Progesterone and recurrent miscarriage are linked in clinical thinking, in patient searches, and in the medical literature. But the relationship is more specific than most guides suggest.
Two large, well-designed trials tested whether giving progesterone to women with recurrent miscarriage history would improve live-birth rates. One enrolled 836 women. The other enrolled more than 4,000. The results matter: blanket progesterone for all women with prior losses did not significantly improve outcomes. A targeted subgroup, women with current bleeding and prior losses, did show benefit, and the benefit scaled with the number of previous miscarriages.
That distinction is the core of this guide. It is also where the restorative approach begins. Progesterone support is not something to prescribe or withhold based on loss history alone. It is something to consider after identifying whether a luteal phase deficiency, an early-pregnancy bleeding pattern, or another underlying factor is present. This guide covers what the evidence actually shows, where progesterone fits in a full recurrent pregnancy loss (RPL) evaluation, and what a restorative workup looks for beyond the hormone.
Cross-links to recurrent miscarriage, NaProTechnology, and NaProTechnology versus IVF are included where relevant.
Progesterone and recurrent miscarriage at a glance
What the evidence supports, by clinical situation, and how the restorative approach differs from empiric treatment. Each cited finding links to the source.
| Situation | What the evidence shows | The restorative approach |
|---|---|---|
| Recurrent miscarriage, unselected (no specific cause identified, no current bleeding) | PROMISE trial (n=836): first-trimester progesterone started after a positive pregnancy test produced no significant improvement in live-birth rate versus placebo (65.8% vs 63.3%; relative rate 1.04, 95% CI 0.94-1.15). (Coomarasamy et al., NEJM 2015) | Identify a cause before treating. Treating everyone with progesterone by history alone does not change outcomes. A cycle-charting-informed evaluation looks for luteal phase deficiency, ovulatory dysfunction, and other contributing factors before any support is started. |
| Recurrent miscarriage plus current early-pregnancy bleeding | PRISM trial (n=4,153): overall live-birth rate not significantly different. In the prespecified subgroup with prior miscarriage AND current bleeding, progesterone benefit emerged and scaled with number of prior losses. Women with 3 or more prior losses plus current bleeding had a meaningfully higher live-birth rate with progesterone than placebo. (Coomarasamy et al., NEJM 2019) | This is the subgroup where progesterone support in early pregnancy has demonstrated benefit. The clinical question is whether the couple falls into this higher-risk profile: prior losses plus active bleeding. Progesterone support here is not empiric for everyone; it is targeted to a defined risk pattern. |
| Luteal phase deficiency identified through cycle monitoring | Abnormal luteal-phase progesterone profiles are documented in a subset of women with recurrent spontaneous abortion. Identification of deficient profiles is clinically distinct from empiric treatment. (Daya et al., 1988) Major professional societies note that a standardized diagnostic test for luteal phase deficiency is not yet validated; the restorative model uses serial cycle charting plus timed hormonal measurement rather than a single day-21 level to characterize the full luteal profile. | Diagnosis precedes treatment. Serial cycle charting reveals the shape of the luteal phase across multiple cycles. A short or flat progesterone rise across the post-peak phase is a different clinical signal than a single low day-21 number. Support, where indicated, is targeted to the woman's own cycle biology. |
| Recurrent miscarriage with another identified contributing factor (thyroid, anatomy, immune, male factor) | A thorough RPL workup investigates uterine and anatomic factors, endocrine factors (including thyroid antibodies even in euthyroid women), antiphospholipid and thrombophilia factors, parental karyotype, and sperm DNA fragmentation. McQueen and colleagues found sperm DNA fragmentation associated with RPL in a systematic review and meta-analysis (n=1,013 men). (McQueen et al., 2019) Note: the TABLET RCT found levothyroxine did not increase live births in euthyroid women with thyroid antibodies; the workup identifies factors but each intervention requires its own evidence evaluation. | Progesterone is one slice of the evaluation. A restorative workup investigates both partners. Male factor is the sole cause in approximately 20% of couples with recurrent loss and a contributing factor in many more. Identifying and addressing all contributors, rather than prescribing progesterone and stopping, is the structural difference between a restorative evaluation and an empiric protocol. |
| Recurrent pregnancy loss after failed IVF | RRM cohorts document favorable live-birth outcomes in couples who had already failed IVF. Boyle and colleagues followed 403 couples using NaProTechnology after failed IVF and reported a 32.1% live-birth rate. (Boyle et al., 2018) The Sanchez-Mendez NaPro cohort (n=1,310) reported take-home baby rates of 50% at 24 months and 62.1% at 36 months or more. (Sanchez-Mendez et al., 2025) | IVF does not investigate or treat the causes of recurrent loss. Couples who have lost pregnancies through IVF cycles or failed to achieve pregnancy with IVF carry the same undiagnosed underlying factors into each cycle. A restorative evaluation starts from the beginning: identify what is causing the losses, address those causes, and support conception in the context of a worked-up biology. |
When progesterone helps, and when it does not
The clearest evidence for progesterone in recurrent pregnancy loss comes from a specific patient profile: prior miscarriage history AND active bleeding in early pregnancy.
