In Vitro Maturation (IVM) for PCOS: Live Birth vs OHSS Data

What Is In Vitro Maturation, and Why Does PCOS Make It Relevant?

IVM collects eggs while they are still immature, from small antral follicles that have had little or no hormonal stimulation, then finishes maturing them in a lab dish instead of inside the ovary over two weeks of injections. A 2026 review of published IVM protocols describes the technique as built around exactly the ovarian pattern that defines polycystic ovarian morphology: a large resting pool of small follicles, which is also the same pattern that makes ovarian hyperstimulation syndrome (OHSS) the leading procedural risk when those follicles are pushed to grow with gonadotropins. Women with PCOS are described in the medical literature as the prototypical candidates for IVM for that reason — not because their eggs need different handling, but because their ovaries respond to stimulation more dramatically than most, and IVM is a way of not stimulating them in the first place.

That distinction matters because PCOS does not appear to lower egg quality itself — the excess-follicle pattern is a quantity and stimulation-response story, not a quality one. IVM does not exist to fix an egg-quality problem. It exists to remove the stimulation step that turns a quantity advantage into a safety risk, which is a different goal from conventional IVF and worth holding separately from it when reading the outcome numbers below.

You may see PCOS written as polyendocrine metabolic ovarian syndrome (PMOS), after a 2026 global consensus of more than 50 organisations renamed it. Every figure on this page applies under either name — only the label changed.

How Much Does IVM Actually Reduce OHSS Risk?

IVM cut the odds of moderate-to-severe OHSS roughly 8-fold against conventional IVF in the best-controlled evidence available: a 2025 Cochrane review pooling two low-risk-of-bias trials and 739 women with PCOS found an odds ratio of 0.08 (95% CI 0.01–0.67, high-certainty evidence) — if IVF carries a 3.5% OHSS rate, the same population on IVM would be expected to see somewhere between 0% and 2.4%. The randomized trial behind much of that pooled evidence recorded it directly: zero OHSS cases in 273 women on IVM, against two cases in 273 women on conventional IVF, in a 546-woman trial conducted in Vietnam. This is the single most consistent finding in the IVM literature, and it is the entire reason the technique exists for this population.

What Are the Actual Live Birth Rates: IVM vs Conventional IVF?

Live birth after a single embryo transfer occurred in 35.2% of women on IVM against 43.2% on conventional IVF, an absolute gap of 8.1 percentage points, in the largest randomized comparison run to date — the same 546-woman Vietnamese trial cited above, which enrolled women with a high antral follicle count (24 or more follicles across both ovaries), the ovarian pattern central to a PCOS diagnosis, though the trial screened on follicle count rather than requiring a formal PCOS diagnosis. The gap widened over a full year of follow-up: cumulative ongoing pregnancy at 12 months reached 62.6% on conventional IVF against 44.0% on IVM, a statistically significant 18.7-point difference. The trial’s own conclusion was blunt — it could not demonstrate that one cycle of IVM was non-inferior to one cycle of IVF, and the cumulative-outcome gap over a year ran in IVF’s favor more clearly than the single-cycle number alone suggested.

The Cochrane review reached a more cautious, differently worded conclusion on the same underlying question: pooling two low-risk trials in 739 women with PCOS, it found live birth on IVM at an odds ratio of 0.47 against conventional IVF (95% CI 0.17–1.32) — very-low-certainty evidence, wide enough that the true effect could range from meaningfully worse to roughly equivalent. That review also found a real cost on the other side of the ledger: miscarriage per clinical pregnancy was higher with IVM, at an odds ratio of 1.66 (95% CI 1.02–2.70, high-certainty evidence, 378 clinical pregnancies) — if 20.1% of clinical pregnancies on IVF end in miscarriage, the comparable figure on IVM could run as high as 40.4%.

A separate, non-randomized data source points the same direction while adding an important nuance. A retrospective cohort of 1,707 cycles at a tertiary hospital, comparing 463 IVM cycles against 1,244 conventional-stimulation cycles in women predicted to over-respond, found cumulative ongoing pregnancy at 42.8% on IVM against 63.8% on stimulation overall — but that gap narrowed sharply as baseline AMH rose, and among women with AMH above 10 ng/mL specifically, the rates converged to 51.1% on IVM against 60.4% on stimulation. Because this is a retrospective comparison rather than a randomized trial, and 93.7% of the IVM group already carried a Rotterdam PCOS diagnosis against 43.8% of the stimulation group, the two groups likely differed in other ways clinics used to decide who got offered which approach — a confounding pattern this study design cannot fully rule out, unlike the randomized trial above.

