HMG, human menopausal gonadotropin (menotropin), is a medicine made from the urine of postmenopausal women. It contains the two pituitary hormones that steer the gonads: FSH and LH activity. For more than half a century (US approval dates from 1975) it has been used to stimulate the ovaries in fertility treatment, and in men whose pituitary fails to produce its hormones it supplies FSH, without which HCG alone does not fully rebuild sperm production. The evidence is strong — this is an approved medicine backed by dozens of randomised trials — but not unanimous: two successive versions of the same Cochrane review reached different conclusions about whether HMG is as good as, or better than, recombinant FSH. In this review we explain in plain terms what HMG is, what is really inside the vial, how it works in women and men, and where the evidence ends.
What HMG is
After the menopause the ovaries stop responding to signals from the pituitary, which — freed from feedback — secretes large amounts of FSH and LH that end up in the urine. The idea of recovering them from there has a long history: as Lunenfeld (2004) describes, the first gonadotropin extracts were prepared in the 1930s from animal pituitaries, pregnant mare serum, human pituitaries and placenta, and the road to today's purified preparations was, in the author's words, complex and full of perils, all the way to recombinant products. Modern “highly purified” HMG (HP-hMG) contains 75 IU of FSH activity and 75 IU of LH activity per vial (Menopur label); in the USA it is approved for inducing the growth of multiple follicles in women undergoing assisted reproduction.
How it works
FSH and LH act on two different “switches” (receptors) in the gonads. In a woman FSH drives the granulosa cells of the follicles to grow and produce estradiol, while LH activity acts on the theca cells that supply them with raw material (androgens) and finally triggers ovulation. In a man FSH acts on the Sertoli cells, the “nurses” of maturing sperm, and LH on the Leydig cells that make testosterone. HMG delivers both signals at once, which is why it is sometimes called a “complete” gonadotropin — unlike FSH alone (recombinant or purified from urine) and unlike HCG alone, which replaces only LH.
Composition and origin — what is really inside the vial
The label “FSH + LH 1:1” is a simplification. Van de Weijer's team (2003) took an HP-hMG preparation apart chemically and found three gonadotropins: FSH, LH and HCG, with three times as much HCG as LH. Because HCG circulates far longer, about 95 % of the “LH” activity delivered by this preparation actually comes from HCG — added from an external source for standardisation, according to the authors, which is an established practice in the industry. Gonadotropins made up at most 70 % of the protein; the rest were impurities from urine, among which leukocyte elastase inhibitor and protein C inhibitor were identified. It is worth adding that the analysis was performed by Organon, maker of a competing recombinant FSH — it does not by itself prove that the impurities do harm, but it explains why HMG differs from the recombinant product more than the name suggests.
Human data — ovarian stimulation in IVF
This is the best-documented use. In the MERIT trial (Andersen et al., 2006) 731 women were randomised to HP-hMG or recombinant FSH before in-vitro fertilisation. Recombinant FSH yielded more oocytes (11.8 versus 10.0), but HMG gave a higher proportion of top-quality embryos (11.3 % versus 9.0 %); an ongoing pregnancy occurred in 27 % of women after HMG and 22 % after rFSH, a difference that did not reach significance: HMG was shown to be non-inferior, not superior. Meta-analyses of this question conflict, and honesty requires citing both. The 2011 Cochrane review (van Wely et al.; 42 trials, 9,606 couples) found no difference between recombinant FSH and urinary preparations in either live births or ovarian hyperstimulation. Its 2026 update (Berkhout et al.; 59 trials, 18,119 women), which excluded trials of doubtful trustworthiness, reached a different conclusion: live births are probably fewer after recombinant FSH than after HMG (odds ratio 0.83), and ovarian hyperstimulation more frequent (1.42), with moderate certainty of evidence. The authors caution that many trials were industry-sponsored. The difference between the versions stems from new trials and a different selection method, not from the discovery of a new mechanism — and it shows that even with thousands of participants the answer depends on which studies are counted.
Human data — men with hypogonadotropic hypogonadism
In men whose pituitary does not release LH and FSH, HMG is the source of FSH added to HCG. In the classic paper by Finkel et al. (1985) HCG alone normalised semen in men whose condition had begun after puberty, but in only one of fifteen with a pre-pubertal onset; once HMG was added, the sperm count returned to normal in five of seven men without a history of undescended testes. The meta-analysis by Rastrelli et al. (2014) confirms that combining HCG with FSH gives a higher success rate than HCG alone, and that the type of FSH preparation (urinary or recombinant) makes no difference. How long does such rebuilding take? Liu et al. (2002) followed 29 men treated at one centre between 1982 and 1998: the median time to the first sperm was 5.5 months, to 5 million/ml 12.4 months, and to 20 million/ml 29 months; conception occurred in 22 of 43 treatment courses with a median of 20.5 months, often at sperm concentrations around 5 million/ml. The strongest predictors were testicular volume and whether the patient had gone through puberty; previous testosterone therapy made no difference.
How do we know that both hormones are needed? From physiological studies in healthy men. Matsumoto et al. (1983) switched off the pituitary of five volunteers with testosterone and then gave FSH alone: in the four men on the higher dose, sperm concentration rose from 0.3 to 33 million/ml against a baseline of 94, that is to about a third of normal, and no participant returned consistently to his own range; after FSH was stopped it fell again to 0.2. The mirror-image experiment with HCG alone gave about a quarter of normal (we describe it in the HCG article), and only adding FSH to HCG restored near-control concentrations. The popular “50 % of normal” appears in none of these papers; the direction is right, the number is not.