The PROMISE trial established the baseline. Coomarasamy and colleagues randomized 836 women with unexplained recurrent miscarriage to first-trimester progesterone or placebo, started after a positive pregnancy test. Live-birth rate was 65.8% with progesterone versus 63.3% with placebo. The difference was not statistically significant. (Coomarasamy et al., NEJM 2015) Giving progesterone empirically to all women selected only by their loss history, without identifying a luteal deficiency or other specific indication, does not change outcomes.
The PRISM trial added a critical variable: current bleeding. Among more than 4,000 women with bleeding in early pregnancy, the overall live-birth rate with progesterone was not significantly higher than placebo. But in the prespecified subgroup with bleeding AND a history of prior miscarriage, a benefit emerged. And that benefit scaled with the number of previous losses: women with three or more prior miscarriages plus current bleeding had a meaningfully higher live-birth rate with progesterone than placebo. (Coomarasamy et al., NEJM 2019)
A pooled critical evaluation of the randomized evidence confirmed this reading: progesterone has no meaningful benefit for an unselected population, but benefits the targeted subgroup of women with current bleeding and prior losses, with benefit scaling with the number of losses. (Coomarasamy et al., AJOG 2020)
The conclusion is not that progesterone does not work. It is that progesterone works in the right patient, not in every patient. Identifying who that is requires more than a loss history.
Luteal phase defect and progesterone insufficiency
A subset of women with recurrent pregnancy loss have abnormal progesterone profiles across the luteal phase. This is not a theoretical construct. Daya and colleagues documented deficient luteal-phase progesterone patterns in women with recurrent spontaneous abortion and studied outcomes with progesterone treatment in that identified subset. (Daya et al., 1988)
The diagnostic question matters. Major professional societies note that a standardized, validated test for luteal phase deficiency does not yet exist, and the condition remains debated in the mainstream literature. That debate is real, and it is worth naming honestly. But it does not mean the phenomenon is fictional. It means the standard diagnostic approach, typically a single day-21 progesterone level, is an inadequate window.
The ovulatory cycle has a shape. Progesterone rises after an ovulatory event, peaks mid-luteal phase, and then falls before menstruation. A single measurement catches one point on that curve. It misses a short luteal phase, a flat or sluggish rise, or a premature fall. Cycle charting over multiple cycles, combined with timed luteal-phase hormonal measurement, maps the full picture. That is a different diagnostic question than a single blood draw, and it produces a different level of clinical information.
Progesterone is also not a hormone that acts only in the uterus. It is systemic. Its role in supporting early pregnancy extends to the broader endocrine environment of the cycle. A progesterone-centered evaluation is, at its core, an ovulation-centered evaluation: progesterone exists at meaningful levels only because an ovulatory event occurred. Inadequate progesterone production across the luteal phase often reflects a problem with the ovulatory event itself, not just with progesterone in isolation.
This is why the restorative diagnostic approach centers on characterizing the full luteal phase, not on treating a single low number.
The restorative approach: timed monitoring and targeted support
The PROMISE and PRISM trials tested a specific intervention: progesterone given after a positive pregnancy test, to women selected by their loss history, without prior identification of a luteal deficiency or other hormonal cause. The restorative approach operates on a different model, and that difference explains why the negative empiric-trial results and the favorable restorative cohort outcomes are not in contradiction.
Three structural differences separate empiric progesterone from cycle-timed targeted support.
