Summary: If you are not ovulating naturally even after working on recovery, and want to use medication to induce ovulation for pregnancy, Femara (letrozole) is preferable to Clomid (clomiphene)*.
Letrozole, By MindZiper – Own work, CC0, https://commons.wikimedia.org/w/index.php?curid=15991603
In No Period. Now What? (NPNW), our book on hypothalamic amenorrhea recovery, Chapter 21 covers the oral medications that can be used to encourage ovulation. We discuss how soy isoflavones, Femara, Clomid, and tamoxifen reduce estrogen levels to encourage an increase in follicle-stimulating hormone (FSH) that in turn leads to growth and maturation of eggs. These medications can be used when pregnancy is desired, but also to “jump-start” menstrual cycles*. It is really important to note that these meds are unlikely to work without progress having made toward recovery in the form of increased eating, reduced exercise, and reduced psychological stress – all of which you can read about in earlier sections in our book.
Clomiphene, by MarinaVladivostok (Own work) [CC0], via Wikimedia Commons
We also cover research comparing pregnancy rates, likelihood of a multiple-gestation pregnancy, uterine lining thickness, and other metrics between Femara (letrozole) and Clomid (clomiphene citrate) to help women decide which choice is optimal for them. (Tamoxifen is an alternate that is rarely used, so there is not a large body of research to reference, although in many ways it is preferable to Clomid based on fewer effects on the uterine lining.) We also discuss dosing recommendations, what to do if the first cycle doesn’t work, and much more. 🙂
Our conclusion in NPNW is that if one is trying to get pregnant, Femara is likely the better option. Lining tends to be thicker, multiple-gestation rate tends to be lower, and pregnancy rates tend to be higher. Increases in pregnancy rates were observed in both individual studies and meta-analyses, however, the higher pregnancy rates with Femara were not statistically significant. That is, there is a trend toward a larger chance of pregnancy with Femara than Clomid, but that trend isn’t strong enough to argue that it is a sure thing.
Given that the vast majority of research for the book was performed in 2014, I wanted to see if there were any additional updates that might sway the choice in one direction or the other.
I will discuss some of the individual studies below, but based on the results I want to update our recommendations to more strongly encourage use of Femara (letrozole) over Clomid (clomiphene) when pregnancy is desired. While there was one study in which pregnancy rates were marginally higher with Clomid, an two meta-analyses found a statistically significant increase in pregnancy with use of letrozole, and the other studies either showed no difference in pregnancy rate, or a higher rate with letrozole.
A review written by Klement and Casper is an excellent document to bring to your doctor to help make the case for Femara. Granted, as with many of the studies, the target population for the review is women with PCOS, but I strongly believe that the effects and results will be similar in women with HA. This is supported by data from my own survey of women with HA. Also, based on the mechanism of action of Femara, it make sense that Femara would work better in women with HA just as it clearly seems to in women with PCOS.
The most recent meta-analysis I could find included data from studies performed through October 2014. Almost 5000 cycles in 1833 patients showed about a 38% increase in pregnancy rate with letrozole (with the 95% confidence interval between a 5% increase and 83%). That’s a pretty big difference. An older meta-analysis (referenced in Klement and Casper) found a similar increase in pregnancy rate of around 32% (range of 9% to 60%) in 2066 women. In both cases a smaller number of studies reported live births, with even larger differences between letrozole and clomiphene (e.g. 55% increase in live birth in the first study, range of 26% to 90%, and a 63% increase in live birth in the second study, range of 31% to 103%. The miscarriage rate, when reported, jibes well with what we described in NPNW at around 30%). (Note that many of the studies included overlap so one cannot treat these as completely independent, however, the fact that both analyses came to similar conclusions is supportive of a real difference).
This article is referenced in the Klement and Casper review; it shows a 27.5% pregnancy rate with letrozole (103/374 women pregnant) versus 19.1% with clomiphene (72/376), p=0.007. There was also a higher rate of ovulation with letrozole (834 of 1352 treatment cycles [61.7%] vs. 688 of 1425 treatment cycles [48.3%], P<0.001, which especially in women with HA will help the pregnancy rate.
This smaller study (only 100 women) found a 32% pregnancy rate with letrozole vs. 16% with clomiphene, p=0.05 (note that the abstract reverses those numbers but if you read the results and discussion section you find that there is in fact a higher pregnancy rate in the letrozole group.)
Other studies found statistically comparable rates of pregnancy between letrozole and clomiphene, e.g., 13.0% letrozole pregnancy rate vs. 18.3% clomiphene (158 women), p=0.243. letrozole group (58% pregnancy rate) in comparison to CC (53.6%) and control groups (46%), 90 women, p > 0.05.
One large study did find a lower pregnancy rate with letrozole than clomiphene, with a 28.3% pregnancy rate in 300 women using clomiphene versus a 22.3% pregnancy rate in 299 women using letrozole, p=0.10 so borderline significance.
We continue to encourage a combination cycle with letrozole plus injectable gonadotropins if one does not respond to the oral medications alone as there is significant evidence for lower chance of multiple gestation as well as less medication needed. A study that I found while looking into the clomiphene/letrozole question highlighted these differences again with a much higher chance of only growing a single follicle with letrozole + injects (80.2%) versus clomiphene + injects (65.3%) versus inject alone (54.7%), p<0.05.
* If you are not trying to get pregnant but rather to use the oral meds to kick-start normal cycles, having worked on recovery, it makes no difference which medication you choose because lining thickness is immaterial in this case. And if one doesn’t work, it is worth trying the other as they use different mechanisms to encourage follicular growth to start.