Safety and the limits of the evidence
HMG is a medicine given under supervision, and its risks are well described: in women, ovarian hyperstimulation syndrome (OHSS) with possible pulmonary and thromboembolic complications, ovarian torsion, and multiple pregnancies — in the Menopur registration trial in IVF a multiple pregnancy was found in 35.3 % of 85 pregnancies. The label requires monitoring of the ovarian response by ultrasound and/or estradiol measurements. In men the treatment is long and requires regular semen and hormone testing; an additional limitation of urine-derived products is batch variability and the presence of proteins other than gonadotropins, described above. The limits of the evidence are twofold: the data in women are plentiful but largely manufacturer-sponsored, and the data in men are strong on direction yet based on small, mostly uncontrolled series (five, eight, twenty-odd patients). We deliberately give no methods of use and no doses.
The wider context — three ways onto the same axis
HMG, HCG and kisspeptin all act on the hypothalamus–pituitary–gonad axis, but at different levels: kisspeptin prompts the pituitary to release its own hormones, HCG replaces LH, and HMG delivers FSH together with LH activity straight to the gonads. In reproductive medicine HMG competes today with recombinant FSH; in andrology it complements HCG — and in both roles it is a tool under specialist control, not a substance for “self-experimentation”.
Summary
HMG is a preparation of FSH and LH activity purified from the urine of postmenopausal women — the latter supplied, in practice, by added HCG. In women undergoing IVF, large randomised trials show it is no worse than recombinant FSH, and the latest Cochrane review even suggests slightly more live births and less hyperstimulation, although the earlier version of that review saw no difference. In men with hypogonadotropic hypogonadism HMG added to HCG supplies the missing FSH: FSH alone or HCG alone restore semen only partially (about a third and a quarter of normal in physiological studies), and only their combination gives near-full recovery — over a process measured in months and not rarely in years. The state of the evidence: strong as a medicine in specialist hands, with discrepancies between meta-analyses that an honest review has to acknowledge.
Sources
- Berkhout RP, Kostova EB, van Wely M. Recombinant follicle-stimulating hormone (rFSH) versus other recombinant or urinary gonadotropins for ovarian stimulation in assisted reproductive technology cycles. Cochrane Database of Systematic Reviews. 2026;7(7):CD005354. PMID: 42445958. DOI: 10.1002/14651858.CD005354.pub3. pubmed.ncbi.nlm.nih.gov/42445958
- van Wely M, Kwan I, Burt AL, et al. Recombinant versus urinary gonadotrophin for ovarian stimulation in assisted reproductive technology cycles. Cochrane Database of Systematic Reviews. 2011;2011(2):CD005354. PMID: 21328276. DOI: 10.1002/14651858.CD005354.pub2. pubmed.ncbi.nlm.nih.gov/21328276
- Rastrelli G, Corona G, Mannucci E, Maggi M. Factors affecting spermatogenesis upon gonadotropin-replacement therapy: a meta-analytic study. Andrology. 2014;2(6):794–808. PMID: 25271205. DOI: 10.1111/andr.262. pubmed.ncbi.nlm.nih.gov/25271205
- Andersen AN, Devroey P, Arce JC. Clinical outcome following stimulation with highly purified hMG or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Human Reproduction. 2006;21(12):3217–3227. PMID: 16873892. DOI: 10.1093/humrep/del284. pubmed.ncbi.nlm.nih.gov/16873892
- Finkel DM, Phillips JL, Snyder PJ. Stimulation of spermatogenesis by gonadotropins in men with hypogonadotropic hypogonadism. The New England Journal of Medicine. 1985;313(11):651–655. PMID: 3927163. DOI: 10.1056/NEJM198509123131102. pubmed.ncbi.nlm.nih.gov/3927163
- Liu PY, Gebski VJ, Turner L, et al. Predicting pregnancy and spermatogenesis by survival analysis during gonadotrophin treatment of gonadotrophin-deficient infertile men. Human Reproduction. 2002;17(3):625–633. PMID: 11870114. DOI: 10.1093/humrep/17.3.625. pubmed.ncbi.nlm.nih.gov/11870114
- Matsumoto AM, Karpas AE, Paulsen CA, Bremner WJ. Reinitiation of sperm production in gonadotropin-suppressed normal men by administration of follicle-stimulating hormone. The Journal of Clinical Investigation. 1983;72(3):1005–1015. PMID: 6411766. DOI: 10.1172/jci111024. pubmed.ncbi.nlm.nih.gov/6411766
- van de Weijer BH, Mulders JW, Bos ES, et al. Compositional analyses of a human menopausal gonadotrophin preparation extracted from urine (menotropin). Identification of some of its major impurities. Reproductive BioMedicine Online. 2003;7(5):547–557. PMID: 14680547. DOI: 10.1016/s1472-6483(10)62071-8. pubmed.ncbi.nlm.nih.gov/14680547
- Lunenfeld B. Historical perspectives in gonadotrophin therapy. Human Reproduction Update. 2004;10(6):453–467. PMID: 15388674. DOI: 10.1093/humupd/dmh044. pubmed.ncbi.nlm.nih.gov/15388674
For in-vitro laboratory research only. It is not a human medicine and is not for treatment.