What belongs with it: letrozole is prescription only. And it is an agent that is not used in an already existing pregnancy. That is why it is checked before every treatment cycle whether a pregnancy is present, and that is why it is given over only a few days in the cycle and not continuously. Fatigue, dizziness, headache and hot flushes occur. In liver disease and while breastfeeding, separate restrictions apply. All of this belongs in the conversation with the doctor who prescribes it, and no line here replaces that.

One detail of this review I find instructive. In the previous edition from 2018 more studies and more participants were included, 42 studies with 7935 women, and the certainty of evidence was rated moderate. In the 2022 edition there are fewer studies, but the certainty rose to high. The reason: four previously included works were removed because of doubts about data validity.

Fewer data and yet more certainty. That is a good example of why the number of studies is not the decisive thing.

There is a third trial I mention in this section although its main result is negative. Precisely for that reason.

Step three, metformin: the substance with the best reputation and the weakest number

Hardly any medicine has as good a reputation in the PCOS community as metformin. It sounds logical: PCOS often has something to do with insulin, metformin can improve insulin sensitivity, so it ought to do something for fertility.

The logic holds. The numbers are more sobering.

The 2023 guideline draws a differentiated conclusion from this. Metformin alone remains possible, but with the explicit note that more effective agents are available. Clomiphene is to be preferred over metformin monotherapy. The combination of clomiphene and metformin is explicitly provided for in the guideline and can produce better results than either one alone. One point belongs here, because it is spelled out separately for letrozole: the guideline itself states that medicines in PCOS are generally not approved for this condition specifically, and that ovulation induction agents including metformin and clomiphene are off label in many countries. In Germany clomiphene is approved for anovulatory infertility, metformin is not.

That is a clear order, and it matters more than any blanket statement. Metformin is not bad. It simply does not come first when the question is ovulation.

And in pregnancy

The next question comes almost automatically. If metformin is being taken and a pregnancy occurs, continue or not? There is a large trial on this, and it is a textbook example of how easily a result gets summarised wrongly.

Hence the clearest sentence of this section: metformin is prescription only. Whether it is started, continued or stopped, and all the more whether it is continued in a pregnancy, is a medical decision. Nobody stops it on their own, and nobody starts it on their own.

And it comes with conditions. Gastrointestinal complaints are the most frequent side effect and were more frequent with metformin than with clomiphene in the trial above as well. In impaired kidney function it must not be given, because a rare but serious acidification of the blood can otherwise develop. Before examinations with iodinated contrast medium and before operations, separate rules apply. That is why kidney function is checked before the start and regularly afterwards.

One more word on inositol, because the question comes up in every consultation. The Cochrane review on it rates the evidence as low to very low, and the most striking single finding on miscarriage rate rested substantially on one study with an unusually high rate in the control group and did not withstand sensitivity analysis.

Step four, injections and the surgical option

When tablets do not lead to ovulation, this is called clomiphene resistance. The word sounds harsh. All it means is that this one route did not take effect.

From here there are two possibilities, and the guideline places them explicitly side by side, not one after the other.

The first: gonadotropins

Instead of prompting the control centre in the brain, the FSH the follicle is missing is given directly. That is more precise and at the same time more demanding, because in PCOS many follicles can respond at once. This is why monitoring by ultrasound is close.

The second: laparoscopic ovarian drilling

In this procedure a few small points are placed on the surface of the ovary through a laparoscopy. The idea behind it: the androgen producing layer is reduced, the hormonal balance shifts, and afterwards the ovary responds differently to its own control signals.

The procedure has an eventful history. It was the standard for a long time, was then displaced by the medicines, and today it is classified in a differentiated way.

This section needs one clear sentence. Nothing written here is an argument for avoiding an operation that has been recommended to you, nor one for demanding one. The decision is made in the fertility consultation, according to your situation, your age and your course so far.

Step five, IVF and ICSI: the risk that looks different in PCOS

In assisted fertilisation it is not one follicle that gets prompted but many at once. That is exactly the sore point in PCOS.

An ovary with many waiting follicles responds strongly to stimulation. Sometimes too strongly. Then ovarian hyperstimulation syndrome develops, OHSS for short. Fluid leaks out of the vessels into the abdomen, the blood thickens, and in severe cases it becomes dangerous.

This is why this section is not at the margin but in the middle of the text. Not to frighten. Rather so that you know the signs and know that the clinics have long had strategies for it.

There are two further adjusting screws used in fertility centres. Both need explanation, because they are documented with differing certainty.

The first concerns the trigger for egg maturation. Classically hCG is given for this, a hormone that resembles LH and stays in the body for a long time. This long duration of action is exactly what favours hyperstimulation syndrome. Alternatively a GnRH agonist can be used, which briefly triggers the body's own LH peak and disappears again quickly.

The second adjusting screw is metformin as an accompanying medication during treatment. A Cochrane review of 13 studies and 1132 women found a lower hyperstimulation risk across all protocols, risk ratio 0.46 from eleven of those studies with 1091 women, though with low certainty of evidence and with markedly more side effects, risk ratio 3.35. An advantage for live birth was not documented, and in one of the two protocols a single study even pointed the other way.

Seeing ovulation when the cycle is barely a cycle

This is the section I found in almost no German language text. Everyone explains the step by step plan. Hardly anyone explains how a woman with a 62 day cycle is supposed to find out whether anything happened at all.

