A Tangle of Blood Vessels in the Brain, and the New Ability to See It Coming

What families need to know about recent research on vein of Galen malformation, from scans before birth to treatment in the womb and measurements in the nursery

A vein of Galen malformation is a rare tangle of blood vessels deep inside a baby's brain that lets blood rush straight from the arteries into one large vein, forcing the heart to work far harder than it should. Four studies published in 2025 and 2026 show that doctors can now predict before birth which babies will be most severely affected, have begun treating a few of them while they are still in the womb, and can measure the strain on the circulation more precisely after birth.

What the malformation is, and why it affects the heart

In a healthy body, blood leaving the heart travels down arteries that branch into capillaries — vessels so fine that blood slows almost to a trickle before it collects in the veins and returns to the heart. In a vein of Galen malformation, that braking system is missing in one place deep in the brain. Arteries connect directly to a large vein, and blood pours through at high speed and high volume.

Two things follow. The heart must pump an enormous extra volume every minute to keep up, which is why a baby with a brain problem can arrive in the delivery room in heart failure, breathing hard and needing help almost immediately. And so much blood is diverted through the shortcut that other parts of the body — including other parts of the brain — can be left short. Doctors call this "steal," and it is why the brain itself can be injured by what is, at heart, a plumbing fault.

Four recent research papers, taken together, tell a connected story: a study of treating the condition before birth from a hospital in Boston [1], a detailed report of a single baby treated before birth in São Paulo, Brazil [2], a study from Memphis, Tennessee, about measuring the "steal" with a routine heart scan [3], and a large review pulling together every published case of brain vessel malformations with heart problems in children [4].

How families and doctors used to face this diagnosis

It helps to know how recently this was a condition with almost nothing to offer. The malformation forms very early — around the eighth to eleventh week of pregnancy, when a vein that should shrink away and disappear instead stays open because blood is rushing through it [5]. Nothing a parent did or did not do causes this. By the time it can be seen on a scan, usually in the last third of pregnancy, the anatomy is already set.

For a long time, surgeons tried to reach the tangle by opening the skull. In some series more than nine in ten babies died, and the approach was abandoned [2]. From the 1990s a different method took over: threading a very fine tube through a blood vessel in the groin, up into the brain, and blocking the abnormal connections from the inside. This is called embolization, and it transformed survival. Around the same time, a French team at Bicêtre Hospital created a scoring system that grades how well a newborn's heart, brain, lungs, liver and kidneys are working, and that score has guided when to do the procedure ever since — very low scores suggesting treatment cannot help, middle scores meaning treat urgently, and high scores meaning the baby is stable enough to wait several months [6]. The condition is rare — estimates range from about one baby in every 25,000 to 50,000 [2] to fewer than one in 60,000 [1] — but it is the most common malformation of the brain's blood vessels that babies are born with [7].

But progress then stalled. When researchers gathered together the outcomes of babies diagnosed before birth, only about 35 in every 100 survived their first year without significant brain injury [8]. Specialist centres do better than that [9], but the underlying problem remained: for the most severely affected babies, waiting until birth means some damage has already happened. Families were often told that everything depended on how the first days went, with little ability to say in advance which group their baby would fall into.

The measurement that changed the conversation

The breakthrough was surprisingly simple. Blood leaving the malformation drains away through a channel called the falcine sinus, and doctors noticed that the width of that channel on a scan before birth predicted how sick the baby would be afterwards. A width of 8 millimetres or more identified the high-risk group with about 88% accuracy [10]. A later study of 59 pregnancies confirmed that this one measurement was linked to survival, to brain injury on the first scan after birth, and to whether babies were meeting their developmental milestones at one, six and twelve months [11].

That gave families and doctors something they had never had: a number, available before birth, that says roughly what to expect. It also raised an uncomfortable question. If you can identify the babies who will do worst, and you know that much of the harm happens in the hours and days around birth, should you try to act sooner?

Treating before birth: what actually happened

That question became a research study at Boston Children's Hospital and Brigham and Women's Hospital in Boston, Massachusetts, called Fetal Treatment of Galenic Malformations, approved by both hospitals' ethics committees and by the US Food and Drug Administration [12]. Mothers carrying a baby with a falcine sinus of 7 millimetres or more, and without major brain injury already visible, could choose to take part. Using ultrasound to see what they were doing, doctors passed a needle through the mother's abdomen and womb and through the baby's skull, then threaded a fine tube into the abnormal vein and packed it with tiny platinum coils to slow the flow. The first successful case was published in 2023 [13].

