Menstrual Cycle Hormones

Whenever someone asks me about levels of menstrual cycle-related hormones during their cycle and the answer isn’t obvious I go to the figure below from Wikimedia (this article). What I love about it is that it shows the average (bold blue line) and then the biological variability around that average (dark blue shaded area)… and then also how much this can vary between cycles and between women.The figure is based on a study performed in 2006 that measured hormone levels in 20 ‘normally cycling women’ not on birth control pills. The data were reanalyzed in 2014 and these figures created.

One example of when I might refer to these figures is if someone tells me that they’ve just had blood drawn, they don’t know what cycle day they’re on (e.g. have not yet had their first post-HA or first postpartum period), and LH was measured at 17 and estradiol at 215 pg/mL. I can check these figures and see that that most likely corresponds to just before ovulation. If LH is 17 and e2 is 85 pg/mL that might mean that ovulation has just occurred. If LH is 17 and estradiol is 30 pg/mL then I might suggest inducing a bleed and testing a hormone panel to determine if PCOS might be in play.

I think what is unique and particularly helpful in these figures is the inclusion of variability so one has an idea of the typical range for these hormones. Let me know if you have any questions!

Nico

 

 

Do you really ovulate “late”?

Some women, as they recover from hypothalamic amenorrhea, will be worried that they are ovulating “late” in their cycle, maybe Cycle Day 21, 22… but then they get their period around the normal time, maybe CD 28-30.

In some cases, and this was true for me, the issue is the short luteal phase (remember, that’s the time between ovulation and when your period arrives) leading to a seemingly long follicular phase (time between period and ovulation).

When I tracked my ovulations and periods carefully I noticed that my ovulations were routinely 28-30 days apart, as expected. So it wasn’t that the ovulation was messed up, it was more that my period wasn’t aligning with the ovulation as in a normal cycle. A similar phenomenon was observed in a research study where women with short luteal phases were observed to have longer follicular phases (this was not the main focus of the study, but my observation from data reported in Table 2).

My theory is that to some degree, our period and then subsequent ovulation are actually hormonally separate events. The period comes because of a drop in progesterone, that is unrelated to the decrease in estrogen that primes the small increase in FSH to start the new follicle growing. So in a woman with a short luteal phase, the drop in progesterone occurs early, leading to an early period – but that does NOT drive the start of follicular growth, which waits for a few days and then begins, on its regular 28 day cycle.

I know this is confusing, so check out these diagrams. Here’s a normal cycle…the small increase in FSH at the beginning leads to growth of the egg-containing follicle. In the middle of the cycle, around CD12-13, LH spikes leading to ovulation around CD14. After ovulation, the follicular structure collapses into the “corpus luteum” which starts secreting progesterone by around CD16. When progesterone levels fall approximately 12 days later, the period starts, FSH increases, and the whole cycle begins again – with ovulation around CD14 and period around CD28.

Here’s my view of what is happening when one has a short luteal phase. Let’s imagine that the first period in this diagram was induced by Provera and then Clomid is used to start the follicular growth. So the follicular phase proceeds just as above – a small increase in FSH leads to growth of the egg-containing follicle, LH spikes around CD12-13, and ovulation happens on CD14. This is where things aren’t working quite right – the corpus luteum forms, but isn’t making enough progesterone. So the increase in progesterone is lower, and for a shorter amount of time. This leads to an early period – in this diagram, about a week early, for a 6 to 7 day luteal phase.

This leads to what I’m terming the “apparent cycle day” in the figure – menses have started, so ostensibly it’s CD1 again. However. The OTHER hormones involved in the cycle are not at CD1 levels yet. They are still at CD22, 23, 24 levels. The estrogen needs to drop further to instigate the increase in FSH to start the follicular growth… and THOSE hormones are not affected by the lack of proesterone. They continue along their merry way as if it’s CD 22-28, not caring that you’re bleeding already and *think* it’s CD1. So then you get to CD14 and think that ovulation should be happening… but it doesn’t. So you get frustrated (trust me, I know!!) But in reality, your other hormones are on their normal 28-30 day cycle, when when you get to where CD14 would have been if your period had come on time at CD28, that’s when you ovulate. Does that make sense? If not, feel free to drop a question in the comments!

To further illustrate this, here’s a table with my cycle data (this was after my second son was born in September 2008… my first postpartum ovulation was 7/29/2009 while I was still breastfeeding morning and night. I got my period just five days later on 8/3/09, for a four day luteal phase. This cycle isn’t a great example as the cycle is long as is common in initial postpartum or recovery cycles. However the next one (cycle #2) is a perfect example. After only a six-day luteal phase, I got my peiod on 9/15/09 – had it been a normal length LP (e.g., CD14 ovulation, CD28 period start), my period would have come on 9/21/09 instead. I then ovulated on 10/6/09, which based on when my period actually started was apparently CD22… but had my period come when it “should have”… the ovulation would have been CD15.

One anomalous cycle happens on cycle 10 – I had gotten pregnant the cycle before, but unfortunately had a miscarriage that was resolved after two D&Cs. After that I started to use progesterone suppositories to support my luteal phase, which leads to close to normal LPs and pretty close to CD14 ovulation. I did NOT use progesterone on cycles 15 or 16… leading to shorter luteal phase and apparent later ovulation!

You can see from this data set that there’s a fair bit of variability between cycles – compare the days between ovulation and you see that mostly they’re around ~28-30 days, but there are a few that are shorter, and a few that ar longer. I know that some women are like clockwork, but that is certainly not true for me!!

