Progesterone for Luteal Phase Support: Why the Route Changes Everything

If your clinic started you on vaginal progesterone and then declined to check a progesterone level, that was not an oversight. Vaginal progesterone produces serum concentrations that look inadequate on paper while delivering more hormone into the uterine lining than an injection does. The blood test measures the wrong compartment. This guide covers what each route actually delivers, the doses the labels support, what the randomised evidence says about choosing between them, and where the side effects come from.

What luteal phase support is replacing

After ovulation the collapsed follicle reorganises into the corpus luteum and starts producing progesterone. That progesterone converts a proliferative endometrium into a secretory one, which is the only state an embryo can implant into, and then holds the lining in place until the placenta takes over hormone production at roughly week eight to ten of pregnancy.

In a natural cycle this happens without help. In assisted reproduction it frequently does not. Egg retrieval aspirates granulosa cells that would have formed the corpus luteum. Pituitary suppression with a GnRH agonist or antagonist blunts the LH signal the corpus luteum depends on. In a programmed frozen embryo transfer there is no corpus luteum at all, because ovulation never happened. In every one of those situations the progesterone has to come from outside the body.

That is the entire job of luteal phase support: supply a hormone the cycle cannot make, at a dose and by a route that puts it into the endometrium during the days when implantation is possible.

The measurement that misleads almost everyone

Progesterone given vaginally does not behave like progesterone given anywhere else. A portion of it moves directly from the vagina into the uterus through local vascular and lymphatic transfer, bypassing the general circulation on the way. Reproductive pharmacology calls this the first uterine pass effect, and it produces a result that looks contradictory on a lab report.

Under steady state conditions the numbers separate sharply by route:

Route	Daily dose studied	Serum progesterone	Endometrial tissue progesterone
Vaginal micronized, 4 divided doses	800 mg	11.9 ng/mL	11.5 ng/mL
Intramuscular in oil, 2 divided doses	100 mg	69.8 ng/mL	1.4 ng/mL

The injection produces roughly six times the blood level and roughly one eighth the tissue level. Studies sampling endometrium directly at hysterectomy found the ratio of endometrial to serum progesterone was around 14 times higher after vaginal gel than after intramuscular injection, and the effect held up when investigators controlled for residual drug contaminating the vaginal sample.

There is a second consequence that surprises people who try to fix a low reading by taking more. The vaginal route has a limited transfer capacity, so serum levels stop responding to dose increases. Across daily doses from 180 mg up to 800 mg, serum progesterone stays capped at roughly 10 to 15 ng/mL. Doubling a vaginal dose because the blood test looked low mostly produces more discharge, not more hormone in the circulation.

None of this means serum levels are useless. In programmed frozen transfer cycles using injection alone, low serum progesterone on transfer day does predict worse outcomes, and clinics increasingly check it. It means the reference range you would apply to a natural cycle does not transfer to a woman using a vaginal preparation.

Vaginal progesterone and what the label supports

The FDA approved vaginal insert is dosed at 100 mg placed vaginally two or three times daily, started the day after oocyte retrieval and continued for up to 10 weeks in total. The approved indication is support of embryo implantation and early pregnancy as part of an assisted reproductive technology programme, and the label is explicit that efficacy in women aged 35 and over was not clearly established in the registration study.

Two practical rules come straight off that label and get ignored constantly. The first is that other vaginal products, antifungal creams in particular, should not be used alongside it, because they change how the progesterone releases and absorbs. The second is that strong CYP3A4 inducers such as rifampin, carbamazepine, phenytoin and St John's wort speed up progesterone elimination and can reduce the effect.

Absorption kinetics are quick. A 50 mg vaginal preparation reaches a peak serum level of about 8.8 ng/mL within two to three hours, which is why vaginal dosing is split across the day rather than given once.

Gel, pessary, capsule and insert are not interchangeable milligram for milligram. A 90 mg gel and a 200 mg capsule are not the same delivered dose, and switching product without switching the prescribed dose is a common self inflicted error.

Intramuscular progesterone in oil

Injection produces the highest and most reliable blood levels of any route. A 100 mg intramuscular dose typically peaks at 40 to 80 ng/mL and keeps serum progesterone elevated for 48 hours or more, which is why once daily dosing is adequate and why this route is preferred whenever a clinic wants a measurable, defensible serum number.

The cost is local. Progesterone is suspended in sesame or olive oil, injected deep into gluteal muscle with a long needle, most often daily for weeks. Redness, itching, hard lumps at the site, panniculitis and sterile abscess formation are all documented consequences, and they get worse the longer the course runs. Rotating sites, warming the oil, and applying heat afterwards reduce but do not eliminate the problem.

A growing number of clinics now combine routes rather than choosing one, giving injection two or three times a week for circulating levels plus a daily vaginal preparation for tissue levels. That approach reflects the physiology better than either route alone, at the price of a more complicated schedule.

