Can One Embryo Split Into Twins During IVF? Monozygotic Timing, Risks, and Likelihood

Yes—one embryo can split into twins during IVF, though it is uncommon. A single transferred embryo can divide into two separate fetuses, producing monozygotic twins that are genetically nearly identical. This happens after fertilization and usually before implantation, when cells of the inner cell mass separate or two embryonic axes form. In IVF, the likelihood is roughly 1% to 3% of clinical pregnancies after single embryo transfer, higher than the 0.4% to 0.5% background rate in spontaneous conception. The exact trigger remains uncertain, and no current lab test can predict which embryo will split.

Mechanism: Can One Embryo Split Into Twins During IVF?

A single transferred embryo is one organism, derived from one egg and one sperm. In standard IVF, dizygotic twins require two embryos, because two separate eggs are fertilized. Monozygotic twins arise after fertilization, when one embryo’s cells separate into two developmental axes. The split can occur at the cleavage stage, at the morula stage, or inside the blastocyst. It is not a routine laboratory step, and no embryologist intentionally divides an embryo during standard IVF.

The timing of that separation determines the membranes around the twins. A zygote forms on day 1 after fertilization. Cleavage divisions follow on days 2 and 3, the morula forms around day 4, and the blastocyst forms on days 5 and 6. Hatching through the zona pellucida occurs around days 6 and 7, and implantation begins around days 7 to 10. If the split happens before trophoblast differentiation, each twin usually gets its own chorion and amnion, creating dichorionic diamniotic twins. If the split happens after trophoblast formation but before amnion formation, the twins share one chorion and have two amnions. If the split happens after amnion formation, the twins share one chorion and one amnion.

The membrane pattern is clinically important. Dichorionic diamniotic monozygotic twins account for roughly 25% to 30% of monozygotic twin pregnancies. Monochorionic diamniotic twins account for roughly 70% to 75%. Monochorionic monoamniotic twins account for about 1% to 2%. Conjoined twins are much rarer, under 1% of monozygotic twin pregnancies. In plain terms: earlier splits tend to create more membranes, and later splits tend to create fewer membranes and higher risk.

The biological mechanism is not fully settled. One leading explanation is that the inner cell mass splits into two separate embryonic discs before gastrulation. Another explanation involves herniation of the blastocyst through the zona pellucida during hatching, which may create two centers of development. Animal models support both ideas, but human evidence remains limited. Researchers treat these as plausible mechanisms rather than proven causes. The split is a developmental accident, not a designed outcome of IVF.

Timing of Monozygotic Splitting

Monozygotic splitting is a time-dependent process, not a single event at one fixed moment. The stage at which separation occurs predicts the membrane anatomy and the level of shared placental circulation. That is why an early ultrasound matters: it can identify chorionicity before the membrane pattern becomes harder to classify. A first-trimester scan at 7 to 9 weeks can count gestational sacs and look for the lambda sign or T sign. Later scans can still help, but first-trimester assessment is more reliable.

The table shows why “identical twins” is not one uniform category. A monochorionic diamniotic pair shares a placenta but has separate amniotic sacs. A monochorionic monoamniotic pair shares both placenta and sac, which raises the risk of cord entanglement. A dichorionic diamniotic pair has separate placentas and separate sacs, even though the twins came from one embryo. These differences drive surveillance intensity and delivery planning.

Likelihood After Single Embryo Transfer

The likelihood of monozygotic twinning after IVF is higher than the background rate in spontaneous conception, but the absolute risk remains small. Estimates vary by definition, registry, embryo stage, and clinic. A clinical pregnancy rate is not the same as a live-birth rate, and a twin pregnancy diagnosed at 7 weeks may later become a singleton after a vanishing twin. These differences explain why reported ranges overlap. They do not mean that the biology changes from one study to the next.

The overlapping ranges are not contradictory. The 1% to 3% band reflects all single embryo transfers, including cleavage and blastocyst transfers. The 2% to 5% band reflects blastocyst-only cohorts, where extended culture and hatching are more common. A single blastocyst transfer can therefore sit at the upper end of the single-embryo range. At a 2% risk, one in 50 single embryo transfer pregnancies would be a monozygotic twin pregnancy. At a 3% risk, one in 33 would be.

This distinction matters for counseling. Per the 2021 ASRM guidance on embryo transfer limits, single embryo transfer is recommended for many good-prognosis patients because it sharply lowers dizygotic twins and higher-order multiples. It does not reduce monozygotic twinning to zero. A clinic can lower the total twin rate from roughly 20% to 30% after double embryo transfer to about 1% to 3% after single embryo transfer. That is a large absolute reduction, but it is not elimination.

Variations and Edge Cases

Blastocyst transfer is the most discussed variation. Extended culture to day 5 or day 6 allows embryos to reach a more advanced stage before transfer, and it improves selection in some cases. Several studies and reviews report a higher monozygotic twinning rate after blastocyst transfer than after cleavage-stage transfer. The absolute increase is small, and confounding is possible because clinics and patients differ. The hypothesis is that hatching and manipulation of the zona pellucida may increase the chance of inner cell mass splitting. The evidence is supportive but not definitive.

Assisted hatching is another edge case. The procedure creates a small opening in the zona pellucida to help the embryo hatch. Some studies report more monozygotic twins after assisted hatching, while others do not. The effect, if real, is modest and inconsistent. ICSI has also been studied, but it is not consistently linked to monozygotic twinning. Frozen-thawed embryo transfer does not eliminate the possibility. Can one embryo split into twins during IVF after a frozen transfer? Yes. Vitrification and warming change the logistics, not the fundamental developmental possibility.