First, diagnosis precedes treatment. Cycle charting across multiple cycles identifies whether a luteal phase deficiency, ovulatory dysfunction, or other hormonal pattern is present before conception is attempted. Treating a confirmed finding is different from treating a population defined only by their loss history.
Second, timing reflects the individual's biology. Support targeted to the woman's own luteal phase and the specific hormonal pattern identified in her charts is different from a protocol started generically after a positive test. Progesterone given early enough to support the luteal phase is a different intervention than progesterone started at week four or five of pregnancy.
Third, progesterone is one component of a full workup. It is not the whole answer. The restorative evaluation investigates the full range of contributing factors, addresses what is found, and uses hormonal support where indicated as part of a coordinated approach, not as a standalone prescription for anyone who has experienced a loss.
NaProTechnology cohort data reflect this model. The Sanchez-Mendez cohort followed 1,310 couples using NaProTechnology; take-home baby rates reached 50% at 24 months and 62.1% at 36 months or more. (Sanchez-Mendez et al., 2025) Boyle and colleagues followed 403 couples who had already failed IVF and reported a 32.1% live-birth rate using restorative methods. (Boyle et al., 2018) These are observational cohorts without randomized controls, and the populations differ from the PROMISE and PRISM trial populations in important ways. What they demonstrate is that diagnosis-first, cycle-timed care produces outcomes worth comparing, not dismissing.
Recurrent miscarriage is more than progesterone
Recurrent pregnancy loss is not one condition. It is a clinical pattern with multiple possible causes, and progesterone is one factor in a longer differential.
A thorough RPL evaluation investigates both partners and both sides of the biology. On the maternal side: uterine anatomy, endocrine factors (including thyroid function and thyroid antibodies even in euthyroid women), antiphospholipid antibodies, thrombophilia factors, and parental karyotype. On the paternal side: sperm DNA fragmentation. McQueen and colleagues conducted a systematic review and meta-analysis of 1,013 men and found that elevated sperm DNA fragmentation is associated with recurrent pregnancy loss. (McQueen et al., 2019) Male factor is not a peripheral consideration in RPL. It is a central one.
Thyroid evaluation deserves a specific note about evidence honesty. Thyroid autoimmunity, elevated TPO antibodies in women who are otherwise euthyroid, is associated with miscarriage risk. But the TABLET randomized controlled trial, which tested levothyroxine treatment in this population, did not find a significant increase in live births. The workup identifies the factor. Whether and how to treat it requires an individualized evidence evaluation, not a default prescription.
Endometriosis is also a contributing factor in some cases of RPL that is not identified unless specifically looked for. The restorative evaluation does not stop at a single hormonal number. It maps the full landscape of what might be disrupting a pregnancy from implanting and sustaining.
The framing that recurrent miscarriage is "unexplained" often means it has not yet been thoroughly worked up. That is a different problem than a problem without a cause. Most losses have a cause. Finding it requires looking at the full picture, in both partners, across the full cycle, rather than defaulting to a hormone supplement and hoping for a different outcome.
To explore the broader picture of recurrent miscarriage and how restorative care approaches it, see the recurrent miscarriage guide and NaProTechnology.
What to ask your clinician
If recurrent miscarriage has been attributed to "bad luck" or managed with progesterone alone without a prior diagnostic workup, these questions are worth raising with any clinician before the next cycle.
Has anyone mapped my luteal phase across multiple cycles, or only measured a single progesterone level? A single day-21 number does not characterize the full luteal phase. Knowing the shape of the progesterone rise and its timing relative to ovulation is a different level of information.
Has my partner been evaluated? Sperm DNA fragmentation is associated with recurrent pregnancy loss in published meta-analysis. A couple-centered evaluation looks at both partners from the beginning, not after maternal workup is exhausted.
Have we looked for endometriosis, thyroid autoimmunity, antiphospholipid antibodies, or anatomic factors? Any of these can contribute to recurrent loss and each requires specific investigation. "Unexplained" RPL is often the result of a workup that stopped too soon.
Is progesterone being prescribed because a deficiency was identified, or as a trial? The evidence supports progesterone support in specific, identifiable clinical situations. It does not support giving it to every woman with a loss history on the chance it might help.
An RRM clinician approaches recurrent pregnancy loss as a diagnostic problem: identify the cause, address what is found, and support conception in the context of a fully worked-up biology. To find an RRM clinician, visit rrmacademy.org/providers/.