Oral micronized progesterone and why it sits last

Swallowed progesterone reaches peak serum concentration within about three hours, but most of the dose is destroyed by first pass metabolism in the gut wall and liver before it reaches the circulation. The FDA label for oral micronized progesterone states plainly that absolute bioavailability is not known.

What survives that metabolism includes neuroactive metabolites, principally allopregnanolone, which act on GABA receptors. That is why the oral label instructs bedtime dosing: drowsiness and dizziness are expected, and in a small number of women the effects extend to blurred vision, difficulty speaking and unsteady walking.

The approved oral doses are 200 mg at bedtime for 12 sequential days of a 28 day cycle when protecting the endometrium during estrogen therapy, and 400 mg at bedtime for 10 days in secondary amenorrhea. Oral progesterone is a legitimate treatment for both of those. It is the weakest of the three options for getting progesterone into an endometrium preparing for implantation.

Dydrogesterone is not the same molecule

Dydrogesterone is a retroprogesterone, a structural variant of progesterone with the hydrogen at carbon 9 flipped. That change makes it orally active with far better bioavailability than micronized progesterone, and it does not produce the sedating metabolites, so daytime dosing is workable.

It is also selective for the progesterone receptor in a way natural progesterone is not, and it does not suppress the body's own progesterone, so a serum progesterone assay will not detect it. A woman taking dydrogesterone who is told her progesterone is low is being misinformed by the test, not by her body.

Honest caveat on availability: dydrogesterone has never been approved in the United States. It is a mainstream luteal support option across Europe, Asia, the Middle East and Australia, with large randomised trials behind its use in fresh IVF cycles, and it is unavailable through the US prescribing system. That is a regulatory fact, not a verdict on the drug.

What the randomised evidence says about choosing a route

The largest synthesis is the 2015 Cochrane review of luteal phase support in assisted reproduction, which pooled 94 randomised trials covering 26,198 women. On the outcome that matters, live birth or ongoing pregnancy, the comparison of intramuscular against vaginal or rectal progesterone produced an odds ratio of 1.37 with a 95% confidence interval of 0.94 to 1.99 across seven trials. The interval crosses 1, so no superiority was demonstrated.

A 2020 systematic review restricted to that same comparison pooled 15 randomised trials and 5,656 patients. Clinical pregnancy came out at a risk ratio of 0.90 for vaginal against intramuscular, 95% confidence interval 0.80 to 1.00, p equal to 0.06. Ongoing pregnancy was 0.90, 95% confidence interval 0.76 to 1.06.

Read those together and the conclusion is uncomfortable but useful: after thousands of randomised patients, neither route has proven itself better, and the intervals are narrow enough that any real difference is small. Route selection is therefore a decision about tolerability, cost, adherence and the individual cycle, not about a settled efficacy ranking. Anyone telling you injections definitively work better is extending past the data.

Oral progesterone has far less evidence behind it in this setting. Cochrane found a single 40 woman trial comparing intramuscular with oral, rated very low quality, with a confidence interval running from 0.14 to 3.66. That is not a finding, it is an absence of one.

hCG as luteal support, and why it lost ground

Before progesterone preparations were widely available, clinics supported the luteal phase with repeat doses of human chorionic gonadotropin, which keeps the corpus luteum producing its own progesterone. Cochrane found hCG against placebo or no treatment gave an odds ratio of 1.76 for live birth or ongoing pregnancy, 95% confidence interval 1.08 to 2.86, from three trials and 527 women, rated very low quality.

The reason it faded is safety rather than efficacy. hCG, with or without progesterone alongside it, carries a higher rate of ovarian hyperstimulation syndrome than progesterone alone. Progesterone achieves the same endometrial goal without re-stimulating ovaries that are already enlarged. If you are trying to understand why your protocol uses hCG once to trigger and then progesterone afterwards rather than more hCG, that is the reason. Our guide to the hCG trigger shot and its timing covers the trigger itself.

One finding from the same review is worth knowing because it is still under-used: adding a GnRH agonist to progesterone support appeared to improve outcomes across the pooled trials.

When to start and when to stop

In a fresh IVF cycle the standard is to begin the day after retrieval, which is the point at which the luteal phase would normally begin and the corpus luteum has just been compromised. Starting earlier does not help and may advance the endometrium out of sync with the embryo.

Stopping is the question that generates more anxiety than it deserves. The approved vaginal insert is labelled for up to 10 weeks of use in total, which corresponds to the luteoplacental shift, the point at which the placenta produces enough progesterone that supplementation adds nothing. Many clinics stop earlier, some at a positive heartbeat scan, and trial evidence has not shown that continuing past that point improves outcomes. Stopping on a clinic instruction is not a risk you are taking; stopping on your own, without telling anyone, is.

