Oral, Vaginal, Cream, Troche, or Suppository: Which Progesterone Actually Protects Your Uterus?

A woman I started working with has been on an estradiol patch and compounded progesterone cream for almost four years. She feels good. She’s had light spotting for most of a year, and she’s been told not to worry about it.

I worried about it. Her estrogen had been telling her uterine lining to grow every day for four years, and the cream she was using to balance that has never been shown to do the balancing.

If you have a uterus and you’re taking systemic estrogen, you need progesterone. Everyone agrees on that. What almost nobody explains is that the route of administration changes whether it works for uterine protection, and the five routes on the market have wildly different amounts of evidence behind them.

Why does the route change anything?

Estrogen is a growth signal to the endometrium, the lining of your uterus. Left unopposed, that lining keeps building. Over time it can go from normal, to overgrown (hyperplasia), to overgrown with abnormal cells, to cancer.

The scale of that risk is not theoretical. In the PEPI trial, a three year randomized study of 596 postmenopausal women, the group on estrogen alone developed simple hyperplasia at 27.7 percent, complex hyperplasia at 22.7 percent, and atypical hyperplasia at 11.8 percent. In the placebo group, every one of those rates was under 1 percent. Adding a progestogen brought them back to placebo levels (Writing Group for the PEPI Trial, 1996).

Progesterone is what stops the growth signal, but only if enough of it reaches endometrial tissue. Not your bloodstream. Not your saliva. The actual tissue. Does oral progesterone protect the uterus?

Yes, and it has the strongest case of any route.

Oral micronized progesterone (Prometrium in the US, Utrogestan elsewhere) is the only route with long-term randomized data. An international panel that reviewed every micronized progesterone study concluded it protects the endometrium at 200 mg a day for 12 to 14 days a month, for up to five years (Stute et al., 2016). The Menopause Society names the same regimen: “Oral MP should be adequately dosed for prevention of endometrial hyperplasia (eg, 200 mg/d for 12-14 d/mo)” (The North American Menopause Society, 2022).

Continuous daily dosing, the 100 mg nightly approach many women are on, also has support. The REPLENISH trial followed 1,255 women on combined estradiol and progesterone capsules for a year and found zero hyperplasia in any dose group, including 100 mg (Lobo et al., 2018). Good evidence, but one year of it, against three to five years behind the 200 mg cyclic regimen. There’s lots of debate about cycling vs. continuous, and you can read more about that here.

The benefits: FDA-approved for this exact use. Decades of data. Usually covered by insurance. And because oral progesterone gets broken down in the liver into allopregnanolone, a compound that calms the same brain receptors benzodiazepines act on, it tends to help sleep. Most women take it at bedtime for this reason.

The pitfalls: That same liver metabolism means most of what you swallow never arrives as progesterone. Some women get genuinely dizzy or foggy, especially at first. Prometrium capsules are made with peanut oil, so they’re off the table with a peanut allergy.

Does vaginal progesterone protect the uterus?

Probably, and the reason is kind of elegant.

Progesterone placed in the vagina reaches the uterus more or less directly, through local tissue and vessels, instead of traveling the whole bloodstream to get there. It’s called the uterine first-pass effect, and it’s been measured. In women undergoing hysterectomy, vaginal progesterone produced an endometrial tissue-to-serum ratio with a median of 14.1, compared with 1.2 after an intramuscular injection (Cicinelli et al., 2000). The uterus got a concentrated dose while the rest of the body barely registered it.

That’s why blood levels mislead here. In one study, women on vaginal progesterone had plasma levels of only 2.4 to 3.6 ng/mL, well below a normal luteal phase, and yet every dose produced normal secretory transformation of the endometrium (Fanchin et al., 1997). Low blood level, fully changed lining.

The expert panel was more cautious here than for oral: vaginal micronized progesterone “may provide” protection at 45 mg daily for at least 10 days a month, or 100 mg every other day, for three to five years, off-label (Stute et al., 2016). The studies behind that ranged from 9 to 136 women. The British Menopause Society also flags a study finding 10 days of vaginal progesterone at 45 mg a day was not enough to fully oppose 1 mg of oral estradiol (British Menopause Society, 2026).

The benefits: Goes where it’s needed. Skips the liver, so far less grogginess. Reasonable option if oral progesterone knocks you sideways.

The pitfalls: Not FDA-approved for menopausal hormone therapy in the US. Crinone, the gel most often used, is approved for fertility treatment and amenorrhea, not for this. Smaller trials, shorter follow-up. Discharge and leakage are real for some women. And you lose the sleep benefit, which for many women is the best part of oral.

Does progesterone cream protect the uterus?

No. And I want to be direct about this because a lot of women have been told otherwise.

When researchers gave postmenopausal women on continuous estrogen a sequential transdermal progesterone cream at 16, 32, or 64 mg a day and then biopsied their endometrium, the progesterone levels achieved were “insufficient to induce any detectable change in the endometrium.” The cream did not produce secretory change in a proliferative lining (Wren et al., 2000). The systematic review that looked at all five cream studies reached the same conclusion, and documented two cases of complex hyperplasia among women using it (Stute et al., 2016). The British Menopause Society states it plainly: “Transdermal micronised progesterone does not provide sufficient endometrial protection” (British Menopause Society, 2026).

Here’s where the confusion comes from. Cream users often have very high salivary progesterone, sometimes hundreds of times their blood level, which looks like proof of absorption. It isn’t. Salivary glands concentrate progesterone, so a saliva test tells you about the salivary gland, not the uterus. When researchers measured plasma, red blood cell, and salivary progesterone in the same women, saliva was high and wildly variable while red cell levels sat below plasma, which sank the theory that cream progesterone was hiding in the blood cells (Lewis et al., 2002).

One study gets cited on the other side, and it deserves a fair hearing. In a crossover trial, 26 women took conjugated estrogen with either oral medroxyprogesterone or 40 mg a day of progesterone cream, six months each, biopsied at every switch. No hyperplasia was found in any of the 52 biopsies, and 77 percent preferred the cream (Leonetti et al., 2005).

It’s not as robust on a closer look. Every woman was on a progestogen in both arms, with no unopposed estrogen group, so the absence of hyperplasia can’t be credited to the cream. Six months and 26 women can’t detect a problem that takes years to develop. And 19 percent of the cream biopsies came back proliferative, meaning estrogen was still driving the lining. The same 40 mg dose studied for 48 weeks produced complex hyperplasia in 5 percent of women (Stute et al., 2016). Same dose, longer look, different answer.

The National Academies reviewed compounded hormone therapy in 2020 and could not identify research studies that would let them draw any conclusion about the safety of compounded progesterone regarding endometrial cancer risk (National Academies of Sciences, Engineering, and Medicine, 2020). Not reassuring data. No data.

Is there any use for cream? If you’re not taking estrogen, nothing is being stimulated that needs opposing, and using a cream because you feel better on it isn’t dangerous. Expect less than the marketing promises, though. The largest randomized trial, 223 women across four doses over six months, found no symptom benefit over placebo and more headaches in the cream groups (Benster et al., 2009), and a systematic review of the randomized trials concluded the evidence doesn’t support cream for hot flashes (Whelan et al., 2013). Some women swear by it anyway, and anecdotally in my practice, I’ve seen it help many of my clients.