Hormonal IUD (Mirena/Kyleena): History, Development, and How It Works

Where It Started: A Finnish Idea in the 1970s

The levonorgestrel IUD did not emerge from a large pharmaceutical program. It came from a single researcher's question about whether a copper IUD could be made to release a hormone locally and eliminate systemic side effects.

Jaakko Luukkainen at the University of Helsinki began experimenting with progestin-releasing intrauterine devices in the early 1970s, building on earlier work by Antonio Scommegna, who had tested progesterone-releasing IUDs in the late 1960s. Luukkainen's insight was specific: levonorgestrel, a synthetic progestin already in oral contraceptive pills, could be embedded in a polydimethylsiloxane membrane inside a T-shaped frame and made to diffuse at a controlled rate directly into the endometrium.

Why Levonorgestrel and Not Progesterone

Progesterone IUDs, marketed briefly as Progestasert in the US from 1976, required annual replacement because natural progesterone degrades quickly. Levonorgestrel is more potent and more stable. Luukkainen's early devices showed that 20 micrograms per day could suppress endometrial proliferation, thicken cervical mucus, and dramatically reduce menstrual bleeding, all with serum levonorgestrel levels roughly ten times lower than those produced by oral levonorgestrel pills.

The T-Shape and the Rate-Controlling Membrane

The physical design solved a clinical problem that plagued earlier IUDs. Earlier devices like the Dalkon Shield (withdrawn 1974) had multifilament strings that were linked to ascending pelvic infections. The Mirena frame uses a monofilament retrieval thread and a hormone reservoir around a vertical stem. The rate-controlling membrane, a mixture of polydimethylsiloxane and levonorgestrel, releases drug at a near-linear rate that falls only modestly over the approved duration of use, from roughly 20 mcg/day at insertion to about 10 mcg/day at five years.

Clinical Trials That Built the Evidence Base

From the 1980s through the late 1990s, Luukkainen and colleagues published a series of controlled studies documenting efficacy, bleeding patterns, and safety. These formed the backbone of the Bayer (then Schering) regulatory submission.

The Finnish Phase III Data

The original Finnish trials enrolled women across reproductive ages and demonstrated a Pearl Index below 0.2, meaning fewer than 2 pregnancies per 1,000 woman-years of use. The trials also documented a side effect that turned out to be a therapeutic benefit: the majority of users developed amenorrhea or very light bleeding within 12 months. That finding redirected the device's development trajectory toward heavy menstrual bleeding as a second indication.

The NEJM 2013 Trial: LNG-IUS vs. Usual Care for HMB

The most rigorous evidence for the bleeding indication came from the ECLIPSE trial, published in the New England Journal of Medicine in 2013. The randomized trial enrolled 571 women with heavy menstrual bleeding and compared the LNG-IUS to usual medical care (tranexamic acid, mefenamic acid, combined oral contraceptives, or norethisterone). At two years, the LNG-IUS group had significantly greater reductions in menstrual bleeding and higher disease-specific quality-of-life scores. The trial also found that LNG-IUS users were substantially less likely to proceed to surgery. This single trial reshaped UK and US clinical practice and anchored ACOG guidance on abnormal uterine bleeding.

Smaller Devices and Lower Doses: Kyleena, Liletta, Skyla

Regulatory submissions for Skyla (2013), Liletta (2015), and Kyleena (2016) used the same core mechanism but varied the hormone reservoir size and the frame dimensions. Kyleena's frame is 28 mm wide versus Mirena's 32 mm, and it releases approximately 17.5 mcg/day initially, falling to 8 mcg/day at steady state, which means less amenorrhea but also slightly lower contraceptive efficacy at the margins (still <1% annual failure rate). These smaller devices were developed explicitly in response to clinician and patient data showing that nulliparous women and adolescents experienced more insertion pain and more expulsions with the full-sized Mirena frame.

Mechanism of Action: How the LNG-IUS Actually Works

The hormonal IUD works through at least three distinct mechanisms, none of which is primarily ovulation suppression. That distinction matters clinically and matters to women who have concerns about how the device works.

Cervical Mucus Thickening

Levonorgestrel binds progesterone receptors in cervical epithelium and causes rapid thickening of cervical mucus. The mucus becomes viscous and cellular, forming a barrier that sperm cannot penetrate effectively. This effect appears within days of insertion and is considered the primary contraceptive mechanism.

Endometrial Suppression

High local concentrations of levonorgestrel suppress endometrial glandular development and stromal proliferation. The endometrium becomes thin and atrophic, which is why bleeding decreases. This same mechanism underlies the FDA-approved indication for heavy menstrual bleeding and the off-label but guideline-supported use for endometrial protection in women taking systemic estrogen for menopause.

