Twins With PCOS Fertility Treatment: The Real Multiple-Pregnancy Rates

How Much Does the Chance of Twins Go Up With PCOS Fertility Treatment?

It roughly quadruples from the gentlest option to the most aggressive one: a 2022 Cochrane review of 14 trials in 2,247 women with anovulatory PCOS found a multiple-pregnancy rate of 1.6% with letrozole and 2.2% with older oral drugs, while a 2022 meta-analysis of 45 studies comparing embryo-transfer numbers found 16.7% with a two-embryo IVF transfer against 0.7–1.0% with a single embryo. No one had previously lined up the full ladder — oral induction, gonadotropins, IUI, IVF — against primary sources built specifically to measure this outcome. Table 1 does that.

Every number below describes what happened in the population that generated it, not a personal prediction. Read the population column as carefully as the percentage; two “multiple-pregnancy rate” figures from different patient groups are not interchangeable.

Two things worth reading twice in that table. First, the single landmark trial of letrozole and clomiphene found roughly triple the multiple-pregnancy rate that the larger pooled analysis did (7.4% versus 2.2% for clomiphene) — both numbers are real, and the honest reading is that a single trial’s exact figure moves around more than a 14-trial pool does. Second, the IVF row is not PCOS-specific; it is included because it is the largest, cleanest evidence available on how embryo-transfer number alone drives multiple pregnancy, and because the 2023 international guideline explicitly flags gonadotropins and clomiphene as multiple-pregnancy risks in PCOS specifically, without publishing its own percentage.

Why Does the Multiple-Pregnancy Rate Climb as Treatment Gets More Intensive?

Every step up this ladder increases the number of eggs or embryos in play at once, and that number — not the diagnosis of PCOS itself — is what drives the twin rate. Oral agents at standard use aim for one mature follicle per cycle; letrozole’s slightly lower multiple-pregnancy rate than clomiphene in the pooled data above is thought to reflect a shorter half-life and a different hormonal mechanism that makes it somewhat less prone to recruiting a second follicle, though the guideline-level evidence does not resolve this to a precise mechanism. Gonadotropins act directly on the ovary and can recruit more than one follicle per cycle if a course is not closely monitored, which is why the 2023 guideline names gonadotropins as carrying a materially higher multiple-pregnancy rate than laparoscopic ovarian surgery, an alternative second-line option that does not carry the same risk. In the Huang et al. 2018 IUI cohort above, cycles that produced more than one mature follicle resulted in multiple pregnancy 15.5% of the time versus 0.8% for single-follicle cycles — a nearly twentyfold difference that had nothing to do with which drug was used and everything to do with follicle count. In IVF, the mechanism is the most direct of all: transferring two embryos instead of one means two independent chances at implantation, which is exactly what the 16.7%-versus-0.7% gap in Table 1 is measuring. None of this is a protocol to follow — monitoring intensity, cycle cancellation, and how many follicles or embryos to proceed with in a given cycle are decisions for the treating clinician, not choices to make from a table on this page.

Most twins from fertility treatment are fraternal — dizygotic, from two separate eggs or embryos — which is the direct arithmetic consequence described above. A smaller share are identical, and single-embryo transfer does not fully eliminate that possibility, since one embryo can still split after transfer.

What Does a Twin Pregnancy Actually Risk, Beyond Two Babies at Once?

A twin pregnancy carries a materially higher rate of nearly every major complication than a singleton pregnancy does, and treatment-conceived twins carry a further increase on top of that baseline. A 2024 meta-analysis of 111 studies covering 802,462 pregnancies found that twins conceived through assisted reproduction had higher odds of preterm birth before 37 weeks (OR 1.26, 95% CI 1.19–1.33, across 70 studies), preterm birth before 34 weeks specifically (OR 1.33, 95% CI 1.14–1.56, 29 studies), hypertensive disorders including pre-eclampsia (OR 1.29, 95% CI 1.14–1.46, 59 studies), gestational diabetes (OR 1.61, 95% CI 1.48–1.75, 51 studies), and NICU admission (OR 1.24, 95% CI 1.14–1.35, 32 studies) compared with twins conceived without fertility treatment. That is the treatment-specific increase; it sits on top of the already-elevated baseline risk that any twin pregnancy carries, treatment-conceived or not.