Continue exploring
References
- Coomarasamy A, Williams H, Truchanowicz E, et al. A randomized trial of progesterone in women with recurrent miscarriages (PROMISE). New England Journal of Medicine. 2015;373(22):2141-2148. doi:10.1056/NEJMoa1504927. PMID 26605928. View in RRM Library
- Coomarasamy A, Devall AJ, Cheed V, et al. A randomized trial of progesterone in women with bleeding in early pregnancy (PRISM). New England Journal of Medicine. 2019;380(19):1815-1824. doi:10.1056/NEJMoa1813730. PMID 31067371. View in RRM Library
- Coomarasamy A, Devall AJ, Brosens JJ, et al. Micronized vaginal progesterone to prevent miscarriage: a critical evaluation of randomized evidence. American Journal of Obstetrics and Gynecology. 2020;223(2):167-176. doi:10.1016/j.ajog.2019.12.006. PMID 32008730. View in RRM Library
- Daya S, Ward S, Burrows E. Progesterone profiles in luteal phase defect cycles and outcome of progesterone treatment in patients with recurrent spontaneous abortion. American Journal of Obstetrics and Gynecology. 1988;158(2):225-232. PMID 3341399. View in RRM Library
- McQueen DB, Zhang J, Robins JC. Sperm DNA fragmentation and recurrent pregnancy loss: a systematic review and meta-analysis. Fertility and Sterility. 2019;112(1):54-60.e3. doi:10.1016/j.fertnstert.2019.03.003. PMID 31056315. View in RRM Library
- Sanchez-Mendez JI, Lombarte M, Abengozar-Muela R, et al. Natural procreative technology (NaProTechnology) for infertility: take-home baby rate and clinical outcomes in a 5-year single-center cohort of 1,310 couples. Frontiers in Reproductive Health. 2025;7:1696679. doi:10.3389/frph.2025.1696679. PMID 41323405. View in RRM Library
- Andralojc KM, Boyle PC, de Groot T, Parnell TA. Healthy singleton pregnancies from restorative reproductive medicine (RRM) after failed IVF. Frontiers in Medicine. 2018;5:210. doi:10.3389/fmed.2018.00210. PMID 30109231. View in RRM Library
- Practice Committee of the American Society for Reproductive Medicine. Recurrent pregnancy loss: a committee opinion. Fertility and Sterility. 2026. PMID 42062119.
- Carpentier PA, Stanford JB, Boyle PC. Progesterone in women with recurrent miscarriages (letter). New England Journal of Medicine. 2016;374(10):982-983. doi:10.1056/NEJMc1600491. PMID 26962922. View in RRM Library
Frequently Asked Questions
Does progesterone prevent recurrent miscarriage?
In many cases, giving progesterone to all women with prior miscarriage history does not improve live-birth rates. The PROMISE trial randomized 836 women to first-trimester progesterone or placebo and found no statistically significant difference in live births (65.8% vs 63.3%). (Coomarasamy et al., NEJM 2015)
A specific subgroup does benefit: women with current early-pregnancy bleeding and a history of prior losses. The PRISM trial found that in this subgroup, progesterone improved live-birth rates, and the benefit increased with the number of prior miscarriages. (Coomarasamy et al., NEJM 2019)
Progesterone is not a blanket treatment for recurrent miscarriage. It is a targeted support in the right clinical situation, and identifying that situation requires a thorough evaluation.
Consult an RRM clinician for guidance specific to your situation.
What is luteal phase deficiency and how does it relate to miscarriage?
Luteal phase deficiency refers to inadequate progesterone production in the second half of the ovulatory cycle, after ovulation and before menstruation. In a subset of women with recurrent pregnancy loss, abnormal luteal-phase progesterone profiles have been documented. (Daya et al., 1988)
Diagnosing it accurately matters. A single day-21 progesterone measurement captures one point on the luteal curve. Serial cycle charting combined with timed hormonal measurement maps the full luteal phase, including the timing of the rise, its peak, and how long adequate levels are sustained. That full picture is what allows a clinician to identify whether a deficiency is present and whether targeted support is warranted.
Major professional societies note that a standardized, validated diagnostic test for luteal phase deficiency is not yet established. The restorative approach uses cycle-charting-based assessment to characterize the full luteal phase rather than relying on a single measurement.
