A new study has been published looking at the lowest prescription strength of vaginal estrogen (10 mcg estradiol) available in the United States to treat the vaginal discomfort associated with menopause or Genital Syndrome of Menopause (GSM).
GSM affects approximately 50% of women and causes many symptoms including (but not limited to) pain with sex, dryness, itching, irritation and discharge. Pain with sex is typically ranked as the most bothersome symptom with dryness number two, although obviously different women are bothered by different symptoms.
This study compares low dose vaginal estradiol with a vaginal moisturizer, Replens, every three days and a placebo vaginal gel every three days. It is a prospective, randomized double-blinded placebo controlled trial and was funded by the National Institutes of Health/National Institute on Aging (so not Pharma). The placebo gel had a fairly similar pH and viscosity to Replens, but did not have the mucoadherent properties (meaning it wasn’t supposed to stick around and provide lasting benefit). The main outcome was a decrease in severity on a 4 point scale (with 0 being no bother and 3 being the most bother) of the most bothersome symptom between enrollment and 12 weeks. Secondary outcome measures were changes in pH, how the vaginal cells responded to the therapy, and the Female Sexual Function Index (FSFI).
The big surprise? The three regimens were equal, even the supposedly non pharmaceutical placebo. They all reduced the symptom severity of the two most bothersome symptoms, which were pain with vaginal penetration (experienced at enrollment by 60% of the women) and vulvovaginal dryness (experienced by 21% at enrollment) by about 50% for most women. The vaginal estrogen and the Replens were both no better than the placebo lubricant. At 12 weeks, 47% of women in the estradiol group, 29% in the Replens group, and 43% in the placebo group said they were rarely or never distressed by their most bothersome symptom. The only benefit with the estrogen versus the Replens and the placebo gel was an improvement in vaginal pH; 46% of women using the estrogen had an improvement in vaginal pH versus 12% of placebo.
I wasn’t surprised, after all the total amount of estrogen vaginally with the 10 mcg tablets twice a week is very, very low. I see a lot of failures with it, women who think they have failed estrogen but in reality they are just on what amounts to a barely more than homeopathic dose. It also didn’t work for me (I suspect this is TMI, but hey, that is part of my charm). As there is a change in pH it’s doing something, but not much.
Up until now I thought perhaps the 10 mcg dose was enough for most, after all, I run a specialty clinic and so I only see people who have problems with standard therapies. Then again, some people who experience treatment failures just don’t come back. So I started to dig into the story of the 10 mcg vaginal tablets. The brand name is Vagifem (now available as the generic Yuvafem) and when they were first introduced in the 1990s they were 25 mcg. In 2008 two studies were published indicating the 10 mcg worked just as well as the 25 mcg, so we all switched. And then I got curious, were these original studies on par with this newer one that shows vaginal estradiol 10 mcg twice a week is as “effective” as a placebo gel?
The first study (Bachman et. al.) compared 10 mcg of estradiol vagina-to-vagina (it sounds better than head to head, sorry, indulge me) with 25 mcg of estradiol. It did not assess the most bothersome symptom, pain with sex. Yes, you read that right. Why? Because “dyspareunia was not indicated by every participant.” Well, of course not because only 60% of women with GSM have it. This is a huge red flag for me. In medicine when you do not want to know the answer you don’t ask the question. Did they suspect the 10 mcg estradiol dose would be too low for pain with sex? Hard not to think that. What this study used instead was a composite score of dryness, soreness and irritation. It is also a very small study, only 76 participants completed it and only 14 were placebo. The placebo was a tablet so of course it is not possible to know if using a gel of some kind would have been similar to either the 10 mcg or the 25 mcg dose. This study used the same 4 point scale of the bother factor as the Mitchell study and the participants all had a similar drop in how bothered they were by their symptoms, a little more than 50%. There was also improvement in pH and the appearance of the vaginal cells using the estrogen versus placebo. Based on this study it is not possible to say how 10 mcg estradiol works for pain with sex, however, it does seem like it is equivalent symptom wise with the new study, i.e. the same improvement as the placebo gel.
The second study that compares 10 mcg vaginal estradiol vagina to vagina with placebo is Simon et. al. I was very confused by this study. I couldn’t even find a primary outcome measure, so I called in a consult from the amazing Dr. Ken Milne M.D. (check out his fantastic podcast) and the fantastic Dr. Elizabeth Hassan, Ph.D, P.Eng, a statistics whiz. This study also looked at bother factors using the same 4 point scale. Fortunately this study included dyspareunia, unfortunately there was no formal primary or second outcome measure. This is not good. This study reads more like “let’s check a bunch of things and see what works and go backwards from there.” This is not how we do studies, at least not robust ones anyway. Dr. Milne pointed that the industry sponsor did the stats and editorial support, also not good. Both Drs. Milne and Hassan pointed out the high loss to follow-up rates, 33% for placebo and 20% for the drug. There were other issues as well, but basically it’s not a quality study and I am uncomfortable with this study being the reason we all decided 10 mcg of vaginal estradiol was a thing.
The Simon et. al. study shows the vaginal epithelial cells changed in appearance in response to the estrogen (not enough to premenopausal levels, but better than pre estrogen therapy), the pH improved although 28% of women using the estrogen still had an abnormal pH.
Here’s the big verdict on pain with sex with the Simon study – it improved 0.87 on the 4 point scale for placebo and 1.23 with the estrogen. This was statistically significant, but the difference between the two is reported as 0.33 (although 1.23-0.87 is 0.36 so I found that confusing) and in my experience an improvement of 0.33 (or 0.36 for that matter) on a 4 point scale is not clinically meaningful. Most women consider pain with sex binary. Improved is ok, but everyone wants it gone.
How do these results, a 1.23 drop in the bother factor of pain with sex with the Simon et. al. study compare with the newer Mitchell study? In the new Mitchell study the drop is 1.2-1.4, so the same. This means 10 mcg of vaginal estradiol is about as good as using a neutral vaginal gel.
Several years ago I realized in my own clinical practice that I was seeing many women on this ultra low dose estrogen therapy who were referred for pain with sex. These women were told that “estrogen just didn’t work for them,” and of course why would they believe any different because here they were using a standard regimen and it wasn’t working or working enough for them.
An ultra-low dose of vaginal estradiol, 10 mcg nightly for 2 weeks and then twice a week, will improve pH for some women but it is unlikely to resolve the most bothersome factor, pain with sex, or resolve dryness. It will improve it, but only as much as a non irritating gel or Replens. The Replens is a lot cheaper, but it has a high osmolality (higher than recommended by the WHO) and over time that could potentially lead to irritation. The current study only looked at using it for 12 weeks.
If you want to completely treat the most bothersome vaginal symptoms of menopause with estrogen you will need a larger dose, so the vaginal estradiol ring or the cream. The tablets are a nice delivery vehicle, but if they don’t work, well, there isn’t much point in wasting your money when it could be spent on a lubricant, Replens, vaginal hyaluronic acid (another moisturizer), or a dose of estrogen that will actually work.
I’m glad we have this new study which confirms what I have seen clinically, but I’m pretty angry the poor quality studies on the 10 mcg of estradiol were taken as gospel.