Preterm labor and pre-eclampsia are the two complications above worth being able to recognise, not just quote as a percentage. Preterm labor feels like regular tightening or cramping that does not ease with rest, low back ache that comes in waves, increased pelvic pressure, or a change in vaginal discharge — any of these before 37 weeks in a twin pregnancy is worth same-day contact with your maternity team, not a wait-and-see. Pre-eclampsia can be silent early on, which is exactly why blood pressure and urine protein are checked at every antenatal visit, but once it becomes symptomatic it produces a headache that will not lift with paracetamol, visual disturbances such as flashing lights or blurring, sudden swelling in the face or hands, or pain under the right ribs — all reasons to be seen the same day, not at the next scheduled appointment.

Does PCOS Itself Add Risk on Top of the Twin Pregnancy?

Not necessarily, according to the one study that has directly compared PCOS and non-PCOS twin pregnancies against each other. A Danish registry study of 360 women with dichorionic twin pregnancies — 72 with PCOS and 288 without — found comparable rates of pre-eclampsia, gestational diabetes and preterm delivery between the two groups after adjusting for BMI and mode of conception. The PCOS group in that study did have a lower average BMI than the general obstetric comparison group, which the study’s own authors flag as a likely reason the PCOS-specific effect may be underestimated — a healthier-than-typical PCOS population is not the same as PCOS carrying no added risk in a broader population. Read plainly, the honest finding is: once a twin pregnancy exists, most of its added risk appears to come from being a twin pregnancy, not from the underlying PCOS diagnosis — but the evidence for that conclusion is thin, from a single study of 72 women, and has not been repeated in a larger or more diverse cohort.

Why Are Some Clinics Now Defaulting to a Single Embryo Instead of Two?

The gap is large enough that transferring one embryo instead of two lowers the multiple-pregnancy rate by roughly 95% in pooled data: the 2022 meta-analysis of 45 studies found an odds ratio of 0.05 (95% CI 0.04–0.06) for multiple pregnancy with single-embryo transfer compared with double, alongside lower rates of cesarean delivery, low birth weight, and NICU admission in the same review. That evidence, not a change in how well embryos implant, is what has moved clinical practice broadly toward single-embryo transfer as a default in many settings over the past decade — the same review found implantation rates were statistically similar between single and double transfer (34.7% vs 34.5%), meaning the live-birth advantage of transferring two embryos comes specifically from a second independent chance at pregnancy, with a real chance of it being a twin pregnancy rather than a second attempt at a singleton one. Which embryo-transfer number is right for a specific IVF cycle depends on embryo quality, age, and prior history — a decision for the treating clinic, not a rule this page can set.

When Is the Multiple-Pregnancy Risk Actually Lowest?

The lowest risk on this entire ladder belongs to letrozole taken as directed and monitored for a single dominant follicle: 1.6% in the pooled trial evidence above, barely above the general-population multiple-birth rate. Three situations meaningfully lower multiple-pregnancy risk regardless of where someone starts: oral ovulation induction over injectable gonadotropins, a monitored cycle that gets cancelled or converted when more than one follicle responds, and a single-embryo IVF transfer over a two-embryo one. None of these are universally right — a two-embryo transfer can be the better-reasoned choice for someone with a lower-quality embryo pool or a specific clinical history, and that tradeoff belongs in a conversation with a fertility specialist, not a general rule. What this page can say plainly is that the multiple-pregnancy rate is not fixed once someone starts fertility treatment; it moves with specific, nameable choices at each step of the ladder, and every one of those choices carries a tradeoff against the chance of pregnancy at all.

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 still applies under either name — only the label changed.

For what a singleton PCOS pregnancy risks on its own — separate from the multiples question this page covers — the four singleton pregnancy risks worth monitoring in PCOS walks through gestational diabetes, hypertensive disorders, preterm birth and miscarriage with the same kind of primary-source numbers. Anyone weighing where to start on the ladder above may also want the letrozole success-rate breakdown, the clomiphene-versus-letrozole comparison, or the wider PCOS fertility treatment overview this page sits inside, alongside every other page on the PCOS fertility treatment hub.