The report on the first seven pregnancies is honest about both sides. In five of the seven, the coils went in successfully, and the amount of blood the baby's heart had to pump fell by about a third. Judged by their scan measurements, these babies would have been expected to have a 90% chance of dying and only a 9% chance of meeting their six-month milestones with usual care after birth. What actually happened was that three of the seven babies died and three of the five treated babies were alive and developing normally at 8, 18 and 24 months [1].

Those are meaningfully better numbers than expected, but they come with real costs. Five of the seven mothers went into labour or needed delivery sooner than planned, on average about three days after the procedure, and three of those births were premature. In two pregnancies the procedure could not be completed at all. And four of the seven babies still needed the usual treatment after birth — treating the malformation before birth reduces the problem, it does not remove it.

The Brazilian report tells the same story from a single family's perspective [2]. A first-time mother's baby was treated at 33 weeks of pregnancy. The signs of strain on the baby's heart improved measurably over the following days. The baby was still born needing a breathing tube at 36 weeks, needed a second procedure on the second day of life, later needed a shunt to drain fluid from the brain, and at 15 months has developmental delays. This was a technically successful procedure and a hard road afterwards, and both parts are true at once.

Measuring the strain once a baby is born

The fourth idea in this group of papers is the one most likely to affect ordinary care. Doctors in Memphis, Tennessee, studied twelve babies who all had reassuring scores on the Bicêtre system — the group usually considered stable enough to wait. Using a standard heart ultrasound, they measured how much blood was flowing backwards in the main artery of the body, sucked back toward the malformation, compared with how much was going forward. In more than eight out of ten of these "stable" babies, a large amount was flowing backwards [3].

In plain terms: a baby can look settled and still be quietly losing much of their circulation to the malformation. The measurement needs no special equipment, can be repeated as often as needed, and after the first embolization it dropped by about half — and babies whose measurement fell the most were least likely to need further procedures. The doctors are careful to say this is a small study that has not yet been linked to long-term development, and that blocking a high-flow shortcut too abruptly carries its own risks [14].

The last paper gathered 1,042 children from studies published since 1971 and confirmed what families already sense: this is overwhelmingly a newborn's condition, and the vein of Galen type accounts for around three-quarters of cases where a brain vessel malformation occurs alongside heart problems [4]. Its older survival figures should not be read as today's outlook, because they include decades of cases treated before modern techniques existed.

What this means for families, and what comes next

If your baby has been given this diagnosis before birth, it is reasonable to ask your team for the falcine sinus measurement, because that number is what specialists use to gauge severity and to decide whether a referral for treatment before birth is worth discussing. It is also reasonable to ask that the delivery be planned as a high-risk birth with a heart specialist involved, whatever else is decided, because the medicines used to support this particular kind of heart failure are not the ones used for most others [15].

Researchers are working on several things next. Larger numbers are needed before anyone can say whether treating in the womb truly improves outcomes rather than simply appearing to; the Boston team plans to enrol 20 to 25 pregnancies in total. They are also looking at whether performing the procedure closer to 37 weeks, with a planned delivery a day or two later, would avoid the early labour that troubled this first group. Others are testing whether measurements of the backward flow predict how children are walking, thinking and talking at two years of age — one single-centre team has already reported a strong link between the amount of blood being stolen and development at that age [16]. What has genuinely changed is that this diagnosis is no longer a matter of waiting to see. Much of what will happen can now be estimated in advance, discussed with you, and planned for — and for a small number of families, acted on before their baby is born.