Preimplantation genetic testing for aneuploidy, or PGT-A, is a common source of confusion. PGT-A screens an embryo for chromosome number and selects embryos with a normal chromosome count. It does not prevent the embryo from splitting after transfer. Embryo biopsy may be associated with small changes in outcome in some studies, but the data are not consistent enough to claim a clear monozygotic twinning effect. A euploid blastocyst can still split. A mosaic embryo can also split, though transfer decisions for mosaics are separate and individualized.

Monozygotic twins are not truly identical in every cell. Post-zygotic mitotic errors can create copy-number differences between the two fetuses. Epigenetic marks can diverge after separation. Monozygotic twins can be discordant for structural anomalies, chromosome abnormalities, or imprinting disorders. This is why a monochorionic twin pregnancy is not treated as two copies of one low-risk fetus. The shared placenta and the potential for unequal placental sharing create additional problems that a singleton pregnancy does not have.

Risks and Monitoring

Twin pregnancy carries higher maternal and fetal risks than singleton pregnancy, whether the twins are monozygotic or dizygotic. Twin pregnancies have a preterm birth rate before 37 weeks of roughly 50% to 60%, compared with about 10% for singletons in many high-income settings. Low birth weight under 2,500 grams occurs in about half of twins. Preeclampsia, gestational diabetes, cesarean delivery, and postpartum hemorrhage are all more common. These risks are not caused by IVF alone; they are driven largely by carrying two fetuses.

Monochorionic twins carry additional risks because they share a placenta. Vascular anastomoses connect the two fetal circulations, which can cause twin-twin transfusion syndrome in about 10% to 15% of monochorionic diamniotic pregnancies. Twin anemia-polycythemia sequence occurs in roughly 3% to 5%. Selective fetal growth restriction occurs in about 10% to 15%. Twin reversed arterial perfusion sequence is rare, under 1%. Monoamniotic twins face cord entanglement risk, and conjoined twins face complex surgical and neonatal issues. These are the reasons chorionicity assignment is not a technicality.

Monitoring is stratified by chorionicity. A first-trimester ultrasound should assign chorionicity and amnionicity. For monochorionic diamniotic twins, the 2017 ISUOG guidelines recommend ultrasound surveillance every two weeks from 16 weeks of gestation. Surveillance includes assessment of amniotic fluid volume, fetal bladders, growth, and Doppler studies when indicated. Monoamniotic twins usually require more intensive monitoring because of cord entanglement risk. Dichorionic twins generally need less frequent surveillance unless growth discordance or another complication appears.

Misconceptions and Professional Caveats

One common misconception is that single embryo transfer guarantees a singleton pregnancy. It does not. Single embryo transfer removes the risk of dizygotic twins from two separate embryos, but it leaves the 1% to 3% monozygotic twinning risk. Another misconception is that the IVF laboratory splits embryos on purpose. Standard IVF does not intentionally split an embryo. The laboratory handles embryos for fertilization, culture, biopsy, freezing, and transfer, but the split itself is a biological event.

A third misconception is that identical twins are always monochorionic. Roughly 25% to 30% of monozygotic twins are dichorionic diamniotic, meaning they have separate placentas and sacs. A fourth misconception is that PGT-A prevents twinning. PGT-A tests chromosome number; it does not control how the inner cell mass separates after transfer. A fifth misconception is that blastocyst transfer is always safer because it lowers multiple pregnancy. Blastocyst transfer with single embryo transfer lowers total multiple risk compared with double embryo transfer, but blastocyst culture may carry a higher monozygotic twinning rate than cleavage-stage culture.

Can one embryo split into twins during IVF because of assisted hatching? It is possible that hatching-related manipulation contributes in some cases, but the evidence is mixed and causality is unproven. Clinics should counsel patients before transfer without promising zero twin risk. No validated biomarker, embryo score, or lab test predicts which embryo will split. The professional caveat is straightforward: monozygotic twinning is uncommon, biologically plausible, and not fully preventable with current technology.

Practical Actions and Remaining Limits

Patients and clinicians can act on what is measurable. Ask the clinic about its single embryo transfer policy, its reported monozygotic twinning rate, and whether blastocyst transfer is planned. Confirm chorionicity early with a first-trimester ultrasound at 7 to 9 weeks. If monochorionic diamniotic twins are found, arrange maternal-fetal medicine care and follow the two-week surveillance schedule from 16 weeks. Do not rely on bed rest, dietary changes, or supplements to prevent a split; no evidence supports those measures. If a twin pregnancy occurs after single embryo transfer, the right response is risk assessment and monitoring, not blame.

Seek urgent care for these red flags:

• Severe vomiting, dehydration, or inability to keep fluids down
• Rapid abdominal swelling or severe abdominal pain
• Vaginal bleeding or unexplained fluid leakage
• Severe headache, visual changes, or upper abdominal pain
• Reduced fetal movement after 24 weeks
• Fever, uterine contractions, or signs of preterm labor

The remaining limits are real. Incidence estimates vary because studies define monozygotic twinning differently and because clinics transfer different embryo stages. The mechanism is still inferred from animal models and human observational data. No intervention has been shown to prevent a monozygotic split after transfer. The practical takeaway is that one embryo can become two during IVF, the risk is low but not zero, and early ultrasound plus chorionicity-based monitoring is the best available response.