Side effects, sorted by route

Vaginal preparations produce discharge, local irritation, and the gritty residue that gel and capsule users describe. Systemic effects are comparatively mild because circulating levels stay low. Most women who dislike vaginal progesterone dislike the mess rather than the symptoms.

Intramuscular progesterone causes site reactions in a large share of users, ranging from tenderness and bruising to hard nodules that persist for weeks. Sterile abscess is uncommon but real. Systemic effects are more pronounced than with vaginal dosing because blood levels are far higher.

Oral progesterone produces the most central nervous system effects of the three: drowsiness, dizziness, and occasionally more. Breast tenderness, bloating, mood change, headache and fatigue occur with any route and overlap almost completely with early pregnancy symptoms, which is why progesterone supplementation makes the two week wait harder to read, not easier.

Who should not use progesterone support

The absolute contraindications on the approved labels are hypersensitivity to progesterone or any excipient, and vaginal bleeding of undiagnosed cause. Labels for both the vaginal insert and oral capsules warn about arterial and venous thromboembolic events during hormone treatment, and instruct discontinuation if myocardial infarction, cerebrovascular events, thromboembolism, thrombophlebitis or retinal thrombosis is suspected.

One specific trap: some oral micronized progesterone capsules are formulated in peanut oil. Anyone with a peanut allergy needs that checked before the first dose, and it is not something a prescriber always thinks to ask.

Progesterone options stocked at SafeRxPills

Product	Molecule and route	Strength	Price
Susten 100mg Softgel	Micronized progesterone, oral or vaginal	100 mg	$24.00
Susten 200mg Softgel	Micronized progesterone, oral or vaginal	200 mg	$45.00
Susten 400 Softgel	Micronized progesterone, oral or vaginal	400 mg	$82.50
Susten VT 200	Progesterone vaginal tablet	200 mg	$52.00
Susten Gel	Progesterone vaginal gel	8%	$18.75
Susten SR 300mg	Sustained release progesterone	300 mg	$73.00
Susten 100mg Injection	Progesterone in oil, intramuscular	100 mg	$48.00
Gestone 50mg	Progesterone in oil, intramuscular	50 mg	$40.00
Gestone 100mg	Progesterone in oil, intramuscular	100 mg	$48.00
Aqsusten 25mg	Aqueous progesterone, subcutaneous	25 mg	$32.00
Gestoford 200mg	Micronized progesterone	200 mg	$80.00
Duphaston 10mg	Dydrogesterone, oral	10 mg	$65.00
Proluton Depot 250	Hydroxyprogesterone caproate, depot	250 mg	$20.00

The aqueous subcutaneous preparation deserves a note. It delivers progesterone by injection without the oil vehicle, which removes most of the injection site problems that make intramuscular courses hard to finish, using a short needle instead of a long one.

What to hold onto

Three things are worth carrying out of this. A vaginal progesterone level in the blood tells you very little about what is happening in the endometrium, and chasing that number with a higher dose does not work because the route saturates. No route has beaten another on live birth across 26,198 randomised women, so tolerability and adherence are legitimate grounds for choosing, and finishing the course you were given matters more than which one it was. And the symptoms of progesterone supplementation and the symptoms of early pregnancy are effectively the same list, so reading anything into them during the wait is a trap.

If ovulation itself is the problem rather than the luteal phase, the starting point is different. See our guides to clomiphene for ovulation induction and metformin in PCOS, or the overview on female infertility and ovulation induction.

This article is for information only and is not a substitute for advice from a qualified prescriber. Progesterone is a prescription medicine. Route, dose, start date and stop date should be set by the clinician managing your cycle, and changing any of them without telling them can cost you the cycle.

References

• DailyMed and FDA label, progesterone vaginal insert 100 mg: dosage and administration, indications, contraindications, warnings on thromboembolic events, drug interactions with CYP3A4 inducers and other vaginal products
• DailyMed and FDA label, micronized progesterone capsules 100 mg and 200 mg: dosage for endometrial protection and secondary amenorrhea, absorption and bioavailability, central nervous system adverse effects, peanut oil excipient
• Cochrane Database of Systematic Reviews 2015, CD009154.pub3, PubMed PMID 26148507: luteal phase support for assisted reproduction cycles, 94 randomised trials and 26,198 women, route comparisons and hCG versus placebo data
• PubMed PMID 32054365: systematic review and meta-analysis of vaginal versus intramuscular progesterone for luteal phase support, 15 randomised trials and 5,656 patients
• PubMed PMID 10711552: direct transport of progesterone from vagina to uterus, endometrial to serum concentration ratios measured in hysterectomy specimens
• Fertility and Sterility Reports 2021 mini-review on the uterine first pass in programmed frozen embryo transfer: steady state endometrial and serum progesterone by route, and the vaginal dose response ceiling