References

  1. Orbach DB, Shamshirsaz AA, Wilkins-Haug L, et al. In utero embolization for fetal vein of Galen malformation. JAMA. 2025;334(10):878–885. doi:10.1001/jama.2025.12363 ↩
  2. Cavalheiro S, da Costa MDS, Nôvo PC, et al. Galen vein malformation—intrauterine treatment. Childs Nerv Syst. 2026;42(1):181. doi:10.1007/s00381-026-07261-5 ↩
  3. Rimler M, Philip R, Tanner L, Huth H, Elijovich L. The aortic flow reversal ratio: a quantitative adjunct to the Bicêtre score in vein of Galen malformation. J Clin Med. 2026;15(2):748. doi:10.3390/jcm15020748 ↩
  4. Ponchio V, Hussin R, Harati W, et al. The relation between cerebral arteriovenous malformations and heart diseases in the pediatric population: a systematic review and meta-analysis. Childs Nerv Syst. 2026;42(1):227. doi:10.1007/s00381-026-07307-8 ↩
  5. Raybaud CA, Strother CM, Hald JK. Aneurysms of the vein of Galen: embryonic considerations and anatomical features relating to the pathogenesis of the malformation. Neuroradiology. 1989;31(2):109–128. doi:10.1007/BF00698838 ↩
  6. Lasjaunias PL, Chng SM, Sachet M, Alvarez H, Rodesch G, Garcia-Monaco R. The management of vein of Galen aneurysmal malformations. Neurosurgery. 2006;59(5 Suppl 3):S184–S194. doi:10.1227/01.NEU.0000237445.39514.16 ↩
  7. Brevis Nuñez F, Dohna-Schwake C. Epidemiology, diagnostics, and management of vein of Galen malformation. Pediatr Neurol. 2021;119:50–55. doi:10.1016/j.pediatrneurol.2021.02.007 ↩
  8. D'Amico A, Tinari S, D'Antonio F, et al. Outcome of fetal vein Galen aneurysmal malformations: a systematic review and meta-analysis. J Matern Fetal Neonatal Med. 2022;35(25):5312–5317. doi:10.1080/14767058.2021.1878494 ↩
  9. Buratti S, Mallamaci M, Tuo G, et al. Vein of Galen aneurysmal malformation in newborns: a retrospective study to describe a paradigm of treatment and identify risk factors of adverse outcome in a referral center. Front Pediatr. 2023;11:1193738. doi:10.3389/fped.2023.1193738 ↩
  10. Arko L, Lambrych M, Montaser A, Zurakowski D, Orbach DB. Fetal and neonatal MRI predictors of aggressive early clinical course in vein of Galen malformation. AJNR Am J Neuroradiol. 2020;41(6):1105–1111. doi:10.3174/ajnr.A6585 ↩
  11. Lazatin PC, Puglisi EN, Meadows JS, Somboonnithiphol K, See AP, Orbach DB. Mediolateral diameter of the falcine sinus as a predictor of clinical outcomes in fetal vein of Galen malformations. Stroke. 2025;56(7):1758–1768. doi:10.1161/STROKEAHA.125.050694 ↩
  12. See AP, Wilkins-Haug LE, Benson CB, Tworetzky W, Orbach DB. Percutaneous transuterine fetal cerebral embolisation to treat vein of Galen malformations at risk of urgent neonatal decompensation: study protocol for a clinical trial of safety and feasibility. BMJ Open. 2022;12(5):e058147. doi:10.1136/bmjopen-2021-058147 ↩
  13. Orbach DB, Wilkins-Haug LE, Benson CB, et al. Transuterine ultrasound-guided fetal embolization of vein of Galen malformation, eliminating postnatal pathophysiology. Stroke. 2023;54(6):e231–e232. doi:10.1161/STROKEAHA.123.043421 ↩
  14. Bhatia K, Mendes Pereira V, Krings T, et al. Factors contributing to major neurological complications from vein of Galen malformation embolization. JAMA Neurol. 2020;77(8):992–999. doi:10.1001/jamaneurol.2020.0825 ↩
  15. Cory MJ, Durand P, Sillero R, Morin L, Savani R, Chalak L, Angelis D. Vein of Galen aneurysmal malformation: rationalizing medical management of neonatal heart failure. Pediatr Res. 2023;93(1):39–48. doi:10.1038/s41390-022-02064-1 ↩
  16. Schwarz S, Brevis Nuñez F, Dürr NR, et al. Association of disease severity and cerebral interventional complications with long-term neurodevelopmental outcome in neonates with vein of Galen malformation. Pediatr Neurol. 2025;169:5–12. doi:10.1016/j.pediatrneurol.2025.05.002 ↩