Consult an RRM clinician for guidance specific to your situation.
How is the RRM approach to recurrent miscarriage different from standard care?
The structural difference is diagnosis before treatment. Standard care often manages recurrent pregnancy loss with empiric progesterone, aspirin, or other interventions based on loss history alone, without first characterizing the underlying cause. The restorative approach identifies what is disrupting the pregnancy before initiating support.
Cycle charting across multiple cycles reveals the shape of the luteal phase, timing of the ovulatory event, and hormonal patterns that a single clinic visit cannot. The evaluation covers both partners from the outset, because male factor, including sperm DNA fragmentation, is associated with recurrent loss. (McQueen et al., 2019) Progesterone support, where indicated, is one part of a coordinated response to a worked-up problem, not a standalone prescription for everyone with losses.
"Unexplained" recurrent miscarriage is often a workup that stopped too soon. A thorough evaluation of both partners across multiple cycles finds causes that a standard approach misses.
Consult an RRM clinician for guidance specific to your situation.
Is my partner's health relevant to recurrent miscarriage?
In many cases, yes. Sperm DNA fragmentation is associated with recurrent pregnancy loss. A systematic review and meta-analysis of 1,013 men found elevated sperm DNA fragmentation in men whose partners experienced recurrent loss. (McQueen et al., 2019)
Recurrent pregnancy loss is a couple's diagnosis. Male factor is the sole cause of infertility in approximately 20% of couples and a contributing factor in many more. Waiting until all maternal factors have been exhausted before evaluating the male partner delays the answer. A restorative evaluation includes both partners from the first appointment.
Consult an RRM clinician for guidance specific to your situation.
Should thyroid levels be checked after recurrent miscarriage?
Thyroid evaluation is part of a thorough RPL workup. This includes thyroid antibodies (TPO antibodies) even in women whose thyroid function is otherwise normal, because thyroid autoimmunity is associated with miscarriage risk.
An important evidence note: the TABLET randomized controlled trial tested levothyroxine treatment in euthyroid women with thyroid antibodies and did not find a significant increase in live births. The workup identifies the factor. Whether and how to respond to thyroid autoimmunity requires individual clinical evaluation, not a default prescription. Identifying the factor is the first step; the treatment decision follows from a full clinical picture.
Consult an RRM clinician for guidance specific to your situation.
Can recurrent miscarriage be treated without IVF?
In many cases, yes. IVF does not treat the causes of recurrent pregnancy loss. It retrieves eggs, fertilizes them outside the body, and transfers embryos, but leaves the underlying biology, including luteal phase deficiency, uterine factors, immune factors, and male factor, unaddressed. Couples who have experienced losses through IVF cycles carry the same undiagnosed factors into each transfer.
Restorative cohorts document favorable outcomes in populations who had already failed IVF. Boyle and colleagues followed 403 couples using NaProTechnology after failed IVF and reported a 32.1% live-birth rate. (Boyle et al., 2018) The Sanchez-Mendez NaPro cohort (n=1,310) reached 50% take-home baby rate at 24 months and 62.1% at 36 months or more. (Sanchez-Mendez et al., 2025)
A diagnosis-first evaluation that addresses the actual cause of losses is a different path than repeating cycles without investigating why losses are occurring.
Consult an RRM clinician for guidance specific to your situation.
What does a restorative workup for recurrent miscarriage include?
A thorough RPL workup covers multiple potential causes across both partners. On the maternal side, this typically includes uterine and anatomic evaluation, thyroid function and thyroid antibodies, antiphospholipid antibodies and thrombophilia screening, parental karyotype, and characterization of the full luteal phase through cycle charting and timed hormonal measurement. On the paternal side, sperm DNA fragmentation is a specific factor associated with recurrent loss and should be evaluated. (McQueen et al., 2019)
The restorative model uses cycle charting not just as a conception-timing tool but as a diagnostic window. The pattern of mucus, basal temperature, and the post-peak luteal phase across multiple cycles reveals hormonal information that a single clinic visit cannot capture. Progesterone support, where indicated after this evaluation, is one component of the response, targeted to the specific finding rather than prescribed empirically to everyone.
Consult an RRM clinician for guidance specific to your situation.
This content is for educational and reference purposes only and does not constitute medical advice, diagnosis, or treatment. Consult a qualified clinician about your specific situation.