Ovulation: Partially Suppressed, Not Reliably So

Serum levonorgestrel from the IUD is low enough that most women continue to ovulate, particularly with the lower-dose devices. A Swedish study found that ovulation was suppressed in roughly 45% of Mirena users in year one, falling to near zero by year five. Kyleena users ovulate at even higher rates. This means the device is not equivalent to a combined hormonal contraceptive with respect to ovulation suppression. Women with PCOS who rely on anovulation should not assume the LNG-IUS will be sufficient for cycle regulation, though it remains highly effective for contraception.

Local vs. Systemic: The Pharmacokinetic Advantage

The ratio of endometrial tissue concentration to serum concentration is approximately 1,000 to 1 with the LNG-IUS. Serum levels average 150 to 200 picograms per milliliter for Mirena users, compared with roughly 2,000 pg/mL for women taking 150 mcg oral levonorgestrel daily. This explains why systemic progestogenic side effects (acne, mood changes, libido changes) are less frequent than with oral contraceptives, though they do occur in a subset of users because serum exposure is not zero.

The WomanRx Three-Level Mechanism Framework for the LNG-IUS:

| Level | Mechanism | Clinical consequence | |---|---|---| | Cervical | Mucus thickening | Primary contraceptive barrier | | Uterine | Endometrial atrophy | Bleeding reduction, endometrial protection | | Ovarian | Partial, inconsistent ovulation suppression | Not reliable for cycle regulation; reliable for contraception |

This three-level framework helps explain why the same device is used for four different indications, why systemic side effects are less common than with pills, and why women with cycle-dependent conditions like PCOS or endometriosis need additional management beyond the IUD alone.

Sex-Specific Physiology: How the Menstrual Cycle and Hormonal Status Affect the Device

The LNG-IUS interacts with your hormonal environment differently depending on your reproductive life stage. This is one of the least-covered aspects of hormonal IUD education, and it directly affects what to expect.

Reproductive Years (Menstruating, Not TTC)

In regularly cycling women, insertion during the first seven days of the cycle allows the device to provide immediate contraceptive protection, per ACOG guidance on contraceptive initiation. Irregular bleeding in the first three to six months is normal and results from the transition of an estrogen-primed, prolific endometrium to a suppressed one. About 20% of Mirena users report amenorrhea at one year, rising with duration of use.

PCOS

Women with PCOS have a thickened, estrogen-dominant endometrium at baseline because of anovulatory cycles. The LNG-IUS provides local progestin opposition that protects against endometrial hyperplasia and the increased endometrial cancer risk associated with chronic anovulation. It does not treat androgen excess, insulin resistance, or anovulation. For women with PCOS who want contraception AND metabolic management, the LNG-IUS is often combined with metformin or weight management, not used as a standalone therapy.

Perimenopause

Perimenopausal women (typically age 40 to 51) experience erratic ovarian estrogen production and often develop heavy, irregular bleeding. The LNG-IUS is one of the most effective therapies for perimenopausal HMB and simultaneously provides contraception during a life stage when pregnancy is still possible and often unplanned. The Mirena is approved for use up to age 55 or confirmed menopause.

Postmenopause / Systemic HRT

Women taking systemic estrogen for menopause symptoms need endometrial protection if they have a uterus. The LNG-IUS delivers local progestin for this purpose, avoiding the systemic progestogen that causes mood and breast symptoms in some women. The British Menopause Society and NICE have endorsed this off-label use, though regulatory approval varies by country. The Mirena used for endometrial protection in HRT is typically replaced every five years rather than at the eight-year contraceptive interval.

Pregnancy, Lactation, and Contraception Requirements

The LNG-IUS is contraindicated in confirmed or suspected pregnancy. This is a hard stop. If a woman becomes pregnant with an IUD in place, the device should be removed as early as possible because leaving it in place increases the risk of second-trimester pregnancy loss, preterm birth, and chorioamnionitis. Removal itself also carries a small risk of pregnancy loss, so the decision is made collaboratively with the patient.

Ectopic Pregnancy Risk

The absolute risk of ectopic pregnancy is lower in LNG-IUS users than in women using no contraception, because the device is so effective at preventing any pregnancy. However, if pregnancy does occur with an IUD in situ, the proportion that are ectopic is higher than in the general pregnant population. Any woman with an IUD who has a positive pregnancy test should have serum hCG levels and pelvic ultrasound promptly to rule out ectopic implantation.