I hope this helps explain why a short luteal phase and longer apparently follicular phase are associated – again, feel free to ask any questions you may have – or please share if this has been your experience as well – or not!

xox Nico

P.S. Check out Chapter 19 in No Period. Now What? for a LOT more information on luteal phases, why they might be short, and what you can do about it!!

How long will my cycles be?

Have you just had your first period party? The utter thrill when you are working to recover from hypothalamic amenorrhea (HA) and you see blood in your underwear is indescribable. I got my first bleed when I was on vacation. It had been really tough because my sister was announcing to our family in South Africa that she was pregnant; we were supposed to be pregnant together but I hadn’t even gotten a period yet, and here she was – Kate and 2/9 as my Uncle put it. I went upstairs to the bathroom and there was my first glorious period. I felt like a million bucks, I had *done* it!!!!  I could finally start trying to get pregnant myself!

After that celebration, though, comes another wait. This one is so tantalizing; you got a period, you hope that if you just to keep up what you’ve been doing you will  ovulate again and get another… or if you’re trying to get pregnant, you’ll get that oh so magical positive test instead. But for many women newly recovered from HA, cycle day 14 (when ovulation happens in a “normal” menstrual cycle) comes and goes, with nary a fertile sign. This can feel incredibly defeating (I know – in my case I randomly had an ultrasound scheduled on CD13 of my next cycle and we saw a nice juicy 13mm follicle which had my heart soaring, until I went back two days later and it hadn’t grown, and my doctor told me we’d need to do injections…).

But it is extremely common in women with HA for it to take a few cycles for the length to normalize, and this is one reason we say in No Period. Now What? to wait at least three cycle before making any changes in terms of eating, exercise, or stress.

I think that it is helpful to know exactly how common longer cycles are, and how long it typically takes for them to normalize, so I graphed the cycle length data I had from the surveys I did for NPNW. I took the data from each person who had at least one natural cycle, and ended up creating four graphs, split by number of cycles to get pregnant (most of the women who took the survey were trying to get pregnant).

Occasionally a woman will get pregnant on her very first ovulation–before getting her first period–in which case I do not have cycle length data (this happened in about 5% of women who ovulated naturally prior to any treatment). Other than that, in the data from my survey respondents, it took between one additional cycle to up to 12 to get pregnant the first time (in the women who provided cycle length data), so I split the graphs into one cycle to get pregnant (so these are women who had one period then got pregnant on their next ovulation), between 2-4 additional cycles, 5-7 cycles, and 8-12 cycles.

Here’s the first one – those women who took only one additional cycle to get pregnant. You can see that the cycle day of ovulation ranged from CD13 to CD63. (These were natural cycles as my requirement for using data for these graphs was that there was at least one natural cycle; there were certainly people who got pregnant on their first oral med, inject, or IVF cycle, but I will leave that for another day.)

The next graph shows those for whom pregnancy took 2-4 cycles… oh, but first here’s the figure from the book that shows the cumulative pregnancy rate by cycle so you have an idea of how long it is likely to take (if you desire pregnancy at the moment). Not everyone supplied cycle length data, so the numbers on these cycle length graphs don’t give a good sense of how long it took on average. Of my survey respondents, 56% achieved pregnancy within the first three cycles, and 84% within the first six. This is very similar to other studies of similar nature.

Back to those who took 2-4 additional cycles after their first period to get pregnant. This graph is a little more complicated, but if you just look at the first versus second cycle you can see a significant decrease in time to ovulation for those who had long initial cycles. The one case where cycle length increased, the woman in question decided to increase her exercise amount. You can see why we suggest not doing that right away, her cycle length increased from 45 to 60 days. This graph also includes some information about cycle lengths for those who used either oral medications (i.e. Femara or Clomid), or injectables. Some started with treatment but then after failed cycles got pregnant naturally (like the one woman who had ovulation on CD12 with injects, red triangle, on her first cycle, also used injects for cycles 2 and 3, but then got pregnant naturally cycle #4). Others had a natural cycle but then moved to treatment, like the woman whose first post-period ovulation was on CD70, reduced to CD24 with Clomid (and pregnant, filled blue square). Another example of this is the woman whose first post-period ovulation was at CD45, increased exercise and up to CD60, decreased exercise again and ovulated CD45, then used Clomid, ovulated CD18 and was pregnant.

The next graph shows those for whom pregnancy took between 5-7 additional cycles after their first period. Again you can see the commonality of a marked decrease in cycle length for those who had long initial cycles, and also a trend toward shorter cycles when oral meds were used. In this group were seven women who initially used Femara or Clomid to ovulate, but then after failed cycles stopped taking the meds and cycled naturally. Note that the filled red triangle indicates a pregnancy achieved on an injectable cycle (adding the legends to the graphs was taking me too much time).

Finally, those for whom it took between 8-12 cycles to achieve pregnancy. In this case the bump down to “CD 0” indicates no ovulation on that particular cycle. Here you see somewhat of an anomaly, with one woman whose first post-period ovulation was on CD63, which didn’t decrease until she started on oral meds, and another whose subsequent cycles were longer than her initial one. However, as you can see from the other graphs, this pattern is definitely not the norm. In either of these cases I probably would recommend oral meds (assuming that eating and exercise were not more restrictive). And again, filled red triangle indicates pregnancy achieved through injectables, black triangle shows a miscarriage on an injectable cycle.

I know that this is a lot of information, kudos if you’ve made it this far 🙂 I am happy to answer questions, please leave a comment if there’s anything unclear or another way you’d like me to analyze the data.

<3 Nico