What Does the International PCOS Guideline Actually Recommend?

The 2023 international evidence-based guideline addresses IVM directly, and its own wording states plainly that the evidence favors conventional IVF on effectiveness: it says IVM and ICSI “could be considered” as an alternative to a stimulated IVF/ICSI cycle, explicitly “acknowledging there is no risk of ovarian hyperstimulation syndrome, but a lower cumulative live birth rate.” The same guideline adds three conditions rather than a blanket recommendation. First, IVM should only be undertaken in services with sufficient expertise, with the guideline calling for dedicated regional or national centres rather than routine availability at any IVF clinic. Second, it names a specific candidate group: women with a history of severe OHSS, or where the risk of severe OHSS is judged unacceptably high, in a service where that expertise already exists. Third, it states outright that the trial evidence available shows IVM/ICSI performing below standard IVF/ICSI on both clinical pregnancy and live birth rate per patient — the same conclusion the trial and meta-analysis data above independently reached.

Does a Newer IVM Protocol Close the Gap With IVF?

A newer biphasic IVM technique, called capacitation IVM or CAPA-IVM, produced oocyte maturation rates of roughly 60–70% against roughly 50% for standard IVM protocols, and live birth rates after a single retrieval in the 20–48% range, according to a 2026 systematic review of published IVM protocols in PCOS — a range that, at its upper end, sits closer to conventional IVF’s per-cycle live birth rates than older IVM data suggested. The same review reports obstetric, neonatal, epigenetic and child-neurodevelopmental outcomes — now followed to age 5–7 in some cohorts — as reassuring, with no signal of harm distinct from IVF.

Live birth rate after a single retrieval sits at roughly 20–48% across all three protocols in that same review, and the review is explicit that the spread comes from differences in priming strategy, culture system, and embryo transfer policy between clinics — not from one protocol being uniformly superior in every reported cohort.

A more recent randomized trial tested whether CAPA-IVM could be run entirely without hormone priming. Women with PCOS undergoing CAPA-IVM were randomized to either no hormonal priming at all or two days of FSH injections before egg collection, and the hormone-free approach produced a live birth and ongoing pregnancy rate of 38.3% against 31.7% with FSH priming — a difference too small to call statistically significant (risk ratio 1.21, 95% CI 0.74–1.98) with the number of women studied, but notable because it suggests the priming step itself, not just the IVM approach broadly, may not be doing the work some protocols assume.

This is a genuinely small evidence base, and it is worth saying so rather than rounding the newer numbers into a settled improvement. The CAPA-IVM live-birth figures above come from a small number of specialist centres, mostly a single research group in Vietnam and its collaborators, and the fully hormone-free comparison rests on one randomized trial. A technique performing well in the hands of the clinics that developed it is not the same claim as a technique performing that well everywhere it is offered — the guideline’s own insistence on “sufficient expertise” exists precisely because outcomes reported from developer centres do not necessarily generalise.

Who Does IVM Not Help, and What Does It Not Solve?

IVM does not address anything about PCOS other than the stimulation-related OHSS risk during an ART cycle, and it sits within the wider PCOS fertility picture as one option among several, not a replacement for the rest of it. It will not improve egg quality — there is no quality deficit for IVM to correct. It is not a treatment for anovulation outside of an ART cycle, and it is not a substitute for ovulation-induction medication or for IUI as a less invasive first step for couples who have not yet needed IVF-level intervention at all.

It also is not available everywhere. The guideline’s call for dedicated centres of expertise reflects a real, current limitation: IVM culture systems, prematuration protocols and outcomes vary enough by clinic that the 2023 guideline explicitly declines to recommend it as a routine, widely offered alternative. And it does not remove the transfer-timing decision that follows any IVM or IVF cycle — freeze-all versus fresh embryo transfer is a separate question with its own evidence base, not one IVM resolves on its own.

Common Questions