And the question is practical. Without this information you know neither whether a treatment has taken effect nor when there was a fertile window.

Why the test strip can mislead in PCOS

An ovulation test measures LH in urine. It is built to turn positive above a certain concentration, because it is meant to detect a short peak.

Now remember the study from Boston. In three out of four anovulatory women with PCOS, LH was permanently above the 95th percentile of the healthy comparison group anyway. A strip that measures exactly this hormone then looks permanently positive.

On top of that comes something that surprises many people. Even in unremarkable cycles, LH test strips are less accurate than their reputation suggests.

What carries instead

Cycle length itself is, by the way, a crude but useful signal. If your cycles shorten from 90 to 45 to 34 days, that is an observation which means something, even without documenting every single ovulation.

Two sentences on cycle based nutrition, because the question comes up often. The phase logic assumes ovulation, because it orients itself to the second half of the cycle. Without ovulation it is only applicable to a limited extent, and the text on it says so too: cycle based nutrition across the phases. Which hormone values are sensibly measured when is covered here: hormone testing in women.

The half that is almost always missing: the partner, the pregnancy, the environment

Up to here it has been about you. That is the perspective of almost every text on this topic, and it is incomplete.

One note in advance for everyone whose diagnosis came shortly after stopping the pill: the cycle often needs time afterwards, and what is normal in this phase and what is not is covered at length here: coming off the pill, what happens afterwards. The timing of stopping belongs in the conversation with your gynaecologist, especially when fertility treatment is planned in parallel.

First: the partner

A semen analysis belongs at the beginning of the assessment, not at the end.

That is not an opinion of mine. The German language S2k guideline 015-085 lists basic andrological assessment as a component of the couple assessment before assisted reproduction. Not as a subordinate step but as part of the beginning. The guideline is from 2019, its validity has expired according to the register and it is under revision. That changes nothing about this basic statement.

Then there are orders of magnitude. A widely cited review from the Cleveland Clinic assumes that in 20 to 30 percent of cases a male factor alone is present and in a further 20 to 30 percent a combination. This work extrapolated from female data for regions without their own data, which is a methodological weakness and belongs with it.

The strongest argument, however, sits elsewhere, and you already know it. The large letrozole trial only included couples with a sperm concentration of at least 14 million per millilitre. In plain terms: all the numbers this article works with come from populations in which the male side had been checked beforehand.

A semen analysis is quick, inexpensive, without a procedure and without anaesthesia. It costs one morning. What influences sperm quality and what can be changed about it is covered at length here: sperm quality and fertility.

Second: the pregnancy

Third: why this practice also looks for environmental factors

Here I come to the point I explain most often in the consultation. And I say up front what it rests on and what it does not: I am not aware of a study showing that environmental diagnostics in PCOS changes the chance of a pregnancy. What there is, are observational data and mechanisms. I still look. I do not promise you a result for it, and it is no reason to postpone anything.

When you sit in my consulting room and PCOS is in the room, I ask about things that are rarely asked about elsewhere. About damp in the home. About amalgam. About plastic packaging, till receipts, cosmetics. About signs of chronic inflammation.

This is not a reproach to gynaecology. The gynaecological and reproductive medicine assessment is the foundation, and it stays that way. Training priorities and time budgets in routine care are set differently, and the guidelines simply do not offer a recommendation on environmental factors in PCOS. The reason for that is understandable: guidelines work according to GRADE and need intervention studies with patient relevant endpoints. For environmental interventions in PCOS such studies do not exist.

What does exist are two other levels of data. And I separate them cleanly here.

The second level is the mechanistic one. The second scientific statement of the Endocrine Society on endocrine disruptors describes why these substances can do anything hormonally at all: dose response relationships that do not run linearly, effects in the low dose range and a particular vulnerability during developmental windows. The authors state the limitation themselves. Causality is supported by animal experiments and is compatible with correlative human data, but one has to remain cautious about deriving causality from human data.

It explicitly does not follow from this that every hormonal disturbance has an environmental cause. This sentence matters enough to me to set it in bold. PCOS has a strong genetic component, and most of those affected find nothing conspicuous in their surroundings.

What I can tell you is something more modest: there are reasons to look. Asking the question costs nothing. If an investigation comes out of it, that is a self pay service in a private practice, and what it costs we tell you beforehand and in writing. Whether it is worth it in your situation we discuss before and not afterwards. On each of the topics two sentences and then the pointer to where it is covered in full.

On endocrine disruptors in everyday life, meaning plasticisers, bisphenols and what accumulates on a normal day: xenoestrogens in everyday life. On mould toxins that can themselves bind in an estrogen like way, zearalenone above all: zearalenone as a mycoestrogen. If damp is an issue in your home, it starts with the building and not with laboratory values: mould in the home, and for pregnancy separately: mould in pregnancy. On heavy metals and the question of which measurement is meaningful at all: measuring heavy metals in blood or urine.

Three things you can raise at your next appointment

No protocol, no doses, no order that is right for everyone. Just three questions that can change the appointment.

And if the environmental topic occupies you, take it along as a fourth question. Not as a condition. As an addition that runs alongside.

Frequently asked questions

Where this topic connects to the rest

PCOS rarely stands alone. Blood sugar, thyroid, stress and environment are part of it, and trying to conceive is a matter for two people. Depending on where you stand right now, read on here.