Lactation

Levonorgestrel is secreted into breast milk in small amounts. A Cochrane review of progestin-only contraceptive methods in lactating women found no demonstrated adverse effects on infant growth or development. The LNG-IUS is considered compatible with breastfeeding by both the WHO Medical Eligibility Criteria (Category 2 in the first six weeks postpartum, Category 1 thereafter) and ACOG. Insertion is typically deferred to at least four weeks postpartum to reduce expulsion risk in a recently involuted uterus.

Postpartum Timing

Immediate postpartum insertion (within 10 minutes of placenta delivery) is an option endorsed by ACOG for women who choose it, though expulsion rates of 24% have been reported at six months with this timing, compared with about 5% for interval insertion. For women leaving the hospital postpartum without a contraceptive plan, the LNG-IUS at the four-to-six-week visit is a practical and evidence-supported option.

Fertility Return

Fertility returns promptly after removal. In studies of women who had Mirena removed to attempt pregnancy, more than 80% conceived within 12 months, a rate consistent with the background fertility of age-matched women who had never used the device.

Who This Is Right For (and Who Should Think Twice)

No single contraceptive suits every woman. The LNG-IUS has a clear benefit profile in some clinical pictures and genuine concerns in others.

Strong Candidates

• Women wanting long-acting reversible contraception with minimal daily demands
• Women with heavy menstrual bleeding, iron-deficiency anemia from blood loss, or dysmenorrhea
• Women with PCOS needing endometrial protection against hyperplasia
• Perimenopausal women who want contraception and period management in one device
• Women on systemic estrogen HRT who need progestin opposition without oral progestogens

Caution or Contraindication

• Women with unexplained uterine cavity abnormalities or submucosal fibroids distorting the cavity (expulsion and efficacy risk)
• Women with active pelvic inflammatory disease or recent STI in the past three months
• Women with known or suspected endometrial or cervical malignancy
• Women who are pregnant
• Women with a history of ectopic pregnancy are not automatically excluded, but the risk-benefit conversation should happen explicitly with a clinician

Nulliparous Women and Adolescents

Both ACOG and the American Academy of Pediatrics support LNG-IUS use in adolescents and nulliparous women. The smaller-frame devices (Kyleena, Skyla) were developed partly to address concerns about insertion pain in this group. Insertion discomfort is real and should not be minimized; pre-procedure NSAIDs and, in some clinic settings, paracervical block are options worth discussing.

Regulatory Milestones: A Timeline

The path from Luukkainen's lab to a woman's uterus took about 25 years and crossed three continents.

| Year | Event | |---|---| | Late 1960s | Scommegna demonstrates progesterone-releasing IUDs reduce bleeding | | 1970s | Luukkainen develops first levonorgestrel-releasing T-device at University of Helsinki | | 1990 | Mirena approved in Finland and Sweden | | 1997 | European-wide approval | | 2000 | FDA approves Mirena for contraception in the US | | 2009 | FDA approves Mirena for heavy menstrual bleeding | | 2013 | Skyla (smaller frame, 3-year duration) approved in US | | 2013 | ECLIPSE trial published in NEJM; LNG-IUS outperforms usual care for HMB | | 2015 | Liletta approved in US (52 mcg LNG system, lower cost) | | 2016 | Kyleena approved in US (19.5 mg LNG, smaller frame, 5-year duration) | | 2024 | FDA extends Mirena approval to 8 years of use |

What Competitors Miss: The Evidence Gap in Women's Subgroups

Most articles on the hormonal IUD treat "women" as a homogeneous category. The clinical trial data do not support that view.

The Finnish phase III trials enrolled predominantly parous, regularly cycling women in their 30s. Adolescents, women with PCOS, women with endometriosis, and women over 45 were either excluded or severely underrepresented. Women have been historically underrepresented in contraceptive device trials designed to earn regulatory approval, and the subgroup-specific data that do exist are largely observational.

What this means in practice:

• Amenorrhea rates in women with PCOS may differ from the 20% figure quoted from phase III data, because baseline endometrial thickness differs.
• Insertion pain data in adolescents come largely from retrospective chart reviews and small prospective studies, not powered randomized trials.
• Endometrial protection data in perimenopausal HRT users derive mostly from the Mirena as Endometrial Protection (MIRA) studies, which were adequate for regulatory purposes in Europe but did not enroll large numbers of women over 55.

Honest clinical counseling names these gaps. A woman with PCOS asking whether the LNG-IUS will "fix" her periods deserves to know that the 20% amenorrhea figure may not apply to her and that her response depends on her baseline estrogen exposure and endometrial state at insertion.